Brian Mast LOSOM Scoping Comment

MEMORANDUM:
NEPA Scoping Process Public Comment

Prepared by U.S. Congressman Brian Mast
Prepared for the U.S. Army Corps of Engineers
April 22, 2019
TABLE OF CONTENTS
I. AUTHORIZED PROJECT PURPOSES ......................................................................... 2

a. Water Quality ....................................................................................................................... 2
b. Preservation of Everglades National Park ........................................................................... 3
II. PUBLIC HEALTH ............................................................................................................. 3
a. Connection Between Discharges and Public Health Threats ............................................... 3
b. Health Standards .................................................................................................................. 4
c. Additional Impacts ............................................................................................................... 6
III. WATER QUALITY ........................................................................................................... 9
a. Sources of Pollution ........................................................................................................... 10
b. Misinformation Surrounding Septic Tanks ........................................................................ 11
c. Impacts of Freshwater Infusion and Pollution ................................................................... 12
IV. ADDITIONAL ENVIRONMENTAL CONCERNS ..................................................... 14
V. RECREATIONAL ACTIVITIES AND ECONOMIC IMPACTS .............................. 15
VI. ENHANCMENT OF FISH AND WILDLIFE ............................................................... 17
VII. RECOMMENDATIONS ................................................................................................. 18
a. Maximum Flows South of the Lake................................................................................... 19
b. Added Flexibility at the High End of the Operational Band ............................................. 19
c. Added Flexibility at the Low End of the Operational Band .............................................. 20
d. Operate at a Routinely Lower Level than LORS 2008 ...................................................... 20
e. Incorporate Human Health and Safety Protections ............................................................ 21
f. Zero Discharges to the St. Lucie Estuary and Beneficial Flows to the Caloosahatchee ... 22
VIII. UNRESOLVED QUESTIONS ........................................................................................ 22
IX. CONCLUSION ................................................................................................................. 26
1
April 22, 2019
I. AUTHORIZED PROJECT PURPOSES
In a post entitled “Why we release water,” published on June 15, 2018, then Jacksonville District
Commander Col. Jason Kirk acknowledged “that large amounts of freshwater into estuaries that
are supposed to be brackish isn’t good for the aquatic plant life or wildlife,” but according to him
that does not matter because “The LORS and the Corps’ Congressional authorities call for our
paramount consideration to be the reduction of flood risk.”1 This assertion is either misinformed
or willfully ignorant. In fact, it is in direct contravention of Congressional intent. Congress has
not placed any of the authorized project purposes as more important than any others.
LORS itself ignores important Congressionally-authorized project purposes. LORS states that
“the Congressionally-authorized project purposes for Lake Okeechobee, the Okeechobee
Waterway (OWW) and the Everglades Agricultural Area (EAA)…include flood control;
navigation; water supply for agricultural irrigation, municipalities and industry, the Everglades
National Park (ENP), regional groundwater control, and salinity control; enhancement of fish
and wildlife; and recreation.”2 This list of authorized project purposes is insufficient and
incomplete.
Notably absent from the list are the authorized project purposes of Water Quality—derived from
P.L. 90-483 § 203, P.L. 92-500 and P.L. 95-217—and Preservation of Everglades National
Park—derived from P.L. 90-843 § 203, P.L. 101-2293. LOSOM must take into account these
additional authorized project purposes, and to the extent that any authorized project purpose is
considered “paramount,” these must be given at least equal weight.4
a. Water Quality
In a letter dated June 22, 2018, Colonel Jason A. Kirk stated that “while water quality is a
consideration, addressing water quality is not a federally authorized project purpose and not a
primary factor in determining how much water to release.”5 P.L. 90-483 § 203, however,
dictates numerous authorized project purposes for the Central and Southern Florida Plan. These
include consideration of “the quality of the waters of the area.”6 This definition encompasses
whether the water is “generally good for recreational purposes such as swimming and fishing.”7
The definition also includes whether the “river and canal systems” are “generally polluted.”8
The report also includes how the “intrusion of brackish water” is “impeded by control structures
1
Col. Jason Kirk, U.S. Army Corps of Engineers, Why we release water (June 15, 2018),
https://www.saj.usace.army.mil/Media/News-Stories/Article/1552379/why-we-release-water/.
2
U.S. Army Corps of Engineers, Water Control Plan for Lake Okeechobee and Everglades Agricultural Area (Mar
2008), https://usace.contentdm.oclc.org/utils/getfile/collection/p16021coll7/id/8423.
3
See P.L. 90-483 § 203, P.L. 92-500, P.L. 95-217, P.L. 101-229.
4
Colonel Jason A. Kirk, Letter to Congressman Brian Mast (July 5, 2018). See Appendix B.
5
Id.
6
P.L. 90-483 § 203 at 46.
7
Id.
8
Id.
2
April 22, 2019
or fresh-water flow.”9 Each of these categories necessitate that the Army Corps incorporate as a
“paramount” consideration the water quality impacts of discharges into the St. Lucie and
Caloosahatchee Rivers.
P.L. 92-500 and P.L. 95-217 dictate the Army Corps’ requirements under the Clean Water Act.10
b. Preservation of Everglades National Park
P.L. 90-483 and P.L 101-229 both ensure that the preservation of Everglades National Park
generally is an authorized project purpose.11 This authorized project purpose includes taking
“steps to restore the natural hydrological conditions within the park.”12 The discharges of
massive quantities of water to the coasts, or the injection of water underground where it cannot
be recovered, is antithetical to the Army Corps’ authorized project purpose of preserving
Everglades National Park and protecting the water supply of Everglades National Park.
II. PUBLIC HEALTH
NEPA requires the Army Corps to consider “the degree to which the proposed action affects
public health or safety.”13 Although the Army Corps has consistently insisted that they have a
duty to protect public health and safety only as it relates to flood risk management14, NEPA
supersedes the authorized project purposes and requires that public health for all people is
considered before LOSOM can be finalized.15 Moreover, the Army Corps is required to consider
the human environmental impacts, including the human health risks associated with human
interaction with those water bodies.16
a. Connection Between Discharges and Public Health Threats
The flow of toxic algae from Lake Okeechobee to the St. Lucie and Caloosahatchee Rivers
during discharges is well established. For example, a report completed for the Florida legislature
by the Water Institute at the University of Florida found that “freshwater inflows…have
transported toxic microalgae into the St. Lucie estuary, with a substantial bloom of Microcystic
9
Id.
10
See P.L. 92-500, P.L 95-217.
11
See P.L. 90-483, P.L. 101-229.
12
P.L. 101-229 § 103.
13
40 C.F.R. § 1508.27.
14
COL Andrew D. Kelly, Jr., U.S. Army Corps of Engineers, Lake Okeechobee System Operation Manual
(LOSOM) A Component of the Central & Southern Florida (C&SF) System Operating Plan Florida Congressional
Delegation Briefing (Jan 29, 2019). See Appendix C.
15
Id.
16
40 C.F.R. § 1508.14.
3
April 22, 2019
aeruginosa being recorded in 2005.”17 Since then, excessive discharge events in 2013, 2016 and
2018—just to name a few—have resulted in extremely harmful algal bloom events.18
In 2018, the U.S. Geological Survey, in coordination with the U.S. Army Corps of Engineers,
released a report that found that “as freshwater cyanobacteria are transported to brackish and
marine waters, there will be a loss of membrane integrity which will lead to the release of
cellular microsystin into the surrounding waterbody.”19 While numerous additional factors do
contribute to the proliferation of harmful algal blooms, the causal connection between discharges
and the blooms is scientifically indisputable—as evidenced by the U.S. Army Corps of
Engineers own study.
b. Health Standards
The scope of the scientific understanding of public health risks related to cyanobacteria has
likewise increased substantially since LORS was put into place. In fact, the words “algal” or
“algae” appear only five times in LORS and the word “cyanobacteria” never appears.20 In fact,
the Final Supplemental Environmental Impact Statement published in November 2017
incorrectly states that “it is unlikely that discharges from Lake Okeechobee are a prerequisite for
HAB formation.”21 The U.S. Army Corps of Engineers own scientific studies now disprove this
point.22 Moreover, when algae is discussed, LORS discusses only the possibility that algal
blooms may develop and proposes that “short-term high rates of release from Lake Okeechobee
are often effective at breaking up such algal blooms.”23 As we have seen repeatedly over the
past decade, including during horrific outbreaks in 2016 and 2018, this is woefully insufficient.
LOSOM must correct this injustice.
As you know, concentrations of toxins in blue-green algae use a measurement of parts per
billion. According to the World Health Organization, water with more than one part per billion
is unsafe to drink and water with more than 10 parts per billion is unsafe for recreational
17
University of Florida Water Institute, Options to Reduce High Volume Freshwater Flows to the St. Lucie and
Caloosahatchee Estuaries and Move More Water from Lake Okeechobee to the Southern Everglades (Mar 2015),
https://waterinstitute.ufl.edu/faculty/graham/wp-content/uploads/UF-Water-Institute-Final-Report-March-2015.pdf.
See Appendix D.
18
See, e.g., TCPalm, Highly toxic blue-green algae at dam where Lake O waters enter St. Lucie River (Aug 29,
2018), https://www.tcpalm.com/story/news/local/indian-river-lagoon/health/2018/08/29/dep-highly-toxic-blue-
green-algae-dam-leading-st-lucie-river/1131439002/. See Appendix E.
19
U.S. Department of the Interior, Understanding the Effect of Salinity Tolerance on Cyanobacteria Associated with
a Harmful Algal Bloom in Lake Okeechobee, Florida (2018), https://pubs.usgs.gov/sir/2018/5092/sir20185092.pdf.
See Appendix F.
20
LORS, supra note 2.
21
U.S. Army Corps of Engineers, Final Supplemental Environmental Impact Statement: Lake Okeechobee
Regulation Schedule (Nov 2007),
https://www.saj.usace.army.mil/Portals/44/docs/h2omgmt/LORSdocs/ACOE_STATEMENT_APPENDICES_A-
G.pdf.
22
U.S. Department of the Interior, supra note 19.
23
Id at 7-6.
4
April 22, 2019
contact.24 After nearly two months of discharges, at the beginning of August 2018, algae in the
St. Lucie River was ten times too toxic to touch based on this measurement.25 By the end of the
month, algae in the St. Lucie River had tested at nearly 50 times too toxic for recreational
contact.26
In 2015, the United States Environmental Protection Agency (EPA) released the “Algal Toxin
Risk Assessment and Management Strategic Plan for Drinking Water” pursuant to P.L. 114-45.27
The report stated “one of the strongest drivers for changes that may be required to prevent future
HABs, and/or mitigate those that occur, is the threat of serious adverse health effects in exposed
populations.”28
The report also found that “there are adequate health effects data to develop [Health Advisories]
for microcystins.”29 The Health Advisory establishes limits for drinking water exposure and
notes that infants, young children, pregnant women, nursing mothers, the elderly and immune-
compromised individuals are at heightened risk.30 The Health Advisory establishes a maximum
safe level for drinking water levels of .3 parts per billion for bottle fed infants and young children
of pre-school age.31 For school-age children through adults, the standard is set at 1.6 parts per
billion.32 Many countries and states have also adopted the World Health Organization’s standard
of 1 part per billion.33 These include Brazil, China, Czech Republic, Denmark, Finland, France,
Germany, Italy, Japan, Korea, Netherland, Norway, New Zealand, Poland, South Africa, Spain,
Ohio and Oregon.34
24
World Health Organization, Toxic cyanobacteria in water (1999),
https://www.who.int/water_sanitation_health/publications/toxicyanobact/en/. See Appendix G.
25
TCPalm, Blue-green algae bloom in St. Lucie River 10 times too toxic to touch, DEP tests show (Aug 8, 2018),
https://www.tcpalm.com/story/news/local/indian-river-lagoon/health/2018/08/08/dep-st-lucie-river-algae-bloom-10-
times-too-toxic-touch/864668002/. See Appendix E.
26
TCPalm, supra note 17.
27
United States Environmental Protection Agency, Algal Toxin Risk Assessment and Management Strategic Plan for
Drinking Water (Nov 2015), https://www.epa.gov/sites/production/files/2015-11/documents/algal-risk-assessment-
strategic-plan-2015.pdf. See Appendix H.
28
Id.
29
United States Environmental Protection Agency, 2015 Drinking Water Health Advisories for Two Cyanobacterial
Toxins (June 2015), https://www.epa.gov/sites/production/files/2017-06/documents/cyanotoxins-fact_sheet-
2015.pdf. See Appendix H.
30
United States Environmental Protection Agency, Drinking Water Health Advisory for the Cyanobacterial
Microcystin Toxins (June 2015), https://www.epa.gov/sites/production/files/2017-06/documents/microcystins-
report-2015.pdf. See Appendix H.
31
Id.
32
Id.
33
Id.
34
Id.
5
April 22, 2019
c. Additional Impacts
In 2016, the Centers for Disease Control launched an online portal to inform the public about
new research related to the health impacts of harmful algal blooms.35 The CDC website states
that “Harmful algal blooms (HABs) are the rapid growth of algae that can cause harm to animals,
people, or the local ecology….HABs can produce toxins that have caused a variety of illnesses in
people and animals.”36
The website summarizes scientific studies conducted between 2000 and 2008 as follows:
“Harmful algal blooms (HABs) can produce toxins that cause illness in people, companion
animals (dogs, cats), livestock (sheep, cattle), and wildlife (including birds and mammals).
Exposures to the toxins can occur when people or animals have direct contact with contaminated
water by: swimming, breathing in aerosols (tiny airborne droplets or mist that contain toxins)
from recreational activities or wind-blown sea spray, [or] swallowing toxins by drinking
contaminated water or eating contaminated fish or shellfish.”37
According to the Centers for Disease Control, “Anyone who visits a contaminated water body
during a HAB event can be exposed through direct contact with the contaminated water. Skin
irritation and reactions in humans and animals can vary depending on the length of contact with
the contaminated water and the type of HAB toxin present in the water.”38 Moreover, “people
can swallow contaminated water while they are swimming or playing in the water. Active water
sports (like water-skiing) pose a higher risk of accidental ingestion. Swimmers may swallow up
to 16–200 mL of freshwater (the equivalent of 0.5 – 6.8 ounces of water) during one swim.”39
35
Centers for Disease Control and Prevention, CDC Launches updated Harmful Algal Blooms (HABs) website and
One Health Harmful Algal Bloom System OHHABS as hot weather brings extensive HAB events (Oct 24, 2016),
https://www.cdc.gov/nceh/hsb/success_stories/harmful_algal_blooms.html. See Appendix I.
36
Centers for Disease Control and Prevention, Harmful Algal Bloom (HAB)-Associated Illness (Accessed Feb 11,
2019), https://www.cdc.gov/habs/. See Appendix I.
37
Centers for Disease Control and Prevention, Harmful Algal Bloom (HAB)-Associated Illness: Illness & Symptoms
(Accessed Feb 11, 2019), https://www.cdc.gov/habs/illness.html. See Appendix I. See also Fleming L, Backer L,
Rowan A. The epidemiology of human illnesses associated with harmful algal blooms. In: Massaro E, ed. Handbook
of Neurotoxicology: Humana Press;2002:363-81; Kite-Powell HL, Fleming LE, Backer LC, Faustman EM,
Hoagland P, Tsuchiya A, Younglove LR, Wilcox BA, Gast RJ. Linking the oceans to public health: current efforts
and future directions. Environ Health. 2008;7(Suppl 2):S6; Van Dolah FM. Marine algal toxins: origins, health
effects, and their increased occurrence. Environ Health Perspect. 2000;108(Suppl 1):133-41.
38
Centers for Disease Control and Prevention, Harmful Algal Bloom (HAB)-Associated Illness: Sources of Exposure
& Risk Factors (Accessed Feb 11, 2019), https://www.cdc.gov/habs/exposure-sources.html. See Appendix I. See
also Koreivienė J, Anne O, Kasperovičienė J, Burškytė V. Cyanotoxin management and human health risk
mitigation in recreational waters. Environ Monit Assess. 2014;186(7):4443-59; Van Dolah FM. Marine algal
toxins: origins, health effects, and their increased occurrence. Environ health Perspect. 2000;108(Suppl 1):133.
39
Id. See also Lopez CB, Jewett, EB, Dortch Q, Walton BT, Hudnell HK. Scientific assessment of freshwater
harmful algal blooms. Interagency Working Group on Harmful Algal Blooms, Hypoxia, and Human Health of the
Joint Subcommittee on Ocean Science and Technology. Washington, DC. 2008; Chorus I, Bartram J, eds. Toxic
Cyanobacteria in Water: A guide to their public health consequences, monitoring and management. London, United
Kingdom: World Health Organization; Routledge, London; 1999; Lopez CB, Dortch Q, Jewett EB, Garrison D.
Scientific assessment of marine harmful algal blooms. Interagency Working Group on Harmful Algal Blooms,
Hypoxia, and Human Health of the Joint Subcommittee on Ocean Science and Technology. Washington, DC. 2008;
Anderson DM. Approaches to monitoring, control and management of harmful algal blooms (HABs). Ocean and
6
April 22, 2019
Additionally, the Centers for Disease Control notes that “people can be exposed to a HAB or
HAB toxins by inhaling (breathing in) tiny water droplets, mist, or sea spray from a
contaminated body of water. This can occur even if a person does not go into the water …
Individuals who have been on the beach or on a boat during a marine HAB event have reported
breathing difficulties after inhaling air or water particles contaminated with HAB toxins. A
study conducted during a Florida red tide found that marine HAB toxins could be transported in
the air almost 4 miles inland from the water source.” 40 and a 2018 Florida Gulf Coast University
confirmed found that cyanotoxins can become aerosolized and breathed more than a mile away
from water.41 Moreover, a 2018 Florida Atlantic University study found that 100% of Martin
County residents who were testing for cyanotoxins in their nasal passages tested positive for
having breathed in the toxins.42
In addition to human impacts, according to the Centers for Disease Control, “animals can
become sick when they drink contaminated water, groom themselves after swimming in
contaminated water, or eat toxic algae or contaminated fish and shellfish in the water. In fact,
animals are at a higher risk of ingesting contaminated water because they do not avoid water that
Coastal Manag. 2009;52(7):342-347; Koreivienė J, Anne O, Kasperovičienė J, Burškytė V. Cyanotoxin management

and human health risk mitigation in recreational waters. Environ Monit Assess, 2014;186(7):4443-59.
40
Id. See also Koreivienė J, Anne O, Kasperovičienė J, Burškytė V. Cyanotoxin management and human health risk
mitigation in recreational waters. Environ Monit Assess. 2014;186(7):4443-59; Backer LC, McNeel SV, Barber T,
Kirkpatrick B, Williams C, Irvin M, Zhou Y, Johnson TB, Nierenberg K, Aubel M. Recreational exposure to
microcystins during algal blooms in two California lakes. Toxicon. 2010;55(5):909-21; Backer LC, Carmichael W,
Kirkpatrick B, Williams C, Irvin M, Zhou Y, Johnson TB, Nierenberg K, Hill VR, Kieszak SM. Recreational
exposure to low concentrations of microcystins during an algal bloom in a small lake. Mar Drugs. 2008;6(2):389-
406; Backer LC, Kirkpatrick B, Fleming LE, Cheng YS, Pierce R, Bean JA, Clark R, Johnson D, Wanner A, Tamer
R. Occupational exposure to aerosolized brevetoxins during Florida red tide events: effects on a healthy worker
population. Environ Health Perspect. 2005;113(5):644-9; Kirkpatrick B, Fleming LE, Backer LC, Bean JA, Tamer
R, Kirkpatrick G, Kane T, Wanner A, Dalpra D, Reich A. Environmental exposures to Florida red tides: Effects on
emergency room respiratory diagnoses admissions. Harmful algae. 2006;5(5):526-33; Nierenberg K, Hollenbeck J,
Fleming LE, Stephan W, Reich A, Backer LC, Currier R, Kirkpatrick B. Frontiers in outreach and education: the
Florida red tide experience. Harmful Algae. 2011;10(4):374-80; Kirkpatrick B, Pierce R, Cheng YS, Henry MS,
Blum P, Osborn S, Nierenberg K, Pederson BA, Fleming LE, Reich A. Inland transport of aerosolized Florida red
tide toxins. Harmful Algae. 2010;9(2):186-89.
41
News-Press, Algae crisis: Airborne particles of toxic cyanobacteria can travel more than a mile inland, new
FGCU study shows (Mar 15, 2019), https://www.news-press.com/story/tech/science/environment/2019/03/15/new-
health-questions-raised-fgcu-research-toxic-algae-dust/3176195002/. See Appendix E.
42
TCPalm, Health effects of breathing toxic algae in St. Lucie River is focus of Harbor Branch study (Sep 14,
2018), https://www.tcpalm.com/story/news/local/indian-river-lagoon/health/2018/09/14/people-along-st-lucie-river-
breathing-toxin-harbor-branch-study/1270005002/. See Appendix E.
7
April 22, 2019
is discolored or smells bad.”43 In 2018, toxic algae in the St. Lucie River resulted in the death of
one dog and at least five additional dogs becoming sick.44
Harmful algal blooms also pose a serious risk to contamination of water used for tap water.45
The Environmental Protection Agency has suggested that the risk of harmful algal blooms
contaminating drinking water is increasing.46 Moreover, according to the Centers for Disease
Control, “harmful cyanotoxins do not cause changes in tap water taste or odor.”47 As a result, “It
is not possible to determine solely upon visual observation if a bloom is producing toxins, thus
any bloom is potentially dangerous. When blooms occur, the risk of cyanotoxin contamination of
the surface water increases, placing potential risk to drinking water sources.”48 In 2015, the
Environmental Protection Agency warned that “Human exposure to cyanotoxins can result in a
host of adverse health effects, including gastroenteritis, liver damage and kidney damage.”49
The Centers for Disease Control has also warned that “people and animals can be exposed to
HAB toxins by eating fish or shellfish from a freshwater body with a HAB. Freshwater fish can
become contaminated with cyanotoxins when they eat toxin-producing cyanobacteria. … People
and animals can be exposed to HAB toxins by eating seafood from contaminated marine (salt)
water bodies. Shellfish can become contaminated when they filter and concentrate water with a
HAB, and reef fish can become contaminated through the food chain.”50 In 2018, the Ocean
Research and Conservation Association began tracking cyanotoxins in Martin County in the
local food web.51 Analysis of fish found that 27.8 percent of fillets and 69.8 percent of livers
contained microcystin concentrations above ORCA’s established detection limits.52 As the
43
Centers for Disease Control and Prevention, Harmful Algal Bloom (HAB)-Associated Illness: Illness & Symptoms
(Accessed Feb 11, 2019), https://www.cdc.gov/habs/illness.html. See Appendix I. See also Work TM, Barr B,
Allison MB, Fritz L, Quilliam MA, Wright JLC. Epidemiology of domoic acid poisoning in brown pelicans
(Pelecanus occidentalis) and Brandt’s cormorants (Phalacrocorax penicillatus) in California. J Zoo Wildl Med.
1993;24(1):54-62; Beltrán AS, Palafox-Uribe M, Grajales-Montiel J, Cruz-Villacorta A, Ochoa J. Sea bird mortality
at Cabo San Lucas, Mexico: evidence that toxic diatom blooms are spreading. Toxicon. 1997;35(3):447-53;
Landsberg J, Flewelling L, Naar J. Karenia brevis red tides, brevetoxins in the food web, and impacts on natural
resources: Decadal advancements. Harmful Algae. 2009;8(4):598-607; Bossart GD, Baden DG, Ewing RY, Roberts
B, Wright SD. Brevetoxicosis in manatees (Trichechus manatus latirostris) from the 1996 epizootic: gross,
histologic, and immunohistochemical features. Toxicol Path. 1998;26(2):276-82.
44
TCPalm, Toxic algae in St. Lucie River may have killed 1 dog, sickened 5 others, Stuart vet says (Sep 10, 2018),
https://www.tcpalm.com/story/news/local/indian-river-lagoon/health/2018/09/10/toxic-algae-dog-
illnesses/1254048002/. See Appendix E.
45
Centers for Disease Control and Prevention, Harmful Algal Bloom (HAB)-Associated Illness: Sources of Exposure
& Risk Factors (Accessed Feb 11, 2019), https://www.cdc.gov/habs/exposure-sources.html. See Appendix I.
46
Environmental Protection Agency, Monitoring Unregulated Drinking Water Contaminants (Accessed Feb 11,
2019), https://www.epa.gov/dwucmr/first-unregulated-contaminant-monitoring-rule#ucmr.
47
Centers for Disease Control and Prevention, supra note 32..
48
Environmental Protection Agency, Recommendations for Public Water Systems to Manage Cyanotoxins in
Drinking Water (June 2015), https://www.epa.gov/sites/production/files/2017-06/documents/cyanotoxin-
management-drinking-water.pdf. See Appendix H.
49
Id.
50
Centers for Disease Control and Prevention, supra note 32.
51
Ocean Research & Conservation Association, Inc., Tracking Cyanotoxins in the Aquatic Food Web in Martin
County (2019). See Appendix J.
52
Id.
8
April 22, 2019
preliminary publication of the report concludes, “These studies show that fishers are being
exposed to cyanotoxins through fish consumption…ORCA’s data should be used to support the
need to improve water quality in Martin County.”53
In conclusion, scientists and health professionals have found that toxic algae can cause nausea,
vomiting, liver disease and even death.54 Scientists have also linked at least one other toxin in the
algae to neurological diseases such as ALS and Alzheimer's.55 Unlike in 2008, we now know
there is a strong causal link between these algae outbreaks and life-threatening public health
consequences. We also know there is a strong causal link between discharges from Lake
Okeechobee and the occurrences of algae outbreaks. Therefore, any regulation schedule that
does not place a premium on the prevention of these discharges would be in direct violation of
NEPA’s requirement to consider “the degree to which the proposed action affects public health
or safety.”56
III. WATER QUALITY
LORS 2008 has caused substantial damage to the environment and water quality in Florida. The
design has forced water where it is not wanted (and extremely harmful) and starved the areas that
do want the water. For example, a 2017 Florida International University study showed that
during the dry season, Florida Bay receives less than 5 percent of the historical freshwater flow it
needs to avoid seagrass-killing hypersalinity.57 Likewise, during the wet season, the bay receives
less than 33 percent of its historical freshwater flow.58 Meanwhile, the St. Lucie Estuary, which
historically received no freshwater flows, is flooded with billions (or hundreds of billions) of
gallons of polluted fresh water each year.59
53
Id.
54
Zhang et. al, Environmental Health, Cyanobacteria blooms and non-alcoholic liver disease: evidence from a
county level ecological study in the United States (2015),
https://ehjournal.biomedcentral.com/track/pdf/10.1186/s12940-015-0026-7. See Appendix K.
55
Paul Alan Cox, David A. Davis, Deborah C. Mash, James S. Metcalf, Sandra Anne Bannack, Dietary exposure to
an environmental toxin triggers neurofibrillary tangles and amyloid deposits in the brain. The Royal Society
Publishing. (Jan 27, 2016), https://royalsocietypublishing.org/doi/full/10.1098/rspb.2015.2397. See Appendix L.
56
40 C.F.R. § 1508.27.
57
Jennifer S. Rehage, Florida International University, Florida Bay: updated, effects & recovery from Hurricane
Irma (2017). See Appendix Z.
58
Id.
59
Data provided by the U.S. Army Corps of Engineers.
9
April 22, 2019
In addition to supporting human populations, the St. Lucie River and Estuary is one of the most
biologically-diverse estuaries in the nation, described as “rich in habitats and species, with the
greatest species diversity of any estuary in North America.”60
The area also includes areas identified as “Outstanding Florida Waters,” such as the Hobe Sound
National Wildlife Refuge, the St. Lucie Inlet State Preserve, the Indian River State Aquatic
Preserve and the Jensen Beach State Aquatic Preserve.61 Regarding these bodies, Florida law
states that “it shall be the Department policy to afford the highest protection to Outstanding
Florida Waters and Outstanding National Resource Waters. No degradation of water quality…is
to be permitted.”62
The Indian River Lagoon estuary system is also one of only 28 estuaries designated as an estuary
of national significance.63 As a result of this designation, the U.S. Army Corps of Engineers
must carefully evaluate its obligations under NEPA to evaluate adverse impacts on significant
scientific resources and critical habitat.
Unfortunately, the estuary is also one of the most ecologically-stressed river and estuarine
systems in the country as a result of decades of mismanagement of our ecosystem.
a. Sources of Pollution
In 1913, the State of Florida decided to construct a canal between Lake Okeechobee and the St.
Lucie estuary to divert water from the lake and to support development south of the lake.64 The
first discharge of water into the estuary reportedly occurred in 1923.65 Prior to that time, there
was no natural connection between the lake and the St. Lucie estuary.66 By 1930, the
environmental impacts of this poor decision began to be realized and the Martin County
Commissioners requested termination of the discharges.67 For nearly the last century, our
regions environmental health has been sacrificed by Florida and federal agencies in order to
provide flood protection and irrigation benefits to communities south of Lake Okeechobee.68
60
Patti Sime (South Florida Water Management District), WETLANDS, St. Lucie Estuary and Indian River Lagoon
Conceptual Ecological Model (Dec 2005), https://link.springer.com/article/10.1672/0277-
5212(2005)025[0898:SLEAIR]2.0.CO;2. See Appendix M.
61
Dr. Gary Goforth, Suggestions to Reduce the Destructive Discharges to the St. Lucie River and Estuary (Feb 7,
2014). See Appendix N.
62
Florida Administrative Code Chapter 62.302-7090.
63
United States Environmental Protection Agency, National Estuary Program (accessed Apr 16, 2019),
https://www.epa.gov/nep/local-estuary-programs#tab-1.
64
Dr. Gary Goforth, Water Quality Assessment of the St. Lucie River Watershed – Water Year 2018 – DRAFT
(2019), http://www.garygoforth.net/DRAFT%20-
%20Water%20Quality%20Assessment%20of%20the%20SLRW%20-%20Water%20Year%202018.pdf. See
Appendix O.
65
Id.
66
Id.
67
Id.
68
Id.
10
April 22, 2019
In 2016, for example, more than 200 billion gallons of polluted Lake Okeechobee water—
containing nutrients, sediment, toxic blue-green algae and low salinity water—was diverted from
its natural flow and discharged east.69 In 2017, more than 190 billion gallons of polluted water
were discharged to the east.70 A similar pattern of discharges and environmental destruction has
also occurred on Florida’s west coast in the Caloosahatchee estuary, which has received even
higher volumes of flow.71
Over the course of the last water year (May 2017 – April 2018), the largest single source of
phosphorus, nitrogen and sediment pollution to the St. Lucie River and Estuary was Lake
Okeechobee discharges.72 Total phosphorus concentrations in Lake Okeechobee discharges to
the St. Lucie River and Estuary are more than 5 times the lake’s Total Maximum Daily Load in-
lake target concentration of 40 parts per billion.73
By land use, Lake Okeechobee discharges in water year 2018 accounted for 42 percent of the
nitrogen load, 28 percent of the phosphorus load, 39 percent of the surface water flow and 90
percent of the total suspended solid load to the St. Lucie River and Estuary.74 In the time period
where the toxic algal blooms actually occurred in 2018, the numbers are even more jarring. Lake
Okeechobee discharges during this time period accounted for more than half of the water
entering the estuary.75 From June 1, 2018 to October 5, 2018 - during which there were more
than 100 reports of algal blooms in St. Lucie and Martin County and discharges were occurring76
- Lake Okeechobee discharges accounted for 54 percent of the water, local basin agricultural
runoff accounted for 35 percent and local communities (including runoff containing pollutants
from septic tanks and fertilizer) accounted for just 8 percent.77
b. Misinformation Surrounding Septic Tanks
Nonetheless, during 2016, several special interest affiliated groups and people misleadingly
claimed that the primary source of nutrient loads to the St. Lucie River and Estuary was the
“200,000 septic tanks” that line Martin County waterways.78 The Florida Department of Health,
however, notes that there are less than 1/10 that number of septic tanks in Martin County.79
Additionally, a former South Florida Water Management District Board Member falsely claimed
that 80 percent of the nutrient loading to the St. Lucie River and Estuary was due to local septic
tanks.80
69
70
Id.
71
Id.
72
Goforth, supra note 59.
73
Id.
74
Id.
75
Id.
76
Florida Department of Environmental Protection, Algal Bloom Sampling Results (accessed Apr 16, 2019),
https://floridadep.gov/dear/algal-bloom/content/algal-bloom-sampling-results. See Appendix P.
77
78
Id.
79
Id.
80
Id.
11
April 22, 2019
The truth is, though, that unlike discharges from Lake Okeechobee, septic tanks contribute no
toxic algae, no sediment and no pesticides to our waterways.81 Septic tanks contribute negligible
amounts of water and phosphorous.82 From 1980 to 2016, septic tanks contributed
approximately 6 percent of the total estimated nitrogen load to the estuary.83 By comparison,
Lake Okeechobee discharges contributed 29 percent of the total load and runoff from local basin
agricultural land contributed approximately 56 percent.84 In years where there are high levels of
discharges, the numbers are even more stark. During 2016, septic tanks represented
approximately 4 percent of the total nitrogen load, Lake Okeechobee discharges accounted for 48
percent and runoff from local basin agricultural land was 41 percent.85
Moreover, Martin County and the City of Stuart have significantly reduced nutrient loading from
septic systems over the last two decades.86 For example, they have undertaken the conversion of
more than 1,700 septic tanks to centralized sewers, which removed an estimated 15,400 pounds
per year of nitrogen and more than 25 stormwater projects that have removed an estimated
30,000 pounds per year of nitrogen.87 As a result, Martin County has exceeded the nitrogen load
reductions required under the State’s Basin Management Action Plan, and nonetheless, is still
planning additional septic to sewer conversion projects.88
This information is included not to minimize the importance of addressing local sources of
pollution but to highlight the fact that continuing to address septic to sewer conversions without
also addressing the much more harmful impacts of Lake Okeechobee discharges will make only
a small dent in the larger problem.
c. Impacts of Freshwater Infusion and Pollution
Habitats and species diversity in the Indian River Lagoon system are affected greatly by this
decline in water quality. A study conducted in 2005 by staff at the South Florida Water
Management District concluded that “the health of the system is being affected by water
management and land-use development….The major anthropogenic changes in both the St.
Lucie Estuary and the C-25 watersheds are significant alternations in the timing…, distribution,
quality, and volume of fresh water entering the estuary, lagoon, and ocean….In turn, [these
stressors] alters estuary salinity and increases turbidity and color.”89
The study further identified six affected ecological attributes summarized below:
81
Dr. Gary Goforth, Septic Tank Contributions to Pollutant Loading to the St. Lucie Estuary (Feb 4, 2017). See
Appendix Q.
82
Id.
83
Id.
84
Id.
85
Id.
86
87
Id.
88
Id.
89
Sime, supra note 55.
12
April 22, 2019
1. Submerged Aquatic Vegetation (SAV): “Changes in normal salinity ranges can

adversely affect SAV” and “The input of increased levels of nutrients and dissolved
organic matter effect SAV abundance and health” and “Large deposits of much that have
replaced normal substrate in the estuary and portions of the Indian River Lagoon have
contributed to the decrease in extent of SAV beds.”90
2. Oyster Communities: “Oysters and other bivalves, such as mussels and clams, found in
the St. Lucie Estuary and Indian River Lagoon are sensitive to salinity, loss of suitable
hard bottom substrate, and high levels of total suspended solids in the water column.”91
3. Estuarine Fish Communities/Sport and Commercial Fisheries: “Species richness in
many of the fish communities of the estuary and lagoon has decreased since the 1970s
when baseline data were collected. In addition to the general decrease in species
richness, specific fish communities seem to be affected by salinity and habitat changes.”
and “Large freshwater releases from the St. Lucie Canal…produced significant
incidences of fish disease and mortality, promoted toxic dinoflagellate blooms, and
reduced overall biodiversity of estuarine and freshwater fish communities within the
Indian River Lagoon.”92
4. Estuarine Benthic Communities: “Benthic macroinvertebrate communities in the St.
Lucie Estuary and Indian River Lagoon are sensitive to bottom type, water quality, and
salinity fluctuations. A decrease in diversity of benthic organisms and increase in the
numbers of pollution-tolerant micrometeorites…can indicated deteriorating water quality
in the estuary and lagoon.”93
5. Shoreline Habitat: “Mangrove wetlands and the emergent bank vegetation of tributaries
of the St. Lucie Estuary and Indian River Lagoon support fish and macroinvertebrate
communities…these once ubiquitous shoreline habitats have decreased…A significant
portion of the floodplain in the North Fork of the St. Lucie River is completely or
partially cut off from the river’s main branch because of dredging conducting during the
1920s to 1940s….resulting in the loss of natural filtration of water-borne nutrients
originating in the watershed.”94
6. Nearshore Reef: “The nearshore reef has been adversely affected by high level
freshwater discharges and the resulting silt and salinity plumes that are found primarily in
the south of the St. Lucie and Ft. Pierce Inlets.”95
The report stated in conclusion, “altered and unstable estuarine salinity has had the most
significant effect on the estuary…Accompanying the regulatory water releases is the transport of
massive volumes of organic and inorganic sediments, which contribute to deposits of much in
the estuaries. The large accumulations of much covering the bottom of the estuary dramatically
90
Id.
91
Id.
92
Id.
93
Id.
94
Id.
95
Id.
13
April 22, 2019
decrease the quality and quantity of suitable habitat…High volume releases create an oceanic
plume of colored water and suspended solids extending to the nearshore reef in the Atlantic
Ocean, reducing light penetration and exacerbating salinity….The recurring high-flow conditions
in the St. Lucie Estuary have significantly reduced the number of oysters, and the frequency at
which these high flows occur has impeded recovery.”96
In conclusion, while the ecology on Lake Okeechobee remains critically important, no study of
the ecological or water quality impacts of the Lake Okeechobee System Operating Manual is
complete without further analysis of the impacts of discharges on the estuaries. Moreover,
NEPA requires your consideration of these effects on the estuaries.97
IV. ADDITIONAL ENVIRONMENTAL CONCERNS
The management of Lake Okeechobee is a balance between numerous interests with a complex
system of often unpredictable variables, but scientific development over the last decade since
LORS 2008 was written can help us better predict some of these variables.
For example, the Centers for Disease Control and Prevention states that “Climate change might
increase the occurrence, severity, and impact of HABs in fresh, marine, and brackish waters. For
example, warming temperatures in Lake Erie have resulted in more extensive blooms of
cyanobacteria…that last into the early winter months. In the past several years, HABS have
been observed with increasing frequency and in more locations in the United States.”98
In addition to the impact on harmful algal blooms, changing weather patterns as a result of
climate change and other factors also impact the temperature and amount of rain received in
Florida, which is especially vulnerable to these shifting climate trends.99 Scientists recognize
that climate change is a major threat to ecosystems globally, especially wetlands such as the
Everglades.100 Warmer climates, along with changes in precipitation patterns and increases in
carbon dioxide, affect hydrologic conditions and biogeochemistry.101 In Florida specifically,
sea-level has been rising at a faster rate than the also accelerating global average, and recent
trends in Florida suggest further acceleration in this rate moving forward.102
96
Id.
97
40 C.F.R. § 1508
98
Centers for Disease Control and Prevention, Harmful Algal Blooms & the Environment (accessed Apr 16, 2019),
https://www.cdc.gov/habs/environment.html. See Appendix I.
99
National Academies of Sciences, Engineering and Medicine, Progress Toward Restoring the Everglades: The
Fifth Beinnial Review 131 (2014), https://www.nap.edu/catalog/18809/progress-toward-restoring-the-everglades-
the-fifth-biennial-review-2014. See Appendix R.
100
Id.
101
Id.
102
Id.
14
April 22, 2019
As a result of these trends, the National Academies of Sciences has recommended an

Everglades-wide analysis that considers how the ecosystem will change as a result of climate
change and how these changes will influence restoration goals.103 In the meantime, they have
recommended planning that—among other factors—utilizes climate change projections to
inform decisions and makes operational decisions with maximum future flexibility in mind. 104
Finally, since LORS 2008 was written, scientists have determined that the system was even
wetter than originally projected.105 LORS 2008 also overestimated the frequency of drought and
underestimated the frequency of extreme high stage events, leading to discharges in 9 of the last
11 years while drought levels occurred in only one year.106 A more accurate, scientifically-
driven and flexible overall system that better takes into account rainfall and changing weather
patterns could realize actual shared adversity instead of forcing nearly all of the adversity on the
communities around the St. Lucie Estuary and Caloosahatchee Estuary.
V. RECREATIONAL ACTIVITIES AND ECONOMIC IMPACTS
In 2016 the East Coast Florida Regional Planning Council and the Treasure Coast Regional
Planning Council produced a report on the economic valuation of the Indian River Lagoon.107
The report concluded that the total annual economic output or value received from the Indian
River Lagoon in 2014 was approximately $7.6 billion.108 This figure also does not include the
estimated $934 million in annualized real estate value added for property located on or near the
Indian River Lagoon, nor does it include the economic impacts on other regions impacted by the
management of Lake Okeechobee, such as the impacts on tourism and recreation on the west
coast of Florida.109
Moreover, the industry groups in the area that rely on the Indian River Lagoon collectively
support more than 70,000 jobs annually with wages in excess of $1.2 billion and the tourism
industry attracts more than 7 million visitors annually.110 Each additional visitor spends
approximately $162 per day, creating a new job for every 85 visitors to the area.111
103
Id.
104
Id. at 166.
105
Id.
106
107
East Coast Florida Regional Planning Council and Treasure Coast Regional Planning Council, Indian River
Lagoon Economic Valuation Update (Aug 26, 2016),
http://tcrpc.org/special_projects/IRL_Econ_Valu/FinalReportIRL08_26_2016.pdf. See Appendix S.
108
Id.
109
Id.
110
Id.
111
Id.
15
April 22, 2019
The report, however, identified that the “Indian River Lagoon-Dependent” industry was in
decline, and this decline is directly linked to the poor management of our water resources.112 For
example:
• The overall economic impact of commercially harvested clams, oysters, crabs and shrimp
in the area has declined by nearly 80 percent over the last two decades, adjusted for
inflation.113
• Pounds of shellfish harvested during the same period also declined from about 7.1 million
pounds to 2 million pounds.114
• Over the same time period, the overall value of commercial fin fish harvest has declined
by 37 percent after being adjusted for inflation.115
• There has also been a decline in boat registrations for the Indian River Lagoon region
over the last decade. While the population has increased by 12 percent, boat registration
has decreased by 11 percent since 2005.116
These are just a couple economic indicators. As the report concludes, “The IRL and nearshore
coastal resources are significant contributors to the region’s strong economy and extraordinary
quality of life. Sustaining the IRL is important to the region’s economic recovery…and
sustainable growth.”117
This report is backed up by additional studies with similar conclusions. For example, a 2015
study conducted by the Florida Realtors Association concluded that “the ongoing problem of
polluted water in the Caloosahatchee and St. Lucie Rivers and estuaries has indeed resulted in a
negative impact on home values….Many Realtors, for instance, have reported lost sales due to
poor water quality related to discharges from Lake Okeechobee…The greater direct damage to
home values in these markets is the continuing recurrence of these types of events, and the
increasing frequency with which they are occurring. This effect is longer lasting and does not go
away when the algal blooms and murky waters clear up after each incident.”118
As the 2016 economic study stated, however, “the health and sustainability of the IRL will be a
key determinant in the future of the region’s economy, public health and other natural, cultural,
and societal values that are collectively worth far more than what has been quantified in this
report.”119 For example, recreational and commercial fishermen have felt both economic and
non-economic impacts:
112
Id.
113
Id.
114
Id.
115
Id.
116
Id.
117
Id.
118
Florida Realtors Association, The Impact of Water Quality on Florida’s Home Values (Mar 2015),
https://www.floridarealtors.org/ResearchAndStatistics/Other-Research-
Reports/upload/FR_WaterQuality_Final_Mar2015.pdf. See Appendix T.
119
East Coast Florida Regional Planning Council, supra note 102.
16
April 22, 2019
• The American Sportfishing Association estimates that ever year in the 18th Congressional
District of Florida alone 106,885 anglers spend more than $130 million on fishing related
purchases, which supports $206.6 million in economic output and 1,642 jobs.120
• The 2005 South Florida Water Management District Report found that “lowered salinities
can result in atypical freshwater conditions that are conducive to the persistence of fish
pathogens, such as fungi, that have been implicated as a causal factor for lesioned fish
within the St. Lucie Estuary.”121
• The 2018 ORCA study found that “subsistence fishers eat 3 to 4 times more fish than the
average US citizen and depend on fishing for up to seven meals each week. Most fish
caught at the Port Mayaca locks had detectable microcystin….These fishers also face
additional cyanotoxin exposure while they fish through skin contact with contaminated
water and inhalation of aerosolized microsystin.”122
This data is provided not as a comprehensive look at the recreational and economic impacts of
water management decisions, but rather to exemplify the impacts on recreation and the economy
for the communities supported by the St. Lucie and Caloosahatchee estuaries. As the U.S. Army
Corps of Engineers evaluates the authorized project purposes and NEPA requirements, the scope
must include these impacts on the estuaries and surrounding communities, in addition to the
impacts of recreation on Lake Okeechobee itself.
VI. ENHANCMENT OF FISH AND WILDLIFE
Both the authorized project purposes and NEPA require that the impacts on fish and wildlife be
taken into account, not only on Lake Okeechobee or in the Everglades, but also in the estuaries
and other “reasonably foreseeable” future effects.123 These impacts include those discussed
above, as well as impacts on endangered and threaten species and other wildlife.
Notably, the St. Lucie Estuary and Indian River Lagoon are together home to more than 50
endangered or threatened species.124 The St. Lucie River, moreover, is home to more than 500
fish species.125 The North Fork of the St. Lucie River is a state aquatic preserve and home to
more than 650 native species, including 33 species of special concern and an additional 20 rare
species.126 The preserve also contains rookeries for the endangered wood stork.127 The Hobe
120
American Sportfishing Association, Economic Contributions of Recreational Fishing by Congressional Districts
(accessed Apr 16, 2019), https://asafishing.org/economic-contributions-of-recreational-fishing/.
121
Sime, supra note 55.
122
Ocean Research & Conservation Association, Inc., supra note 49.
123
40 C.F.R. § 1508
124
South Florida Water Management District, North Fork of the St. Lucie River Water Restoration (Apr 2015),
https://www.sfwmd.gov/sites/default/files/documents/jtf_slr_res.pdf. See Appendix U.
125
Id.
126
Id.
127
Id.
17
April 22, 2019
Sound National Wildlife Refuge is home to nearly 40 species listed as either threatened,
endangered, or of special concern.128 The Caloosahatchee estuary is likewise home to species
protected by the Endangered Species Act.129 These species collectively are routinely threatened
by the Army Corps’ actions and far more emphasis must be placed on this area than was done in
LORS 2008.
Additionally, the U.S. Army Corps of Engineers must take into account effects that are
“reasonably foreseeable” as part of its NEPA requirements.130 For example, statistics from the
Florida Fish and Wildlife Commission say that red tide and blue-green algae are to blame for a
substantial rise in manatee deaths in 2018.131 Moreover, numerous scientific studies have
connected the cyanobacterial neurotoxin BMAA to damaging impacts on sharks and dolphins.132
Most recently, a 2019 study found that toxins produced by blue-green algae were found in dead
dolphins that showed signs similar to an Alzheimer’s-like brain disease.133 Not only is this
concerning for the similar impacts that humans could experience, but it necessitates review by
the U.S. Army Corps of Engineers as part of the NEPA process.
VII. RECOMMENDATIONS
The U.S. Army Corps of Engineers must institute increased flexibility to manage the lake at a
lower level under the new Lake Okeechobee System Operating Manual (LOSOM) that balances
human health needs; the ecological needs of the St. Lucie Estuary, the Caloosahatchee Estuary
and Lake Okeechobee; the economic and recreational impacts of the estuary regions and on Lake
Okeechobee; reducing the risk of Dike failure; navigation on the Lake; water flows south to the
Everglades and Florida Bay; and public-use water supply such as for municipalities and drinking
water.
This increased flexibility must ensure that the ecological, human health and economic needs of
the estuary regions are finally considered as part of the equation. Moreover, periodic lower lake
levels would be ecologically beneficial to the estuaries and often to the lake as well. When these
interests align, the system should not be forced to be operated in a way that restricts
commonsense water management decisions.
128
U.S. Fish and Wildlife Service, Nathaniel P. Reed Hobe Sound National Wildlife Refuge: About the Refuge
(accessed Apr 16, 2019), https://www.fws.gov/refuge/Hobe_Sound/about_the_refuge.html.
129
U.S. Fish and Wildlife Service, Caloosahatchee National Wildlife Refuge Fact Sheet (accessed Apr 16, 2019),
https://www.fws.gov/southeast/pdf/fact-sheet/caloosahatchee-national-wildlife-refuge.pdf.
130
40 C.F.R. § 1508
131
Florida Fish and Wildlife Conservation Commission, 2018 Preliminary Red Tide Manatee Mortalities (Mar 7,
2019), https://myfwc.com/media/18471/2018preliminaryredtide.pdf. See Appendix V.
132
See, e.g., Mondo et. al, Food and Chemical Toxicology, Environmental Neurotoxins b-N-methylamino-L-alanine
(BMAA) and mercury in shark cartilage dietary supplements (Feb 25, 2014); Hammerschlag et. al, Toxins,
Cyanobacterial Neurotoxin BMAA and Mercury in Sharks (Apr 28, 2016); Davis et. al, Cyanobacterial neurotoxin
BMAA and brain pathology in stranded dolphins (Mar 20, 2019). See Appendix W.
133
Davis et. al, Cyanobacterial neurotoxin BMAA and brain pathology in stranded dolphins (Mar 20, 2019). See
Appendix W.
18
April 22, 2019
Below are several recommendations to accomplish this goal:

a. LOSOM should include maximum flows south of the lake:
1. Although available STA treatment acreage has increased dramatically—more than
15-fold since 1995134—the volume of lake releases sent south has not kept pace.
LORS 2008 recommended a dramatic decrease in the amount of regulatory
releases sent south to the STAs. LORS 2008 assumed a limit on Lake
Okeechobee releases to the STAs of 60,000 acre-feet per year, which results in
substantial underutilization of these structures.135 Instead, LOSOM should set
aggressive minimum targets for releases to the STAs/WCAs before any regulatory
releases are sent to the estuaries (excluding the ecologically beneficial flows to
the Caloosahatchee).136
2. Moreover, as a publicly-accountable agency, South Florida Water Management
District needs to operate with the transparency necessary to demonstrate that they
are utilizing the maximum capacity south of the lake (such as in the STAs, the
WCAs and in the wildlife management areas), including full utilization of the
EAA and L-8 flow equalization basins. The U.S. Army Corps of Engineers can
help facilitate this communication with South Florida Water Management District
to ensure maximum flows south.
b. LOSOM should include added flexibility at the high end of the operational band:
1. Now that the dike rehabilitation efforts have reduced risk of breach of the Hebert
Hoover Dike, the U.S. Army Corps of Engineers should raise the upper water
level limit of flexibility to help prevent discharges. This should not raise the
“target” level of the lake, but rather raise the level at which discharges are
mandated to minimize the need for discharges and ensure a more equitable
balance of the impacts to the estuaries. In short: the level at which preemptive
lake releases are mandated should be raised substantially and the U.S. Army
Corps of Engineers should be granted further flexibility to minimize the need of
discharges.
2. When the schedule does call for regulatory discharges, the U.S. Army Corps of
Engineers should be empowered with additional flexibility to reduce inflows to
Lake Okeechobee. For example, the schedule could call for temporarily raising
the operational levels of canals surrounding the lake. The schedule could also call
for reducing the use of pump stations that move water into the lake (it is
operationally self-defeating to pump water into the lake from the EAA at the same
time that the water is being discharged out of the lake).
134
Dr. Gary Goforth, supra note 56.
135
Id.
136
See, e.g., Dr. Gary Goforth, supra note 56.
19
April 22, 2019
c. LOSOM should include added flexibility at the low end of the operational band:
1. Numerous environmental groups, scientists and engineers have recognized the
ecological benefits to Lake Okeechobee of periodic reductions in lake levels to
allow the regrowth of vegetation.137 Lower lake levels in the spring have also
been linked to reduced algal bloom proliferation on the lake in the summer, and of
course, lower lake levels heading into wet season also helps mitigate the need for
discharges during wet season.138
2. The U.S. Army Corps of Engineers should not be restricted by an inflexible
“target” that requires lake levels be held above a certain level if it is beneficial to
water allocation and quality to reduce the levels further in any given year. Thus,
the U.S. Army Corps of Engineers should have the built-in flexibility at the lower
end of the schedule to reduce lake levels (e.g., below 11 feet heading into wet
season) when it would be beneficial to do so, weighed with the need to provide
water supply for municipal use and drinking water.
3. Just as at the high-end of the operational band, at the low-end of the operational
band the U.S. Army Corps of Engineers should be empowered to utilize flexibility
to protect impacted communities. For example, the U.S. Army Corps of
Engineers should ensure that drinking water and other municipal water uses are
protected during times of scarcity by accommodating these uses prior to for-profit
users of water.
4. Moreover, LOSOM should include flexibility to adjust operational protocols as
additional CERP and CEPP storage projects, such as the EAA Southern Storage
Reservoir, are brought online.
d. LOSOM should aim to operate at routinely lower levels than LORS 2008:
1. Under LORS 2008, extreme high stage events have occurred at a far higher
frequency than extreme low stage events. Over the last 11 years, discharges have
occurred during 9 years while drought levels have occurred during only 1 year.139
As a result, LORS 2008 has balanced the hypothetical, largely unrealized risk of
drought against the certainty of economic and environmental devastation resulting
from discharge. In turn, the lake and the region south of the lake have realized
nearly all of the benefits of LORS 2008 while the communities surrounding the
St. Lucie Estuary and Caloosahatchee Estuary have realized nearly all of the
negative impacts.
2. To institute true shared adversity, LOSOM should lower the “target” level of the
lake to mitigate the need for estuary discharges, while also protecting the ecology
of the lake and other operational priorities. Perhaps most importantly, this must
137
See, e.g., Audubon Florida, Lake Okeechobee Discharges Letter to Colonel Andrew Kelly and Drew Bartlett
(Apr 5, 2019). See Appendix X.
138
Statement by Dr. Susan Gray, Lake O Fact Check #8: How Water Levels Impact Algae Blooms (Apr 11, 2019),
https://mast.house.gov/blog?ID=75BF27B7-38A2-4C5B-B534-3742453EE8DD.
139
20
April 22, 2019
be done with much more flexibility than under LORS 2008 so that the U.S. Army
Corps of Engineers is not stuck with rigid protocols that force discharges to the
estuaries in pursuit of a specific high (or low) water level when it is inadvisable or
unnecessary due to other monthly or yearly predictive factors (e.g., weather
patterns).
e. LOSOM should incorporate human health and safety protections:

1. The U.S. Army Corps of Engineers’ contention that the “paramount
consideration” of the management of Lake Okeechobee is “the reduction of flood
risk” is in direct contravention of Congressional intent, which has set forth
numerous, equal operational priorities.140 That assertion is also in direct
contravention to the actual operational decisions made over the last decade, which
have routinely held excess water on Lake Okeechobee directly threatening the
stability of the Herbert Hoover Dike. The U.S. Army Corps of Engineers must
cease operating the system in complete and total deference to the water control
needs of the Everglades Agricultural Area, and instead must give equal priority to
water quality, Everglades restoration, public-use water supply, human health and
safety needs of the rest of Florida.
2. The U.S. Army Corps of Engineers should incorporate into operational protocols
efforts to mitigate the proliferation of harmful algal blooms. While LORS 2008
envisioned that pulse releases would accomplish this goal141, the realities of the
last decade have proven otherwise. The U.S. Army Corps of Engineers must
work with its partners at the South Florida Water Management District, Florida
Department of Environmental Protection, U.S. Environmental Protection Agency,
Centers for Disease Control and Prevention, Department of the Interior and other
agencies to build in true public health protections and deterrents to the
proliferation of harmful algal blooms. With the overwhelming evidence cited
above, claiming ignorance to the health impacts or attempting to deflect to other
agencies is not an acceptable approach.
3. When discharges are necessary, the U.S. Army Corps of Engineers, South Florida
Water Management District and Florida Department of Environmental Protection
must work together as part of LOSOM to provide notification to impacted
communities about the health impacts of contact with the cyanobacteria-infested
water being released by the U.S. Army Corps of Engineers.
140
Col. Jason Kirk, supra note 1.
141
LORS, supra note 2.
21
April 22, 2019
f. LOSOM can accomplish all of these priorities with zero discharges to the St. Lucie
estuary and beneficial dry-season flows to the Caloosahatchee:
1. The St. Lucie estuary was connected to Lake Okeechobee by man and has zero
ecological need for any additional discharges, no matter the time of the year.142
The current recommended 200 cfs flow to the S-80 is not grounded in science and
should be reduced to zero. Moreover, the system should operate with the goal of
never necessitating discharges to the St. Lucie estuary.
2. The U.S. Army Corps of Engineers should meet the minimum flow requirements
of 650 to 1000 cfs to the Caloosahatchee estuary (without crossing the high flow
harm threshold). The communities to the West of Lake Okeechobee want the
water in these amounts during the dry season, and it’s negligent to not give it to
them when it is beneficial to the rest of the system also.143 Likewise with the St.
Lucie estuary, the system should be operated with the goal of never necessitating
harmful wet-season discharges to the Caloosahatchee estuary.
3. Moreover, when discharges are necessitated, the U.S. Army Corps of Engineers
should prioritize which areas (and in which amounts) receive harmful flows from
the operations of the system based on which areas receive actual benefit from the
operations of the system (i.e. water supply for irrigation, etc.). An area that
receives little to no benefit, such as the communities surrounding the St. Lucie
estuary, should not be burdened with the majority of the harm. This is the only
way to ensure true and equitable “shared adversity.”
4. Widespread risk to public health is simply not a consequence that we should be
willing to accept as a result of decades of operational mismanagement.
The bottom line is that the use of taxpayer dollars to benefit private business interests cannot be
placed above the public good, such as health, safety, drinking water and the environment. As the
U.S. Army Corps of Engineers undertakes their rewrite of the Lake Okeechobee Regulation
Schedule, changes can and should be made that prioritize public health and public use of our
water.
VIII. UNRESOLVED QUESTIONS
1. How will LOSOM take into account the authorized project purpose of Water Quality?
2. How will LOSOM take into account the authorized project purpose of Preservation of
Everglades National Park?
3. How will LOSOM take into account the overarching policy priority of protecting human
health and safety as it applies not only to preventing flood risk but also to prevent
harmful algal bloom related death and illness?
142
Dr. Gary Goforth¸ supra note 59.
143
See, e,g,, Appendix Y.
22
April 22, 2019
4. Does the U.S. Army Corps of Engineers acknowledge that the science draws a direct
causal link between Lake Okeechobee discharges and the proliferation of blue-green
algae in the St. Lucie and Caloosahatchee estuaries?
5. If so, what actions will the U.S. Army Corps of Engineers take to mitigate these impacts
and protect the public health and safety of the communities surrounding these
waterbodies?
6. Is cyanobacteria a toxin?
7. At what point, if any, is water flowing from Lake Okeechobee via the S-308 too polluted,
dirty or toxic to send into the St. Lucie or Caloosahatchee estuaries?
8. Does the Indian River Lagoon contribute any nutrients or pollutants to Lake
Okeechobee?
9. Given that the Environmental Protection Agency and World Health Organization have
established Health Advisories for cyanobacteria, what steps will the U.S. Army Corps of
Engineers take to comply with these advisories?
10. Given that the Environmental Protection Agency has a Strategic Plan for managing algal
toxin risk, what steps will the U.S. Army Corps of Engineers take to comply with this
strategic plan?
11. Given that the Centers for Disease Control states that Harmful Algal Blooms produce
toxins that “cause illness in people,” what actions will the U.S. Army Corps of Engineers
take to prevent the poisoning of communities impacted by discharges?
12. Given that toxic algae in the St. Lucie River has already caused animals to die and people
to get sick, what proactive steps will the U.S. Army Corps of Engineers take in LOSOM
to warn the public of these health threats prior to discharging cyanobacteria-polluted
water?
13. What steps will the U.S. Army Corps of Engineers take to comply with NEPA’s
requirement to consider “the degree to which the proposed action affects public health or
safety”?
14. Does the U.S. Army Corps of Engineers consider Lake Okeechobee and the St. Lucie
River/Estuary one body of water or does it acknowledge that these are separate bodies of
water unnaturally connected by man?
15. What steps will the U.S. Army Corps of Engineers take to safeguard areas identified as
“Outstanding Florida Waters”?
16. What steps will the U.S. Army Corps of Engineers take to safeguard the Indian River
Lagoon, which has been designated an “estuary of national significance”?
17. How does the diversion of billions of gallons of polluted and toxic water comply with the
U.S. Army Corps of Engineers mandate to take into consideration the human health
impacts of the people surrounding the estuaries to the east and west of the lake?
18. What scientific reviews has the U.S. Army Corps of Engineers completed to study the
degree to which the proposed action affects public health of the communities surrounding
the St. Lucie and Caloosahatchee estuaries?
23
April 22, 2019
19. What scientific information is the U.S. Army Corps of Engineers relying upon to
determine whether the proposed action complies with the authorized project purpose of
protecting water quality in the estuaries?
20. When considering the water quality impacts of the proposed action, does the U.S. Army
Corps of Engineers give at least equal weight to the impacts on the St. Lucie and
Caloosahatchee estuaries as compared to Lake Okeechobee and other bodies of water in
Florida?
21. To what extent does the U.S. Army Corps of Engineers include climate change modeling
in your decision-making process?
22. When completing economic analysis, with what information (please cite sources) and
how does the U.S. Army Corps of Engineers balance the actual realized harm to the
communities surrounding the St. Lucie and Caloosahatchee estuaries with the entirely
hypothetical, unrealized harm to the agricultural communities south of Lake
Okeechobee?
23. How does the taxpayer-funded U.S. Army Corps of Engineers weigh public-use benefits
of water against private, for-profit uses of water when making decisions about water
supply?
24. Does the U.S. Army Corps of Engineers acknowledge that it is a risk to the communities
surrounding Lake Okeechobee when the water levels on the lake are held too high?
25. What actions are being taken to protect the endangered and threatened species in the St.
Lucie and Caloosahatchee estuaries?
26. How does the proposed action of releasing billions of gallons of polluted water comply
with the Endangered Species Act and the U.S. Army Corps of Engineers NEPA
requirement to consider impacts on wildlife?
27. How does the proposed action of releasing billions of gallons of polluted water comply
with the authorized project purpose of enhancement of fish and wildlife?
28. Does the U.S. Army Corps of Engineers, based on the large amount of peer-reviewed and
scientifically-studied material, acknowledge that it is “reasonably foreseeable” that if
actions are not taken to reduce the proliferation of toxic cyanobacteria that fish and
wildlife will continue to be killed off?
29. What steps are being taken as part of this proposed action to ensure full utilization of
taxpayer-funded storage and treatment projects?
30. What impact will the repairs of the Herbert Hoover Dike have on the proposed action?
31. What flexibility will be built into the proposal to ensure that future storage constructed
under CERP and CEPP are taken into account?
32. Does the U.S. Army Corps of Engineers acknowledge that the scientific consensus is that
periodic reductions in lake levels are beneficial both to the ecology of Lake Okeechobee
and the health of the estuaries?
no discharges are beneficial to the ecology of the St. Lucie estuary and river? If the U.S.
Army Corps of Engineers disagrees with this assessment, please explain why and what
your assessment is.
24
April 22, 2019
minimum dry-season flow levels of 650 to 1000 cfs would be beneficial to the ecology of
the Caloosahatchee estuary? If the U.S. Army Corps of Engineers disagrees with this
assessment, please explain why and what your assessment is.
35. Will the science of what amount of fresh water discharge out of Lake Okeechobee is
good for, or harmful to, each estuary be taken into consideration?
36. Under LOSOM, what specific procedures will be in place to consider the proliferation of
harmful algal blooms on Lake Okeechobee before discharging water into the St. Lucie or
Caloosahatchee estuaries?
37. In considering alternative to the proposed action, will the U.S. Army Corps of Engineers
consider a model that includes permanently removing the St. Lucie River as an option for
discharging water out of Lake Okeechobee? If so, what were the impacts of doing so on
the health of the St. Lucie estuary and the broader CS&F system?
38. What percentage of water entering the taxpayer funded and publicly managed stormwater
treatment and water conservation areas comes off of privately-owned land in the
Everglades Agricultural Area compared to water flowing from Lake Okeechobee? If
private land owners in the Everglades Agricultural Area were required to provide
independent retention and treatment of water used on their land how would this policy
change impact the amount of water available to flow south out of Lake Okeechobee and
the amount of water required to be discharged to the east and west out of the lake?
39. In the case of an extended drought period, what options are available to the U.S. Army
Corps of Engineers to ensure water supply is available to public utilities, such as for
drinking water?
40. In the case of high-water events what options are available to the U.S. Army Corps of
Engineers to mitigate the need to discharge water into the St. Lucie and Caloosahatchee
estuaries?
41. Why is the U.S. Army Corps of Engineers exempt from needing a permit under the Clean
Water Act, other federal law or state law to discharge polluted water from Lake
Okeechobee to the St. Lucie or Caloosahatchee estuary?
42. How is back pumping water from the Everglades Agricultural Area into Lake
Okeechobee during extreme weather events consistent with the U.S. Army Corps of
Engineers’ mission as it relates to the stability of the Hebert Hoover Dike?
43. Has the U.S. Army Corps of Engineers ever been required to or actually did obtain a
permit for discharging water as part of the CS&F project?
44. Over the operational history of LORS 2008, did the U.S. Army Corps of Engineers make
operational suggestions to the South Florida Water Management District that would have
increased the U.S. Army Corps of Engineers capacity to move water south from Lake
Okeechobee or otherwise decreased the need to discharge water to the estuaries that were
not implemented by South Florida Water Management District? If so, what were they?
45. Does the U.S. Army Corps of Engineers acknowledge that the results of LORS 2008
created 9 years of discharge and only 1 year of drought levels? If so, does the U.S. Army
Corps of Engineers believe this is an equitable distribution of shared adversity?
25
April 22, 2019
46. Does the U.S. Army Corps of Engineers believe that widespread risk to public health,
such as is created by the discharge of toxic algal blooms in the estuaries, is an acceptable
result of LOSOM?
IX. CONCLUSION
In conclusion, LORS 2008 has caused substantial damage to the environment in Florida and also
to the public health of Florida’s inhabitants. The design has forced water where it is not wanted
(and extremely harmful) and starved the areas that do want the water.144 As a result, the St.
Lucie estuary, which historically received no freshwater flows, is flooded with billions (or
hundreds of billions) of gallons of polluted fresh water each year.145
The U.S. Army Corps of Engineers operational priorities, along with the requirements under
NEPA, provide important guardrails to ensure that a system as fundamentally flawed as LORS
2008 is never allowed to happen again. Instead, the U.S. Army Corps of Engineers should
design a system that includes substantially added flexibility ensuring maximum flows south of
the lake and minimum flows to the estuaries. This can be accomplished through added flexibility
at both the high and low ends of the system, and an overall lower operational level for Lake
Okeechobee than under LORS 2008. Moreover, the U.S. Army Corps of Engineers must
incorporate specific and stringent safeguards to protect Florida communities from the dangerous
health impacts of exposure to the toxins contained in these discharges. Ultimately, the goal must
be zero discharges to the St. Lucie and only the beneficial dry-season flows requested by the
Caloosahatchee.
The public health consequences of the project you are undertaking could not be more severe.
This is truly a life or death proposition for Florida’s environment, our economy, the people that
call Florida home and the millions of visitors we receive every year.
144
Rehage, supra note 57.
145
26
APPENDICES
CONTENTS
APPENDIX A: Constituent Scoping Comments
APPENDIX B: Letter From Col. Jason A. Kirk
APPENDIX C: U.S. Army Corps Of Engineers LOSOM Presentation
APPENDIX D: University of Florida Water Institute Study
APPENDIX E: News Reports
APPENDIX F: U.S. Department Of The Interior Study
APPENDIX G: World Health Organization Study
APPENDIX H: EPA Documents
APPENDIX I: CDC Documents
APPENDIX J: ORCA Study Summary
APPENDIX K: Environmental Health Study
APPENDIX L: The Royal Society Publishing Study
APPENDIX M: Wetlands Conceptual Ecological Model
APPENDIX N: Dr. Gary Goforth Letter To U.S. Army Corps Of Engineers
APPENDIX O: Dr. Gary Goforth Water Year 2018 Analysis
APPENDIX P: FDEP 2018 Algal Bloom Sampling Results
APPENDIX Q: Dr. Gary Goforth Septic Tank Analysis
APPENDIX R: National Academies Of Science Everglades Report
APPENDIX S: Indian River Lagoon Economic Valuation Update
APPENDIX T: Florida Realtors Association Study
APPENDIX U: South Florida Water Management District Fact Sheet
APPENDIX V: Florida Fish And Wildlife Manatee Mortality Statistics
APPENDIX W: BMAA Studies
APPENDIX X: Audubon Letter
APPENDIX Y: Letters Regarding Minimum Flow Levels For The Caloosahatchee
APPENDIX Z: Florida International University Study
APPENDIX A:
CONSTITUENT SCOPING COMMENTS
APPENDIX B:
LETTER FROM COL. JASON A. KIRK
APPENDIX C:
U.S. ARMY CORPS OF ENGINEERS LOSOM
PRESENTATION
APPENDIX D:
UNIVERSITY OF FLORIDA WATER INSTITUTE
STUDY
Options to Reduce High Volume Freshwater Flows to the St. Lucie
and Caloosahatchee Estuaries and Move More Water from Lake
Okeechobee to the Southern Everglades
An Independent Technical Review by the University of Florida Water Institute
March 2015
Authors
Wendy D. Graham
Director, Water Institute
Professor and Carl S. Swisher Eminent Scholar Chair in Water Resources
University of Florida
Mary Jane Angelo

Director, Environmental and Land Use Law Program
Professor, Levin College of Law, University of Florida
Thomas K. Frazer
Director, School of Natural Resources and Environment
Professor, School of Forest Resources and Conservation, Institute of Food and Agricultural
Sciences, University of Florida
Peter C. Frederick
Research Professor, Wildlife Ecology and Conservation, Institute of Food and Agricultural
Karl E. Havens
Director, Florida Sea Grant
Professor, School of Forest Resources and Conservation, Institute of Food and Agricultural
K. Ramesh Reddy
Graduate Research Professor and Chair, Soil and Water Science, Institute of Food and
Agricultural Sciences, University of Florida
This report provides a review of relevant plans, projects and related technical documents as part of a broader
effort to evaluate options to move more fresh water from Lake Okeechobee to the Everglades. The review was
stimulated, in large part, by concern that recent regulatory releases of fresh water from Lake Okeechobee to
the St. Lucie and Caloosahatchee estuaries have resulted in substantial negative ecological and economic
impacts. Findings and opinions expressed herein are the collective work of a team of faculty affiliated with the
University of Florida Water Institute and are based solely on pre-existing information.
© University of Florida, March 2015
2
Table of Contents
Executive Summary ................................................................................................................................... 5
I. Background and Technical Review Team Charge...................................................................... 11
II. Challenges to Reducing High Volume Freshwater Flows to the Estuaries and Moving More Water
South from Lake Okeechobee to the Southern Everglades ...................................................... 15
1. A Complex Inter-Connected System ........................................................................................... 15
2. Climatological, Topographic and Geologic Constraints ............................................................ 16
3. Engineering Constraints due to System Design ........................................................................ 17
4. Need for Sustained, Long-term, Holistic Planning and Execution .......................................... 18
5. Legal Context .................................................................................................................................. 19
III. Existing State and Federal Plans to Reduce High Volume Freshwater Flows to the Estuaries and
Move More Water South from Lake Okeechobee to the Southern Everglades ..................... 28
1. Introduction ...................................................................................................................................... 28
2. Planning Efforts ............................................................................................................................... 33
a. Comprehensive Everglades Restoration Plan (CERP) .......................................................... 33
b. Northern Everglades and Estuaries Protection Program (NEEPP)...................................... 34
c. River of Grass (ROG) Planning Process .................................................................................. 36
3. Construction Project Status........................................................................................................... 38
a. CERP Indian River Lagoon-South (IRL-S) Project ................................................................. 38
b. CERP Caloosahatchee River (C-43) West Basin Storage Projects..................................... 41
c. CERP Aquifer Storage and Recovery (ASR) Pilot Projects and Regional Study ............... 42
d. NEEPP Lake Okeechobee Watershed Construction Project Phase II Technical Plan ..... 43
e. NEEPP St. Lucie River Watershed Protection Plan ............................................................... 48
f. NEEPP Caloosahatchee River Watershed Protection Plan ................................................... 49
g. State of Florida Restoration Strategies ..................................................................................... 51
h. CERP Central Everglades Planning Project (CEPP) ............................................................. 52
4. Summary .......................................................................................................................................... 56
IV. Existing State Plans to Improve Water Quality ............................................................................ 60
1. Greater Everglades Watershed Descriptions ............................................................................. 60
2. Greater Everglades Watershed Nutrient Loads ......................................................................... 63
3. Current and Planned Management of Greater Everglades Watershed Nutrient Loads ...... 72
4. Summary .......................................................................................................................................... 84
3
V. Options (beyond approved projects) to Reduce High Volume Freshwater Flows to the Estuaries
and Move More Water South from Lake Okeechobee to the Southern Everglades.............. 85
1. Introduction ...................................................................................................................................... 85
2. Storage and Treatment North of Lake Okeechobee ................................................................. 87
3. Additional Storage, Treatment and Conveyance South of Lake Okeechobee ..................... 89
a. Plan 6 and Other Flow-way Options ......................................................................................... 90
b. Other South of the Lake Storage, Treatment and Conveyance Options........................... 101
c. Summary ..................................................................................................................................... 106
4. Deep Well Disposal of Excess Flows ........................................................................................ 107
5. Operational Changes ................................................................................................................... 110
a. Modification of Lake Okeechobee Regulation Schedule ..................................................... 110
b. Modification of Holey Land and Rotenberger Regulation Schedules ................................ 116
VI. Future Uncertainties....................................................................................................................... 120
1. Potential Effects of Climate Change .......................................................................................... 120
2. Effects of Changes in Human Population Size, Location and Land Use ............................. 125
3. Other Possible Future Changes ................................................................................................. 125
4. Uncertainty in Future Funding .................................................................................................... 127
5. Summary ........................................................................................................................................ 129
VII. Summary and Conclusions ........................................................................................................... 130
VIII. List of Abbreviations....................................................................................................................... 135
IX. References ...................................................................................................................................... 137
X. Review Team Biographical Sketches.......................................................................................... 142
4
Executive Summary
Background and Technical Review Team Charge
It is widely recognized that the flood control and water delivery system that serves Florida’s urban and
agricultural interests has substantially and adversely impacted natural ecosystems in south Florida,
including the St. Lucie and Caloosahatchee estuaries, Lake Okeechobee and the Everglades
Protection Area (EPA). The environmental problems stem from periods when there is too much water,
periods when there is too little water, and a regional delivery system that quickly transports nutrients
from upstream agricultural and urban sources to natural systems where adverse impacts occur. When
South Florida receives a large amount of rainfall, there are often damaging freshwater discharges to
both east coast and west coast estuaries, whereas prolonged drought strains the capacity of the
regional system to deliver sufficient water to its full complement of end users. Regardless of the
regional hydrologic regime, much of the EPA remains chronically deprived of fresh water necessary to
sustain remnant habitats and native biota.
For decades, planning has been underway by federal and state agencies as well as key stakeholders
to develop solutions to water related issues in the South Florida Ecosystem. The Comprehensive
Everglades Restoration Plan (CERP), Northern Everglades and Estuaries Protection Program
(NEEPP) and River of Grass Planning Process (ROG) are exemplars of important planning efforts
that guide contemporary restoration activities. Despite these extensive planning efforts, however, little
has been done to solve the regional problems identified above.
In response to stakeholder concerns about the timing and completion of South Florida Ecosystem
restoration, the 2014 Florida Senate authorized an independent technical review of options to reduce
high volume freshwater flows to the St. Lucie and Caloosahatchee estuaries and move more water
from Lake Okeechobee to the Everglades, to be conducted by the University of Florida (UF) Water
Institute. Specifically, the interdisciplinary UF Technical Review Team was charged with reviewing
existing documents that have set forth plans and projects to reduce regulatory discharges from Lake
Okeechobee to the St. Lucie and Caloosahatchee estuaries and increase the flow of water from the
lake to the southern Everglades; identifying scientific, engineering, legal and institutional constraints to
implementing the identified plans and projects; and identifying options for accelerated and more
effective protection of the estuaries and restoration of the Everglades. Key findings of the Technical
Review Team are summarized below.
Challenges to Reducing High Volume Freshwater Flows to the Estuaries and Moving More
Water South from Lake Okeechobee to the Southern Everglades
After extensive interviews with experts and evaluation of existing plans, supporting studies and other
documentation, the UF Technical Review Team identified a number of fundamental challenges to
5
reducing the frequency and duration of freshwater discharges to the St. Lucie and Caloosahatchee
estuaries while at the same time increasing the flow of water south of the lake. These challenges
include the complex and inter-connected nature of the Greater Everglades system with its flat
topography, porous geology and highly variable climate; the reduced footprint of the Everglades
system to approximately half its original size; the much larger capacity of canals and structures that
provide inflow to Lake Okeechobee relative to those that provide outflow; the much smaller capacity of
outflow canals and structures to carry water south of Lake Okeechobee versus east and west;
flooding risks in agricultural and urban areas surrounding and southeast of Lake Okeechobee; legal
limits for phosphorus loading to the EPA and Everglades National Park (ENP); regulation schedules
for the Water Conservation Areas (WCAs) in the EPA intended to protect ridge, slough and tree island
habitats and associated wildlife; constraints imposed by the existing and sometimes conflicting rights
of legal water users; and the need to comply with existing laws and court orders.
Despite these challenges, the Technical Review Team concludes that relief to the estuaries and the
ability to move more water south of Lake Okeechobee can be accomplished using existing
technology. The solution is enormous increases in storage and treatment of water both north and
south of the lake. Existing and currently authorized storage and treatment projects are insufficient to
achieve these goals. The path forward requires significant long-term investment in the infrastructure of
the South Florida hydrologic system.
Options to Reduce High Volume Freshwater Flows to the Estuaries and Move More Water
South from Lake Okeechobee to the Southern Everglades
To reduce damage to the St. Lucie and Caloosahatchee estuaries freshwater inflow and nutrient loads
from both Lake Okeechobee and the local basins must be reduced. On average, 70-80% of the
freshwater discharge and 65-80% of the nutrient load to the St. Lucie and Caloosahatchee estuaries
originates in the local basins, with the remaining balance contributed from Lake Okeechobee.
Previous CERP, NEEPP and ROG planning exercises have all identified that providing large volumes
of regional storage is essential to reduce freshwater discharges to the estuaries. The most recent
estimates of required storage include:
 400,000 acre-feet of water storage within the Caloosahatchee River watershed,
 200,000 acre-feet of water storage within the St. Lucie River watershed, and
 approximately 1,000,000 acre-ft of water storage distributed north and south of Lake
Okeechobee.
In spite of the repeated demonstrated need for large volumes of water storage, very little new storage
has been designed or constructed in the system. For example, in the St. Lucie watershed it is
estimated that approximately 200,000 acre-ft of storage is required. However, only one 40,000 acre-ft
surface reservoir is currently under construction. In the Caloosahatchee watershed, it is estimated that
6
approximately 400,000 acre-ft of storage is needed, but currently only one 170,000 acre-ft surface
reservoir is being designed, and state and federal funds for its construction have not yet been
appropriated. Furthermore, although at least one million acre-ft of storage is required either north or
south of Lake Okeechobee, currently only four Flow Equalization Basins (FEBs) that provide 168,000
acre-ft of shallow storage are planned and they are sited south of Lake Okeechobee. Two of the FEBs
(totaling 101,000 acre-ft) currently are under construction by the State and are scheduled to be
completed by 2016. State construction of a third 11,000 acre-ft FEB will not begin until after 2018. The
fourth CERP FEB has yet to be authorized by the US Congress.
Based on review and analyses, the Technical Review Team identified the following options to reduce
damaging discharges to the St. Lucie and Caloosahatchee estuaries and move more water south
from Lake Okeechobee:
1. Accelerate funding and completion of existing approved projects
To provide substantial improvement to the St. Lucie and Caloosahatchee estuaries, accelerate the
funding and completion of existing federally authorized CERP projects designed specifically to provide
relief to St. Lucie and Caloosahatchee Basins, i.e.:
 Indian River Lagoon-South (IRL-S) Project: Accelerate construction of the C-44 reservoir and
associated Stormwater Treatment Area (STA). Aggressively pursue state and federal
appropriations needed to design and construct remainder of the IRL-S project (including C-23,
24, 25 reservoirs and associated STAs, and restoration of over 90,000 acres of upland and
wetland areas).
 C-43 Reservoir: Accelerate the design and aggressively pursue state and federal
appropriations needed to design and construct project.
Current Basin Management Action Plans (BMAPs) will not achieve Florida Department of
Environmental Protection (FDEP) approved Total Maximum Daily Loads (TMDLs). To achieve water
quality standards in Lake Okeechobee, the St. Lucie estuary and the Caloosahatchee estuary, more
aggressive BMAPs are required. New field-verified agricultural and urban Best Management Practices
(BMPs) that protect water quality, advanced in situ treatment technologies, and the strategic
placement of additional FEB-STAs in priority basins will be essential to achieve State and Federal
water quality standards. Beyond existing and planned approaches, the substantial reservoir of legacy
phosphorus in the Northern Everglades watersheds will necessitate new and more aggressive
strategies to combat the mobility of phosphorus.
7
To substantially increase the volume of water moving from Lake Okeechobee to the Southern
Everglades accelerate funding and completion of the State of Florida Restoration Strategies and the
CERP Central Everglades Planning Project (CEPP), i.e.:
 Obtain federal authorization for CEPP,
 Accelerate the design and obtain state and federal appropriations for the construction of
CEPP,
 Accelerate State funding and completion of Restoration Strategies,
 Conduct a careful analysis of CEPP project construction phasing to determine which CEPP
features can be constructed as soon as possible and to develop a plan for completion of as
many CEPP features as possible during the construction phase of Restoration Strategies, and
 Reconsider using the Talisman property for a deep storage reservoir with STA rather than the
current design which uses the Talisman property for shallow FEBs.
Additional efforts, beyond the approved projects listed above, will be required to reduce Lake
Okeechobee-triggered high discharges and nutrient loads to the St. Lucie and Caloosahatchee
estuaries and to achieve dry season Everglades demand targets. Studies indicate that after the
completion of the IRL-S, C-43, Restoration Strategies, and CEPP projects, lake-triggered high
discharges to the St. Lucie and Caloosahatchee estuaries will be reduced by less than 55% and less
than 75% of the dry season Everglades demand target will be delivered to the EPA. A series of
options, beyond currently approved projects, to more fully achieve restoration objectives are
summarized below.
2. Provide Water Storage and Treatment North of Lake Okeechobee
Conduct a strategic planning exercise to provide additional water storage and treatment north of Lake
Okeechobee similar to the ROG Planning Process that was conducted south of the lake. The NEEPP
Lake Okeechobee Phase II Technical Plan (LOP2TP) and the ROG Planning Process provide a
sound foundation from which to plan, design, and build the additional storage and treatment needed
north of Lake Okeechobee. A new strategic planning exercise would necessarily include a regional
modeling effort that takes into account lessons learned and information gained since the CERP,
NEEPP and ROG planning exercises. Examples of new information gained include the permitting
requirements, engineering feasibility and costs, and inter-annual storage benefits associated with
Aquifer Storage and Recovery (ASR), deep storage reservoirs, shallow water impoundments and
dispersed water management (DWM), as well as the water quality benefits of Stormwater Treatment
Areas (STAs) and other treatment technologies. New data gathering efforts and model developments
will be required to simulate the cumulative impacts of a regional DWM system north of the lake on the
quality, quantity and timing of flows into Lake Okeechobee as a function of climatic conditions, spatial
8
location and density of DWM features on the landscape, and operation of the regional canal system.
The Technical Review Team expects that the strategic plan will show that, while DWM on private
lands may provide some benefits, DWM will fall short of providing the additional storage and treatment
needed, even if fully implemented. Additional land north of Lake Okeechobee will need to be acquired
for that purpose.
3. Provide Additional Water Storage, Treatment and Conveyance South of Lake Okeechobee
Develop a strategic plan for the next increment of south-of-lake storage, treatment and conveyance to
pursue beyond CEPP to take advantage of new north-of-lake storage and treatment, and more closely
meet the performance targets of both the estuaries and the Everglades ecosystems. Independent
assessments suggest that an expansive gravity-driven wet flow-way throughout the Everglades
Agricultural Area (EAA) may not be feasible or provide maximal benefits to the estuaries. However,
the ROG planning process demonstrated that there are several possible options involving
combinations of deep and shallow storage, and wet- and dry- flow-ways, coupled with STAs and
enhanced conveyance that could provide significant benefit both for the estuaries and the Everglades,
far beyond the benefits provided by the Kissimmee River Restoration (KRR), IRL-S, C-43, Restoration
Strategies and CEPP projects. Achieving substantial reduction in lake-triggered discharges to the
estuaries and substantial improvement toward the dry season Everglades demand target will require
additional land between the lake and the EPA, e.g., the current U.S. Sugar land purchase option,
lands from other willing sellers, and/or use of existing state-owned land (e.g., Holey Land and
Rotenberger Wildlife Management Areas (WMAs)).
4. Deep Well Disposal of Excess Flows
Deep well disposal could be part of a long-term solution to reducing damaging discharges from Lake
Okeechobee to the St. Lucie and Caloosahatchee estuaries, or it could provide an interim solution
until additional water storage, treatment and conveyance capacity can be constructed south of the
lake. If sufficient inter-annual storage and treatment north of the lake is determined to be economically
or politically infeasible, or the analyses indicate that the captured water cannot be efficiently treated
and conveyed south of the lake for use in subsequent dry seasons, the option of constructing a
system of large injection wells to permanently dispose of excess flows from Lake Okeechobee in the
deep Boulder Zone, rather than discharging to the estuaries, should be explored.
5. Operational Changes
Adjustments within the current Lake Okeechobee regulation schedule (LORS 2008) are unlikely to
have a substantive effect on the occurrence of damaging high discharges to the estuaries. However, a
substantially revised regulation schedule that provides more storage in the lake might provide those
benefits. Developing a new regulation schedule requires completion of the on-going U.S. Army Corps
9
of Engineers (USACE) Dam Safety Modification Study and guidance about the safety of the
rehabilitated Herbert Hoover levee and operational structures in light of any new safety standards.
The USACE should accelerate completion of the Dam Safety Modification Study so that modification
of the Lake Okeechobee regulation schedule, if warranted, can occur as soon as possible.
Development of a new regulation schedule will require balancing benefits of holding additional water
in the lake for the express purpose of reducing damaging discharges to the estuaries and increasing
agricultural, urban and ecosystem water supply versus potential adverse impacts to the lake’s
ecology.
In the interim, to provide incremental estuarine relief, Lake Okeechobee operations could be modified
within the discretionary bands of LORS 2008. Increasing the dynamic range of storage in the lake
could allow some additional water to be moved south to the EPA, and also provide increased dry
season flows to the Caloosahatchee estuary and EAA. In addition, the regulation schedules of the
Holey Land and Rotenberger WMAs could be modified to allow more water storage during both the
wet and dry seasons. This modification of the WMA regulation schedules could be in keeping with
current goals to restore natural hydroperiods, but will require the inflow/outflow infrastructure be
upgraded to allow dynamic water level manipulations.
Future Uncertainties
Failure to draw on information about the range of possible future conditions risks the success of
restoration project outcomes. Substantive research indicates clearly that climate change, changes in
human demographics, energy costs and land use will affect Florida’s future, yet there is little evidence
that salient information is being incorporated into restoration project plans. Even if the future of these
variables is highly uncertain, the possibility of future changes needs to be acknowledged, effects on
restoration outcomes assessed, and flexibility incorporated into projects so that they can have positive
outcomes over a broad range of conditions.
Path Forward
Even in the face of uncertainty, many existing plans and projects have been fully vetted and can be
expected to yield substantial benefits to the citizens of Florida. Most of the projects are delayed
because of a lack of funding. In the interim, the coupled human-ecological system is continuing to
degrade in ways that may not be reversible. Monitoring and assessment of system performance is
essential to guide projects and to detect and adapt to future surprises. Increased and sustained State
and Federal funding is critical to provide additional water storage and treatment before the system
becomes so degraded that major attributes reach tipping points that cannot be reversed.
10
I. Background and Technical Review Team Charge
An extensive network of man-made canals, levees and water control structures permeates the south
Florida landscape. The land has been ditched, drained and otherwise reconfigured to provide flood
protection and fresh water for a current population of more than eight million residents while
simultaneously serving the needs of a multi-billion dollar agricultural industry (Hodges et al, 2014).
Major projects in south Florida include the Herbert Hoover dike around Lake Okeechobee which was
initiated in 1930, and the massive Central and Southern Florida (C&SF) Project, begun in 1948. From
an engineering perspective, this regional water distribution and delivery system is highly effective at
meeting its intended project purposes. Improved human welfare and economic prosperity are tangible
consequences of these large projects targeted mainly at improving flood control and water supply.
The water supply and flood control functions of the C&SF continue to be of critical importance to south
Florida. The originally authorized C&SF Project also provided for conservation of natural resources
and, in particular, the preservation of fish and wildlife. In this regard, however, the C&SF Project has
underperformed and, in fact, has resulted in considerable ecological decline over time.
It now is widely recognized that the flood control and water delivery system that serves Florida’s
human population and agricultural interests has substantially and adversely impacted natural
ecosystems in south Florida, including the St. Lucie and Caloosahatchee estuaries, Lake
Okeechobee and the Everglades Protection Area (EPA) which includes the Water Conservation Areas
(WCAs) and Everglades National Park (ENP) (Figure I-1). The environmental problems stem from
periods when there is too much water, periods when there is too little water, and a regional delivery
system that quickly transports nutrients from upstream agricultural and urban sources to natural
systems where adverse impacts occur. When south Florida receives a large amount of rainfall, there
are often damaging freshwater discharges to both east coast and west coast estuaries, whereas
prolonged drought strains the capacity of the regional system to deliver sufficient water to its full
complement of end users. During times of high rainfall, especially those following droughts, nutrients
are flushed from soils and wetlands into Lake Okeechobee, the estuaries and the EPA. Yet, except for
the periods of highest rainfall, much of the EPA including ENP, remains chronically deprived of fresh
water necessary to sustain remnant habitats and native biota.
Reducing the frequency and duration of damaging freshwater discharges to the St. Lucie and
Caloosahatchee estuaries while at the same time increasing the flow of water south through the
Everglades and into Florida Bay is a complicated task. In the system’s pre-engineered state, during
high water events, the vast majority of water coming into Lake Okeechobee overflowed the southern
rim of the lake and was carried south into the Everglades as sheet flow. However, urban and
suburban development along the eastern and western margins of the historic Everglades, and
conversion of marsh land south of Lake Okeechobee into agricultural production in what is now the
Everglades Agricultural Area (EAA), have reduced the Everglades to approximately one half of its
11
original size. As a consequence, the volume of water that can flow out of Lake Okeechobee to the
south without causing harm to agricultural or urban/suburban areas has been reduced significantly.
Figure I-1. Map of the Greater Everglades Ecosystem showing the extent of the Northern Everglades, Southern Everglades,
Everglades Agricultural Area (EAA), Everglades Protection Area (EPA) and Water Conservation Areas (WCAs)
(SFWMD, 2015)
12
A number of additional constraints limit the amount and timing of water that can be discharged to the
south from Lake Okeechobee. These constraints include the much larger capacity of canals and
structures that provide inflow to the lake than those that provide outflow; much smaller capacity of
outflow canals and structures to carry water south of Lake Okeechobee versus east and west;
flooding risks in agricultural and urban areas; legal limits for phosphorus loading to the EPA and ENP;
and regulation schedules for the Water Conservation Areas (WCAs) in the EPA intended to protect
ridge, slough and tree island mosaics and wildlife. These are familiar obstacles to those engaged in
the process of Everglades restoration (see section II below for more detail).
For decades, planning has been underway to develop solutions to problems associated with the
C&SF Project. In the 1990s, the US Army Corps of Engineers (USACE) carried out a Reconnaissance
Study to evaluate potential approaches to regional restoration. That study and the resultant
recommendations (USACE, 1994) led to the C&SF Restudy and ultimately to the Comprehensive
Everglades Restoration Plan (CERP; USACE,1999) which is being implemented by the USACE and
its local partner the South Florida Water Management District (SFWMD). Recently, these
management entities have focused their efforts on a component of CERP, the Central Everglades
Planning Project (CEPP; USACE, 2014b), which was designed to direct more water through the
Everglades Protection Area, and deliver it ultimately to the estuaries of Everglades National Park and
Florida Bay.
In 2013, after a particularly wet season in the region and large regulatory discharges from Lake
Okeechobee to the St. Lucie and Caloosahatchee estuaries, concerns were raised again about the
timing and completion of CERP and other restoration projects. Interest focused on accelerated
construction and completion of projects that would reduce damaging freshwater releases to the
estuaries and send more of that water, cleaned of phosphorus, to the Everglades. A broad suite of
stakeholders has, in fact, questioned whether there are more immediate solutions, especially to the
problem of high discharges of nutrient-laden fresh water to the estuaries. In response to the
recommendations of the Florida Senate Select Committee on Indian River Lagoon and Lake
Okeechobee Basin, the 2014 Florida Legislature appropriated $232 million to accelerate projects
intended to take pressure off the estuaries and restore the Everglades.
Everglades Technical Review Team Study Objectives and Approach
In addition to appropriating funds to accelerate construction projects, the 2014 Florida Senate
authorized an independent technical review of options to reduce damaging discharges to the St. Lucie
and Caloosahatchee estuaries and move water from Lake Okeechobee to the Everglades, to be
conducted by the University of Florida (UF) Water Institute. Specifically, the interdisciplinary UF
Technical Review Team was charged with reviewing existing documents that have set forth plans and
projects to reduce regulatory discharges from Lake Okeechobee to the St. Lucie and Caloosahatchee
13
estuaries and increase the flow of water from the lake to the southern Everglades; identifying
scientific, engineering, legal and institutional constraints to implementing the identified plans and
projects; and identifying options for accelerated and more effective protection of the estuaries and
restoration of the Everglades.
After extensive interviews with experts and evaluation of existing plans, supporting studies and other
documentation, the Technical Review Team identified a number of fundamental challenges to
reducing the frequency and duration of damaging freshwater discharges to the St. Lucie and
Caloosahatchee estuaries while at the same time increasing the flow of water south through the
Everglades and ultimately into Florida Bay. We present an analysis of these challenges in Section II of
this report as they are crucial for understanding the constraints on potential solutions to water
management issues in south Florida. Section III presents analyses of water storage needs for
reducing freshwater discharge to the estuaries, and a summary of existing State and Federal
approved plans that would achieve some measure of relief. Section IV provides insight into the
treatment capacity needed to improve water quality and reduce nutrient delivery to Lake Okeechobee,
the St. Lucie and Caloosahatchee estuaries, and also to move water south from Lake Okeechobee to
the EPA in compliance with regulatory standards. Section V provides a series of options, beyond
currently approved projects, to reduce damaging discharges to the estuaries and move more water
south. Section VI summarizes future uncertainties that may impact restoration of the Greater
Everglades Ecosystem and Section VII summarizes the findings and conclusions of the review team
effort.
14
II. Challenges to Reducing High Volume Freshwater Flows to the
Estuaries and Moving More Water South from Lake Okeechobee to
the Southern Everglades
The Comprehensive Everglades Restoration Plan (CERP; USACE, 1999) was developed to provide
regional solutions to problems associated with the Central and South Florida (C&SF) Project. The
overarching goals of the plan are to achieve the restoration, preservation and protection of the south
Florida ecosystem while providing for other water related needs of the region. Ecosystem restoration
goals include: 1) improving the volume, timing and quality of water entering Lake Okeechobee and
managing the lake as an ecological resource; 2) improving the volume, timing and quality of water
delivered to the Caloosahatchee and St. Lucie estuaries; 3) improving the volume, timing, spatial
distribution and quality of water entering the Everglades Protection Area (EPA); and 4) improving
freshwater flows to Florida Bay. Additional goals include improving urban and agricultural water
supplies.
CERP is generally accepted as the appropriate framework for dealing with decades of accumulated
water related issues, however even after decades of planning and some project implementation,
much of CERP has yet to be constructed and major regional problems remain. This section of the
report discusses some of the key reasons that there continues to be major issues with regulatory
discharges to the estuaries and a lack of an adequate volume of clean freshwater delivered to the
Everglades.
1. A Complex Inter-Connected System
The Greater Everglades ecosystem is vast and complex. Its footprint is similar in size to the state of
New Jersey, and many decades of engineering coupled with urban and agricultural development have
added layers of complexity and expectations for performance that were not there even 40 years ago.
The size, intricately balanced nature, and especially the many competing interests, make it very
challenging for holistic solutions to emerge. The following sections illustrate the specific physical,
ecological, design, and legal features that constrain solutions to almost any hydrologic issue in south
Florida.
The first feature that constrains south Florida is the strong interconnectedness of the hydrologic
system that serves so many interests. All parts of the Greater Everglades ecosystem share a common
water resource, and the entire area is therefore a strongly connected hydrologic system. For example,
the water stored in Lake Okeechobee is used to grow crops in the Everglades Agricultural Area
(EAA), recharge aquifers used for public water supplies, provide dry season flows to the
Caloosahatchee estuary, and to hydrate the Everglades Protection Area. Accordingly, actions taken to
address storage, conveyance or distribution in one part of the system invariably affect other parts of
the system. Further, the competing needs are typically synchronous - virtually all parts of the system
are concerned with flood reduction during the rainy season, and all are competing for scarce water
15
during the dry season. Compounding this situation is the fact that conversion of natural lands to
agricultural and urban uses has increased nutrient loads creating a broad suite of negative impacts. In
undertaking any restoration or re-planning of this water system, it is crucial to recognize that in the
context of a strongly shared resource and high connectivity, trade-offs are usually unavoidable and
difficult policy choices will have to be made.
2. Climatological, Topographic and Geologic Constraints
One of the most obvious drivers of hydrology and water management in the south Florida region is the
marked variability in seasonal and annual rainfall that inexorably leads to patterns of overabundance
and scarcity of water. Within the average year, 70% of rainfall comes in just five months, with a
pronounced dry season during the rest of the year. Further, annual rainfall is highly variable from year
to year, varying as much as 42 inches (82% of mean) between very wet and very dry years. This is in
part due to large swings in global weather patterns that are driven by processes well outside the south
Florida region. In winter months, the El Niño phase of the El Niño Southern Oscillation (ENSO) cycle
can result in 2 – 7 times as much winter rainfall than during the La Niña phase (Abtew and Trimble
2010) and in the summer, rainfall is 40% higher in the warm phase of the Atlantic Multidecadal
Oscillation than in the cold phase (Enfield et al., 2012). Furthermore, in summer months, tropical
storms and hurricanes, which also are strongly driven by global weather patterns, can deliver up to 25
inches of rainfall (half the total average annual precipitation) to the region in periods of only a few
days. Our ability to predict the occurrence or magnitude of any of these dominant, globally driven
weather scenarios, even one or two months in advance, is poor, and hence there is limited ability to
make proactive operational changes in the system to mitigate the resulting massive hydrological
variability. While the annual wet-dry cycle is somewhat predictable from a management point of view,
the interannual swings are not, and the need to accommodate large, unpredictable fluctuations in
rainfall will always be an overarching concern for human populations in south Florida.
Another major driver of water management in south Florida is topography and geology. The land is
extremely flat, making it difficult to move large quantities of water quickly by gravity. Even with some
of the largest pump stations in the world, the flat topography places clear constraints on the speed
with which any response can occur. Further, the flat landscape also offers virtually no natural storage
areas – without boundaries water spreads out, which means that the ability to store water must be
created by constructing surface impoundments or aquifer storage and recovery systems. The ability to
contain water in any space is uniquely limited in south Florida by the underlying parent rock – highly
porous limestone. The construction of dikes or ditches from this material is inherently problematic
because of the high transmissivity, and rapid movement of large volumes of water out of surface
impoundments and into nearby areas through groundwater seepage can only be controlled with
constructed and expensive seepage barriers. Finally, the ability to store water underground through
aquifer storage and recovery systems is constrained by the existence of locally intact confining layers
above the target storage zone, as well as the hydraulic and geochemical characteristics of the target
16
storage zone. In south Florida, this storage zone varies in extent, competence, and hydrologic
connectivity with other parts of the aquifer. This means that underground storage is only effective in
certain places, and the characteristics of storage and recovery are ultimately difficult to predict.
The highly variable rainfall and flat, porous landscape serve to create a highly variable hydrologic
system (Duever et al., 1994), in which periods of water scarcity and flood are fundamental
characteristics. The porous geology and extremely flat topography, in combination, impose important
constraints on the ability to control water and store it.
3. Engineering Constraints due to System Design
Historically, water entered Lake Okeechobee mainly from the Kissimmee River to the north,
Fisheating Creek to the west and from various smaller rivers, creeks and broad seepage fronts. The
vast majority of water exited the lake at the southern end into a large number of natural creeks that
dispersed into a dense forest of pond apple and then sawgrass plains. In its currently engineered
condition Lake Okeechobee receives substantially more water inflow from the north than can be
discharged south (to the EAA, the L-8 basin or the WCAs), and the Everglades itself has been
reduced to approximately half its former footprint. This is a largely planned hydrologic design with a
primary purpose to provide flood control for agricultural and urban areas around and to the southeast
of the lake, (USACE, 1955a). In the current system, with a levee surrounding the lake, and relatively
small canals to the south, the only outlets with adequate capacity to quickly release large amounts of
water from the lake are the canals that discharge to the St. Lucie and Caloosahatchee estuaries, C-44
and C-43, respectively.
In addition, there are constraints on the volume of water that can be held in the lake itself. Historically
the lake’s margin fluctuated considerably, and higher flows simply meant that the lake edge was
farther away from the center than in lower flow years. During the wet season, the lake typically
overflowed its southern boundary and the vast majority of outflow was to the Everglades. Now the
lake is more like a bathtub and the upper limit of water level, and storage capacity of the lake, are
determined by the design constraints of the Herbert Hoover Dike. When water levels rise, the lake
does not expand. Instead, it becomes deeper and floods its internal marsh or littoral zone (Aumen,
1995). A number of different US Army Corps of Engineer (USACE) regulation schedules have
regulated when water is discharged from the lake. Between 2000 and 2008 an operating schedule
called WSE (Water Supply and Environment) was followed. It was intended to provide the necessary
amount of flood protection while also allowing the lake to meet the needs of water users and minimize
environmental effects on the lake, estuaries, WCAs and Everglades. Most recently, a schedule called
Lake Okeechobee Regulation Schedule 2008 (LORS 2008) has been used, which aims to hold the
lake between 12.5 and 15.5 ft. surface elevation (above mean sea level) to minimize the risk of dike
failure during a period when the USACE is reinforcing the levee and water control structures around
the lake. While there is operational flexibility built into this schedule (see Section V.5) changes in
17
water storage created by this flexibility are small compared to the volumes of water that must be
accommodated in high rainfall years.
This basic problem of constrained outflows from Lake Okeechobee is fundamental to the issue of
regulatory releases to the St. Lucie and Caloosahatchee estuaries. To reduce flows to the estuaries, a
substantially greater amount of regional storage and enhanced water conveyance are needed.
Furthermore, because legal constraints resulting from water quality concerns (see Section II.5), it is
currently impossible to simply flow much of this water south of Lake Okeechobee without reducing its
phosphorus content. Thus, it is inescapable that additional regional storage of large volumes of water
is an essential part of any solution to reducing damaging discharges from Lake Okeechobee to the St.
Lucie and Caloosahatchee estuaries. Furthermore, moving the stored water south to the Everglades
Protection Area will require substantial additional treatment and improved conveyance.
4. Need for Sustained, Long-term, Holistic Planning and Execution
Historically, policy decisions regarding water supply and flood control planning in south Florida have
been driven by reaction to individual major weather events and water issues. The Herbert Hoover
Dike, for example, was one of the most massive public works projects of its time and was a direct and
rapid response to the 1926 hurricane that destroyed towns south of Lake Okeechobee. Similarly, the
1947-48 floods were a major driver for protecting the east coast with levees that eventually expanded
into forming the WCAs. Drought and fires during the early 1930s, 1940s, and 1960s were similarly
influential in the notion of impounding water for urban and recreational use, and at one point
impoundments were even a design feature of Everglades National Park. Yet storing water in the
WCAs to combat the effects of floods and drought resulted in hydroperiods that were not within the
range of tolerance for plants and animals typical of the Everglades. While each of these historical
responses was perhaps appropriate at the time, they either ignored or led to unintended
consequences, often of equal magnitude to the issue being addressed.
This pattern of reacting to individual events rather than planning for a full range of possible outcomes
over the long term, leads to the commitment of money and resources to individual “fixes.” In an era of
reduced funding for public works projects, this can greatly reduce the ability to respond to other
aspects that may not have been as prominent at that particular time. One of the most significant and
related barriers to Everglades restoration success has been the lack of a sustained commitment of
human and financial resources by both state and federal partners, a topic discussed in greater detail
in Section VI.
The history of south Florida water management has shown that short-term, highly directed solutions
often create other problems of large magnitude, either because of premeditated tradeoffs, or due to
unintended consequences. Solutions to the issues of damaging releases to the St. Lucie and
Caloosahatchee estuaries and insufficient flows south of Lake Okeechobee to the Everglades
18
Protection Area are strongly connected to a complex set of problems and competing stakeholders
throughout the region. Thus regional planning with long time horizons, and sustained, long term
execution are key elements needed to achieve long term resolution to the current issues.
5. Legal Context
An unavoidable reality of any action designed to meet the objective of reducing discharge to the
estuaries and moving additional water south to the EPA is that it must take place within the context of
a complex set of statutes, regulations, consent decrees, and other legally binding agreements. An
array of federal and state statutes and regulations, as well as compacts with Tribes and local
government regulations govern activities that occur within the greater Everglades ecosystem. In
addition, decades of federal and state litigation under a variety of statutes and regulatory schemes
have resulted in a number of court opinions, court orders, and consent decrees that impose further
legal constraints on activities in the Everglades. Although the legal issues surrounding Everglades
restoration are far too numerous and complex to be covered fully in this report, a discussion of several
key legal constraints is warranted. These legal constraints relate to a few key issues, including public
health and safety, water supply, flood protection, water quality, endangered species, and the right to
use water.
Issued Addressed by C&SF authorized purposes
Much of what governs activities in the Everglades is set forth in a number of federal laws authorizing
C&SF projects, which articulate prescribed “authorized purposes” for each project. These purposes,
together with statutory requirements imposing restriction on the local sponsor of the projects set forth
certain limitations on activities that can take place. The specific authorized purposes for the public
laws authorizing projects within the C&SF dating back to the 1930s include: flood control (Public Laws
71-250, 80-858, 87-874 and 90-483); navigation (Public Laws 71-520 and 80-858); water supply
(including for agricultural, and municipal, and industrial uses) (Public Laws 80-858, 90-843 and 87-
874); preservation of fish and wildlife (Public Laws 80-858, 85-624, 90-843 and 930205); drainage
and water control (Public Laws 80-858, 87-874 and 90-843); preservation of the Everglades National
Park (Public Laws 90-843 and 101-229); water supply for the Everglades National Park (Public Laws
80-858, 90-843, 91-282, 98-181, 99-190, 101-229 and 100-676); recreation (Public Laws 90-843 and
78-534); protection of water quality (Public Laws 90-843, 92-500 and 95-217); prevention of saltwater
intrusion (Public Laws 80-858, 90-843 and 87-874); and groundwater recharge (Public Laws 80-858
and 87-874). Projects authorized by Congress for these specific purposes are carried out and
maintained in a manner that achieves the specified purposes. For example, a project that was
authorized for flood control purposes would not be modified in a manner that would subordinate flood
control to recreational purposes. These historic Congressional authorizations, along with more recent
federal activities such as Water Resources Development Act of 2000 (WRDA 2000) (Pub. Law 106-
541) and the adoption of the LORS 2008 regulation schedule, clearly place protection of public health
19
and safety from events such as dam failure or flooding as paramount. Through a series of these
authorizations dating back to 1930, Congress sought to accomplish flood control by authorizing the
construction of the large canal systems to discharge excess water to the St. Lucie and
Caloosahatchee estuaries. Likewise, protection of water supply ranks as a high priority in many of
these authorizations.
Beyond the authorized purposes set forth in federal law, state law also includes similar purposes. For
example, Section 373.1501, F.S., which authorizes SFWMD as the local sponsor of the C&SF project,
provides that, in its role as local sponsor, SFWMD shall, among other things, “consider all applicable
water resource issues, including water supply, water quality, flood protection, threatened and
endangered species, and other natural systems and habitat needs”. Section 373.1502, F.S., requires
that permits issued for component projects comply with water quality standards; that discharges from
project components not pose a serious danger to public health, safety or welfare; and that any
impacts to wetlands or threatened or endangered species be avoided, minimized, and mitigated.
This multitude of “purposes” articulated by both federal and state law for various aspects of
Everglades Restoration can, in some cases, create conflicts and challenges. Federal and State
agencies responsible for carrying out aspects of Everglades Restoration must continually work within
this framework to try to find solutions that meet a wide range of objectives that can be difficult to
reconcile.
Rights of Existing Legal Users of Water
State law rather than federal law generally governs the use of water. In Florida, water use is regulated
by the water management districts pursuant to chapter 373 of Florida Statutes (F.S.). Section
373.219, F.S. authorizes the water management districts to issue permits for the consumptive use of
water. Other than for the domestic consumption of water by individual users, permits are required for
all uses of water. Once a permit is issued, the permittee maintains the right to use the amount of
water authorized by permit under the conditions set forth in the permit for the duration of the permit.
Permits are typically issued with 20-year durations. Water management districts may not issue
permits that will interfere with any presently existing legal user of water. (373.223(1)(b), F.S.). In other
words, the districts are prohibited from issuing new permits that would reduce the amount of
previously allocated water available to existing permit holders. In the context of Everglades
restoration, the rights of existing legal users of water through water management district consumptive
use permits must be recognized. Restoration projects cannot interfere with these legal rights.
Accordingly, in some situations existing consumptive use permits may impose constraints on
restoration projects. In addition to these protections of existing legal users set forth in the general
water use permitting provisions, Florida law expressly addressing Everglades-related matters further
directs SFWMD to “provide reasonable assurances that the quantity of water available to existing
legal users shall not be diminished by implementation of [C&SF] project components so as to
20
adversely impact existing legal users, that existing level of service for flood protection will not be
diminished . . . and that water management practices will continue to adapt to meet the needs of the
restored natural environment.” (373.1501(5)(d), F.S.
The requirement that CERP projects, in particular, protect existing legal users of water and provide
flood protection is further codified in WRDA 2000. Section 601(h)(1) of this Act states that the
overarching objective of CERP is the "restoration, preservation, and protection of the South Florida
ecosystem while providing for other water-related needs of the region, including water supply and
flood protection." Section 601(h)(5) provides the savings clause which makes clear that neither the
USACE nor the local sponsor, SFWMD, should eliminate or transfer existing legal sources of water,
including those for agricultural or urban water supply, allocation or entitlement to the Seminole Indian
tribe of Florida, the Miccosukee Indian tribe of Florida, water supply for the Everglades National Park,
or water supply for fish and wildlife. Also included in the savings clause is a requirement that
implementation of CERP shall not reduce levels of service for flood protection that were in existence
on the date of enactment of WRDA 2000 and in accordance with applicable law. Finally, the savings
clause states that nothing in the Everglades portion of WRDA 2000 amends, alters, prevents, or
otherwise abrogates rights of the Seminole Indian Tribe under the compact among the tribe, the state,
and the SFWMD.
Water Quality
Water Quality Issues in the EPA

Currently, the majority of the legal constraints on moving water south into the EPA revolve around
water quality concerns. Although the primary focus of Everglades restoration is to “get the water right”
by restoring the hydrology of the system, water quantity and water quality are inextricably linked and
any effort to address water quantity concerns cannot occur in isolation from water quality concerns.
One of the significant constraints to moving water out of Lake Okeechobee to the south is the need to
ensure that water discharge to the Everglades Protection Area complies with federal and state water
quality requirements. In particular, numeric phosphorus standards have been established for the EPA
and for the ENP must be met. Currently, water discharged from Lake Okeechobee through STAs into
WCAs is not in compliance with water quality standards due to high phosphorus levels (see Section
IV). The history of Everglades issues related to compliance with water quality standards has been the
subject of two significant and ongoing lawsuits. Both of these cases make it clear that discharging
water into the Everglades Protection Area in a way that does not comply with U.S. Environmental
Protection Agency (USEPA)-approved state water quality standards is considered to be a violation of
federal law.
The first of the two significant water quality-related federal cases began in 1988 when the U.S.
Attorney for the Southern District of Florida filed a lawsuit against the State of Florida and the SFWMD
alleging that Florida’s lack of enforcement of water quality laws and discharges through state-
21
controlled structures threatened the water quality of the Everglades National Park and the
Loxahatchee National Wildlife Refuge (United States v. South Florida Water Management District, No.
88-1886S.D. Fla.). In 1991 Florida conceded liability and settled the case via a consent decree, which
was approved by the court in 1992. Under the terms of the consent decree, the State of Florida
agreed to take actions to ensure all discharges to the Park and the Refuge met the long-term
phosphorus levels set forth in the consent decree by 2002. To accomplish this, the state committed to
build and operate a minimum of 32,600 acres of Stormwater Treatment Areas (STAs) and to establish
a regulatory program that would require farmers in the EAA to implement Best Management Practice
(BMPs) to reduce nutrient waste from the farms. The parties continued to actively pursue aspects of
this litigation, which ultimately resulted in a modified consent decree which, among other things,
extended the deadline for compliance with long-term phosphorus standards from 2002 to 2006. The
modified consent decree remains in effect and under the jurisdiction of the court. Pursuant to
Appendix A of the consent decree, water entering the Everglades National Park is governed by a
prescribed compliance methodology. Failure to comply with the long-term phosphorus levels as
determined by the Appendix A compliance methodology is considered a violation.
The second of the two significant federal lawsuits was initiated when the Miccosukee tribe petitioned
the USEPA to review the Everglades Forever Act (EFA) as a change in Florida water quality
standards. Under the federal Clean Water Act (CWA) section 303(c), EPA is required to review and
approve state water quality standards and changes to the standards. (33 U.S.C. § 1313(c)). After a
series of USEPA decisions that were challenged and the resulting court decisions, USEPA ultimately
reviewed the EFA and found it was in compliance with the CWA. Subsequently, in 2003, the EFA was
amended, which revived the lawsuit. Finally, in an April 14, 2010 Order, Judge Gold ruled that
USEPA’s 2009 “Determination”, that the EFA complied with the CWA, failed to address a prior ruling
of the court and directed USEPA and FDEP to take certain steps to comply with their mandatory
duties under the CWA. Ultimately, the lawsuit resulted in USEPA issuing a document known as the
September 2, 2010 “Amended Determination.” In the Amended Determination, USEPA stated that it
was “notifying FDEP that the narrative and numeric nutrient criteria in the State’s water quality
standards are not being met for the Everglades Protection Area.” The Amended Determination was
intended to provide an enforceable plan for ensuring that the water quality entering the Everglades
Protection Area (EPA) from the Everglades Agricultural Area (EAA) and the C-139 Basin complied
with the narrative and numeric phosphorus criteria, which were already in place for the EPA.
The Amended Determination specifically addressed each of the directives ordered by Judge Gold.
These actions include: (1) revisions to USEPA’s 2009 Determination; (2) directions to Florida for
correcting deficiencies in both Florida’s Phosphorus Rule and the Amended Everglades Forever Act;
(3) provisions for the “manner and method for obtaining enforceable WQBEL within time certain”; (4)
requirements to measure and submit annual reports on cumulative impacts until water quality
standards are attained; (5) directions to Florida to conform all National Pollutant Discharge Elimination
System (NPDES) permits under the CWA and EFA permits pursuant to both the Court’s 2008 order
22
and the 2010 order by eliminating all nonconforming language and by including the WQBEL
presented in the Amended Determination; (6) establishment of an “enforceable framework for
ensuring compliance with the CWA and Applicable Regulations.”
The Amended Determination, for the first time, established a Water Quality-Based Effluent Limitation
(WQBEL) that was required to be included in all permits for discharges from STAs. This WQBEL is
intended to ensure that water leaving the STAs is of high enough quality to ensure compliance with
narrative and numeric nutrient criteria. The WQBEL provided that total P concentrations in the
discharge from the STAs may not exceed either: 10 ppb as an annual geometric mean in more than
two consecutive years; or 18 ppb as an annual flow-weighted mean. USEPA maintains that
“[c]ompliance with both parts of the WQBEL is necessary to assure that the STA discharges will not
cause an exceedance of the long-term criterion of 10 ppb.” The Amended Determination also provides
detailed instructions to the State of Florida on how to meet the WQBEL, including specific milestones
that must be met. The Amended Determination also provided an opportunity for the state to develop
an alternative proposal for achieving water quality standards in the EPA. Ultimately, the state devised
its own plan, known as “Restoration Strategies” to ensure compliance with water quality standards. In
this plan the state committed to several projects that will create more than 6,500 acres of new STAs
and creation of Flow Equalization Basins (FEBs), which will provide 112,000 acre-ft of additional water
storage. The plan also includes additional source control. USEPA approved this plan as well as a
revised WQBEL that requires that the flow-weighted mean (FWM) Total Phosphorus (TP)
concentrations at STA discharge points not exceed (1) an annual FWM of 13 μg/L in more than three
out of five years and (2) an annual FWM of 19 μg/L in any one year, as an alternative to the plan and
WQBEL set forth in the Amended Determination.
In 2012, FDEP issued both NPDES discharge permits and EFA watershed permits that incorporate
this WQBEL. Water leaving the STAs and entering the WCAs must meet WQBEL set forth in these
permits. Because WQBEL's are not expected to be met until the completion of Restoration Strategies,
the SFWMD and the FDEP have entered consent orders in which the FDEP has articulated its finding
that it is clearly in the public interest to exercise its enforcement discretion to allow the continued
operation of STAs while the corrective actions required by the consent order are implemented.
Water Quality Issues in the St. Lucie and Caloosahatchee Estuaries
Unlike the discharges to the EPA, discharges to the St. Lucie and Caloosahatchee estuaries are not
subject to strict quantitative effluent limitations. Discharges to the EPA require both NPDES and EFA
permits, which, as described above, both contain a numeric WQBEL for phosphorus. Discharges from
Lake Okeechobee to the aforementioned estuaries are not subject to NPDES or EFA permitting
requirements and consequently, legally-binding numeric effluent limitations do not apply. State
adopted and USEPA-approved water quality standards and Total Maximum Daily Loads (TMDLs),
however, have been established for both the St. Lucie and Caloosahatchee estuaries. A TMDL
represents, in essence, the amount of a particular pollutant that a particular water body can assimilate
23
without resulting in a violation of a water quality standard. TMDLs are the means by which water
quality criteria can be translated into water-quality based effluent limitations in NPDES permits, or
other types of pollution limitations under state regulatory programs.
The State of Florida implements TMDLs through its Basin Management Action Plans (BMAPs), which
serve as "blueprints" for restoring impaired waters by reducing pollutant loadings. (403.067(7)(a),
F.S.). BMAPs contain a comprehensive set of strategies, some regulatory and some nonregulatory,
designed to meet TMDLs. Regulatory strategies include providing WQBELs in NPDES or other
pollutant discharge permits. For sources of pollutant loading that are not subject to discharge
permitting requirements, such as for the lake discharges to the St. Lucie and Caloosahatchee
estuaries, BMAPs include non-regulatory strategies, such as urban and agricultural best management
practices and conservation programs. These broad-based plans are developed with local
stakeholders—they rely on local input and local commitment—and they are adopted by Secretarial
Order to be enforceable. BMAPs have been established for both the St. Lucie and Caloosahatchee
estuaries. In addition, a BMAP has been established for Lake Okeechobee to bring lake water into
compliance with the established TMDL for phosphorus.
The Endangered Species Act and Migratory Bird Treaty Act
Another federal law that has significantly affected Everglades restoration is the Endangered Species
Act (ESA) (16 U.S.C. §§ 1531 to 1544). The ESA is the primary federal statute governing activities
that may affect threatened or endangered species. It is administered and enforced by two separate
agencies. The United States Fish and Wildlife Service (FWS) within the Department of the Interior
implements the ESA with regard to freshwater and terrestrial species. The National Marine Fisheries
Service (NMFS) within the Department of Commerce implements the ESA with regard to marine and
anadromous species. These two agencies, referred to collectively as “the Services,” are responsible
for listing species as endangered or threatened and implementing regulations to protect the listed
species and their habitats. Two substantial protections are afforded species listed pursuant to the
ESA. The first, set forth in Section 9 of the ESA, prohibits the “taking” of listed species (16 U.S.C.
§1538). The statute defines the term “take” broadly to include to “harass, harm, pursue, hunt, shoot,
wound, kill, trap, capture, or collect or attempt to engage in any such conduct” 16 U.S.C. § 1532(19).
The Services have further defined the term “harm” to include acts that involve significant habitat
modification or degradation where it actually kills or injures wildlife by significantly impairing essential
behavior patterns, including breeding, feeding, or sheltering. This broad interpretation has been
upheld by the U.S. Supreme Court. Moreover, additional habitat protection is afforded where the
Services have designated “critical habitat” for a particular listed species. The ESA authorizes the
Services to assess penalties for unauthorized “takes” of listed species and authorizes courts to award
injunctive relief to prevent the takes from occurring or continuing. Moreover, the ESA contains a
“citizen suit” provision, which authorizes citizens to bring suit and act as private attorneys general to
enforce the law under where the Services have failed to do so. Because the take prohibition applies to
24
“any person,” any of the activities related to Everglades restoration that are carried out by the State or
the Federal government must ensure that unauthorized takes do not occur. Section 10 of the ESA
authorizes the Services to issue permits to authorize specific “takes” of a listed species, if the “taking
is incidental to, and not the purpose of, the carrying out of an otherwise lawful activity,” and “will not
appreciably reduce the likelihood of the survival and recovery of the species in the wild.” A permit
applicant seeking an incidental take permit must develop a “habitat conservation plan” that minimizes
and mitigates impact of the taking to the maximum extent practicable.
The other significant regulatory program under the ESA is the consultation requirement set forth in
Section 7(a)(2) of the Act (16 U.S.C. § 1536). Unlike the “take” prohibition, which applies to ”any
person”, the consultation requirement applies only to federal agencies. This section requires that, prior
to engaging in any federal agency action that “may affect” listed species, the federal agency must
consult with the Services to “insure that any action authorized, funded, or carried out by such agency
is not likely to jeopardize the continued existence of any endangered species or threatened species or
result in the destruction or adverse modification of [critical habitat] of such species.” The phrase
“jeopardize the continued existence of a listed species” includes actions that can reasonably be
expected, directly or indirectly, to appreciably reduce the likelihood of both the survival and recovery
of a listed species in the wild by reducing the reproduction, numbers, or distribution of that species.
Formal consultation can be avoided only where the agency determines, with written concurrence of
the Services, that the proposed action is “not likely to adversely affect” listed species. The purpose of
the consultation process is for the Services to determine whether the federal agency action is likely to
jeopardize the continued existence of listed species. The consultation process culminates in the
issuance of a Biological Opinion (BiOp). If a jeopardy finding is made, the BiOp must include
“reasonable and prudent alternatives” (RPAs) that if implemented would avoid jeopardy. Although the
action agency has discretion to choose whether to implement the RPAs, if the Agency’s action results
in a take, the Agency will be liable under section 9, unless such a take is provided for by an incidental
take statement (ITS) in the BiOp. An ITS describes actions that will not be considered a prohibited
take and which sets forth “reasonable and prudent measures” which must be complied with to be
covered by the ITS (40 C.F.R. Part 402). Finally, Section 7(a)(1) of the ESA imposes an obligation on
federal agencies to use their existing authorities to conserve endangered and threatened species.
Because the C&SF and CERP are federal projects, the USACE is subject the consultation
requirement of Section 7of the ESA.
The Migratory Bird Treaty Act (MBTA) implements four international treaties that are aimed at
protecting migratory birds (16 U.S.C. § 703). The scope of the MBTA is quite broad and covers almost
all native North American birds. Some, but not all, migratory birds covered by the MBTA are also a
listed species under the ESA and, thus, both Acts would apply to those species. Many of the
endangered species in the everglades are birds that are also protected by the MBTA. As with the
ESA, the MBTA prohibits “takes” of covered species. Although the MBTA does not define the term
25
“take,” regulations define it to mean “pursue, hunt, shoot, wound, kill, trap, capture, or collect, or
attempt” any of the foregoing (50 C.F.R. 10.12). Although not well-defined, the MBTA’s definition of
“take” appears to be narrower than the definition under the ESA, which, as described above, may
include significant habitat modification or degradation where it actually kills or injuries wildlife. Thus,
although some general habitat changes may not be violations of the MBTA, actions such as
construction of projects or flooding areas in a way that actually kill birds could be violations of the
MBTA.
Because of its subtropical nature and geographical location, south Florida is home to a large number
of endangered species, many of which are highly dependent upon the ecological conditions that were
characteristic of an un-drained south Florida landscape. At least four of these are of primary concern
in considering hydrological restoration and the routing of water within the EPA – Cape Sable Seaside
Sparrow (Ammodramus maritimus sociabilis), Wood Stork (Mycteria americana), Snail Kite
(Rostrhamus sociabilis) and American Crocodile (Crocodylus acutus). CERP was designed in part to
improve conditions for many of these species through restoration of habitat quality, largely through
hydrological change. However, the road to achieving many of those restored conditions involves
massive construction and temporary conditions that may temporarily negatively impact some species.
In addition, the Everglades ecosystem has been reduced in area by approximately half and thus,
habitat available for endangered species has been dramatically reduced. Moreover, many of the most
imperiled species use habitat that is artificial, and thus, may not contain all of the features and buffers
that were typical of historical natural habitat.
The ESA is designed to err on the side of protection of species, and it becomes most protective in the
case of species with very low population size. The global population of Cape Sable Seaside Sparrows
exists entirely within ENP and may consist of fewer than 3,000 individuals, and many fewer breeding
individuals. The Cape Sable Seaside Sparrow requires relatively dry conditions, and it now nests in
parts of Everglades National Park that may be over-drained relative to historical conditions in order to
protect those nests. Restoring hydrological conditions in the park could therefore present a direct
threat to one or more of the current sub-populations of this bird, unless the population moves to
another location.
Actions in one part of the highly compartmentalized south Florida ecosystem strongly affect conditions
in others. The ability to keep the Cape Sable Seaside Sparrows habitat dry enough during the high
water periods of the mid-1990s, for example, required deep, long-term storage of water in WCA 3
(see Figure I-1 for WCA 3 location). Long-term, deep conditions in WCA 3 strongly affects the viability
of tree islands there, many of which are used by the endangered Snail Kite. Both conditions triggered
lawsuits from different parties using the ESA as a tool. Especially with populations that are already
critically low, the ESA is sensitive to even temporary conditions that may impair the species in
question.
26
USACE has consulted with the FWS on numerous projects related to Everglades restoration dating
back to 1983 and a number of BiOps have been issued. One of the most significant BiOps was the
2010 Biological Opinion for Everglades Restoration Transition Plan (ERTP). The ERTP was designed
to provide a flexible multispecies approach to water management and operations in WCA-3 by
balancing the water needs of multiple species. The ERTP BiOp, among other things, evaluated effects
of WCA-3A operation and regulatory releases on, and set forth reasonable and prudent measures
considered necessary and appropriate to minimize the take of, the Cape Sable Seaside Sparrow, the
Everglades Snail Kite and the Wood Stork. Most recently, in March of 2014, the FWS issued
“Programmatic Biological Opinion and Select Concurrence to the Central Everglades Planning Project
on Effects to Threatened or Endanger Species and Critical Habitat”. This document sets forth
reasonable and prudent measures (RPMs) that the FWS considers necessary and appropriate to
minimize the take of the Cape Sable seaside sparrow, the Everglades Snail Kite, the Wood Stork, and
the Eastern Indigo Snake (Drymarchon couperi). The RPMs in these and other BiOps influence
whether and the manner under which restoration projects and operations are carried out.
Summary
The ability to change water management in substantial ways in south Florida is constrained by a
number of laws that protect the rights of existing legal users of water, including urban populations,
native American tribes, agriculture, and Federal land holders, as well as by laws protecting
endangered species, migratory birds and water quality. Large-scale changes in the volume of flow
south of Lake Okeechobee affects many of these users, and any flows south of the lake toward the
Everglades Protection Area are particularly constrained by the need to meet stringent water quality
standards. Thus, the ability to ameliorate negative effects of freshwater releases to the St. Lucie and
Caloosahatchee estuaries is strongly linked to the need to clean that water to meet legal requirements
and to avoid conflicts with existing users.
27
III. Existing State and Federal Plans to Reduce High Volume Freshwater
Flows to the Estuaries and Move More Water South from Lake
Okeechobee to the Southern Everglades
1. Introduction
As discussed in Sections I and II, reducing the frequency and duration of damaging freshwater
discharge to the St. Lucie and Caloosahatchee estuaries and increasing the flow of water south is a
complicated task because 1) Lake Okeechobee inflow capacity currently exceeds outflow capacity by
as much as 4-6 fold depending on hydrologic conditions, and 2) the capacity of outflow canals and
structures to carry water south of the Lake is currently much smaller than the capacity of outflow
canals and structures to carry water east and west to the St. Lucie and Caloosahatchee estuaries,
respectively. Figure III-1 presents the distribution of flows released from Lake Okeechobee for the
May 1997-April 2014 time period, for Water Year (WY) 2014 (May 2013-April 2014), and for the WY
2014 wet season (June 2013-October 2013). These figures illustrate that, over the long-term, annual
average east/west releases to the estuaries represented 66% of the Lake Okeechobee outflow (45%
to Caloosahatchee and 21% to St. Lucie), while southerly releases to the EAA, L-8 basin and WCAs
represented 34% of the Lake Okeechobee outflow (Figure III-1a). For the extremely wet WY 2014 the
annual percentages of east/west versus southerly flows did not change appreciably (Figure III-b),
however, during the WY 2014 wet season the estuaries received 87% of the outflow from Lake
Okeechobee, while only 13% went south. It should be noted that these releases include both water
supply and regulatory releases.
WY 1998-2014 Annual Average: WY 2014 Annual Total: WY 2014 Wet Season:

1.52 maf 2.44 maf 1.55 maf
21%
18%
50% 60% 27%
45% 32%
34%
13%
To Caloosahatchee To St. Lucie To EAA and L8 To Caloosahatchee To St. Lucie To EAA and L8 To Caloosahatchee To St. Lucie To EAA and L8
Figure III-1 Distribution of Lake Okeechobee Releases in million acre-ft (maf) for a) Long-term annual average for May
1997-April 2014, b) Annual total for WY 2014 (May 2013-April 2014), c) Wet season totals for WY 2014 (June 2013-
Ocotober 2014). Data from SFWMD.
Compounding this problem of high releases of freshwater from Lake Okeechobee, both the St. Lucie
and the Caloosahatchee basins contribute significant additional local runoff to their estuaries (Figure
III-2). For the St. Lucie the long-term annual average (WY1997-WY2014) contribution of local basin
runoff to the total freshwater inflows was 77%, while the contribution during WY2014 was 70%. For
28
the Caloosahatchee the long-term average (WY1997-WY2014) contribution of the local basin runoff to
the total freshwater inflows was 68%, while the contribution during WY2014 was 62%. For both basins
local runoff is a larger contributor of freshwater inflows than Lake Okeechobee. Thus, to reduce
damaging high freshwater discharge to the estuaries, inflows from both Lake Okeechobee and the
local basins must be reduced.
Total Inflow Basin Contribution LO Contribution

3.5
Annual Discharge (maf)
3
2.5
2 62%
1.5
70% 68%
1 77%
0.5
0
WY1997- WY2014 WY1997- WY2014
WY2014 WY2014
St Lucie Caloosahatchee
Figure III-2 Total Freshwater Inflow, Lake Okeechobee Contribution and Local Basin Contribution to the St. Lucie River
and Caloosahatchee River Basins in million acre-ft (maf). Data from SFWMD (SFWMD, 2015).
Freshwater inflows generate direct and indirect effects in the Caloosahatchee and St. Lucie estuaries
(Barnes, 2005; Sime, 2005). Low salinity stresses many biotic components of these estuarine
systems. For example, prolonged exposure to low salinity increases mortality, disease and loss of
recruits for oysters, which can lead to degraded reefs and loss of infauna, epifauna and fish
associated with these reefs (Tolley et al., 2006; Parker et al., 2013). Lower salinity also increases
physiological stress on seagrasses, which represents a second, key structural habitat (Buzzelli et al.,
2012). This direct stress combines with increased shading, an indirect result of turbidity transported in
freshwater inflows, phytoplankton blooms transported to the estuaries or blooms stimulated by
increased nutrient loads, to lead to loss of seagrasses and subsequent changes in invertebrate and
fish assemblages (Gilmore, 1995; Millie et al., 2004; Buzzelli et al., 2012; Phlips et al., 2012; Wang et
al., 2012). Freshwater inflows also have transported toxic microalgae into the St. Lucie estuary, with a
substantial bloom of Microcystis aeruginosa being recorded in 2005 (Phlips et al., 2012). To reduce
these impacts of freshwater inflows on the estuaries, performance criteria have been established
which set limits on the frequency and magnitude of high monthly discharges to the estuaries.
The frequency, magnitude and duration of freshwater discharges to the St. Lucie and Caloosahatchee
estuaries that exceed established flow targets between 1994 and 2014 are shown in Figures III-3 and
29
III-4. In the last 20 years, total flows to the St. Lucie estuary have exceeded high mean monthly flow
targets (occurrences between 2000 cfs and 3000 cfs) 10% of the time and very high monthly mean
flow targets (>3000cfs ) 12% of the time. Restoration targets, on the other hand, are 6.3% and 1.4%,
respectively (SFWMD, 2009a; Table III-1). Similarly, in the last 20 years flows to the Caloosahatchee
estuary have exceeded high targets (>2800cfs) 28% of the time and very high targets (>4500cfs) 15%
of the time versus restoration targets 0.7% and 0%, respectively (SFWMD, 2009b, Table III-2).
Damaging high discharges typically occur in June through October when significant rainfall in the
Kissimmee River Basin results in large inflows to Lake Okeechobee and the structural, ecological and
legal constraints discussed in Section II limit the amount of water that can be discharged south of the
Lake. Most recently this occurred in the summer of 2013 (WY 2014) when Lake Okeechobee received
2.4 million acre-ft of inflow during the months of June through October. This, together with high rainfall
in the local basins, resulted in 5 consecutive months of damaging discharges to the Caloosahatchee
estuary (2.1 million acre-ft, of which 45% was from Lake Okeechobee and 55% was from local basin
runoff) and four consecutive months of damaging discharges to the St. Lucie estuary (1.0 million acre-
ft, of which 37% was from Lake Okeechobee and 63% was from local basin runoff).
Low flows are also a concern in the Caloosahatchee basin. In the last 20 years flows to the
Caloosahatchee estuary have fallen below the low flow target (450 cfs) 28% of the time versus a
restoration target of 0%. Low flows typically occur during the November to May dry season when
rainfall is reduced and agricultural and urban water demands compete with the Caloosahatchee
estuary for releases from Lake Okeechobee. Low flows currently are less of a concern in the St. Lucie
estuary.
30
Total LO inflow Total Flow SLE SLE high SLE very high
1400000
1200000
1000000
Monthly Discharge (acre-ft)
800000
600000
400000
200000
0
Nov-94
Nov-95
Nov-96
Nov-97
May-98
Nov-98
Nov-99
Nov-00
Nov-01
Nov-02
Nov-03
Nov-04
Nov-05
May-06
Nov-06
Nov-07
Nov-08
Nov-09
Nov-10
Nov-11
Nov-12
Nov-13
May-14
May-94
May-95
May-96
May-97
May-99
May-00
May-01
May-02
May-03
May-04
May-05
May-07
May-08
May-09
May-10
May-11
May-12
May-13
Figure III- 3: Total monthly flows into Lake Okeechobee and discharges to the St. Lucie estuary. Dashed horizontal line indicates damaging high discharges and
solid horizontal line indicates damaging very high discharges to the estuaries. The WSE Lake Okeechobee regulation schedule was in effect from July 2000 to
March 2008. The LORS 2008 regulation schedule has been in effect from April 2008 to present. Data from SFWMD.
31
Total LO inflow Total Flow CE CE high CE very high CE low
1400000
1200000
1000000
Monthly Discharge (acre-ft)
800000
600000
400000
200000
0
Nov-94
Nov-95
Nov-96
Nov-97
May-98
Nov-98
Nov-99
Nov-00
Nov-01
Nov-02
Nov-03
Nov-04
Nov-05
May-06
Nov-06
Nov-07
Nov-08
Nov-09
Nov-10
Nov-11
Nov-12
Nov-13
May-14
May-94
May-95
May-96
May-97
May-99
May-00
May-01
May-02
May-03
May-04
May-05
May-07
May-08
May-09
May-10
May-11
May-12
May-13
Figure III-4: Total flows into Lake Okeechobee and discharges to the Caloosahatchee estuary. Dashed horizontal line indicates damaging high discharges and
solid horizontal line indicates damaging very high discharges to the estuaries. Dotted horizontal line indicates damaging low discharges. The WSE Lake
Okeechobee regulation schedule was in effect from July 2000 to March 2008. The LORS 2008 regulation schedule has been in effect from April 2008 to present.
Data from SFWMD.
32
2. Planning Efforts
a. Comprehensive Everglades Restoration Plan (CERP)
As discussed in Section II, the overarching objective of the Comprehensive Everglades
Restoration Plan (CERP; USACE, 1999) which resulted from the Central and Southern Florida
(C&SF) Project Comprehensive Review Study (known as the Restudy) was the restoration,
preservation and protection of the south Florida ecosystem while providing for other water
related needs of the region. In particular, the impacts to the Caloosahatchee and St. Lucie
estuaries that the Restudy aimed to mitigate were extreme fluctuations in discharge, i.e. too
much or too little fresh water.
Modeling analyses conducted using the peer-reviewed South Florida Water Management Model
(SFWMM; SFWMD, 2005) during the Restudy demonstrated that large amounts of storage are
needed in the system to reduce damaging high volume discharges of fresh water to the St.
Lucie and Caloosahatchee estuaries, as well as to maintain minimum dry season flows to the
Caloosahatchee estuary and to move more water south from Lake Okeechobee. The Restudy
recommended a comprehensive plan containing over sixty project features to improve the
quality, quantity, timing and distribution of water supply throughout the South Florida ecosystem
and increase water supply for urban and agricultural needs. Specific projects intended to
improve conditions in the estuaries and move more water south of Lake Okeechobee included:
 A 200,000 acre-ft above-ground reservoir in the Kissimmee River region north of Lake
Okeechobee to reduce the amount of runoff entering Lake Okeechobee during the wet
season, shorten the duration of damaging high water levels within the lake, reduce the
frequency of high volume discharges to the Caloosahatchee and St. Lucie estuaries, and
supplement flows to the lake during the dry season.
 A 50,000 acre-ft above ground reservoir and a 20,000 acre-ft stormwater treatment area
in the Taylor Creek/Nubbin Slough Basin to attenuate flows to Lake Okeechobee and
reduce the amount of nutrients flowing to the lake.
 200 Aquifer Storage and Recovery (ASR) wells around Lake Okeechobee to store up to
1.0 billion gallons per day (bgd) of excess lake water (that would either be lost through
discharge to tide or create harmful high water conditions in the lake) and return the
stored water to the lake for use during drought years.
 A 160,000 acre-ft above-ground reservoir, and 44 ASR wells with a total capacity of 220
million gallons per day (mgd), in the Caloosahatchee River C-43 basin to capture basin
runoff and releases from Lake Okeechobee to reduce damaging discharges and provide
water quality benefits to the estuary in the wet season, and to provide environmental
base flows to the estuary and other water supply benefits during the dry season.
33
 A 40,000 acre-ft above-ground reservoir in the St. Lucie C-44 basin and 349,400 acre-ft
of above-ground reservoir storage in the C-23/C-24/C-25 basins to attenuate local basin
runoff and provide water quality benefits to the St. Lucie estuary in the wet season and
provide a supplemental source of water for irrigation and environmental base flows to
the estuary in the dry season.
 A 360,000 acre-ft above-ground reservoir in the Everglades Agricultural (EAA) to provide

storage and reduce damaging flood releases from the EAA to the WCAs, reduce Lake
Okeechobee regulatory releases to the estuaries, meet EAA irrigation and Everglades
water demands, and increase flood protection in the EAA.
b. Northern Everglades and Estuaries Protection Program (NEEPP)

In 2007, outside of the CERP planning process, the State of Florida Legislature initiated the
Northern Everglades and Estuaries Protection Program (NEEPP; Section 373.4595, F.S.). The
goal of NEEPP was to promote a comprehensive, interconnected watershed approach to
protecting Lake Okeechobee and the Caloosahatchee and St. Lucie rivers and estuaries that
specifically addressed both water quality and the quantity, timing and distribution of water to the
natural systems. The program resulted in the development of the Lake Okeechobee Watershed
Construction Project Phase II Technical Plan (SFWMD, 2008), the St. Lucie River Watershed
Protection Plan (SFWMD, 2009a), and the Caloosahatchee River Watershed Protection Plan
(SFWMD, 2009b).
The Lake Okeechobee Watershed Construction Project Phase II Technical Plan (LOP2TP;
SFWMD, 2008) identified construction projects, along with on-site measures intended to prevent
or reduce pollution at its source to achieve water quality targets for the lake. In addition, the plan
included projects for increasing water storage north of Lake Okeechobee to achieve more
desirable lake levels and reduce harmful, high-volume freshwater discharges to the
Caloosahatchee and St. Lucie estuaries. Components of the Lake Okeechobee Plan included:
 Implementing Best Management Practices (BMPs) on more than 1.7 million acres of
farmland.
 Building treatment wetlands to clean water flowing into the lake.
 Using other innovative nutrient control technologies to reduce phosphorus loads from the
watershed.
 Creating between 900,000 and 1.3 million acre-ft of water storage north of the lake
through a combination of above-ground reservoirs, underground storage and alternative
water storage projects on public and private lands (note this estimate included existing
storage and planned state and CERP projects).
34
The St. Lucie River Watershed Protection Plan (SFWMD 2009a) was developed to reduce
nutrient loads to meet Total Maximum Daily Loads (TMDLs) and reduce the frequency and
duration of undesirable salinities in the estuary while also meeting other water related needs
such as water supply and flood protection. Building on the expected results from implementation
of the LOP2TP, and incorporating the authorized CERP Indian River Lagoon - South (IRL-S)
Final Integrated Project Implementation Report projects (USACE, 2004; see description in
Section III.2.a below) the St. Lucie River Watershed Plan included:
 Implementation of BMPs on more than 297,000 acres of agricultural lands and on nearly
84,000 acres of urban lands.
 Construction of approximately 11,800 acres of reservoirs and more than 8,500 acres of
STAs.
 Restoration of approximately 95,000 acres of wetlands and natural areas within the St.
Lucie River watershed.
 Removal of more than 8 million cubic yards of muck sediment from the St. Lucie
Estuary.
 Provision for approximately 200,000 acre-ft of water storage within the St. Lucie River
watershed (in addition to the 900,000 acre-ft per year minimum storage needs identified
in the Lake Okeechobee watershed).
The Caloosahatchee River Watershed Protection Plan (SFWMD, 2009b) identified a

combination of watershed storage and water quality projects needed to help improve the quality,
timing and distribution of water in the natural ecosystem. Working in concert with the expected
results from implementing the LOP2TP, and incorporating the planned CERP C-43 reservoir
(USACE, 2010; see description in Section III.2.b below), the Caloosahatchee River Watershed
Protection Plan included:
 Implementation of BMPs on more than 430,000 acres of agricultural lands and 145,000
acres of urban lands.
 Construction of approximately 36,000 acres of reservoirs and 15,000 acres of STAs and
Water Quality Treatment Areas.
 Restoration of more than 2,000 acres of wetlands within the Caloosahatchee River
Watershed.
 Provision of approximately 400,000 acre-ft of water storage within the Caloosahatchee

River Watershed (in addition to the 900,000 acre-feet minimum storage needs identified
in the Lake Okeechobee Watershed).
35
It should be noted that these plans assumed that all water related needs of the region including
water supply and flood protection must continue to be met. They also assumed Everglades
water deliveries within the constraints of the system with the Acceler8 projects constructed (C-
43 Reservoir, C-44 Reservoir and STA, Broward Water Preserve Areas, Site 1 Reservoir, and
EAA Phase A-1 Reservoir; SFWMD, 2004) and a version of the Modified Water Deliveries to
ENP was operational. However, full restoration of Everglades flows south of Lake Okeechobee
were not provided by these plans.
c. River of Grass (ROG) Planning Process

In 2008, the SFWMD had an option to purchase a large amount of land in the Everglades
Agricultural Area (EAA) from willing sellers, with the possibility to carry out land trades to
achieve a contiguous corridor through the EAA for storage, treatment and conveyance of water
to the Everglades. During this time the volume of storage needed to reduce lake-triggered
damaging discharges to the Caloosahatchee and St. Lucie estuaries and move more water
south of Lake Okeechobee was re-evaluated as part of the River of Grass (ROG) Phase 1
planning process. Screening-level modeling studies were conducted using the peer-reviewed
Reservoir Sizing and Operations Screening model (RESOPS, SFMWD 2009c). These modeling
analyses, conducted for the 41-year (1965-2005) period of record, assumed that the Kissimmee
River Restoration (KRR) was complete, the C-43 and C-44 reservoirs were constructed (for
details see Sections III.2a, b below), and the Lake Okeechobee Regulation Schedule was
optimized using the upper bound from LORS 2008 (for details see Section V.5.a below). Various
alternatives were evaluated with regard to their effectiveness in meeting a set of standard
performance measures including percent reduction in lake-triggered high discharges to the St.
Lucie (> 2000 cfs) and the Caloosahatchee (>2800 cfs) estuaries, and the ability to deliver the
target dry season Everglades demand. The performance maps below (Figures III-5a-c; taken
from SFWMD 2009d) show the percent reduction in lake-triggered high flow discharges to the
estuaries, the percent of the dry season Everglades demand met, and the increase in mean
annual flow to the Everglades for various combinations of storage north and south of Lake
Okeechobee. According to these screening level estimates approximately one million acre-ft
(maf) of storage can achieve a 90% reduction in lake-triggered discharge to the estuaries, meet
90% of the Everglades dry season target, and provide approximately 350,000 additional acre-ft
of annual flow to the Everglades. Furthermore, a variety of north and south of lake storage
configurations can effectively be used to meet these restoration objectives. It is important to
note that this analysis focused on lake-triggered high discharge events, and thus the 1 maf
estimate does not include the additional storage required within the Caloosahatchee and St.
Lucie basins to attenuate local discharges to the estuaries. In addition to evaluating storage
needs the ROG also estimated the additional treatment capacity required for the additional flows
to meet Everglades legal water quality standards. For details on additional plans and projects to
achieve water quality goals see Section IV.
36
.
Figure III-5: Percent Reduction in Lake Triggered High Discharges to the Estuaries (a, top left), Percent Achievement of Dry
Season Everglades Demand Target (b, top right) and Increase in Mean Annual Flows to the Everglades Achieved (c, bottom
left) by alternative storage configurations proposed as part of the River of Grass modeling efforts. (adapted from SFWMD,
2009d). Various options to achieve 90% reduction in performance targets, and the increase in flows to the Everglades that this
represents are noted on the figures.
37
In summary, the CERP, NEEPP and the ROG planning exercises all identified the need for an
extremely large volume of new storage in the Greater Everglades system. The most recent
estimates include:
 approximately 1,000,000 acre-feet of water storage distributed north and south of Lake
Okeechobee.
Additional water quality treatment projects needed to discharge the stored water to the lake,
estuaries and EPA are described in Section IV.
3. Construction Project Status
Each of the planning efforts described above indicated clearly that large volumes of storage are
required to reduce damaging freshwater discharges to the estuaries and move more water
south of Lake Okeechobee. The following section summarizes the status and expected benefits
of current CERP and NEEPP construction projects resulting from the planning efforts described
above that are intended to provide portions of the required storage. The ROG process did not
progress past the planning stage, and did not result in any planned construction projects outside
of CERP.
a. CERP Indian River Lagoon-South (IRL-S) Project

Project Summary: The IRL-S Project is a CERP Project located within Martin and St. Lucie
Counties (Figure III-6). The purpose of the project is to improve surface-water management in
the C-23/C-24, C-25, and C-44 basins for habitat improvement in the St. Lucie River Estuary
and southern portions of the Indian River Lagoon (USACE 2004). Project features include the
construction and operation of four above ground reservoirs to capture water from the C-44, C-
23, C-24, and C-25 canals for increased storage (130,000 acre-ft), the construction and
operation of four STAs (approximately 8,700 acres) to reduce sediment, phosphorus, and
nitrogen to the estuary and lagoon, the restoration of over 90,000 acres of upland and wetland
habitat, the redirection of water from the C-23/24 basin to the north fork of the St. Lucie River to
attenuate freshwater flows to the estuary, and muck removal from the north and south forks of
the St. Lucie River and middle estuary.
Expected Benefits: The IRL-S project is expected to provide significant water-quality

improvement benefits to both the St. Lucie River and Estuary and Indian River Lagoon by
reducing the load of nutrients, pesticides, and suspended materials from basin runoff. The
project is also expected to reduce damaging freshwater runoff from the local basin to the
estuary. Modeling studies using the peer-reviewed SFWMM (SFWMD, 2005) and the Natural
38
Systems Model for the St. Lucie (NSM-IRL) estimated, based on the 1965 to 1995 simulation
period, that the number of occurrences of mean monthly flow discharged to the estuary between
2,000 cfs and 3,000 cfs would be reduced from 7.5% of the time for the 1995 base condition
(which assumes no CERP projects are in place) to 4.3% of the time if the entire IRL-S project,
but no other CERP projects were constructed (Table III-1). Furthermore, modeling results
estimated that the number of occurrences of mean monthly flow above the very high target of
3,000 cfs would be reduced from 3.5% to 1.9% if the entire project were constructed (Table III-
1). The modeled performance of the IRL-S project achieves the restoration targets published by
USACE (2004) and comes very close to achieving the restoration targets published by SFWMD
(2009a).
Status: The IRL-S Project was authorized by the US Congress in 2007. Funding has been
appropriated for the first phase of the project, the C-44 reservoir (approximately 40,000 acre-ft
storage) and an associated STA (approximately 6,300 acre treatment area). To expedite
completion of the C-44 project, in July 2014 the SFWMD Governing Board agreed to construct
the C-44 STA, the pump station and a portion of the system discharge canal and the USACE
agreed to construct the reservoir. The agencies entered into an amended Project Partnership
Agreement in July 2014 to reflect this agreement, and the USACE completed the first
construction contract of the project in the same month. In August 2014, the SFWMD Governing
Board awarded the contract for the construction of a spillway that will serve as the single point
of water movement out of the entire C-44 project. The estimated date of completion for the C-44
project is 2020. Funds have not yet been appropriated for the C-23, C-24, or C-25 reservoirs
and associated STAs, or the habitat restoration projects.
Table III-1: Existing Conditions, Restoration Targets and Expected Benefits IRL-S Project. Percentages indicate
percent of time the monthly flow condition occurs under specific scenarios. Red-colored percentages have not
achieved restoration goals. Green-colored percentages have achieved restoration goals.
Condition St. Lucie

# mean monthly flows between 2000 # mean monthly flows >3000 cfs
and 3000 cfs
Restoration Targets 1965-1995 <18 mo in 31 years3 <10 mo in 31 years3
(USACE, 2004) (4.8%) (2.7%)
Restoration Targets (1970-2005 <21 mo in 36 years1 <6 mo in 36 years1
(SFWMD, 2009a) (6.3%) (1.4%)
Historical Performance 1994-2014 24 mo in 20 years2 29 mo in 20 years2
(10%) (12%)
1995 base condition with no CERP 28 mo in 31 years3 13 mo in 31 years3
projects (7.5%) (3.5%)
Future with complete IRL-S project 16 mo in 31 years3 7 mo in 31 years3
(4.3%) (1.9%)
1
based on simulations from 1970-2005 period of record (Figure 6.5-1, SFWMD 2009a)
2
based on observed flows 1994-2014
3
based on simulations for 1965-1995 period of record (Table 6-6, USACE 2004)
39
Figure III-6 Map showing locations of Greater Everglades Ecosystem Restoration Construction Projects that are
completed (green), under construction (orange) or in Design or Conceptual Phase (blue) (from SFWMD
http://www.sfwmd.gov/restorationprogress).
40
b. CERP Caloosahatchee River (C-43) West Basin Storage Projects
Project Summary: The Caloosahatchee River (C-43) West Basin Storage Reservoir Project is
a CERP project that provides approximately 170,000 acre-ft of above-ground storage in a two-
cell reservoir in Hendry County (Figure III-6). The primary functions of the Caloosahatchee
reservoir are to capture excess basin runoff and discharges from Lake Okeechobee during
periods of high volume flows, and to provide an additional source of water to maintain desirable
salinity in the Caloosahatchee Estuary during periods of low flow (USACE, 2010).
Expected Benefits: Modeling studies using the SFWMM (SFWMD, 2005) and the Freshwater
Caloosahatchee River Basin Model (FCRB) estimated that, based on the 1965 to 2000 time
period, the project will reduce the number of months that flows to the estuary are below the
restoration target of 450 cfs from 40% of the time for a future without project condition (which
assumed that the KRR project was operational, but no CERP projects had been completed) to
28% of time if all CERP projects, including the selected C-43 project, were constructed (Table
III-2). Furthermore, the number of months that flows to the estuary are above mean monthly
high flow target of 2,800 cfs are estimated to be reduced from 9.3% of the time for the future
without project condition to 2.7% if all CERP projects were constructed, and the number of
months that flows to the estuary exceed the very high mean monthly flow target of 4,500 cfs are
estimated to be reduced from 2.7% of the time to 0% (USACE, 2010). The modeled
performance of the C-43 project does not achieve the low flow or high flow restoration targets,
but does achieve the very high restoration target published by SFWMD (2009b, Table III-2).
Status: In June 2014, the federal government authorized this project via the 2014 Water
Resources Reform and Development Act (WRRDA), however, federal funds have not yet been
appropriated. In the 2014 Legislative Session, the State of Florida appropriated $18 million
dollars to design and construct an early start interim project that will provide approximately
2,200 acre-ft of static storage.
41
Table III-2: Existing Conditions, Restoration Targets and Expected Benefits of C-43 Project. Percentages indicate
achieved restoration goals. Green-colored percentages have achieved restoration goals.
Condition Caloosahatchee
# mean monthly flows < 450 # mean monthly flows > # mean monthly flows >
cfs 2800 cfs 4500 cfs
Restoration Targets 0 mo in 36 years1 <3 months in 36 years1 0 mo in 36 years1
(0%) (0.7%) (0%)
Historical Performance 68 mo in 20 years2 68 mo in 20 years2 37 mo in 20 years2
1994-2014 (28%) (28%) (15%)
Future without any CERP 43 mo in 9 years3 10 mo in 9 years3 3 mo in 9 years3
projects simulated for C-43 (40%) (9.3%) (2.7%)
Project (2010)
Future with complete CERP 30 mo in 9 years3 3 mo in 9 years3 0 in 9 years3
including C-43 Project (28%) (2.7%) (0%)
(2010)
1
based on simulations from 1970-2005 period of record (Figures 6.5-1 and 6.5-2, SFWMD 2009b)
2
3
based on simulations for the 1965-2000 period of record, however only 9 years containing 3 wet, 3 dry and 3
normal years out of the 36 year simulation were analyzed (Table 5-10, USACE 2010).
c. CERP Aquifer Storage and Recovery (ASR) Pilot Projects and Regional Study
Project Summary: Aquifer storage and recovery (ASR) is a mechanism for providing inter-
annual water storage underground through injection wells in the wet season to be withdrawn in
subsequent dry seasons for beneficial purposes as surface water. The CERP Restudy (USACE,
1999) called for a total of 333 ASR wells, including 200 located around Lake Okeechobee
intended to store up to 1.0 billion gallons per day (bgd) of excess lake water that would either be
lost through discharge to tide or create harmful high water conditions in the lake. The stored
water is intended to be returned to the lake for use during drought years. Due to the
unprecedented magnitude of ASR implementation proposed in CERP, three ASR pilot projects
and an ASR regional study were proposed to evaluate its technical and regulatory feasibility.
Status: The ASR pilot facilities were used to conduct scientific and engineering studies
regarding the technical performance and regulatory compliance of the ASR concept. Three 5
mgd pilot studies were proposed in the CERP Restudy: the Kissimmee River ASR Pilot, the
Hillsboro ASR Pilot and the Caloosahatchee River (C-43) basin ASR Pilot. ASR facilities
consisting of recharge and recovery pumps, filtration and UV pre-treatment systems, ASR and
monitoring wells, and on-site storage ponds were designed, constructed and operated at the
Kissimmee River and Hillsboro sites between 2007 and 2013. The third pilot site in the
Caloosahatchee River (C-43) basin was abandoned because geotechnical testing and
exploratory well construction indicated the site was not suitable for development of an ASR
storage zone.
42
Results from the two successful pilot facilities indicated that recovery efficiencies were 100% at
the Kissimmee River facility and ranged from 20-40% at the Hillsboro facility where the Floridan
Aquifer water was brackish (USACE, 2013). Operating pressures did not compromise the
overlying Hawthorn confining unit, and no pressure or water quality effects were observed in the
surficial aquifer, at either site. The recharge water quality (primarily color) caused problems with
the UV disinfection system at both sites resulting in frequent detections of coliform bacteria in
the treated water. Thus, a more robust disinfection system will be required in future permanent
facilities. Issues associated with arsenic mobilization in the recovered water resolved
themselves over time during the pilot studies. Furthermore, mercury and methylmercury
concentrations declined during the storage phase of the ASR cycle, and phosphorus
concentrations also declined (e.g., from a mean of 66 g/l +/- 42 g/l in recharged water to a
mean of 10.8g/l +/- 11.6 g/l in recovered water at the Kissimmee River site (USACE, 2013).
Although the operational testing costs at the pilot sites were considered to be high, several
recommendations were made to improve cost-effectiveness in future permanent facilities
(USACE, 2013).
The ASR regional study is an on-going data collection and regional modeling effort to analyze
the effect of the ASR well network on water levels and water quality within the regional aquifer
systems, and on existing water users and surface-water bodies. In the ASR Regional Model
Production Scenario Report (USACE, 2014a), it was suggested that operation of 333 wells in
the South Florida region would likely produce undesirable hydrogeological impacts. The report
recommended a smaller system consisting of approximately 94 ASR wells in the Upper Floridan
Aquifer (with an assumed 70% recovery), 37 ASR wells in the Avon Park Permeable Zone (with
an assumed 30% recovery) and additional 101 wells in the deep Boulder Zone (with an
assumed 0% recovery). Of these wells, 139 were located in the Lake Okeechobee basin (47
Upper Floridan, 31 Avon Park and 61 Boulder Zone), 27 were located in the Caloosahatchee
Basin (9 in the Upper Floridan, 1 in the Avon Park and 17 in the Boulder Zone) and the
remaining 66 were located in basins southeast of Lake Okeechobee. The Final Report for the
ASR regional study is currently under peer review and is scheduled to be completed in 2015.
d. NEEPP Lake Okeechobee Watershed Construction Project Phase II Technical Plan
Plan Summary: The LOP2TP required 900,000 to 1,300,000 acre-ft of storage in the Lake
Okeechobee watershed to manage lake stages in Lake Okeechobee within a desirable range,
and reduce damaging freshwater releases to the Caloosahatchee and St. Lucie estuaries. Plan
elements for capturing and storing water in the Lake Okeechobee watershed included the
following above-ground reservoirs, Aquifer Storage and Recovery (ASR) wells, and Dispersed
Water Management (DWM) projects:
43
 Above-ground Storage Reservoirs: Although the LOP2TP specified that 900,000 to
1,300,000 acre-ft of storage was needed, only three storage reservoirs with a total
storage capacity of approximately 440,800 acre-ft, were identified in the initial
implementation plan (Table III-3, from SFWMD (2008)). While these facilities were
modeled and evaluated as surface water reservoirs, the plan indicated that the storage
benefits derived from these types of projects might also be attained through smaller-
scale DWM on private lands, ASR, surface water reservoirs, or a combination of these.
The plan specified that the appropriate mix of storage would become more defined as
results from ASR pilots and the ASR Regional Study became available.
Table III-3 Location and Capacity of LOP2TP Storage Features (from SFWMD 2008)
Storage ID Sub-watershed Storage Capacity TP load Source Water

(acre-ft) reduction to Lake
(metric tons/yr)
Kissimmee East Lower Kissimmee 200,000 6.5  Receives flows from and
Storage discharge back to the
Kissimmee River
 Stored water can potentially
also be diverted to the TCNS
Sub-watershed for additional
treatment
Kissimmee Lower Kissimmee 161,263 12  Receives flows from and
Storage discharges to Kissimmee
River
 CERP-LOW Project feature
Istokpoga Istokpoga Indian 79,560 7  Receives flows from Lake
Storage Prairies Istokpoga/Indian Prairie and
discharges to Indian Prairie
 CERP-LOW Project feature
 Aquifer Storage and Recovery: To complement surface storage the LOP2TP also
identified four ASR facilities, with a total capacity of 66 mgd (Table III-4, from SFWMD,
2008)). These ASR facilities were expected to be completed during the initial 2008-2010
implementation stage. It was anticipated that additional ASR features would be identified
in the future to help meet the storage goal of the plan after the Lake Okeechobee ASR
pilots and the ASR Regional Study were completed.
 Dispersed Water Management (DWM) Program: The goals and objectives of the DWM
Program are to provide shallow water storage to enhance Lake Okeechobee and
estuary health by reducing discharge volumes, reducing nutrient loads to downstream
receiving waters, and expanding groundwater recharge opportunities. The four main
categories of projects under the DWM Program include storage and retention projects
on private lands, storage and retention projects on public lands, Northern Everglades
Payment for Environmental Services (NE-PES) projects on ranch lands, and Water
44
Farming Payment for Environmental Services (WF-PES) pilot projects on fallow citrus
lands. The LOP2TP included the implementation of approximately 100,000 acre-ft of
alternative water storage projects on public and private lands in the Lake Okeechobee
watershed during the initial 2008-2010 implementation phase.
Table III-4. Location, water storage and TP load reduction for LOP2TP ASR (from SFWMD 2008)
ASR ID Sub-watershed Storage TP load Comments

(acre-ft) reduction
(metric tons/yr)
Kissimmee Pilot Lower Kissimmee 3,780 0.1  One 5 mgd ASR well built,
(5 million gallons per operated and tested between
day (mgd)) 2007 and 2013
Paradise Run Lower Kissimmee 22,950 (50 mgd) 1.4  Maximum pumping capacity of
10-well System up to 50 mgd
Seminole 3,780 0.8  One 5 mgd ASR well system
Brighton (5 mgd) to be located along the C-41
Reservoir ASR Canal on the western edge of
System the Seminole Brighton
Reservation in Glades County
Taylor Creek Taylor 5,400 1.2  Assessment and eventual
ASR Creek/Nubbin (6 mgd) reactivation of the TCNS ASR
Reactivation Slough Sub- system
watershed  One 6 mgd well system is
proposed adjacent to the L-
63N Canal
Expected Benefits: Modeling analyses were conducted using the Northern Everglades
Regional Simulation Model (NERSM) for the 1970 to 2005 time period. The 2005 base condition
(which assumed that no CERP or LOP2TP projects were in place and that Lake Okeechobee
releases to the estuaries and WCAs were based on the Water Supply/Environmental (WSE)
regulation schedule) was compared to the Future Base Condition (which assumed that the KRR
and Acceler8 projects, i.e. C-43 reservoir, C44 reservoir and STA, Broward Water Preserve
Areas, Site 1 Reservoir, Modified Water Deliveries to Everglades National Park, limited version
of Everglades rainfall deliveries, and EAA Phase A-1 Reservoir, were in place) and the Future
with LOP2TP Condition which assumed that, in addition to the Future Base, the LOP2TP was
fully implemented (including 914,000 acres of reservoir storage, 54,000 acres of STAs and 66
mgd of ASR) (Table III-5).
Modeling results indicated that the occurrences of undesirable high flows to the St. Lucie
estuary between 2,000 cfs and 3,000 cfs would be reduced from 8.5% of the time over the 36
year simulation period for 2005 base conditions to 7.6% for the Future with LOP2TP, versus a
restoration target of 4.9% (Table III-5). Furthermore, modeling analyses estimated that
undesirable very high flows greater than 3,000 cfs would be reduced from 6.4% of the time for
45
the 2005 base condition to 4.2% of the time for the Future with LOP2TP versus a restoration
target of 1.4%.
For the Caloosahatchee estuary, modeling analyses estimated that the occurrences of
undesirable high flows above 2,800 cfs would be reduced from 19% of the time over the 36 year
simulation period for 2005 base conditions to 12% of the time for the Future with LOP2TP,
versus a restoration target of 0.7% (Table III-5). Furthermore, modeling analyses estimated that
undesirable very high flows greater than 4,500 cfs would be reduced from 8.6% to 4.2% versus
a restoration target of 0%, and that undesirable low flows less than 450 cfs would be reduced
from 44% to 4.2%, versus a restoration target of 0%. Thus, while the LOP2TP is estimated to
achieve substantial improvement of estuary conditions it does not achieve the restoration
targets published by SFWMD (2009a,b).
Table III-5: existing conditions, restoration targets and expected benefits of NEEPP projects. Percentages indicate
achieved restoration goals.
Condition St. Lucie Caloosahatchee

# mean monthly #mean # mean # mean # mean
flows between monthly flows monthly flows monthly flows monthly flows
2000 and 3000 > 3000 cfs < 450 cfs > 2800 cfs > 4500 cfs
cfs
Restoration Targets <21 mo in 36 <6 mo in 36 0 mo in 36 <3 months in 0 mo in 36
years1 years1 years2 36 years2 years2
(4.9%) (1.4%) (0%) (0.7%) (0%)
Historical Performance 24 mo in 20 29 mo in 20 68 mo in 20 68 mo in 20 37 mo in 20
1994-2014 years3 years3 years3 years3 years3
(10%) (12%) (28%) (28%) (15%)
2005 base condition 37 mo in 36 28 mo in 36 189 mo in 36 80 mo in 36 37 in 36 years2
(with no CERP or years1 years1 years2 years2
NEEPP projects) (8.5%) (6.4%) (44%) (19%) (8.6%)
Future base condition 38 mo in 36 21 mo 36 32 mo in 36 55 mo in 36 25 mo in 36
(including KRR and years4 years4 years4 years4 years4
Acceler8 projects) (8.8%) (4.2%) (4.2%) (12%) (4.2%)
Future with LOP2TP 33 mo in 36 18 mo 36 18 mo in 36 51 mo in 36 18 mo in 36
years4 years4 years4 years4 years4
(7.6%) (4.2%) (4.2%) (12%) (4.2%)
Future with LOP2TP, 25 mo in 36 17 mo in 36 4 mo in 36 44 mo in 36 16 mo in 36
St. Lucie and years1 years1 years2 years2 years2
Caloosahatchee River (5.8%) (3.9%) (0.9%) (10%) (3.7%)
Protection Plans
1
2
3
4
based on simulations from 1970-2005 period of record (Figures 8-6, 8-7and 8-8, SFWMD 2008)
46
Status:
 None of the above-ground reservoirs have been sited, authorized, designed or
constructed. Initial planning and design work on the two reservoirs associated with the
CERP Lake Okeechobee Watershed (LOW) Project began in 2004 with the goal of
having a Project Implementation Report by 2009. However work was suspended in 2010
due to unresolved policy concerns about cost-sharing water quality/stormwater treatment
area features proposed and included in the tentatively selected plan.
 One 5 mgd pilot ASR facility has been designed, constructed and successfully tested in
the Kissimmee River basin. None of the additional planned ASR facilities have been
sited, designed or constructed.
 As of 2014, the total DWM either in operation or under construction in the Lake
Okeechobee watershed was roughly estimated through preliminary modeling studies to
be approximately 60,500 acre-ft (Table III-6, SFWMD (2015)).
Summaries, expected benefits and status of projects to improve water quality in the Lake
Okeechobee watershed are described in Section IV.
Table III-6. Dispersed Water Management Projects in the Lake Okeechobee Watershed (from SFWMD, 2015).
Project Name Category Status Estimated Storage
Benefits (acre-ft/yr)
Buck Island Ranch NE-PES 1 Operational 1,573
Dixie West NE-PES 1 Operational 315
Dixie Ranch NE-PES 1 Operational 856
Lost Oak Ranch NE-PES 1 Operational 374
Triple A Ranch NE-PES 1 Construction 397
Willaway Cattle and NE-PES 1 Operational 229
Sod
XL Ranch NE-PES 1 Operational 887
Blue Head Ranch NE-PES 2 Construction 3,462
West Waterhole Private Lands Operational 5,000
Pasture FRESP
Rafter T Ranch Private Lands Operational 1,145
FRESP
Payne and Son Private Lands WRP Operational 932
Ranch FRESP/WRP
Williamson Cattle Private Lands WRP Operational 150
Company FRESP/WRP
Nicodemus Slough Private Lands Construction 34,000
Avon Park Air Force Public Lands Operational 10,000
Range
Sumica Tract Public Lands Operational 281
Istokpoga Marsh Public Lands Construction 950
Watershed
Improvement District
47
e. NEEPP St. Lucie River Watershed Protection Plan
Plan Summary: Specific elements of the St. Lucie Watershed Protection Plan intended to
provide increased water storage capacity required to reduce damaging discharges to the St.
Lucie estuary and improve low flow conditions included:
 The LOP2TP elements described in Section III.2.d above.

 The regional CERP IRL-S project described in Section III.2.a above.
 The CERP Ten Mile Creek Water Preserve Area Critical Project located in St. Lucie
County at the headwaters of the North Fork of the St. Lucie River along Ten Mile Creek.
This project, as originally designed, consisted of an aboveground reservoir of
approximately 550 acres, designed to store up to 6,000 acre-ft of water. The project also
included a 100-acre STA to treat flows from the reservoir.
 Local alternative DWM projects developed on private, public, and tribal lands to prevent
runoff from reaching the regional drainage system or improve the timing of its delivery.
Expected Benefits: Modeling analyses conducted for the St. Lucie Watershed Protection Plan
using the Northern Everglades Regional Simulation Model (NERSM) for the 1970-2005 period
of record estimated that the proposed plan would reduce the occurrences of undesirable high
flows between 2,000 cfs and 3,000 cfs from 8.5% of the time during the 36 year simulation
period for 2005 base conditions (which assumed that no CERP or LOP2TP projects were in
place and that Lake Okeechobee releases to the estuaries and WCAs were based on the WSE
regulation schedule) to 5.8% of the time for the Future with Plan (which assumed the KRR, all
Acceler8 projects described above, and all elements of both the LOP2TP and the St. Lucie
Watershed plans were fully constructed). Furthermore, modeling analyses estimated that
undesirable very high flows greater than 3,000 cfs would be reduced from 6.4% of the time to
3.9% for the 36 year simulation period. While the St. Lucie Watershed Protection Plan is
estimated to provide some improvement in performance over the LOP2TP it still does not
achieve the performance targets published by SFWMD (2009a).
Status:
 The LOP2TP elements: See Section III.2.d above.
 CERP IRL-S project: See Section III.2.a above.
 Ten Mile Creek Project: Construction of this project was completed by the USACE in
2005, however, it did not perform as designed. In an August 11, 2014 letter, the SFWMD
requested that the USACE “deauthorize the project, terminate the Project Cooperation
Agreements, and extinguish any real estate certifications associated with the project.
Following these steps, the SFWMD will take ownership and responsibility for converting
48
the Ten Mile Creek Project into a functional facility designed to provide necessary
storage and water treatment options.”
 Dispersed Water Management: As of 2014, the DWM either in operation or under

construction in the St. Lucie watershed was roughly estimated through preliminary
modeling studies to be approximately 23,000 acre-ft (Table III-7, SRWMD (2015)). This
includes contributions from the U.S. Department of Agriculture Natural Resources
Conservation Service (USDA NRCS) Wetland Reserve Program (WRP) and other
programs, the Florida Department of Agriculture and Consumer Affairs (FDACS) BMP
program, agricultural landowners, agricultural organizations, non-governmental
organization, and local governments.
Summaries, expected benefits and status of projects to improve water quality in the St. Lucie
watershed are described in Section IV.
Table III-7: SFWMD Dispersed Water Management Projects in the St. Lucie Watershed (from SFWMD, 2015).
Benefits (ac-ft/yr)
Alderman-Deloney Ranch NE-PES 1 Operational 147
Caulkins Citrus Water Farming Operational 6,780
Evans Properties (Alt. E-1) Water Farming Construction 3,635
Spur Land and Cattle/Bull Hammock Water Farming Construction 870
Ranch
Harbour Ridge Private Lands Operational 667
Indiantown Citrus Growers Phase I and II Private Lands Operational 3,550
Adams Ranch Cattle and Citrus Public Lands Operational 190
Operations (ARCCO) (C-23/C-24
Complex)
C-23 Interim Storage (Section D - PC55) Public Lands Operational 110
C-23 Interim Storage (Section C) Public Lands Construction 212
Williamson Ranch/Turnpike Dairy Public Lands WRP Operational 547
Allapattah Parcels A and B - Phase I Public Lands WRP Operational 3,500
Allapattah Parcels A and B - Phase II Public Lands WRP Construction 1,243
Allapattah H Canal Public Lands Operational 1,610
f. NEEPP Caloosahatchee River Watershed Protection Plan

Plan Summary: Specific elements of the Caloosahatchee River Watershed Protection Plan
intended to provide increased water storage capacity to reduce damaging high discharges to
the Caloosahatchee estuary and improve low flow conditions included:
 The LOP2TP described above.
 CERP C-43 basin storage reservoir project described above.
 Additional Reservoir Storage Sites
o C-43 Distributed Reservoirs
49
o Harns Marsh Improvements
o West Lake Hicpochee Project
o Yellowtail Structure Construction
o Hendry County Storage
o Hendry Extension Canal Widening
o East Caloosahatchee Storage
o Additional Caloosahatchee Storage (beyond the C-43 reservoir)
 Aquifer Storage and Recovery sites including the Cape Coral and Lee County Well Field
site projects.
 DWM in the Barron Water Control District, Recyclable Water Containment Areas, and
Recycled Water Containment Area in the S-4 Basin.
Expected Benefits: Modeling analyses conducted for the Caloosahatchee River Watershed
Protection Plan exercise estimated that, based on the 1970 to 2005 time period, the proposed
plan would reduce the occurrences of undesirable high flows above 2,800 cfs from 19% of the
time under 2005 base conditions (which assumed that no CERP or LOP2TP projects were in
place and that Lake Okeechobee releases to the estuaries and WCAs were based on the WSE
regulation schedule) to 10% for the Future with plan (that assumed the KRR, all Acceler8
projects described above, and all elements of both the LOP2TP and the Caloosahatchee
Watershed plans were fully constructed). Furthermore, modeling analyses estimated that
undesirable very high flows greater than 4,500 cfs would be reduced from 8.6% to 3.7% of the
time, and undesirable low flows less than 450 cfs would be reduced from 44% to 0.9% of the
time. While the Caloosahatchee watershed plan is estimated to achieve improvements in
performance over the LOP2TP it still does not achieve the performance targets published by
SFWMD (2009b).
Status:
 The LOP2TP elements: See Section III.2.d above.
 CERP C-43 basin storage reservoir project: See Section III.2.b above.
 Lake Hicpochee Hydrologic Enhancement Project - North: Construction is scheduled to
begin in FY2015.
 Dispersed Water Management: The total DWM either in operation or under construction
in the Caloosahatchee River watershed is estimated through preliminary modeling
studies to be approximately 9,900 acre-ft (Table III-8, from SFWMD (2015)). This
includes contributions from the U.S. Department of Agriculture Natural Resources
Conservation Service (USDA NRCS) Wetland Reserve Program (WRP) and other
programs, the FDACS BMP program, agricultural landowners, agricultural organizations,
non-governmental organization, and local governments.
50
Summaries, expected benefits and status of projects to improve water quality in the
Caloosahatchee watershed are described in Section IV.
Table III-8. SFWMD Dispersed Water Management Projects in the Caloosahatchee Watershed (from SFWMD,
2015).
Benefits (ac-ft/yr)
Mudge Ranch NE-PES 2 Operational 392
Barron Water Control Public Lands Operational 5,000
District
ECWCD Mirror Public Lands Operational 1,000
Lakes/Halfway Pond
Phase I
BOMA Public Lands Operational 836
Six Mile Cypress Slough Public Lands Construction 1,400
North
C-43 Interim Temporary Public Lands Operational 1,250
Storage
g. State of Florida Restoration Strategies
Plan Summary: To address water quality concerns associated with existing flows to the
Everglades Protection Area (EPA), the South Florida Water Management District (SFWMD),
Florida Department of Environmental Protection (FDEP), and United States Environmental
Protection Agency (USEPA) established a Water Quality-Based Effluent Limit (WQBEL)
intended to achieve compliance with the State of Florida’s numeric phosphorus criterion in the
EPA. In addition, they identified “Restoration Strategies”, a suite of additional water quality
projects to work in conjunction with the existing Everglades Stormwater Treatment Areas
(STAs) to meet the WQBEL (SFWMD, 2012; see details of legal requirements in Section II.4).
The Restoration Strategies projects primarily consist of Flow Equalization Basins (FEBs), STA
expansions, and associated infrastructure and conveyance improvements along three flow
paths (Eastern, Central, Southern). The primary purpose of FEBs is to attenuate peak
stormwater flows prior to delivery to STAs and provide dry season benefits, while the primary
purpose of STAs is to use biological processes to reduce phosphorus concentrations to achieve
the WQBEL (See Section IV for more details on water quality problems and potential solutions).
Additional water quality projects for the eastern flow path include the L-8 FEB in the S-5A Basin
with approximately 45,000 acre-ft of storage and an STA expansion of approximately 6,500
acres that will operate in conjunction with STA-1W. The additional project in the central flow
path is the A-1 FEB with approximately 56,000 acre-ft of storage on the A-1 Talisman property
that will attenuate peak flows to STA-3/4, and STA-2 and Compartment B. In the western flow
path an FEB with approximately 11,000 acre-ft of storage and approximately 800 acres of
effective treatment area within STA-5 are being added (Figure III-7).
51
Figure III-7: Location of Restoration Strategies FEBs and STA Expansions (SFWMD, 2012).
Expected Benefits: Modeling using the Dynamic Model for Stormwater Treatment Areas
(DMSTA) (Walker and Kadlec, 2011) and the peer-reviewed SFWMM (SFWMD 2005) indicates
that the proposed additional FEBs and STAs will allow approximately 1.5 million acre-ft of
inflows to the EPA to achieve the WQBEL. DMSTA has not undergone a formal, external peer-
review. As stated by NRC (2012) “External peer review is important, particularly for models that
are used extensively in the planning process, and peer review of the DMSTA is a high
priority." For more details about the expected benefits of Restoration Strategies see Section IV.
Status: The 45,000 acre-ft eastern flowpath FEB (L-8) and the 56,000 acre-ft Central Flowpath
FEB (A-1) are under construction and expected to be complete in 2016. The design of the STA
1W expansion is in process. The entire Restoration Strategies project is projected to be
completed in 2025.
h. CERP Central Everglades Planning Project (CEPP)
Plan Summary: The overarching purpose of the Central Everglades Planning Project (CEPP) is
to improve the quantity, quality, timing and distribution of water flows to the Caloosahatchee and
St. Lucie estuaries, Central Everglades, and Florida Bay while increasing water supply for
municipal, industrial and agricultural users. In general, ecosystem restoration objectives were
focused on providing additional water to the Everglades by capturing freshwater discharges
from Lake Okeechobee that would otherwise have gone to the St. Lucie and Caloosahatchee
estuaries. Timing of deliveries and distribution of the flows to the Everglades and improvements
to water supply for municipal, agricultural, and Tribal use were also evaluated. Options for
52
achieving restoration were constrained to maximize use of existing infrastructure capacity (i.e.
canals, STAs and Restoration Strategies projects), and to use existing, previously-impacted,
publicly-owned land in the EAA acquired for the purpose of environmental restoration (e.g., the
Talisman property).
Initial screening efforts (USACE, 2014b) resulted in two cost-effective options to achieve
restoration objectives:
 Two 14,000 acre Flow Equalization Basins (FEBs) with an ability to store a maximum of
112,000 acre-ft, i.e. a new FEB on the A-2 Talisman parcel and the previously planned
FEB on the A-1 Talisman parcel that was included in Restoration Strategies (described
in Section III.2.g), with Lake Okeechobee operations optimized for agricultural water
supply in the EAA. This option was estimated to provide approximately 210,000 acre-ft
of additional water annually to the Everglades system at an expected cost of $360-550
million.
 A 12-foot deep 21,000 acre reservoir with an ability to store a maximum of 252,000 acre-
ft and a 7,000 acre STA constructed on both the A-1 and A-2 Talisman parcels, also with
Lake Okeechobee operations optimized for agricultural water supply in the EAA, at a
cost of approximately $2 billion. This configuration provided the greatest benefits to the
Caloosahatchee and St. Lucie estuaries and delivered 240,000 ac-ft of additional water
to the Everglades rather than 210,000 ac-ft, under the FEB plan. However, as a result of
the 400-600% increase in cost to provide approximately 20% greater benefits over the
FEB option, the 12-foot reservoir configuration was eliminated from further
consideration.
The final recommended CEPP project is expected to provide a substantial increase in flow of
clean water to the Everglades. Water will be routed to two 14,000 acre FEBs which will provide
storage capacity, attenuate high flows, and may provide incidental water quality benefits prior to
delivery to State owned and operated STAs. The STAs will reduce phosphorus concentrations
in the water to meet required water quality standards. Re-routing this treated water south and
redistributing it across spreader canals is intended to facilitate hydropattern restoration in WCA
3A (Figure I-1). This, in combination with Miami Canal backfilling and other CERP components,
is intended to re-establish a 500,000-acre flowing system through the northern most extent of
the remnant Everglades. The treated water will be distributed through WCA 3A to WCA 3B and
ENP via structures and creation of the Blue Shanty Flowway. The Blue Shanty Flowway is
intended to restore continuous sheetflow and re-connection of a portion of WCA 3B to ENP and
Florida Bay. A seepage barrier wall and pump station are designed to manage seepage to
maintain levels of flood protection and water supply in the urban and agricultural areas east of
the WCAs and ENP.
53
Expected Benefits: CEPP Modeling, using the peer-reviewed RESOPS model (SFWMD,
2009c) for the 1965-2005 time period, indicated that the final recommended plan would send an
additional 210,000 acre-ft/year of water from Lake Okeechobee south to the Everglades
Protection Area on an annual average basis compared to a “Future without Project” condition
which assumed that the IRL-S, C-43, KRR, and Restoration Strategies projects were completed
(USACE, 2014b). This additional flow is approximately 70% of the 300,000 acre-ft/year
additional flow envisioned in the 1999 Restudy, and 30% of the 700,000 acre-ft/year additional
flow identified in the revised pre-drainage target used in the River of Grass Planning Process.
For the St. Lucie estuary, modeling based on the 41 year (1965-2005) time period estimated
that high mean monthly flows ( between 2,000 cfs and 3,000 cfs ) from Lake Okeechobee would
occur approximately 11% of the time for Existing Condition Baseline (ECB, which assumes no
CERP projects are constructed), approximately 11% of the time for the Future Without Project
Condition (FWO which assumes IRL-S, C-43, KRR and Restoration Strategies are constructed)
and reduced to 7.1% of the time for the Future with Project Condition (FWP, that assumes
CEPP is also constructed) (Table III-9). Furthermore modeling estimated that very high mean
monthly flows (>3,000 cfs) to the St. Lucie estuary would be reduced from 8.7% for the EBC to
6.3% for the FWO condition, to 5.7% for the FWP (USACE, 2014b).
For the Caloosahatchee estuary modeling estimated that the project would reduce high mean
monthly flows (>2,800 cfs) from Lake Okeechobee from 19% for the EBC, to 16% for the FWO
to 14% for the FWP (Table III-9). Similarly very high flows (>4,500 cfs) were estimated to be
reduced from 8.7% to 6.7% to 6.3% for the EBC, FWO and FWP, respectively. Low flows to the
Caloosahatchee estuary were estimated to be reduced from 24% of the time for the EBC, to
5.2% of the time for the FWO, to 4.7% of the time for the RWP condition (USACE, 2014b).
Thus, the modeling results indicate that the CEPP project, although producing significant benefit
for the Central Everglades, produces only relatively modest improvements in high flow
conditions and almost no improvement in very high flow conditions for the St. Lucie and
Caloosahatchee over that obtained from the IRL-S, C-43 and KRR projects. Modeling results
were not presented for the CEPP project alone (i.e. without completion of the IRL-S, C-43 and
KRR); however, it is likely that virtually no benefits would be realized for the St. Lucie and
Caloosahatchee estuaries without the increased storage and flow attenuation provided by those
projects.
54
Table III-9: existing conditions, restoration targets and expected benefits of proposed projects.
Percentages indicate percent of time the monthly flow condition occurs under specific scenarios. Red-
colored percentages have not achieved restoration goals.
Plan St. Lucie Caloosahatchee
# mean monthly #mean # mean # mean # mean
flows between monthly flows monthly flows monthly flows monthly flows
2000 and 3000 > 3000 cfs < 450 cfs > 2800 cfs > 4500 cfs
cfs
Restoration Targets <21 mo in 36 <6 mo in 36 0 mo in 36 <3 months in 0 mo in 36
years1 years1 years2 36 years2 years2
(4.9%) (1.4%) (0%) (0.7%) (0%)
Historical Performance 24 mo in 20 29 mo in 20 68 mo in 20 68 mo in 20 37 mo in 20
1994-2014 years3 years3 years3 years3 years3
(10%) (12%) (28%) (28%) (15%)
CEPP Existing 52 mo in 41 43 mo in 41 116 mo in 41 94 mo in 41 43 mo in 41
Condition Baseline, i.e. years4 years4 years4 years4 years4
no CERP projects (11%) (8.7%) (24%) (19%) (8.7%)
(2014)
Future without CEPP 54 mo in 41 31 mo in 41 27 mo in 41 81 mo in 41 33 mo in 41
project (includes IRL-S, years4 years4 years4 years4 years4
C-43, KRR and Rest. (11%) (6.3%) (5.2%) (16%) (6.7%)
Strat.) (2014)
Future with CEPP 35 mo in 41 28 mo in 41 23 mo in 41 68 mo in 41 31 mo in 41
selected plan (2014) years4 years4 years4 years4 years4
(7.1%) (5.7%) (4.7%) (14%) (6.3%)
1
2
3
4
based on simulations for 1965-2005 period of record (Tables 4-1 and 5.1-2, Figures 6-7 and 6-8, USACE
2014b)
Status: The CEPP PIR and EIS Report has been prepared and signed by the Chief of the US
Army Corps of Engineers. It is now undergoing additional review by the Secretary of the Army
(Civil Works) and the Office of Management and Budget. It will be formally transmitted to the US
Congress upon completion of those reviews.
The achievement of CEPP benefits for the Central Everglades depends on the successful
completion of SFWMD’s Restoration Strategies project. Once Restoration Strategies is
complete it is expected that the STAs that are part of Restoration Strategies will adequately
treat the existing flows plus additional water provided by CEPP (total 1.5 million acre-ft/year) to
meet the established WQBEL. The agencies have taken the position that construction of CEPP
features cannot commence until Restoration Strategies is complete and found to be in
compliance with the WQBEL. Pursuant to this position, even using the most optimistic
assumptions, CEPP is not estimated to be complete for a minimum of 24 years. This long time
frame for project completion is due, in part, to the need for coordinated phasing of individual
projects and some unavoidable constraints. Nevertheless, some of the constraints cited as
55
limiting CEPP project construction appear to be discretionary rather than mandatory. In
particular, the USACE PIR states that "all features of the State’s Restoration Strategies must be
completed and meet state water quality standards prior to initiating construction of most CEPP
project features." It is clear that any water leaving any of the STAs must comply with the
WQBEL, or prior to completion of Restoration Strategies, must be in compliance with the
NPDES and Everglades Forever Act permits and accompanying consent orders. Consequently,
CEPP will not be able to function until Restoration Strategies is complete and in compliance with
state water quality standards. However, this does not necessarily mean that construction of
most CEPP features cannot commence prior to completion of Restoration Strategies.
Construction of certain features of CEPP could begin during the time Restoration Strategies
projects are being constructed, and greatly shorten the aggregate time necessary for both
projects to be operational.
A careful analysis of CEPP project construction phasing should be conducted to determine

which of the CEPP features could usefully be initiated prior to the completion of Restoration
Strategies, and to develop a plan for completion of as many CEPP features as possible during
the construction phase of Restoration Strategies. Other CEPP features may need to be
constructed or completed after Restoration Strategies is complete, however, it is likely that
some of the work can be done simultaneously with Restoration Strategies construction thereby
potentially significantly reducing the number of years before CEPP can be fully implemented. In
its 2014 report, Progress Toward Restoring the Everglades (NRC, 2014), the National Research
Council concluded that the current timeline for completion of CEPP construction is
“unreasonable and undesirable” and recommended that USACE and SFWMD “look for creative
implementation strategies to reduce existing constraints.” Specifically, the NRC recommended
that the agencies “investigate implementation, and permitting alternatives that would enable
[CEPP] to move forward as quickly as possible with WQBEL-compliant discharges.”
4. Summary
Repeated planning exercises have shown that large volumes of inter-annual storage are
required north, south, east and west of Lake Okeechobee to manage Lake Okeechobee levels
within a desirable range, reduce damaging high and low flows to the St. Lucie and
Caloosahatchee estuaries, and move more water south for agricultural, urban and ecosystem
uses. In spite of this, very little new storage has been designed or constructed. In the St. Lucie
watershed approximately 200,000 acre-ft of storage is required according to the most recent
estimates, yet only one approximately 40,000 acre-ft surface reservoir currently is under
construction (Table III-10). In the Caloosahatchee watershed approximately 400,000 acre-ft of
storage is required according to most recent estimates yet only one 170,000 acre-ft surface
reservoir currently is being designed and federal funds for its construction have not yet been
appropriated. According to the River of Grass Planning effort at least one million acre-ft of
storage is required in some combination north or south of Lake Okeechobee, yet only 168,000
56
acre-ft of shallow storage in FEBs has been sited south of Lake Okeechobee, with 101,000
acre-ft currently under construction. Throughout the Lake Okeechobee, St. Lucie and
Caloosahatchee watersheds approximately 92,000 acre-ft of DWM projects have been
established, but the efficacy of this type of local, typically intra-seasonal, water retention for
managing Lake Okeechobee levels within a desirable range, reducing damaging discharges to
the St. Lucie and Caloosahatchee estuaries, or providing long-term storage of water that can be
used for agricultural, urban or environmental uses in subsequent dry seasons is unknown.
If the currently authorized CERP projects, i.e. the complete IRL-S project and the C-43 project,
are constructed, and the Kissimmee River Restoration Project is completed, it is estimated that
very high volume damaging freshwater discharges to the St. Lucie estuary would be reduced
from 8.7% of the time to 6.3% of the time versus a restoration target of 1.4%. Similarly, very
high volume damaging freshwater discharges to the Caloosahatchee estuary would be reduced
from 8.7% to 6.7% if these authorized projects were constructed, versus a restoration target 0%.
If the CEPP project is also authorized and constructed, the occurrences of very high volume
damaging freshwater discharges will be marginally reduced over these estimates, i.e. to 5.7%
for the St. Lucie and to 6.3% for the Caloosahatchee. To further reduce very high freshwater
discharges at approximately one million acre-ft of additional storage must be constructed, either
north or South of Lake Okeechobee, as envisioned in both the LOP2TP and the River of Grass
Planning Process.
As indicated above, the CEPP modeling analyses indicated that the deep reservoir-STA option
on the Talisman property provided 140,000 acre-ft more storage and higher benefits to both the
estuaries and the Everglades than the selected FEB plan, but was rejected due to cost
considerations. Given that 1) it is clear that large volumes of storage are needed throughout the
South Florida System to reach restoration objectives, 2) the cost of deep storage is unlikely to
decrease, and 3) the Talisman property is currently the only state-owned non-conservation
property suitable for deep storage in the EAA, this is a decision that could be revisited during
the development of the detailed design phase for the CEPP FEB.
57
Table III-10: Summary of Estimated Storage Required by Plans and Provided by Approved Projects1
Estimated Storage Required to Achieve Restoration Targets
Plan North of Lake Caloosahatchee
EAA St. Lucie Watershed
Okeechobee Watershed
CERP Restudy 360,000 acre-ft deep 250,000 acre-ft deep 389,400 acre-ft 160,000 acre-ft deep
Plan (1999) reservoir reservoirs, deep reservoirs reservoir
1000 mgd ASR 220 mgd ASR
NEEPP Plans KRR and all Acceler8 900,000-1,300,000 200,000 acre-ft 400,000 acre-ft
(2008-2009) projects assumed acre-ft reservoirs, reservoirs, DWM reservoirs, ASR,
operational ASR, DWM DWM
ROG plan (2009- 1,000,000 - 1,350,000 acre-ft C-44 Reservoir and C-43 Reservoir
2010) deep reservoirs, shallow reservoirs, shallow dry STA assumed assumed operational
storage areas, FEBs operational
Project 1 Existing Approved Projects to Provide Components of Required Storage
Authorized IRL-S 130,000 acre-ft deep
Project reservoirs
Designed C-44 40,000 acre-ft deep
Project reservoir
(component of
IRL-S)
Authorized C-43 170,000 acre-ft deep
Project reservoir
Restoration 112,000 acre-ft FEB2
Strategies
CEPP Project 56,000 acre-ft FEB
1
None of the projects in this table are operational at the time of report writing. CEPP is currently awaiting Federal
authorization. See previous sections for details on status of each project.
2
Includes 45,000 acre-ft in eastern flow-way, 56,000 acre-ft in central flowway and 11,000 acre-ft in western flow-
way
To provide substantial improvement to the St. Lucie and Caloosahatchee estuaries accelerate
the funding and completion of existing authorized CERP projects designed specifically to
provide relief to St. Lucie and Caloosahatchee Basins, i.e. :
 IRL-S Project: Accelerate construction of C-44 reservoir and STA. Aggressively pursue
state and federal appropriations needed to design and construct remainder of the project
(including C-23, 24, 25 reservoirs and STAs, and restoration of over 90,000 acres of
upland and wetland areas).
 C-43 Reservoir: Accelerate design and aggressively pursue state and federal
appropriations needed to design and construct this project.
To substantially increase the volume of water moving south from Lake Okeechobee to the
Everglades, accelerate funding and completion of the State of Florida Restoration Strategies
and the CERP Central Everglades Planning Project (CEPP) . i.e.:
58
 Obtain federal authorization for CEPP.
CEPP.
 Accelerate state funding and completion of Restoration Strategies.
 Conduct a careful analysis of CEPP project construction phasing to determine which
CEPP features can be constructed as soon as possible and to develop a plan for
completion of as many CEPP features as possible during the construction phase of
Restoration Strategies.
 Reconsider using the Talisman property for a deep storage reservoir with STA rather
than for shallow FEBs during the CEPP detailed design phase.
All modeling studies that assess restoration potential for CERP, NEEPP and ROG plans and
IRL-S, C-43, and CEPP projects assume that the KRR has been completed. However, no
modeling studies have isolated the potential of the KRR to affect discharges to the northern
estuaries, or to affect the ability to move water south from Lake Okeechobee. Although it is not
possible from existing model runs to discern the individual contribution of the KRR project to
these larger restoration goals, the Technical Review Team expects that it is likely to be relatively
minor. Nevertheless, the KRR is expected to provide significant benefits within the Lake
Okeechobee watershed and should be completed as soon as possible.
As indicated in Table III-10, even after these existing approved projects are constructed
significant additional water storage, treatment and conveyance will be required to fully achieve
restoration goals. Section IV reviews existing state plans and projects to improve water quality.
Section V reviews additional options (beyond approved projects) to provide the additional
required water storage, treatment and conveyance.
59
IV. Existing State Plans to Improve Water Quality
The need to store substantial volumes of water both north and south of Lake Okeechobee is
well recognized to both reduce high freshwater flows to the Caloosahatchee and St. Lucie
estuaries and move more water south of Lake Okeechobee to the Everglades Protection Area
(EPA) (Section III). Increased storage capacity in and of itself is not sufficient, however, to
reduce damage to the estuaries and the EPA which is due to both problems with the volume,
timing and distribution of freshwater flows and high nutrient loads. Even if all storage needs
were achieved, water quality remains a significant problem for the St. Lucie and
Caloosahatchee estuaries and a direct legal hurdle for moving water south of the lake into the
EPA. Thus, a holistic and coordinated approach to providing both storage and treatment is
needed to maintain water quality and protect these systems and the ecosystem services they
provide.
Overall, water quality concerns represent a major management concern, and they have a
profound influence on any decision or plan to store and convey water in the Greater Everglades
ecosystem. State and federal agencies have relied heavily on scientific literature to evaluate
best management practices and available treatment technologies, and to identify on-site
management practices and construction projects that will help reduce nutrient loads in the
discharged water. In this section, we review water quality related issues in the context of
existing state and federal restoration plans to reduce damaging discharges to the
Caloosahatchee and St. Lucie estuaries, and move more water south of Lake Okeechobee.
From a water quality perspective, phosphorus (P) is unambiguously the nutrient of greatest
concern in Lake Okeechobee and the Southern Everglades. For the St. Lucie and
Caloosahatchee estuaries nitrogen (N) is also a concern.
1. Greater Everglades Watershed Descriptions
Northern Everglades-Lake Okeechobee Watershed
Lake Okeechobee, the heart of the Greater Everglades Ecosystem, receives water that
originates in a 3.4 million acre watershed. The watershed is comprised of six sub-basins, Upper
Kissimmee, Lower Kissimmee, Taylor Creek/Nubbin Slough, Lake Istokpoga, Indian Prairie, and
Fisheating Creek (Figure IV-1; SFWMD, 2015). Three sub-basins situated west, east, and south
of the lake, in general, do not contribute water to Lake Okeechobee; i.e. water is discharged
away from the lake. In combination, the aforementioned six sub-basins deliver approximately
88% of the freshwater that enters Lake Okeechobee (SFWMD, 2014). Land use in the
watershed is primarily agriculture (51%) with natural areas (31%) and urban development (10%)
accounting for the majority of the balance. Approximately 18% and 15% of the Lake
Okeechobee watershed exists as wetlands and open water bodies, respectively, with these
60
systems being the final recipients of discharges from adjacent terrestrial systems (FDEP,
2014a).
Figure IV-1. The Lake Okeechobee watershed (delineated by the red line) and color-coded sub-basins (Source:
SFWMD).
Northern Everglades- Caloosahatchee River and St. Lucie River Watersheds
To the west and east of Lake Okeechobee, lie the Caloosahatchee River watershed (CRW) and
the St. Lucie River watershed (SLRW). Both watersheds have been extensively altered as a
result of changes in land use and hydrology. Flows from Lake Okeechobee and surface water
runoff from these two watersheds have a major influence on the physics, chemistry, biology,
and ecology of both the Caloosahatchee River and St. Lucie River estuaries, with freshwater
flows and nutrient inputs representing two major influences (Barnes, 2005; Sime, 2005).
The CRW (C-43 Basin) is comprised of five sub-basins (Figure IV-2). Approximately 55% of the
watershed is represented by the S-4, East Caloosahatchee (adjacent to Lake Okeechobee) and
West Caloosahatchee sub-basins. The remaining 45% of the watershed includes the Tidal
61
Caloosahatchee and Coastal Caloosahatchee sub-basins. The total watershed area is
1,090,381 acres, with 34.6% attributed to agriculture, 18.5% to urban and developed lands;
15.9% to wetlands; 12.5% to upland forests; 11.7% to water, and 5.6% to rangelands. The
remaining area is identified as used for transportation, communication, and utilities, and barren
land (SFWMD, 2015).
The SLRW is comprised of seven sub-basins that drain into the St. Lucie River or its estuary
(SLRE; Figure IV-2). These include: C-44; C-23; C-24; C-25; Basin 4, 5, and 6; North Fork;
South Fork; and a portion of South Coastal sub-basins. The total watershed area is 537,805
acres, with 53.9% of the area used for agriculture. Other land uses/classifications include urban
and built up (19.1%); wetlands (10.9%); upland forests (6.8%); water (3.3%); and rangelands
(3.2%). The remaining area is identified as being used for transportation, communication,
utilities, and barren land (SFWMD, 2015).
Figure IV-2. Map showing Caloosahatchee River watershed (CRW) (left panel) and St. Lucie River
watershed (SLRW) (right panel) (source: SFWMD 2015).
Southern Everglades- Everglades Agricultural Area and C-139 Basins
South of Lake Okeechobee lay the EAA and C-139 basin (Figure IV-3). The EAA is ~ 700,000
acres (Buzzelli et al., 2012) and is dedicated largely to the production of sugarcane, vegetables
and sod. A protective levee separates the EAA from northwest Palm Beach County and Miami-
Dade County. The C-139 basin (170,000 acres) is located west of the EAA, with pasture, row
crops, citrus, sugarcane and agriculture being the dominant land uses.
62
Figure IV-3. Map showing C-139 (left panel) and the Everglades Agricultural Area (right panel) watersheds (source:
SFWMD)
2. Greater Everglades Watershed Nutrient Loads
Northern Everglades-Lake Okeechobee Watershed
Freshwater flows and associated nutrient loads to Lake Okeechobee and the Caloosahatchee
and St. Lucie estuaries are driven, in large part, by regional rainfall patterns, which can be quite
variable, especially in the wake of tropical storms and hurricanes. Phosphorus (P) is a limiting
nutrient and primary determinant of productivity within the terrestrial (upland), wetland and
aquatic ecosystems of the Greater Everglades. For the past several decades P imported into
the basin, primarily to improve agricultural production, has largely accumulated in soils and
sediments. In addition to current P imports, the legacy P in the basin has become a constant
source of P to the lake and estuaries. As such, these ecosystems, whether they exist north,
east, west or south of Lake Okeechobee, are sensitive to anthropogenic nutrient loads.
Increased P loads have, in fact, accelerated the eutrophication process across the south Florida
landscape. Although much attention is focused on phosphorus as the nutrient of concern,
nitrogen too is delivered to Lake Okeechobee and the loading rates are, as expected, highly
correlated.
To avoid excessive P delivery to Lake Okeechobee and the associated algal blooms and growth
of exotic and nuisance aquatic plants in the littoral zone, FDEP established a Total Maximum
Daily Load (TMDL) of 140 metric tons of P per year (FDEP, 2014a). This load includes 35 metric
tons of P per year from atmospheric deposition, resulting in an allowable P load from the
watershed of 105 metric tons per year. Since 1974, annual total P loads to Lake Okeechobee
have exceeded 500 metric tons nearly 50% of the time (Figure IV-4). Averaged over the 41-year
63
period of record, the annual P load is approximately 3.6 times the annualized TMDL. Thus,
annual average P loads will have to be reduced by more than 350 metric tons per year to meet
the current TMDL for the lake.
Figure IV-4. Annual and 5 year-moving average of total phosphorus (TP) loads to Lake Okeechobee between 1974
and 2014. The TMDL target of 140 metric tons (mt) (red dashed line) is indicated for comparison (Data source:
SFWMD, 2015).
Annual P loads are highly variable and quite clearly influenced by climatic conditions
(precipitation in particular), along with changes in land use and water management activities.
During wet years, large volumes of water (~ 3 million acre-feet) are delivered to the lake, while
in dry years the volume of water delivered to the lake can be less than 1 million acre-ft. Annual
P loads to Lake Okeechobee reflect this variation, with high P loads observed during wet years
and low P loads during dry years (Figures IV-4 and IV-7). In fact, Jawitz and Mitchell (2011)
reported that 80% of the P load to Lake Okeechobee is delivered in approximately 73 days in
any particular year. This unequal load distribution presents enormous challenges for water
resource managers.
Overall, ~11,000 metric tons of P per year are imported into the Lake Okeechobee watershed
(1,392,874 ha), with slightly more than half (51%) being exported and 49% (5,350 metric tons
per year) remaining in the system (HDR, 2010). Phosphorus that remains in the watershed and
is stored in soils is considered a legacy pool that can contribute to overall P load even after P
imports are decreased. Legacy P can substantially extend the time required for a wetland or
aquatic system to recover from an impaired state and/or revert to a stable condition that is
compliant with an adopted water quality standard (Figure IV-5). This legacy P can be classified
as either reactive or non-reactive. The reactive pool is comprised of P that potentially can be
released from the watershed, whereas non-reactive P is considered to exist in a stable pool and
is essentially immobile. The reactive P pool represents approximately 65% of the legacy P in
64
soils (Reddy et al., 2011). Recent reports estimate that 170,000 metric tons of P is stored in
Lake Okeechobee watershed soils (Reddy et al., 2011; SWET, 2008a,b). This estimate equates
to approximately 110,500 metric tons of reactive P which can potentially be leached out of the
watershed. Thus, legacy P in the Lake Okeechobee watershed could sustain contemporary P
loading rates, i.e. 500 metric tons per year, for more than two centuries. Clearly, there is need to
fully recognize the potential contribution of legacy P to any future P-loading scenario.
Water Column Phosphorus External Load

Reduction
Background
Internal Level
Memory
Lag time for Recovery
Time - Years
Figure IV-5. Schematic showing the effects of internal phosphorus memory in the watershed on ecosystem recovery
(Reddy and DeLaune, 2008).
The six sub-basins within the Lake Okeechobee watershed contribute differing quantities of
water and P (Figure IV-6), but, in all cases, overall P loads are highly correlated with the
combined sub-basin water discharge (Figure IV-7). High water discharge to the lake and
correspondingly high P loads (> 900 metric tons per year) resulted from hurricanes that affected
the region in 2005 and 2006. Low water discharges associated with drier years, e.g., WY1997,
WY2001, WY2007, WY2008 and WY2011, generated markedly lower average P loads (155
metric tons per year). As a means of differentiating the six sub-basins, P loads are compared for
the following conditions: (1) baseline average P loads during WY1991-WY2005, (2) baseline
average P loads during WY2001-2012, (3) the most recent drought year – WY2011, and (4) the
most recent wet year – WY2014 (Figure IV-8).
Phosphorus originating in the Indian Prairie, Fisheating Creek, and Lake Istokpoga sub-basins
increased between the two baseline periods (WY1991- WY2005 and WY2001-WY2012) by 16,
27, and 52%, respectively. During the same two time periods, P loads decreased by 3, 19, and
20%, respectively, for the Upper Kissimmee, Lower Kissimmee, and Taylor Creek/Nubbin
Slough sub-basins. During WY2011, a drought year, P loads were reduced in all of the sub-
basins relative to either of the baseline periods. In combination, only 177 metric tons of P were
delivered to Lake Okeechobee in WY2011 (SFWMD, 2015). In more recent years, total P loads
to the lake have increased markedly. In WY2012, WY2013, and WY2014, 377, 569, and 609
metric tons of P were delivered to Lake Okeechobee, respectively. The increases, as noted
above, can be attributed to an increase in rainfall in the watershed and subsequent discharges
65
to the lake. Long-term P loads (1991-2012) from Taylor Creek/Nubbin Slough, Indian Prairie,
and Fisheating Creek contributed more than 55% of the P load, but only comprised 27% of the
available watershed area. These observations suggest that these latter sub-basins be
considered as having high priority in establishing P control strategies to reduce P loads.
Further evidence for a need to focus on certain sub-basins arises from a consideration of
phosphorus loads normalized per unit area (Figure IV-9). These data provide additional insight
into the relative stability of stored P and its capacity to be released into the surface water
system. On a per unit area basis, the Taylor Creek/Nubbin Slough sub-basin released P at
higher rates than other sub-basins, regardless of the time frame considered. Notable is the
observation that the P loads originating in the Taylor Creek/Nubbin Slough basin, both total and
per unit area, declined markedly between the two baseline periods reflecting some success in
nutrient remediation efforts. In contrast, P loads from the Indian Prairie basin increased between
the two baseline periods and represent a management concern.
Figure IV-6. Relative proportion of WY2014 freshwater delivery and total phosphorus load to Lake Okeechobee
from each of the following sub-basins: Upper Kissimmee, Lower Kissimmee, Lake Istokpoga, Indian Prairie, and
Taylor Creek/Nubbin Slough. Note that the East, West and South sub-basins contribute little to water delivery and P
load to the lake relative to their combined area (Data source: SFWMD).
66
Figure IV-7. Relationship between water discharge and total P loads between WY1995 and WY2014 (Data source:
SFWMD).
Figure IV-8. Annual total phosphorus loads from sub-basins within the Lake Okeechobee watershed. Note that these
loads do not include atmospheric deposition of phosphorus (Data source: SFWMD, 2015).
Figure IV-9. Annual phosphorus release rates (kg P/ha) from sub-watersheds to Lake Okeechobee (Data source:
SFWMD, 2015).
67
All currently developed nutrient source control programs for the Lake Okeechobee watershed
primarily focus on P reduction, but consideration should be given to nitrogen control as well
because the ratios of these nutrients can yield variable effects on eutrophication of waterbodies
(Jacoby and Frazer, 2009) and are particularly important in estuaries. Loads of N to Lake
Okeechobee measured between WY2000 and WY2014 ranged from 2,500 to 8,800 metric tons
N per year (Figure IV-10).
Figure IV-10. Relationship between water discharge and total N loads during WY2000 to WY2014 (Data source:
SFWMD, 2015).
During the past 15 years, 11 water years corresponded to wet years and high N loading rates
(range: 4,620 to 8,775 metric tons per year) were recorded, including WY2005 and WY2006 in
which hurricanes resulted in substantial regional rainfall. Drought conditions during the period of
record occurred in WY2001, 2007, 2008, and 2011. The average annual N load to Lake
Okeechobee during this time was only 2,947 metric tons (range: 2,517 to 3,393 metric tons per
year). Furthermore, during WY2000 to WY2014, the N load per unit of P load or the ratio of N to
P loads ranged from 10 to 13 during wet years as compared to 14 to 19 during dry years. Such
variation highlights differences in the cycling of these two important elements. Unlike P, N
undergoes several biogeochemical transformations including the conversion of nitrate to
gaseous end products such as nitrous oxide and nitrogen gas via the denitrification process and
loss of ammonia as a consequence of volatilization (Reddy and Delaune, 2008). The
denitrification process is facilitated by wet or flooded soil conditions. It is likely that during wet
years, substantial nitrogen is lost to the atmosphere as a consequence of denitrification in the
Lake Okeechobee watershed, thus reducing the amount of N delivered to the lake relative to P.
Under wet soil conditions, P accumulated in soils can be solubilized and exported with water
moving through the system (Reddy et al., 2011), thus exacerbating a shift toward lower N to P
ratios. During drier years, high N to P load ratios are reflective of a slower release of P and rapid
mineralization of organic N to ammonium and nitrate. In fact, up to 90% of the nitrogen present
68
in the water delivered to the lake is organic in nature, and it must be mineralized to inorganic N
by microbial organisms before it is made available to other biota.
Northern Everglades-Caloosahatchee and St. Lucie River Watersheds
Both phosphorus and nitrogen loads in the Caloosahatchee and St. Lucie River watersheds are
significantly correlated with water discharge, with coefficients of determination (r2) of 0.70 and
0.99, respectively. The long-term average annual P load to the CRE (WY1997-WY2014) was
estimated to be 282 metric tons with 23% derived from Lake Okeechobee (Figure IV-11).
Figure IV-11. Phosphorus loads to St. Lucie River and Caloosahatchee River estuaries (SFWMD, 2015).
The long-term average annual N load to the CRE (WY1997-WY2014) was estimated to be
2,952 metric tons with 34% derived from the lake (Figure IV-12). With regard to the SLRE, the
long-term average annual P load (WY1997-WY2014) was 312 metric tons with 18% derived
from Lake Okeechobee, and the long-term average annual N load (WY1997-WY2014) was
1,715 metric tons with 32% derived from the lake. Substantial precipitation and subsequent high
hydraulic loading during WY2014 increased P loads to the CRE to 425 metric tons with 26%
contributed from Lake Okeechobee, P loads to the SLRE to 375 metric tons of P with 19%
contributed from Lake Okeechobee, N loads to the CRE to 5,100 metric tons with 37%
contributed from Lake Okeechobee, and N loads to the SLRE to 2,200 metric tons with the
relative contribution from the lake remaining approximately equal to the long-term average; i.e.
31%. In combination, these results suggest that approximately 65-80% of the nutrient load to
CRE and SLRE is derived from their respective watersheds, with only 20-35% contributed by
Lake Okeechobee.
69
Figure IV-12. Nitrogen loads to St. Lucie River and Caloosahatchee River estuaries (SFWMD, 2015).
Southern Everglades
Moving water south, especially into the Everglades Protection Area (EPA), requires adherence
to stringent water quality standards (see Section II-5), especially as they relate to P. Agricultural
Best Management Practices (BMPs) and Stormwater Treatment Areas (STAs) are intended to
reduce overall P loads to the EPA. In fact, the Everglades Forever Act (EFA) mandates
implementation of BMPs followed by monitoring to assess the effectiveness of source control
programs to achieve required P loads at the basin scale and reduce P loads to the STAs. The
largest source of P and other nutrients to the EPA, is the Everglades Agricultural Area (EAA)
(Figure IV-13). Total P concentration (flow-weighted mean (FWM) concentration) in the outflow
from the EAA over the period of record averaged 132 µg P/L with a range of 64 to 243 µg P/L.
During WY2014, P concentrations in EAA outflow were 94 µg P/L, which is markedly lower than
35-year average value.
As mandated by the EFA, the EAA basin is required to achieve a 25% reduction in TP loads
discharged when compared to the designated baseline period, with adjustments to account for
variation in rainfall (see SFWMD, 2015). Since agricultural BMPs were implemented on 640,000
acres south of Lake Okeechobee estimated load reductions have surpassed the required 25%
load reduction, for example achieving a 63% reduction of observed loads versus modeled loads
estimated assuming no BMPs were implemented. This equates to an average of approximately
180 metric tons per year of modeled P reduction, with 105 metric tons per year released to
STAs for further treatment (SFWMD, 2015). Since WY1996, modeling estimates indicate that
agricultural BMPs have resulted in a total cumulative reduction in total P load of approximately
2,853 metric tons, which represents a long-term reduction of 55 percent (SFWMD, 2015).
70
Figure: IV-13. Long-term (WY1980-WY2014) phosphorus loads from the EAA Basin (SFWMD, 2015).
For the EPA, another source of P and other nutrients is the C-139 basin (170,000 acres). The
goal for this basin is to maintain P loads at or below historic baseline loads. The 35-year
average annual P load (WY1980 through WY2014) was estimated at 37 metric tons, while the
average target load for the same period was 30 metric tons P per year (Figure IV-14). From
WY1980 through WY2014, total P concentrations (FWMC) in outflows from the C-139 basin
averaged 189 µg P/L, and concentrations ranged from 69 to 363 µg P/L (SFWMD, 2015).
Recent reduced P loads are consistent with the implementation of the comprehensive BMP plan
in 2011. Loads have exceeded the target in only 2 of the most recent four years.
Figure IV-14: Long-term (WY1980-WY2014) phosphorus loads from the C-139 Basin (Data source:
SFWMD).
71
3. Current and Planned Management of Greater Everglades Watershed Nutrient Loads
The State of Florida has invested more than $2 billion during the last two decades to improve
water quality in south Florida ecosystems including both the northern and southern Everglades
(Figure IV-15). A review of current and planned nutrient reduction efforts in all regional
watersheds provides insights into challenges associated with moving clean water south.
Figure IV-15. Watersheds in the Northern Everglades as well as ECP and non-ECP basins in the Southern
Everglades Ecosystem (Source: SFWMD).
72
Specific guidance for current and planned management activities is provided in a suite of
assessments and plans. In 2001 the FDEP established a Total Maximum Daily Load (TMDL) for
the Lake Okeechobee watershed. TMDLs were established for the St. Lucie Basin and the
Caloosahatchee Estuary in 2008 and 2009, respectively. As discussed in Section III.2.b, in 2007
the State of Florida Legislature initiated the Northern Everglades and Estuaries Protection
Program (NEEPP) to promote a comprehensive, interconnected watershed approach to
protecting Lake Okeechobee and the Caloosahatchee and St. Lucie rivers and estuaries that
specifically addressed both water quality and the quantity, timing and distribution of water to the
natural systems. The program resulted in the development of the Lake Okeechobee Watershed
Construction Project Phase II Technical Plan (SFWMD, 2008), the St. Lucie river Watershed
Protection Plan (SFWMD, 2009a), and the Caloosahatchee River Watershed Protection Plan
(SFWMD, 2009b) which identified construction projects, along with on-site measures intended
to prevent or reduce pollution at its source to achieve water quality targets for the Lake
Okeechobee and the Caloosahatchee and St. Lucie estuaries. Since the NEEPP reports were
prepared, Basin Management Action Plans (BMAPs) have been developed for the Lake
Okeechobee, St. Lucie and Caloosahatchee basins (FDEP, 2012, 2013, 2014a). These BMAPs
provide the blueprint to meet the TMDLs established by the FDEP to address water quality
issues in the Northern Everglades. Planning efforts and projects that address water quality
issues in the Southern Everglades include the Florida Water Resources Act, Comprehensive
Everglades Restoration Plan, and Everglades Forever Act.
Northern Everglades - Lake Okeechobee Basin Management Action Plan (BMAP)
Several state (SFWMD, FDEP, FDACS) and federal (USACE, DOI) agencies, local
governments, and stakeholders have been working cooperatively to address inter-related issues
that affect reductions in TP loads and enhance ecosystem services provided by Lake
Okeechobee. In 2008, the Lake Okeechobee Phase II Technical Plan (LPO2TP) identified
construction projects, along with on-site measures (Best Management Practice, BMPs) intended
to prevent or reduce pollution at its source to achieve water quality targets for the lake. In
December 2014, the FDEP adopted the Lake Okeechobee BMAP (FDEP, 2014a). BMAP
projects are designed for phased implementation to achieve the TMDL established for Lake
Okeechobee, with initial effort focused on the six sub-basins that contribute approximately 89%
of the total P load to the lake: Fisheating Creek; Indian Prairie, Lake Istokpoga, Lower
Kissimmee, Upper Kissimmee, and Taylor Creek/Nubbin Slough (Figure IV-4). Phase 1
proposed BMAP projects for these sub-basins include urban and agricultural BMPs, Dispersed
Water Management (DWM), Stormwater Treatment Areas (STAs), Floating Aquatic Vegetation
Treatment (FAVT), Hybrid Wetland Treatment Technology (HWTT), natural systems restoration,
and education and outreach (Table IV-1). Descriptions of these projects and their expected
water quality benefits are summarized below and in Table IV-2. Phase I BMAP projects will be
completed over the next 10 years. These projects are expected to reduce total P by
73
approximately 150 metric tons per year when completely implemented, out of a required total P
reduction of approximately 370 metric tons per year.
Table IV-1. Basin Management Action Plan for Lake Okeechobee Watershed (source: FDEP, 2014a). Sub-basins:
Fisheating Creek (FC); Indian Prairie (IP); Lake Istokpoga (LI); Lower Kissimmee (LK); Upper Kissimmee (UK);
Taylor Creek/Nubbin Slough (TCNS).
Entities Project/Program Type Sub-basins

FDACS BMPs (voluntary or cost-share) – Agricultural lands FC, IP, LI, LK, UK, and TCNS
Counties BMPs – Urban and Municipal IP, LI, LK, UK, and TCNS
SFWMD Dispersed Water Management (DWM) FC, IP, LI, LK, UK and TCNS
SFWMD, FDACS, and FDEP Stormwater Treatment Areas (STAs) IP and TCNS
FDACS Floating Aquatic Vegetation Treatment (FAVT) FC, IP, LI, LK, UK, and TCNS
FDACS Hybrid Wetland Treatment Technology (HWTT) TCNS
SFWMD Restoration LK and UK
Counties Public education/outreach FC, IP, LI, LK, UK, and TCNS
Best Management Practices (BMPs)
Implementation of Best Management Practices as outlined in BMAPs can (1) improve water
quality and reduce P loads from sub-basins to Lake Okeechobee; (2) increase coordinated
efforts to address surface water quality issues by state and local governments; (3) engage
stakeholders in decision-making and priority setting processes; (4) increase understanding of
basin hydrology, sources of pollutants, legacy P, and water quality; and (5) increase public
awareness of complex issues (FDEP, 2014a). Appropriate BMPs have been proposed for
agricultural and urban lands, but it is often difficult to assess their effectiveness with regard to
improving water quality and reducing loads of P due to climate variability, landscape
heterogeneity and limited monitoring funds. It is estimated that the current BMP programs will
remove approximately 32.8 metric tons P per year (FDEP, 2014a); however these model
predictions lack robust validation due to a dearth of data. Furthermore, none of the proposed
BMPs address the issue of legacy P. Chemical amendments and treatment of hot spots may be
needed to reduce the release of legacy P.
Dispersed Water Management (DWM)
Dispersed water management refers to the distribution of water across a highly parceled
landscape using relatively simple structures (FDEP, 2014a). The SFWMD initiated a DWM
program entitled the Northern Everglades Payment for Environmental Services (NE-PES).
Under this program, cattle ranchers are compensated for providing water storage and nutrient
retention on private lands (Bohlen et al., 2009; Lynch and Shabman, 2011). The DWM program
is planned to extend to all six sub-basins (FDEP, 2014a). This approach provides shallow water
storage, and, in some cases, it can reduce P and N loads. However, flooding of lands and
resulting anaerobic conditions in soils can potentially solubilize some of the stable, legacy P and
74
increase its mobility in surface or groundwater. In addition, flooding may cause a shift in
vegetation to water tolerant plants that may not be grazed readily by cattle. However, the
benefits of storing water on ranchlands during peak flows might offset such negative effects. It is
estimated that the current DWM programs will remove approximately 10.2 metric tons of P per
year (FDEP, 2014a). However, as discussed in more detail in Section V.2, new data gathering
and modeling efforts are required to more accurately simulate the cumulative impacts of DWM
on the quality, quantity and timing of flows into Lake Okeechobee.
Hybrid Wetland Treatment Technology (HWTT)
Hybrid Wetland Treatment Technology relies on a combination of chemical treatments and

wetlands to remove P at sub-basin and parcel levels. Currently, five HWTT systems are
operating in the Taylor Creek/Nubbin Slough sub-basin, with approximate removal of 7.6 metric
tons of P per year (FDEP, 2014a). It should be noted that these systems require less land than
STAs but more intensive management of vegetation. Disposal of the floating aquatic vegetation
(in this case, water hyacinths) and potential insect damage to the systems’ monocultures remain
unknowns that hamper evaluation of the long-term sustainability of P removal by such systems.
Further evaluation is needed to determine the long-term sustainability of expected P removal
rates and cost of operating these systems.
Floating Aquatic Vegetation Tilling Systems (FAVT)
Similar to HWTT systems, Floating Aquatic Vegetation Tilling Systems grow floating aquatic
vegetation (such as water hyacinths) in shallow reservoirs, but the biomass and associated
nutrients are tilled into the soil during dry seasons rather than being harvested. Currently, one
FAVT is operating in the Fisheating Creek sub-basin, with P removal estimated at 8.6 metric
tons per year (FDEP, 2014a). At present, there is not adequate information to evaluate the long-
term sustainability of P removal by this system because biomass incorporated into the soil
undergoes rapid decomposition and it releases P and other nutrients that can enter the water
column once the soil is flooded. Further evaluation is needed to determine the long-term
sustainability of expected P removal rates and cost of operating these systems.
Stormwater Treatment Areas (STAs):
The lessons learned and knowledge gained during the operation of Southern Everglades STAs
over the last two decades suggest that STAs also can be designed and constructed in the sub-
basins north of Lake Okeechobee to reduce P loads. Furthermore Flow Equalization Basin
(FEB)-STA technology offers substantial promise as a P load reduction tool and may aid also in
meeting some needed water storage needs (see Section III). At present, four sub-basins,
TCNS, IP, FC, and LK, generate approximately 80% (400 metric tons per year) of the P load to
Lake Okeechobee. Currently, three STAs are in various phases of implementation in the TCNS
75
sub-basin with a total 3,671 acres (1,486 ha) and 3,200 acres of effective treatment area. It is
estimated that the current STA projects will remove approximately 22 metric tons P per year
(FDEP, 2014a).
Table IV-2. Total phosphorus loads and reduction targets for the Lake Okeechobee watershed. (source: FDEP,
2014a).
Category Total Phosphorus, metric tons/year

Total P load [1975-2014]1 503
Baseline P load [1991-2005]1 517
Baseline P load [2001-2012]1 512
TMDL (including contribution from atmospheric deposition) 140
Reductions needed to achieve TMDL 372
Projects identified in 6 sub-basins 103
Fisheating Creek2 [18.9]
Indian Prairie2 [11.8]
Lake Istokpoga2 [1.8]
Lower Kissimmee2 [23.3]
Taylor Creek/Nubbin Slough2 [40.8]
Upper Kissimmee2 [6.8]
Projects under development in Indian Prairie; Taylor 46-48
Creek/Nubbin Slough; Lake Istokpoga; Upper Kissimmee3
Additional P reduction needed to meet TMDL 221-223
1
Data Source: Table 8-2 (SFWMD, 2015) numbers include measured loads plus atmospheric deposition contribution
2
Data Source: Table 24 (FDEP, 2014a)
3
Data source: Table 22 (FDEP, 2014a)
To reduce P loads to Lake Okeechobee by the additional approximately 220 metric tons P per
year needed to achieve the TMDL approximately 27,000 acres (11,000 ha) of STAs and
associated FEBs would be needed, assuming an average total P removal of 20 kg/ha (18
lb/acre) P per year. The long-term (21 year) average P removal of STA-1W (south of the lake) is
estimated to be approximately 10 kg/ha (9 lb/acre) P per year with an average effluent quality of
< 50 ug/L (SFWMD, 2015). The P removal in pilot scale STAs in Taylor Creek/Nubbin Slough
(north of the lake) is estimated between 16-21 kg/ha (14 to 19 lb/acre P) per year. North of the
lake P load reduction is more important than P concentration in the outflow water, in contrast to
STAs south of the lake where outflow concentrations are regulated by the WQBEL. With
appropriate modifications in design and management strategies, including the addition of FEBs,
P removal rates for STAs north of the lake are likely to achieve 20 kg/ha (18 lb /acre) P per
year.
The Phase 1 BMAP projects described above are expected to achieve a total P reduction of
approximately 150 metric tons per year when completely implemented, out of a total required
reduction of approximately 370 metric tons per year (Table IV-2). Additional strategies such as
new field-verified BMPs that protect water quality, in situ technologies that immobilize legacy P
76
(e.g. by chemical amendments) and additional FEBs and STAs will be needed to achieve the
Lake Okeechobee TMDL. FEB-STAs, based on existing information, are a logical choice for
enhanced treatment. Approximately 27,000 additional acres of STAS and associated FEBs
would be needed to reduce P loads by 220 metric tons per year.
Northern Everglades - Caloosahatchee Estuary BMAP
The Caloosahatchee Estuary has been determined to be impaired by nutrients, and in 2009 the
FDEP adopted the Caloosahatchee Estuary TMDL for total nitrogen (TN), which has been
linked to high chlorophyll-a concentrations. Current estimated TN loading to the Caloosahatchee
estuary is 11,490,281 lbs (5,212 metric tons) per year (FDEP, 2012). Approximately 85% of the
TN load (9,736,039 lb/year, 4416 metric tons/year) is contributed by basins upstream of S-79
and 15% (1,754,242 lb/year, 796 metric tons/year) is contributed by basins downstream of S-79
(FDEP, 2012). The Caloosahatchee Estuary BMAP and its associated projects are only
intended to address TN reductions in the basins downstream of S-79. Because domestic
wastewater loads are estimated to be less than 1% of the TN loading downstream of S-79, the
BMAP focuses on stormwater load reductions. TN sources above S-79 will be addressed by
other efforts.
The Caloosahatchee Estuary Basin BMAP was adopted in 2012 (FDEP, 2012), following
completion of a stakeholder driven processes that was focused on identifying projects that had
been constructed since 2000, or are planned to be built within the first five years after BMAP
adoption (2012-2017). These projects include structural urban stormwater projects, street
sweeping, public education and fertilizer ordinances, conservation land purchases, hydrologic
restoration and agricultural BMPs. Overall, the first five-year iteration (2012-2017) BMAP
projects proposed by stakeholders are expected to reduce stormwater TN loads by 155,488 lbs
(71 metric tons), or approximately 40 percent of the 388,719 lbs (176 metric tons) per year
stormwater TN reduction required to achieve the TMDL downstream of S-79. Estimated load
reduction as of November 2013, including those projects given credit before BMAP adoption,
was 140,465 lbs (64 metric tons) per year of TN (Figure IV-16).
77
Figure IV-16. Progress toward the Caloosahatchee Estuary Total Nitrogen TMDL through November 30, 2014. The
first bar is the starting load for urban and agricultural stormwater runoff in the basin. The second bar shows the
current estimated loading after the implementation of proposed BMAP projects. The third bar is the total load
allowed for urban and agricultural runoff to meet the TMDL (Source: FDEP, 2014c) .
The Caloosahatchee River Watershed Protection Plan (SFWMD, 2009b) identified a

combination of watershed storage and water quality projects to improve the quality, timing and
distribution of flows throughout the system. In addition to urban and agricultural BMPs, the plan
recommended construction of approximately 15,000 acres of STAs, for a total potential
reduction of 1,840 metric tons (approximately 4,000,000 lbs) of TN per year and 166 metric tons
(361,000 lbs) of P per year (SFWMD, 2009b). If this plan were implemented, it would remove a
significant portion (35%) of the TN loads from both upstream and downstream of S-79; however,
no STAs are currently designed or sited in the Caloosahatchee basin. The first stage of the C-
43 Water Quality Treatment and Test Facility, intended to investigate and test wetland treatment
for removing TN and other constituents, is anticipated to begin in fiscal year 2015. In addition, a
540 acre site to test the feasibility of using Floating Aquatic Vegetative Tilling to remove TN is
operating in the East Caloosahatchee Watershed. At present there is no requirement to reduce
to P loads to the Caloosahatchee estuary.
78
Northern Everglades - St. Lucie River and Estuary BMAP
Long-term St. Lucie River Watershed TN and TP loads (WY1997-WY2014), exclusive of Lake
Okeechobee contributions, are estimated at approximately 2,429,832 lbs (1,102 metric tons) TN
per year and 597,552 lbs (271 metric tons) TP per year, respectively (SFWMD, 2014b). The
TMDL requires a reduction to 1,136,633 lb (515 metric tons) TN per year and 127,016 lbs (58
metric tons) TP per year (Figure IV-17). Lake Okeechobee loads are assumed to be addressed
by the Lake Okeechobee BMAP.
The St. Lucie River and Estuary Basin BMAP was adopted in 2013 (FDEP, 2013) following
completion of a stakeholder-driven process that was focused on identifying projects that have
been constructed since 2000, or are planned to be built within the first five years after BMAP
adoption (2013-2018). The first iteration BMAP projects include agricultural BMPs, public
education and fertilizer ordinances, street sweeping, and urban stormwater and wastewater
projects. These first iteration projects are expected to achieve reductions of approximately
316,024 lbs (143 metric tons) N per year and 121,250 lbs (55 metric tons) P per year (Figure IV-
17). It is estimated that total load reductions as of June 2014 were 477,789 lbs/year (217 metric
tons) TN and 130,542 lbs/year (60 metric tons) TP, exceeding the first iteration BMAP target.
Nevertheless these load reductions represent only approximately 36% (TP) and 28% (TN) of the
TMDL required reductions, thus additional nutrient reduction strategies are required to meet the
TMDL.
In addition to these BMAP projects the CERP C-44 project includes construction of a 7,300 acre
STA with an effective treatment area of 6,300 acres that is expected to remove approximately
180,780 lbs (82 metric tons) TN per year and 57,320 lbs (26 metric tons) TP per year (SFWMD,
2015). These removal estimates appear to be based on the P removal rate of approximately 10
kg /ha per year that has been achieved south of the lake. However, with effective management
this STA can likely be loaded at a higher rate than those south of the lake, and thereby achieve
an increased total P removal of up to 20 kg/ha per year. There is one Hybrid Wetland Treatment
Technology test facility currently operating in the St. Lucie watershed and one is under
development.
Based on the assumptions made for the additional STAs required for the Lake Okeechobee
Watershed (i.e. removal of 20kg P/ha per year), approximately 16,000 acres (6400 ha) of STAs
(beyond the C-44 STA and current BMAP projects) are needed in the St. Lucie watershed to
remove the additional 284,000 lb (128 metric tons) TP per year required to meet the TMDL.
Based on nitrogen to phosphorus load ratios in the St. Lucie basin, this acreage of STAs will
remove approximately 1,280,000 lb (580 metric tons) total N per year.
79
Figure IV-17. Progress toward the St. Lucie River and Estuary total nitrogen TMDL through June 30, 2014.
(Source: FDEP, 2014b).
Figure IV-18. Progress toward the St. Lucie River and Estuary total phosphorus TMDL through June 30, 2014
(Source: FDEP, 2014b).
Current and planned BMAP projects in the Caloosahatchee and St. Lucie basins will not reduce
nutrient loads to the level needed to achieve their respective TMDLs. Rather, in the first five
years of BMAP implementation the FDEP expects (SFWMD, 2015):
80
 Modest improvements in water quality trends in the watershed tributaries as well as each
estuary (St. Lucie and Caloosahatchee);
 Decreased loading of the applicable target pollutants (TP, TN, and BOD);
 Increased coordination between state and local governments and within divisions of
local governments in problem solving for surface water quality restoration;
 Determination of effective projects through the stakeholder decision-making and priority-

setting processes;
 Enhanced public awareness of pollutant sources, pollutant impacts on water quality, and
corresponding corrective actions;
 Enhanced understanding of basin hydrology, water quality, and pollutant sources.
Additional strategies such as new field-verified BMPs that protect water quality, in situ
technologies that immobilize legacy P (e.g. by chemical amendments) and additional FEBs and
STAs are needed. Approximately 15,000 acres of STAs are needed in the Caloosahatchee
River Watershed for the potential reduction of 1,840 metric tons of total N and 166 metric tons of
total P per year. Approximately 16,000 acres of STAs are needed in the St. Lucie River
Watershed for potential reduction of 130 metric tons of total P per year and 585 metric tons of
total N per year. The N to P load ratio from the Caloosahatchee River Watershed is
approximately 10 as compared to a ratio of 4 to 5 for St. Lucie River Watershed. These ratios
suggest that the St. Lucie River Watershed has much more legacy P than the Caloosahatchee
River Watershed.
Southern Everglades - Restoration Strategies
The Everglades Forever Act (EFA; Section 373.4592, F.S.) required that strategies to decrease
nutrient loads in water discharged downstream be developed for both the Everglades
Construction Project (ECP) and non-Everglades Construction Project (non-ECP) basins in the
Southern Everglades. As a result, 57,000 acres of constructed wetlands, known as Stormwater
Treatment Areas (STAs), were built on former agricultural lands at the interface of the EAA and
the WCAs to reduce excess TP from surface waters prior to discharging that water into the
Everglades Protection Area (EPA) (Figure IV-19). The STAs currently in operation include: STA-
1E; STA-1W; STA-2; STA-3/4; and STA-5/6.
81
Figure IV-19. Map showing the Everglades Agricultural Area Basins and C-139 Basin, and associated Stormwater
Treatment Areas (SFWMD, 2015).
As discussed in Section II.5, in response to USEPA’s 2010 “Amended Determination” in the

federal litigation over EPA’s approval of Florida’s water quality standards, SFWMD, USEPA,
and FDEP established a Water Quality-Based Effluent Limit (WQBEL) for TP in discharges from
STAs order to achieve compliance with the 10 μg P/L long-term geometric mean standard in the
EPA. The WQBEL requires that the flow-weighted mean (FWM) TP concentrations at STA
discharge points not exceed (1) an annual FWM of 13 μg/L in more than three out of five years
or (2) an annual FWM of 19 μg/L in any one year (SFWMD, 2012). To achieve the WQBEL in
2012 the State of Florida developed a plan known as the “Restoration Strategies Regional
Water Quality Plan” (Restoration Strategies, SFWMD, 2012). Restoration Strategies included a
suite of projects to improve water quality and a 10-year, $50 million research program to
investigate critical factors that regulate the sustainable removal of phosphorus by STAs.
The planned suite of Restoration Strategies projects includes more than 6,500 acres of new
STAs and 110,000 acre-feet of additional water storage in FEBs. These projects are located in
the Eastern Flow Path (STA-1E and STA-1W), Central Flow Path (STA-2, and STA3/4), and
Western Flow Path (STA-5 and STA-6) (Figure IV-20).
82
Figure IV-20. Phosphorus loads to western, central, and eastern flowpath (Data source: SFWMD, 2012).
The Eastern Flow Path will receive a 45,000 acre-ft FEB adjacent to L-8 canal to attenuate peak
flows and optimize hydraulic loading rates to STA-1W and STA-1E. The addition of
approximately 6,500 acres near STA-1W will increase total treatment capacity to 422,300 acre-ft
of runoff water according to modeling results using the Dynamic Model for Stormwater
Treatment Areas (DMSTA) model (Walker and Kadlec, 2011) and the peer-reviewed South
Florida Water Management Model (SFWMM). In the Central Flow Path, a 54,000 acre-ft FEB
adjacent to the North New River Canal will attenuate peak flows and optimize hydraulic loading
rates to STA-3/4. In addition, approximately 800 additional acres for STA-5 will increase total
treatment capacity to 877,300 acre-ft of water according to modeling results using DMSTA and
SFWMM. The Western Flow Path will receive an 11,000 acre-feet FEB that will attenuate peak
flows and optimize hydraulic loading rates for a total of 197,600 acre-ft of runoff from the C-139
Basin to STA-5/6 (DMSTA and SFWMM). In total, current plans include FEBs and STAs with
the potential to treat approximately 1.5 million acre-ft of water, with the DMSTA model predicting
long-term, FWM outflow concentrations of 12 µg/L, which suggests compliance with the WQBEL
target.
The Science Plan will investigate critical factors that influence phosphorus treatment and
performance, especially at low levels of total P. Specific investigations will include the effects of
microbial activity, phosphorus flux, inflow volumes and timing, inflow phosphorus loading rate
83
and concentrations on phosphorus outflow, phosphorus removal by specific vegetation
speciation, and the stability of accreted phosphorus. The DMSTA model does not currently
consider internal dynamics (such as short circuiting, vegetation die-off, internal regeneration of
P, and extreme events), but research conducted as part of the Science Plan could provide
information to include internal dynamics in the model. The combination of FEBs and STAs show
promise for meeting the WQBEL requirements. The FEBs should help to attenuate peak
stormwater flows prior discharge into STAs, thus improving the long-term sustainability of STAs
to improve the water quality. New approaches (including improved models) developed as a
result of this effort may provide additional innovative strategies to improve the performance of
STAs to achieve the WQBEL.
4. Summary
In the context of the Everglades restoration, water storage and water quality issues are
inextricably linked. Along with increased storage, increased treatment capacity is needed to
achieve the Lake Okeechobee, St. Lucie and Caloosahatchee TMDLs, and if the stored water is
to be moved south from Lake Okeechobee and discharged from the STAs into the EPA, it must
comply with the established WQBEL. Legacy P in the Lake Okeechobee watershed is of
particular concern because current efforts to achieve the Lake Okeechobee TMDL have proven
inadequate. None of the current BMAPs for the Lake Okeechobee, St. Lucie or Caloosahatchee
watersheds will achieve their respective TMDLs within the next 5 years. Therefore, additional
controls, such as FEBs, STAs, and aggressive BMPs that include in-situ immobilization of
legacy P by chemical amendments, will be needed to meet TMDL targets. These shortcomings
have even more substantial consequences for the Caloosahatchee River and St. Lucie River
estuaries because the BMAPs established for both of these systems assume that the targets for
Lake Okeechobee have been met. Furthermore, substantially more FEB-STA treatment
capacity will be required to move water south from Lake Okeechobee to the EPA if Lake
Okeechobee’s TMDL is not met.
To achieve water quality standards in Lake Okeechobee, the St. Lucie estuary, and the
Caloosahatchee estuary, more aggressive BMAPs must be developed. New field-verified BMPs
that protect water quality, advanced in situ technologies that immobilize legacy P, and the
strategic placement of FEB-STAs in the Northern Everglades priority basins will be essential to
achieve water quality targets. Beyond existing and planned approaches, the substantial
reservoir of legacy P in the Northern Everglades watersheds will necessitate new and more
effective strategies to combat the mobility of P. Furthermore, as discussed in Section III, the
funding and completion of the FEBs and STAs specified by Restoration Strategies should be
accelerated to achieve compliance with the EPA WQBEL more quickly.
84
V. Options (beyond approved projects) to Reduce High Volume
Freshwater Flows to the Estuaries and Move More Water South
from Lake Okeechobee to the Southern Everglades
1. Introduction
Repeated planning exercises, described in Section III, have shown that large volumes of inter-
annual storage are required north, south, east and west of Lake Okeechobee to manage Lake
Okeechobee levels within a desirable range, reduce damaging high and low flows to the St.
Lucie and Caloosahatchee estuaries, and move more water south for agricultural, urban and
environmental uses. Furthermore, as described in Section IV, to discharge this stored water to
Lake Okeechobee and the St. Lucie and Caloosahatchee estuaries in compliance with their
respective Total Maximum Daily Loads (TMDLs), and to the Everglades Protection Area (EPA)
in compliance with its Water Quality Based Effluent Limit (WQBEL), significant additional
treatment will be required. The red star on Figure V-1 shows the restoration performance of the
system which can be expected after the completion of the Kissimmee River Restoration (KRR),
and the construction of the C-43, C-44, Restoration Strategies and Central Everglades Planning
Project (CEPP) projects, based on the River of Grass modeling study (SFWMD, 2009d). The
KRR project is expected to attenuate peak flows to Lake Okeechobee; the C-43 and C-44
projects are expected to significantly reduce local-basin triggered high flows and improve dry
season flows to the St. Lucie and Caloosahatchee estuaries; and the Restoration Strategies and
CEPP projects together are expected to increase the delivery of clean water to the EPA to an
average of 1.5 million acre-ft per year. Nevertheless, Figure V-1 indicates that even after these
projects are constructed, lake-triggered high discharges to the St. Lucie and Caloosahatchee
estuaries will be reduced by less than 55%, and less than 75% of the dry season Everglades
demand target will be delivered to the EPA over the 41-year simulation period. Two possible
configurations that provide a 90% reduction in lake-triggered discharges and delivery of 90% of
the dry season Everglades demand are shown with green stars on Figure V-1. These two
example configurations require approximately 1) 750,000 acre-ft new storage north of the lake
and 132,000 acre-ft additional storage south of the Lake (i.e. beyond the CEPP/Restoration
Strategies Flow Equalization Basin (FEBs)), or 2) 300,000 acre-ft storage north of the lake and
507,000 acre-ft additional storage south of the Lake.
The following sections present options for additional efforts, beyond the approved projects, to
provide the additional storage and treatment required to more fully achieve the restoration
objectives of reducing lake-triggered high discharges to the estuaries and moving more water
south.
85
a)
b)
Figure V-1: a) Percent Reduction in Lake Triggered High Discharges to the St. Lucie and Caloosahatchee estuaries
(top) and b) Percent Achievement of Dry Season Everglades Demand Target (bottom) by alternative storage
configurations (adapted from SFWMD, 2009d) . System performance after Restoration Strategies and CEPP add
112,000 ac-ft of storage south of the lake is indicated by the red star. Green stars indicate two possible future
configurations that would achieve 90% restoration.
86
2. Storage and Treatment North of Lake Okeechobee
Figure V-1, produced as part of the River of Grass planning process (SFWMD, 2009d), shows
that storage can be effective at reducing damaging discharges to the St. Lucie and
Caloosahatchee estuaries whether it is constructed north or south of the lake. Storage north of
the lake is effective for managing lake levels within a desirable range and thus reducing
damaging discharges to the estuaries. Furthermore, water storage and treatment is needed
north of the lake to meet the Lake Okeechobee TMDL. However, due to the extended time it
takes to route water from north of the lake to the Water Conservation Areas (WCAs), northern
storage is not likely to be as effective as southern storage in meeting the timing and distribution
objectives of water deliveries to the EPA. Furthermore, it is likely that water stored north of the
lake, if passed through the Lake or through perimeter canals subject to agricultural runoff, may
need to undergo additional water quality treatment to meet applicable standards before it is
released to the EPA. Thus, the additional required storage will be needed to be distributed both
north and south of the lake to achieve all restoration objectives. The green asterisks on Figure
V-1 indicate that between 300,000 and 750,000 acre-ft of additional storage north of the lake will
improve achievement of restoration goals in the St. Lucie and Caloosahatchee estuaries and
the Everglades considerably. Options to provide additional storage north of the lake include
combinations of ASR wells, deep storage reservoirs, shallow storage impoundments, and
dispersed water management. The advantages and disadvantages of each of these types of
storage are briefly summarized below. Options to treat the stored water before it is released to
Lake Okeechobee to help achieve the TMDL are discussed in Section IV.
Aquifer Storage and Recovery (ASR):
ASR wells inject and store excess water underground in the wet season with the goal of
extracting the water in subsequent dry seasons for water supply. Advantages of ASR wells
include 1) in sufficient numbers they are able to provide for substantial inter-annual storage, 2)
they require a minimal land surface footprint, and 3) they have been shown to provide a
reduction in phosphorus concentrations in recovered water in CERP pilot studies.
Disadvantages of ASR wells include 1) high operation and maintenance costs due to pumping
and required treatment prior to injection (filtration and UV disinfection), 2) highly colored water
may reduce UV disinfection performance to below regulatory requirements for injection, and 3)
recovery efficiency is site specific depending on local hydrogeological characteristics (e.g.
~100% at the Kissimmee River Pilot site, ~ 20-40% at the Hillsboro Pilot site, and not feasible at
the Caloosahatchee River Basin Pilot site) For more details on the ASR pilot study see Section
III.3.c.
87
Surface Reservoirs:
Above-ground storage reservoirs are used to capture and hold normal and peak flows during
wet seasons. Water is then discharged during dry seasons when flows are needed for urban,
agricultural or natural system uses. Water depths in above-ground reservoirs proposed to date
typically range from 4-12 feet, with vegetation management and dam safety concerns being the
limiting factors. Deep 12-ft storage reservoirs have relatively high construction costs
(approximately $9,900 per acre-ft, excluding land costs estimated in the CEPP planning effort)
compared to shallow 4-ft water impoundments (approximately $4,900 per acre-ft, excluding land
costs estimated in CEPP planning effort) due to additional dam safety requirements. Both deep
reservoirs and shallow impoundments are operationally flexible and offer the potential to
improve the timing and distribution of water to the natural system. Storage reservoirs and
shallow impoundments are allowed to experience dry-outs during extended drought periods and
are not intended to provide substantial fish and wildlife habitat value or water quality treatment
capability. In fact, reservoirs and impoundments may cause water quality problems, such as
algae blooms when water is held for substantial periods of time and may release soil
phosphorus following dry-outs. Deep reservoirs are advantageous in that they have a reduced
land footprint, e.g., a 12-ft reservoir requires one-third of the land that a 4-ft impoundment
requires for the same amount of storage. Shallow reservoirs are advantageous in that they are
not required to meet dam safety standards, however, they have the potential for higher
operation and maintenance issues related to exotic vegetation management within large shallow
footprints. Using the construction costs estimated for the CEPP planning effort, and assuming
sufficient land is available for purchase, shallow storage will be cheaper than deep storage on a
per acre foot basis as long as land costs remain below $30,000 per acre.
Dispersed Water Management:
Dispersed water management (DWM) provides short-term (intra-seasonal) local water retention,
peak flow attenuation, and onsite hydrologic restoration, typically on private lands or in publicly
owned wetlands management or wildlife refuge areas. Advantages of DWM include that it 1)
improves natural habitat, and 2) it has the potential to increase local storage and thus attenuate
the magnitude of peak flows to the lake, 3) it has the potential to increase evapotranspiration
and thus reduce total flows to the lake, and 4) it provides economic incentives to ranchers and
citrus growers to maintain the land in its current use rather than selling the land or converting it
into more intensive uses which have the potential to create more nutrient loading and
exacerbate water quantity issues. Disadvantages of DWM include 1) it is a non-permanent,
land-owner implemented solution with associated monitoring and maintenance issues, 2) there
is high uncertainty related to long-term (inter-annual) storage benefits because stored water will
likely evaporate or infiltrate within the season and thus not be available for use in subsequent
droughts, 3) there is high uncertainty in water quality treatment capability due to the potential for
increased phosphorus release upon re-wetting after dry-downs, and 4) limited modeling tools
88
currently exist to evaluate hydrologic and water quality performance. A regional data collection
and modeling study, similar to that being conducted for ASR (see Section III.3.c above), is
needed to evaluate the cumulative impact of a regional DWM system north of the Lake on the
quality, quantity and timing of flows Lake Okeechobee, and its potential for inter-annual storage
of water that could be used in subsequent dry years, as a function of climatic conditions, spatial
location and density of DWM features on the landscape, and operation of the regional canal
system.
Summary:
The Lake Okeechobee Phase II Technical Plan and the River of Grass Planning Process
provide a sound foundation from which to plan, design, and build the additional storage and
treatment needed north of Lake Okeechobee. To determine the most cost-effective means of
providing the required storage and treatment a strategic planning exercise should be conducted
north of Lake Okeechobee similar to the River of Grass Planning Process conducted south of
the lake. The goal of this exercise should be to determine the best combinations and locations,
and logical phasing of the various types of storage that will provide the desired benefits to the
lake, St. Lucie and Caloosahatchee estuaries and EPA. This will require a regional modeling
effort that takes into account lessons learned and information gained since the CERP, NEEPP
and ROG planning exercises regarding the permitting requirements, engineering feasibility and
costs, and inter-annual storage benefits associated with ASR, deep storage reservoirs, shallow
water impoundments and dispersed water management. New data gathering efforts and model
developments will be required to simulate the cumulative impacts of a regional DWM system
north of the Lake on the quality, quantity, and timing of flows into Lake Okeechobee as a
function of climatic conditions, spatial location and density of DWM features on the landscape,
and operation of the regional canal system. The Technical Review Team expects that the
modeling study will show that, while DWM on private lands may provide some benefits,
providing the additional storage and treatment needed will require acquisition of additional land
north of the Lake (i.e. from approximately 25,000 acres if 300,000 acre-ft is provided by deep
storage, up to approximately 187,500 acres if 750,000 acre-ft is provided by shallow water
impoundments).
3. Additional Storage, Treatment and Conveyance South of Lake Okeechobee
In addition to the north of lake storage discussed above, the green asterisks on Figure V-1
indicate that provision of between 132,000 and 507,000 acre-ft of additional storage, treatment
and conveyance south of the lake (i.e. in the EAA) will improve achievement of restoration goals
for both the St. Lucie and Caloosahatchee estuaries and the Everglades considerably. Storage
areas within the EAA have the advantage of being able to store excess water from within the
EAA basin and upstream sources (i.e. Lake Okeechobee and its inflow sources). Because of
the existing canal system in the EAA, storage located between the Miami River and North New
89
River Canals is strategically located to store excess runoff from significant portions of the EAA
basin and makes maximum use of existing infrastructure. Storage and treatment in the EAA can
effectively be used to meet Everglades targets, presuming the construction of new
outflow/delivery infrastructure within the EPA. Options to provide additional storage south of the
lake include combinations of deep storage reservoirs, shallow water impoundments, wet flow-
ways, and dry-flow-ways. Each of these storage options will also require additional STA acreage
to provide the required water treatment. In the next section we review past evaluations of flow-
way and flow-way like plans and evaluate their efficacy and feasibility to provide significant
restoration benefits in the current, highly modified regional ecosystem.
a. Plan 6 and Other Flow-way Options
Introduction
One concept for restoring more flow to the Everglades, which was first identified in a report of
the Jacksonville District of the US Army Corps of Engineers (USACE, 1955a) and referred to as
‘Plan 6’ (Figure V-2) is a ‘flood-way from Lake Okeechobee to Conservation Area 3 … that …
would discharge by gravity with a spillway structure at the lake end to regulate discharge.’ In this
report the USACE noted that ‘benefits from provision of the flood-way (flow-way) would consist
of reduction in maximum lake stages and hence reduction of the amount of water diverted
through the St. Lucie Canal, which would reduce damages along the lower St. Lucie River.’ The
report also noted that ‘water supply to Water Conservation Area 3 would not be available during
drought periods’ and that ‘discharge from the flood-way would cause water level in Conservation
Area 3 to rise during flood periods, necessitating increases in levee heights and additional water
control structures.’ In the report it also was noted that ‘a spillway would be required at the
centerline of the Lake Okeechobee levee to control discharge from the lake and, during
hurricanes, to limit discharges through the flood-way as required to prevent damages from wind
tides and waves.’ In that report the USACE identified the need for a spillway designed to handle
a maximum of 20.6 ft of water in the lake. The anticipated diversion capacity from the lake to the
south via the flood-way system was 4,800 cfs at lake stage 17.4 ft and 16,800 cfs at lake stage
20.6 ft.
The Technical Review Team was unable to secure documentation as to why this particular plan
was not carried forward. Nevertheless, it is clear that engineers at that time were working under
a very different set of conditions than exist at present. First, they assumed that water could flow
across the EAA by gravity. Today, due to soil subsidence, this is no longer possible and would,
in fact, require a more engineered configuration that includes reservoirs and pumps to move
water south over that same expanse of land. Second, the engineers envisioned a lake reaching
very high levels (up to 20.6 ft) that today we know would negatively impact the lakes ecology,
the littoral zone in particular (Havens, 2002; Havens and Gawlik, 2005).
90
Figure V-2. Plan 6, a flow-way from Lake Okeechobee through the EAA to the Everglades, as illustrated in USACE
(1955b). Note that north is to the left in the orientation of this illustration.
91
The USACE Reconnaissance Report
The US Army Corps of Engineers performed an initial evaluation of Everglades restoration plans
in the early 1990s, and provided results in a Reconnaissance Report (USACE, 1994). That
study emphasized that the aim was to formulate restoration plans ‘with the premise that the
hydrological restoration would create a system that could function similarly to the way it
functioned under pre-drainage conditions’ (the pre-drainage system was a continuous broad
flow path from the Upper Kissimmee to Florida Bay).
One of the plans considered in the Reconnaissance study was Plan 6. In this case, there was
not a spillway at the lake, and the flow-way started two miles north of the Bolles Canal, varying
in width (east-west) from 13.1 miles to 7.2 miles, and with a length of 22.5 miles (Figure V-4). It
was bounded on the west and east by the Miami and North New River Canals, respectively.
Water would be introduced by two pump stations – with capacities of 4,170 cfs (from the Miami
Canal) at the northwest corner of the flow-way and 2,460 cfs (from the North New River Canal)
at the northeast corner of the flow-way. The capacity of the flow-way itself to deliver water south
was estimated at 6,600 cfs.
During the Reconnaissance study, there were public workshops where preliminary results of
restoration plans were presented. It was noted in the Reconnaissance Report that there was
strong opposition to Plan 6, however, it was kept in the mix of plans for further analysis because
of its potential technical merits.
Relative to other plans evaluated as part of the Reconnaissance Report, Plan 6 allowed for the
largest spatial extent of wetlands (1,815,000 acres; RR Table 11, Page 202), the greatest
amount of sheet-flow (1,658,000 acres; RR Table 11, page 202) and the greatest amount of
storage capacity (8,777,000 acre-ft RR Table 11, Page 202). On the other hand, Plan 6 had a
projected implementation cost that was among the highest of all the plans considered ($592
million; RR Table 7, Page 192) and the highest equivalent average annual cost ($114 million;
RR Table 9, Page 196). Plan 6, because it involved unconstrained gravity flow, did not perform
as well as other more managed plans that aimed to restore natural flow patterns across the
Tamiami Trail (RR, page 177). Likewise, Plan 6 performed poorly with regard to its projected
community suitability index for the marl prairie habitat in the southern Everglades (RR Figure
18, Page 182). The plan was determined to have “good” performance with regard to reducing
water discharges to the St. Lucie and Caloosahatchee Estuaries, although not as “good” as
several other more engineered plans (RR Figure 18, Page 182).
In the Summary presented on RR Page 188 it was stated that one of the ‘most important
elements in the final plan to restore the Everglades will be … enlargement of conveyance
capacity from Lake Okeechobee through the EAA’ and in an accompanying table that listed
92
project elements recommended for consideration in the subsequent C&SF Restudy (page 229,
Table 12), a ‘flow-way within the area between Miami and North New River Canals’ was
included as one of those elements.
Figure V- 4 The EAA flow-way (stippled area on this map between the Miami and North New River Canals)
identified in Plan 6 of the USACE Reconnaissance Report (USACE, 1994).
The C&SF Restudy (CERP, 1999)
In spite of the recommendation provided in the Reconnaissance Report, Plan 6 was not
included as an alternative formally evaluated in the C&SF Restudy, for the following reason
(from Appendix B CERP, 1999):
“A flow-way is generally described as a broad shallow marsh area that is used to

flow water freely from Lake Okeechobee to one or more of the WCAs. The concept
includes creation of an Everglades-type environment having both storage and water
quality benefits.
Evaluation of the concept shows several erroneous assumptions about the

feasibility. Problems with soil subsidence, ET, seepage management, vegetation,
timing of flows, and lack of flow events is evident. Addressing other EAA issues
would be required if a flow-way cuts through and dividing the area, including
93
numerous roads, bridges, and railroad relocations. Soil subsidence in the EAA has
substantially reduced the hydraulic head that drives the southward flow of water;
hence, velocities and flow rates are greatly reduced. By spreading the water over
shallower areas (as opposed to reservoirs) and maintaining proper hydration of a
functioning marsh habitat, the ET loss could easily be doubled. A long, rectangular
configuration can have a 75 percent longer levee than a squared one, thus
increasing seepage management features.
Because nutrient-laden soil would be flooded for the flow-way, cattails would most
likely dominate the vegetation and not the desirable Everglades habitat. Flow-ways
would not be able to hold back water going to the WCAs. The continuous delivery of
that water would exacerbate the already high stages in the northern parts of the
WCAs.
Thus, the timing of flows from flow-ways would not be manageable or beneficial for
the remaining Everglades. Perhaps the most crucial element–water flowing from the
Lake to the WCAs–is not present in dry or even normal years! For example, during
the long periods from 1970-1982 or 1985-1994, no significant excess Lake water
was available for the flow-way. Only demand releases to the Everglades were made
from the Lake during those periods. Water delivered to the Everglades on a demand
basis through a flow-way would not be effective because of increased travel times
and increased ET losses. The only years where water could flow for a long duration
are wet periods similar to 1969-1970, 1982-1983, and 1994-1995. In those years,
the stages in the WCAs are already too high and additional flow from flow-ways
would be damaging, not beneficial.
Summarizing, the flow-way is a concept that creates a water supply burden on the
system without clear hydrologic benefits. The need for flow-ways would have to be
justified for other reasons rather than hydrology alone.”
One major difference in the EAA region where the Plan 6 flow-way was sited between 1955,
when the USACE first developed the concept, and the time of the C&SF Restudy is the large
amount of subsidence in the EAA due to loss of peat (Aich et al., 2013). Today’s EAA is much
like a shallow basin and water removal during high rainfall periods requires pumping. A gravity
flow-way in the EAA would no longer have a north to south elevation gradient, but rather a drop
and then increase in elevation, resulting in an area that would hold water, but not passively
allow it to flow south. This situation conflicts directly with one of the major concepts of a flow-
way as a near-natural solution that could depend on gravity flow and not require the use of
energy (diesel, electricity, etc.) to move water south.
94
South of Lake storage and conveyance alternatives identified in the River of Grass
Planning Process
As discussed in Section III.1.c, in 2008 the SFWMD had an option to purchase a large amount
of land in the Everglades Agricultural Area (EAA) from willing sellers, with the possibility to carry
out land trades to achieve a contiguous corridor through the EAA for conveyance of water to the
Everglades. In facilitated meetings, various stakeholders were asked to provide concepts for
how land in the EAA could be used to construct storage, treatment and conveyance projects
that would reduce harmful discharges of water to the estuaries and increase the amount of
freshwater that could be delivered to the Everglades. The peer-reviewed RESOPS model
(SFWMD, 2009c) was used to evaluate various stakeholder proposed alternatives with regard to
their effectiveness in meeting a set of standard performance measures, including the percent
reduction in high lake-triggered discharges to the St. Lucie (>2000cfs) and Caloosahatchee
(>2800cfs) estuaries and the ability to provide the target Everglades dry season demand.
Calculations also were carried out for each alternative regarding the amount of STA acreage
(and cost) that would be needed to meet Everglades P standards, under conditions of Lake
Okeechobee outflow water ranging in concentration from 40 to 200 ppb TP.
The River of Grass stakeholder alternatives (Table V-1, SFWMD 2009e) included various
configurations of deep and shallow reservoirs, wet and dry flow-ways, widened and reconfigured
EAA canals, and STAs. The Everglades River of Grass Northern Expansion option (Figure V-5),
included large wet flow-ways through the EAA with no engineered reservoirs or STAs, quite
similar in concept to Plan 6 considered in the USACE Reconnaissance Report (USACE, 1994).
As indicated in Table V-1, alternatives that performed best for the estuaries were the most
engineered and required large, deep storage. For example, the Chain of Lakes alternative,
which included a number of inter-connected reservoirs (Figure V-6), reduced damaging
discharges to the estuaries by 94% with 500,000 acre-ft of storage north of Lake Okeechobee.
In contrast, the Everglades River of Grass Northern Expansion alternative (Figure V-5), with its
large passive flow-ways in the EAA and approximately the same volume of storage north of the
lake (550,000 acre-ft), reduced damaging discharges by just 77%. The River of Grass Northern
Expansion alternative also had considerably poorer performance than the more engineered
Chain of Lakes alternative in regard to meeting the Everglades dry season water demands. In
fact, the River of Grass Northern Expansion alternative was one of the least effective plans
evaluated in regard to helping the estuaries with the issue of regulatory discharges or the
Everglades with the issue of too little water in the dry season. Thus similar results, from an
entirely different modeling approach, mirrored those in the USACE Reconnaissance Report
(1994), i.e. unmanaged gravity flow of water south of Lake Okeechobee is less beneficial to the
estuaries and Everglades than more managed scenarios.
95
It is notable that in the River of Grass planning process, an alternative called the Marshall Plan
Element 6 (not to be confused with Plan 6 of the USACE because of its very different
configuration) that included both a large storage reservoir and a large wet flow-way (Figure V-7)
in the EAA had good performance for the estuaries (95% high flow reduction) and the
Everglades (89 score, Table V-1). However, this plan included 650,000 acre-ft of storage north
of the lake, so it is not possible to discern whether the benefits came from that added northern
storage or the deep storage elements within the EAA.
96
Table V-1. Planning alternatives considered in the River of Grass planning process, with information on
the size of north-of-lake storage, a summary of the plan components and the performance measure results
for the estuaries (percent reduction in damaging regulatory releases) and Everglades (standardized score,
higher is better). The Everglades Benefit is the ability to deliver needed water south in the dry season
(SFWMD, 2009e).
Alternative Summary North of Lake Estuary Benefit Everglades Benefit

Storage (acre- ft)
Estuary Driven Deep storage, large 1,000,000 96% 98
Everglades dry flow-ways, below
Restoration ground canals and
STAs
Everglades River of Large gravity wet 550,000 77% 80
Grass Northern flow-ways in central
Expansion and northeast EAA
Chain of Lakes Large shallow lakes 500,000 94% 91
connected by wide
canals and an
ecoslough
Florida Crystals Large, dry flow-way 500,000 76% 73
and large STAs
Restoration Plus Large deep storage 300,000 95% 96
Employment and STAs
Marshall Plan Large, deep storage, 650,000 95% 89
Element 6 managed wetland,
large, wet flow-way,
and large STAs
Performance Large, deep storage 200,000 95% 95
and large STAs
Performance – Cost Large, deep storage, 300,000 94% 91
Plan large, shallow
storage, large STAs,
wide below ground
conveyance canals
Reservoir within Large deep reservoir 0 93% 86
Lake Okeechobee within Lake
Okeechobee,
dispersed storage
west of Lake
Okeechobee wide
below ground
conveyance canals,
STAs
97
Figure V-5. The ‘Everglades River of Grass Northern Expansion’ alternative in the River of Grass planning process did not yield favorable outcomes with regard
to reducing damaging high regulatory water discharges from Lake Okeechobee to the estuaries, or delivering water south to the Everglades. This alternative
included 550,000 acre-ft of north-of-lake storage (from SFWMD ROG planning process).
98
Figure V-6: The Chain of Lakes alternative in the River of Grass planning process that had good performance with regard to reducing damaging high discharges
of water from Lake Okeechobee to the estuaries. This alternative included 500,000 acre-ft of north-of-lake storage (from SFWMD ROG planning process).
99
Figure V-7. The Marshall Plan Element 6 alternative in the River of Grass planning process included a large reservoir south of the lake and a large flow-way.
This alternative scenario had good outcomes for the estuaries and Everglades, however it included 650,000 acre-ft of storage north of the lake (from SFWMD
ROG planning process).
100
Summary
The concept of creating a flow-way to carry water from Lake Okeechobee through the EAA to
the WCAs has existed since 1955. While it may seem intuitive that re-establishing a broad flow-
way from near Lake Okeechobee to the northern Everglades is a sound restoration strategy,
independent assessments indicate that modifications of the landscape have, to a large degree,
compromised options to do so. In both the USACE Reconnaissance Report (1994) and the
SFWMD River of Grass planning process (2009), results indicated that a passive EAA flow-way
is not the optimal approach for addressing problems of too much water going to the estuaries in
the wet season or too little water going to the Everglades in the dry season. Furthermore,
creation of a large wet flow-way in the EAA region creates additional wetlands restoration
objectives and additional water demands in a system that is already heavily constrained.
b. Other South of the Lake Storage, Treatment and Conveyance Options

Although an expansive passive wet flow-way throughout the EAA may not be the best solution,
the River of Grass planning process demonstrated that there are several possible options
involving combinations of deep and shallow storage, and wet- and dry- flow-ways, coupled with
STAs and enhanced conveyance to provide significant benefit for both the estuaries and the
Everglades, beyond CEPP (Figure V-1 and Table V-1). Figure V-1 indicates that if between
300,000 to 750,000 acre-ft of storage can be provided north of the lake, between 132,000 to
507,000 acre-ft of additional storage south of the lake will be sufficient to provide 90% reduction
in lake-triggered high flows to the estuaries and achieve 90% of the dry season target. If this
required storage were to be provided strictly though deep 12-ft reservoirs, new land area
between approximately 11,000 and 43,000 acres would be required south of Lake Okeechobee.
If the required storage were provided strictly through shallow 4-ft impoundments, the land area
requirement triples to approximately 33,000 to 129,000 acres.
To provide the required land area for additional storage, conveyance and treatment south of the
Lake there are a number of options that should be considered: 1) purchase of private land in the
EAA (including the current U.S. Sugar land purchase option or land from other willing sellers) to
use or trade for deep and/or shallow storage and treatment opportunities, 2) develop a cost-
sharing program for on-farm or sub-regional multi-farm shallow storage and treatment
opportunities, and 3) use existing state owned wildlife management areas such as Holey Land
and Rotenberger for storage and treatment opportunities. There is some evidence to indicate
that creating new storage and conveyance along the western boundary of the EAA, where Lake
Okeechobee water is cleaner and excess treatment capacity exists in STA 5/6, may be a
promising option.
101
Purchase of Private Lands
Currently, the state of Florida has an option to purchase approximately 46,000 acres in the EAA
(Figure V-8). The option is set to expire in October 2015. Thus, the state has a limited window of
opportunity to purchase this land at market prices. Given the limited opportunity and the
uncertainty of any future similar opportunities to purchase large acreages of lands in the EAA,
the state should consider this time-limited option. The particular 46,000 acres at issue may be
useful for additional storage and treatment or may serve as lands that the state could trade with
other agricultural interests in the area if land in different locations are needed.
Figure V-8. Map of U.S. Sugar Options Lands. Dark green-shaded lands have already been acquired. The option for
the brown-shaded lands expires in October 2015 (from SFWMD
http://www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/rog_map_2010_0804.pdf).
102
Develop a Cost-Sharing Program for on-farm or multi-farm storage
Another option for additional storage and treatment south of Lake Okeechobee is to increase
on-farm or sub-regional multi-farm storage opportunities through cost-sharing or Payment for
Environmental Services programs similar to those offered north of the Lake. Utilizing existing
farmlands, either individually or in the aggregate, it is possible that wet-season storage could be
constructed to provide water supply for agriculture during the dry season and thus reduce
competition between agricultural water demands in the EAA and water needs in the
Caloosahatchee estuary. The feasibility, costs and benefits of such a program should be
examined.
Convert Holey Land and Rotenberger Wildlife Management Areas to Storage and/or STAs
Holey Land and Rotenberger are designated Wildlife Management Areas (hereafter HLWMA
and RWMA) and constitute approximately 30,000 and 33,000 acres, respectively. Once part of a
vast sawgrass plain, these areas exist presently as impounded marshes situated just south and
west of the EAA and immediately adjacent to STAs 3/4 and 5/6 (Figure V-9). Because of their
location within a conceptual and historical flow path of water from Lake Okeechobee south,
these marsh lands have been suggested as potential water storage areas at various times in the
past. They are large, situated close to canals, and, from a geographic perspective, are centrally
located in the complex of lands in which water from Lake Okeechobee, and the EAA are
cleaned and stored.
103
Figure V-9: Map of Holy-Land and Rotenberger Wildlife Management Areas (from SFWMD).
These lands are owned by the State of Florida and have been under the management authority
of the Florida Fish and Wildlife Conservation Commission (FWC) (and its predecessor the
Florida Game and Freshwater Fish Commission) since they were established in 1968 (HLWMA)
and 1975 (RWMA). Both HLWMA and RWMA have been historically over-drained, and have
undergone considerable changes with regard to soil elevation and vegetation type, with a clear
movement toward more woody and shrubby vegetation. This is generally believed to be a
consequence of a reduced hydroperiod. Although degraded compared to pristine marsh, these
areas still maintain many characteristics of the sawgrass plain, and are probably restorable to
that condition with proper water management.
Both HLWMA and RWMA are currently managed by the FWC for recreation, hunting and wildlife
viewing. They are designated as Outstanding Florida Waters, and each is designated for
restoration as part of the Everglades Forever Act of 1994. Both HLWMA and RWMA are
identified as targets for ecological restoration in the CERP, however, they are not included in the
Everglades Protection Area.
These lands have previously been considered for use as water supply impoundments as part of
an effort to reduce agricultural back-pumping to Lake Okeechobee (see Barnett, 1986). A
feasibility report prepared by the SFWMD (Barnett, 1986) concluded that while impounded
water storage in HLWMA and RWMA could provide considerable dry season water supply to the
104
EAA, there was little effect of wet season storage, largely because both areas were already
flooded to capacity during the summer. As a consequence of opposition by environmental
groups and state agencies, the proposal to use HLWMA and RWMA for water storage was
voted down by the SFWMD in March 1983 and has not been formally revived since. Several
formal and informal plans that incorporate flow-way features south of Lake Okeechobee have
either included HLWMA and/or RWMA as part of the flow-way, for FEBs and/or for STAs,
however, none of these options have been included in any currently approved restoration plan.
As mentioned above, together the HLWMA and RWMA comprise approximately 63,000 acres,
and if flooded to a depth of 4 feet, could accommodate a maximum of approximately 250,000
acre-ft of storage. If flooded to a maximum depth of 12 feet, they could accommodate a
maximum of approximately 750,000 acre-ft While these storage figures are rough and there are
important engineering constraints that may push these numbers downward, this illustrates that
these lands have the potential to store a substantial amount of water.
However, there are a significant number of infrastructural and political obstacles that would have
to be overcome to use these lands. For example, neither HLWMA nor RWMA currently has the
infrastructure necessary to handle large volumes of inflows or outflows, and the dikes are
currently not designed to hold large volumes. Furthermore, the RWMA is situated on the
western edge of the former sawgrass plain, and water would need to be pumped to and away
from it to make that storage useful. Use of these WMAs for storage would destroy the current
vegetative structure. This was a clearly defined outcome identified in the Barnett (1986)
feasibility report, and all deeply impounded areas of former Everglades have shown similar
responses – loss of tree islands and plant diversity, lack of regeneration of sawgrass and trends
toward an open water system.
The use of HLWMA and RWMA as storage and treatment areas would be a reversal of the
restoration processes envisioned for these areas in CERP, and a conversion would result in a
net reduction of restorable Everglades habitat. If the entire area were used as deep storage,
these lands would become almost completely incompatible with their current designation and
usage. However, the areas outside HLWMA and RWMA currently being managed as STAs
have proven to be highly productive in terms of bird abundance and biodiversity. These
biological functions are compatible with certain kinds of hunting, birding and recreation, and it is
possible that these human activities could be accommodated in STAs within the HLMA and
RWMA. However, it should be clear that conversion to STA or storage would be incompatible
with restoration of these areas to anything like their original sawgrass plain habitat, and given
the investment in infrastructure, that recreation and hunting would probably become secondary
rather than primary objectives for STAs within HLWMA and RWMA.
105
As a result, the conversion of these WMAs to STAs or FEBs would need to clear significant
political and cultural hurdles, and require significant investment for infrastructure. The
designation of both areas as Outstanding Florida Waters (OFWs) may need to change if they
were being used explicitly to improve water quality. While this designation can be removed, it
would be a clear signal of change in goals and philosophy regarding the use of these state
lands.
All of the current STAs and FEBs have been placed on previously impacted (farmed) lands
purchased by the state for the purpose of cleaning water from the EAA. Using public recreation
lands, and lands that remain restorable to Everglades habitat for cleaning runoff from private
agricultural lands has in the past been forcefully opposed. One possible way to address this
complaint and potential legal hurdle would be to provide an unambiguous means of using
HLWMA and RWMA to only store and treat water from Lake Okeechobee, without any input
from the EAA. This would present a substantial infrastructure investment, since the existing
canals connecting the two WMAs to Lake Okeechobee can handle only small volumes at
present and pass through agricultural lands However, the capacity of STA 5/6 is thought to be
under-utilized with current water sources for the same reason, and creation of infrastructure to
bring Lake water to 5/6 could probably also be sized to serve HLWMA and RWMA.
In summary, the HLWMA and RWMA have the capacity to provide between 250,000 – 750,000
acre-ft of storage with associated STAs for water treatment. However, significant infrastructural
investment would have to be made to use either of these areas for storage or treatment
purposes and this usage would require re-designation of use, which has historically run into
strong public and agency opposition. Nevertheless, given the scarcity of available land from
willing sellers in the EAA, and the strategic location of HLWMA and RWMA on the boundary of
the EPA, it is an option that should be considered.
c. Summary
Based on results of the River of Grass Phase 1 screening analyses, and the finding that CEPP
provides only modest relief to the problems of high regulatory discharges to the estuaries, a
strategic plan should be developed for the next increment of south-of-the-lake storage,
treatment and conveyance to pursue beyond CEPP, to more closely meet the performance
targets of both the estuaries and Everglades ecosystems. Depending on the type of storage
provided, and how much storage can be constructed north of the lake, more closely achieving
the performance targets will require between 11,000 to 129,000 acres of additional land
between the lake and the EPA, e.g. the US Sugar option lands, lands from other willing sellers,
and/or use of existing state owned land (e.g. Holey Land and Rotenberger WMAs). There is
some evidence to indicate that creating new storage and conveyance along the western
boundary of the EAA, where Lake Okeechobee water is cleaner and excess treatment capacity
exists in STA 5/6, may be a promising option.
106
If sufficient inter-annual storage and treatment north of the lake is found to be economically or
politically infeasible, or the analysis indicates that the captured water cannot be efficiently
treated and conveyed south of the lake for use in subsequent dry seasons, the option of
constructing a system of large deep disposal wells to permanently dispose of excess flows from
Lake Okeechobee could be explored. While deep injection wells alone could not entirely prevent
damaging discharges to the St. Lucie and Caloosahatchee estuaries in an extreme wet season,
they could operate in conjunction with other storage elements to help reduce estuarine
discharges.
Deep injection wells involve disposing of fluids via injection wells deep below the earth’s surface
and have been used extensively in the State of Florida for more than 20 years (USEPA, 2005).
Deep injection wells are classified by the USEPA as belonging to one of five classes, depending
upon the nature of the fluid to be discharged and the depth of the well. The requirements for
siting, permitting, and monitoring, and the costs for construction and operation vary significantly
by well class.
Class I deep injection Wells are used to inject nonhazardous waste such municipal wastewater
effluent or stormwater runoff below the lowermost underground source of drinking water
(USDW). A USDW is defined as an aquifer that contains a total dissolved solids concentration of
less than 10,000 milligrams per liter. Class I injection wells must be constructed, maintained,
and operated so that the injected fluid remains in the injection zone, and interchange of water
between aquifers is prevented. According to the FDEP Underground Injection Control website
there are more than 180 active Class I wells in Florida (FDEP 2014d,
http://www.dep.state.fl.us/water/uic/).
Permitting requirements for Class I deep injection wells are generally easier to meet than those
for ASR wells, because ASR wells typically inject into drinking-water aquifers, whereas deep
injection wells typically inject into aquifers containing salty water. Deep injection wells also have
the added advantage of permanent disposal of stormwater that contains excess nutrients.
Additionally, injection wells can typically be operated at higher pumping rates than ASR wells
because water is injected into an aquifer of higher transmissivity (i.e. the cavernous Boulder
Zone in South Florida which has extremely high transmissivities on the order of 3,000,000
ft2/day, Johnson and Bush (1988)). The primary disadvantage of using deep injection wells is
that once the water is injected it cannot be typically be recovered. Figure V-10 shows a deep
injection well system compared with a typical ASR well and water well.
A typical deep injection well is 24-30 inches in diameter and can dispose of 10-30 MGD of
excess water per well (SFWMD, 2009a; Maliva and Missimer, 2006). Maliva and Meissimer
107
(2006) proposed consideration of fifty 30 MGD Class I deep injection wells for South Florida that
would provide 1.5 bgd of disposal capacity. Similarly, the ASR regional modeling study
(USACE, 2014) recommended one hundred and one 10 MGD injection wells into the Boulder
Zone, for a total disposal capacity of 1 bgd. If a network of deep injection wells with a combined
disposal capacity of 1 bgd were operated continuously for a period of 100 days it could dispose
of approximately 307,000 acre-ft of excess water which could reduce the depth of water in Lake
Okeechobee by almost a foot. However, disposing of excess water with this network would have
to take place proactively -- managed in concert with other storage elements, lake levels, and in
anticipation of extreme events -- since it would not provide sufficient pumping capacity to quickly
dispose of extremely large wet season flows (such as occurred in WY 2014).
Summary
If sufficient inter-annual storage and treatment north of the lake is found to be economically or
politically infeasible, or the analysis indicates that the captured water cannot be efficiently
treated and conveyed south of the lake for use in subsequent dry seasons, the option of
constructing a system of large deep disposal wells to permanently dispose of excess flows from
Lake Okeechobee should be explored. Deep injection wells could be part of a long-term solution
to reduce damaging discharges to the St. Lucie and Caloosahatchee estuaries, or they could
provide an interim solution until additional storage, treatment and conveyance capacity can be
constructed south of Lake Okeechobee.
108
Figure V-10. Typical Municipal Class I Injection Well, Aquifer Storage and Recovery Well, and Water Well in
Southeast Florida ( from FDEP http://www.dep.state.fl.us/water/uic/docs/ASR_mun.pdf, accessed Feb 12, 2014).
109
a. Modification of Lake Okeechobee Regulation Schedule

Introduction
Lake Okeechobee was designed to be the central water storage feature of the Central and
South Florida (C&SF) Flood Control Project when the latter was constructed in the 1950s and
60s. However, the lake has an unusual configuration for a flood control reservoir: (a) it has no
spillway to allow water to exit when water levels reach a level that is unsafe for the perimeter
levee; and (b) the inflow capacity of water exceeds the outflow capacity by as much as 4 to 6-
fold (depending on hydrologic conditions, rainfall and antecedent water levels in the lake). When
water levels in the lake are high and a large amount of rainfall occurs in the basin north of the
lake, large inflows can occur and the lake can rise several feet in just a short period of time,
presenting a potentially dangerous scenario in which the structural integrity and function of the
levee are compromised.
The large canals to the St. Lucie and Caloosahatchee estuaries are the main ‘relief valves’ for
removing water from the lake when water levels are high, because canals leading south have a
much smaller conveyance capacity, are limited in their ability to take water by regulation
schedules of the WCAs, and are limited by additional factors including threatened and
endangered species and by rules regarding maximal amounts of phosphorus that can be
discharged into the Everglades (see Section II.5).
Dike Rehabilitation
In 2007, the US Army Corps of Engineers (USACE) initiated a greater than $300 million effort to
reinforce the Herbert Hoover Dike that surrounds Lake Okeechobee. Components of this project
include installation of a cutoff wall inside the levee (dike) and replacing water control structures
around the lake. A phased work plan was proposed where areas of the dike that were
determined from geotechnical studies to be at greatest risk of breach under condition of high
water and/or tropical storms would be retrofitted first. However, in 2011 the USACE changed its
approach. After completing the construction of a cut-off wall between Belle Glade and Port
Mayaca in 2013, the USACE opted next to replace 32 water control structures around the lake,
which were considered the next greatest risk of levee failure (SFWMD 2013). Concurrently, the
USACE is conducting a comprehensive Dam Safety Modification Study on the Herbert Hoover
Dike to determine what additional structural and non-structural alternatives are required to
reduce risk to an acceptable level, and to develop a roadmap for their implementation.
110
Regulation Schedule
Because the underlying reason for levee rehabilitation was concern regarding risk of structural
failure, as evidenced by extensive piping and sinkhole formation after high water years, the
USACE formally adopted a new regulation schedule for the lake in April 2007 and implemented
it in 2008. The Lake Okeechobee Regulation Schedule of 2008 (LORS 2008) was designed to
hold the lake at what the USACE considered to be safer levels (from a flood control perspective)
until the Dam Safety Modification Study is completed, the risk of failure more rigorously
evaluated, and modifications required to reduce the risk to an acceptable level are completed.
The LORS 2008 has many features that are similar to the earlier Water Supply/Environmental
(WSE) regulation schedule: (a) it has seasonally varying ‘bands’ that identify different zones
related to the amount of water to be released from the lake to protect the levee; (b) there is a
water shortage zone below the bottom band of LORS 2008 where the South Florida Water
Management District has the authority to implement water use restrictions; and (c) releases of
water from the lake for flood control purposes are determined, in part by rainfall outlook,
tributary hydrologic conditions and multi-season climate outlooks (USACE 2008).
One major difference between LORS 2008 and the earlier schedule is the aim of LORS 2008 to
hold Lake Okeechobee at a lower level, by allowing for regulatory discharges to the St. Lucie
and Caloosahatchee Estuaries and to the Everglades Water Conservation Areas. According to
the USACE ‘one of the primary goals of LORS is to maintain a lake level between 12.5 and 15.5
feet.’ This is done by managing water within bands shown in Figure V-11, as follows:
High lake management band - includes lake levels above 16 feet in advance of the wet
season, or levels above 17.25 feet during the dry season. Within this band, operations are
focused on reducing the lake level, freeing up additional capacity for runoff from future heavy
rain events. Maximum water releases typically occur when the lake is in this band.
Operational bands – these are five sub-bands that guide decisions to balance the needs of all
users, while maintaining a lake level in a 12.5 to 15.5 ft range prescribed in the USACE Water
Control Manual. Toward the lower end of the range, the USACE relies heavily on input from the
SFWMD to assist with water allocations. In the upper bands regulatory releases are made by
the USACE for the purpose of flood protection.
Water shortage management band - this band includes lake levels below 10.5 feet in advance
of the wet season, or levels below 13 feet at the start of the dry season. In this band, the
USACE typically defers decisions on water releases to the SFWMD.
111
When water levels are in the lower operational bands and in the water shortage zone, the
SFWMD balances human and environmental needs and makes recommendations to the
USACE about water releases (e.g., environmental water deliveries to the Caloosahatchee
Estuary) based on Adaptive Protocols, which are procedures adopted by the SFWMD
Governing Board in September 2010 (SFWMD 2010).
Figure V-11. The 2008 Lake Okeechobee Regulation Schedule (LORS 2008), showing operational bands referenced
in the text as well as the zones where the SFWMD may implement water use restrictions because the lake is very
low. Source: USACE Water Control Manual for Lake Okeechobee, March 2008.
Effects of the Lower Schedule
Because LORS 2008 aims to hold water levels lower in Lake Okeechobee relative to the WSE
schedule, the storage capacity of the lake has been functionally reduced. Thus, both regulatory
discharges to the estuaries and water shortage restrictions are expected to occur more
frequently, and be of greater magnitude and duration, for like climatic and hydrologic conditions
(EIS, USACE 2007).
Modeling studies conducted for the Environmental Impact Statement (EIS) produced pursuant
to the requirements of the National Environmental Policy Act (NEPA) using the same 41 year
climatic record for LORS 2008 and WSE showed that for the St. Lucie Estuary, “All alternatives
(of LORS) had high flows (>3,000 cfs) of longer duration than the No Action Alternative (WSE).
The 14-day moving average total inflow exceeded 3,000 cfs for more than ten weeks (five two-
week periods) in Alternatives A through E while the No Action Alternative (WSE) had no such
occurrences. When the high flow events increase substantially in duration impacts to the
112
estuary can be more adverse. Oysters and sea grasses (sic) may be negatively affected. These
sessile species cannot move to areas of preferred salinity ranges although they can tolerate low
salinity levels for short durations. These species become more susceptible to disease and
predation as the duration of extreme high flow events increase.”
The same negative impacts to the Caloosahatchee Estuary were projected by modeling in the
EIS: “All alternatives (of LORS) had high flows (>4,500 cfs) of longer duration than the No
Action Alternative (WSE). This was reflected in the total number of weeks of high flows greater
than five weeks and at times, Alternatives Band C, had very long duration flows in the 13 to 16
week range. The No Action Alternative had the fewest number of weeks of high flow greater
than five weeks.”
According to the EIS, holding the lake lower also has potential adverse consequences for
consumptive use of lake water, which is used to irrigate agricultural crops to the southeast,
south and southwest of the lake, and to maintain a freshwater head that helps prevent saltwater
intrusion to urban well fields along the lower east coast metropolitan area of Florida. Modeling
results identified that the average water cutback volume to permitted users in 8 drought years
was, cumulatively, 124,000 acre-ft with WSE and 223,000 acre-ft with LORS 2008 for the same
41-year period of record. Likewise, the number of months with adequate water to maintain a
desired minimal flow to protect submerged vegetation in the upper Caloosahatchee Estuary
(450 cfs) was reduced from 216 under WSE to 143 with the LORS 2008 (USACE 2007).
EIS modeling scenarios project what the effects of a change in regulation schedule would be if
all conditions were held constant except for regulatory release decisions. However, attributing
observed estuarine effects across different time periods to differences in the regulation
schedules is difficult because differences in regulatory releases are, in fact, governed by a large
number of factors aside from the release schedule. These factors include the amount and timing
of lake inflows, antecedent lake levels, contemporaneous conditions downstream of the lake
that constrain the ability to move water south, and local basin runoff.
This point is illustrated in Figures III-3 and III-4 (Section III) which show the Lake Okeechobee
inflows and the estuary outflows spanning both the period that the WSE schedule was in place
(July 2000 to March 2008) and the period that LORS 2008 has been in place (April 2008 to
present). During the WSE time period freshwater inflows to the St. Lucie estuary exceeded its
very high target (>3,000 cfs) 12% of the time versus 5% of the time for the LOR2008 time
period. Similarly, freshwater inflow to the Caloosahatchee estuary exceeded its very high target
(>4,500 cfs) 17% of the time during the WSE time period versus 10% of the time for the LORS
2008 period. Thus, even though WSE performs better than LORS 2008 in simulations using
identical conditions, high discharges to the estuaries actually occurred more frequently under
WSE than under LORS 2008 due to the differences in weather patterns and hydrologic
113
conditions. Given the extreme rainfall conditions that occurred in the summer of 2013 it is
unlikely that any reasonable regulation schedule could have avoided damaging estuarine
discharges given the current C&SF design.
Opportunities for Holding More Water in the Lake
In general, it is the timing and duration of extreme rainfall events that determines whether a
regulatory release must take place under the current C&SF system design. Therefore, small
adjustments within the discretionary bands of LORS 2008 may be able to shave the peaks off
damaging high flows to the estuary or perhaps reduce their duration, but they are unlikely to
change their frequency of occurrence. However, the completion of the Dam Safety Modification
Study, and the execution of its recommendations for reducing the risk of failure, could provide
an opportunity for substantive changes to the Lake Regulation Schedule that could produce
significant benefits for the estuaries and move more water south of Lake Okeechobee.
Clearly, a decision to change the regulation schedule to hold more water in the lake needs to
evaluate tradeoffs between benefits to the estuaries and water supply and adverse impacts to
the lake’s littoral zone, which has been shown to have optimal ecological structure and function
in a water level range of 12 to 15 ft (Havens, 2002). Furthermore, it may be prudent to
simultaneously re-evaluate the lake’s minimum water level (ML) of 11 ft, to determine if it is
scientifically sound. This ML was set largely to protect food resources of the federally-
endangered Everglades Snail Kite, to prevent loss of peat from exposed lake sediments and to
prevent woody upland plants from becoming established in the littoral zone (SFWMD 2000).
However, in 2000-01 the lake experienced an extreme drought, with water levels falling as low
as 9.1 ft and remaining near or below 10 ft for over 7 months (Havens et al., 2007). The system
was highly resilient and did not suffer long term adverse effects from the low water levels during
that event. Thus, today lake ecologists consider periodic low lake level conditions to be highly
beneficial for native biota including commercially important sport fish (SFWMD 2014). However,
recurrent extremely low lake levels may impact the littoral zone, submerged vegetation and
fisheries (Havens and Steinman, 2013).
Based on the stage-storage curve for Lake Okeechobee (SFWMD, 1986), raising the allowable
maximum lake stage (trigger points in every operational band and upper band) by just 0.5 ft
could add approximately 225,000 acre-ft of static storage, and potentially a greater amount of
dynamic storage over the course of a year if that water was released from the lake in a
beneficial manner. Furthermore, lowering the minimum level of the lake from 11 to 10 ft would
free up an additional approximately 330,000 acre-ft of water for use by agriculture, urban areas
and for downstream ecological benefits (e.g., for support of aquatic plants, reducing salinity and
flushing of algal blooms in the upper Caloosahatchee Estuary).
114
Based on research conducted in the late 1990s and early 2000s, and summarized in Havens
and Gawlik (2005) and Havens (2002), there are some attributes of Lake Okeechobee
hydrology that have a high degree of certainty in regard to their ecosystem impacts. For
example:
 Lake stages in the range of 15.5 ft or higher result in mixing of phosphorus and
sediment-rich water from the center of the lake with waters in the nearshore region,
which cause adverse impacts to submerged plants and increases in algal blooms.
 Lake stages in the range of 17 ft or higher result in a high level of erosion of the littoral
shoreline during times of high wind/wave energy, and also to large incursions of
phosphorus into the littoral zone, which in the past has contributed to an expansion of
cattail.
 A spring recession of lake water levels to near 12 ft lake stage is beneficial for wading
birds, snail kites, submerged plants and the lake’s fishery, as long as it is not quickly
followed by a rise in water level.
Based on that same research, an optimal ‘lake stage envelope’ of 12 to 15 ft was established,
and it is here where more uncertainty occurs. There are not sufficient data to discern whether
similar ecological benefits would occur from a yearly range of 11.5 to 15.5 ft or 12.5 to 15.5 ft (or
some other combination of low and high stages). As noted above, there also is uncertainty, now
that the recovery from a severe drought has been observed, about whether a minimum water
level of 11 ft is overly protective of the lake ecosystem.
Summary
Given the possibility of reducing estuary discharges by storing more water in the lake, either
incrementally by modifying operations with the discretional bands of LORS 2008, or
substantially by establishing a new lake regulation schedule, a rigorous analysis should be
conducted that weighs the benefits (reduced damaging high and low estuary discharges,
reduced water shortage emergencies) vs. costs (lake impacts) of adjustments to the lake’s
minimal level, and associated changes in water allocation rules, as well as the location of bands
in the USACE regulation schedule.
Based on the available information, it does not appear that adjustments within the current LORS
2008 schedule will have a substantive effect on the occurrence of high damaging discharges to
the estuaries. However, a substantially revised regulation schedule that provided considerably
more dynamic storage in the lake could provide those benefits and should be considered, along
with trade-offs including potential adverse impacts to the lake’s littoral zone, submerged
vegetation and fisheries. Developing a new regulation schedule first requires completion of the
USACE Dam Safety Modification Study and guidance about the safety of the Herbert Hoover
115
levee and operational structures. The USACE should be urged to accelerate completion of the
Dam Safety Modification Study so that the modification of the Lake Okeechobee regulation
schedule can occur as soon as possible. In the interim, to provide incremental estuarine relief,
Lake Okeechobee operations could be modified within the discretionary bands of LORS 2008 to
allow some additional water to be moved south to the EPA, and also provide increased dry
season flows to the Caloosahatchee estuary and EAA.
b. Modification of Holey Land and Rotenberger Regulation Schedules

Section V.3.b above considered the possibility of converting both the HLWMA and the RWMA
for use as shallow storage and STAs to increase the storage, treatment and conveyance
capacity south of Lake Okeechobee. This section considers the less radical option of modifying
their regulation schedules under their current management regime, to contribute to reductions in
harmful discharge to the St. Lucie and Caloosahatchee estuaries.
Under this option HLWMA and RWMA would be used in their current form and designation to
receive clean water, and to increase transient storage within those areas during times of high
discharge. Under some high discharge conditions, the WCAs become too full to receive more
water from the STAs, creating a bottleneck. Additional post-treatment storage could help to
alleviate this problem. Both WMAs are in close proximity to one or more STAs, and are
therefore in a reasonable geographic position for this purpose. The current hydrologic
management plans for both areas call for water depths that mimic historical flooding regimes,
with an explicit goal of moving vegetative communities from the over-dried, shrubby and woody
condition, towards a more historically typical sawgrass marsh with occasional embedded
sloughs and tree islands (FWC, 2007). In both areas, this means a move toward longer
hydroperiods than have been typical of the past 35 years.
While RWMA may be at or close to the desired longer hydroperiod, longer hydroperiods in
HLWMA have not been achieved because of structure-limited ability to move water in and out of
the impounded marsh. Inflows occur at a single point (G200A see Figure V-12), which has been
out of operation because of hurricane damage in 2006, and as a result HLWMA has been
rainfall-driven, or close to it, for nearly a decade. During 2014, temporary pumps were installed
that allow for the inflows at that location. However, this is a temporary fix and more permanent
infrastructure would be needed in the long term. The second and perhaps more important long
term problem in HLWMA is outflows, which currently occur at several points along the L5 borrow
canal. These outflows are poorly placed for efficient movement of water since they are nearly
two feet higher than the lowest edge (northeast corner, see Fig V-13). In other words, much of
HLWMA must remain flooded after the water level has dropped below the level of the current
outflow structures. New outflow structures have been proposed for the much lower northeast
corner, but construction is currently constrained by lack of funding.
116
Figure V-12. Northwest corner of Holey Land Wildlife Management Area, showing inflow structures. Figure
courtesy of the Florida Fish and Wildlife Conservation Commission.
Figure V-13. Contour map of Holey Land Wildlife Management Area. Figure courtesy of the Florida Fish and
Wildlife Conservation Commission.
117
At present, water levels in HLWMA are generally above its regulation schedule in the wet
season, and therefore it is not able to receive wet season flow (Fig V-13). However, the lack of
effective outflow structures from HLWMA means that especially during the wet season, much of
the water can only leave by evapotranspiration. Thus, any water added to the system during the
wet season must stay for long periods, resulting in high probability of damaging the vegetative
communities through long lasting flooding. The current water management schedule is clearly
designed to avoid this condition. With infrastructure that allows for more active manipulation of
inflows and outflows, water storage could be much more dynamic, allowing the ability to remove
water relatively quickly after short term storage. Storing an additional 1 – 2 feet of water for a
period of 2 – 3 weeks, for example, may not be permanently damaging to vegetation. What is
important is that there is a means to quickly remove the water after that time – which can only
be achieved through more efficient inflow and outflow structures.
There also appears to be some potential for increased dry season storage in HLWMA based on
the optimum hydrograph under the current management plan (Figure V-14). Historical average
stages during the dry season from 1990 to 2010 were between 0.5 and 1.0 feet lower than the
preferred stage, indicating that during the dry season this WMA could store a minimum of
31,000 acre-ft of water more than it currently does. This offers a modest increase in dry season
storage, and in some years it is possible that this storage could reduce some releases to the
estuaries and allow FEBs, STAs, and Lake Okeechobee to be maintained at lower dry season
levels. It is unclear how often such dry season storage would be useful, and, in the general
context of the requirement for long-term regional storage, it is not interannual, nor is it very
large.
118
Figure V-14: Holey Land WMA monthly stage levels for 2013, 2044, and 1990-2013 average. Also shown are
ground elevation and high and low closure criteria. Figure courtesy of the Florida Fish and Wildlife Conservation
Commission.
Summary
In summary, modest increases in dry season storage (approximately 31,000 acre-ft) appear to
be achievable and desirable in HLWMA, which under some conditions could provide buffer for
ensuing wet season capacity upstream. There is also the potential for increased dynamic
storage (approximately 66,000 – 130,000 acre-ft) during the wet season in HLWMA, but only
over short periods of several weeks at a time. Both dry and wet season storage would require
funding and construction of new water control structures at inflow and outflows of HLWMA.
119
VI. Future Uncertainties
Success in restoring a more natural timing, distribution and volume of clean water to the
Everglades, while reducing harmful discharges to the St. Lucie and Caloosahatchee estuaries,
depends on the short and long-term outcomes of operational decisions and construction
projects. The performance of those decisions and projects is subject to variation depending on
future conditions.
CERP was developed under the principle of ‘stationarity,’ which is assuming that the future will
be like the past. This is best exemplified by the fact that all of the modeling runs done to
determine hydrologic performance in CERP were based on a pattern of rainfall and surface
water movements that happened in the prior 30-year period. CERP also made assumptions
about the rate of population growth and water demands in southeast Florida, assumed no
changes in land use and water demands in the agricultural areas around the lake, and assumed
no changes in per capita water use.
If any of these variables that affect the availability of water should change substantially in future
decades, this could greatly alter the real outcomes of regional restoration. Research to date
suggests that major changes are likely to occur.
1. Potential Effects of Climate Change
In 2013, a group of scientists from Florida universities and state/federal agencies held a series
of workshops, culminating in a scenario planning exercise. The same approach that was used to
screen restoration alternatives in the Restudy was adopted, using the South Florida Water
Management Model (SFWMM, 2005), driven by projections of what climate and sea-level rise
might be in the year 2060. Obeysekera et al. (2014) describe the details of the selection of
climate change scenarios and the use of that regional model.
In short, the modeling considered three possible future scenarios: (1) future base condition, i.e.,
no changes in hydrology other than those resulting from completion and operation of CERP
projects; (2) future base condition plus a 1.5 degree C rise in temperature with the resulting rise
in evapotranspiration and a 1.5 ft rise in sea level; and (3) future base conditions plus increased
temperature and evapotranspiration, and a change in annual rainfall of ±10 %, plus a 1.5 ft rise
in sea level. These temperature and precipitation scenarios reflect the range of projections from
Obeysekera et al. (2011) (Figure IV-1). The sea-level rise scenario is a mid-range estimate from
those being used by the USACE in their regional planning for projects to be constructed in south
Florida (Figure VI-2).
120
Figure VI-1. Predictions of the changes in temperature and rainfall from an ensemble of climate models run for
different IPCC future scenarios. The average increase in temperature using this approach is 1.5 degrees C by 2060,
and rainfall is not well predicted, with estimates varying from plus to minus 10 percent (Source: Obeyesekera, 2011).
Figure VI-2. The range of sea-level rise that is considered by the USACE in planning coastal projects in south
Florida. A rise of 1.5 ft is in the middle of the range for 2060 and was the value used in the modeling conducted by
Obeysekera (2014).
The SFWMM assumed no changes to the current regulatory schedule for Lake Okeechobee
and did not account for other measures (physical or operational) that might be adopted to
ameliorate adverse effects of water availability caused by climate change. For this exercise,
potential evapotranspiration (ET) was estimated using a simple method (Abtew et al.,2011)
where an increase in temperature is directly translated into an effect on potential ET. Further,
the modeling did not assume any changes in the seasonal dynamics of rainfall or ET. While
such changes may occur, there currently are no regional projections to guide such scenarios.
Output from the SFWMM included a water mass balance, 41-year hydrographs (1965–2006),
121
and stage duration curves (which indicate the percent of time a particular surface elevation is
exceeded in the particular modeling scenario). These results were produced for all the major
regions of the greater Everglades, from Lake Okeechobee to Florida Bay, and were evaluated
by the same experts who were involved in alternatives evaluations during the Restudy, to
project what the outcomes would be with regard to ecological and societal impacts. In the future
scenario with just an increase in rainfall (no change in temperature or ET), the regional system
would be wetter (Figure VI-3, left panel). However, this is a highly unlikely future scenario and in
fact, Figure VI-1 indicates that none of the climate models predict this scenario will occur. In a
more likely scenario with a warmer future ET would be greater, and with no change in rainfall
the entire regional system would experience a water deficit. In a hot and dry future, i.e.,
increased temperature and evapotranspiration and 10% less rain, the result is extreme water
deficits and severe impacts to both the ecological systems and the built environment (Figure VI-
3, right panel).
Figure VI-3. Two extreme scenarios for 2060 based on the regional simulation modeling performed by Obeysekera
(2014). On the left, the entire system is wetter than today, however this only would occur if rainfall increases by ten
percent AND the future is not warmer. The latter is unlikely. On the right, the entire system is extremely dry due to
warmer temperatures, higher ET and less rainfall, and the lower east coastal areas are inundated by rising seas.
In the case of Lake Okeechobee (Havens and Steinman, 2013), by comparing 40-year lake
stage hydrographs, which show daily changes in water levels in the lake, it can be seen that in a
future with increased temperature and more rain, water levels are identical to the future base
122
(the two lines are on top of one another in Figure IV-4). If it becomes warmer and ET increases,
but rainfall does not also increase, lake levels drop considerably, as do outflows to the estuaries
(St. Lucie regulatory releases are reduced by 62% and Caloosahatchee regulatory releases are
reduced by 56%). If it becomes warmer and rainfall declines by ten percent, the lake becomes
very low, staying below what is considered an ecologically optimal zone of 12 to 15 ft nearly 80
percent of the time, and sometimes it drops as low as 5 ft. Under this scenario the lake is
extremely low for prolonged periods of time, during which it would be impossible for vessels to
navigate across the lake or for water from the lake to meet downstream needs.
In the hot/dry future scenario, discharges to the estuaries are greatly reduced. Regulatory
discharges to the St. Lucie Estuary decline by 95% and regulatory discharges to the St. Lucie
Estuary decline by 90%. While this scenario may seem extreme, it is a possible future state of
south Florida.
Figure VI-4 Historical (41-year) hydrographs for Lake Okeechobee, based on output from the South Florida Water
Management Model and assuming four different regional conditions: Future base – 2060 conditions of projected
water demands based on population estimates and anticipated configuration of the regional flood control system
including new projects built as part of the Comprehensive Everglades Restoration Plan; +ET—same conditions as
the future base, with increased regional evapotranspiration calculated from a temperature increase of 1.5 degrees C;
+RF+ET - same as +ET scenario, with a 10% increase in rainfall compared to the historic conditions; -RF+ET -
same as the +ET scenario, with a 10% decrease in rainfall. The two horizontal dashed lines delineate the zone that
has been identified as optimal for lake elevation in most years to support a diverse assemblage of plants, fish,
wading birds, and a variety of ecosystem services provided by the lake. The +RF+ET hydrograph is hidden beneath
the base, as lake stages are nearly identical for the entire period of record. (Havens and Steinman 2013).
Other regions of the Everglades landscape responded in a similar manner to the future climate
scenarios (Nungesser et al., 2014) with a substantial increase in future rainfall being required to
counter-balance the anticipated increases in temperature and ET. Futures without that
additional rainfall experience prolonged periods of drying of the Water Conservation Areas and
123
Everglades, are more at risk for fires that can destroy peat, and suffer losses to native
vegetation including sawgrass and tree islands and the fish and wildlife that they support. In the
most extreme situation of a future with higher temperatures and less rain, there are long periods
during which there is no water in much of the Everglades, which also would risk coastal well
fields, because of the lack of fresh water hydraulic head to hold back salt water. This problem is
exacerbated because there is at the same time a projected rise of sea level by 1.5 ft and thus a
greater need for fresh water from the regional ecosystem to maintain the fresh water-salt water
interface at a desirable location.
Warmer surface waters, both in Lake Okeechobee, the St. Lucie and Caloosahatchee rivers and
the estuaries, can be expected to result in more frequent and severe blooms of potentially toxic
algae, because warmer water has synergistic positive effects with nutrients on the growth of the
algae that cause those blooms (Moss et al., 2011). This means that in the Caloosahatchee,
where algal blooms already are a common summer occurrence, there will be a greater need for
environmental pulse releases from the lake to push the blooms out to sea in a warmer future –
yet in a drier future, the water may not be available.
Any climate induced change in hydrology will also have a profound influence on the
biogeochemical processes that underlie the structure and function of uplands, wetlands, and
aquatic ecosystems within the Greater Everglades (Delpla et al., 2009; Orem et al., 2014;
Reddy and Delaune, 2008; Reddy et al., 2010; Twilley, 2007). Extreme variations in
temperature and precipitation resulting from climate change, for example, can affect transport of
particulate matter, nutrients, and other constituents from uplands/wetlands to downstream
receiving waters and exacerbate eutrophication related issues. Warmer soils and irregular
precipitation patterns will likely increase the rates of microbial activity, organic matter
decomposition and nutrient regeneration, and export of pollutants downstream. During dry
conditions P is often bound to soils and is relatively stable, but under prolonged flooded
conditions P can be solubilized and exported to downstream resulting in increased nutrient
loads (Reddy et al., 2010). Dry conditions can also promote the release of nitrous oxide from
soils, while flooded conditions result in the production of methane gas. Both nitrous oxide and
methane are well known greenhouse gasses. Dry conditions in wetlands can lead to the
oxidation of sulfides and a decrease in soil pH, which, in turn, can result in solubilization of P
(Reddy and Delaune, 2008). During flooded conditions, sulfate can be reduced to sulfide; a
process that promotes methyl mercury formation. Although many of the processes noted above
currently occur in the Everglades ecosystem, a change in process rates can have substantial
consequences for nutrient reduction and remediation efforts. Ultimately managers and
regulators involved in Everglades restoration activities should consider more fully how climate
change may affect nutrient reduction goals.
124
2. Effects of Changes in Human Population Size, Location and Land Use
Future changes in population size are highly uncertain, because as seen in recent years, they
can rise and fall with the economy. However, it is expected that over the next few decades,
several million more people will live in Florida, adding to the current population that has just
reached 20 million. Researchers at the Massachusetts Institute of Technology (MIT) conducted
a project called ‘addressing the challenge of climate change in the greater Everglades
landscape’ with funding from the US Fish and Wildlife Service and the US Geological Survey.
They examined possible future scenarios with high vs. low sea-level rise, high vs. low financial
resources used to mitigate impacts of climate change, pro-active planning vs. no advance action
and a doubling of population size over the next 50 years vs. continued growth at the current rate
(Vargas-Moreno and Flaxman, 2010).
The scenarios show a great range of different outcomes with regard to how land use will change
and where people will live in 50 years (Figure VI-6). Both factors have a large potential influence
on water demands, and are not considered at this time in any Everglades restoration planning.
3. Other Possible Future Changes
Per-capita consumption of water is another unknown, and if it declines markedly, could offset
some of the adverse consequences of a future with less available freshwater. However, it is
unknown to what extent water conservation measures will be implemented over the next 50
years, either by choice of users or by legal mandates.
Another uncertainty that could affect ecosystem restoration is the price of energy. The current
C&SF and the proposed restoration plans require large amounts of energy to operate. The
Department of Defense has in recent years considered energy to be its single largest future
constraint, and as such a critical threat to national security (U.S. Army. 2014). Thus the annual
energy costs of pumping water in a fully restored South Florida ecosystem could be substantial.
125
Figure VI-6. Two scenarios of land use and human population distribution in 2060 under different assumptions
(upper scenario diagram), showing a wide range of outcomes, from one with a large amount of natural area north
and south of Lake Okeechobee in a future with a high level of pro-active planning and small amount of climate
change (right-hand panel) to a near-complete loss of natural areas north of the lake in a future following little pro-
active planning (left-hand panel) (from Vargas-Moreno and Flaxman, 2010)
126
4. Uncertainty in Future Funding
Dealing with the issues of damaging freshwater releases to the St. Lucie and Caloosahatchee
Estuaries and insufficient flow of clean fresh water to the Everglades requires a large
expenditure of federal and state funds. In a future with a growing human population and
unknown changes in their geographic distribution and water needs, it is reasonable to expect
that future costs of addressing these issues will rise, particularly if the cost of land increases
substantially at the same time that there is an increased need for buying large areas of land to
store and treat water.
Figure VI-7 illustrates the historical trend in spending for CERP projects from both state and
federal sources and Figure VI-8 illustrates state and federal spending for both CERP and non-
CERP projects in South Florida. Between 2002 and 2008 Florida’s spending substantially
outpaced federal spending, however Florida’s spending has declined significantly since 2008
(NRC, 2014). Furthermore, federal spending has also declined since 2010 (NRC, 2014). If these
patterns continue it may be impossible to ameliorate the problems facing the estuaries and
Everglades.
It is notable that in FY14 the Florida Legislature appropriated over $200M in new funds for
restoration projects, including a variety of projects that will help to address the issues facing the
Everglades and estuaries. A long-term commitment of federal and state funds will be required to
substantively address the issues in the long term and in the context of the uncertainties noted
above.
127
Figure VI-7. Spending on CERP projects by federal and state governments (from NRC, 2014).
NOTE:* Requested.
Figure VI-8 Federal and state spending related to South Florida ecosystem restoration activities, including CERP
and non-CERP projects and related expenditures (from NRC, 2014)
NOTE:* Requested
128
5. Summary
There are major uncertainties associated with the outcomes of Everglades restoration. These
include uncertainties related to climate change, human population size, geographic location of
population centers, changes in land use and funding. Substantive research and modeling has
been done to identify how climate change, changes in human demographics and changes in
land use may affect Florida’s future, yet there is little evidence that the information from this
research and modeling is being incorporated into restoration project plans. Projects appear to
be planned and executed with the assumption of stationarity (the world of the future is like the
world of the past) and this likely will not be true for critical drivers including rainfall,
evapotranspiration, temperature, sea level and where people live and how they use the land.
Failure to draw on information about future conditions risks successful project outcomes. Even if
the future of the aforementioned parameters is highly uncertain, the possibility of future changes
needs to be acknowledged, effects on restoration outcomes assessed, and flexibility be
incorporated into projects so that they can have positive outcomes over a broad range of
conditions.
Even in the face of uncertainty, many existing plans have been fully vetted and can be expected
to yield substantial benefits to the citizens of Florida. Most of those plans are not proceeding
because of a lack of funding. In the interim, there is considerable evidence that the ecological
system is continuing to degrade in ways that may not be reversible. Monitoring and assessment
of system performance is essential to guide projects and to detect and adapt to future surprises.
Increased and sustained State and Federal funding is critical to achieve restoration before the
system becomes so degraded that major attributes reach tipping points that cannot be reversed.
The recent decline in funding should be addressed by the lead agencies, Congress and the
Florida Legislature so that sufficient financial and human resources are available to complete
critical restoration elements in an accelerated and effective manner.
129
VII. Summary and Conclusions
The Technical Review Team concludes that relief to the estuaries and the ability to move water
south of Lake Okeechobee can be accomplished using existing technology. The solution is
enormous increases in storage and treatment of water both north and south of the lake. Existing
and currently authorized storage and treatment projects are insufficient to achieve these goals.
The path forward requires significant long-term investment in the infrastructure of the South
Florida hydrological system.
To reduce damage to the St. Lucie and Caloosahatchee estuaries freshwater inflow and nutrient
loads from both Lake Okeechobee and the local basins must be reduced. On average, 70-80%
of the freshwater discharge and 65-80% of the nutrient load to the St. Lucie and
Caloosahatchee estuaries originates in the local basins, with the remaining balance contributed
from Lake Okeechobee. Previous Comprehensive Everglades Restoration Plan (CERP),
Northern Everglades and Estuaries Protection Program (NEEPP) and River of Grass (ROG)
planning exercises have all identified that providing large volumes of regional storage is
essential to reduce freshwater discharges to the estuaries. The most recent estimates of
required storage include:

 approximately 1,000,000 acre-ft of water storage distributed north and south of Lake
Okeechobee.
In spite of the repeated demonstrated need for large volumes of water storage, very little new
storage has been designed or constructed in the system. For example, in the St. Lucie
watershed it is estimated that approximately 200,000 acre-ft of storage is required. However,
only one 40,000 acre-ft surface reservoir is currently under construction. In the Caloosahatchee
watershed, it is estimated that approximately 400,000 acre-ft of storage is needed, but currently
only one 170,000 acre-ft surface reservoir is being designed, and state and federal funds for its
construction have not yet been appropriated. Furthermore, although at least one million acre-ft
of storage is required either north or south of Lake Okeechobee, currently only four Flow
Equalization Basins (FEBs) that provide 168,000 acre-ft of shallow storage are planned and
they are sited south of Lake Okeechobee. Two of the FEBs (totaling 101,000 acre-ft) currently
are under construction by the State and are scheduled to be completed by 2016. State
construction of a third 11,000 acre-ft FEB will not begin until after 2018. The fourth CERP FEB
has yet to be authorized by the US Congress.
130
Based on review and analyses, the Technical Review Team identified the following options to
reduce damaging discharges to the St. Lucie and Caloosahatchee estuaries and move more
water south from Lake Okeechobee:
1. Accelerate funding and completion of existing approved projects
To provide substantial improvement to the St. Lucie and Caloosahatchee estuaries, accelerate
the funding and completion of existing federally authorized CERP projects designed specifically
to provide relief to St. Lucie and Caloosahatchee Basins, i.e.:
 Indian River Lagoon-South (IRL-S) Project: Accelerate construction of the C-44 reservoir
and associated Stormwater Treatment Area (STA). Aggressively pursue state and
federal appropriations needed to design and construct remainder of the IRL-S project
(including C-23, 24, 25 reservoirs and associated STAs, and restoration of over 90,000
acres of upland and wetland areas). Total Storage Provided in St. Lucie watershed:
130,000 acre-ft of 200,000 acre-ft required.
 C-43 Reservoir: Accelerate the design and aggressively pursue state and federal
appropriations needed to design and construct project. Total Storage Provided in
Caloosahatchee watershed: 170,000 acre-ft of 400,000 acre-ft required.
Current Basin Management Action Plans (BMAPs) will not achieve FDEP approved Total
Maximum Daily Loads (TMDLs). To achieve water quality standards in Lake Okeechobee, the
St. Lucie estuary and the Caloosahatchee estuary, more aggressive BMAPs are required. New
field-verified agricultural and urban Best Management Practices (BMPs) that protect water
quality, advanced in situ treatment technologies, and the strategic placement of additional FEB-
STAs in priority basins will be essential to achieve State and Federal water quality standards.
Beyond existing and planned approaches, the substantial reservoir of legacy phosphorus in the
Northern Everglades watersheds will necessitate new and more aggressive strategies to
combat the mobility of phosphorus.
To substantially increase the volume of water moving from Lake Okeechobee to the Southern
Everglades, accelerate funding and completion of the State of Florida Restoration Strategies
and the CERP Central Everglades Planning Project (CEPP), i.e.:
 Obtain federal authorization for CEPP,

CEPP,
 Accelerate State funding and completion of Restoration Strategies,
131
 Conduct a careful analysis of CEPP project construction phasing to determine which
CEPP features can be constructed as soon as possible and to develop a plan for
completion of as many CEPP features as possible during the construction phase of
Restoration Strategies, and
 Reconsider using the Talisman property for a deep storage reservoir with STA rather
than the current design which uses the Talisman property for shallow FEBs.
Total Treatment and Conveyance Capacity to Everglades Protection Area (EPA) after CEPP
and Restoration Strategies: 1.5 million acre-ft per year of 1.7 million acre-ft per year 1999 CERP
target and 2.0 million acre-ft per year 2009 River of Grass target.
Additional efforts, beyond the approved projects listed above, will be required to reduce Lake
Okeechobee-triggered high discharges and nutrient loads to the St. Lucie and Caloosahatchee
estuaries and to achieve dry season Everglades demand targets. Studies indicate that after the
completion of the IRL-S, C-43, Restoration Strategies, and CEPP projects, lake-triggered high
discharges to the St. Lucie and Caloosahatchee estuaries will be reduced by less than 55% and
less than 75% of the dry season Everglades demand target will be delivered to the EPA. A
series of options, beyond currently approved projects, to more fully achieve restoration
objectives are summarized below.
2. Provide Water Storage and Treatment North of Lake Okeechobee
Conduct a strategic planning exercise to provide additional water storage and treatment north of
Lake Okeechobee similar to the ROG Planning Process that was conducted south of the lake.
The NEEPP Lake Okeechobee Phase II Technical Plan (LOP2TP) and the ROG Planning
Process provide a sound foundation from which to plan, design, and build the additional storage
and treatment needed north of Lake Okeechobee. A new strategic planning exercise would
necessarily include a regional modeling effort that takes into account lessons learned and
information gained since the CERP, NEEPP and ROG planning exercises. Examples of new
information gained include the permitting requirements, engineering feasibility and costs, and
inter-annual storage benefits associated with Aquifer Storage and Recovery (ASR), deep
storage reservoirs, shallow water impoundments and dispersed water management (DWM), as
well as the water quality benefits of Stormwater Treatment Areas (STAs) and other treatment
technologies. New data gathering efforts and model developments will be required to simulate
the cumulative impacts of a regional DWM system north of the lake on the quality, quantity and
timing of flows into Lake Okeechobee as a function of climatic conditions, spatial location and
density of DWM features on the landscape, and operation of the regional canal system. The
Technical Review Team expects that the strategic plan will show that, while DWM on private
lands may provide some benefits, DWM will fall short of providing the additional storage and
treatment needed, even if fully implemented. Additional land north of Lake Okeechobee will
132
need to be acquired for that purpose (i.e. from approximately 25,000 acres if 300,000 acre-ft is
provided by deep storage to approximately 187,500 acres if 750,000 acre-ft is provided by
shallow water impoundments).
3. Provide Additional Water Storage, Treatment and Conveyance South of Lake

Okeechobee
Develop a strategic plan for the next increment of south-of-lake storage, treatment and
conveyance to pursue beyond CEPP to take advantage of new north-of-lake storage and
treatment, and more closely meet the performance targets of both the estuaries and the
Everglades ecosystems. Independent assessments suggest that an expansive gravity-driven
wet flow-way throughout the Everglades Agricultural Area (EAA) may not be feasible or provide
maximal benefits to the estuaries. However, the ROG planning process demonstrated that there
are several possible options involving combinations of deep and shallow storage, and wet- and
dry- flow-ways, coupled with STAs and enhanced conveyance that could provide significant
benefit both for the estuaries and the Everglades, far beyond the benefits provided by the
Kissimmee River Restoration (KRR), IRL-S, C-43, Restoration Strategies and CEPP projects.
Achieving substantial reduction in lake-triggered discharges to the estuaries and substantial
improvement toward the dry season Everglades demand target will require between 11,000 and
129,000 acres of additional land between the lake and the EPA, depending on the mix of
storage and treatment options. This land could be obtained by purchase of the current U.S.
Sugar option lands, purchase of lands from other willing sellers, and/or the use of existing state-
owned land (e.g., Holey Land and Rotenberger Wildlife Management Areas (WMAs)).
Deep well disposal could be part of a long-term solution to reducing damaging discharges from
Lake Okeechobee to the St. Lucie and Caloosahatchee estuaries, or it could provide an interim
solution until additional water storage, treatment and conveyance capacity can be constructed
south of the lake. If sufficient inter-annual storage and treatment north of the lake is determined
to be economically or politically infeasible, or the analyses indicate that the captured water
cannot be efficiently treated and conveyed south of the lake for use in subsequent dry seasons,
the option of constructing a system of large injection wells to permanently dispose of excess
flows from Lake Okeechobee in the deep Boulder Zone, rather than discharging to the
estuaries, should be explored.
Adjustments within the current Lake Okeechobee regulation schedule (LORS 2008) are unlikely
to have a substantive effect on the occurrence of damaging high discharges to the estuaries.
However, a substantially revised regulation schedule that provides more storage in the lake
133
might provide those benefits. Developing a new regulation schedule requires completion of the
on-going U.S. Army Corps of Engineers (USACE) Dam Safety Modification Study and guidance
about the safety of the rehabilitated Herbert Hoover levee and operational structures in light of
any new safety standards. The USACE should accelerate completion of the Dam Safety
Modification Study so that modification of the Lake Okeechobee regulation schedule, if
warranted, can occur as soon as possible. Development of a new regulation schedule will
require balancing benefits of holding additional water in the lake for the express purpose of
reducing damaging discharges to the estuaries and increasing agricultural, urban and
ecosystem water supply versus potential adverse impacts to the lake’s ecology.
In the interim, to provide incremental estuarine relief, Lake Okeechobee operations could be
modified within the discretionary bands of LORS 2008. Increasing the dynamic range of storage
in the lake could allow some additional water to be moved south to the EPA, and also provide
increased dry season flows to the Caloosahatchee estuary and EAA. In addition, the regulation
schedules of the Holey Land and Rotenberger Water Management Areas (WMAs) could be
modified to allow more water storage during both the wet and dry seasons. This modification of
the WMA regulation schedules could be in keeping with current goals to restore natural
hydroperiods, but will require the inflow/outflow infrastructure be upgraded to allow dynamic
water level manipulations.
Future Uncertainties
Failure to draw on information about the range of possible future conditions risks the success of
restoration project outcomes. Substantive research indicates clearly that climate change,
changes in human demographics, energy costs and land use will affect Florida’s future, yet
there is little evidence that salient information is being incorporated into restoration project
plans. Even if the future of these variables is highly uncertain, the possibility of future changes
needs to be acknowledged, effects on restoration outcomes assessed, and flexibility
incorporated into projects so that they can have positive outcomes over a broad range of
conditions.
Path Forward
Even in the face of uncertainty, many existing plans and projects have been fully vetted and can
be expected to yield substantial benefits to the citizens of Florida. Most of the projects are
delayed because of a lack of funding. In the interim, the coupled human-ecological system is
continuing to degrade in ways that may not be reversible. Monitoring and assessment of system
performance is essential to guide projects and to detect and adapt to future surprises. Increased
and sustained State and Federal funding is critical to provide additional water storage and
treatment before the system becomes so degraded that major attributes reach tipping points
that cannot be reversed.
134
VIII. List of Abbreviations
acre-ft acre-feet
ASR Aquifer Storage and Recovery
BiOp Biological Opinion
bgd billion gallons per day
BMAP Basin Management Action Plan
BMP best management practice
cfs cubic feet per second
C&SF Central and Southern Florida Project
CEPP Central Everglades Planning Project
CERP Comprehensive Everglades Restoration Plan
CSSS Cape Sable Seaside Sparrows
CWA Clean Water Act
DWM Dispersed Water Management
EAA Everglades Agricultural Area
EFA Everglades Forever Act
EIS Environmental Impact Statement pursuant to NEPA
ENP Everglades National Park
EPA Everglades Protection Area
ESA Endangered Species Act
FAVT Floating Aquatic Vegetation Treatment
FDACS Florida Department of Agriculture and Consumer Services
FDEP Florida Department of Environmental Protection
FEB Flow Equalization Basin
FWC Florida Fish and Wildlife Conservation Commission
FWM flow-weighted mean
FWS U.S. Fish and Wildlife Service
HLWMA Holey Land Wildlife Management Area
HWTT Hybrid Wetland Treatment Technology
IFAS Institute of Food and Agricultural Sciences
IRL Indian River Lagoon
IRL-S Indian River Lagoon-South
ITS Incidental Take Statement
LOP2TP Lake Okeechobee Phase II Technical Plan
135
LORS Lake Okeechobee Regulation Schedule
LORS 2008 2008 Lake Okeechobee Regulation Schedule
maf million acre feet
MBTA Migratory Bird Species Act
mgd million gallons per day
NEEPP Northern Everglades and Estuaries Protection Program
NEPA National Environmental Policy Act
NPDES National Pollutant Discharge Elimination System
NPS National Park Service
PIR Project Implementation Report of USACE
ppb parts per billion
ROG River of Grass Planning Process
RPA Reasonable and Prudent Alternative
RPM Reasonable and Prudent Measure
RWMA Rotenberger Wildlife Management Area
SAV Submerged Aquatic Vegetation
SFER South Florida Environmental Report
SFWMD South Florida Water Management District
STA Stormwater Treatment Area
SWET Soil and Water Engineering and Technology, Inc.
TN Total nitrogen
TP Total phosphorus
UF University of Florida
US United States, or U.S.
USACE U.S. Army Corps of Engineers
USEPA U.S. Environmental Protection Agency
USGS U.S. Geological Survey
WCA Water Conservation Area
WMA Wildlife Management Area
WQBEL Water Quality-Based Effluent Limit
WRP Wetland Reserve Program
WRDA Water Resources Development Act
136
IX. References
Abtew, W. and P. Trimble. 2010. El Niño-Southern and its role in shaping the Everglades
Oscillation link to south Florida hydrology and ecosystem. In Everglades: The Ecosystem and
water management applications. Water Resource Its Restoration. Boca Raton, Fla.: St. Lucie
Management 24: 4255-4271. Press.
Abtew, W., J. Obeysekera and N. Iricanin. 2011. Pan Enfield, D. B., A. M. Mestas-Nunez and P. J. Trimble.
evaporation and potential evapotranspiration 2012. The Atlantic Multidecadal Oscillation and
trends in south Florida. Hydrologic Processes 25: its relation to rainfall and river flows in the
958-969. continental U.S. Geophysical Research Letters
28: 2077-2080.
Aich, S., C. W. McVoy, T.W. Dreschel and F.
Santamaria. 2013. Estimating soil subsidence FDEP. 2012. Basin management action plan for the
and carbon loss in the Everglades Agricultural implementation of total maximum daily loads for
Area, Florida using geospatial techniques. nutrients and dissolved oxygen by the Florida
Agriculture, Ecosystems and Environment 171: Department of Environmental Protection in the
125-133. Caloosahatchee Estuary Basin. Florida
Department of Environmental Protection.
Aumen, N. G. 1995. The history of human impacts,
lake management, and limnological research on FDEP. 2013. Basin management action plan for the
implementation of total maximum daily loads for
Lake Okeechobee, Florida (USA). Archiv für
nutrients and dissolved oxygen by the Florida
Hydrobiologie, Advances in Limnology 45: 1-16. Department of Environmental Protection in the
St. Lucie and Estuary Basin. Florida Department
Barnes, T. 2005. Caloosahatchee Estuary of Environmental Protection.
conceptual ecological model. Wetlands 25:884-
897. FDEP. 2014a. Basin management action plan for the
implementation of total maximum daily loads for
Barnett, B. 1986. Holey Land regional storage total phosphorus by the Florida Department of
reservoir: an option feasibility report. Report Environmental Protection in Lake Okeechobee.
prepared for the Everglades Agricultural Area
Subcommittee of the Lake Okeechobee FDEP. 2014b. Progress report for the St. Lucie River
Technical Committee. and Estuary Basin Management Action Plan.
Florida Department of Environmental Protection.
Bohlen, P. J., S. Lynch, L. Shabman, M. Clark, S. Retrieved from
Shukla, and H. Swain. 2009. Paying for http://www.dep.state.fl.us/water/watersheds/docs/
environmental services from agricultural lands: bmap/StLucieRiverEstuaryBMAP-APR-2014.pdf
an example from the northern Everglades.
Frontiers in Ecology and the Environment 7:46– FDEP. 2014c. Progress report for the
55. Caloosahatchee Estuary Basin Management
Action Plan. Florida Department of Environmental
Buzzelli, C., R. Robbins, P. Doering, Z. Chen, D. Protection. Retrieved from
Sun, Y. Wan, B. Welch, and A. Schwarzschild. http://www.dep.state.fl.us/water/watersheds/docs/
2012. Monitoring and modeling of Syringodium bmap/caloosa-estuary-bmap-apr2013.pdf
filiforme (manatee grass) in southern Indian River
Lagoon. Estuaries and Coasts 35:1401-1415. FDEP. 2014d. Underground Injection Control
Program. Last modified November 18, 2014.
Congressional Research Service. 2014. Everglades Retrieved from
Restoration: Federal Funding and http://www.dep.state.fl.us/water/uic.
Implementation Progress. CRS Report 42007.
FWC. 2007. Position Paper on the restoration of
Delpla, I., A. V. Jung, E. Baures, M. Clement, and O. Holey Land and Rotenberger Wildlife
Thomas. 2009. Impacts of climate change on Management Areas. Florida Fish and Wildlife
surface water quality in relation to drinking water Conservation Commission.
production. Environmental International 35:1225-
1233. Gilmore, R. G. 1995. Environmental and
biogeographic factors influencing ichthyofaunal
Duever, M. J., J.F Meeder, L. C. Meeder, and J. M. diversity: Indian River Lagoon. Bulletin of Marine
McCollom. (1994). The climate of South Florida Science 57:153-170.
137
Havens, K. E. 2002. Development and application of Maliva, R. G. and T. M. Missimer. 2006. Deep
hydrologic restoration goals for a large injection wells and sustainable water
subtropical lake. Lake and Reservoir development in Florida. Proceedings of the 2006
Management 18: 285-292. Florida Section of the American Water Works
Association conference.
Havens, K. E. and A. D. Steinman. 2013. Ecological
responses of a large shallow lake (Okeechobee, Millie, D. F., H. J. Carrick, P. H. Doering, and K. A.
Steidinger. 2004. Intra-annual variability of water
Florida) to climate change and potential future
quality of the St. Lucie River Estuary, Florida
hydrologic regimes. Environmental Management (USA): a quantitative assessment. Estuarine,
DOI 10.1007/s00267-013-0189-3. Coastal and Shelf Science 61:37-149.
Havens, K. E. and D. E. Gawlik. 2005. Lake Moss , B., S. Kosten, M. Meerhoff, R. W. Batterbee,
Okeechobee conceptual ecosystem model. E. Jeppesen, N. Mazzeo, K. Havens, G. Lacerot,
Wetlands 25: 908-925. Z. Liu, L. De Meester, H. Paerl and M. Scheffer.
2011. Allied attack: climate change and
Havens, K. E. 2002. Development and application of eutrophication. Inland Waters 1: 101-105.
hydrologic restoration goals for a large
subtropical lake. Lake and Reservoir National Research Council. 2012. Progress Toward
Management 18:285-292. Restoring the Everglades, The Fourth Biennial
Review. Washington, DC: National Academies
Havens, K. E., T. L. East and J. R. Beaver. 2007. Press.
Zooplankton response to extreme drought in a
large subtropical lake. Hydrobiologia 589: 187- National Research Council. 2014. Progress Toward
198. Restoring the Everglades, The Fifth Biennial
Review. Washington, DC: National Academies
HDR Team. 2010. Task 2 – Nutrient budget analysis Press.
for the Lake Okeechobee Watershed. Task 4.
Final report submitted May 2010 to SFWMD, Nungesser, M., C. Sanders, C. Coronado-Molina, J.
West Palm Beach, Fla. Obeysekera, J. Johnson, C. McVoy and B.
Benscoter. 2014. Potential effects of climate
Hodges, A. W., Rahmani, M., and Stevens, T. J. change on Florida’s Everglades. Environmental
2014 Economic contributions of agriculture, Management, in press.
natural resources, and related food industries in
Florida in 2012. EDIS document FE954, Food Obeysekera, J., J. Barnes and M. Nungesser. 2014.
and Resource Economics Department, UF/IFAS Climate sensitivity runs and regional hydrologic
Extension. Retrieved from: modeling for predicting the response of the
http://edis.ifas.ufl.edu/pdffiles/FE/FE95400.pdf. greater Florida Everglades ecosystem to climate
change. Environmental Management DOI
Jacoby, C. A. and T. K. Frazer. 2009. Eutrophication: 10.1007/s00267-014-0315-x.
time to adjust expectations. Science 324:723–
724. Obeysekera, J., J. Park. M. Irizarry-Ortiz, P. Trimble,
J. Barnes, J. Van Arman, W. Said and E.
Jawitz, J. W. and J. Mitchell. 2011. Temporal Gadzinski. 2011. Past and projected trends in
inequality in catchment discharge and solute climate and sea level for south Florida. West
export. Water Resources Research. 47: W00J14, Palm Beach Fla.: South Florida Water
DOI:10.1029/2010WR010197. Management District.
Johnson, R. H. and P. W. Bush. 1988. Summary of Orem. W, S, Newman, T.Z Osborne, K.R. Reddy.
the hydrology of the Floridan aquifer system in 2014. Projecting changes in Everglades soil
Florida and in parts of Georgia, South Carolina, biogeochemistry for carbon and other key
and Alabama. USGS Professional Paper 1403-A. elements, to possible 2060 climate and
Retrieved from: hydrologic scenarios. Environmental
http://sofia.usgs.gov/publications/papers/pp1403a Management, 1-23.
/.
Parker, M. L., W. S. Arnold, S. P. Geiger, P. Gorman
Lynch, S. and L. Shabman. 2011. Designing a E. H. Leone. 2013. Impacts of freshwater
payment for environmental services program for management activities on eastern oyster
the northern Everglades. National Wetlands (Crassostrea virginica) density and recruitment:
Newsletter 33:12–15. recovery and long-term stability in seven Florida
138
estuaries. Journal of Shellfish Research 32:695- SFWMD. 2009b. St. Lucie River Watershed
708. Protection Plan. Retrieved from:
http://www.sfwmd.gov/portal/page/portal/xreposit
Phlips, E. J., S. Badylak, J. Hart, D. Haunert, J. ory/sfwmd_repository_pdf/ne_slrwpp_main_1231
Lockwood, K. O’Donnell, D. Sun, P. Viveros, and 08.pdf
M. Yilmaz. 2012. Climatic influences on
autochthonous and allochthonous phytoplankton SFWMD. 2009c. Draft Documentation Report for the
blooms in a subtropical estuary, St. Lucie Reservoir Sizing and Operations Screening
Estuary, Florida, USA. Estuaries and Coasts (RESOPS) Model.
35:335-352.
SFWMD. 2009d. River of Grass Performance
Reddy, K. R. and R. D. DeLaune. 2008. Summary Maps, PowerPoint presentation.
Biogeochemistry of Wetlands: Science and Retrieved from
Applications. Boca Raton, Fla.: CRC Press. http://www.sfwmd.gov/portal/page/portal/xreposit
ory/sfwmd_repository_pdf/20_performance_sum
Reddy, K. R., R. DeLaune, and C. B. Craft. 2010. mary_maps.pdf.
Nutrients in wetlands: Implications to water
quality under changing climatic conditions. Final SFWMD. 2009e. River of Grass: methodology for
Report submitted to U. S. Environmental evaluating water quality performance of team
Protection Agency. EPA Contract No. EP-C-09- configurations. South Florida Water Management
001. District (Power Point presentation to the SFWMD
Governing Board). Retrieved from
Reddy, K.R., S. Newman, T.Z., Osborne, J.R. White, http://www.sfwmd.gov/portal/page/portal/xreposit
and H.C. Fitz. 2011. Phosphorus cycling in the ory/sfwmd_repository_pdf/rogws_060209_final.p
Everglades ecosystem: legacy phosphorus df
implications for management and restoration.
Critical Reviews in Environmental Science and SFWMD. 2010. Final Adaptive Protocols for Lake
Technology, 41: 149–186. Okeechobee Operations. South Florida Water
Management District, West Palm Beach, FL. 14
SFWMD. 1986. Regional Routing Model. Technical pp (and multiple appendices).
Publication 86-3. West Palm Beach, Fla.: South
Florida Water Management District. SFWMD. 2012. Restoration Strategies Regional
Water Quality Plan. South Florida Water
SFWMD. 2000. Minimum Flows and Levels for Lake Management District, April 2012.
Okeechobee, the Everglades and the Biscayne
Aquifer. West Palm Beach, Fla.: South Florida SFWMD. 2013. Assessing the Capacity to Discharge
Water Management District. Excess Lake Okeechobee Water South: Review
of System Operations (January through mid-June
SFWMD. 2004, Accerler8- An Overview: Stepping 2013). South Florida Water Management District,
Up the Pace to Restore America’s Everglades. West Palm Beach, FL.
Retrieved from
https://my.sfwmd.gov/acceler8/documentation/pr SFWMD. 2014. 2014 South Florida Environmental
oj_wpa.pdf. Report. Volume 1. The South Florida
Environment. South Florida Water Management
SFWMD. 2005. Documentation of the South Florida District, West Palm Beach, FL. Retrieved from:
Water Management Model Version 5.5. South http://www.sfwmd.gov/portal/page/portal/pg_grp_
Florida Water Management District, West Palm sfwmd_sfer/portlet_prevreport/2014_sfer/v1/vol1
Beach, FL. _table_of_contents.html
SFWMD. 2008. Lake Okeechobee Watershed SFWMD. 2015. Draft 2015 South Florida
Construction Project. Phase II Technical Plan. Environmental Report. The South Florida
Retrieved from: Environment. South Florida Water Management
http://www.sfwmd.gov/portal/page/portal/xreposit District, West Palm Beach, FL. Retrieved from
ory/sfwmd_repository_pdf/lakeo_watershed_cons http://www.sfwmd.gov/portal/page/portal/pg_grp_
truction%20proj_phase_ii_tech_plan.pdf . sfwmd_sfer/portlet_prevreport/2015_sfer_draft/20
15_SFER_draft_toc.html.
SFWMD. 2009a. Caloosahatchee River Watershed
Protection Plan. Retrieved from: Sime, P. 2005. St. Lucie Estuary and Indian River
http://www.sfwmd.gov/portal/page/portal/xreposit Lagoon conceptual ecological model. Wetlands
ory/sfwmd_repository_pdf/ne_crwpp_main_1231 25:898-907.
08.pdf
139
Soil Water Engineering and Technology (SWET), Environmental Impact Statement. US Army
Inc. 2008a. Legacy phosphorus abatement plan Corps of Engineers and South Florida Water
for project entitled “Technical Assistance in Management District.
Review and Analysis of Existing Data for
Evaluation of Legacy Phosphorus in the Lake USACE. 2007. Final Supplemental Environmental
Okeechobee Watershed.” West Palm Beach, Impact Statement, Including Appendices A
Fla.: South Florida Water Management District. through G, Lake Okeechobee Regulation
Schedule. US Army Corps of Engineers,
Soil Water Engineering and Technology (SWET), Jacksonville, FL.
Inc. 2008b. Task 2: Evaluation of Existing
Information. For Project Entitled: Technical USACE. 2008. Central and Southern Florida Flood
Assistance in Review and Analysis of Existing Control Project. Water Control Plan for Lake
Data for Evaluation of Legacy Phosphorus in the Okeechobee and Everglades Agricultural Area.
Lake Okeechobee Watershed. Final Report to US Army Corps of Engineers, Jacksonville, FL.
SFWMD, West Palm Beach, FL. 31 pp (and multiple appendices).
Tolley, S. G., A. K. Volety, M. Savarese, L. D. Walls, USACE. 2010. Central and Southern Florida Project
C. Linardich, and E. M. Everham III. 2006. Comprehensive Everglades Restoration Plan,
Impacts of salinity and freshwater inflow on Caloosahatchee River (C-43) West Basin
oyster-reef communities in southwest Florida. Storage Reservoir Project, Final Integrated
Aquatic Living Resources 19:371-387. Project Implementation Report and Final
Environmental Impact Statement. US Army
Twilley, R. R. 2007. Coastal wetlands and global Corps of Engineers and South Florida Water
climate change: Gulf coast wetland sustainability Management District.
in a changing climate. Report to Pew Center on
Climate Change. USACE. 2013. Central and Southern Florida Project
Comprehensive Everglades Restoration Plan,
U.S. Army. 2009. Army energy security Final Technical Data Report, Comprehensive
implementation strategy. Army Senior Energy Everglades Restoration Plan Aquifer Storage and
Council and the Office of the Deputy Assistant Recovery Pilot Project, US Army Corps of
Secretary of the Army for Energy and Engineers and South Florida Water Management
Partnerships. Retrieved from District.
http://www.asaie.army.mil/Public/Partnerships/En
ergySecurity/docs/Army_Energy_Security_Imple USACE. 2014a. Regional Model Production Scenario
mentation_Strategy_AESIS_January-2009.pdf Report, Aquifer Storage and Recovery Modeling
Study, US Army Corps of Engineers.
USACE. 1955a. Central and South Florida Project for
flood control and other purposes. US Army Corps USACE. 2014b. Comprehensive Everglades
of Engineers, Jacksonville, FL. Restoration Plan, Central Everglades Planning
Project Final Integrated Project Implementation
USACE. 1955b. Central and Southern Florida Project Report Environmental Impact Statement. US
for Flood Control and other Purposes. Part IV. Army Corps of Engineers and South Florida
Lake Okeechobee and Outlets. Supplement 2 – Water Management District.
Hydrology and Hydraulic Design. US Army Corps
of Engineers, Jacksonville, FL. USEPA. 2005. USEPA provides a regulatory
alternative for Class I Municipal Disposal Wells in
USACE. 1994. Central and Southern Florida Project, specific counties in Florida. Fact Sheet (USEPA
Reconnaissance Report, Comprehensive Review 815-F-05-033). Office of Water.
Study. US Army Corps of Engineers,
Jacksonville, FL. 237 pp. Vargas-Moreno, J. C., and Flaxman, M. 2010.
Addressing the challenges of climate change in
USACE. 1999. Central and Southern Florida Projects the greater Everglades landscape. Project Sheet
Comprehensive Review Study, Final Integrated November, 2010. Department of Urban Studies
Feasibility Report and Programmatic and Planning. MIT.
Environmental Impact Statement. US Army
Corps of Engineers and South Florida Water Walker, W. W and R. H. Kadlec. 2011. Modeling
Management District. phosphorus dynamics in the Everglades wetlands
and stormwater treatment areas. Critical Reviews
USACE. 2004. Central and Southern Florida Project Environ. Sci. Technol. 41:430-446.
Indian River Lagoon South, Final Integrated
Project Implementation Report and
140
Wang, M., C.J. Nim, S. Son, and W. Shi. 2012.
Characterization of turbidity in Florida’s Lake
Okeechobee and Caloosahatchee and St. Lucie
estuaries using MODIS-Aqua measurements.
Water Research 46:5410-5422.
141
X. Review Team Biographical Sketches
Wendy D. Graham, Ph.D., is the Carl S. Swisher Eminent Scholar in Water Resources in the
Department of Agricultural and Biological Engineering at the University of Florida, and Director
of the University of Florida Water Institute. She holds a bachelor’s degree from the University of
Florida in environmental engineering. Her PhD is in civil and environmental engineering from the
Massachusetts Institute of Technology. She conducts research in the areas of coupled
hydrologic-water quality modeling; water resources evaluation and remediation; evaluation of
impacts of agricultural production on surface and groundwater quality; and evaluation of impacts
of climate variability and climate change on water resources. She has served on the National
Research Council Committee on Independent Scientific Review of Everglades Restoration
Progress, and the Committee on Review of EPA's Economic Analysis of Final Water Quality
Standards for Nutrients for Lakes and Flowing Waters in Florida.
Mary Jane Angelo, J.D., is professor and director of the Environmental and Land Use Law
Program at the University of Florida. She holds a bachelor’s degree in biological sciences from
Rutgers University, and a master’s degree in entomology and a J.D. in law from the University
of Florida. Her academic and teaching focus is in environmental, water, biotechnology, and
wildlife protection law, as well as in agricultural policy and the environment. Prior to joining the
University of Florida, she practiced as an environmental lawyer for many years. Angelo served
in the U.S. Environmental Protection Agency Office of the Administrator and Office of General
Counsel in Washington, D.C., and as Senior Assistant General Counsel for the St. Johns River
Water Management District in Florida. She has served on the National Research Council
Committee on Independent Scientific Review of Everglades Restoration Progress.
Thomas K. Frazer, Ph.D., is a professor of ecology in the fisheries and aquatic sciences
program and director of the School of Natural Resources and Environment at the University of
Florida. He holds a bachelor’s degree in fisheries from Humboldt State University and a
master’s degree in fisheries from the University of Florida. His Ph.D. is in biological sciences
from the University of California, Santa Barbara. The overarching goals of his individual and
collaborative research efforts are to develop and transfer into management a mechanistic
understanding of the effects of anthropogenic activities on the ecology of both freshwater and
marine ecosystems. His research is, by nature, interdisciplinary and involves collaborators from
disparate disciplines and is carried across broad space and time scales in order to most
effectively address contemporary and emerging environmental issues.
Peter C. Frederick, Ph.D., is a research professor in the Department of Wildlife Ecology and
Conservation at the University of Florida. He holds a bachelor’s degree in biology from
Swarthmore College, and a Ph.D. in zoology from the University of North Carolina, Chapel Hill.
His research interests are in the ecology and conservation of wetlands, and particularly of
142
wetland vertebrates. His research lab is focused on understanding wetland processes both for
their own sake and as guides to restoration and conservation activities, and much of this work
has focused on the Everglades for the past 28 years. Frederick has served on the committees
of more than 40 graduate students, and he has authored or co-authored more than 80
publications on wetlands topics.
Karl E. Havens, Ph.D., is a professor of limnology in the fisheries and aquatic sciences
program at the University of Florida, and the director of the Florida Sea Grant College Program.
He holds a bachelor’s degree from the State University of New York, Buffalo, and a master’s
degree and Ph.D. from West Virginia University. He has 30 years of professional experience in
aquatic research, education and outreach, and has worked with Florida aquatic ecosystems and
the use of objective science in their management for the past 15 years. His area of research
specialty is plankton ecology and the eutrophication of lakes and estuaries. At the University of
Florida, he has served as chair of the UF Oil Spill Task Force, and currently is chair of the UF
Oyster Recovery Team, which is dealing with the collapse of a historic oyster fishery in
Apalachicola Bay, Florida.
K. Ramesh Reddy, Ph.D., is graduate research professor and chair of the Department of Soil
and Water Science at the University of Florida. He holds a Ph.D. in soil science with
specialization in biogeochemistry from Louisiana State University. He conducts research on
coupled biogeochemical cycling of nutrients and other contaminants in wetlands and aquatic
systems, as related to water quality, carbon sequestration, ecosystem productivity, and
restoration. He has worked on Florida’s wetlands and aquatic systems for more than 30 years.
He has served on numerous advisory committees at state, national, and international levels. He
has served on the National Research Council Committee on Soil Science and the Committee on
Independent Scientific Review of Everglades Restoration Progress. He also served on several
U.S. Environmental Protection Agency committees including the Science Advisory Board
Ecological Effects Committee, Wetland Connectivity Panel, and Lake Erie Phosphorus
Objective Panel.
143
APPENDIX E:
NEWS REPORTS
APPENDIX F:
U.S. DEPARTMENT OF THE INTERIOR STUDY
Prepared in cooperation with the U.S. Army Corps of Engineers
Understanding the Effect of Salinity Tolerance on

Cyanobacteria Associated with a Harmful Algal Bloom in
Lake Okeechobee, Florida
Scientific Investigations Report 2018–5092
U.S. Department of the Interior

U.S. Geological Survey
Cover. Dolichospermum circinale stained with SYTOX® Green. Photograph by Barry H. Rosen, U.S. Geological Survey, July 17, 2015, using an
epifluorescence microscope.
Understanding the Effect of Salinity
Tolerance on Cyanobacteria Associated
with a Harmful Algal Bloom in Lake
Okeechobee, Florida
By Barry H. Rosen, Keith A. Loftin, Jennifer L. Graham, Katherine N. Stahlhut,

James M. Riley, Brett D. Johnston, and Sarena Senegal
Prepared in cooperation with the U.S. Army Corps of Engineers
Scientific Investigations Report 2018–5092

RYAN K. ZINKE, Secretary

James F. Reilly II, Director
U.S. Geological Survey, Reston, Virginia: 2018
For more information on the USGS—the Federal source for science about the Earth, its natural and living
resources, natural hazards, and the environment—visit https://www.usgs.gov or call 1–888–ASK–USGS.
For an overview of USGS information products, including maps, imagery, and publications,
visit https://store.usgs.gov.
Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the
U.S. Government.
Although this information product, for the most part, is in the public domain, it also may contain copyrighted materials
as noted in the text. Permission to reproduce copyrighted items must be secured from the copyright owner.
Suggested citation:
Rosen, B.H., Loftin, K.A., Graham, J.L., Stahlhut, K.N., Riley, J.M., Johnston, B.D., and Senegal, S., 2018,
Understanding the effect of salinity tolerance on cyanobacteria associated with a harmful algal bloom
in Lake Okeechobee, Florida: U.S. Geological Survey Scientific Investigations Report 2018–5092, 32 p.,
https://doi.org/10.3133/sir20185092.
ISSN 2328-031X (print)

ISSN 2328-0328 (online)
ISBN 978-1-4113-4247-7
iii
Acknowledgments
This work was supported by the U.S. Army Corps of Engineers, the U.S. Geological Survey
(USGS) Priority Ecosystems Greater Everglades Program, and the USGS Environmental Health
Program. Assistance with laboratory processing was provided by Stephanie Pena.
iv
v
Contents
Acknowledgments.........................................................................................................................................iii
Abstract............................................................................................................................................................1
Introduction.....................................................................................................................................................1
Methods...........................................................................................................................................................3
Initial Bloom Material............................................................................................................................3
Biomass Indicator—Chlorophyll-a.....................................................................................................3
Biomass Indicator—Cell Numbers.....................................................................................................3
Live Organism Physiological Assessment.........................................................................................5
Microcystin Analysis by Enzyme-Linked Immunosorbent Assay..................................................5
Additional Studies..................................................................................................................................5
Results ..............................................................................................................................................................5
Initial Bloom Material............................................................................................................................5
Biomass Indicators................................................................................................................................5
Microcystin.............................................................................................................................................6
Physiological Response to Salinity...................................................................................................12
Physiological Response as Determined by SYTOX® Green Visualization..................................12
Physio-Chemical Conditions..............................................................................................................21
Discussion......................................................................................................................................................30
Summary and Conclusions..........................................................................................................................31
References.....................................................................................................................................................31
Figures
1. Map showing the location of Lake Okeechobee, the tributary system in
and out of the lake, and Eagle Bay, where samples were taken for this study,
July 7, 2017......................................................................................................................................2
2. Photograph showing area where water and bloom materials were collected
for bloom bioassay, 2017 Eagle Bay, Lake Okeechobee, Florida...........................................4
3. Photograph showing a 1-liter graduated cylinder with initial bloom material....................4
4. Photographs showing the initial cyanobacterial community in the water
sample collected July 7, 2017, from Eagle Bay, Lake Okeechobee, Florida........................6
5. Graph showing chlorophyll-a concentrations during the bioassay......................................7
6. Graph showing Microcystis aeruginosa abundance in response to salinity......................8
7. Graph showing Dolichospermum circinale abundance in response to salinity.................9
8. Graphs showing microcystin response to salinity.................................................................10
9. Photographs showing colony morphology of Microcystis aeruginosa in various
salinity treatments after 1 day of exposure.............................................................................12
10. Photographs showing colony morphology of Microcystis aeruginosa in various
salinity treatments after 4 days of exposure...........................................................................13
11. Photographs showing disintegration of some of the Microcystis aeruginosa
colonies in salinity treatments greater than 10 psu after 1 or more days of
exposure........................................................................................................................................13
12. Photographs showing response of Dolichospermum circinale filament
morphology to salinity after 1 day of exposure.......................................................................14
vi

morphology to salinity after 2 days of exposure.....................................................................14
morphology to salinity after 8 days of exposure.....................................................................14
15. Photographs showing extracellular mucilage, at arrows, in
Dolichospermum circinale, day four at 25 psu and Microcystis aeruginosa,
day three at 15 psu......................................................................................................................15
16. Photographs showing response of Microcystis aeruginosa after 1 day of
exposure to treatments with different salinities.....................................................................16
17. Photographs showing response of Microcystis aeruginosa after 2 days of
21. Photographs showing response of Dolichospermum circinale after 1 day
of exposure to treatments with different salinities................................................................22
22. Photographs showing response of Dolichospermum circinale after 8 days
of exposure to treatments with different salinities................................................................23
23. Graph showing high frequency, optical sensor data for dissolved oxygen
concentrations in three treatments..........................................................................................24
24. Graph showing high frequency sensor data for temperature.............................................25
25. Graph showing high frequency sensor data for specific conductance.............................26
26. Graph showing high frequency sensor data for pH...............................................................27
27. Graph showing high frequency, optical sensor data for chlorophyll-a
fluorescence.................................................................................................................................28
28. Graph showing high frequency, optical sensor data for phycocyanin
fluorescence.................................................................................................................................29
vii
Conversion Factors
International System of Units to U.S. customary units
Multiply By To obtain
Length
centimeter (cm) 0.3937 inch (in.)
millimeter (mm) 0.03937 inch (in.)
meter (m) 3.281 foot (ft)
kilometer (km) 0.6214 mile (mi)
kilometer (km) 0.5400 mile, nautical (nmi)
meter (m) 1.094 yard (yd)
Area
square kilometer (km2) 247.1 acre
square kilometer (km2) 0.3861 square mile (mi2)
Volume
liter (L) 33.81402 ounce, fluid (fl. oz)
liter (L) 2.113 pint (pt)
liter (L) 1.057 quart (qt)
liter (L) 0.2642 gallon (gal)
liter (L) 61.02 cubic inch (in3)
Flow rate
meter per second (m/s) 3.281 foot per second (ft/s)
Mass
gram (g) 0.03527 ounce, avoirdupois (oz)
U.S. customary units to International System of Units
Multiply By To obtain
Flow rate
foot per second (ft/s) 0.3048 meter per second (m/s)
Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as

°F = (1.8 × °C) + 32
Supplemental Information
Specific conductance is given in microsiemens per centimeter at 25 degrees Celsius (µS/cm at
25 °C).
Concentrations of chemical constituents in water are given in either milligrams per liter (mg/L)
or micrograms per liter (µg/L).
viii
Abbreviations
cells/mL cells per milliliter
FDEP Florida Department of Environmental Protection
LM light microscopy
μWm-2 microwatts per square meter
MCLR microcystin-LR
mL milliliters
μL microliter
μg microgram
µm micrometer
NaCl sodium chloride
PRSD percent relative standard deviation
psu practical salinity units
RFU relative fluorescent units
UV ultraviolet
WB wide blue
Understanding the Effect of Salinity Tolerance on
Cyanobacteria Associated with a Harmful Algal Bloom in
Lake Okeechobee, Florida
By Barry H. Rosen,1 Keith A. Loftin,1 Jennifer L. Graham,1 Katherine N. Stahlhut,1 James M. Riley,2
Brett D. Johnston,1 and Sarena Senegal1
Abstract threatens the integrity of the dike, water is released to the

Atlantic Ocean through the St. Lucie Canal and to the Gulf of
In an effort to simulate the survival of cyanobacteria as Mexico through the Caloosahatchee River. Releases follow
they are transported from Lake Okeechobee to the estuarine the Lake Okeechobee Regulation Schedule, developed and
habitats that receive waters from the lake, a bioassay implemented by the U.S. Army Corps of Engineers and the
encompassing a range of salinities was performed. An overall South Florida Water Management District (SFWMD), to
decline in cyanobacteria health in salinity treatments greater ensure the lake is at a low enough stage to accommodate
than 18 practical salinity units (psu) was indicated by loss of summer season runoff and rainfall (U.S. Army Corps of
cell membrane integrity based on SYTOX® Green staining, Engineers, 2008).
but this loss varied by the kind of cyanobacteria present. Lake Okeechobee is classified as a moderately eutrophic
Microcystis aeruginosa was tolerant of salinities up to 18 psu; (Brezonik and Engstrom, 1998) shallow waterbody that has
however, higher salinities caused leaking of microcystin from undergone ecological changes because of external nutrient
the cells. Dolichospermum circinale, another common bloom- loading from agriculture (Havens and others, 1996) and, more
former in this system, did not tolerate salinities greater than recently, by internal loading of phosphorus from lakebed
7.5 psu. Stimulation of mucilage production was observed and sediments (Pollman and James, 2011). Ample phosphorus and
is likely a mechanism used by both species to protect organism other nutrients create the ideal conditions for cyanobacterial
viability. At 7.5 psu, microcystin increased relative to blooms and have been documented in the lake since 1970 by
chlorophyll-a, providing some evidence of biosynthesis when the SFWMD (1989). The first large-scale bloom occurred in
M. aeruginosa is exposed to this salinity. This study indicates 1986, was dominated by Dolichospermum circinale (formerly
that as freshwater cyanobacteria are transported to brackish Anabaena circinalis), and covered approximately 310 km2.
and marine waters, there will be a loss of membrane integrity Some of the conditions that lead to cyanobacterial
which will lead to the release of cellular microcystin into the blooms are warm temperatures, sunlight, water-column
surrounding waterbody. Additional research would be needed stability, and sufficiently high concentrations of nitrogen
to determine the exact effect of salinity on this relationship. and (or) phosphorus. While all cyanobacteria need ample
phosphorus to thrive, inorganic nitrogen limitation allows a
subset of cyanobacteria, those capable of using atmospheric
nitrogen, to be more successful. Dolichospermum (formerly
Introduction Anabaena) has the ability to fix nitrogen (Rosen and others,
2017), compared to Microcystis, a nonnitrogen fixing genus,
Lake Okeechobee (located at 27º north latitude and that needs to acquire nitrogen from inorganic sources.
81º west longitude) is a large shallow lake (1,900 square During the summer of 2016, regulatory discharges
kilometers [km2]) that receives inflow from the Kissimmee from Lake Okeechobee into the Caloosahatchee River
River and other smaller tributaries (fig. 1). The bloom and the St. Lucie Canal and Estuary occurred during an
material for this study came from Eagle Bay on the north extensive cyanobacterial bloom (National Aeronautics and
side of the lake. An earthen dike that surrounds the lake was Space Administration, 2016). The Florida Department of
constructed to control adjacent flooding. When lake stage Environmental Protection (FDEP), in conjunction with the
Florida Fish and Wildlife Commission, reports cyanobacteria
species composition and microcystin concentrations weekly in
1
U.S. Geological Survey. Lake Okeechobee and attending waterways (FDEP, 2018). In
2
U.S. Army Corps of Engineers. late June 2016, a large cyanobacterial accumulation occurred
2 Effect of Salinity Tolerance on Cyanobacteria Associated with a Harmful Algal Bloom in Lake Okeechobee, Florida
81°30' 81° 80°30'

BREVARD
Ind
ian
OSCEOLA
Riv
er
FLORIDA
POLK ek
La
Cre ATLANTIC OCEAN
ress
go
yp
ue C
on
Bl INDIAN RIVER
Study
area
95
Ar
bu
ck
27°30'
le
HARDEE
Fiv
Cr
OKEECHOBEE
emi
ee
k
ga
kpo
le C
Isto k STLUCIE
reek
Ta
e
Cre
ylo
Fis
r
HIGHLANDS
Cr
h
eat
eek
ing
K is s
Cr
me
im
eek
Riv
er
DESOTO Eagle Bay
MARTIN l
an a
ie C
ee e
uc
Cr airi
S t. L
Pr
Port Mayaca
27°
GLADES
LAKE OKEECHOBEE
CHARLOTTE
Caloosahatchee River
Caloosahatchee
River PALMBEACH
LEE HENDRY
26°30'
COLLIER
BROWARD
75
Base from U.S. Census Bureau-TIGER, 1990 and U.S. Geological Survey EXPLANATION 0 5 10 15 20 KILOMETERS
digital data, 1999, 1:100,000
Sample sites
0 5 10 15 20 MILES
Figure 1. The location of Lake Okeechobee, the tributary system in and out of the lake, and Eagle Bay, where samples were taken
for this study, July 7, 2017.
laf18-0875_fig 01
Methods 3
in the St. Lucie Estuary, prompting the Governor of Florida bloom materials were blended and mixed thoroughly in the
to issue a state of emergency for three affected counties. The laboratory and distributed to eighteen 1-L glass graduated
FDEP data indicated that Lake Okeechobee and (or) the canal cylinders (fig. 3). The 1-L graduated cylinders were used to
system had existing populations of cyanobacteria (FDEP, simulate a water column approximately 35 centimeters (cm)
2017) that were likely transported downstream when water in depth to allow the diurnal migration typically exhibited by
was discharged from Lake Okeechobee. Local and coastal planktonic cyanobacteria. Cylinders were incubated in a west-
basin nutrient runoff, with cyanobacterial proliferation in the northwest window, and they were treated as batch reactors.
backwater areas, also contributed organisms to estuarine and Daily temperature and solar irradiance was recorded by using
marine habitats, mainly in the St. Lucie Estuary and Indian a Li-Cor LI-200/R pyranometer. Average temperature ranged
River Lagoon. In 2005, a similar event occurred with a large from 23 to 26 degrees Celsius (ºC), and light readings yielded
cyanobacteria population in the St. Lucie Estuary, which 8.9 microwatts per square meter (µWm-2; plus or minus one
occurred during regulatory discharges from Lake Okeechobee standard deviation).
(Phlips and others, 2012). Approximately 20 hours after collection, triplicate
Salinity tolerance determines how long cyanobacteria cylinders were dosed with sodium chloride (NaCl) to increase
originating from freshwater habitats will survive at the salinity by 7.5, 10, 15, or 18 g L-1 to yield 7.5, 10, 15, or
salinities typical of estuarine and marine environments 18 psu; NaCl was not added to three control cylinders. The
(Batterton and Van Baalen, 1971; Orr and others, 2004; Tonk time zero samples were taken just prior to dosing the cylinders
and others, 2007), and collectively, these studies indicate with the NaCl. Time one began after 24 hours of exposure,
that many freshwater cyanobacteria experience mortality at at approximately 9:00 every morning. The water in the
salinities between 15 and 21 practical salinity units (psu). In cylinders was thoroughly mixed by completely inverting the
the Dutch delta and its associated impoundments, such as Lake cylinders 5 times, and the volume needed for all the analytical
Volkerak in the Netherlands, water management strategies procedures was quickly poured from the cylinder. During
have been developed for Microcystis blooms (Verspagen and the time period of the experiment, no attached algae grew on
others, 2016) based on salinity tolerance in their systems. the cylinder walls. Although not part of the original planned
Using cyanobacteria isolated from Lake Okeechobee, bioassay, some of the original sample was placed into water
the U.S. Geological Survey (USGS), in cooperation with with higher salinities (20, 25, 30, and 35 psu) in graduated
the U.S. Army Corps of Engineers, conducted a 4-day cylinders 2 days after the main bioassay was started. The
laboratory bioassay with four different salinity treatments assessment of these treatments was limited to microscopic
and a control. Response variables included morphological observations and physiological assessment.
changes to cyanobacterial filaments and colonies, changes
in growth rate, physiological indicators, intracellular and
extracellular (released) microcystin, and an approximation Biomass Indicator—Chlorophyll-a
of cell membrane integrity. Collectively, these data provide
converging lines of evidence about how Lake Okeechobee Chlorophyll provided an overall indication of algal
cyanobacteria might respond to increased salinity under biomass. Ten-milliliter (mL) subsamples collected from
natural conditions as water mixes in downstream estuaries. each cylinder daily were filtered as described in Hambrook
Additional data related to this study can be found in King Berkman and Canova (2007), stored (frozen with desiccant),
and others (2018a, b). and quantified by using a modification of U.S. Environmental
Protection Agency Method 445.0 (Arar and Collins, 1997).
Instead of acetone extraction, samples were extracted in
heated ethanol (Sartory and Grobbelar, 1986), and the
Methods fluorometer was modified with a flow-through cell (Knowlton,
1984). Samples were analyzed in duplicate, and the results
The organisms used for the study were collected from were reported as an average.
Lake Okeechobee and brought into the laboratory in Orlando,
Fla., for this study.
Biomass Indicator—Cell Numbers
Initial Bloom Material A 10-milliliter (mL) sample from each of the 18 cylinders
was collected by 10-mL graduated pipette dispensed into
Water and floating bloom material were collected a 15-mL screw-cap plastic centrifuge tube, preserved with
from the upper 10 cm of the water surface at noon on 100 microliters (µL) of Lugol’s iodine, and refrigerated at
July 7, 2017, from Eagle Bay, Lake Okeechobee, Fla. (latitude 4 ºC until it was processed. For counting, the samples were
27°11'42.19"N, longitude 80°49'46.24"W). Approximately homogenized by vigorous shaking to disperse the colonies, an
35 liters (L) of water and bloom material (fig. 2) were aliquot was withdrawn, and the precise weight was determined
collected in two large carboys and immediately transported (x 0.0001 gram [g]) with an Ohaus Explorer EX224 Analytical
to the laboratory, which was approximately 3 hours away, Electronic Balance. The weight was considered equivalent
without being cooled or held in the dark. All water and to the volume (1 mL = 1 g) and evenly distributed under a
Figure 2. Area where water and bloom

materials were collected for bloom bioassay,
2017 Eagle Bay, Lake Okeechobee, Florida.
(Photograph by Barry H. Rosen, U.S. Geological
Survey.)
Figure 3. A 1-liter graduated cylinder with

initial bloom material. (Photograph by Barry H.
Rosen, U.S. Geological Survey.)
Results 5
22-square-millimeter (mm2) glass coverslip. Counting was Additional Studies

conducted microscopically at 400x by enumerating the
cyanobacteria observed in a linear strip along the full length The influence of circulation was evaluated in unamended
of the coverslip. The diameter of the strip was measured with water and bloom material. Three 4-L glass beakers were filled
a stage micrometer. Using the volume distributed under the with 2.5 L of water and bloom material, set atop magnetic
coverslip per unit area, the number of organisms per unit stir plates, and separately circulated at velocities of 0.00 foot
volume was calculated. For Microcystis, its dense, colonial per second (ft/s; “nonstirred”), 0.27 ft/s, and 0.95 ft/s. The
beakers were sealed with plastic to prevent evaporation.
form does not allow an exact direct cell count. We were able
Velocity was measured with a USGS pygmy current meter
to enumerate the number of colonies and optically image
by using the 0.6-depth method (Turnipseed and Sauer, 2010).
a subset of colonies to obtain an average number of cells
High-frequency sensor data were collected at 5-minute
per colony. This average number of cells per colony was intervals. Data were collected by using a multiparameter
multiplied by the number of colonies found in each sample to sonde (Xylem/YSI EXO2), calibrated in accordance with
get estimated cell count. manufacturer protocols (YSI Incorporated, 2017) to measure
water temperature, specific conductance, dissolved oxygen,
pH, chlorophyll-a fluorescence, and phycocyanin fluorescence.
Live Organism Physiological Assessment The high-resolution visualization resulting from these data
Approximately 0.020 mL of live material taken from provides insight into the biogeochemical processes in relation
to the results from less frequently collected discrete analyses
each salinity treatment was examined daily to assess the
(Downing and others, 2017).
health of cells by following the protocol of Rosen and others
(2010). Microscopic observations were made with a BX51
Olympus microscope with differential interference contrast
and epifluorescence using ultraviolet and wide blue excitation Results
sources. SYTOX® Green (Invitrogen, Carlsbad, Calif.), a
The majority of the results center on the salinity bioassay
deoxyribonucleic acid (DNA) stain excluded from live cells,
conducted; however, ancillary data were collected from the
was added (Rosen and others, 2010) to assess if cells lost
same water and bloom material to provide some understanding
cellular membrane integrity. of how stirred conditions affect physio-chemical parameters.
Microcystin Analysis by Enzyme-Linked Initial Bloom Material

Immunosorbent Assay
The bloom material collected from Lake Okeechobee
For analysis of total microcystin, a 10-mL subsample on July 7, 2017, contained three species of Microcystis: M.
was taken daily from each cylinder and frozen. For analysis aeruginosa, M. flosaquae, and M. wesenbergii. Microcystis
of dissolved microcystin, the filtrate (Millipore Type TSTP aeruginosa was the most frequently observed species of
3.0-micrometer [µm], 25-millimeter [mm] diameter) from this genus. Also present were Dolichospermum circinale,
a 10-mL subsample was collected daily from each cylinder Planktolyngbya contorta, Planktolyngbya limnetica, and
Cuspidothrix tropicalis (fig. 4). Microcystin concentration in
and frozen. These samples were processed and analyzed as
the surface scum collected directly from Lake Okeechobee
described in Loftin and others (2016), except that an Abraxis
was 560 µg/L, indicating the presence of microcystin-
streptavidin amplified enhanced sensitivity (SAES) enzyme-
producing species.
linked immunosorbent assay kit was used with a 5-psu sodium Although rare in the initial community of organisms,
chloride microcystin-LR (MCLR) calibration curve custom two additional cyanobacteria genera grew during the
made by Abraxis, Inc. (Warminster, Pa.). A four-parameter experimental treatments. Planktothrix was found on day
curve-fit was used for calibration. Measurements of kit diluent eight of the experiment in water with salinity of 15 psu, and
and laboratory reagent water were below the kit minimum Cuspidothrix tropicalis was found on day four in water with
reporting level (0.10 microgram per liter [µg/L]), 0.75-µg/L salinity up to 18 psu.
MCLR kit controls were analyzed every 10th sample, and
28 percent of samples were laboratory replicates. Mean kit
Biomass Indicators
control recovery was 106 percent, mean percent relative
standard deviation (PRSD) was 9.6 percent, and laboratory Chlorophyll-a concentrations and cyanobacterial
replicate PRSD was 20 percent. Particulate microcystin abundance were used to quantify changes in algal biomass
was calculated by subtracting dissolved concentration during the bioassay (figs. 5–7). For chlorophyll-a, all
from total concentration. concentrations declined from day zero to day four, with the
A B C
D E F
Figure 4. The initial cyanobacterial community in the water sample collected July 7, 2017, from Eagle Bay, Lake Okeechobee, Florida.
A, Microcystis aeruginosa (most abundant); B, M. flosaquae; C, M. wesenbergii; D, Dolichospermum circinale; E, Planktolyngbya
contorta; F, Cuspidothrix tropicalis; G, Planktolyngbya limnetica.
greatest declines in the control (0 psu) and 7.5-psu treatments The dissolved phase (fig. 8B) is a combination of
(fig. 5). The 10, 15, and 18 psu treatments showed greater microcystin leaked from live or unhealthy cells and from
amounts of chlorophyll-a than the lower salinity treatments. those cells that died and released microcystin. In the control,
Microcystis aeruginosa and Dolichospermum circinale the amount of dissolved-phase microcystin remained less
were the dominant cyanobacteria in the initial community than 2 µg/L, with the largest amounts in the 15- and 18-psu
used for the bioassay. For Microcystis, all of the treatments treatments (12–13 µg/L) and in the 7.5- and 10-psu treatments
showed an initial increase in the number of cells during the (8–10 µg/L). At all the salinities greater than the control, the
first time interval from time zero through the first 24 hours, amount of dissolved microcystin increased over time (fig. 8B),
and then the number of cells decreased during the next indicating that some of the cells in the treatments were
24-hour time period (fig. 6), including the control treatment leaking cellular microcystin and cell lysis was occurring. The
with no added salt. Because the bloom collected from Eagle partitioning of microcystin between the amount retained in
Bay had been thriving in full sunlight, this adaptation to a cells (calculated particulate) and the dissolved phase (fig. 8C)
laboratory setting was an anticipated response. has a similar pattern as the total microcystin (fig. 8A), with
Dolichospermum circinale abundance in the control and more retained or produced in the 7.5- and 10-psu treatments.
all of the treatments substantially decreased during the first The calculated particulate microcystin to chlorophyll-a
24 hours of the bioassay (fig. 7). ratio was examined (fig. 8D) to normalize the toxin
concentration to a cell abundance surrogate (chlorophyll-a).
Cell abundance was not used for this calculation because
Microcystin chlorophyll-a was a more precise measurement then cell
abundance, given the difficulty and potential variability in
Microcystin was reported as a total concentration in estimating cell numbers. In addition, particulate microcystin
whole water and as a dissolved phase in the water after the and particulate chlorophyll-a are both intracellular
cells were removed by filtration. Total concentration (fig. 8A) constituents. Using this calculation, by day four, the
decreased to less than 20 µg/L in the control. The 7.5- and 7.5- and 10-psu treatments showed an increase in this
10-psu salinity treatments initially declined but increased to ratio that is approximately double the ratios in the 15- and
more than 40 µg/L by the end of the 4-day bioassay. 18-psu treatments.
Results 7
450
400
Chlorophyll-a, in micrograms per liter
350
300
250
200
150
100
0 1 2 3 4
Days
EXPLANATION
Standard error of 7.5 psu

the mean
10 psu
0 Practical salinity 15 psu
units (psu)
18 psu
Figure 5. Chlorophyll-a concentrations during the bioassay.

1,200,000
1,000,000
800,000
Cells, per milliliter
600,000
400,000
200,000
0
0 1 2 3 4
Days
EXPLANATION

the mean
10 psu
units (psu)
18 psu
Figure 6. Microcystis aeruginosa abundance in response to salinity. The error bars are standard error of the mean.
Results 9
180,000
160,000
140,000
120,000
Cells, per milliliter
100,000
80,000
60,000
40,000
20,000
0
0 1 2 3 4
Days
EXPLANATION

the mean
10 psu
units (psu)
18 psu
Figure 7. Dolichospermum circinale abundance in response to salinity. The error bars are standard error of the mean.
A
50
45
Microcystin concentration, in micrograms per liter
40
35
30
25
20
15
10
0
0 1 2 3 4
Days
B
16
14
12
10
0
0 1 2 3 4
Days
EXPLANATION

the mean
10 psu
units (psu)
18 psu
Figure 8. Microcystin response to salinity: A, total microcystin; B, dissolved phase; C, calculated particulate; D, calculated
particulate microcystin:chlorophyll-a ratio. The error bars are standard error of the mean.
Results 11
C
50
45
40
35
30
25
20
EXPLANATION
15
10 the mean
10 psu
5 units (psu)
18 psu
0
0 1 2 3 4
Days
D
0.25
Ratio of particulate microcystin to chlorophyll-a
0.2
0.15
0.1
0.05
0
0 1 2 3 4
Days
EXPLANATION

the mean
10 psu
units (psu)
18 psu
Figure 8. Microcystin response to salinity: A, total microcystin; B, dissolved phase; C, calculated particulate; D,
calculated particulate microcystin:chlorophyll-a ratio. The error bars are standard error of the mean.—Continued
Physiological Response to Salinity penetrate, providing an indication of cell health. In figures

16–22, the denotation is (1) LM—organisms illuminated
Overall, the Microcystis aeruginosa colonies maintained with differential interference light microscopy for overall cell
integrity during the first 24 hours of exposure at salinities and colony structure; (2) UV—organisms illuminated with
up to 25 psu (fig. 9; 30 and 35 psu were not photographed at ultraviolet light, with variations in pigment color an indicator
this time interval). After 4 days, complete disintegration of of cell health; (3) WB—organisms illuminated with “wide
all Microcystis aeruginosa colonies did not occur, even up to blue” as the baseline color before adding SYTOX® Green; (4)
35 psu (fig. 10). Although most colonies remained intact, a few SYTOX® Green—organisms stained with this dye are bright
disintegrated to small clusters and individual cells (fig. 11). green when DNA is present (Rosen and others, 2010). The
Dolichospermum circinale was much less abundant, photographs are representative of the overall condition of the
leaving a less complete picture of salinity effects. Intact colonies or filaments in the treatments.
filaments were present after 1 day; however, the cells seem to After 1 day of exposure, the SYTOX® Green is only seen
be separating in the 20- and 25-psu treatments (fig. 12). After outside of the Microcystis colony, with no penetration of the
2 days of exposure, Dolichospermum has additional separation stain into the cells in all treatments (fig. 16). After 2 days of
of cells at higher salinity (fig. 13). Dolichospermum circinale exposure, the SYTOX® Green penetrated the Microcystis cells
was tolerant of 7.5 psu and was in good condition after 8 days in the 18-, 20-, and 25-psu treatments as seen in figure 17
(fig. 14), but was not found at any salinity greater than 7.5 psu where the cells are green compared to the 15-psu treatment.
by the end of the bioassay. After 3 days of exposure, the SYTOX® Green is seen only
In addition to overall morphological changes in colonies
external to the Microcystis colonies, with no penetration of
and filaments of these two genera of cyanobacteria, evidence
the stain into the cells in all treatments, with the exception
of adaptation was observed microscopically. The production
of the 15-psu treatment (fig. 18). In the 15-psu treatment, the
of mucilage was observed enveloping the filaments and the
SYTOX® Green penetration and the pigment shift to yellow
colonies (fig. 15).
under UV indicate that this colony was no longer viable. After
4 days of exposure, with one of two Microcystis colonies in
Physiological Response as Determined by the 15-psu treatment and all of the colonies in treatments of
SYTOX® Green Visualization 20 psu or greater salinity, the SYTOX® Green penetration
and the pigment shift to yellow under UV indicate that these
Cell health was analyzed by using epifluorescence colonies were no longer viable (fig. 19). After 8 days of
microscopy in conjunction with SYTOX® Green, a DNA stain exposure, the 18-psu treatments had some Microcystis colonies
that emits green light in the presence of DNA. If the cellular that were still viable and some that were not, while all lower
membrane of an organism is intact, SYTOX® Green does not salinity treatments had viable colonies (fig. 20).
0 psu 7.5 psu 10 psu 15 psu 18 psu
20 psu 25 psu
Figure 9. Colony morphology of Microcystis aeruginosa in various salinity treatments after 1 day of exposure.
Results 13
0 psu 7.5 psu 10 psu 15 psu 18 psu
20 psu 25 psu 30 psu 35 psu
Figure 10. Colony morphology of Microcystis aeruginosa in various salinity treatments after 4 days of exposure.
10 psu, 72 hours 18 psu, 24 hours
20 psu, 48 hours 25 psu, 48 hours
Figure 11. Disintegration of some of the Microcystis aeruginosa colonies in salinity treatments greater than 10 psu after 1 or more
days of exposure.
7.5 psu 10 psu 20 psu 25 psu
Figure 12. Response of Dolichospermum circinale filament morphology to salinity after 1 day of exposure.
0 psu 20 psu 25 psu
Figure 13. Response of Dolichospermum circinale filament morphology to salinity after 2 days of exposure.
0 psu 7.5 psu
Figure 14. Response of Dolichospermum circinale filament morphology to salinity after 8 days of exposure.
Results 15
Figure 15. Extracellular mucilage, at arrows, in A, Dolichospermum circinale, day four at 25 psu; B, Microcystis
aeruginosa, day three at 15 psu.
0 psu
7.5 psu
10 psu
15 psu
18 psu
Figure 16. Response of

Microcystis aeruginosa
after 1 day of exposure
to treatments with
20 psu different salinities
(psu, practical salinity
units; LM, differential
interference microscopy;
UV, epifluorescence
microscopy; WB, wide-
blue epifluorescence
microscopy; WB and
25 psu SYTOX® Green, wide-
blue microscopy with
the DNA stain SYTOX®
Green added. Blank boxes
indicate that data were not
available).
LM UV WB WB and SYTOX® Green
Results 17
0 psu
7.5 psu
10 psu
15 psu
18 psu

20 psu after 2 days of exposure
to treatments with
different salinities
UV, epifluorescence
25 psu blue epifluorescence
microscopy; WB and
SYTOX® Green, wide-
blue microscopy with the
DNA stain SYTOX® Green
added).
0 psu
7.5 psu
10 psu
15 psu
18 psu

20 psu after 3 days of exposure
to treatments with
UV, epifluorescence
25 psu blue epifluorescence
microscopy; WB and
SYTOX® Green, wide-
blue microscopy with
the DNA stain SYTOX®
Green added).
Results 19
0 psu
7.5 psu
10 psu
15 psu
18 psu
20 psu
25 psu

after 4 days of exposure
to treatments with
30 psu
UV, epifluorescence
35 psu
blue epifluorescence
microscopy; WB and
SYTOX® Green, wide-blue
microscopy with the DNA
LM UV WB WB and SYTOX® Green stain SYTOX® Green added).
0 psu
7.5 psu
10 psu
15 psu
18 psu
Figure 20. Response of Microcystis aeruginosa after 8 days of exposure to treatments with different salinities (psu, practical
salinity units; LM, differential interference microscopy; UV, epifluorescence microscopy; WB, wide-blue epifluorescence
microscopy; WB and SYTOX® Green, wide-blue microscopy with the DNA stain SYTOX® Green added. Blank boxes indicate that
data were not available).
Results 21
After 1 day of exposure, Dolichospermum was much less Temperature varied diurnally from a low of 23 ºC
abundant in the treatments, hence the sparse observations (fig. to a high slightly greater than 27 ºC in the nonstirred
21). The Green was observed to have penetrated one of two treatment (fig. 24). Water in both stirred beakers (on a
filaments observed after 1 day of exposure at 20 psu. magnetic plate with a stir bar) had elevated temperatures
Dolichospermum filaments were seen in the control and that were 2 to 4 ºC higher than temperatures in the
in the 7.5-psu treatment, but were absent in salinity greater nonstirred treatment.
than 7.5 psu after day one. After 8 days, the only treatments In the nonstirred treatment, specific conductance was
307 microsiemens per centimeter at 25 degrees Celsius (µS/
with any Dolichospermum were the control and the 7.5-psu
cm at 25 °C) on the first day of the test; specific conductance
treatment, and Dolichospermum were rare in both samples
increased to 322 µS/cm by the end of the test (fig. 25). The
(fig. 22). The few filaments observed in both of these
stirred treatments had the same initial specific conductance as
treatments were viable and appeared healthy. that of the nonstirred treatment; however, specific conductance
was approximately 30 µS/cm greater in the stirred treatments
Physio-Chemical Conditions than in the nonstirred treatment by the end of the test.
The pH peaked at just over 7.3 in the nonstirred treatment
Ancillary to the bioassay itself, a continuously recording (fig. 26). In the stirred treatments, pH values were higher on
sonde was used to collect physical and water-chemistry data average. In the treatment that was stirred at a rate of 0.27 ft/s,
from water associated with the initial cyanobacteria sample the maximum pH was 7.6. In the treatment that was stirred at a
collection and a simulation of water flow without any salinity rate of 0.95 ft/s, the maximum pH was 7.9.
For all treatments, chlorophyll-a fluorescence exhibited
treatments. Material was placed in 4-L beakers that were either
the greatest decrease during the first day as the bloom was
stirred or not stirred. The nonstirred treatment is the most
moved from Lake Okeechobee to the laboratory setting
similar to the actual bioassay in the sense that the graduated
(fig. 27). The decrease of chlorophyll-a fluorescence in
cylinders used for the bioassay were not stirred. Dissolved the nonstirred treatment leveled off for the duration of the
oxygen initially showed a small increase to 4 milligrams per experiment; however, fluorescence in both stirred treatments
liter (mg/L), then decreased and maintained a fairly constant was higher than in the nonstirred treatment, and chlorophyll-a
level of 3 mg/L in the nonstirred treatment. A slight diurnal fluorescence increased in the stirred treatments after day three.
pattern was present that dissipated after day two (fig. 23). Phycocyanin fluorescence decreased substantially across
Both stirred treatments had higher dissolved oxygen than the all treatments over the course of the experiment, though
nonstirred treatment, with the highest concentration of oxygen fluorescence did remain higher in the treatment that was
at the fastest rate of stirring, 0.95 ft/s, reaching 8 mg/L. stirred at the fastest rate (0.95 ft/s; fig. 28).
0 psu
7.5 psu
10 psu
20 psu
25 psu
Figure 21. Response of Dolichospermum circinale after 1 day of exposure to treatments with different salinities (LM, differential
interference microscopy; UV, epifluorescence microscopy; WB, wide-blue epifluorescence microscopy; WB and SYTOX® Green,
wide-blue microscopy with the DNA stain SYTOX® Green added. Blank boxes indicate that data were not available).
Results 23
0 psu
7.5 psu
Figure 22. Response of Dolichospermum circinale after 8 days of exposure to treatments with different salinities (filament at
blue arrow for image at upper right, among the Microcystis colonies; psu, practical salinity units; LM, differential interference
microscopy; UV, epifluorescence microscopy; WB, wide-blue epifluorescence microscopy; WB and SYTOX® Green, wide-blue
microscopy with the DNA stain SYTOX® Green added).
7
Dissolved oxygen, in milligrams per liter
2 EXPLANATION
Treatment, stirred at
0.95 foot per second
1
0
07/10/17 07/11/17 07/11/17 07/12/17 07/12/17 07/13/17 07/13/17 07/14/17
12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Date and time
Figure 23. High frequency, optical sensor data for dissolved oxygen concentrations in three treatments: one not stirred, one
stirred at 0.27 foot per second (ft/s), and one stirred at 0.95 ft/s.
laf18-0875_fig 23
Results 25
35
33
31
29
Temperature, in degrees Celsius
27
25
23
21
EXPLANATION
19
17 0.27 foot per second
15
07/10/17 07/11/17 07/11/17 07/12/17 07/12/17 07/13/17 07/13/17 07/14/17
12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Date and time
Figure 24. High frequency sensor data for temperature.
laf18-0875_fig 24
335
Specific conductance, in microsiemens per centimeter at 25 degrees Celsius
330
325
320
315
EXPLANATION
310
305
07/10/17 07/11/17 07/11/17 07/12/17 07/12/17 07/13/17 07/13/17 07/14/17
12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Date and time
Figure 25. High frequency sensor data for specific conductance.
laf18-0875_fig 25
Results 27
8.1
8.0
7.9
7.8
7.7
7.6
pH
7.5
EXPLANATION
7.4
7.3 0.00 foot per second
7.2
7.1
07/10/17 07/11/17 07/11/17 07/12/17 07/12/17 07/13/17 07/13/17 07/14/17
12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Date and time
Figure 26. High frequency sensor data for pH.
laf18-0875_fig 26
6.5 EXPLANATION
Chlorophyll-a, in relative fluorescence units (RFU)
5.5
4.5
3.5
2.5
1.5
07/10/17 07/11/17 07/11/17 07/12/17 07/12/17 07/13/17 07/13/17 07/14/17
12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Date and time
Figure 27. High frequency, optical sensor data for chlorophyll-a fluorescence.
laf18-0875_fig 27
Results 29
30
EXPLANATION
25 0.95 foot per second
Phycocyanin, in relative fluorescence units (RFU)
20
15
10
0
07/10/17 07/11/17 07/11/17 07/12/17 07/12/17 07/13/17 07/13/17 07/14/17
12:00 00:00 12:00 00:00 12:00 00:00 12:00 00:00
Date and time
Figure 28. High frequency, optical sensor data for phycocyanin fluorescence.
laf18-0875_fig 28
Discussion released to the water as a result of salinity stress. Another

major bloom-forming organism from the lake (Rosen and
It is frequently assumed that when cyanobacteria from others, 2017), Dolichospermum circinale, was present at
freshwater lakes, reservoirs, and rivers reach saline estuaries, an initial density that was more than 140,000 cells/mL. The
they will die off after some period of time. The hydrologic response of these two organisms provided information on the
dynamics in these transition areas influence the salinity colonial and filamentous forms of cyanobacteria, Microcystis
conditions that these organisms experience. In a study that and Dolichospermum, respectively.
included a period of augmented flow from Lake Okeechobee Generally, cyanobacteria health declined in salinity
in 2005, Phlips and others (2012) found that Microcystis treatments greater than 18 psu as indicated by a loss of
aeruginosa persisted in the St. Lucie Estuary. The Phlips cell membrane integrity. In any given waterbody, the
study reported that the South Fork of the St. Lucie Estuary cyanobacteria colonies and filaments are not identical; rather,
had dense surface accumulations of cyanobacteria, with they are in various physiological states. Some colonies and
chlorophyll-a reaching 2,863 µg/L–1 at the surface, and the filaments contain healthy cells that are likely better able to
cyanobacteria seem to have come from the St. Lucie Canal. withstand stress, such as salinity changes, while others are
Only a few studies have attempted to define the effects of unhealthy and more easily disrupted. Mucilage production,
salinity on cell growth, cell death, toxicity, pigmentation, and a mechanism commonly observed in response to desiccation
cell morphology (Batterton and Van Baalen, 1971; Orr and in cyanobacteria (Pereira and others, 2009; Rosen and
others, 2004). Tonk (Tonk and others, 2007) found the salinity Mareš, 2016) and salinity (Ozturk and Aslim, 2010), was
tolerance of Microcystis did not exceed 20 psu. also an indicator of response to salinity in Microcystis and
The bloom used for this bioassay was collected Dolichospermum (fig. 15). For mucilage production to
from Eagle Bay, Lake Okeechobee, on July 7, 2017, and occur, the filaments/colonies have to be alive and able to
brought into a laboratory setting. The bloom was moved synthesize compounds that are excreted from the cells. The
from an open-water habitat where it experienced wind and extracellular polymeric substance is mainly composed of
wave activity and nutrient and light variability associated complex polysaccharides that serve as a protective boundary
with an open waterbody. Moving the community from between the cells and the water that surrounds them (Kehr
its natural environment into the laboratory and into batch and Dittmann, 2015). Nitrogen limitation has also been found
growth conditions (the graduated cylinders) altered many to increase polysaccharide production in Microcystis cells,
parameters that would equate to physiological stress on the and light intensity seems to play an important role (Yang and
organisms; however, the approach used in this study ensured others, 2012).
physiological variables were similar across the treatments, Initially, individual Microcystis colonies contained
leaving salinity changes as the only variable. The responses of hundreds of cells (fig. 4A), and the effect of salinity on
dissolved oxygen and pH indicate the loss of a typical diurnal colony integrity did not show a single pattern. Some colonies
pattern that likely existed in the lake (figs. 23 and 26). The maintain their overall shape with several hundreds of cells,
increase in specific conductance (fig. 25) may be explained even after 4 days of exposure in treatments up to 35 psu
by greater evaporation in the stirred treatments, which were (fig. 10), while other colonies broke up into individual cells at
also warmer than the nonstirred treatment (fig. 24) because 10 psu or greater (fig. 11). Although some cells of Microcystis
of the heat produced by the stir plates. On a cellular level, if lost integrity after 2 days of exposure to salinity greater than
the cells were being disrupted and leaking cellular contents 18 psu, as indicated by being stained green with SYTOX®
mineralized by bacteria, it could also explain the increase in Green (fig. 17), other cells and colonies were left intact even
specific conductance, as well as being a possible factor for the after 4 days of exposure to salinity up to 20 psu. Cell mortality
observed decline in fluorescence. occurred after 4 days of exposure to salinity of 25 psu or
Overall, the control and treatments showed a downward greater (fig. 19), even if the colonies appeared to be intact.
trend in chlorophyll-a and phycocyanin fluorescence over This range in colony breakdown in response to salinity is
time (figs. 27 and 28), and numerous possibilities exist likely because of the natural variability in the health of each
for this phenomenon, including moving organisms from a individual colony in a bloom, similar to the concept of the
high-light environment to a laboratory setting and changing “paradox of the plankton” (Hutchinson, 1961).
nutrient dynamics that affect pigment fluorescence (Rosen and Dolichospermum circinale cells decreased substantially
Lowe, 1984). in all treatments during the first 24 hours, indicating this
The water collected contained the targeted organism, species is not tolerant of the salinities in the ranges tested
Microcystis aeruginosa, with a density of approximately (fig. 7). Filaments were initially coiled (fig. 4D), but became
400,000 cells per milliliter (cells/mL). This original sample shortened fragments within 1 day of salinity exposure
contained ample microcystin (560 µg/L), which allowed us to (fig. 12). Separation of the cells in a filament was observed
monitor the presence of microcystin in cells and the amount and likely the cause of fragmentation. Dolichospermum
References 31
tolerated the lowest salinity treatment, 7.5 psu (fig. 14B),

for 8 days, but filaments were not observed in the higher
References
salinities in this timeframe. From these observations, this other
common Lake Okeechobee bloom-forming organism is much Arar, E.J., and Collins, G.B., 1997, U.S. Environmental
less tolerant of salinity compared to Microcystis. Although Protection Agency Method 445.0, in vitro determination of
some of the fragments were stained with SYTOX® Green chlorophyll-a and pheophytin-a in marine and freshwater
algae by fluorescence, revision 1.2: Office of Research and
after 2 days of exposure to the 7.5-psu treatment, this salinity
Development, 22 p.
does not appear to cause complete mortality, as indicated
by filamentous colonies that were still viable after 8 days at Batterton, J.C., Jr., and Van Baalen, C., 1971, Growth
7.5 psu. Dolichospermum circinale did not tolerate conditions responses of blue-green algae to sodium chloride
greater than 7.5 psu. concentration: Archiv fur Mikrobiologie, v. 76, p. 151–165.
Total, dissolved, or calculated microcystin
concentrations, as total, dissolved, or calculated particulate, Brezonik, P.L., and Engstrom, D.R., 1998, Modern and
increased over time and were greater in the salinity treatments historic accumulation rates of phosphorus in Lake
compared to the control. The amount of microcystin was Okeechobee, Florida: Journal of Paleolimnology, v. 20,
greatest in the dissolved phase at the two highest salinities p. 31–46.
(fig. 8B), indicating that either a greater amount leaked from Downing, B.D., Bergamaschi, B.A., and Kraus, T.E.C.,
the cells in general or that cellular contents were released 2017, Synthesis of data from high-frequency nutrient and
by dead cells. The particulate microcystin and chlorophyll-a associated biogeochemical monitoring in the Sacramento–
give us information about the cellular content of organisms, San Joaquin Delta, northern California: U.S. Geological
hence we gain a more direct understanding of the cellular Survey Scientific Investigations Report 2017–5066, 28 p.
microcystin response as a function of salinity. The 7.5- and
10-psu treatments contained a greater amount of microcystin Florida Department of Environmental Protection
per unit of chlorophyll, indicating toxin stimulation at these [FDEP], 2017, June 2016 sampling results, accessed
concentrations. In contrast, in the 15- and 18-psu salinity January 29, 2018, at https://floridadep.gov/comm/press-
treatments, the microcystin to chlorophyll ratios were similar office/documents/june-2016-sampling-results.
to or lower than in the control, suggesting the inhibition
Florida Department of Environmental Protection [FDEP],
of microcystin at these salinities. Additional experiments
2018, Algal bloom monitoring and response, accessed
are needed to determine the mechanisms by which salinity
February 8, 2018, at https://floridadep.gov/dear/algal-
stimulates or inhibits microcystin production. bloom.
Hambrook Berkman, J.A., and Canova, M.G., 2007, Algal

biomass indicators (ver. 1.0): U.S. Geological Survey
Summary and Conclusions Techniques of Water-Resources Investigations, book 9,
chap. A7, section 7.4, accessed December 10, 2017, at
This study documents an overall decline in cyanobacteria
https://water.usgs.gov/owq/FieldManual/Chapter7/7.4.pdf.
health in salinity treatments greater than 18 psu. A dominant
bloom in this system, Microcystis aeruginosa, was tolerant Havens, K.E., Aumen, N.G., James, R.T., and Smith, V.H.,
of salinities up to 18 psu; however, higher salinities caused 1996, Rapid ecological changes in a large subtropical
leaking of microcystin from the cells. Dolichospermum lake undergoing cultural eutrophication: Ambio, v. 25,
circinale, another common bloom-former in this system, did p. 150–155.
not tolerate salinities greater than 7.5 psu. Both organisms
produced extracellular mucilage as a protective response Hutchinson, G.E., 1961, The paradox of the plankton: The
to salinity, which is a common ecological strategy in American Naturalist, v. 95, no. 882, p. 137–145.
cyanobacteria undergoing desiccation. At 7.5 psu, microcystin Kehr, J.C., and Dittmann, E., 2015, Biosynthesis and function
increased relative to chlorophyll-a, providing some evidence of extracellular glycans in cyanobacteria: Life (Basel), v. 5,
of biosynthesis when M. aeruginosa is stressed at these no. 1, p. 164–180.
salinities. This study indicates that as freshwater cyanobacteria
are transported to brackish and marine waters, there will be King, L.R., Rosen, B.H., Loftin, K.A., Graham, J.L., Stahlhut,
a loss of membrane integrity which will lead to the release K.N., Johnston, B.D., and Senegal, Sarena, 2018a,
of cellular microcystin into the surrounding waterbody. Microcystin, chlorophyll, and cell-count data for assessing
Additional research would be needed to determine the exact the effect of salinity tolerance on cyanobacteria associated
effect of salinity on this relation. with a harmful algal bloom in Lake Okeechobee, Florida,
July 9 to 17, 2017: U.S. Geological Survey data release,
https://doi.org/10.5066/P9FGIVCD.
King, L.R., Rosen, B.H., Loftin, K.A., Graham, J.L., Stahlhut, Rosen, B.H., Loftin, K.A., Smith, C.E., Lane, R.F., and
K.N., Johnston, B.D., and Senegal, Sarena, 2018b, Velocity Keydel, S.P., 2010, Microphotographs of cyanobacteria
test data for assessing the effect of salinity tolerance on documenting the effects of various cell-lysis techniques:
cyanobacteria associated with a harmful algal bloom in Lake U.S. Geological Survey Open-File Report 2010–1289,
Okeechobee, Florida, July 10 to 13, 2017: U.S. Geological 203 p.
Survey data release, https://doi.org/10.5066/P996GAZP.
Rosen, B.H., and Lowe, R.L., 1984, Physiological and
Knowlton, M.F., 1984, Flow-through microcuvette for ultrastructural responses of Cyclotella meneghiniana
fluorometric determination of chlorophyll: Water Resources (Bacillariophyta) to light intensity and nutrient limitation:
Bulletin, v. 20, p. 1198–1205. Journal of Phycology, v. 20, p. 173–183.
Loftin, K.A., Graham, J.L., Hilborn, E.D., Lehmann, S.C., Rosen, B.H., and Mareš, Jan, 2016, Catalog of microscopic
Meyer, M.T., Dietze, J.E., and Griffith, C.B., 2016, organisms of the Everglades, Part 1—The cyanobacteria:
Cyanotoxins in inland lakes of the United States: Occurrence U.S. Geological Survey Open-File Report 2016–1114,
and potential recreational health risks in the EPA National 108 p.
Lakes Assessment 2007: Harmful Algae, v. 56, p. 77–90.
Sartory, D.P., and Grobbelar, J.U., 1986, Extraction of
National Aeronautics and Space Administration, 2016, chlorophyll-a from freshwater phytoplankton for
Earth Observatory, Bloom in Lake Okeechobee, accessed spectrophotometric analysis: Hydrobiologia, v. 114,
October 18, 2017, at https://earthobservatory.nasa.gov/ p. 117–187.
NaturalHazards/view.php?id=88311.
South Florida Water Management District [SFWMD], 1989,
Orr, P.T., Jones, G.J., and Douglas, G.B., 2004, Response Interim Surface Water Improvement and Management
of cultured Microcystis aeruginosa from the Swan River, (SWIM) Plan for Lake Okeechobee, Part 1: Water Quality,
Australia, to elevated salt concentration and consequences Part VII: Public Information, p. 150.
for bloom and toxin management in estuaries: Marine and
Freshwater Research, v. 55, p. 277–283. Tonk, L., Bosch, K., Visser, P.M., and Huisman, J., 2007,
Salt tolerance of the harmful cyanobacterium Microcystis
Ozturk, S., and Aslim, B., 2010, Modification of aeruginosa: Aquatic Microbial Ecology, v. 46, p. 117–123.
exopolysaccharide composition and production by three
cyanobacterial isolates under salt stress: Environmental Turnipseed, D.P., and Sauer, V.B., 2010, Discharge
Science and Pollution Research International, v. 17, measurements at gaging stations: U.S. Geological Survey
p. 595–602. Techniques and Methods book 3, chap. A8, 87 p.
Pereira, S., Zille, A., Micheletti, E., Moradas-Ferreira, P., U.S. Army Corps of Engineers, 2008, Central and Southern
De Philippis, R., and Tamagnini, P., 2009, Complexity of Florida Project—Water control plan for Lake Okeechobee
cyanobacterial exopolysaccharides: Composition, structures, and Everglades Agricultural Area, accessed October 18,
inducing factors and putative genes involved in their 2017, at http://www.saj.usace.army.mil/Portals/44/docs/
biosynthesis and assembly: FEMS Microbiology Reviews, v. h2omgmt/LORSdocs/2008_LORS_WCP_mar2008.pdf.
33, p. 917–941. Verspagen, J.M., Passarge, J., Jӧhnk, K.D., Visser, P.M.,
Phlips, E.J., Badylak, S., Hart, J., Haunert, D., Lockwood, J., Peperzak, L., Boers, P., Laanbroek, H.J., and Huisman,
O’Donnell, K., Sun, D., Viveros, P., and Yilmaz, M., 2012, J., 2016, Water management strategies against toxic
Climatic influences on autochthonous and allochthonous Microcystis blooms in the Dutch delta: Ecological
phytoplankton blooms in a subtropical estuary, St. Lucie Applications, v. 16, no. 1, p. 313–327.
Estuary, Florida, USA: Estuaries and Coasts, v. 35, Yang, Z., Linlin, G., Wang, W., and Zhang, J., 2012,
p. 335–352. Combined effects of temperature, light intensity, and
Pollman, C.D., and James, R.T., 2011, A simple model of nitrogen concentration on the growth and polysaccharide
internal loading of phosphorus in Lake Okeechobee: Lake content of Microcystis aeruginosa in batch culture:
and Reservoir Management, v. 27, no. 1, p. 15–27. Biochemical Systematics and Ecology, v. 41, p. 130–135.
Rosen, B.H., Davis, T.W., Gobler, C.J., Kramer, B.J., and YSI Incorporated, 2017, EXO user manual, accessed April 4,
Loftin, K.A., 2017, Cyanobacteria of the 2016 Lake 2018, at https://www.ysi.com/File%20Library/Documents/
Okeechobee Waterway harmful algal bloom: U.S. Geological Manuals/EXO-User-Manual-Web.pdf.
Survey Open-File Report 2017–1054, 34 p., accessed April
4, 2018, at https://doi.org/10.3133/ofr20171054.
For more information about this publication, contact
Director, Caribbean-Florida Water Science Center

4446 Pet Lane
Lutz, Florida 33559
For additional information visit https://fl.water.usgs.gov/
Publishing support provided by

Lafayette Publishing Service Center
Rosen and others—Effect of Salinity Tolerance on Cyanobacteria Associated with a Harmful Algal Bloom in Lake Okeechobee, Florida—SIR 2018–5092
ISBN 978-1-4113-4247-7
ISSN 2328-031X (print)

ISSN 2328-0328 (online) 9 7 8 1 4 1 1 3 4 2 4 7 7
Printed on recycled paper
https://doi.org/10.3133/sir20185092
APPENDIX G:
WORLD HEALTH ORGANIZATION STUDY
Toxic Cyanobacteria in Water:
A guide to their public health consequences,

monitoring and management
Edited by Ingrid Chorus and Jamie Bartram
E & FN Spon
An imprint of Routledge
London and New York
First published 1999 by E & FN Spon, an imprint of Routledge

11 New Fetter Lane, London EC4P 4EE
© 1999 WHO
Printed and bound in Great Britain by

St Edmundsbury Press, Bury St Edmunds, Suffolk
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any
form or by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any information storage or retrieval system,
without permission in writing from the publishers.
The publisher makes no representation, express or implied, with regard to the accuracy
of the information contained in this book and cannot accept any legal responsibility or
liability for any errors or omissions that may be made.
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data

A catalog record for this book has been requested.
ISBN 0-419-23930-8
Also available from E & FN Spon
Agricultural Pollution
M. Redman
Determination of Organic Compounds in Natural and Treated Waters

T.R. Crompton
Ecological Effects of Waste Water, 2nd edition
Applied limnology and pollutant effects
E.B. Welch
Integrated Approaches to Water Pollution Problems

Edited by J. Bau, J.D. Henriques, J. de Oliviera Raposo and J.P. Lobo Ferreira
International River Water Quality

Pollution and restoration
Edited by G. Best, E. Niemirycz and T. Bogacka
Microbiology and Chemistry for Environmental Scientists and Engineers

J.N. Lester and J.W. Birkett
The Coliform Index and Waterborne Disease

Problems of microbial drinking water assessment
C. Gleeson and N. Gray
Water and the Environment

Innovative issues in irrigation and drainage
Edited by L.S. Pereira and J. Gowing
Water and Wastewater Treatment, Fourth edition

R. Bardolet
Water: Economics, management and demand

Edited by B. Kay, L.E.D. Smith and T. Franks
Water Policy
Allocation and management in practice
Edited by P. Howsam and R.C. Carter
Water Pollution Control

A guide to the use of water quality management principles
R. Helmer and I. Hespanhol
A Water Quality Assessment of the Former Soviet Union

Edited by V. Kimstach, M. Meybeck and E. Baroudy
Water Quality Assessments, 2nd edition

A guide to the use of biota, sediments and water in environmental monitoring
Edited by D. Chapman
Water Quality Monitoring

A practical guide to the design and implementation of freshwater quality studies and
monitoring programmes
Edited by J. Bartram and R. Ballance
Water Resources
Health, environment and development
Edited by B. Kay
Water Wells: Monitoring, maintenance, rehabilitation

Proceedings of the International Groundwater Engineering Conference, Cranfield
Institute of Technology, UK
Edited by P. Howsan
Ordering information
Toxic Cyanobacteria in Water

A Guide to their Public Health Consequences, Monitoring, and Management
published on behalf of WHO by:
F & FN Spon
11 New Fetter Lane
London EC4) 4EE
Telephone: +44 171 583 9855
Fax: +44 171 843 2298
Order online: http://www.earthprint.com

Table of Contents
Foreword
Acknowledgements
Chapter 1. Introduction
1.1 Water resources

1.2 Eutrophication, cyanobacterial blooms and surface scums
1.3 Toxic cyanobacteria and other water-related health problems
1.4 Present state of knowledge
1.5 Structure and purpose of this book
1.6 References
Chapter 2. Cyanobacteria in the environment
2.1 Nature and diversity

2.2 Factors affecting bloom formation
2.3 Cyanobacterial ecostrategists
2.4 Additional information
2.5 References
Chapter 3. Cyanobacterial toxins
3.1 Classification
3.2 Occurrence of cyanotoxins
3.3 Production and regulation
3.4 Fate in the environment
3.5 Impact on aquatic biota
3.6 References
Chapter 4. Human health aspects
4.1 Human and animal poisonings

4.2 Toxicological studies
4.3 References
Chapter 5. Safe levels and safe practices
5.1 Tolerable exposures

5.2 Safe practices
5.3 Other exposure routes
5.4 Tastes and odours
5.5 References
Chapter 6. Situation assessment, planning and management
6.1 The risk-management framework

6.2 Situation assessment
6.3 Management actions, the Alert Levels Framework
6.4 Planning and response
6.5 References
Chapter 7. Implementation of management plans
7.1 Organisations, agencies and groups

7.2 Policy tools
7.3 Legislation, regulations, and standards
7.4 Awareness raising, communication and public participation
7.5 References
Chapter 8. Preventative measures
8.1 Carrying capacity

8.2 Target values for total phosphorus within water bodies
8.3 Target values for total phosphorus inputs to water bodies
8.4 Sources and reduction of external nutrient inputs
8.5 Internal measures for nutrient and cyanobacterial control
8.6 References
Chapter 9. Remedial measures
9.1 Management of abstraction

9.2 Use of algicides
9.3 Efficiency of drinking water treatment in cyanotoxin removal
9.4 Chemical oxidation and disinfection
9.5 Membrane processes and reverse osmosis
9.6 Microcystins other than microcystin-LR
9.7 Effective drinking water treatment at treatment works
9.8 Drinking water treatment for households and small community supplies
9.9 References
Chapter 10. Design of monitoring programmes
10.1 Approaches to monitoring programme development

10.2 Laboratory capacities and staff training
10.3 Reactive versus programmed monitoring strategies
10.4 Sample site selection
10.5 Monitoring frequency
10.6 References
Chapter 11. Fieldwork: site inspection and sampling
11.1 Planning for fieldwork

11.2 Site inspection
11.3 Sampling
11.4 Nutrients, cyanobacteria and toxins
11.5 On-site analysis
11.6 Field records
11.7 Sample preservation and transport
11.8 References
Chapter 12. Determination of cyanobacteria in the laboratory
12.1 Sample handling and storage

12.2 Cyanobacterial identification
12.3 Quantification
12.4 Determination of biomass using chlorophyll a analysis
12.5 Determination of nutrient concentrations
12.6 References
Chapter 13. Laboratory analysis of cyanotoxins

13.2 Sample preparation for cyanotoxin determination and bioassays
13.3 Toxicity tests and bioassays
13.4 Analytical methods for cyanotoxins
13.5 References
Foreword
Concern about the effects of cyanobacteria on human health has grown in many
countries in recent years for a variety of reasons. These include cases of poisoning
attributed to toxic cyanobacteria and awareness of contamination of water sources
(especially lakes) resulting in increased cyanobacterial growth. Cyanobacteria also
continue to attract attention in part because of well-publicised incidents of animal
poisoning.
Outbreaks of human poisoning attributed to toxic cyanobacteria have been reported in

Australia, following exposure of individuals to contaminated drinking water, and in the
UK, where army recruits were exposed while swimming and canoeing. However, the
only known human fatalities associated with cyanobacteria and their toxins occurred in
Caruaru, Brazil, where exposure through renal dialysis led to the death of over 50
patients. Fortunately, such severe acute effects on human health appear to be rare, but
little is known of the scale and nature of either long-term effects (such as tumour
promotion and liver damage) or milder short-term effects, such as contact irritation.
Water and health, and in particular drinking water and health, has been an area of
concern to the World Health Organization (WHO) for many years. A major activity of
WHO is the development of guidelines which present an authoritative assessment of the
health risks associated with exposure to infectious agents and chemicals through water.
Such guidelines already exist for drinking water and for the safe use of wastewater and
excreta in agriculture and aquaculture, and are currently being prepared for recreational
uses of water. In co-operation with the United Nations Educational, Scientific and
Cultural Organization (UNESCO), United Nations Environment Programme (UNEP) and
the World Meteorological Organization (WMO), WHO is also involved in the long-term
monitoring of water through the GEMS/Water Programme; and in the monitoring of water
supply and sanitation services in co-operation with the United Nations Children's Fund
(UNICEF). The World Health Organization supports the development of national and
international policies concerning water and health, and assists countries in developing
capacities to establish and maintain healthy water environments, including legal
frameworks, institutional structures and human resources.
The first WHO publication dealing specifically with drinking water was published in 1958
as International Standards for Drinking-Water. Further editions were published in 1963
and 1971. The first edition of WHO's Guidelines for Drinking-Water Quality was
published in 1984-1985. It comprised three volumes: Volume 1: Recommendations;
Volume 2: Health criteria and other supporting information; Volume 3: Drinking-water
quality control in small-community supplies. The primary aim of the Guidelines for
Drinking-Water Quality is the protection of public health. The guidelines provide an
assessment of the health risks associated with exposure to micro-organisms and
chemicals in drinking water. Second editions of the three volumes of the guidelines were
published in 1993, 1996 and 1997 respectively and addenda to Volumes 1 and 2 were
published in 1998.
Through ongoing review of the Guidelines for Drinking-water Quality, specific micro-
organisms and chemicals are periodically evaluated and documentation relating to
protection and control of drinking-water quality is prepared. The Working Group on
Protection and Control of Drinking-Water Quality identified cyanobacteria as one of the
most urgent areas in which guidance was required. During the development by WHO of
the Guidelines for Safe Recreational-water Environments, it also became clear that
health concerns related to cyanobacteria should be considered and were an area of
increasing public and professional interest.
This book describes the present state of knowledge regarding the impact of
cyanobacteria on health through the use of water. It considers aspects of risk
management and details the information needed for protecting drinking water sources
and recreational water bodies from the health hazards caused by cyanobacteria and
their toxins. It also outlines the state of knowledge regarding the principal considerations
in the design of programmes and studies for monitoring water resources and supplies
and describes the approaches and procedures used.
The development of this publication was guided by the recommendations of several

expert meetings concerning drinking water (Geneva, December 1995; Bad Elster, June
1996) and recreational water (Bad Elster, June 1996; St Helier, May 1997). An expert
meeting in Bad Elster, April 1997, critically reviewed the literature concerning the toxicity
of cyanotoxins and developed the scope and content of this book. A draft manuscript
was reviewed at an editorial meeting in November 1997, and a further draft was
reviewed by the working group responsible for updating the Guidelines for Drinking-
water Quality in March 1998.
Toxic Cyanobacteria in Water is one of a series of guidebooks concerning water

management issues published by E & FN Spon on behalf of WHO. Other volumes in the
series include:
Water Quality Assessments (D. Chapman, Ed., Second Edition, 1996)

Water Quality Monitoring (J. Bartram and R. Ballance, Eds, 1996)
Water Pollution Control (R. Helmer and I. Hespanhol, Eds, 1997)
It is hoped that this volume will be useful to all those concerned with cyanobacteria and
health, including environmental and public health officers and professionals in the fields
of water supply and management of water resources and recreational water. It should
also be of interest to postgraduates in these fields as well as to those involved in
freshwater ecology and special interest groups.
Acknowledgements
The World Health Organization wishes to express its appreciation to all those whose
efforts made the production of this book possible. Special thanks are due to the editors,
Dr Ingrid Chorus, German Federal Environmental Agency, Berlin, Germany, who co-
ordinated the development of the book and to Dr Jamie Bartram, Division of Operational
Support in Environmental Health, WHO, Geneva, Switzerland (formerly of the WHO
European Centre for Environment and Health, Rome, Italy), who managed the process
of preparing the manuscript.
An editorial advisory group assisted in guiding the development of this book, particularly
through co-ordination and review of specific sections. Special thanks are due to
Professor Wayne Carmichael, USA; Professor Geoffrey Codd, UK; Professor Ian
Falconer, Australia; Dr Gary Jones, Australia; Dr Tine Kuiper-Goodman, Canada; and Dr
Linda Lawton, UK, for their dedication and support.
An international group of experts provided material and, in most cases, several authors
and their collaborators contributed to each chapter. Because numerous contributions
were spread over several chapters it is difficult to identify precisely the contribution made
by each individual author and therefore the principal contributors are listed together
below:
Dr Sandra Azevedo, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil (Box 4.3
and Section 5.3.1)
Dr Jamie Bartram, World Health Organization, Geneva, Switzerland (Chapters 1 and 5-7)
Dr Lee Bowling, Department of Land and Water Conservation, Parramatta, New South
Wales, Australia (Chapter 7)
Dr Michael Burch, Cooperative Research Centre for Water Quality and Treatment,
Salisbury, South Australia, Australia (Chapters 5, 6, 9 and 10, Section 8.5.8)
Professor Wayne Carmichael, Wright State University, Dayton, Ohio, USA (Chapter 1,
Box 4.4 and Section 5.3.3)
Dr Ingrid Chorus, Institute for Water, Soil and Air Hygiene, Federal Environmental
Agency, Berlin Germany (Chapters 1, 5, 8, 10 and 12)
Professor Geoffrey Codd, University of Dundee, Dundee, Scotland (Chapters 5, 7 and

10, Section 8.5.8)
Dr Mary Drikas, Australian Water Quality Centre, Adelaide, South Australia, Australia
(Chapter 9)
Professor Ian Falconer, University of Adelaide, Adelaide, Australia (Chapters 4-7)

Dr Jutta Fastner, Institute for Water, Soil and Air Hygiene, Federal Environmental
Agency, Berlin, Germany (Chapter 11 and Figure 13.5)
Dr Jim Fitzgerald, South Australian Health Commission, Adelaide, South Australia,

Australia (Chapter 4)
Dr Ross Gregory, Water Research Centre, Swindon, Wiltshire, England (Chapter 9)
Dr Ken-Ichi Harada, Meijo University, Nagoya, Japan (Chapter 13)
Dr Steve Hrudey, University of Alberta, Edmonton, Alberta, Canada (Chapter 9)
Dr Gary Jones, Commonwealth Scientific and Industrial Research Organization (Land

and Water), Indooroopilly, Brisbane, Queensland, Australia (Chapters 1, 3, 6 and 7,
Figure 5.1, Table 5.2, Box 8.3)
Dr Fumio Kondo, Aichi Prefectural Institute of Public Health, Nagoya, Japan (Chapter 13)
Dr Tine Kuiper-Goodman, Health Canada, Ottawa, Ontario, Canada (Chapters 4 and 5,

Box 6.1)
Dr Linda Lawton, Robert Gordon University of Aberdeen, Aberdeen, Scotland (Chapters

12 and 13)
Dr Blahoslav Marsalek, Institute of Botany, Brno, Czech Republic (Sections 3.5.1 and
3.5.4, Chapter 12)
Dr Luuc Mur, University of Amsterdam, Amsterdam, Netherlands (Chapters 2 and 8)
Dr Judit Padisák, Institute of Biology, University of Veszprém, Veszprém, Hungary

(Chapter 12)
Dr Kaarina Sivonen, University of Helsinki, Helsinki, Finland (Chapter 3)
Dr Olav Skulberg, Norwegian Institute for Water Research, Oslo, Norway (Chapters 1
and 2, Figures 2.1 and 12.1, Box 7.5)
Dr Hans Utkilen, National Institute for Public Health, Oslo, Norway (Section 5.4, Chapter
11, Figure 13.2)
Dr Jessica Vapnek, Food and Agriculture Organization of the United Nations, Rome,
Italy (Chapter 7)
Dr Yu Shun-Zhang, Institute of Public Health, Shanghai, China (Box 5.2)
Acknowledgements are also due to the following contributors: Dr Rainer Enderlein,

United Nations Economic Commission for Europe (UN ECE), Geneva, Switzerland (Box
7.4); Dr Michelle Giddings, Health Canada, Ottawa, Ontario, Canada (Box 6.1); Dr Nina
Gjølme, National Institute for Public Health, Oslo, Norway (Figures 2.3-2.5); Dr Rita
Heinze, Institute for Water, Soil and Air Hygiene, Federal Environmental Agency, Bad
Elster, Germany (Section 13.3.5); Dr Peter Henriksen, National Environmental Research
Institute, Roskilde, Denmark (Figure 3.4); Dr Elke Pawlitzky, Institute for Water, Soil and
Air Hygiene, Federal Environmental Agency, Berlin, Germany (Section 12.5.1); and Dr
Maria Sheffer, Health Canada, Ottawa, Ontario, Canada (Box 6.1).
The World Health Organization also thanks the following people, who reviewed the text:
Dr Igor Brown, Kiev, Ukraine; Dr Maurizio Cavalieri, Local Agency for Electricity and
Water Supply, Rome, Italy; Dr Gertrud Cronberg, Lund University, Lund, Sweden; John
Fawell, Water Research Centre, Medmenham, Buckinghamshire, England; Dr Gertraud
Hoetzel, La Trobe University, Wodonga, Victoria, Australia; Dr Jaroslava Komárková,
Hydrobiological Institute of the Czech Academy of Sciences, Ceské Budejovice, Czech
Republic; Dr Andrea Kozma-Törökne, National Institute for Public Health, Budapest,
Hungary; Dr Peter Literathy, Water Resources Research Centre (VITUKI), Budapest,
Hungary; Dr Gerry Moy, Programme of Food Safety and Food Aid, WHO, Geneva,
Switzerland; staff of the Norwegian Institute for Water Research, Oslo, Norway; and Dr
Stephen Pedley, University of Surrey, Guildford, Surrey, England.
Thanks are also due to Dr Deborah Chapman, the series editor, for editorial assistance,
layout and production management, and to Ms Grazia Motturi and Ms Sylvaine Bassi, for
secretarial and administrative assistance. We are also grateful to Alan Steel for
preparation of illustrations, to A. Willcocks and L. Willcocks for typesetting assistance
and to Stephanie Dagg for preparation of the index.
Special thanks are due to the Ministries of Environment and Health of Germany and the
Institute for Water, Soil and Air Hygiene of the Federal Environmental Agency, Berlin,
which provided financial support for the book. The meetings at which the various drafts
of the manuscript were reviewed were supported by the Ministry of Health of Italy, the
States of Jersey and the United States Environment Protection Agency.
Toxic Cyanobacteria in Water: A guide to their public health consequences,
© 1999 WHO
ISBN 0-419-23930-8
Chapter 1. INTRODUCTION
This chapter was prepared by Jamie Bartram, Wayne W. Carmichael, Ingrid Chorus,
Gary Jones, and Olav M. Skulberg
"A pet child has many names". This proverb is well illustrated by such expressions as
blue-greens, blue-green algae, myxophyceaens, cyanophyceans, cyanophytes,
cyanobacteria, cyanoprokaryotes, etc. These are among the many names used for the
organisms this book considers. This apparent confusion in use of names highlights the
important position that these organisms occupy in the development of biology as a
science. From their earliest observation and recognition by botanists (Linné, 1755;
Vaucher, 1803; Geitler, 1932), and onwards to their treatment in modem textbooks
(Anagnostidis and Komárek, 1985; Staley et al., 1989), the amazing combination of
properties found in algae and bacteria which these organisms exhibit, have been a
source of fascination and attraction for many scientists.
The cyanobacteria also provide an extraordinarily wide-ranging contribution to human

affairs in everyday life (Tiffany, 1958) and are of economic importance (Mann and Carr,
1992). Both the beneficial and detrimental features of the cyanobacteria are of
considerable significance. They are important primary producers and their general
nutritive value is high. The nitrogen-fixing species contribute globally to soil and water
fertility (Rai, 1990). The use of cyanobacteria in food production and in solar energy
conversion holds promising potential for the future (Skulberg, 1995). However,
cyanobacteria may also be a source of considerable nuisance in many situations.
Abundant growth of cyanobacteria in water reservoirs creates severe practical problems
for water supplies. The development of strains containing toxins is a common
experience in polluted inland water systems all over the world, as well as in some
coastal waters. Thus cyanobacterial toxins, or "cyanotoxins", have become a concern for
human health.
Prior to the first acute cyanotoxin poisoning of domestic animals documented in the
scientific literature (Francis, 1878), reports of cyanobacteria poisonings were largely
anecdotal. Perhaps one of the earliest is from the Han dynasty of China. About 1,000
years ago, General Zhu Ge-Ling, while on a military campaign in southern China,
reported losing troops from poisonings whilst crossing a river. He reported that the river
was green in colour at the time and that his troops drank from the green water (Shun
Zhang Yu, Pers. Comm.). Codd (1996) reported that human awareness of toxic blooms
existed in the twelfth century at the former Monasterium Virdis Stagni (Monastery of the
Green Loch), located near the eutrophic, freshwater Soulseat Loch near Stranraer in
south west Scotland. In more recent times, several investigators have noted that local
people in China, Africa, North and South America and Australia, who use water from
water bodies where green scums are present, will dig holes (soaks) near the water's
edge in order to filter the water through the ground and thus prevent the green material
from contaminating drinking-water supplies. This practice is similar to that of developing
wells next to surface waters in order to use the filtering capacity of the soil to remove
organisms and some chemicals from the surface waters - a technique known as
bankside filtration.
1.1 Water resources

The hydrological cycle represents a complex interconnection of diverse water types with
different characteristics, each subject to different uses. Recent developments have
shown the importance of water resource management in an integrated manner and of
recognising interconnections, especially between human activities and water quality.
Most of the world's available freshwater (i.e. excluding that in polar ice-caps, snow and
glaciers) exists as groundwater. This ready supply of relatively clean and accessible
water has encouraged use of this resource, and in many regions groundwater provides
drinking water of excellent quality. However, in some areas, geological conditions do not
allow the use of groundwater or the supplies are insufficient. Thus, where groundwater
supplies are insufficient or of unsuitable quality, surface water must be used for
purposes such as drinking-water supply. Compared with surface waters, groundwaters
have a high volume and low throughput. Over-abstraction is therefore common.
This book is concerned principally with inland, surface freshwaters, and to a lesser
extent with estuarine and coastal waters where cyanobacteria can grow, and under
suitable conditions, form water blooms or surface scums. Cyanobacteria are a frequent
component of many freshwater and marine ecosystems. Those species that live
dispersed in the water are part of the phytoplankton whilst those that grow on sediments
form part of the phytobenthos. Under certain conditions, especially where waters are rich
in nutrients and exposed to sunlight, cyanobacteria may multiply to high densities - a
condition referred to as a water bloom (see Chapter 2).
The composition of freshwaters is dependent on a number of environmental factors,

including geology, topography, climate and biology. Many of these factors vary over
different time scales such as daily, seasonally, or even over longer timespans. Large
natural variations in water quality may therefore be observed in any given water system.
Eutrophication is the enhancement of the natural process of biological production in

rivers, lakes and reservoirs, caused by increases in levels of nutrients, usually
phosphorus and nitrogen compounds. Eutrophication can result in visible cyanobacterial
or algal blooms, surface scums, floating plant mats and benthic macrophyte
aggregations. The decay of this organic matter may lead to the depletion of dissolved
oxygen in the water, which in turn can cause secondary problems such as fish mortality
from lack of oxygen and liberation of toxic substances or phosphates that were
previously bound to oxidised sediments. Phosphates released from sediments
accelerate eutrophication, thus closing a positive feedback cycle. Some lakes are
naturally eutrophic but in many others the excess nutrient input is of anthropogenic origin,
resulting from municipal wastewater discharges or run-off from fertilisers and manure
spread on agricultural areas. Losses of nutrients due to erosion and run-off from soils
may be low in relation to agricultural input and yet high in relation to the eutrophication
they cause, because concentrations of phosphorus of less than 0.1 mg l-1 are sufficient to
induce a cyanobacterial bloom (see Chapter 8).
Hydrological differences between rivers, impoundments and lakes have important

consequences for nutrient concentrations and thus for cyanobacterial growth. Rivers
generally have a significant flushing rate. The term "self-purification" was adopted to
describe the rapid degradation of organic compounds in rivers where turbulent mixing
effectively replenishes consumed oxygen. This term has been applied, mistakenly, to
any process of removing undesirable substances from water but does not actually
eliminate the contaminants, including processes such as adsorption to sediments or
dilution. Substances bound to sediments may accumulate, be released back into the
water, and may be carried downstream. This process is important for phosphorus. Lakes
generally have long water retention times compared with rivers, and by their nature lakes
tend to accumulate sediments and the chemicals associated with them. Sediments
therefore act as sinks for important nutrients such as phosphorus, but if conditions
change the sediments may also serve as sources, liberating the nutrient back into the
water where it can stimulate the growth of cyanobacteria and algae.
Surface water systems world-wide are now often highly regulated in efforts to control
water availability, whether for direct use in irrigation, hydropower generation or drinking
water supplies or to guard against the consequences of floods and droughts. Many
major rivers (such as the Danube in Europe or the Murray in Australia) may be viewed
as a cascade of impoundments. This trend in regulation of flow has an impact upon the
quality and the quantity of water. It alters sediment transport and, as a result, the
transport of substances attached to sediments, such as plant nutrients which may
enhance cyanobacterial growth. By increasing retention times and surface areas
exposed to sunlight, impoundments change the growth conditions for organisms and
promote opportunities for cyanobacterial growth and water-bloom formation through
modifications to river discharges. For many estuarine and coastal systems, human
impact on hydrological conditions and nutrient concentrations is also now extensive.
Figure 1.1 Schematic representation of the development of surface water pollution
with pathogens, oxygen-consuming organic matter, phosphorus and
cyanobacteria in north-western Europe and in North America
Changes in the nature and scale of human activities have consequences both for the
qualitative and quantitative properties of water resources. Historically, the development
of society has involved a change from rural and agricultural to urban and industrial water
uses, which is reflected in both water demands and water pollution as illustrated in
Figure 1.1. The general trend has been an increase in concentrations of pollutants in
surface waters together with increases in urbanisation. Construction of sewerage first
enhanced this trend by concentrating pollutants from latrines (which can leak into
groundwater or surface waters). After some decades, construction of sewage treatment
systems began extensively in the 1950s. Originally these systems comprised only a
biological step which degraded the organic material which otherwise had led to dramatic
oxygen depletion in the receiving water bodies. Pathogens were also reduced to some
extent, but phosphate remained unaffected. Upgrading treatment systems to remove
phosphorus only began in the 1960s and also had the side-effect of further reducing
pathogens. A resultant decline in eutrophication, and thus of cyanobacterial blooms, is
lagging behind the decline of phosphorus inputs to freshwaters because phytoplankton
growth becomes nutrient-controlled only below threshold concentrations (see Chapter 8).
It is unclear whether the historical shift in water demand from rural to urban will continue
in the future, although a number of influences are apparent. The anticipated food crises
of the early twenty first century will place increasing demands upon irrigated agriculture -
a process that already accounts for about 70 per cent of water demand world-wide. By
contrast, many industries have successfully developed processes with substantial water
economy measures, and their demand upon water resources per unit of activity is now
decreasing in some countries. Domestic water consumption tends to increase with
population and affluence, but development of lower consumption appliances and control
of losses from water mains may stabilise, or even reduce, demand in the future.
Nevertheless, overall trends point to an increasing total demand for water, driven
principally by global population growth.
1.2 Eutrophication, cyanobacterial blooms and surface scums

Eutrophication was recognised as a pollution problem in many western European and
North American lakes and reservoirs in the middle of the twentieth century (Rohde,
1969). Since then, it has become more widespread, especially in some regions; it has
caused deterioration in the aquatic environment and serious problems for water use,
particularly in drinking-water treatment. A recent survey showed that in the Asia Pacific
Region, 54 per cent of lakes are eutrophic; the proportions for Europe, Africa, North
America and South America are 53 per cent, 28 per cent, 48 per cent and 41 per cent
respectively (ILEC/Lake Biwa Research Institute, 1988-1993). Eutrophication also
affects slow flowing rivers, particularly if they have extended low-flow periods during a
dry season. Practical measures for prevention of nutrient loading from wastewater and
from agriculture have been developed. In some regions preventative measures are
being implemented more and more. During the 1990s, increasing introduction of nutrient
removal during sewage treatment in North America and in north western Europe has
begun to show success in reducing phosphorus concentrations; in a few water bodies,
algal and cyanobacterial blooms have actually declined. Technical measures for
reduction of nutrients already present in lakes are also available but have not been
widely applied (see Chapter 8).
Wherever conditions of temperature, light and nutrient status are conducive, surface
waters (both freshwater and marine) may host increased growth of algae or
cyanobacteria. Where such proliferation is dominated by a single (or a few) species, the
phenomenon is referred to as an algal or cyanobacterial bloom. Problems associated
with cyanobacteria are likely to increase in areas experiencing population growth with a
lack of concomitant sewage treatment and in regions with agricultural practices causing
nutrient losses to water bodies through over-fertilisation and erosion.
There are important differences in algal and cyanobacterial growth between tropical and
temperate areas. A characteristic pattern of seasonal succession of algal and
cyanobacterial communities is, for example, diatoms in association with rapidly growing
small flagellates in winter and spring, followed by green algae in late spring and early
summer, and then by species which cannot easily be eaten by zooplankton, such as
dinoflagellates, desmids and large yellow-green algae (in moderately turbulent waters
also diatoms) in late summer and autumn. In eutrophic and hypertrophic waters,
cyanobacteria often dominate the summer phytoplankton. As winter approaches, in most
water bodies, increasing turbulence and the lack of light during the winter leads to their
replacement by diatoms. In the tropics, seasonal differences in environmental factors are
often not great enough to induce the replacement of cyanobacteria by other
phytoplankton species. If cyanobacteria are present or even dominant for most of the
year, the practical problems associated with high cyanobacterial biomass and the
potential health threats from their toxins increase. High cyanobacterial biomass may also
contribute to aesthetic problems, impair recreational use (due to surface scums and
unpleasant odours), and affect the taste of treated drinking water.
Phosphorus is the major nutrient controlling the occurrence of water blooms of
cyanobacteria in many regions of the world, although nitrogen compounds are
sometimes relevant in determining the amount of cyanobacteria present. However, in
contrast to planktonic algae, some cyanobacteria are able to escape nitrogen limitation
by fixing atmospheric nitrogen. The lack of nitrate or ammonia, therefore, favours the
dominance of these species. Thus, the availability of nitrate or ammonia is an important
factor in determining which cyanobacterial species become dominant.
Cyanobacterial blooms are monitored using biomass measurements coupled with the
examination of the species present. A widely-used measure of algal and cyanobacterial
biomass is the chlorophyll a concentration. Peak values of chlorophyll a for an
oligotrophic lake are about 1-10 µg l-1, while in a eutrophic lake they can reach 300 µg l-1.
In cases of hypereutrophy, such as Hartbeespoort Dam in South Africa, maxima of
chlorophyll a can be as high as 3,000 µg l-1 (Zohary and Roberts, 1990).
Trophic state classifications, such as that adopted by the Organisation for Economic Co-
operation and Development (OECD), combine information concerning nutrient status
and algal biomass (OECD, 1982). They provide a basis for the evaluation of status and
trends for management and they facilitate international information exchange and
comparison.
1.3 Toxic cyanobacteria and other water-related health problems

The contamination of water resources and drinking water supplies by human excreta
remains a major human health concern, just as it has been for centuries. By contrast, the
importance of toxic substances, such as metals and synthetic organic compounds, has
only emerged in the latter half of the twentieth century. Although eutrophication has been
recognised as a growing concern since the 1950s, only recently have cyanobacterial
toxins become widely recognised as a human health problem arising as a consequence
of eutrophication. The importance of such toxins, relative to other water-health issues,
can currently only be estimated. A significant proportion of cyanobacteria produce one or
more of a range of potent toxins (see Chapter 3). If water containing high concentrations
of toxic cyanobacteria or their toxins is ingested (in drinking water or accidentally during
recreation), they present a risk to human health (see Chapter 4). Some cyanobacterial
substances may cause skin irritation on contact.
The relationship between water resources and health is complex. The most well
recognised relationship is the transmission of infectious and toxic agents through
consumption of water. Drinking water has therefore played a prominent role in concerns
for water and human health. Diseases arising from the consumption of contaminated
water are generally referred to as "waterborne". Globally, the waterborne diseases of
greatest importance are those caused by bacteria, viruses and parasites, such as
cholera, typhoid, hepatitis A, cryptosporidiosis and giardiasis. Most of the pathogens
involved are derived from human faeces and the resulting diseases are generally
referred to as "faecal-oral" diseases; however they can also be spread by means other
than contaminated water, such as by contaminated food. Waterborne diseases also
include some caused by toxic chemicals, although many of these may only cause health
effects some time after exposure has occurred and may therefore be difficult to
associate directly with the cause.
The second major area of interaction between water and human health concerns its role
in personal and domestic hygiene, through which it contributes to the control of disease.
Because hygiene is a key measure in the control of faecal-oral disease, such diseases
are also "water hygiene" diseases. Other water hygiene diseases include skin and eye
infections and infestations, such as tinea, scabies, pediculosis and trachoma. All of
these diseases occur less frequently when adequate quantities of water are available for
personal and domestic hygiene. It is important to note that the role of water in control of
water hygiene diseases depends on availability and use, and water quality is therefore a
secondary consideration in this context.
"Water contact diseases" are the third group of water-related diseases and occur
through skin contact. The most important example world-wide is schistosomiasis
(bilharzia). In infected persons, eggs of Schistosoma spp. are excreted in faeces or urine.
The schistosomes require a snail intermediate host and go on to infect persons in
contact with water by penetrating intact skin. The disease is of primary importance in
areas where collection of water requires wading or direct contact with contaminated
surface waters such as lakes or rivers. The water contact diseases also include those
diseases arising from non-infectious agents in the water, that may give rise, for example,
to allergies and to skin irritation or to dermatitis.
The fourth principal connection between water and human health concerns "water
habitat vector" diseases. These are diseases transmitted by insect vectors that spend all
or part of their lives in or near water. The best-known examples are malaria (transmitted
by mosquito bites and caused by Plasmodium spp.) and filariasis (transmitted by
mosquito bite and caused by microfilaria).
The classification of water-related disease into four groups (waterborne disease, water
hygiene disease, water contact disease and water habitat vector disease) was originally
developed in order to associate groups of disease more clearly with the measures for
their transmission and control and has contributed greatly to furthering this
understanding. Because of its importance to the global burden of disease, the
classification is based upon infectious disease. Nevertheless, the principal groups of
diseases related to chemicals occurring in water may also be categorised in a similar
way. However, there are a number of water-health associations that fall outside these
categories. These include deficiency-related diseases and recreational uses of water.
For recreational water use, the principal area of concern relating to faecal-oral disease
transmission may be classified reasonably alongside other waterborne disease
transmission. However, concern related to transmission of, for example, eye and ear
infections does not readily fit into the classification system, nor does the increased
transmission of diseases arising from the effect of immersion compromising natural
defence systems (such as those of the eye).
Public health concern regarding cyanobacteria centres on the ability of many species
and strains of these organisms to produce cyanotoxins. Cyanotoxins may fall into two of
the four groups of water-related diseases. They may cause waterborne disease when
ingested, and water contact disease primarily through recreational exposure. In hospitals
and clinics, exposure through intravenous injection has led to human fatalities from
cyanotoxins (see Chapter 4). These toxins pose a challenge for management. Unlike
most toxic chemicals, cyanotoxins only sometimes occur dissolved in the water - they
are usually contained within cyanobacterial cells. In contrast to pathogenic bacteria,
these cells do not proliferate within the human body after uptake, only in the aquatic
environment before uptake.
Cyanotoxins belong to rather diverse groups of chemical substances (see Chapter 3),
each of which shows specific toxic mechanisms in vertebrates (see Chapter 4). Some
cyanotoxins are strong neurotoxins (anatoxin-a, anatoxin-a(s), saxitoxins), others are
primarily toxic to the liver (microcystins, nodularin and cylindrospermopsin), and yet
others (such as the lipopolysaccharides) appear to cause health impairments (such as
gastroenteritis) which are poorly understood. Microcystins are geographically most
widely distributed in freshwaters. Recently, they have even been identified in marine
environments as a cause of liver disease in net-pen reared salmon, although it is not
clear which organism in marine environments contains these toxins. As with many
cyanotoxins, microcystins were named after the first organism found to produce them,
Microcystis aeruginosa, but later studies also showed their occurrence in other
cyanobacterial genera.
The hazard to human health caused by cyanotoxins can be estimated from toxicological
knowledge (see section 4.2) in combination with information on their occurrence (see
section 3.2). However, although the information clearly indicates hazards, there are few
documented cases of human illness unequivocally attributed to cyanotoxins (see section
4.1). In a number of cases, investigation of cyanobacteria and cyanotoxins was carried
out only several days after patients had been exposed and had developed symptoms.
This was because diagnosis moved on to considering cyanobacteria only after other
potential causative agents had proved negative, or even years later when knowledge of
cyanobacterial blooms in a water body was connected with the information on an
outbreak of symptoms of unidentified cause.
The number of quantitative surveys on cyanotoxin occurrence is low, and the level of
cyanotoxin exposure through drinking water or during recreational activities largely
unknown. Surveys on cyanobacteria and cyanotoxins have been primarily ecological and
biogeographical. Early surveys in a number of countries including Australia, Canada,
Finland, Norway, South Africa, Sweden, the UK and the USA involved toxicity testing of
scum samples by mouse bioassay. Surveys during the 1990s have tended to employ
more sensitive and definitive methods for characterisation of the toxins, such as
chromatographic or immunological methods (see Chapter 3). These studies provide an
improving basis for estimating the range of concentrations to be expected in a given
water body and season. However, monitoring cyanotoxin concentration is more difficult
than many other waterborne disease agents, because variations in cyanobacterial
quantities, in time and space, is substantial, particularly if scum-forming species are
dominant (see section 2.2). Wind-driven accumulations and distribution of surface scums
can result in concentrations of the toxin by a factor of 1,000 or more (or even result in
the beaching of scums) and such situations can change within very short time periods,
i.e. the range of hours. Therefore, discontinuous samples only provide a fragmentary
insight into the potential cyanotoxin dose for occasional swimmers and into the amount
entering drinking water intakes.
Very few studies of cyanotoxin removal by drinking water treatment processes have
been published (see Chapter 9), although some water companies have carried out
unpublished studies. Thus, a reliable basis for estimation of cyanotoxin exposure
through drinking water is lacking. In regions using surface waters affected with
cyanobacteria as a source for drinking water, actual toxin exposure will depend strongly
on method of water abstraction and treatment.
In comparing the available indications of hazards from cyanotoxins with other water-
related health hazards, it is conspicuous that cyanotoxins have caused numerous fatal
poisonings of livestock and wildlife, but no human fatalities due to oral uptake have been
documented. Human deaths have only been observed as a consequence of intravenous
exposure through renal dialysis. Cyanotoxins are rarely likely to be ingested by humans
in sufficient amounts for an acute lethal dose. Thus, cyanobacteria are less of a health
hazard than pathogens such as Vibrio cholerae or Salmonella typhi. Nevertheless, dose
estimates indicate that a fatal dose is possible for humans, if scum material is swallowed.
However, swallowing such a repulsive material is likely to be avoided. The combination
of available knowledge on chronic toxicity mechanisms (such as cumulative liver
damage and tumour promotion by microcystins) with that on ambient concentrations
occurring under some environmental conditions, shows that chronic human injury from
some cyanotoxins is likely, particularly if exposure is frequent or prolonged at high
concentrations.
1.4 Present state of knowledge

Research into developing further understanding of the human health significance of
cyanobacteria and individual cyanotoxins, and into practical means for assessing and
controlling exposure to cyanobacteria and to cyanotoxins, is a priority. A major gap also
lies in the synthesis and dissemination of the available information.
Information concerning the efficiency of cyanotoxin removal in drinking water treatment

systems is limited. Especially, simple, low-cost techniques for cyanobacterial cell
removal, such as slow sand filtration, should be investigated and developed further.
More information is also needed on the capability of simple disinfection techniques, such
as chlorine, for oxidising microcystins and cylindrospermopsin (Nicholson et at., 1994). If
this is found to be applicable, or if "conventional" treatments are found to be effective if
properly operated, these approaches would provide a practical tool for removing
cyanotoxins in many situations.
Whilst cyanobacterial blooms remain sporadic or occasional events, most emphasis is

still placed upon the protection of drinking water supplies through the preparation of
contingency plans and their activation when appropriate. Early warning systems and
predictive models can facilitate this and should be based upon available information on
the conditions leading to cyanobacterial bloom development and on occurrence,
localisation and movement of scums.
Epidemiological evidence is of particular value in determining the true nature and

severity of human health effects (and therefore the appropriate response) but is
generally lacking in relation to human exposures to cyanobacteria. The limited studies
undertaken to date in relation to recreational exposure require further substantiation.
Opportunistic studies into exposures through drinking water may provide further valuable
insights. Information from experimental toxicology also needs to be strengthened. In
particular, long-term exposure studies (of at least one year or longer) should be carried
out to assess the chronic toxicity of microcystins and cylindrospermopsins. Uptake
routes (e.g. through nasal tissues and mucous membranes) require further investigation.
Further systematic studies are also required into the suggested tumour-promoting
effects of some cyanotoxins, particularly in the dose range of potential oral uptake with
drinking or bathing water.
Lipopolysaccharide (LPS) endotoxins from cyanobacteria pose a potential health risk for
humans, but knowledge of the occurrence of individual LPS components, their toxicology,
and their removal in drinking water treatment plants, is so poor that guidelines cannot be
set at present. Further bioactive cyanobacterial metabolites are also identified frequently
and the health significance of these requires investigation.
1.5 Structure and purpose of this book

The structure of this book follows a logical progression of issues as outlined in Figure 1.2.
Because of the lack of comprehensive literature in the field of cyanotoxins, this book
aims to give background information as well as practical guidance. Some parts of the
text will mainly be of interest to particular readers. Chapters 2 and 3 provide the
background for understanding the behaviour of cyanobacteria and their toxin production
in given environmental conditions. Chapter 4 reviews the evidence regarding health
impacts, primarily for public health experts establishing national guidelines or academics
identifying and addressing current research needs. Chapters 5-7 provide guidance on
safe practices in the planning and management of drinking water supplies and
recreational resorts. Readers who access the book with specific questions regarding
prevention of cyanobacterial growth or their removal in drinking water treatment will find
Chapters 8 and 9 of direct relevance. Guidance on the design and implementation of
monitoring programmes is given in Chapter 10, and Chapters 11-13 provide field and
laboratory methods for monitoring cyanobacteria, their toxins and the conditions which
lead to their excessive growth. As far as is possible, individual chapters have been
written to be self-contained and self-explanatory. However, substantial cross-referencing,
particularly between Chapters 10 to 13, requires that these chapters should be used
jointly. Where chapters call upon information presented elsewhere in the text, this has
been specifically noted.
Figure 1.2 Aspects of monitoring and managing toxic cyanobacteria in water as
discussed in the various chapters of this book
1.6 References
Anagnostidis, K. and Komárek, J. 1985 Modem approach to the classification system of
cyanophytes. 1 Introduction. Arch. Hydrobiol. Suppl. 71, Algological Studies, 38/39, 291-
302.
Codd, G.A. 1996 Harmful Algae News. IOC of UNESCO, 15, 4, United Nations
Educational, Scientific and Cultural Organization, Paris.
Francis, G. 1878 Poisonous Australian lake. Nature 18,11-12.
Geitler, L. 1932 Cyanophyceae. In: L. Rabenhorst [Ed.] Kryptogamen-Flora. 14. Band.

Akademische Verlagsgesellschaft, Leipzig.
ILEC/Lake Biwa Research Institute [Eds] 1988-1993 Survey of the State of the World's
Lakes. Volumes I-IV. International Lake Environment Committee, Otsu and United
Nations Environment Programme, Nairobi.
Linné, C. 1753 Species Plantarum. Tom II, Stockholm, 561-1200.
Mann, N.H. and Carr, N.G. [Eds] 1992 Photosynthetic Prokaryotes. Biotechnology
Handbooks, Volume 6, Plenum Press, London, 275 pp.
Nicholson, B.C., Rositano, J. and Burch, M.D. 1994 Destruction of cyanobacterial

peptide hepatotoxins by chlorine and chloramine. Wat. Res. 28, 1297-1303.
Rai, A.N. 1990 CRC Handbook of Symbiotic Cyanobacteria. CRC Press, Boca Raton,
253 pp.
Rodhe, W. 1969 Crystallization of eutrophication concepts in North Europe. In:

Eutrophication, Causes, Consequences, Correctives. National Academy of Sciences,
Washington D.C., Standard Book Number 309-01700-9, 50-64.
Skulberg, O.M. 1995 Biophotolysis, hydrogen production and algal culture technology. In:
Y. Yürüm [Ed.] Hydrogen Energy System. Production and Utilization of Hydrogen and
Future Aspects. NATO ASI Series E, Applied Sciences, Vol. 295, Kluwer Academic
Publishers, Dordrecht, 95-110.
Staley, J.T., Bryant, M.P., Pfennig, N. and Holt, J.G. [Eds] 1989 Bergey's Manual of
Systematic Bacteriology. Volume 3, Williams & Wilkins, Baltimore.
Tiffany, L.H. 1958 Algae. The Grass of Many Waters. Charles C. Thomas Publisher,
Springfield, 199 pp.
Vaucher, J.P. 1803 Historie des Conferves déau douce. Geneva.
OECD 1982 Eutrophication of Waters, Monitoring, Assessment and Control.

Organisation for Economic Co-operation and Development, Paris.
Zohary, T. and Roberts, R.D. 1990 Hyperscums and the population dynamics of
Microcystis aeruginosa. J. Plankton Res., 12, 423.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 2. CYANOBACTERIA IN THE ENVIRONMENT
This chapter was prepared by Luuc R. Mur, Olav M. Skulberg and Hans Utkilen
For management of cyanobacterial hazards to human health, a basic understanding of

the properties, the behaviour in natural ecosystems, and the environmental conditions
which support the growth of certain species is helpful. This chapter provides information
on how cyanobacteria are structured and the abilities which they posses that support
their proliferation in aquatic ecosystems.
2.1 Nature and diversity

2.1.1 Systematics
Plants and animals possess consistent features by which they can be identified reliably
and sorted into recognisably distinct groups. Biologists observe and compare what the
organisms look like, how they grow and what they do. The results make it possible to
construct systematic groupings based on multiple correlations of common characters
and that reflect the greatest overall similarity. The basis for such groupings is the fact
that all organisms are related to one another by way of evolutionary descent. Their
biology and phylogenetic relationships makes the establishment of systematic groupings
possible (Minkoff, 1983).
However, microbial systematics has long remained an enigma. Conceptual advances in

microbiology during the twentieth century included the realisation that a discontinuity
exists between those cellular organisms that are prokaryotic (i.e. whose cells have no
nucleus) and those that are eukaryotic (i.e. more complexly structured cells with a
nucleus) within the organisation of their cells. The microalgae investigated by
phycologists under the International Code of Botanical Nomenclature (ICBN) (Greuter et
al., 1994) included organisms of both eukaryotic and prokaryotic cell types. The blue-
green algae (Geitler, 1932) constituted the largest group of the latter category. The
prokaryotic nature of these organisms and their fairly close relationship with eubacteria
made work under provisions of the International Code of Nomenclature of Bacteria
(ICNB) (Sneath, 1992) more appropriate (Rippka et al., 1979; Waterbury, 1992).
The prevailing systematic view is that comparative studies of the genetic constitution of
the cyanobacteria will now contribute significantly to the revision of their taxonomy.
Relevant classification should reflect as closely as possible the phylogenetic
relationships as, for example, encoded in 16S or 23S rRNA sequence data (Woese,
1987). The integration of phenotypic, genotypic and phylogenetic information render
possible a consensus type of taxonomy known as polyphasic taxonomy (Vandamme et
al., 1996).
The names "cyanobacteria" and "blue-green algae" (Cyanophyceae) are valid and
compatible systematic terms. This group of micro-organisms comprises unicellular to
multicellular prokaryotes that possess chlorophyll a and perform oxygenic
photosynthesis associated with photosystems I and II (Castenholz and Waterbury, 1989).
2.1.2 Occurrence in nature
The majority of cyanobacteria are aerobic photoautotrophs. Their life processes require
only water, carbon dioxide, inorganic substances and light. Photosynthesis is their
principal mode of energy metabolism. In the natural environment, however, it is known
that some species are able to survive long periods in complete darkness. Furthermore,
certain cyanobacteria show a distinct ability for heterotrophic nutrition (Fay, 1965).
Cyanobacteria are often the first plants to colonise bare areas of rock and soil.
Adaptations, such as ultraviolet absorbing sheath pigments, increase their fitness in the
relatively exposed land environment. Many species are capable of living in the soil and
other terrestrial habitats, where they are important in the functional processes of
ecosystems and the cycling of nutrient elements (Whitton, 1992).
The prominent habitats of cyanobacteria are limnic and marine environments. They
flourish in water that is salty, brackish or fresh, in cold and hot springs, and in
environments where no other microalgae can exist. Most marine forms (Humm and
Wicks, 1980) grow along the shore as benthic vegetation in the zone between the high
and low tide marks. The cyanobacteria comprise a large component of marine plankton
with global distribution (Wille, 1904; Gallon et al., 1996). A number of freshwater species
are also able to withstand relatively high concentrations of sodium chloride. It appears
that many cyanobacteria isolated from coastal environments tolerate saline
environments (i.e. are halotolerant) rather than require salinity (i.e. are halophilic). As
frequent colonisers of euryhaline (very saline) environments, cyanobacteria are found in
salt works and salt marshes, and are capable of growth at combined salt concentrations
as high as 3-4 molar mass (Reed et al., 1984). Freshwater localities with diverse trophic
states are the prominent habitats for cyanobacteria. Numerous species characteristically
inhabit, and can occasionally dominate, both near-surface epilimnic and deep, euphotic,
hypolimnic waters of lakes (Whitton, 1973). Others colonise surfaces by attaching to
rocks or sediments, sometimes forming mats that may tear loose and float to the surface.
Cyanobacteria have an impressive ability to colonise infertile substrates such as volcanic

ash, desert sand and rocks (Jaag, 1945; Dor and Danin, 1996). They are extraordinary
excavators, boring hollows into limestone and special types of sandstone (Weber et al.,
1996). Another remarkable feature is their ability to survive extremely high and low
temperatures. Cyanobacteria are inhabitants of hot springs (Castenholz, 1973),
mountain streams (Kann, 1988), Arctic and Antarctic lakes (Skulberg, 1996a) and snow
and ice (Kol, 1968; Laamanen, 1996). The cyanobacteria also include species that run
through the entire range of water types, from polysaprobic zones to katharobic waters
(Van Landingham, 1982).
Cyanobacteria also form symbiotic associations with animals and plants. Symbiotic
relations exist with, for example, fungi, bryophytes, pteridophytes, gymnosperms and
angiosperms (Rai, 1990). The hypothesis for the endosymbiotic origin of chloroplasts
and mitochondria should be mentioned in this context. The evolutionary formation of a
photosynthetic eukaryote can be explained by a cyanobacteria being engulfed and co-
developed by a phagotrophic host (Douglas, 1994).
Fossils of what were almost certainly prokaryotes are present in the 3,450 million year
old Warrawoona sedimentary rock of north-western Australia. Cyanobacteria were
among the pioneer organisms of the early earth (Brock 1973; Schopf, 1996). These
photosynthetic micro-organisms were, at that time, probably the chief primary producers
of organic matter, and the first organisms to release elemental oxygen into the primitive
atmosphere. Sequencing of deoxyribonucleic acid (DNA) has given evidence that the
earliest organisms were thermophilic and thus able to survive in oceans that were
heated by volcanoes, hot springs and bolide impacts (Holland, 1997).
2.1.3 Organisation, function and behaviour
The structure and organisation of cyanobacteria are studied using light and electron
microscopes. The basic morphology comprises unicellular, colonial and multicellular
filamentous forms (Figure 2.1).
Unicellular forms, for example in the order Chroococcales, have spherical, ovoid or
cylindrical cells. They occur singly when the daughter cells separate after reproduction
by binary fission. The cells may aggregate in irregular colonies, being held together by
the slimy matrix secreted during the growth of the colony. By means of a more or less
regular series of cell division, combined with sheath secretions, more ordered colonies
may be produced.
Figure 2.1 Basic morphology of cyanobacteria
Unicellular, isopolar (Order: Chroococcales)
Pseudoparenchymatous (Order: Pleurocapsales)

Unicellular, heteropolar (Order: Chamaesiphonales)
Multicellular, trichal, heterocysts not present (Order: Oscillatoriales)

Multicellular, trichal, with branches, heterocysts present (Order: Stigonematales)
Multicellular, trichal, heterocysts present (Order: Nostocales)
A particular mode of reproduction, which may supplement binary fission, distinguishes
cyanobacteria in the order Chamaesiphonales and Pleurocapsales. In the
Chamaesiphonales exospores are budded off from the upper ends of cells. In the
second order, the principal mode of replication is by a series of successive binary
fissions converting a single mother cell into many minute daughter cells (baeocytes or
endospores).
Filamentous morphology is the result of repeated cell divisions occurring in a single

plane at right angles to the main axis of the filament. The multicellular structure
consisting of a chain of cells is called a trichome. The trichome may be straight or coiled.
Cell size and shape show great variability among the filamentous cyanobacteria.
Species in the order Oscillatoriales, with unseriated and unbranched trichomes, are
composed of essentially identical cells. The other orders with a filamentous organisation
(orders Nostocales and Stigonematales) are characterised with trichomes having a
heterogeneous cellular composition. Vegetative cells may be differentiated into
heterocysts (having a thick wall and hyaline protoplast, capable of nitrogen fixation) and
akinetes (large thick-walled cells, containing reserve materials, enabling survival under
unfavourable conditions). In the order Stigonematales, the filaments are often
multiseriated, with genuine branching. Both heterocysts and akinetes are present.
The only means of reproduction in cyanobacteria is asexual. Filamentous forms

reproduce by trichome fragmentation, or by formation of special hormogonia.
Hormogonia are distinct reproductive segments of the trichomes. They exhibit active
gliding motion upon their liberation and gradually develop into new trichomes.
In contrast to eukaryotic microalgae, cyanobacteria do not possess membrane-bound

sub-cellular organelles; they have no discrete membrane-bound nucleus; they possess a
wall structure based upon a peptidoglycan layer; and they contain 70 S rather than 80 S
ribosomes (Fay and Van Baalen, 1987; Bryant, 1994).
The photosynthetic pigments of cyanobacteria are located in thylakoids that lie free in
the cytoplasm near the cell periphery. Cell colours vary from blue-green to violet-red.
The green of chlorophyll a is usually masked by carotenoids (e.g. beta-carotene) and
accessory pigments such as phycocyanin, allophycocyanin and phycoerythrin
(phycobiliproteins). The pigments are embodied in phycobilisomes, which are found in
rows on the outer surface of the thylakoids (Douglas, 1994). All cyanobacteria contain
chlorophyll a and phycocyanine.
The basic features of photosynthesis in cyanobacteria have been well described

(Ormerod, 1992). Cyanobacteria are oxygenic phototrophs possessing two kinds of
reaction centres, PS I and PS II, in their photosynthetic apparatus. With the accessory
pigments mentioned above, they are able to use effectively that region of the light
spectrum between the absorption peaks of chlorophyll a and the carotenoids. The ability
for continuous photo-synthetic growth in the presence of oxygen, together with having
water as their electron donor for CO2 reduction, enables cyanobacteria to colonise a
wide range of ecological niches (Whitton, 1992). Phycobiliprotein synthesis is particularly
susceptible to environmental influences, especially light quality. Chromatic adaptation is
largely attributable to a change in the ratio between phycocyanin and phycoerythrin in
the phycobilisomes. Thus, cyanobacteria are able to produce the accessory pigment
needed to absorb light most efficiently in the habitat in which they are present.
Cyanobacteria have a remarkable ability to store essential nutrients and metabolites

within their cytoplasm. Prominent cytoplasmic inclusions for this purpose can be seen
with the electron microscope (e.g. glycogen granules, lipid globules, cyanophycin
granules, polyphosphate bodies, carboxysomes) (Fay and Van Baalen, 1987). Reserve
products are accumulated under conditions of an excess supply of particular nutrients.
For example, when the synthesis of nitrogenous cell constituents is halted because of an
absence of a usable nitrogen source, the primary products of photosynthesis are
channelled towards the synthesis and accumulation of glycogen and lipids.
Dinitrogen fixation is a fundamental metabolic process of cyanobacteria, giving them the

simplest nutritional requirements of all living organisms. By using the enzyme
nitrogenase, they convert N2 directly into ammonium (NH4) (a form through which
nitrogen enters the food chain) and by using solar energy to drive their metabolic and
biosynthetic machinery, only N2, CO2, water and mineral elements are needed for growth
in the light. Nitrogen-fixing cyanobacteria are widespread among the filamentous,
heterocyst forming genera (e.g. Anabaena, Nostoc) (Stewart, 1973). However, there are
also several well documented examples of dinitrogen fixation among cyanobacteria not
forming heterocysts (e.g. Trichodesmium) (Carpenter et al., 1992). Under predominantly
nitrogen limited conditions, but when other nutrients are available, nitrogen fixing
cyanobacteria may be favoured and gain growth and reproductive success. Mass
developments (often referred to as "blooms") of such species in limnic (e.g. eutrophic
lakes, see Figure 2.2 in the colour plate section) and marine environments (e.g. the
Baltic Sea) are common phenomena world-wide.
Many species of cyanobacteria possess gas vesicles. These are cytoplasmic inclusions
that enable buoyancy regulation and are gas-filled, cylindrical structures. Their function
is to give planktonic species an ecologically important mechanism enabling them to
adjust their vertical position in the water column (Walsby, 1987). To optimise their
position, and thus to find a suitable niche for survival and growth, cyanobacteria use
different environmental stimuli (e.g. photic, gravitational, chemical, thermal) as clues.
Gas vesicles become more abundant when light is reduced and the growth rate slows
down. Increases in the turgor pressure of cells, as a result of the accumulation of
photosynthate, cause a decrease in existing gas vesicles and therefore a reduction in
buoyancy. Cyanobacteria can, by such buoyancy regulation, poise themselves within
vertical gradients of physical and chemical factors (Figures 2.3A and 2.3B). Other
ecologically significant mechanisms of movement shown by some cyanobacteria are
photomovement by slime secretion or surface undulations of cells (Häder, 1987; Paerl,
1988).
Figure 2.3A Vertical distribution of Anabaena sp. in a thermally stratified eutrophic

lake during bloom conditions
The presence of very small cells of cyanobacteria (in the size range 0.2-2 µm) has been
recognised as a potentially significant source of primary production in various freshwater
and marine environments. These cyanobacteria constitute a component of the
picoplankton in pelagic ecosystems. Cells can be recognised and estimates of their
abundance made by using epifluorescence microscopy (e.g. observing the orange
fluorescence due to phycoerythrin). The unicellular genus Synechococcus is one of the
most studied, and geographically most widely distributed, cyanobacteria in the
picoplankton. Toxigenic strains of Synechococcus have been reported (Skulberg et al.,
1993).
Figure 2.3B Vertical distribution of Planktothrix sp. in a thermally stratified meso-

oligotrophic lake during bloom conditions
2.1.4 Biological diversity
Although cyanobacteria probably evolved as a group of organisms about 2,000 million

years before the advent of eukaryotes, they comprise fewer taxa than eukaryotic
microalgae (Bisby, 1995). The concept of species in the cyanobacteria has, however, no
distinct boundaries. The situation is similar for most organisms, except for those that are
sexually reproductive. Depending on the classification system used, the number of
species recognised varies greatly. Based on the International Code of Botanical
Nomenclature the class Cyanophyceae, for example, contains about 150 genera and
2,000 species (Hoek et al., 1995).
Chemotaxonomic studies include the use of markers, such as lipid composition,

polyamines, carotenoids and special biochemical features. The resulting data support
the more traditional examinations of phenotypic and ecological characteristics.
Physiological parameters are conveniently studied using laboratory cultures (Packer and
Glazer, 1988).
The diversity of cyanobacteria can be seen in the multitude of structural and functional
aspects of cell morphology and in variations in metabolic strategies, motility, cell division,
developmental biology, etc. The production of extracellular substances and cyanotoxins
by cyanobacteria illustrates the diverse nature of their interactions with other organisms
(i.e. allelopathy) (Rizvi and Rizvi, 1992).
A molecular approach to the systematics of cyanobacteria may be most fruitful for

inferring phylogenetic relationships. Macromolecules, such as nucleic acids and proteins,
are copies or translations of genetic information. The methods applied involve direct
studies of the relevant macromolecules by sequencing, or indirectly by electrophoresis,
hybridisation, or immunological procedures (Wilmotte, 1994). Nucleic acid technologies,
especially the polymerase chain reaction (PCR), have advanced to the point that it is
feasible to amplify and sequence genes and other conserved regions from a single cell.
To date, 16S rRNA has given the most detailed information on the relationships within
the cyanobacteria (Rudi et al., 1997). However, the molecular results obtained should be
integrated with other characteristics as the base for a polyphasic taxonomy (Vandamme,
et al., 1996). A considerable morphological, as well as a genotypical, polymorphy exists
in the cyanobacteria, although as data from rRNA sequencing indicates they are
correlated to a high degree.
The phylogenetic relationship of cyanobacteria is the rationale behind the meaningful

systematic groupings. However, it is difficult to set up a system of classification that
serves both the everyday need for practical identification, and offers an expression of the
natural relationship between the organisms in question (Mayr, 1981). Meanwhile, it will
be necessary to use the available manuals and reference books to help in these
investigations and with the proper identification of the cyanobacteria. Table 12.1 shows
examples of how cyanobacteria with toxigenic strains are treated for determinative
purposes according to the prevailing classification systems.
Because they are photoautotrophs, cyanobacteria can be grown in simple mineral media.
Vitamin B12 is the only growth factor that is known to be required by some species.
Media must be supplemented with the essential nutrients needed to support cell growth,
including sources of nitrogen, phosphorus, trace elements, etc. Toxigenic strains of
cyanobacteria are deposited in international-type culture collections (Rippka, 1988;
Sugawara et al., 1993). Clonal cultures are distributed for research, taxonomic work and
teaching purposes.
2.1.5 Practical scope
The cyanobacteria have both beneficial and detrimental properties when judged from a
human perspective. Their extensive growth can create considerable nuisance for
management of inland waters (water supply, recreation, fishing, etc.) and they also
release substances into the water which may be unpleasant (Jüttner, 1987) or toxic
(Gorham and Carmichael, 1988). The water quality problems caused by dense
populations of cyanobacteria are intricate, many and various (Skulberg, 1996b) and can
have great health and economic impacts. As a consequence, the negative aspects of
cyanobacteria have gained research attention and public concern.
The properties that make the cyanobacteria generally undesirable are also the
qualifications for possible positive economic use. Blue-greens are the source of many
valuable products (Richmond, 1990) and carry promising physiological processes,
including light-induced hydrogen evolution by biophotolysis (Skulberg, 1994). Extensive
research has taken place in the relevant fields of biotechnology. Cyanobacteria may be
used for food or fodder because some strains have a very high content of proteins,
vitamins and other essential growth factors and vital pigments of interest can also be
produced (Borowitzka and Borowitzka, 1988). Cyanobacteria are also sources for
substances of pharmaceutical interest (such as antibiotics) (Falch et al., 1995). These
examples are only a few of the possible applications of cyanobacteria for economic
development and their exploitation is among the many challenges for biotechnology for
the next millennium. Also in this context, their secondary metabolites and health
relationships will become important.
2.2 Factors affecting bloom formation

Cyanobacteria have a number of special properties which determine their relative
importance in phytoplankton communities. However, the behaviour of different
cyanobacterial taxa in nature is not homogeneous because their ecophysiological
properties differ. An understanding of their response to environmental factors is
fundamental for setting water management targets. Because some cyanobacteria show
similar ecological and ecophysiological characteristics, they can be grouped by their
behaviour in planktonic ecosystems as "ecostrategists" typically inhabiting different
niches of aquatic ecosystems. A number of properties and reactions to environmental
conditions are discussed below in order to describe these ecostrategists and to aid the
understanding of their specific behaviour.
2.2.1 Light intensity
Like algae, cyanobacteria contain chlorophyll a as a major pigment for harvesting light
and conducting photosynthesis. They also contain other pigments such as the
phycobiliproteins which include allophycocyanin (blue), phycocyanin (blue) and
sometimes phycoerythrine (red) (Cohen-Bazir and Bryant, 1982). These pigments
harvest light in the green, yellow and orange part of the spectrum (500-650 nm) which is
hardly used by other phytoplankton species. The phycobiliproteins, together with
chlorophyll a, enable cyanobacteria to harvest light energy efficiently and to live in an
environment with only green light.
Many cyanobacteria are sensitive to prolonged periods of high light intensities. The
growth of Planktothrix (formerly Oscillatoria) agardhii is inhibited when exposed for
extended periods to light intensities above 180 µE m-2 s-1. Long exposures at light
intensities of 320 µE m-2 s-1 are lethal for many species (Van Liere and Mur, 1980).
However, if exposed intermittently to this high light intensity, cyanobacteria grow at their
approximate maximal rate (Loogman, 1982). This light intensity amounts to less than
half of the light intensity at the surface of a lake, which can reach 700-1,000 µE m-2 s-1.
Cyanobacteria which form surface blooms seem to have a higher tolerance for high light
intensities. Paerl et al. (1983) related this to an increase in carotenoid production which
protects the cells from photoinhibition.
Cyanobacteria are further characterised by a favourable energy balance. Their

maintenance constant is low which means that they require little energy to maintain cell
function and structure (Gons, 1977; Van Liere et al., 1979). As a result of this, the
cyanobacteria can maintain a relatively higher growth rate than other phytoplankton
organisms when light intensities are low. The cyanobacteria will therefore have a
competitive advantage in lakes which are turbid due to dense growths of other
phytoplankton. This was demonstrated in an investigation measuring growth of different
species of phytoplankton at various depths in a eutrophic Norwegian lake (Källqvist,
1981). The results showed that the diatoms Asterionella, Diatoma and Synedra grew
faster than the cyanobacterium Planktothrix at 1 m depth, while the growth rate was
about the same for all these organisms at 2 m depth. At the very low light intensities
below 3 m only Planktothrix grew. The ability of cyanobacteria to grow at low light
intensities and to harvest certain specific light qualities, enables them to grow in the
"shadow" of other phytoplankton. Van Liere and Mur (1979) demonstrated competition
between cyanobacteria and other phytoplankton. Whereas the green alga
(Scenedesmus protuberans) grew faster at high light intensities, growth of the
cyanobacterium (Planktothrix agardhii) was faster at low light intensities (Figures 2.4A
and 2.4B). If both organisms were grown in the same continuous culture at low light
intensity, Planktothrix could out-compete Scenedesmus (Figure 2.4A). At high light
intensities, the biomass of the green alga increased rapidly, causing an increase in
turbidity and a decrease in light availability. This increased the growth rate of the
cyanobacterium, which then became dominant after 20 days (Figure 2.4B). Although
cyanobacteria cannot reach the maximum growth rates of green algae, at very low light
intensities their growth rate is higher. Therefore, in waters with high turbidity they have
better chances of out-competing other species. This can explain why cyanobacteria
which can grow under very poor nutritional conditions (see section 2.2.4) often develop
blooms in nutrient-rich eutrophic waters.
Figure 2.4A Competition for light between a cyanobacterium and a green alga.
Growth rates of Planktothrix agardhii Gomont and Scenedesmus protuberans
Fritsch as a function of average light intensities at pH 8.0, 20 °C with continuous
illumination (Redrawn after Van Liere, 1979)
The light conditions in a given water body determine the extent to which the
physiological properties of cyanobacteria will be of advantage in their competition
against other phytoplankton organisms (Mur et al., 1978). The zone in which
photosynthesis can occur is termed the euphotic zone (Zeu). By definition, the euphotic
zone extends from the surface to the depth at which 1 per cent of the surface light
intensity can be detected. It can be estimated by measuring transparency with a Secchi
disk (see Chapter 11) and multiplying the Secchi depth reading by a factor of 2-3. The
euphotic zone may be deeper or more shallow than the mixed, upper zone of a thermally
stratified water body, the depth of which is termed the epilimnion (Zm) (Figure 2.5). Many
species of planktonic algae and cyanobacteria have little, or only weak, means of active
movement and are passively entrained in the water circulation within the epilimnion.
Thus, they can be photosynthetically active only when the circulation maintains them in
the euphotic zone. In eutrophic waters, phytoplankton biomass is frequently very high
and causes substantial turbidity. In such situations, the euphotic zone is often more
shallow than the epilimnion, i.e. the ratio Zeu/Zm is < 1, and phytoplankton spend part of
the daylight period in the dark. Thus, the Zeu/Zm ratio is a reasonable (and easy to
measure) approach to describing the light conditions encountered by the planktonic
organisms.
Figure 2.4B Competition for light between a cyanobacterium and a green alga.
Outcome of competition between Planktothrix agardhii Gomont and Scenedesmus
protuberans Fritsch in continuous cultures at two different light intensities and
dilution rates (Redrawn after Van Liere, 1979)
2.2.2 Gas vesicles
Many planktonic cyanobacteria contain gas vacuoles (Walsby, 1981). These structures
are aggregates of gas-filled vesicles, which are hollow chambers with a hydrophilic outer
surface and a hydrophobic inner surface (Walsby, 1978). A gas vesicle has a density of
about one tenth that of water (Walsby, 1987) and thus gas vesicles can give
cyanobacterial cells a lower density than water.
2.2.3 Growth rate
The growth rate of cyanobacteria is usually much lower than that of many algal species
(Hoogenhout and Amesz, 1965; Reynolds, 1984). At 20 °C and light saturation, most
common planktonic cyanobacteria achieve growth rates of 0.3-1.4 doublings per day,
while diatoms reach 0.8-1.9 doublings per day and growth rates of up to 1.3-2.3
doublings per day have been observed for single-celled green algae (Van Liere and
Walsby, 1982). Slow growth rates require long water retention times to enable a bloom
of cyanobacteria to form. Therefore cyanobacteria do not bloom in water with short
retention times. A. comprehensive overview of mechanisms determining the growth
rates of planktonic algae and cyanobacteria under different field conditions is available in
Reynolds (1997).
Figure 2.5 Vertical extension of the euphotic zone (Zeu) in relation to depth of the
epilimnion (Zm) in situations with different turbidity.
A. Euphotic zone is deeper than epilimnion;
B. Euphotic zone is not as deep as epilimnion. Secchi depth (Zs) is included as

rough measure of euphotic depth (Zeu) (Zs × 2.5 ≅ Zeu)
2.2.4 Phosphorus and nitrogen
Because cyanobacterial blooms often develop in eutrophic lakes, it was originally

assumed that they required high phosphorus and nitrogen concentrations. This
assumption was maintained even though cyanobacterial blooms often occurred when
concentrations of dissolved phosphate were lowest. Experimental data have shown that
the affinity of many cyanobacteria for nitrogen or phosphorus is higher than for many
other photosynthetic organisms. This means that they can out-compete other
phytoplankton organisms under conditions of phosphorus or nitrogen limitation.
In addition to their high nutrient affinity, cyanobacteria have a substantial storage

capacity for phosphorus. They can store enough phosphorus to perform two to four cell
divisions, which corresponds to a 4-32 fold increase in biomass. However, if total
phosphate rather than only dissolved phosphate is considered, high concentrations
indirectly support cyanobacteria because they provide a high carrying capacity for
phytoplankton. High phytoplankton density leads to high turbidity and low light availability,
and cyanobacteria are the group of phytoplankton organisms which can grow best under
these conditions.
A low ratio between nitrogen and phosphorus concentrations may favour the
development of cyanobacterial blooms. A comparison between the optimum N:P ratios
for eukaryotic algae (16-23 molecules N:1 molecule of P) with the optimum rates for
bloom-forming cyanobacteria (10-16 molecules N: 1 molecule P), shows that the ratio is
lower for cyanobacteria (Schreurs, 1992).
2.2.5 Population stability
While many planktonic algae are grazed by copepods, daphnids and protozoa,
cyanobacteria are not grazed to the same extent, and the impact of grazing by some
specialised ciliates and rhizopod protozoans is usually not substantial. Cyanobacteria
are attacked by viruses, bacteria and actino-mycetes, but the importance of these
natural enemies for the breakdown of populations is not well understood. Because they
have few natural enemies, and their capacity for buoyancy regulation prevents
sedimentation, the loss rates of cyanobacterial populations are generally low. Thus, their
slow growth rates are compensated by the high prevalence of populations once they
have been established.
2.2.6 Temperature
Maximum growth rates are attained by most cyanobacteria at temperatures above 25 °C

(Robarts and Zohary, 1987). These optimum temperatures are higher than for green
algae and diatoms. This can explain why in temperate and boreal water bodies most
cyanobacteria bloom during summer.
2.3 Cyanobacterial ecostrategists

The physiological properties of cyanobacteria discussed above vary between different
species. As a consequence, different "ecostrategists" inhabit different types of water
bodies. A preliminary approach to describing these ecostrategists, based on
ecophysiological laboratory work together with field observations (largely from north-
western Europe), is described below. This information may be useful for management,
because it helps to predict which cyanobacteria can be expected to occur under certain
conditions. Further development of this approach will be possible as more data on
occurrence of cyanobacteria under different growth conditions are collected from other
continents.
2.3.1 Scum-forming ecostrategists
During the vegetation period, a number of cyanobacteria develop large aggregates

(colonies) of coccoid cells or filaments which are not homogeneously distributed over the
water column. Important genera showing this development are Microcystis, Anabaena
and Aphanizomenon. At the water surface the rate of photosynthesis of the colonies is
high and the cells store large quantities of carbohydrates. Although the cells contain gas
vesicles, the heavy carbohydrates acts as ballast and induce sinking within the colonies.
According to Stoke's Law the sinking rate is dependent on the difference in density
between the water and the cells, and on the square of the colony size (d2). Large
colonies sink faster than small ones, and single cells hardly show any vertical migration.
By sinking, colonies move out of the euphotic zone into the deeper, dark water layers,
where they use their carbohydrates during respiration and synthesise new gas-vesicles
(Utkilen et al., 1985). They then become buoyant again and return to the euphotic zone.
Buoyancy regulation enables the colonies to position themselves in light conditions
which are optimal for their growth. A prerequisite is that the water body is not too
turbulent. During the night, all colonies may become buoyant and some of the population
may be accumulated on the water surface where they can be blown together by wind,
forming stable scums along downwind shores. Vertical movement by buoyancy
regulation is illustrated in Figure 2.6. The frequency of vertical migration is dependent
upon colony size.
In temperate regions, as temperatures decline in the autumn, photosynthesis becomes

more rapid than respiration, and the carbohydrate "ballast" is not consumed. The
colonies therefore sink to the bottom of the water body where they may survive the
winter, gradually consuming their carbohydrate stores by respiration or fermentation.
Cells which re-ascend from the bottom in the spring are unicellular or formed into very
small colonies. During this period Microcystis spp. is difficult to recognise in plankton
samples, and only becomes more conspicuous when the colonies increase in size
during early summer.
Buoyancy regulation can be a substantial advantage in competition with other

phytoplankton organisms. However, this type of regulation is only possible in water
bodies with a shallow euphotic zone in relation to the depth of vertical mixing (Zeu < Zm).
Therefore, in temperate climates, blooms of Microcystis spp. are found particularly in
water bodies deeper than 3 m, because the euphotic zone is likely to be substantially
more shallow than the mixed depth. However, even in shallow lakes, where they do not
have the competitive advantage of vertical migration, Microcystis spp. may become
dominant and form substantial blooms, as has been reported from Hungary, Australia,
and particularly from subtropical and tropical regions. Reynolds (1997) characterises
Microcystis spp. as notoriously and overwhelmingly dominant in some lakes of the lower
latitudes that exhibit diel stratification.
Figure 2.6 Effect of colony size on vertical movement of Microcystis aeruginosa
by buoyancy regulation (simulation). Colonies with diameters <20 µm scarcely
migrate, colonies with diameters <160 µm accomplish less than one migration per
day, and colonies up to 1,600 µm diameter can migrate down to 10 m depth and
back up to the surface three times per day
Many cyanobacteria cannot survive high light intensities over longer periods. This may
limit their distribution to more turbid, eutrophic ecosystems. However, Microcystis
species are less sensitive to high light intensities because buoyancy regulation enables
them to find light conditions that are optimal for their growth. This means that the
presence of Microcystis cannot be related strictly to the level of eutrophication. The
genus is therefore found in mesotrophic, eutrophic and in hypertrophic waters. However,
the amount of biomass that this species can attain depends on the level of
eutrophication. Most Microcystis blooms are found in lakes with an average summer
chlorophyll a concentration of 20-50 µg l-1 and a Secchi transparency of 1-2 m.
2.3.2 Homogeneously dispersed ecostrategists
This ecotype comprises filamentous species, such as Planktothrix (Oscillatoria) agardhii

and Limnothrix (Oscillatoria) redekei. These species are extremely sensitive to high light
intensities and do not form colonies (Reynolds, 1987). Because the filaments are quite
small, vertical migration by buoyancy regulation is less pronounced than their passive
entrainment by water circulation. Therefore, these species are homogeneously
dispersed throughout the epilimnion.
Figure 2.7 Distribution of phytoplankton in Lake Veluwe, The Netherlands,
showing an almost permanent bloom of cyanobacteria during the years 1978-83.
The dominant bloom of Planktothrix agardhii was only interrupted by periods of
ice cover (Redrawn from RIZA, 1996)
This type of ecostrategist is found in eutrophic and hypertrophic shallow lakes. Many
lakes with blooms of dispersed ecotypes have a depth of not more than 3 m and
chlorophyll concentrations of 50 µg l-1 and, in extreme cases, greater than 200 µg l-1. The
filaments are hardly grazed and do not sediment. Blooms of this type often lead to virtual
monocultures which can prevail year-round for many years (Figure 2.7). Population
dynamics in such lakes can be limited. In temperate regions, the autumn population can
even survive under ice in winter. In such situations, the spring population starts growth
with a relatively high density and thus has an advantage in competition with other
species (Visser, 1990). By causing high turbidity, these cyanobacterial populations
effectively suppress the growth of other phytoplankton species. Thus, the next summer
population establishes itself almost without any seasonal succession between different
species of phytoplankton. This high stability of the population precludes any
redistribution of phosphorus and nitrogen to other components of the ecosystem and this
can cause a resilience effect in lake restoration projects (see Chapter 8).
Figure 2.2A Micrograph of Figure 2.2B Micrograph of Anabaena
Planktothrix (Oscillatoria) agardhii lemmermannii
Figure 2.2C Micrograph of Figure 2.2D Surface bloom of Planktothrix

Microcystis aeruginosa (Oscillatoria)
Figure 2.2E Surface bloom of Figure 2.2F Aerial photograph with infrared
Microcystis colour film of a freshwater bloom of
cyanobacteria
Figure 2.2G Use of a barrier or boom to keep surface scums of algae and
cyanobacteria away from water offtake structures (Photograph courtesy of Peter
Baker, Australian Water Quality Centre)
2.3.3 Stratifying ecostrategists
Representatives of this ecotype develop stable summer populations in the intermediate

zone of thermally stratified lakes and reservoirs known as the metalimnion see (see
Figure 2.3). The organisms contain the red pigment phycoerythrin to absorb the green
light, which is the prevailing wave length at this depth. The most common of these
species is Planktothrix (Oscillatoria) rubescens, but red varieties of other Planktothrix
species can also form metalimnic populations (Aune et al., 1997).
The single filaments of these species hardly show any vertical migration. However, in
late autumn at the end of the growing season, the cells can become buoyant and then
form red surface scums (Walsby et al., 1983). The niche of this type of Planktothrix is
very limited. It needs sufficient light in the metalimnetic zone, but may be inhibited by too
much light. Most metalimnetic blooms are found at light intensities of 1-5 per cent of the
surface irradiance and in a range of Zeu/Zm between 0.7 and 1.2.
2.3.4 Nitrogen fixing ecostrategists
The mass development of species capable of fixation of atmospheric nitrogen (species

of the genera Anabaena, Aphanizomenon, Cylindrospermopsis, Nodularia, and Nostoc)
can often be related to periodic nitrogen limitation. Examples are found in deep, as well
as in shallow, systems. However, while these ecostrategists often dominate in
ecosystems with low levels of inorganic dissolved nitrogen, the reverse does not
necessarily apply. Numerous lakes with clear nitrogen limitation are not dominated by
nitrogen-fixing cyanobacteria. Low light availability may be the reason for this, because
nitrogen fixation requires high amounts of energy. In turbid lakes, insufficient light energy
may be available for effective nitrogen fixation (Zevenboom and Mur, 1980). A number of
nitrogen fixing species can form colonies and possess gas vesicles. This means that
they can regulate buoyancy, like Microcystis, and can form stable scums along
downwind shores.
Restoration measures which simultaneously reduce phosphate and nitrogen loading

(sewage diversion, isolation) may strengthen prevailing nitrogen-limiting conditions and
hence the probability of large populations of nitrogen-fixing cyanobacteria.
2.3.5 Small, colony-forming taxa
Cases of large populations of the small, colony-forming genus Aphanothece have been
reported. Little information is available on buoyancy regulation and scum formation by
the species involved. In several water bodies, Aphanothece dominance has occurred
after a decrease of Planktothrix rubescens populations. Reynolds (1997) reported them
as the only cyanobacteria present in the summer plankton of small, intermittently flushed
lakes in England. The dominance of this group is not strictly related with phosphate or
nitrogen limitation, and there are no obvious relationships that can explain the sudden
dominance of these cyanobacteria. They seem to dominate in an intermediate state
during lake recovery after restoration measures have been taken; their ecology is
unknown.
2.3.6 Benthic cyanobacteria
Besides the planktonic ecostrategists described above, cyanobacteria may grow on the
bottom sediments of water bodies which are sufficiently clear to allow light penetration to
these surfaces. These benthic species may form coherent mats. Especially high rates of
photosynthesis by such mats sometimes leads to trapping of the photosynthetically
produced oxygen as bubbles within the mats; parts of the mats may then become
sufficiently buoyant to tear loose and rise to the surface. For monitoring and
management of toxic cyanobacteria, awareness of these is important because
cyanotoxin problems are usually not expected in clear, oligotrophic waters. However,
toxic benthic cyanobacteria have caused animal deaths in Scotland, where beached
mats along the shore of a clear loch were scavenged by dogs (Gunn et al., 1992), and in
Switzerland where toxic benthic populations of Oscillatoria limosa were ingested by
cattle drinking from pristine mountain lakes (Metz et al., 1997, 1998).
2.4 Additional information

It is beyond the scope of this book to give a detailed account of the taxonomy and
ecology of cyanobacteria. However, in addition to the references cited in the previous
sections of this chapter there are many useful texts that are widely available. Taxonomy
and species identification are covered in some detail by Anagnostidis and Komárek
(1985), Staley et al. (1989), Larsen and Moestrup (1990) and Waterbury (1992). Detailed
accounts of plankton ecology, including cyanobacteria, are available in Sommer (1989)
and Reynolds (1997) and cyanobacterial ecophysiology is described by Mur (1983).
2.5 References
Anagnostidis, K. and Komárek, J. 1985 Modem approach to the classification system of
cyanophytes. 1 Introduction. Arch. Hydrobiol. Suppl. 71, Algological Studies, 38/39, 291-
302.
Aune, T., Ramstad, H., Skulberg, O.M., Underdal, B., Yndestad, M. and Østensvik, Ø.
1997 Cyanotoksiner og edelkreps - toksinproduserende blågrønnalger i Steinsfjorden
sommeren 1997. Norwegian Institute for Water Research, Oslo.
Bisby, F.A. 1995 Characterization of biodiversity. In: V.H. Heywood and R.T. Watson
[Eds] Global Biodiversity Assessment. Cambridge University Press, Cambridge, 21-106.
Borowitzka, M.A. and Borowitzka, L.J. 1988 Micro-algal Biotechnology. Cambridge

University Press, Cambridge, 477 pp.
Brock. T.D. 1973 Evolutionary and ecological aspects of the cyanophytes. In: N.G. Carr
and B.A. Whitton [Eds] The Biology of the Blue-Green Algae. Blackwell Scientific
Publications, Oxford, 487-500.
Bryant, D.A. [Ed.] 1994 The Molecular Biology of Cyanobacteria. Kluwer Academic
Publishers, Dordrecht, 879 pp.
Carpenter, E.J., Capone, D.G. and Reuter, J.G. [Eds] 1992 Marine Pelagic
Cyanobacteria: Trichodesmium and other Diazotrophs. NATO ASI Series C,
Mathematical and Physical Sciences, Vol. 362. Kluwer Academic Publishers, Dordrecht.
Castenholz, R.W. 1973 Ecology of blue-green algae in hot springs. In: N.G. Carr and
B.A. Whitton [Eds] The Biology of Blue-Green Algae. Blackwell Scientific Publications,
Oxford, 379-414.
Castenholz, R.W. and Waterbury, J.B. 1989 In: J.T. Staley, M.P. Bryant, N. Pfennig and
J.G. Holt [Eds] Bergey's Manual of Systematic Bacteriology. Vol. 3, Williams & Wilkins,
Baltimore, 1710-1727.
Cohen-Bazire, G. and Bryant, D.A. 1982 Phycobilisomes: composition and structure. In:
N.G. Carr and B.A. Whitton [Eds] The Biology of Cyanobacteria. Blackwell Scientific
Publications, Oxford.
Dor, I. and Danin, A. 1996 Cyanobacterial desert crusts in the Dead Sea Valley, Israel.
Arch. Hydrobiol. Suppl. 117, Algological Studies, 83, 197-206.
Douglas, S.E. 1994 Chloroplast origins and evolution. In: D.A. Bryant [Ed.] The
Molecular Biology of Cyanobacteria., Kluwer Academic Publishers, Dordrecht. 91-118.
Falch, B.S., König, G.M., Wright, A.D., Sticher, O., Angerhofer, C.K., Pezzuto, J.M. and
Bachmann, H. 1995 Biological activities of cyanobacteria: evaluation of extracts and
pure compounds. Planta Med. 61, 321-328.
Fay, P. 1965 Heterotrophy and nitrogen fixation in Chlorogloea fritschii. J. Gen. Microbiol.
39, 11-20.
Fay, P. and Van Baalen, C. [Eds] 1987 The Cyanobacteria. Elsevier, Amsterdam, 534
pp.
Gallon, J.R., Jones, D.A. and Page, T.S. 1996 Trichodesmium, the paradoxial
diazotroph. Arch. Hydrobiol. Suppl., Algological Studies, 83,215-243.
Geitler, L. 1932 Cyanophyceae. In: L. Rabenhorst [Ed.] Kryptogamen-Flora. 14. Band.

Akademische Verlagsgesellschaft, Leipzig, 1196 pp.
Gorham, P.R. and Carmichael, W.W. 1988 Hazards of freshwater blue-green algae
(Cyanobacteria). In: C.A. Lembi and J.R. Waaland. [Eds] Algae and Human Affairs.
Cambridge University Press, Cambridge, 403-432.
Gons, H.J. 1977 On the light-limited growth of Scenedesmus protuberans Fritsch. Thesis,
University of Amsterdam.
Greuter, W., Barrie, F., Burdet, H.M., Chaloner, W.G., Demoulin, V., Hawksworth, D.L.,
Jørgensen, P.M., Nicholson, D.H., Silva, P.C., Trehane, P. and McNeill, J. [Eds] 1994
International Code of Botanical Nomenclature (Tokyo Code). (Regnum Vegetabile No
131), Koeltz Scientific Books, Königstein.
Gunn, G.J., Rafferty, A.G., Rafferty, G.C., Cockburn, N., Edwards, C., Beattie, K.A. and
Codd, G.A. 1992 Fatal canine neurotoxicosis attributed to blue-green algae
(cyanobacteria). Vet. Rec., 4, 301-302.
Häder, D.P. 1987 Photomovement. In: P. Fay and C. Van Baalen [Eds] The
Cyanobacteria. Elsevier, Amsterdam, 325-345.
Hoek, C. van den, Mann, D.G and Jahns, H.M. [Eds] 1995 Algae. An Introduction to
Phycology. Cambridge University Press, Cambridge, 623 pp.
Hoogenhout, H. and Amesz, J. 1965 Growth rates of photosynthetic microorganisms in

laboratory cultures. Arch. Microbiol., 50, 10-15.
Holland. H.D. 1997 Evidence for life on earth more than 3,850 million years ago. Science,
275, 38-39.
Humm, H.J. and Wicks, S.R. 1980 Introduction and Guide to the Marine Bluegreen
Algae. John Wiley & Sons, New York, 194 pp.
Jaag, O. 1945 Untersuchungen über die Vegetation und Biologie der Algen des nackten
Gesteins in den Alpen, im Jura und im schweizerischen Mittelland. Kryptogamenflora der
Schweiz, Band IX, Heft 3. Kommissionsverlag Buchdruckerei Büchler and Co., Bern.
Jüttner, F. 1987 Volatile organic substances. In: P. Fay and C. Van Baalen [Eds] The
Cyanobacteria. Elsevier, Amsterdam, 453-469.
Källqvist, T. 1981 Hydroecological field experiment 1981. Incubation of Natural

Phytoplankton in Lake Gjersjøen. Norwegian Institute for Water Research, F-80402,
Oslo, 21 pp.
Kann, E. 1988 Zur Autøkologie benthischer Cyanophyten in reinen europäischen Seen

und Fliessgewässern. Arch. Hydrobiol. Suppl. 80, Algological Studies, 50-53, 473-495.
Kol, E. 1968 Kryobiologie. I. Kryovegetation. In: H.J. Elster and W. Ohle [Eds] Die
Binnengewässer, Band XXIV. E. Schweizerbart'sche Verlagsbuchhandlung, Stuttgart,
216 pp.
Laamanen, M. 1996 Cyanoprokaryotes in the Baltic Sea ice and winter plankton. Arch.
Hydrobiol. Suppl. 117, Algological Studies, 83, 423-433.
Larsen, J. and Moestrup, O. 1990 Guide to Toxic and Potentially Toxic Marine Algae.
The Fish Inspection Service, Ministry of Fisheries, Copenhagen. ISBN 87-983238-1-4,
200 pp.
Loogman, J.G. 1982 Influence of photoperiodicity on algal growth kinetics. Thesis,

Mayr, E. 1981 Biological classification: toward a synthesis of opposing methodologies.

Science, 214, 510-516.
Metz, K., Beattie, K.A., Codd, G.A., Hanselmann, K., Hauser, B., Naegeli, H.P. and
Preisig, H.R. 1997 Identification of a microcystin in benthic cyanobacteria linked to cattle
deaths on alpine pastures in Switzerland. Eur. J. Phycol. 32, 111-117.
Metz, K., Hanselmann, K. and Preisig, H.R. 1998 Environmental conditions in high
mountain lakes containing toxic benthic cyanobacteria. Hydrobiologia,. (In Press).
Minkoff, E.C. 1983 Evolutionary Biology. Addison-Wesley Publishing Company, Reading,

627 pp.
Mur, L.R. 1983 Some aspects of the ecophysiology of cyanobacteria. Ann. Microbiol.,
134B, 61-72.
Mur, L.R., Gons, H.J. and Van Liere, L. 1978 Competition of the green alga
Scenedesmus and the blue-green alga Oscillatoria in light limited environments. FEMS
Microbiol. Letters 1, 335-338.
Ormerod, J.G. 1992 Physiology of the photosynthetic prokaryotes. In: N.H. Mann and
N.G. Carr [Eds] Photosynthetic Prokaryotes. Plenum Press, New York, 93-120.
Packer, L. and Glazer, A.N. [Eds] 1988 Cyanobacteria. Methods in Enzymology. Volume
167, Academic Press, New York, 915 pp.
Paerl, H.W. 1988 Growth and reproductive strategies of freshwater blue-green algae. In:
C.D. Sandgren [Ed.] Growth and Reproductive Strategies of Freshwater Phytoplankton.
Cambridge University Press, Cambridge, 261-315.
Paerl, H.W., Tucker, J. and Bland, P.T. 1983 Carotenoid enhancement and its role in
maintaining blue-green (Microcystis aeruginosa) surface blooms. Oceanogr. 28, 847-857.
Rai, A.N. 1990 CRC Handbook of Symbiotic Cyanobacteria. CRC Press, Boca Raton,
253 pp.
Reed, R.H., Chudek, J.A., Foster, R. and Stewart, W.D.P. 1984 Osmotic adjustment in
cyanobacteria. Arch. Microbiol., 138, 333-337.
Reynolds, C.S. 1984 The Ecology of Freshwater Phytoplankton. Cambridge University

Press, Cambridge.
Reynolds, C.S. 1987 Cyanobacterial waterblooms. In: P. Callow [Ed.] Advances in

Botanical Research, 13, Academic Press, London, 17-143.
Reynolds, C.S. 1997 Vegetation Processes in the Pelagic. A Model for Ecosystem
Theory. Ecology Institute, D-21385 Oldendorf/Luhe, ISSN 0932-2205.
Richmond, A. 1990 Large scale microalgal culture and applications. Prog. Phycol. Res.,
7, 269-330.
Rippka, R. 1988 Recognition and identification of cyanobacteria. In: L. Packer and A.N.
Glazer [Eds] Cyanobacteria. Methods in Enzymology, Volume 167, Academic Press,
New York, 28-67.
Rippka, R., Deruelles, J., Waterbury, J.B., Herdman, M. and Stanier, R.Y. 1979 Generic
assignments, strain histories and properties of pure cultures of cyanobacteria. J. Gen.
Microb., 111, 1-61.
RIZA 1986 Bestrijding van de Eutrofiëring van het Veluwe-Drontermeer. Rijkswaterstaat

dienst binnenwateren, Lelystad.
Rizvi, S.J.H. and Rizvi, V. 1992 Allelopathy. Basic and Applied Aspects. Chapman & Hall,
London, 480 pp.
Robarts, R.D. and Zohary, T. 1987 Temperature effects on photosynthetic capacity,

respiration, and growth rates of bloom-forming cyanobacteria. N.Z. J. Mar. Freshwat.
Res., 21, 391-399.
Rudi, K., Skulberg, O.M., Larsen, F. and Jakobsen, K.S. 1997 Strain characterization
and classification of oxyphotobacteria in clone cultures on the basis of 16S rRNA
sequences from the variable regions V6, V7 and V8. Appl. Environ. Microbiol., 63(7),
2593-2599.
Schopf, J.W. 1996 Cyanobacteria. Pioneers of the early Earth. In: A.K.S.K, Prasad, J.A.
Nienow and V.N.R Rao [Eds] Contributions in Phycology. Nova Hedwigia, Beiheft 112, J.
Cramer, Berlin, 13-32.
Schreurs, H. 1992 Cyanobacterial dominance, relation to eutrophication and lake

morphology. Thesis, University of Amsterdam.
Skulberg, O.M. 1994 Oscillatoialean cyanoprokaryotes and their application for algal
culture technology. Arch. Hydrobiol. Suppl. 105, Algological Studies, 75, 265-278.
Skulberg, O.M. 1996a Terrestrial and limnic algae and cyanobacteria In: A. Elvebakk
and P. Prestrud [Eds] A Catalogue of Svalbard Plants, Fungi, Algae and Cyanobacteria.
Part 9, Norsk Polarinstitutt Skrifter 198, 383-395.
Skulberg, O.M. 1996b Toxin produced by cyanophytes in Norwegian inland waters -

health and environment. In: J. Låg [Ed.] Chemical Data as a Basis of Geomedical
Investigations. The Norwegian Academy of Science and Letters, Oslo, 197-216.
Skulberg, O.M., Carmichael, W.W., Codd, G.A. and Skulberg, R. 1993 Taxonomy of
toxic Cyanophyceae (Cyanobacteria). In: I. R. Falconer [Ed.] Algal Toxins in Seafood
and Drinking Water. Academic Press Ltd., London, 145-164.
Sneath, P.H.A. 1992 International Code of Nomenclature of Bacteria, 1990 Revision.

American Society for Microbiology, Washington, D.C.
Sommer, U. 1989 The role of competition for resources in plankton succession. In: U.
Sommer [Ed.] Plankton Ecology: Succession in Plankton Communities. Springer Verlag,
New York, 57-106.
Staley, J.T., Bryant, M.P., Pfennig, N. and Holt, J.G. [Eds] 1989 Bergey's Manual of
Systematic Bacteriology. Vol. 3, Williams & Wilkins, Baltimore.
Stewart, W.D.P. 1973 Nitrogen fixation by photosynthetic microorganisms. Ann. Rev.

Microbiol., 27,283-316.
Sugawara, H., Ma, J., Miyazaki, S., Shimura, J. and Takishima, Y. [Eds] 1993 World
Directory of Collections of Cultures of Microorganisms., ISBN 4-930813-56-5, 1152 pp.
Utkilen, H.C., Oliver, L.R. and Walsby, A.E. 1985 Buoyancy regulation in a red
Oscillatoria unable to collapse gas vacuoles by turgor pressure. Arch. Hydrobiol., 102,
319-329.
Van Liere, L. and Mur, L.R. 1979 Chapter 9. Some experiments on the competition
between a green alga and a cyanobacterium. In: L. Van Liere, Thesis, University of
Amsterdam.
Van Liere, L., Mur, L.R., Gibson, C.E. and Herdman, M. 1979 Growth and physiology of
Oscillatoria agardhii and some related species, a survey. Dev. Hydrobiol., 2, 67-77.
Van Liere, L. and Mur, L.R. 1980 Occurrence of Oscillatoria agardhii and some related
species, a survey. Dev. Hydrobiol., 2, 67-77.
Van Liere, L. and Walsby, A.E. 1982 Interactions of cyanobacteria with light. In: N.G.
Carr and B.A. Whitton [Eds] The Biology of the Cyanobacteria. Blackwell Science
Publications, Oxford, 9-45.
Vandamme, P., Pot, B., Gillis, M., De Vos, P., Kersters, K. and Swings, J. 1996
Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiol. Rev.,
60, 407-438.
Van Landingham, S.L. 1982 Guide to the Identification, Environmental Requirements

and Pollution Tolerance of Freshwater Blue-Green Algae (Cyanophyta). United States
Environmental Protection Agency, Cincinnati, Ohio, 341 pp.
Visser P.M. 1990 De primaire productie van het Markermeer. Microbiology Laboratory,
Walsby, A.E. 1978 The gas vesicles of aquatic prokaryotes. In: R.Y. Stanier, H.J. Rogers
and J.B. Ward [Eds] Relations Between Structure and Functioning in the Prokaryotic Cell.
28th Symposium of the Society for General Microbiology. Cambridge University Press,
338 pp.
Walsby, A.E. 1981 Cyanobacteria: planktonic gas-vaculated forms. In: M. Star, H. Stolp,
A. Balowes and H.G. Schlegel [Eds] The Prokaryotes. Springer Verlag, New York, 224-
235.
Walsby, A.E. 1987 Mechanisms of buoyancy regulation by planktonic cyanobacteria with

gas vesicles. In: P. Fay and C. Van Baalen [Eds] The Cyanobacteria. Elsevier,
Amsterdam, 377-414.
Walsby, A.E., Utkilen, H.C. and Johnsen, I. J. 1983 Buoyancy changes of a red coloured
Oscillatoria agardhii in lake Gjersjøen, Norway. Arch. Microbiol., 97,18-38.
Waterbury, J.B. 1992 The cyanobacteria isolation, purification and identification. In: A.
Balows, H.G., M. Trüper, M. Dworkin, W. Harder and K.H. Schleifer [Eds] The
Prokaryotes. Second edition, Volume II, Springer-Verlag, New York, 2058-2078.
Weber, B., Wessels, D.C.J. and Büdel, B. 1996 Biology and ecology of cryptoendolithic
cyanobacteria of a sandstone outcrop in the Northern Province, South Africa. Arch.
Hydrobiol., Suppl. 117, Algological Studies, 83,565-579.
Whitton, B.A. 1973 Freshwater plankton, In: G.E. Fogg, W.D.P. Stewart, P. Fay and A.E.
Walsby [Eds] The Blue-Green Algae. Academic Press, London, 353-367.
Whitton, B.A. 1992 Diversity, ecology and taxonomy of the cyanobacteria. In: N.H. Mann
and N.G. Carr [Eds] Photosynthetic Prokaryotes. Plenum Press, New York, 1-51.
Wille, N. 1904 Die Schizophyceen der Plankton Expedition. Ergebnisse der Plankton
Expedition der Humboldt-Stiftung. Bd. II. M.f. Verlag von Lipsius & Tischer, Kiel, 88 pp.
Wilmotte, A. 1994 Molecular evolution and taxonomy of the cyanobacteria. In: D.A.
Bryant [Ed.] The Molecular Biology of Cyanobacteria. Kluwer Academic Publishers,
Dordrecht, 1-25.
Woese, C.R. 1987 Bacterial evolution. Microbiol. Rev., 51,221-271.
Zevenboom, W. and Mur, L. 1980 N2 fixing cyanobacteria, why they do not become
dominant in dutch hypertrophic lakes. Dev. Hydrobiol., 2,123-131.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 3. CYANOBACTERIAL TOXINS

This chapter was prepared by Kaarina Sivonen and Gary Jones
The cyanotoxins are a diverse group of natural toxins, both from the chemical and the
toxicological points of view. In spite of their aquatic origin, most of the cyanotoxins that
have been identified to date appear to be more hazardous to terrestrial mammals than to
aquatic biota. Cyanobacteria produce a variety of unusual metabolites, the natural
function of which is unclear, although some, perhaps only coincidentally, elicit effects
upon other biota. Research has primarily focused on compounds that impact upon
humans and livestock, either as toxins or as pharmaceutically useful substances. Further
ranges of non-toxic products are also being found in cyanobacteria and the biochemical
and pharmacological properties of these are totally unknown. An overview of the
currently identified cyanotoxins is given in section 3.1 and their toxicological properties
are discussed in Chapter 4.
Studies on the occurrence, distribution and frequency of toxic cyanobacteria were

conducted in a number of countries during the 1980s using mouse bioassay. Analytical
methods suitable for quantitative toxin determination only became available in the late
1980s, but studies of specific cyanotoxins have been increasing since then. The results
of both approaches indicate that neurotoxins are generally less common, except
perhaps in some countries where they frequently cause lethal animal poisonings. In
contrast, the cyclic peptide toxins (microcystins and nodularins) which primarily cause
liver injury are more widespread and are very likely to occur if certain taxa of
cyanobacteria are present. Section 3.2 presents an overview of the data currently
available on the occurrence of cyanotoxins. It is noteworthy, however, that current
knowledge is clearly biased by the inconsistent distribution of research effort around the
world, with studies from Asia, Africa and South America beginning to appear in the
1990s. Because the ecological role of the toxins is unclear, it is not possible to use a
functional approach to study the factors that enhance toxicity. Section 3.3 looks at the
available data on relationships between environmental factors and toxin content and at
the emerging understanding of genetic regulation of toxin production. Research into
toxin production by cyanobacteria is increasing, and a better understanding of toxin
function may provide a basis for predicting occurrence of toxicity in the future.
For assessing the health risk caused by cyanotoxins, an understanding of their

persistence and degradation in aquatic environments is of crucial importance. Section
3.4 gives an overview of the current understanding of these processes. Because effects
on aquatic biota may be relevant issues for water managers, and because public
concern could raise questions in this field for practitioners, section 3.5 briefly introduces
the limited state of knowledge of cyanotoxin impacts on other aquatic organisms.
3.1 Classification
Mechanisms of cyanobacterial toxicity currently described and understood are very
diverse and range from hepatotoxic, neurotoxic and dermatotoxic effects to general
inhibition of protein synthesis. To assess the specific hazards of cyanobacterial toxins it
is necessary to understand their chemical and physical properties, their occurrence in
waters used by people, the regulation of their production, and their fate in the
environment.
Cyanotoxins fall into three broad groups of chemical structure: cyclic peptides, alkaloids
and lipopolysaccharides (LPS). An overview of the specific toxic substances within these
broad groups that have been identified to date from different genera of cyanobacteria,
together with their primary target organs in humans, is given in Table 3.1.
3.1.1 Hepatotoxic cyclic peptides - microcystins and nodularins
Globally the most frequently found cyanobacterial toxins in blooms from fresh and
brackish waters are the cyclic peptide toxins of the microcystin and nodularin family.
They pose a major challenge for the production of safe drinking water from surface
waters containing cyanobacteria with these toxins. In mouse bioassays, which
traditionally have been used to screen toxicity of field and laboratory samples,
cyanobacterial hepatotoxins (liver toxins) cause death by liver haemorrhage within a few
hours of the acute doses (see Chapter 4). Microcystins have been characterised from
planktonic Anabaena, Microcystis, Oscillatoria (Planktothrix), Nostoc, and Anabaenopsis
species, and from terrestrial Hapalosiphon genera. Nodularin has been characterised
only from Nodularia spumigena.
The cyclic peptides are comparatively large natural products, molecular weight (MW) ≈
800-1,100, although small compared with many other cell oligopeptides and
polypeptides (proteins) (MW > 10,000). They contain either five (nodularins) or seven
(microcystins) amino acids, with the two terminal amino acids of the linear peptide being
condensed (joined) to form a cyclic compound. They are water soluble and, except
perhaps for a few somewhat more hydrophobic microcystins, are unable to penetrate
directly the lipid membranes of animal, plant and bacterial cells. Therefore, to elicit their
toxic effect, uptake into cells occurs through membrane transporters which otherwise
carry essential biochemicals or nutrients. As will be outlined in section 4.2, this restricts
the target organ range in mammals largely to the liver. In aquatic environments, these
toxins usually remain contained within the cyanobacterial cells and are only released in
substantial amounts on cell lysis. Along with their high chemical stability and their water
solubility, this containment has important implications for their environmental persistence
and exposure to humans in surface water bodies (see section 3.4).
able 3.1 General features of the cyanotoxins
Primary target organ

Toxin group1 Cyanobacterial genera2
in mammals
Cyclic peptides
Microcystins Liver Microcystis, Anabaena, Planktothrix
(Oscillatoria), Nostoc, Hapalosiphon,
Anabaenopsis
Nodularin Liver Nodularia
Alkaloids
Anatoxin-a Nerve synapse Anabaena, Planktothrix (Oscillatoria),
Aphanizomenon
Anatoxin-a(S) Nerve synapse Anabaena
Aplysiatoxins Skin Lyngbya, Schizothrix, Planktothrix (Oscillatoria)
3
Cylindrospermopsins Liver Cylindrospermopsis, Aphanizomenon,
Umezakia
Lyngbyatoxin-a Skin, gastro-intestinal Lyngbya
tract
Saxitoxins Nerve axons Anabaena, Aphanizomenon, Lyngbya,
Cylindrospermopsis
Lipopolysaccharides Potential irritant; affects All
(LPS) any exposed tissue
1
Many structural variants may be known for each toxin group - see section 3.1 for details
2
Not produced by all species of the particular genus
3
Whole cells of toxic species elicit widespread tissue damage, including damage to
kidney and lymphoid tissue
The first chemical structures of cyanobacterial cyclic peptide toxins were identified in the
early 1980s and the number of fully characterised toxin variants has greatly increased
during the 1990s. The first such compounds found in freshwater cyanobacteria were
cyclic heptapeptides (that is they contain seven peptide-linked amino acids) with the
general structure of:
cyclo-(D-alanine1-X2-D-MeAsp3-Z4-Adda5-D-glutamate6-Mdha7)
in which X and Z are variable L amino acids, D-MeAsp3 is D-erythro-β-methylaspartic
acid, and Mdha is N-methyldehydroalanine (Figure 3.1 A). The amino acid Adda,
(2S,3S,8S,9S)-3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-dienoic acid, is the
most unusual structure in this group of cyanobacterial cyclic peptide toxins.
These compounds were first isolated from the cyanobacterium Microcystis aeruginosa
and therefore the toxins were named microcystins (Carmichael et al., 1988). Structural
variations have been reported in all seven amino acids, but most frequently with
substitution of L-amino acids at positions 2 and 4, and demethylation of amino acids at
positions 3 and/or 7 (Figure 3.1A). About 60 structural variants of microcystins have
been characterised so far from bloom samples and isolated strains of cyanobacteria
(Table 3.2).
In one species of brackish water cyanobacterium, an identically acting and structurally
very similar, cyclic pentapeptide occurs. It has been named as nodularin after its
producer, Nodularia spumigena. The chemical structure of nodularin is cyclo-(D-MeAsp1-
L-arginine2-Adda3-D-glutamate4-Mdhb5), in which Mdhb is 2-(methylamino)-2-
dehydrobutyric acid (Figure 3.1B). A few naturally occurring variations of nodularins
have been found: two demethylated variants, one with D-Asp1 instead of D-MeAsp1, the
other with DMAdda3 instead of Adda3; and the non-toxic nodularin which has the 6Z-
stereoisomer of Adda3 (Namikoshi et al., 1994). The equivalent 6Z-Adda3 stereoisomer of
microcystins is also non-toxic. In the marine sponge, Theonella swinhoei, a nodularin
analogue called motuporin has been found. It differs from nodularin only by one amino
acid, having hydrophobic L-Val in place of the polar L-Arg in nodularin (de Silva et al.,
1992). The toxin might be cyanobacterial in origin because the sponge is known to
harbour cyanobacterial symbionts.
The mammalian toxicity of microcystins and nodularins is mediated through their strong
binding to key cellular enzymes called protein phosphatases (see Chapter 4). In solution,
microcystins and nodularins adopt a chemical "shape" that is similar, especially in the
Adda-glutamate part of the cyanotoxin molecule (Rudolph-Böhner et al., 1994; Annila et
al., 1996). Recent studies have shown that this region is crucial for interaction with the
protein phosphatase protein molecule, and hence it is crucial for the toxicity of these
cyanotoxins (Barford and Keller, 1994; Goldberg et al., 1995). Microcystins show an
additional characteristic of forming a covalent bond between the Mdha residue and the
protein phosphatase molecule.
Figure 3.1 The structure of cyclic peptide toxins and cylindrospermopsin.
A. General structure of microcystins (MCYST), cyanobacterial heptapeptide

hepatotoxins, showing the most frequently found variations. X and Z are variable
L-amino acids (in MCYST-LR, X = L-Leusine (L) and Z = L-Arginine (R)); R1 and R2
are H (demethylmicrocystins) or CH3; D-MeAsp is D-erythro-β-methylaspartic acid;
Adda is (2S,3S,8S,9S)-3-amino-9-methoxy-2,6,8-trimethyl-10-phenyldeca-4,6-
dienoic acid and Mdha is N-methyldehydroalanine (Dha = dehydroalanine) (see
Table 3.2 for known microcystins); General structure of microcystins cyclo-(D-
Ala1-X2-D-MeAsp3-Z4-Adda5-D-Glu6-Mdha7)
B. Structures of nodularins (Z = L-arginine) and motuporin (Z = L-Valine). Mdhb is

N-methyldehydrobutyrin; Nodularin R1,R2 = CH3; D-Asp1Nodularin R1 = H, R2 = CH3;
DMAdda3 Nodularin R1 = CH3, R2 = H; Structure of nodularins cyclo-(D-MeAsp1-Z2-
Adda3-D-Glu4-Mdhb5)
C. Structure of cylindrospermopsin; Cylindrospermopsin MW 415; C15H21N5O7S
Table 3.2 The microcystins (MCYST) reported in the scientific literature
Molecular Toxicity
Microcystin Organism3 Reference(s)
weight LD502
MCYST-LA 909 50 M. aeruginosas, Botes et al., 1984; Kaya
M. viridiss and Watanabe, 1990
MCYST-LAba 923 NR M. aeruginosas Gathercole and Thiel,
1987
MCYST-LL 951 + M. aeruginosab Craig et al., 1993
b
MCYST-AR 952 250 Microcystis spp. Namikoshi et al., 1992a
MCYST-YA 959 NR M. aeruginosas Botes et al., 1985
3 7 s
[D-Asp ,Dha ]MCYST-LR 966 + M. aeruginosa , Harada et al., 1991b;
Anabaena sp.s Sivonen et al., 1992a
[D-Asp3,Dha7]MCYST- 969 + Anabaena sp.s Namikoshi et al., 1998
EE(OMe)
MCYST-VF 971 NR M. aeruginosas Bateman et al., 1995
3 s
(D-Asp ]MCYST-LR 980 160-300 A. flos-aquae , Krishnamyrthy et al.,
M. aeruginosas, 1989; Cremer and
M. viridisb, O. Henning, 1991; Harada
agardhiis et al., 1990b; 1991a;
Luukkainen et al., 1993
[Dha7]MCYST-LR 980 250 M. aeruginosas, Harada et al., 1991b;
Anabaena sp.s, Sivonen et al., 1992a;
O. agardhiis Luukkainen et al., 1993
[DMAdda5]MCYST-LR 980 90-100 Microcystis Namikoshi et al., 1992a;
spp.b, Nostoc Sivonen et al., 1992b
sp.s
[Dha7]MCYST-EE(OMe) 983 + Anabaena sp.s Namikoshi et al., 1998
3 7 s
[D-Asp ,Dha ]MCYST- 983 + Anabaena sp. Namikoshi et al., 1998
E(OMe)E(OMe)
MCYST-LF 985 + M. aeruginosas Azevedo et al., 1994
MCYST-LR 994 50 M. aeruginosas, Botes et al., 1985;
A. flos-aquaes Rinehart et al., 1988;
M. viridiss Krishnamyrthy et al.,
1989; Watanabe et al.,
1988
[D-Asp3,D-Glu(OCH3)6]MCYST- 994 NR A. flos-aquaes Sivonen et al., 1992d
LR
[(6Z)-Adda5]MCYST-LR 994 >1,200 M. viridisb Harada et al., 1990a,b
7 s
[Dha ]MCYST-E(OMe)E(OMe) 997 + Anabaena sp. Namikoshi et al., 1998
7 s
[L-Ser ]MCYST-LR 998 + Anabaena sp. Namikoshi et al., 1992c
s
MCYST-LY 1,001 90 M. aeruginosa Stoner et al., 1989
7 s
[L-Ser ]MCYST-EE(OMe) 1,001 + Anabaena sp. Namikoshi et al., 1998
3 7 s
[D-Asp ,Ser ]MCYST- 1,001 + Anabaena sp. Namikoshi et al., 1998
E(OMe)E(OMe)
MCYST-HilR 1,008 100 Microcystis spp.b Namikoshi et al., 1995
[D-Asp3,ADMAdda5]MCYST-LR 1,008 160 Nostoc sp.s Sivonen et al., 1990a;
Namikoshi et al., 1990
[D-Glu(OCH3)6]MCYST-LR 1,008 >1,000 A. flos-aguaes, Sivonen et al., 1992d;
Microcystis sp.s Bateman et al., 1995;
Rinehart et al., 1994
[D-Asp3,Dha7]MCYST-RR 1,009 + O. agardhiib, Krishnamyrthy et al.,
Anabaena sp.s, 1989; Sivonen et al.,
M. aeruginosas 1992a; Luukkainen et
al., 1994
[D-Asp3, ADMAdda5, 1,009 + Nostoc sp.s Beattie et al., 1998
Dhb7]MCYST-LR
[L-MeSer7]MCYST-LR 1,012 150 Microcystis spp.b Namikoshi et al., 1992a;
1995
[Dha7]MCYST-FR 1,014 NR Microcystis sp.s Luukkainen et al., 1994
7 s
[L-Ser ]MCYST- 1,015 + Anabaena sp. Namikoshi et al., 1998
E(OMe)E(OMe)
[ADMAdda5]MCYST-LR 1,022 60 Nostoc sp.s Sivonen et al., 1990a;
[D-Asp3,ADMAdda5]MCYST- 1,022 + Nostoc sp.s Sivonen et al., 1992b
LHar
[D-Asp3]MCYST-RR 1,023 250 O. agardhiis, Meriluoto et al., 1989;
M. aeruginosas Luukkainen et al., 1994
[Dha7]MCYST-RR 1,023 180 M. aeruginosas, Kiviranta et al., 1992;
O. agardhiis Luukkainen et al., 1993
MCYST-LW 1,024 NR M. aeruginosas Bateman et al., 1995
b
MCYST-FR 1,028 250 Microcystis spp. Namikoshi et al., 1992a
MCYST-M(O)R 1,028 700-800 Microcystis spp.b Namikoshi et al., 1992a
[Dha7]MCYST-HphR 1,028 + Anabaena sp.s Namikoshi et al., 1992b
3 7 s
[D-Asp , Dha ]MCYST-HtyR 1,030 + Anabaena sp. Namikoshi et al., 1992b
[Dha7]MCYST-YR 1,030 + M. aeruginosas Sivonen et al., 1992c
3 b
[D-Asp ]MCYST-YR 1,030 + Microcystis spp. Namikoshi et al., 1992d
MCYST-YM(O) 1,035 56 M. aeruginosab Botes et al., 1985;
Elleman et al., 1978
[ADMAdda5]MCYST-LHar 1,036 60 Nostoc sp.s Sivonen et al., 1990a;
MCYST-RR 1,037 600 M. aeruginosas, Kusumi et al., 1987;
M. viridiss, Painuly et al., 1988;
Anabaena sp.s Watanabe et al., 1988;
Sivonen et al., 1992a
[(6Z)-Adda5]MCYST-RR 1,037 >1,200 M. viridisb Harada et al., 1990a,b
1 5 s
[D-Ser , ADMAdda ]MCYST-LR 1,038 + Nostoc sp. Sivonen et al., 1992b
5 7 s
[ADMAdda ,MeSer ]MCYST- 1,040 + Nostoc sp. Sivonen et al., 1992b
LR
[L-Ser7]MCYST-RR 1,041 + Anabaena sp.s, Namikoshi et al., 1992c;
M. aeruginosas/b Luukkainen et al., 1994
[D-Asp3,MeSer7]MCYST-RR 1,041 + O. agardhiis Luukkainen et al. 1993
s
MCYST-YR 1,044 70 M. aeruginosa , Botes et al., 1985;
M. viridiss Watanabe et al., 1988
[D-Asp3]MCYST-HtyR 1,044 160-300 A. flos-aquaes Harada et al., 1991a
7 s
[Dha ]MCYST-HtyR 1,044 + Anabaena sp. Namikoshi et al., 1992b
b
MCYST-(H4)YR 1,048 NR Microcystis spp. Namikoshi et al., 1995
[D-Glu- 1,052 >1,000 Microcystis spp.b Namikoshi et al., 1992a
OC2H3(CH3)OH6]MCYST-LR
[D-Asp3,ADMAdda5, 1,052 + Nostoc sp.s Beattie et al., 1998
Dhb7]MCYST-RR
MCYST-HtyR 1,058 80-100 A. flos-aquaes Harada et al., 1991a
7 s
[L-Ser ]MCYST-HtyR 1,062 + Anabaena sp. Namikoshi et al., 1992b
b
MCYST-WR 1,067 150-200 Microcystis spp. Namikoshi et al., 1992a
[D- 1,073 + Nostoc sp.s Beattie et al., 1998
Asp3,ADMAdda5,Dhb7]MCYST-
HtyR
[L-MeLan7]MCYST-LR 1,115 1,000 Microcystis spp.b Namikoshi et al., 1995
Aba Aminoisobutyric acid

ADMAdda O-Acetyl-O-demethylAdda
Dha Dehydroalanine
Dhb Dehydrobutyrine
DMAdda O-DemethylAdda
E(OMe) Glutamic acid methyl ester ∆
(H4)Y 1,2,3,4,-tetrahydrotyrosine
Har Homoarginine
Hil Homoisoleucine
Hph Homophenylalanine
Hty Homotyrosine
MeLan N-Methyllanthionine
M(O) Methionine-S-oxide
MeSer N-Methylserine
(6Z)-Adda Stereoisomer of Adda at the ∆6 double bond
1
Several partial structures of microcystins have been reported in addition to those
shown in this table (see Boland et al., 1993; Craig et al., 1993; Jones et al., 1995;
Sivonen et al., 1995)
2
Toxicity determined i.p. mouse (µg kg-1); the LD50 value is the dose of toxin that kills
50% of exposed animals; a '+' denotes a toxic result in a non-quantitative mouse
bioassay or inhibition of protein phosphatase and 'NR' denotes 'Not reported'
3
An 's' denotes toxins isolated from culture samples and a 'b' denotes toxins isolated
from bloom samples
Most of the structural variants of microcystin and nodularin are highly toxic within a
comparatively narrow range (intra-peritoneal (i.p.) mouse toxicities largely in the range
50-300 µg kg-1 body weight (bw); see Table 3.2 and section 4.2). Only a few non-toxic
variants have been identified. In general, any structural modifications to the Adda-
glutamate region of the toxin molecule, such as a change in isomerisation of the Adda-
diene (6(E) to 6(Z)) or acylation of the glutamate, renders microcystins and nodularins
non-toxic (Harada et al., 1990 a,b; Rinehart et al., 1994). Linear microcystins and
nodularin are more than 100 times less toxic than the equivalent cyclic compounds. The
linear microcystins are thought to be microcystin precursors and/or bacterial breakdown
products (Choi et al., 1993; Rinehart et al., 1994; Bourne et al., 1996).
Microcystins and nodularin have been characterised from axenic cyanobacterial strains
(i.e. strains free of contaminating bacteria) and thus the cyanobacterial origin of these
compounds is clear. At the present time, it is known that microcystins are produced by
bloom forming species of Microcystis, Anabaena, Oscillatoria (Planktothrix), and Nostoc
(see Table 3.2), by a species of Anabaenopsis and by a soil isolate of Haphalosiphon
hibernicus. Nodularins have been found, with the exception of the marine sponge
Theonella already mentioned, only in Nodularia spumigena (see section 3.2 for more
details). Further species may yet be demonstrated to produce microcystin.
3.1.2 Neurotoxic alkaloids - anatoxins and saxitoxins
Mass occurrences of neurotoxic cyanobacteria have been reported from North America,
Europe and Australia, where they have caused animal poisonings. In mouse bioassays
death by respiratory arrest occurs rapidly (within 2-30 minutes) (see Chapters 4 and 13).
Three families of cyanobacterial neurotoxins are known:
• anatoxin-a and homoanatoxin-a, which mimic the effect of acetyl choline,
• anatoxin-a(S), which is an anticholinesterase, and
• saxitoxins, also known as paralytic shellfish poisons (PSPs) in the marine literature,
which block nerve cell sodium channels.
Anatoxin-a has been found in Anabaena, Oscillatoria and Aphanizomenon,
homoanatoxin-a from Oscillatoria, anatoxin-a(S) from Anabaena, and saxitoxins from
Aphanizomenon, Anabaena, Lyngbya and Cylindrospermopsis. Sixteen confirmed
saxitoxins from cyanobacterial samples have been reported, some of which (e.g. the
decarbamoyl-gonyautoxins) may be chemical breakdown products in some species (see
section 3.4.2).
The alkaloid toxins are diverse, both in their chemical structures and in their mammalian
toxicities. Alkaloids, in general, are a broad group of heterocyclic nitrogenous
compounds (i.e. they contain ring structures with at least one carbon-nitrogen bond)
usually of low to moderate molecular weight (< 1,000). They are produced, in particular,
by plants and by some bacteria, and are invariably bioactive and commonly toxic. The
non-sulphated alkaloid toxins of freshwater cyanobacteria (anatoxins and saxitoxin) are
all neurotoxins. The sulphated PSPs, C-toxins and gonyautoxins (sulphated derivatives
of saxitoxin) are also neurotoxins, but the sulphated alkaloid cylindrospermopsin blocks
protein synthesis with a major impact on liver cells. Some marine cyanobacteria also
contain alkaloids (lyngbyatoxins, aplysiatoxins) which are dermatoxins (skin irritants), but
have also been associated with gastro-enteritis and more general symptoms such as
fever (see Chapter 4).
Alkaloids have varying chemical stabilities, often undergoing spontaneous

transformations to by-products which may have higher or lower potencies than the
parent toxin. Some are also susceptible to direct photolytic degradation (see section 3.4).
Anatoxin-a
Anatoxin-a is a low molecular weight alkaloid (MW = 165), a secondary amine, 2-acetyl-
9-azabicyclo(4-2-1)non-2-ene (Figure 3.2) (Devlin et al., 1977). Anatoxin-a is produced
by Anabaena flos-aquae, Anabaena spp. (flos-aquae-lemmermannii group), Anabaena
planktonica, Oscillatoria, Aphanizomenon and Cylindrospermum (see section 3.2 for
details). Homoanatoxin-a (MW = 179) is an anatoxin-a homologue isolated from an
Oscillatoria formosa (Phormidium formosum) strain (Figure 3.2). It has a propionyl group
at C-2 instead of the acetyl group in anatoxin-a (Skulberg et al., 1992). The LD50 (lethal
dose resulting in 50 per cent deaths) of anatoxin-a and homoanatoxin-a are 200 - 250 µg
kg-1 bw (Devlin et al., 1977; Carmichael et al., 1990; Skulberg et al., 1992).
Anatoxin-a(S)
Anatoxin-a(S) is a unique phosphate ester of a cyclic N-hydroxyguanine (MW = 252)

(Figure 3.2) produced by Anabaena flos-aquae strain NRC 525-17 (Matsunaga et al.,
1989). It has more recently been identified in blooms and isolated strains of Anabaena
lemmermannii (Henriksen et al., 1997; Onodera et al., 1997a). The LD50 of anatoxin-a(S)
is 20 µg kg-1 bw (i.p. mouse) (Carmichael et al., 1990). Structural variants of anatoxin-
a(S) have not been detected.
Saxitoxins
Saxitoxins are a group of carbamate alkaloid neurotoxins which are either non-sulphated
(saxitoxins - STX), singly sulphated (gonyautoxins - GTX) or doubly sulphated (C-toxins)
(Figure 3.2 and Table 3.3). In addition, decarbamoyl variants and several new toxins
have been identified in some species.
Figure 3.2 The chemical structures of cyanobacterial neurotoxins, anatoxin-a,

homoanatoxin-a, anatoxin-a(S), and the general structure of saxitoxins. Sixteen different
saxitoxins have been reported from cyanobacteria (for details see Table 3.3). MW =
molecular weight
Anatoxin-a MW 165; C10H15NO
Homoanatoxin-a MW 179; C11H17NO

Anatoxin-a(S) MW 252; C7H17N4O4P
General structure of saxitoxins
Saxitoxins were originally isolated from shellfish where they are concentrated from
marine dinoflagellates (so called "red tide" algae) and have caused deaths in humans
(Anderson, 1994) (see section 4.1). Saxitoxins have been found in the cyanobacteria
Aphanizomenon flos-aquae, Anabaena circinalis, Lyngbya wollei and
Cylindrospermopsis raciborskii. The North American Aphanizomenon flos-aquae strains
NH-1 and NH-5 contain mostly neosaxitoxin and less saxitoxin (plus a few unidentified
neurotoxins). Anabaena circinalis strains (from Australia) contain mostly C1 and C2
toxins, with lesser amounts of gonyautoxins 2 and 3. The freshwater cyanobacterium
Lyngbya wollei produced three known and six new saxitoxin analogues.
Cylindrospermopsis raciborskii in Brazil was found to contain mostly neosaxitoxin and a
smaller amount of saxitoxin.
Table 3.3 Saxitoxins reported from cyanobacterial strains and bloom samples (for the
chemical structure see Figure 3.2)
Name of toxin Variable chemical groups in toxins Cyanobacteria

R1 R2 R3 R4 R5 Aph1 Ana2 Lyn3 Cyl4
STX H H H CONH2 OH + + +
-
GTX2 H H OSO3 CONH2 OH +
-
GTX3 H OSO3 H CONH2 OH +
-
GTX5 H H H CONHSO3 OH +
- -
C1 H H OSO3 CONHSO3 OH +
- -
C2 H OSO3 H CONHSO3 OH +
NEO OH H H CONH2 OH + +
-
GTX1 OH H OSO3 CONH2 OH *
-
GTX4 OH OSO3 H CONH2 OH *
-
GTX6 OH H H CONHSO3 OH *
dcSTX H H H H OH + +
-
dcGTX2 H H OSO3 H OH + +
-
dcGTX3 H OSO3 H H OH + +
3 -
LWTX1 H OSO3 H COCH3 H +
3 -
LWTX2 H OSO3 H COCH3 OH +
3 -
LWTX3 H H OSO3 COCH3 OH +
3
LWTX4 H H H H H +
3
LWTX5 H H H COCH3 OH +
3
LWTX6 H H H COCH3 H +
STX Saxitoxin
GTX Gonyautoxins
C C-toxins
dcSTX Decarbamoylsaxitoxin
LWTX Lyngbya-wollei-toxins
1
Toxins found in Aphanizomenon flos-aquae, New Hampshire, USA (Ikawa et al., 1982;
Mahmood and Carmichael, 1986)
2
Toxins reported in an Anabaena circinalis strain and bloom samples, Australia (Hump-
age et al., 1994; Negri et al., 1995; Negri et al., 1997). dcGTX2 and dcGTX3 are
probably break down products of C1 and C2 in this species (Jones and Negri, 1997). An
asterisk in this column denotes toxins reported by Humpage et al., 1994 for Anabaena
circinalis based on retention time data, but not confirmed by mass spectrometry, and not
found in subsequent studies
3
Toxins detected in Lyngbya wollei, USA (Onodera et al., 1997b)
4
Toxins thus far found in Cylindrospermopsis raciborskii, Brazil (Lagos et al., 1997)
Other neurotoxic cyanobacteria
In marine Trichodesmium blooms from the Virgin Islands, a neurotoxic factor has been
reported which was not anatoxin-a or anatoxin-a(S) but remains to be characterised
(Hawser et al., 1991).
3.1.3 Cytotoxic alkaloids
In tropical and subtropical waters of Australia, the alkaloid hepatotoxin

cylindrospermopsin with a completely different mechanism of toxicity has caused health
problems in drinking water supplies (see section 4.1). It is a cyclic guanidine alkaloid
with a molecular weight of 415 (Figure 3.1C). It is produced by Cylindrospermopsis
raciborskii (Hawkins et al., 1985, 1997), Umezakia natans (Harada et al., 1994) and
Aphanizomenon ovalisporum (Banker et al., 1997). In pure form, cylindrospermopsin
mainly affects the liver, although crude extracts of C. raciborskii injected or given orally
to mice also induce pathological symptoms in the kidneys, spleen, thymus and heart
(see Chapter 4 for more details). Pure cylindrospermopsin has an LD50 in mice (i.p.) of
2.1 mg kg l-1 bw at 24 h and 0.2 mg kg l-1 bw at 5-6 days (Ohtani et al., 1992). Recently,
new structural variants of cylindrospermopsin have been isolated from an Australian
strain of C. raciborskii, with one being identified as demethoxy-cylindrospermopsin
(Chiswell et al., 1999).
Figure 3.3 The chemical structures of debromoaplysiatoxin and lyngbiatoxin-a
Debromoaplysiatoxin
Lyngbyatoxin A
3.1.4 Dermatotoxic alkaloids - aplysiatoxins and lyngbyatoxin
Benthic marine cyanobacteria such as Lyngbya, Oscillatoria and Schizothrix may

produce toxins causing severe dermatitis among swimmers in contact with the filaments
(see section 4.2). The inflammatory activity of Lyngbya is caused by aplysiatoxins and
debromoaplysiatoxin (Figure 3.3) which are potent tumour promoters and protein kinase
C activators (Mynderse et al., 1977; Fujiki et al., 1990). Another strain of Lyngbya
majuscula has caused dermatitis and severe oral and gastrointestinal inflammation. It
was found to contain lyngbyatoxin-a (see Figure 3.3) (Cardellina et al., 1979).
Debromoaplysiatoxin along with other toxic compounds has also been isolated from
other Oscillatoriaceae, such as Schizothrix calcicola and Oscillatoria nigroviridis.
3.1.5 Irritant toxins - lipopolysaccharides
Weise et al. (1970) were the first to isolate LPS from the cyanobacterium Anacystis
nidulans and numerous reports of endotoxins in cyanobacteria have followed.
Lipopolysaccarides are generally found in the outer membrane of the cell wall of Gram
negative bacteria, including cyanobacteria, where they form complexes with proteins and
phospholipids. They are pyrogenic and toxic (Weckesser and Drews, 1979).
Lipopolysaccarides, as the name implies, are condensed products of a sugar, usually a
hexose, and a lipid, normally a hydroxy C14-C18 fatty acid. The many structural variants
of LPS are generally phylogenetically conserved, i.e. particular orders, genera and
occasionally species, have identical or similar fatty acid and sugar components
contained in their cell wall LPS. It is generally the fatty acid component of the LPS
molecule that elicits an irritant of allergenic response in humans and mammals.
Lipopolysaccharides are an integral component of the cell wall of all Gram negative
bacteria, including cyanobacteria, and can elicit irritant and allergenic responses in
human and animal tissues that come in contact with the compounds. There is
considerable diversity of LPS composition among the cyanobacteria, but differences are
largely related to phylogeny. Thus, different genera typically have distinct LPS
compositions that are largely conserved within that genus (Kerr et al., 1995).
Cyanobacterial LPS are considerably less potent than LPS from pathogenic gram-
negative bacteria such as, for example, Salmonella (see Chapter 4). The chemical
stability of cyanobacterial LPS in surface waters is unknown.
Structurally, LPS is a complex polymer composed of four regions. Region I, the O-

antigen region, consists of repeating oligosaccharide units that may vary in structure,
with numerous combinations of different sugar residues and associated glycosidic
linkages. As suggested by its name, the O-antigen also exhibits several antigenic
determinants that constitute the receptor sites for a number of lysogenic bacteriophages.
Regions II and III are the outer core and backbone of a core polysaccharide. There is
generally only minor variation in core structure between species. The backbone of the
polysaccharide is connected to a glycolipid, lipid A (Region IV), via a short link normally
composed of 3-deoxy-D-mannoocmiosonic acid (KDO). Lipid A is a disaccharide of
glucosamines highly substituted with phosphate, fatty acids and KDO, although the
proportion of KDO is low or absent in cyanobacteria compared with other bacterial LPS.
The lipid A component is also acetylated with amide and ester-linked hydroxy fatty acids.
Recent studies of the fatty acid composition of Australian species of cyanobacteria (Kerr
et al., 1995) show a range of β-OH fatty acids ranging in size from C10 to C22. Normal,
saturated and branched chain acids have been detected. There was a stark
predominance of straight chain 14:0 and 18:0 β-OH acids in Microcystis strains that was
quite distinct from Anabaena and Nodularia strains where 16:0 β-OH predominated the
LPS fatty acid fraction.
Although comparatively poorly studied, cell wall components, particularly LPS

endotoxins from cyanobacteria may contribute to human health problems associated
with exposure to mass occurrences of cyanobacteria. The few results available indicate
that cyanobacterial LPS is less toxic than the LPS of other bacteria, such as Salmonella
(Keleti and Sykora, 1982; Raziuddin et al., 1983) (see also section 4.2). More studies
are needed to evaluate the chemical structures and health risks of cyanobacterial LPS.
3.1.6 Other bioactive compounds
Cyanobacteria are known to produce several other bioactive compounds, some of which
are of medical interest, as well as compounds toxic to other cyanobacteria, bacteria,
algae and zooplankton (see section 3.5). Severe intoxication of fish embryos by crude
extracts of Planktothrix agardhii has been reported by Oberemm et al. (1997). Skulberg
et al. (1994) reported the presence of an unidentified "protracted toxic effect" in
cyanobacterial samples that caused death within 4-24 hours in mice. Whether this effect
was due to a specific cyanotoxin is unclear.
Cyanobacteria have been found to be a rich source of biomedically interesting

compounds and therefore screening programmes for new bioactivities are underway.
Cyanobacteria are known to produce antitumour, antiviral, antibiotic and antifungal
compounds. Of the cyanobacterial extracts screened by a Hawaiian research group, 0.8
per cent showed solid tumour selective cytotoxicity (Moore et al., 1996). Depsipeptides
(peptides with an ester linkage) called cryptophycins isolated from a cyanobacterium,
Nostoc sp. strain GSV 224, are promising candidates for an anticancer drug (Trimurtulu
et al., 1995). Recently, several new cyclic or linear peptides and depsipeptides from
cyanobacteria have been characterised. Some are protease inhibitors, but the biological
activity of the others remains to be characterised (Namikoshi and Rinehart, 1996). Many
of the cyanobacterial bioactive compounds possess structural similarities to natural
products from marine invertebrates.
3.2 Occurrence of cyanotoxins

3.2.1 Mass occurrences of toxic cyanobacteria
The toxicity of cyanobacterial mass occurrences (blooms) was originally brought to the
attention of scientists through reports of animal poisonings by farmers and veterinarians,
with the first well documented case being reported from Australia in 1878 (Francis, 1878).
In most, if not all, reported cases since that time, afflicted animals consumed water from
water bodies where there was an obvious presence of a cyanobacterial scum on the
water surface (see Ressom et al. (1994) and Yoo et al. (1995) for a list of reported
animal poisonings). More recent measurements of cyanobacterial toxins using sensitive
modem analytical methods have often revealed high frequencies of toxic blooms even
when animal poisonings have not been reported (Table 3.4).
Throughout the world, it appears that liver-toxic (hepatotoxic, microcystin-containing)

freshwater blooms of cyanobacteria are more commonly found than neurotoxic blooms.
Liver-toxic blooms have been reported from all continents and almost every part of the
world where samples have been collected for analysis. Nevertheless, mass occurrences
of neurotoxic cyanobacteria are common in some countries and these have been
reported from North America, Europe and Australia, where they have caused several
animal poisonings. Blooms which have caused both liver and kidney damage due to the
toxin cylindrospermopsin (and possibly related cyanotoxins) have been reported in
Australia, Japan, Israel and Hungary.
In recent years, surveys have been carried out in a number of countries in South
America, Africa, Australasia, Asia and Europe. The conclusion that can be drawn from
these surveys is that toxic cyanobacteria are internationally ubiquitous, and that as
further surveys are carried out more toxic cyanobacterial blooms and new toxic species
will be discovered. This is particularly true of tropical and subtropical regions that are
currently under-represented in the literature. It seems likely that every country in the
world will have water bodies which support blooms of toxic cyanobacteria at some time
or another. It is also important to note that mass occurrences of toxic cyanobacteria are
not always associated with human activities causing pollution or "cultural eutrophication"
(see Chapter 2). For example, massive blooms of toxic cyanobacteria have been
reported in Australian reservoirs with pristine or near-pristine catchments (watersheds),
and toxic benthic cyanobacteria have killed cattle drinking from oligotrophic, high-alpine
waters in Switzerland.
3.2.2 Species composition and variation among toxic blooms
Cyanobacterial populations may be dominated by a single species or be composed of a

variety of species, some of which may not be toxic. Even within a single-species bloom
there may be a mixture of toxic and non-toxic strains. A strain is a specific genetic
subgroup within a particular species, and each species may encompass tens or
hundreds of strains, each with slightly different traits. Some strains are much more toxic
than others, sometimes by more than three orders of magnitude. This can mean that one
highly toxic strain, even when occurring in minor amounts amongst larger numbers of
non-toxic strains, may render a bloom sample toxic (Sivonen et al., 1989a,b; Bolch et al.,
1997; Vezie et al., 1998).
Table 3.4 Frequencies of mass occurrences of toxic cyanobacteria in freshwaters
No of samples % of toxic Type of

Country Reference
tested samples toxicity
Australia 231 42 Hepatotoxic Baker and Humpage,
Neurotoxic 1994
1
Australia 31 84 Neurotoxic Negri et al., 1997
Brazil 16 75 Hepatotoxic Costa and Azevedo,
1994
Canada, Alberta 24 66 Hepatotoxic Gorham, 1962
Neurotoxic
Canada, Alberta 39 95 Hepatotoxic Kotak et al., 1993
1
Canada, Alberta (3 226 74 Hepatotoxic Kotak et al., 1995
lakes)
Canada, 50 10 Hepatotoxic Hammer, 1968
Saskatchewan Neurotoxic
China 26 73 Hepatotoxic Carmichael et al.,
1988b
Czech and Slovak 63 82 Hepatotoxic Maršálek et al., 1996
Rep.
Denmark 296 82 Hepatotoxic Henriksen et al.,
SDF 1996b
Neurotoxic
Former German 10 70 Hepatotoxic Henning and Kohl,
1981
Dem. Rep. SDF
1
Germany 533 72 Hepatotoxic Fastner, 1998
Germany 393 22 Neurotoxic Bumke-Vogt, 1998
Greece 18 ? Hepatotoxic Lanaras et al., 1989
Finland 215 44 Hepatotoxic Sivonen, 1990a
Neurotoxic
1
France, Brittany 22 73 Hepatotoxic Vezie et al., 1997
Hungary 50 66 Hepatotoxic Törökné, 1991
Japan 23 39 Hepatotoxic Watanabe and Oishi,
1980
Netherlands 10 90 Hepatotoxic Leeuwangh et al.,
1983
Norway 64 92 Hepatotoxic Skulberg et al., 1994
Neurotoxic
SDF
Portugal 30 60 Hepatotoxic Vasconcelos, 1994
Scandinavia 81 60 Hepatotoxic Berg et al., 1986
Sweden 331 47 Hepatotoxic Willén and Mattsson,
Neurotoxic 1997
UK 50 48 Hepatotoxic Codd and Bell, 1996

1
28
USA, Minnesota 92 53 Unspecified Olson, 1960
Neurotoxic
USA, Wisconsin 102 25 Hepatotoxic Repavich et al., 1990
Neurotoxic
Mean 59
1
HPLC was used to detect the toxin content of the samples
Some of the studies shown in the table have been conducted over several years while
others lasted only one season. The relative share of cyanobacteria in the samples varied;
low frequency of cyanobacteria led to low percentages of toxic samples in some studies.
In most of the studies the method used to detect toxicity is mouse bioassay, normally
with a 4-hour time limit (or longer when slow death factors (SDF) have been included).
SDF may indicate low hepatotoxicity of samples or other unknown toxicity.
Toxic and non-toxic strains from the same cyanobacterial species cannot be separated
by microscopic identification. The use of molecular genetic methods, in particular the use
of molecular probes and primers that target specific toxin production genes, will lead to
the development of more precise identification methods for toxic cyanobacteria in the
future. To confirm that a particular cyanobacterial strain is a toxin-producer, it is
important to isolate a pure culture of that strain, preferably free of other bacteria; then to
detect and quantify toxin concentrations in the pure culture (either by bioassay or
chemical analysis); and, where possible, to purify and characterise fully the toxins (for
such examples see Tables 3.2 and 3.3). It is likely that the list of confirmed toxic species
will increase in the future due to the isolation of new species and strains, and because of
the use of improved isolation, culturing and analytical methods.
Microcystis sp., commonly Microcystis aeruginosa, are linked most frequently to

hepatotoxic blooms world-wide (see Tables 3.2 and 3.5 for details and references for all
toxic species). Microcystis viridis and Microcystis botrys strains also have been shown to
produce microcystins. As noted in section 2.2, Microcystis is a non-nitrogen-fixing genus
which is often dominant under nutrient-rich conditions (especially where there is a
significant supply of ammonia), although it also forms blooms in less polluted waters.
Microcystin-producing Anabaena sp. have been reported from Canada, Denmark,
Finland, France and Norway. A recent study from Egypt revealed that 25 per cent of 75
Anabaena and Nostoc strains isolated from soil, rice fields and water bodies contained
microcystins. Planktothrix agardhii and Planktothrix rubescens (previously called
Oscillatoria agardhii and O. rubescens) are common microcystin producers in the
Northern Hemisphere; toxic strains of these have been isolated from blooms in Denmark,
Finland and Norway. In addition, these species were frequently shown to be dominant in
microcystin containing blooms in China, in Germany and in Sweden. In Swiss alpine
lakes, Oscillatoria limosa, which is benthic (i.e. it grows attached to sediments and
rocks), is a microcystin producer. In spite of the widespread occurrence of
cyanobacterial blooms in Australia, Planktothrix blooms are rare there. This may be due
to the higher temperature and tendency for elevated clay-derived turbidity in Australian
water bodies.
Table 3.5 Toxic cyanobacteria species and their geographical distribution
Toxic species Cyanotoxin Location Reference(s)

Anabaena flos-aquae Microcystins Canada Krishnamurthy et al., 1989;
Harada et al., 1991
Anabaena ? Microcystins Denmark Henriksen et al., 1996b
Anabaena spp. Microcystins Egypt Yanni and Carmichael, 1997
Anabaena spp. (flos-aquae, Microcystins Finland Sivonen et al., 1990b; 1992a
lemmermannii, circinalis)
Anabaena circinalis Microcystins France Vezie et al., 1998
Anabaena flos-aquae Microcystins Norway Sivonen et al., 1992a
Microcystis aeruginosa Microcystins Worldwide Several; see Rinehart et al.,
1994 for a summary
M. viridis Microcystins Japan Kusumi et al., 1987; Watanabe
et al., 1986
M. botrys Microcystins Denmark Henriksen et al., 1996b
Planktothrix agardhii Microcystins China Ueno et al., 1996a
P. agardhii Microcystins Denmark Henriksen et al., 1996b
P. mougeotii Microcystins Denmark Henriksen et al., 1996b
P. agardhii Microcystins Finland Sivonen, 1990b; Luukkainen et
al., 1993
P. agardhii Microcystins Norway Krishnamurthy et al., 1989;
Meriluoto et al., 1989
Oscillatoria limosa Microcystins Switzerland Mez et al., 1996
Nostoc sp. Microcystins Finland Sivonen et al., 1990a, 1992b
Nostoc sp. Microcystins England Beattie et al., 1998
Anabaenopsis millerii Microcystins Greece Lanaras and Cook, 1994
Haphalosiphon hibernicus Microcystins USA Prinsep et al., 1992
(soil isolate)
Nodularia spumigena Nodularins Australia Baker and Humpage, 1994;
Jones et al., 1994
N. spumigena Nodularins Baltic Sea Sivonen et al., 1989b
N. spumigena Nodularins New Carmichael et al., 1988a;
Zealand Rinehart et al., 1988
Aphanizomenon Cylindrospermopsin Israel Banker et al., 1997
ovalisporum
Cylindrospermopsis Cylindrospermopsin Australia Hawkins et al., 1985; 1997
raciborskii
C. raciborskii Cylindrospermopsin Hungary Törökné, 1997
Umezakia natans Cylindrospermopsin Japan Harada et al., 1994
Anabaena flos-aquae Anatoxin-a Canada Carmichael et al., 1975; Devlin
et al., 1977
Anabaena spp. Anatoxin-a Finland Sivonen et al., 1989a
Anabaena blooms Anatoxin-a Germany Bumke-Vogt, 1998
Anabaena sp. Anatoxin-a Ireland James et al., 1997
Anabaena sp. Anatoxin-a (minor Japan Park et al., 1993a
amounts)
Anabaena planctonica Anatoxin-a Italy Bruno et al., 1994
bloom
Aphanizomenon sp. Anatoxin-a Finland Sivonen et al., 1989a
Aphanizomenon blooms Anatoxin-a Germany Bumke-Vogt, 1998
Cylindrospermum sp. Anatoxin-a Finland Sivonen et al., 1989a
Microcystis sp. Anatoxin-a (minor Japan Park et al., 1993a
amounts)
Oscillatoria sp. benthic Anatoxin-a Scotland Edwards et al., 1992
Oscillatoria sp. ? Anatoxin-a Ireland James et al., 1997
Planktothrix sp. Anatoxin-a Finland Sivonen et al., 1989a
Planktothrix formosa Homoanatoxin-a Norway Skulberg et al., 1992
Anabaena flos-aquae Anatoxin-a(S) Canada Matsunaga et al., 1989;
Mahmood and Carmichael,
1987
A. lemmermannii Anatoxin-a(S) Denmark Henriksen et al., 1997;
Onodera et al., 1997a
Anabaena circinalis Saxitoxins Australia Humpage et al., 1994; Negri et
al., 1995; 1997
Aphanizomenon flos-aquae Saxitoxins USA Jackim and Gentile, 1968;
Ikawa et al., 1982; Mahmood
and Carmichael, 1986
Cylindrospermopsis Saxitoxins Brazil Lagos et al., 1997
raciborskii
Lyngbya wollei Saxitoxins USA Carmichael et al., 1997;
Onodera et al., 1997b
The toxicity of the species listed in the table is in most cases verified by laboratory
studies with isolated strains. A few bloom samples are also included from the new areas
of occurrence where toxicity of the species is not verified by strain isolation but the toxins
are determined in the bloom samples. The authors have suggested the listed species as
the probable toxin producer (based on their dominance) but these reports should be
treated as tentative until pure strains are studied.
Nostoc rivulare blooms in Texas, USA have caused poisoning of domestic and wild
animals (Davidson, 1959) and, more recently, two unidentified Nostoc strains were
shown to produce microcystins (Table 3.5).
The hepatotoxin, cylindrospermopsin, has been found in Cylindro-spermopsis raciborskii

in Australia and Hungary, in Umezakia natans in Japan, and in Aphanizomenon
ovalisporum in Israel (Table 3.5). In spite of their occurrence in Europe, it appears that
cylindrospermopsin-producing genera most commonly form toxic blooms in subtropical,
tropical or arid zone water bodies. However, there have been reports of increasing
occurrences of Cylindrospermopsis raciborskii in Europe and the USA (Padisák, 1997).
The neurotoxin, anatoxin-a, was first shown to be produced by Anabaena flos-aquae

strains originating from Canada, and later by Finnish strains of unidentified Anabaena
species, and in individual species of Oscillatoria, Aphanizomenon, and Cylindrospermum,
by benthic Oscillatoria from Scotland, and by Anabaena and Oscillatoria in Ireland. It
also was present in Anabaena planctonica blooms in Sardinia, Italy, in Anabaena and
Aphanizomenon blooms in Germany, and in minor amounts in some Japanese bloom
samples, as well as in Anabaena strains. Homoanatoxin-a has been characterised from
an Oscillatoria formosum (Phormidium formosum) strain from Norway (see Table 3.5).
To date, anatoxin-a(S) has been found only from Anabaena species: A. flos-aquae in the
USA and Scotland, and A. lemmermannii in Denmark.
Aphanizomenon flos-aquae blooms and strains were found to contain saxitoxins in the
USA and this species was for a long time the only known saxitoxin producer amongst
the cyanobacteria. More recently, saxitoxins have been shown to be common in
Australian rivers and reservoirs and to be produced by Anabaena circinalis. In North
America, a benthic freshwater Lyngbya wollei was found to produce saxitoxins, as was a
strain of Cylindrospermopsis raciborskii in Brazil (Table 3.5).
3.2.3 Cyanotoxin patterns in strains and species of cyanobacteria
Cyanobacteria may produce several toxins simultaneously. In general, more than one
microcystin has been characterised from the strains listed in Table 3.2. Among
neurotoxic strains, several PSPs are found in the same strain, although there are
considerable variations between species (Table 3.3). Furthermore, simultaneous
neurotoxin and hepatotoxin production has been noted; the best example studied being
the Anabaena flos-aquae strain NRC 525-17 which produces anatoxin-a(S) (Matsunaga
et al., 1989) and several microcystins (Harada et al., 1991a).
Microcystin
Although many strains produce several microcystins simultaneously, usually only one or
two of them are dominant in any single strain. Qualitative variation in the microcystins
present is most frequently found among strains of Anabaena but also in Microcystis
(Sivonen et al., 1995). Some taxa have a number of microcystins in common, such as
planktonic Anabaena, Microcystis and Planktothrix (Oscillatoria). However, there is
evidence of microcystin variants that are typical for certain cyanobacterial taxa.
Planktothrix and some strains of Anabaena produce only demethylmicrocystins (Table

3.2). Planktothrix (Oscillatoria) isolates from Finland (13 strains studied and toxins fully
characterised) seem to produce one of two types of microcystin (D-Asp3-RR or Dha7-RR)
(Sivonen et al., 1995).
German field samples dominated by Planktothrix have also shown these microcystins,
with dominance of one major demethylated microcystin in populations of P. rubescens,
and two or three of these variants in populations of P. agardhii (Fastner et al., 1998).
Microcystis strains from Japan appear to contain chiefly microcystin-LR, -RR and -YR,
with some cultures showing all three variants, and some strains being dominated by one
of them. These three microcystins are the only variants reported in several studies on M.
aeruginosa and M. viridis (e.g. Watanabe, 1996). Many of the microcystins listed in
Table 3.2 have been found only in minor amounts or, to date, have been found to be
produced only by individual isolates.
In natural samples which usually contain many strains, or more than one toxin-producing
species, different combinations of microcystins can be found. For example, in a
Microcystis bloom from Homer Lake, USA, 19 different microcystins were characterised
(Namikoshi et al., 1992a, 1995) and in one Australian bloom of Microcystis aeruginosa,
23 microcystins were detected by high pressure liquid chromatography (HPLC), none of
which were microcystin-LR (Jones et al., 1995).
Microcystin-LR is often mentioned as the most frequently occurring microcystin, although

such observations may be biased by the fact that a chemical standard for the analysis of
microcystin-LR was the earliest to be commercially available (see Chapter 13). It has
been reported to be the major toxin in bloom and strain samples from Portugal
(Vasconcelos et al., 1995. 1996), France (Vezie et al., 1997), Canada (Kotak et al., 1993)
and frequently co-occurring with microcystin-RR and -YR in Japan (Watanabe et al.,
1988, 1989).
Geographical patterns are indicated by some survey results. Wide variation among the
L-amino acids of microcystins has been reported for South Africa (Scott, 1991), frequent
presence of demethylmicrocystins (RR and LR) has been seen in Finnish strains
(Sivonen et al., 1995) and microcystins in Danish samples show wide variation
(Henriksen, 1996a). In part, these patterns probably reflect regional differences in
dominance of cyanobacterial species or strains. Water bodies with regular dominance of
specific taxa are likely to exhibit characteristic patterns of microcystin variants.
Nodularin
The cyanotoxin, nodularin, is found in waters where Nodularia spumigena is present; the
most prominent areas being the Baltic Sea and brackish water estuaries and coastal
lakes of Australia and New Zealand. However, the best known Nodularia spumigena
bloom location, Lake Alexandrina, Australia, has salinities which are only slightly
elevated above normal river water and at levels still suitable for drinking water. The
presence of variants of nodularin in environmental samples is usually rather insignificant.
In the Baltic Sea, the collection of samples for several years has shown nodularin to be
the major compound present. The same is true for the almost 90 hepatotoxic Nodularia
strains isolated from the same source (Sivonen et al., 1989b; Lehtimäki et al., 1997).
Analyses of several strains isolated from blooms across Australia have revealed similar
results, with nodularin variants being found rarely, and then only at low relative
abundance (Jones et al., 1994; Blackburn et al., 1997).
Saxitoxins
There is much diversity of saxitoxin distribution in cyanobacteria from around the world
(Table 3.3). In addition, the lack of analytical standards for many saxitoxins has probably
restricted findings in some countries. Saxitoxin-producing Anabaena circinalis blooms
are widespread in Australian rivers and reservoirs, and the relative abundance of
individual toxins is remarkably constant in toxin-producing strains, which is quite
opposite to the microcystin-producing cyanobacteria. In all healthy Anabaena circinalis
cultures examined in Australia, the relative composition of individual saxitoxins is very
similar and dominated by C-toxins (Blackburn et al., 1997). As blooms and cultures age,
the proportion of decarbamoyl-gonyautoxins breakdown products increases at the
expense of the C-toxins (see section 3.4). In North American Aphanizomenon flos-aquae
and in Brazilian Cylindrospermopsis raciborskii samples, mostly neosaxitoxin and
smaller amounts of saxitoxin have been detected. In the case of the Aphanizomenon
flos-aquae from North America, only bloom samples and strains from New Hampshire,
USA have been found to contain saxitoxins. Mat-forming Lyngbya wollei from Alabama,
USA was found to produce decarbamoyl saxitoxin (dcSTX), decarbamoylgonyautoxin-2
and-3 (dcGTX2 and dcGTX3) and six new saxitoxins (see section 3.1).
3.2.4 Concentrations of cyanotoxins in surface waters
Information on the concentrations of cyanotoxins in surface waters has been appearing

only recently in the international literature. In early studies (pre-1990s), the toxicity of
bloom samples was determined by mouse bioassays, but this method is unsuitable for
measuring the low concentrations of cyanotoxins that usually prevail in cyanobacterial
populations when they do not accumulate in scums. The development of better
analytical methods, in the first instance HPLC and more recently enzyme linked
immunosorbent assay (ELISA), (and for microcystins and nodularins also the protein
phosphatase assay) has made the quantification of total and individual toxins possible
(see Box 3.1 and Chapter 13).
The quantitative determination of toxin concentrations is mostly performed from

lyophilised (freeze-dried) cultures, bloom samples or seston (particulate material
suspended in water, which contains not only cyanobacterial cells but usually other algae,
some zooplankton, and possibly inorganic material such as soil and sediment particles)
(see Chapter 13). Results are usually expressed as milligrams or micrograms of toxin
per gram dry weight (dw). Whereas in cultures and bloom samples, the dry weight
originates from cyanobacteria, it will encompass further particles (seston) in plankton
samples taken outside of scum areas. The highest published concentrations of
cyanotoxins from cyanobacterial bloom samples, measured by HPLC, are (see Table
3.6):
• microcystin - 7,300 µg g-1 dw from China and Portugal,

• nodularin - 18,000 µg g-1 dw from the Baltic Sea,
• cylindrospermopsin - 5,500 µg g-1 dw from Australia,
• anatoxin-a - 4,400 µg g-1 dw from Finland,
• saxitoxins - 3,400 µg g-1 dw from Australia,
• anatoxin-a(S) - 3,300 µg g-1 dw from the USA.
Box 3.1 Toxins and toxicity: what's in a name?
There is often a misunderstanding of the terms "toxicity" or "toxin content" when applied to
cyanobacteria or a water sample that contains cyanotoxins. In particular, the terms are often
taken to be synonymous, which they are not. Furthermore, the measurement units in which data
are reported are not always carefully considered.
To some extent, the problem lies in the different ways in which toxin data can be expressed, and
the way in which people think about the cyanobacterial or water sample, particularly in the context
of their own professional background. For example, a water treatment plant operator may assume
or expect a "toxicity" value to refer to the toxin concentration per litre of drinking water, a biologist
may think of the same term to mean the amount of toxin per mass of cyanobacteria, whereas a
toxicologist will normally take the term to reflect the amount of cyanobacteria needed to kill an
animal.
Toxicity. In the strict sense, toxicity refers only to animal testing data, and is expressed as the
amount of cyanobacteria lethal to an animal (usually normalised per kilogram of body weight).
The commonly reported LD50 value, a measure of toxin potency, is the amount of cyanobacteria
or pure toxin needed to kill 50 per cent of animals in an experimental trial, again normalised per
kilogram of body weight. Thus, the lower the LD50 the more potent the cyanobacterial sample or
pure toxin (see Table 3.2 for examples). Note, however, that cell growth assays may express
results as EC50 (concentration reducing growth rate by 50 per cent).
Toxin concentration and toxin content. Toxin concentration can refer either to the amount of toxin
per litre of water or the amount of toxin per mass of cyanobacterial bloom material. Therefore,
particular attention must be paid to the units in which the data are reported. In the published
literature on toxic cyanobacteria, the amount of toxin per mass of bloom material is often referred
to as the toxin content. Strictly speaking, this is incorrect; the correct term should be the
gravimetric, or per cyanobacterial mass, toxin concentration.
Toxin quota. This refers to the amount (mass or moles) of toxin per cyanobacteria cell.
Cell bound versus free toxin. In most circumstances almost all toxins are cell-bound (within the
cells) and little or no toxin is found in solution except where a bloom is senescent or where an
algicide has been applied.
Higher concentrations have been reported in unpublished studies and may be expected
under some circumstances.
For the purposes of water treatment and public health management, toxin concentration
per litre of water is often a more relevant unit, for example for relating ambient
concentrations to guideline levels. Concentration per litre usually refers to toxins
contained within the cells as well as dissolved in water, and taken from a defined volume
of water. Very high concentrations of microcystins per litre of water (rather than per
mass of cyanobacteria), have been reported up to 25,000 µg l-1 microcystin and up to
3,300 µg l-1 anatoxin-a(S) (Table 3.6). It should be noted, however, that these very high
concentrations of toxins would be from scums or from very dense accumulations of
cyanobacteria.
In one or two studies where many toxin concentration measurements have been made
from more or less randomly collected plankton samples, the distribution of data revealed
a small number of samples with high toxin concentrations, some samples with moderate
concentrations, and many samples with low or zero concentrations (Figure 3.4).
However, such surveys tend to include a large proportion of samples where the particle
content is not dominated by cyanobacteria. If cyanobacteria constitute only a small
fraction of the total dry weight, toxin concentration per gram dry weight will necessarily
be low. In contrast, if mass developments of toxic cyanobacterial species are
investigated, cyanobacteria will constitute a large fraction of the total dry weight. In these
cases, toxin content is often high. Recent German data show that the relative frequency
of high, moderate and low toxin concentrations per gram dry weight is dependent on the
particular species that is dominant (Figure 3.5).
For management, the option of estimating toxin content from the biomass of a dominant
cyanobacterial species (i.e. the "toxin quota" per cell or biovolume) can be helpful.
Survey data from Germany show that microcystin toxin quotas differ between taxa, but
variation within most of the samples dominated by the same taxon is only moderate
(two- to fivefold) (Figure 3.5). In contrast, an Australian investigation with other species
of cyanobacteria showed substantial variation of toxin concentrations per dry weight,
even in cases with dominance of the same cyanobacterial species (Negri et al., 1997).
Table 3.6 Toxin concentrations reported in cyanobacterial bloom or water samples
No. of
Range of total
toxic
Period concentrations
samples Toxins Analysis
Location of (µg g-1 dw, Reference
(total no. identified method
study unless otherwise
of
indicated)
samples)
Microcystins
Australia 1991 4 Microcystins, 2,100-4,1002 HPLC Jones et al.,
24 unidentified 1995
Canada, 1990 37(50) Microcystin-LR 4-610 HPLC Kotak et al.,
Alberta 1993
Canada, 1990- 168(226) Microcystin-LR 1-1,550 HPLC Kotak et al.,
Alberta (3 93 1995
lakes)
China 1988 5(10) Microcystin- 200-7,300 HPLC Zhang et al.,
RR, -LR 1991
Czech and 1995- (63) Microcystin-LR 4-6,835 HPLC Maršálek et
Slovak 96 al., 1996
Rep.
Denmark 1992- Microcystin- 3-2,800 HPLC Christoffersen,
94 RR, -LR 1996
Denmark 1993- 198(296) Microcystins 5-1,900 HPLC Henriksen et
95 al., 1996b
Finland 1994- 17(20) Microcystin-LR > 10-800 HPLC Lahti et al.,
95 1997
France 1994 16(22) Microcystins 70-3,970 HPLC Vezie et al.,
1997
France, L. 1994 19(30) Microcystins 30-230 HPLC Vezie et al.,
Grand- 1998
Lieu
Germany 1992 8(15) Microcystin-LR 36-360 HPLC Fastner, 1994
1,2
Germany 1993 17(18) Microcystins 0.15-36 ELISA Ueno et al.,
1996b
Germany 1995- 385(533) Microcystins 1-5,000 HPLC Fastner, 1998
96
Germany 1997 34 Microcystins, 1-25,000 1 HPLC Chorus et al.,
several 1998
Japan 1990 12(14) Microcystin- 160-950 HPLC Watanabe et
RR, -YR, -LR al., 1992
Japan 1988- 11(19) Microcystin- 70-1,610 HPLC Park et al.,
92 RR, -YR, -LR, 1993a
[Dha7] -LR
Japan, 1980- 13 Microcystin- 30-2,100 HPLC Park et al.,
Lake 91 RR, -YR, -LR 1993b
Suwa
Japan 1986- 4(4) Microcystin- 100-860 HPLC Shirai et al.,
88 RR, -YR, -LR 1991
Japan 1992- 18(22) Microcystin- 0.04-480 1 HPLC Tsuji et al.,
95 RR, -YR, -LR 1996
Japan 1993- 46(57) Microcystins 0.05-1,300 1,2 ELISA Ueno et al.,
95 1996b
Japan 1993- 12(17) Microcystins 0.06-94 1,2 ELISA Nagata et al.,
94 1997
Japan 1989- 10(10) Microcystins 300-15,600 1,2 ELISA Nagata et al.,
94 1997
330-19,500 1,2 HPLC
Portugal 1989- 12(12) Microcystin-LR 1,000-7,100 HPLC Vasconcelos
92 plus six known et al., 1996
and three
unidentified
microcystins
Portugal 1994- 28(29) Microcystins 0.1-37 1,2 ELISA Ueno et al.,
95 1996b
South 1985- Microcystin- 5-420 HPLC Wicks and
Africa 86 FR, -LR, -YR, Thiel, 1990
-LA, -YA, -
LAba
South 1988- 9(9) Microcystin- 40-630 HPLC Scott, 1991
Africa 89 YR, -LR, -FR,
-YA, -LA, -
LAba
UK 1992 3(3) Microcystins, 3 17-131 1,2 HPLC Codd et al.,
1995
USA, 1993 9 Microcystins 1,900-12,800 2 ELISA McDermott et
Wisconsin al., 1995
Nodularins
Baltic Sea 1985- 17(23) Nodularin < 100-2,400 HPLC Sivonen et al.,
87 1989b
Baltic Sea 1990- 6(16) Nodularin 300-18,000 HPLC Kononen et
91 al., 1993
Tasmania, 1992- 7(9) Nodularin 2,000-3,500 HPLC Jones et al.,
Orielton 93 1994
Lagoon
Anatoxins
Denmark 1993- 9(10) Anatoxin-a(S) 4-3,300 3 ChE Henriksen et
95 inhibition al., 1997
assay
Finland 1985- 13(30) Anatoxin-a 10-4,400 GC/MS Sivonen et al.,
87 1989a
Finland 3(3) Anatoxin-a 0.4-2,600 LC/MS Harada et al.,
and Japan 1993
Germany 1995- 10(45) Anatoxin-a 0.02-0.36 1 GC-ECD Bumke-Vogt,
96 1998
Ireland 1995 2(2) Anatoxin-a 10-100 HPLC James et al.,
(fluorimetric) 1997
Japan 1988- 9(14) Anatoxin-a 0.3-16 TSP-LC/MS Park et al.,
92 1993a
Saxitoxins
Australia 1990- 11(11) Saxitoxins 85-2,040 HPLC and Humpage et
92 FAB-MS al., 1994
Australia 1992- 24(31) Saxitoxins 50-3,400 HPLC Negri et al.,
94 1997
USA 1994 7(8) Saxitoxins 5-60 4 HPLC/AOAC Carmichael et
al., 1997
dw Dry weight
HPLC High pressure liquid chromatography
ELISA Enzyme linked immunosorbent assay
GC/MS Gas chromatography/mass spectrometry
LC/MS Liquid chromatography/mass spectrometry
GC-ECD Gas chromatography-electron capture detection
TSP-LC/MS Thermospray-liquid chromatography/mass spectrometry
FAB-MS Fast atom bombardment-mass spectrometry
AOAC Mouse bioassay done according to the Association of Official Analytical Chemists
1
Given as µg l-1
2
Microcystin-LR used as standard
3
Measured by enzyme inhibition
4
Micrograms of STX equivalents
Figure 3.4 Cell-bound total microcystin content (measured by HPLC) of samples
taken in Denmark between 1993 and 1995 and dominated by different
cyanobacteria. Boxes show median values and the values within the 50 percentile
range; bars indicate the 10th and 90th percentile; n = number of samples (Modified
from Henriksen, 1996)
Further regional investigations are needed to clarify the toxins and species for which
toxin quotas may be reasonably stable and, thus, predictable.
The further development of quantitative, pre-concentration methods coupled with the

highly sensitive analytical methods noted above (see also Chapter 13) has also enabled
the measurement of very low concentrations of extracellular, dissolved toxins in water.
Data for dissolved toxin concentration, as reported in the scientific literature, are given in
Table 3.7. Prevalence and degradation of dissolved toxins in water are discussed in
section 3.4. Concentrations of microcystins dissolved in water vary from trace
concentrations up to 1,800 µg l-1 or higher, following the collapse of a large, highly toxic
bloom.
3.2.5 Seasonal variations in bloom toxin concentration
The timing and duration of the bloom season of cyanobacteria depends largely on the
climatic conditions of the region. In temperate zones, mass occurrences of
cyanobacteria are most prominent during the late summer and early autumn and may
last 2-4 months. In regions with more Mediterranean or subtropical climates, the bloom
season may start earlier and persist longer.
Figure 3.5 Cell-bound total microcystin content (measured by HPLC) of samples
taken in Germany between 1995 and 1996 dominated by different cyanobacteria. A.
On a dry weight basis; B. On a chlorophyll a basis; C. On a volume basis. (Micro.
spp = Microcystis spp.; Plankto. agardhii = Planktothrix agardhii; Plankto.
rubescens = Planktothrix rubescens) Boxes show median values and the values
within the 50 percentile range; bars indicate the 10th and 90th percentile; n =
number of samples (Modified from Fastner et at., 1998)
Table 3.7 Dissolved (extracellular) toxin concentrations measured in water samples
No. of toxic
Period
samples (total Toxin Concentration Analysis
Location of Reference
no. of identified (µg l-1) method
study
samples)
Australia 24 Microcystins 3-1.8001 HPLC Jones and
Orr, 1994
China 1993-94 130(835) Microcystins 0.05-1.6 ELISA Ueno et al.,
1996a
Finland 1993-94 38(38) Microcystin- 0.06-0,21 HPLC and Lahti et al.,
LR ELISA 1997
Germany 1993-94 11(19) Microcystins 0.07-0.76 ELISA Ueno et al.,
1996b
Japan 1992-95 9(22) Microcystin- 0.02-3.8 HPLC Tsuji et al.,
RR, -YR, -LR 1996
Japan 1993-95 26(38) Microcystins trace-5.6 ELISA Ueno et al.,
1996b
Japan 1993-94 4(13) Microcystins 0.08-0.8 ELISA Nagata et
al., 1997
Thailand 1994 7(10) Microcystins 0.08-0.35 ELISA Ueno et al.,
1996b
USA, 1993 27(27) Microcystins 0.07-200 ELISA McDermott
Wisconsin et al., 1995
HPLC High pressure liquid chromatography

ELISA Enzyme linked immunosorbent assay
1
High range concentrations following treatment of a large bloom with algicide, which
released intracellular microcystins (see section 3.4)
In France, four months is not uncommon, and in Japan, Portugal, Spain, South Africa
and southern Australia blooms may occur for up to six months or longer. By contrast, in
dry years, in tropical or subtropical areas of China, Brazil and Australia, cyanobacterial
blooms may occur almost all year round, perhaps waning only briefly during reservoir
overturn.
In shallow lakes, particularly in north-western Europe, populations of Planktothrix

agardhii (Oscillatoria agardhii) may prevail perennially for many years. In deeper,
thermally stratified lakes and reservoirs with moderate nutrient pollution, Planktothrix
rubescens (Oscillatoria rubescens) may form blooms at the interface between the
warmer upper and colder deeper layers of water during summer, but maintain high,
evenly distributed density throughout the entire water body during winter. Both
Planktothrix species may contain high amounts of microcystins (see Figure 3.5). Blooms
of cyanobacteria, especially Planktothrix agardhii, have been found in winter under ice in
Scandinavian and German lakes and can thus be an all year round problem.
Although toxic cyanobacteria occur in a large number of lakes, reservoirs and rivers in
the world, quantitative reports on seasonal variation of cyanobacterial species
composition and toxin concentration are rare. Only a few studies on seasonal, spatial
and diel (day to night) variations in lakes have been published. Carmichael and Gorham
(1981) showed a high degree of spatial variation of bloom toxicity that was due mostly to
variations in the relative amounts of toxic Microcystis aeruginosa throughout the lake,
rather than to substantial variations in cell toxin content. Other measurements of toxin
concentrations in lakes have revealed similar trends; samples taken at the same time
from different parts of the lake may show wide divergence in cyanotoxin content
(Ekman-Ekebom et al., 1992; Kotak et al., 1995; Vezie et al., 1998). A study in Alberta,
Canada, showed considerable variation in toxin concentrations among the three lakes
studied, both within and between years, even though the lakes were located within the
same climatic region (Kotak et al., 1995).
In any year or season, individual water bodies have their own populations of
cyanobacteria and algae, the dominance of which is dependent not only on the weather,
but on the specific geochemical conditions of the lake. If there are no major changes in
these conditions, toxic blooms are likely to recur annually in those lakes that have a
history of toxic blooms (Wicks and Thiel, 1990; Ekman-Ekebom et al., 1992). Certain
species, including the highly toxic Planktothrix agardhii and P. Rubescens, are known to
produce maximum mass occurrences deep in the water column and which may be
overlooked by surface monitoring of waters. Such situations may also cause problems
for water treatment (see Chapters 2 and 9) (Lindholm and Meriluoto, 1991).
Studies over prolonged periods usually show that toxin concentration per gram dry
weight may vary substantially over a time scale of weeks to months, but rarely from day
to day as is sometimes reported. The maximum toxin concentration per gram dry weight
is usually reported in summer or autumn, when cyanobacterial biomass dominates dry
matter (Wicks and Thiel, 1990; Watanabe et al., 1992; Park et al., 1993b; Kotak et al.,
1995; Maršálek et al., 1995; Vezie et al., 1998). However, the time of toxin concentration
maximum and biomass maximum are not necessarily coincident. Thus, there can be
significant variation in the amount of toxin per mass of cyanobacteria over time,
independently of changes in the size of the cyanobacterial population. The explanations
for this are twofold. Firstly, there may be a waxing and waning of species or strains of
quite different toxin quotas (i.e. toxin content per cell). Secondly, the toxin quotas may
change up to five-fold in response to changes in environmental conditions (see section
3.3). A study by Kotak et al. (1995) found substantially higher concentrations of
microcystin in blooms during the day than at night, whereas a study from Australia found
no variation in microcystin content when samples were incubated during 24 hours at
different depths in a reservoir (Jones and Falconer, 1994). Both findings need to be
explored further.
High regional, seasonal, spatial, temporal and diel variations of toxin concentrations
indicate that predicting or modelling the occurrence of toxin concentrations requires a
comprehensive understanding of population (strain) development in different types of
aquatic ecosystems, as well as of the variability of their toxin quotas. Data bases for
such predictive models have yet to be compiled.
3.3 Production and regulation
Laboratory studies with pure strains of cyanobacteria have found that environmental
factors can induce changes in toxicity or toxin concentration (on a per unit biomass
basis), but usually by a factor of no more than three or four. On a per cell basis, the
changes in toxin content are probably even smaller. These environmentally-induced
changes are far less than the range of more than three orders of magnitude in toxin
content measured between individual strains grown in culture under identical conditions.
This lends support to the assumption that much, if not most, of the variation in toxicity of
"monospecific" natural blooms is the waxing and waning of strains of the same species,
but with varying toxin quotas. The factors that control the growth and toxin content of
individual strains are as yet unknown, but clearly the genetic regulation of cyanotoxin
production is an important area for further study and understanding (see section 3.3.3).
Both toxigenic (toxin producing) and non-toxigenic strains exist within many species of
cyanobacteria. When grown in the laboratory, particular strains always produce much
greater amounts of toxins than others. Indeed, the difference may be as much as three
orders of magnitude or more (Bolch et al., 1997). Several attempts to differentiate toxin
producing (toxigenic) from non-toxic strains of the same species using microscopic
methods have failed. The use of molecular biological methods to characterise toxic and
non-toxic planktonic isolates has been initiated recently and will, in future, help to clarify
the taxonomic status of these organisms, as well as the ecology of individual strain types
(see section 3.3.3).
3.3.1 Regulation by chemical and physical factors
The production of toxin by a single cyanobacterial strain seems to be consistent and the
spontaneous and permanent loss of toxin production has been seldom reported. The
effects of several environmental factors on growth and toxin production by cyanobacteria
have been studied in batch and continuous culture experiments. Culture age in batch
cultures, and temperature, are the parameters most frequently examined, followed by
light, nutrients, salinity, pH and micronutrient concentrations. Studies have been done
with hepatotoxic Microcystis, Oscillatoria (Planktothrix), Anabaena and Nodularia;
anatoxin-a producing Anabaena, Aphanizomenon and Planktothrix; and saxitoxin
producing Aphanizomenon and Anabaena circinalis (Table 3.8). Microcystins and
anatoxin-a are largely retained within cells when the conditions for the growth of the
organism are favourable. The amount of microcystin in a culture increases during the
logarithmic growth phase, being highest in the late logarithmic phase. The maximum
anatoxin-a concentration is found during the logarithmic growth phase (Sivonen, 1996;
Watanabe, 1996). While the variants of microcystins produced by a particular strain are
rather constant, the ratios of individual microcystins may change with time, or under
conditions of different temperatures and light. Environmental factors affect toxin content
of cyanobacteria, but only within a range of less than an order of magnitude. The
majority of studies indicate that cyanobacteria produce most toxins under conditions
which are most favourable for their growth. For example, different cyanobacterial
species have different light requirements: Planktothrix prefers low light intensities for
growth, Anabaena moderate and Aphanizomenon high light intensities. All strains
produce most toxin when grown under their optimum light conditions (Table 3.8). Two- to
threefold differences in toxin content (on a per unit biomass basis) have been reported in
relation to light conditions.
Strains and species also differ slightly in their optimum growth temperatures. The toxin
content in most studies was highest at temperatures between 18 °C and 25 °C, whereas
low (10 °C) or very high temperatures (30 °C) decreased toxin content. Temperature
gradients caused two- to threefold differences in toxin content.
In a study using mouse bioassay to detect effects of pH on toxin production, cells were
found to be more toxic when grown at high and low pH (Van der Westhuizen and Eloff,
1983).
Table 3.8 Laboratory studies on cellular toxin concentrations in cyanobacteria
Changes in
Toxin(s)/ Highest/lowest
toxin
Parameter Organism analysis toxin Reference(s)
concentrations
method production
(dw)
Temperature (°C)
12.5-30 Anabaena spp. Microcystins 3.5-30 fold1 Highest at 25, Rapala et al.,
(2 strains), batch HPLC lowest at 30; 1997
cultures different toxins
at different
temperatures
10-28 Anabaena spp. Microcystins 3-10 fold Lowest at 10, Rapala and
(2 strains), HPLC highest at 25 Sivonen, 1998
continuous
cultures
15-30 Anabaena spp. Anatoxin-a 3 fold Lowest at 30, Rapala et al.,
(2 strains), batch HPLC highest at 20 1993
cultures
10-28 Anabaena spp. Anatoxin-a 4-7 fold Highest at 19- Rapala and
(2 strains), HPLC 21, lowest at 10 Sivonen, 1998
continuous and 28
cultures
15-30 Aphanizomenon Anatoxin-a 3 fold Lowest at 30, Rapala et al.,
sp. (1 strain), HPLC highest at 20 1993
batch cultures
10,25,34 Microcystis Microcystins 5 fold Highest toxicity Codd and
aeruginosa (1 mouse at 25, lowest at Poon, 1988
strain), batch bioassay 10
cultures
15-35 Microcystis Microcystins 4 fold Highest toxicity van der
aeruginosa (1 mouse at 20; different Westhuizen
strain) batch bioassay toxins at and Eloff,
cultures HPLC different 1985; van der
temperatures Westhuizen et
al., 1986
18,25,35 Microcystis Microcystins 1.4 fold Highest toxicity Watanabe
aeruginosa (1 mouse at 18, lowest at and Oishi,
strain), batch bioassay 32 1985
cultures
10-30 Nodularia Nodularin 3-4 fold Highest at 20, Lehtimäki et
spumigena (2 HPLC lowest at 10 or al., 1994
strains), batch 30
cultures
7-28 Nodularia Nodularin 3 fold Highest at 19 Lehtimäki et
spumigena (1 HPLC al., 1997
strain), batch
cultures
15-30 Oscillatoria Microcystins 7 fold Strain Sivonen,
agardhii (2 HPLC dependent; 1990b
strains), batch lowest at 30
cultures
Light (µmol m-2 s-1)
2-100 Anabaena spp. Microcystins 3 fold Highest at 25 Rapala et al.,
continuous (2 strains), batch HPLC 1997
cultures
7, 19, 42 Anabaena spp. Microcystins 2.5-15 Lowest at 10, Rapala and
continuous (2 strains), HPLC highest at 25 Sivonen, 1998
continuous
cultures
2-128 Anabaena spp. Anatoxin-a 3 fold Highest at 26- Rapala et al.,
continuous (2 strains), batch HPLC 44, lowest at 2 1993
cultures
7, 19, 42 Anabaena spp. Anatoxin-a No effect Highest at 19, Rapala and
continuous (2 strains), HPLC lowest at 7 Sivonen, 1998
continuous
cultures
2-128 Aphanizomenon Anatoxin-a 4 fold Highest at 128, Rapala et al.,
continuous sp. (1 strain), HPLC lowest at 2 1993
batch cultures
5-50 Microcystis Microcystins 2.4 fold Highest toxicity Codd and
continuous aeruginosa (1 mouse at 20 Poon, 1988
strain), batch bioassay
cultures
20-75 Microcystis Microcystins 2.5 fold Highest at 40 Utkilen and
continuous aeruginosa (1 HPLC Gjølme, 1992
strain),
continuous
cultures
21-205 Microcystis Microcystins 1.2 fold Highest toxicity van der
continuous aeruginosa (1 mouse at 142, lowest at Westhuizen
strain), batch bioassay 21 and Eloff,
cultures 1985
7.5, 30, 75 Microcystis Microcystins 3.8 fold Highest toxicity Watanabe
continuous aeruginosa (1 mouse at 30, lowest at and Oishi,
strain), batch bioassay 7.5 1985
cultures
25, 50, 80 Nodularia Nodularin No difference Lehtimäki et
continuous spumigena (2 HPLC al., 1994
strains), batch
cultures
2-155 Nodularia Nodularin 50 fold1 Higher at high Lehtimäki et
continuous spumigena (1 HPLC irradiances, al., 1997
strain), batch minimal at 2
cultures
12-95 Oscillatoria Microcystins 2.5 fold Highest at 12-44 Sivonen,
continuous agardhii (2 HPLC 1990b
strains), batch
cultures
Phosphorus (mg P l-1)
0.05-5.5 Anabaena spp. Microcystins 5 fold Highest at 5.5, Rapala et al.,
(2 strains), batch HPLC lowest at 0.05 1997
cultures
0.05-5.5 Anabaena spp. Anatoxin-a No difference No statistically Rapala et al.,
(2 strains), batch HPLC significant 1993
cultures differences
0.05-5.5 Aphanizomenon Anatoxin-a 2 fold Lowest at 0.05- Rapala et al.,
sp. (2 strains), HPLC 0.1, highest at 1993
batch cultures 0.5-5.5
BG-11 and Microcystis Microcystins 1.7 fold Higher toxicity Codd and
medium aeruginosa (1 Mouse without P Poon, 1988
without P strain), batch bioassay
cultures
0.0025, 0.025 Microcystis Microcystins 2.3 fold More toxin at Utkilen and
aeruginosa (1 HPLC 0.025 Gjølme, 1995
strain),
continuous
cultures
MA medium Microcystis Microcystins Less than 1 Highest toxicity Watanabe
1/1; dilutions aeruginosa (1 Mouse with the original and Oishi,
1/10, 1/20 strain), batch bioassay medium 1985
cultures
0.1-5.5 Oscillatoria Microcystins 1.8-2.5 fold Lowest toxin at Sivonen,
agardhii (2 HPLC 0.1 1990b
strains), batch
cultures
0.3, 0.6, 1.0 Nodularia Nodularin Less than 1 fold Lowest at 0.3 Lehtimäki et
spumigena (2 HPLC al., 1994
strains), batch
cultures
0-5.5 Nodularia Nodularin 4 fold Lowest at 0- Lehtimäki et
spumigena (1 HPLC 0.02, highest at al., 1997
strain), batch 0.2-5.5
cultures
Nitrogen (mg N l-1)
BG-11 Microcystis Microcystins 5 fold Higher toxicity Codd and
medium, aeruginosa (1 mouse with the medium Poon, 1988
medium strain), batch bioassay containing N
without N cultures
0.05-1 Microcystis Microcystins 3 fold Higher at high N Utkilen and
aeruginosa (1 HPLC Gjølme, 1995
strain),
continuous
cultures
MA medium Microcystis Microcystins 2.5 fold Highest toxicity Watanabe
1/1; dilutions aeruginosa (1 mouse with the original and Oishi,
1/10, 1/20 strain), batch bioassay medium 1985
cultures
0.42-84 Oscillatoria Microcystins 5 fold Higher at high Sivonen,
agardhii (2 HPLC N, lowest at low 1990b
strains), batch N
cultures
Micronutrients
Al, Cd, Cr, Cu, Microcystis Microcystins 1.7 fold Less toxins at Lukac and
Fe, Mn, Ni, Sn, aeruginosa (1 HPLC low Fe Aegerter,
Zn; various strain), batch concentrations 1993
concentrations cultures
0.1-3.4 µg Fe l- Microcystis Microcystins 1.5 fold More toxin at Utkilen and
1
aeruginosa (1 HPLC high Fe Gjølme, 1995
strain), concentrations
continuous
cultures
0.03-1.2 µg Fe Microcystis Microcystins 0-3 fold Less toxin at Lyck et al.,
l-1 aeruginosa (1 HPLC low Fe 1966
strain), concentrations
continuous
cultures
Salinity (‰)
3, 5, 8, 11 Nodularia Nodularin No difference No statistical Lehtimäki et
spumigena (2 HPLC difference al., 1994
strains), batch
cultures
0-30 Nodularia Nodularin 8 fold Highest at 15, Lehtimäki et
spumigena (1 HPLC lowest at 0 and al., 1997
strain), batch 30
cultures
0-35 Nodularia Nodularin 5 fold Highest at 12, Blackburn et
spumigena (6 HPLC lowest at 35 al., 1996
strains), batch
cultures
CO2
BG-11 Microcystis Microcystins 6 fold Higher toxicity Codd and
medium, aeruginosa (1 mouse with the medium Poon, 1988
medium strain), batch bioassay containing CO2
without CO2 culture
pH
1-14 Microcystis Microcystins 1.8 fold Toxicity highest van der
aeruginosa (1 mouse at low and high Westhuizen
strain), batch bioassay pH and Eloff,
cultures 1983
dw Dry weight
HPLC High performance liquid chromatography
1
When the growth of the strains was poor the amount of toxins was also very low (less
than 0.1 mg g-1 dw of cells); when these cases were compared to maximal toxin
production more than ten-fold differences could be seen
In high concentrations of phosphorus, hepatotoxic strains produced more toxins, but for
anatoxin-a production phosphorus had no effect (Table 3.8). The differences induced by
low and high phosphorus concentrations vary between two- and fourfold. Similarly, in
field studies, a positive correlation of total phosphorus with microcystin-LR concentration
in cells of Microcystis aeruginosa (Kotak et al., 1995) or in bloom material of Microcystis
spp. (Lahti et al., 1997b) has been found. Non-nitrogen fixing species, like Microcystis
and Oscillatoria, produce more toxins under nitrogen-rich conditions. Nitrogen fixing
species are not dependent on the nitrogen in the media for their toxin production (Rapala
et al., 1993; Lehtimäki et al., 1997). In continuous cultures when the toxins were
expressed in relation to cell protein rather than to dry weight, Utkilen and Gjølme (1995)
found that nitrogen and phosphorus limited conditions had no effect on the toxin content
of Microcystis aeruginosa.
Indications regarding the role of iron are contradictory (Lukac and Aegerther, 1993;
Utkilen and Gjølme, 1995; Lyck et al., 1996). While studying the effect of trace metals on
growth and on toxin content of Microcystis aeruginosa, Lukac and Aegerter (1993) found
that in batch cultures only zinc was required for both optimal growth and toxin production.
Orr and Jones (1998) have unified many of these rather disparate studies on the effect
of growth conditions on cyanotoxin production. They showed that the rate of microcystin
production by a cyanobacterial population in culture is directly proportional to its growth
rate, no matter what environmental factor was limiting growth. Moreover, they showed
that the amount of microcystin contained by a single cell of Microcystis aeruginosa (i.e.
on a per cell or cell quota basis) is constant within a narrow range (two- to threefold).
3.3.2 Biosynthesis
To understand how cyanotoxins are produced, it is necessary to study the biochemical

and genetic basis of toxin production. Knowledge of the biosynthetic pathways of
cyanotoxins is in its early stage and no complete biochemical pathways are known.
Biosynthesis of several cyanotoxins has been studied by feeding labelled precursors to a
cyanobacterial culture and following their incorporation into the carbon skeleton of the
toxins. Shimizu et al. (1984) used an Aphanizomenon flos-aquae strain to study
biosynthesis of saxitoxin analogues. They proposed a new pathway for neosaxitoxin
biosynthesis, the key steps of which are the condensation of an acetate unit, or its
derivative, to the amino group bearing an α-carbon of arginine or an equivalent, and a
subsequent loss of the carboxyl carbon and imidazole ring formation on the adjacent
carbonyl carbon. They established the origin of all the carbons in the toxin alkaloid ring
system. The side-chain carbon was derived from methionine (Shimizu, 1986).
Anatoxin-a is related structurally to the tropane class of alkaloids found in higher plants.
Based mainly on 14C-labelled precursors and enzymatic studies, Gallon et al. (1990) and
Gallon et al. (1994) suggested the biosynthesis of anatoxin-a to be analogous to that of
tropanes. Anatoxin-a was proposed to be formed from ornithine/arginine via putrescine,
which is oxidised to pyrroline, a precursor of anatoxin-a. Labelling experiments using 13C
NMR (nuclear magnetic resonance spectrometry) indicated that the carbon skeleton of
anatoxin-a is derived from acetate and glutamate. The studies showed that C-1 of
glutamic acid is retained during the transformation of anatoxin-a and not lost by
decarboxylation, a finding incompatible with the tropane alkaloid theory (Hemscheidt et
al., 1995b). All of the carbons of anatoxin-a(S) are derived from amino acids. Three
methyl carbons arise from L-methionine or other donors to the tetrahydrofolate C1 pool.
L-arginine accounts for C-2, C-4, C-5 and C-6 carbons of the toxin (Moore et al., 1992,
1993). The intermediate in the biosynthesis of anatoxin-a(S) from L arginine is (2S,4S)-4
hydroxyarginine (Hemscheidt et al., 1995a). The structure of the cylindrospermopsin
suggests a polyketide origin for the toxin (Moore et al., 1993).
The origin of carbons in microcystin (Moore et al., 1991) and in nodularin (Choi et al.,
1993; Rinehart et al., 1994) have been studied by following the incorporation of labelled
precursors into the toxins by NMR. Carbons C1-C8 of Adda in nodularin are acetate
derived and the remaining carbons presumably originate from phenylalanine. Methyl
groups in carbons 2, 4, 6, 8, and the O-methyl group in the Adda unit, originated from
methionine. The D-Glu and L-Arg carbons C4-C5 were acetate derived, with C1-C2
being from glutamate. Methyldehydrobutyrine was possibly formed from threonine, its
methyl group coming from methionine. The β-methylaspartic acid was found to originate
from condensation of pyruvic acid (C3-C4) and acetyl-CoA (C1-C2) (Rinehart et al.,
1994). The studies on the carbon skeleton of nodularin, with some minor differences,
agree with work on microcystin-LR by Moore et al. (1991). In their study, L-Leu and D-
Ala units in microcystin had acetate incorporation. The dehydroamino acid in microcystin
has been proposed to be formed from serine rather than from threonine (Rinehart et al.,
1994). Rinehart's group found linear nodularin, which was shown by culture experiments
to be a precursor of cyclic nodularin. Three additional linear peptides were isolated from
a bloom sample, one of them was possibly a precursor of cyclic microcystin-LR and the
others possibly degradation products (Rinehart et al., 1994).
3.3.3 Genetic regulation of cyanotoxin production
The genes and enzymes involved in cyanotoxin production are still mostly unknown. The
first molecular biological studies on toxic cyanobacteria investigated the possible
involvement of plasmids in toxin production. Four toxic strains of Microcystis aeruginosa
contained plasmids, and no plasmid could be shown in one toxic and in several non-
toxic strains (Schwabe et al., 1988). More recently, a similar study in Australia found no
evidence for plasmid involvement in microcystin synthesis (Bolch et al., 1997). Gallon et
al. (1994) studied an anatoxin-a producing Anabaena strain NCR 44-1, which
spontaneously became non-toxic. They found that the size of a plasmid was reduced in
that non-toxic clone, but this work has not been repeated or confirmed.
Multi-enzyme complexes and peptide synthetase genes are involved in hepatotoxin
production. Several cyclic and linear peptides, often with D-amino acids, are known to be
produced, non-ribosomally, by multi-domain peptide synthetases via the so-called
thiotemplate mechanism in bacteria and lower eukaryotes. The best characterised are
the synthesis of gramicidin S and tyrocidin by Bacillus. Peptide synthetase genes have
been detected and sequenced (partly) in Microcystis aeruginosa (Meissner et al., 1996)
and in Anabaena (Rouhiainen et al., 1994). Analogous polymerase chain reaction (PCR)
products to the peptide synthetase genes have been shown by using DNA from
Microcystis (Jacobs et al., 1995; Arment and Carmichael, 1996) and Nodularia as a
template. Dittman et al. (1997) showed, in knockout experiments, that peptide
synthetase genes are responsible for microcystin production.
At least some strains which produce hepatotoxins also produce other small cyclic
peptides (Namikoshi and Rinehart, 1996; Weckesser et al., 1996) which are likely to be
produced by nonribosomal peptide synthesis.
3.4 Fate in the environment

3.4.1 Partitioning between cells and water
It appears likely that cyanotoxins are produced and contained within the actively growing
cyanobacterial cells (i.e. they are intracellular or particulate). Release to the surrounding
water, to form dissolved toxin, appears to occur mostly, if not exclusively, during cell
senescence, death and lysis, rather than by continuous excretion.
In laboratory studies, where both intracellular and dissolved toxins

(microcystins/nodularin and saxitoxins) have been measured, it is generally the case that
in healthy log phase cultures, less than 10-20 per cent of the total toxin pool is
extracellular (Sivonen, 1990b; Lehtimäki et al., 1997; Negri et al., 1997; Rapala et al.,
1997). As cells enter stationary phase the increased rate of cell death may lead to an
increase in the extracellular dissolved fraction. Even during log-phase cell growth in
culture, a small percentage of cells in the population may be dying and lysing (and
releasing intracellular toxins), even though there is an overall positive population growth.
There are some indications that anatoxin-a may leak out of cells during growth
especially in low light conditions. High concentrations of anatoxin-a, sometimes
exceeding the intracellular pool of toxins, have been found in media in a batch culture
study (Bumke-Vogt et al., 1996).
In the field, healthy bloom populations produce little extracellular toxin. The range of
measured concentrations for dissolved cyanotoxins, in all cases except those where a
major bloom is obviously breaking down, is 0.1-10 µg l-1 (Lindholm and Meriluoto, 1991;
Jones and Orr, 1994; Tsuji et al., 1996; Ueno et al., 1996b; Lahti et al., 1997b) Cell-
bound concentrations are several orders of magnitude higher (see Tables 3.6 and 3.7).
In lakes or rivers, toxins liberated from cells are rapidly diluted by the large mass of
water, especially if mixing of water by wind action or currents is vigorous (Jones and Orr,
1994). However, the concentration of dissolved toxins may be much higher in ageing or
declining blooms. This is an important consideration for water treatment plant operators,
because it means that removal of healthy cyanobacterial cells intact from the raw water
supply may obviate or substantially reduce the need for additional adsorptive (activated
carbon) or oxidative (ozone or chlorine) toxin removal processes (see Chapter 9).
The release of toxins from cells is enhanced by chemical treatments for the eradication
of cyanobacteria, especially the use of algicides (either copper-based or organic
herbicides). Treatment of a bloom with copper sulphate, for example, may lead to
complete lysis of the bloom population within three days and release of all the toxins into
the surrounding water (Berg et al., 1987; Kenefick et al., 1992; Jones and Orr, 1994).
The efficacy of copper sulphate treatment is, however, very much dependent on water
chemistry, especially alkalinity, pH and dissolved organic content (see sections 8.5 and
9.2).
3.4.2 Chemical breakdown
The four main groups of cyanotoxins: microcystins, anatoxins, PSPs and

cylindrospermopsins, exhibit quite different chemical stabilities and biological activities in
water.
Microcystins
Microcystins, being cyclic peptides, are extremely stable and resistant to chemical
hydrolysis or oxidation at near neutral pH. Microcystins and nodularin remain potent
even after boiling. In natural waters and in the dark, microcystins may persist for months
or years. At high temperatures (40 °C) and at elevated or low pH, slow hydrolysis has
been observed, with the times to achieve greater than 90 per cent breakdown being
approximately 10 weeks at pH 1 and greater than 12 weeks at pH 9 (Harada et al.,
1996). Rapid chemical hydrolysis occurs only under conditions that are unlikely to be
attained outside the laboratory, e.g. 6M HCl at high temperature.
Microcystins can be oxidised by ozone and other strong oxidising agents, and degraded
by intense ultra violet (UV) light. These processes have relevance for water treatment
and are discussed in Chapter 9, although they are unlikely to contribute to degradation
occurring in the natural environment.
In full sunlight, microcystins undergo slow photochemical breakdown and isomerisation,

with the reaction rate being enhanced by the presence of water-soluble cell pigments,
presumably phycobiliproteins (Tsuji et al., 1993). In the presence of pigments the
photochemical breakdown of microcystin in full sunlight can take as little as two weeks
for greater than 90 per cent breakdown, or longer than six weeks, depending on the
concentration of pigment (and presumably toxin, although this has not been tested). A
more rapid breakdown under sunlight has been reported in the presence of humic
substances (which can act as photosensitisers) in field concentrations ranging from 2-16
mg l-1 dissolved organic carbon (DOC). Approximately 40 per cent of the microcystins
was degraded per day under summer conditions of insolation (Welker and Steinberg,
1998). In deeper or muddy waters, the rate of breakdown is likely to be considerably
slower.
Anatoxins
Anatoxin-a is relatively stable in the dark, but in pure solution in the absence of pigments
it undergoes rapid photochemical degradation in sunlight. Breakdown is further
accelerated by alkaline conditions (Stevens and Krieger, 1991). The half-life for
photochemical breakdown is 1-2 hours. Under normal day and night light conditions at
pH 8 or pH 10, and at low initial concentrations (10 µg l-1), the half-life for anatoxin-a
breakdown was found to be approximately 14 days (Smith and Sutton, 1993). Anatoxin-
a(S) decomposes rapidly in basic solutions but is relatively stable under neutral and
acidic conditions (Matsunaga et al., 1989).
Saxitoxins
In the dark at room temperature, saxitoxins undergo a series of slow chemical hydrolysis
reactions. The C-toxins lose the N-sulphocarbamoyl group to form decarbamoyl
gonyautoxins (dc-GTXs); while the dc-GTXs, GTXs and STXs slowly degrade to, as yet
unidentified, non-toxic products. The half-lives for the breakdown reactions are in the
order of 1-10 weeks, with more than three months often being required for greater than
90 per cent breakdown (Jones and Negri, 1997). Because dc-GTXs are much more toxic
than C-toxins (by a factor of 10-100), a solution or water body containing a natural
mixture of C-toxins and GTXs, for example from the lysis of an Australian bloom of
Anabaena circinalis, will actually increase in toxicity over a period of up to three weeks,
before toxicity begins to abate during the succeeding 2-3 months. Boiling an extract of
Anabaena with predominant C-toxins may also substantially increase toxicity. Similar
transformation reactions occur in living cells as they age in culture or in a natural bloom
(Negri et al., 1997). No detailed studies have been carried out on saxitoxin breakdown in
sunlight, either with or without pigments.
Cylindrospermopsins
Cylindrospermopsin is relatively stable in the dark, with slow breakdown occurring at

elevated temperature (50 °C) (Chiswell et al., 1999). In sunlight and in the presence of
cell pigments, breakdown occurs quite rapidly being more than 90 per cent complete
within 2-3 days (Chiswell et al., 1999). Pure cylindrospermopsin is relatively stable in
sunlight.
3.4.3 Removal on natural sediments and soils
Microcystins appear to be retained only weakly on natural suspended solids in rivers and
reservoirs; usually no more than 20 per cent of the total microcystin concentration is
adsorbed. In a laboratory experiment, some of the dissolved anatoxin-a and microcystins
were reported by Rapala et al. (1993) to be adsorbed on lake sediments. Percolation
through clay soils may provide some cyanotoxin removal, but this will depend greatly on
the type of clay, surface charge, cation concentration of the water, etc. Cyanobacterial
cells and microcystins were retained in soil columns, but less efficiently in sediment
columns, in laboratory experiments simulating the fate of cyanobacterial toxins in
artificial recharge of groundwater and bank filtration (Lahti el al., 1996). No data are
available for other cyanobacterial toxins, but some removal may be expected, again
depending on the chemical conditions of soil and water.
Sedimentation of living cells without lysis, for example through grazing by zooplankton
and sinking of faecal pellets, may lead to accumulation and persistence of toxin material
in sediments, although this process has received little scientific attention. As discussed
in more detail below, microcystins retained in intact cells may persist for several months.
Cells deposited in sediments may be subject to fairly rapid breakdown by sediment
bacteria and protozoa, with the resultant release of toxins.
3.4.4 Biodegradation
Microcystins
In spite of their chemical stability and resistance to eucaryotic and many bacterial
peptidases, microcystins are susceptible to breakdown by aquatic bacteria found
naturally in rivers and reservoirs. These bacteria appear to be reasonably common and
widespread. Degradative bacteria have been found in sewage effluent (Lam et al., 1995),
lake water (Jones et al., 1994; Cousins et al., 1996; Lahti et al., 1997a), lake sediment
(Rapala et al., 1994; Lahti et al., 1997a) and river water (Jones et al., 1994).
Nonetheless, one Finnish study showed a complete lack of degradation of microcystin
over a three-month period by an inoculum taken in winter from the Vantaanjoki River
(Kiviranta et al., 1991). There is usually an initial lag phase with little loss of microcystin
and this period can be as short as two days or more than three weeks, depending on the
water body, climatic conditions, the concentration of dissolved microcystin and in some
cases, although not all, the previous bloom history of a lake (Jones et al., 1994; Rapala
et al., 1994; Lahti et al., 1997b). Once the biodegradation process commences, removal
of microcystin can be more than 90 per cent complete within 2-10 days. This may vary
depending on the water body, initial microcystin concentration and water temperature
(Jones et al., 1994; Lahti et al., 1997b).
Jones et al. (1994) isolated a species of aquatic Sphingomonas that initiated ring-
opening of microcystin-LR to produce linear (acyclo-)microcystin-LR as a transient
intermediate (Bourne et al., 1996). This compound was nearly 200 times less toxic than
the parent toxin. The products of complete bacterial degradation were non-toxic to mice
at doses up to 500 µg kg-1 (compared with an LD50 for microcystin-LR of about 60 µg kg-1).
The same bacterium, however, did not degrade the closely related cyclic pentapeptide
nodularin. In a strain of Pseudomonas aeruginosa from a Japanese lake, microcystin
degradation appeared to proceed by attack on the Adda side chain of microcystin
(Takenaka and Watanabe, 1997). Several bacteria were isolated from lake water and
sediment in Finland capable of degradation of microcystins and some strains also
degraded nodularin. One strain was identified as a Sphingomonas sp. and two of the
strains belonged to the beta-subgroup of Proteobacteria, although the genera remains to
be determined (Lahti et al., 1997a).
Other cyanobacterial toxins
Little work has been undertaken on the biodegradation of anatoxins, saxitoxins or

cylindrospermopsin. Anatoxin-a may be readily degraded by bacteria associated with
cyanobacterial filaments. Laboratory studies using non-axenic strains of cyanobacteria
found low concentrations of dissolved anatoxin-a in the culture medium (Kiviranta et al.,
1991; Rapala et al., 1993) whereas high concentrations of anatoxin-a were found in the
medium of a continuous culture using an axenic strain (free of contaminating bacteria) of
the same species (Rapala and Sivonen, 1998). A Pseudomonas sp. strain able to
degrade anatoxin-a at a rate of 6-10 µg ml-1 per three days was isolated by Kiviranta et al.
(1991). In the presence of lake sediment and natural bacteria, the half-life for breakdown
of anatoxin-a in the laboratory was about five days (Smith and Sutton, 1993). In a recent
study by Jones and Negri (1997) no bacterially-mediated degradation of saxitoxins from
Anabaena circinalis was observed in a range of surface water samples.
3.4.5 Bioaccumulation
Microcystins bioaccumulate in common aquatic vertebrates and invertebrates, including

fish (Carbis et al., 1997; Beattie et al., 1998), mussels (Eriksson et al., 1989; Falconer et
al., 1992; Prepas et al., 1997; Watanabe et al., 1997) and zooplankton (Watanabe et al.,
1992). In mussels, the highest microcystin concentrations are found in the
hepatopancreas, and in vertebrates they are found in the liver. Williams et al., (1997)
have shown covalent binding and accumulation of microcystin-LR in salmon liver and
crab larvae. Whether the levels of microcystin accumulation are sufficient to pose a risk
to humans is uncertain, and will depend on levels of consumption and the severity of
toxic blooms in the area where fish or shellfish are caught or collected. Common advice
given by water authorities is that the viscera of the fish should not be eaten, but caution
should be taken in all cases where major toxic blooms occur.
Saxitoxins from marine "red tide" dinoflagellates are well known for their propensity to
bioaccumulate in marine vertebrates and invertebrates, often with disastrous
consequences for animals and humans that consume them. Similarly, saxitoxins from
the freshwater cyanobacterium Anabaena circinalis may bioaccumulate in an Australian
species of freshwater mussel to concentrations exceeding international guidelines
(Shumway et al., 1995) during as little as seven days exposure to a cell density of
100,000 cells per ml of a toxigenic strain (Negri and Jones, 1995). This cell density is
commonly encountered in natural blooms of this species.
3.4.6 Persistence and stability in cells
Culture studies indicate that microcystins and nodularin degrade only very slowly (time
scale of weeks), if at all, whilst contained within living cells (Sivonen, 1990b; Lehtimäki et
al., 1994, 1997; Rapala et al., 1997; Orr and Jones, 1998). Similarly, scums of
Microcystis aeruginosa that dry on the shores of lakes may contain high concentrations
of microcystin for several months (Jones et al., 1995). These toxins are released back
into the water body when re-immersed. Thus there is the potential for significant
localised concentrations of dissolved microcystin even in the absence of living cells or a
recently collapsed bloom.
In a lake study carried out over two summer - autumn periods, Lahti et al. (1997b) found
that dissolved microcystin was more persistent than particulate toxin, with 30 and 15
days respectively required for 90 per cent degradation to occur.
3.5 Impact on aquatic biota
Direct cyanobacterial poisoning of animals can occur by two routes: through
consumption of cyanobacterial cells from the water, or indirectly through consumption of
other animals that have themselves fed on cyanobacteria and accumulated cyanotoxins.
As was outlined in section 3.4.5, cyanotoxins are known to bioaccumulate in common
aquatic vertebrates and invertebrates, including fish, mussels and zooplankton.
Consequently, there is considerable potential for toxic effects to be magnified in aquatic
food chains. Such toxicity biomagnification is well known for anthropogenic pollutants
such as heavy metals and pesticides. There is no reason to suspect that the situation
would be any different with natural cyanotoxins.
It is difficult to ascribe the deaths of natural populations of aquatic animals, especially

fish, unequivocally to cyanotoxin poisoning. One of the main reasons for this is because
the collapse of a large cyanobacterial bloom can lead to very low concentrations of
oxygen in the water column as a consequence of bacterial metabolism; consequent fish
deaths may be due to the anoxia. The best evidence for the potential for toxic effects on
aquatic organisms comes from controlled laboratory trials with exposure of animals to
toxic cyanobacteria or cell-free solutions of cyanotoxins.
3.5.1 Effects on aquatic bacteria
The influence of cyanobacterial toxins on bacteria is not fully understood and the
scientific literature gives a number of contradictory statements. According to some
authors neither an extract of Microcystis aeruginosa nor pure microcystin-LR have a
biocidal effect on Bacillus subtilis, Staphylococcus aureus, Escherichia coli or
Pseudomonas hydrophila (Foxall and Sasner, 1988). However, these limited tests
should not be seen as general indicators of the potential impacts of cyanotoxins on
aquatic bacteria. The majority of aquatic bacteria are yet to be cultured, and studies with
common mammalian pathogens or "laboratory" bacteria should not be taken as all
encompassing. It is quite possible that cyanotoxins impact on some species of aquatic
bacteria and not others. Certainly, microcystins are not toxic to all bacteria because
several species are known to degrade quite high concentrations of these toxins (see
section 3.4.4). It is even possible that the slow release of cyanotoxins from the cell
surface or from senescent cells may stimulate associations of particular bacterial types
(see section 3.4.4) which may even act as symbionts.
Attempts have been made to use bacterial toxicity tests (based on inhibition of bacterial
phosphorescence) to screen for the presence of cyanotoxins, especially microcystins.
However it appears that the inhibition of bacterial phosphorescence is not related to the
commonly known cyanotoxins. It has been suggested that the negative effect may be
related to the presence of unidentified LPS endotoxins in the cell wall of the
cyanobacterial cells (see sections 3.1 and 13.3).
3.5.2 Effects on zooplankton
Evidence of the potential effects of cyanotoxins on zooplankton from numerous studies,

mostly in laboratory situations, is complex and inconsistent. The vast majority of
published studies has been based on mouse bioassay data describing cyanobacterial
toxicity, with only a few more recent studies having used analytical methods such as
HPLC (see section 13.4) to quantify individual toxins. Overall, it appears that
cyanobacteria may exhibit a deleterious effect on zooplankton, but the effect is highly
variable between genera and species, and even between clones of individual
zooplankton species. One of the main questions yet to be resolved is whether the
observed inhibitory effects are due to the putative poor nutritional value of cyanobacteria,
to the known cyanotoxins, or to other unidentified compounds. There is evidence in the
literature to support all three effects as being significant, at least with particular species
under experimental growth conditions. A major difference in study design is whether
organisms are exposed to cyanotoxins dissolved in water, or fed with toxic
cyanobacteria. The latter is likely to lead to a substantially higher dose. Furthermore,
Jungmann and Benndorf (1994) reported that exposure of Daphnia to dissolved
microcystins showed effects only at concentrations several orders of magnitude above
those found in field samples. They did, however, observe toxicity to Daphnia by
unidentified metabolites other than microcystins from Microcystis.
There is dramatic variation among zooplankton species in their response to toxic (and
even non-toxic) cyanobacteria. For example, DeMott et al. (1991) showed that the four
species of zooplankton differed in their sensitivity to hepatotoxins by almost two orders
of magnitude, but toxicity was observed only at very high concentrations that are
scarcely encountered in natural water bodies (48 h LC50 ranging from 450 to 21,400 µg
of microcystin per litre). Snell (1980) found that there was a genotype-dependent
response of the rotifer Asplanchna girodi to toxic Anabaena flos-aquae and Lyngbya sp.
Hietala et al. (1997) observed a variation in susceptibility of more than three orders of
magnitude in the acute toxicity of Microcystis aeruginosa to 10 clones of Daphnia pulex.
Both DeMott et al. (1991) and Laurén-Määttä et al. (1997) suggested that clone and
species differences between zooplankton susceptibilities to toxic cyanobacteria may
lead to selection pressures in favour of resistant strains or species in water bodies where
toxic cyanobacteria occur frequently.
Benndorf and Henning (1989) found that the toxicity of a field population of Microcystis
was increased by the feeding activity of Daphnia galeata over a period of a few months.
A possible explanation for this phenomenon is offered by DeMott et al. (1991) who
demonstrated that a number of zooplankton species will avoid grazing on toxic
cyanobacteria, but continue to graze on non-toxic species. Similar results have also
been shown for grazing by the phytoplanktivorous fish Tilapia and silver carp. Thus,
grazing pressure from zooplankton and some fish may lead to the selective enrichment
of toxic cyanobacterial strains over time.
It is likely that under natural conditions in water bodies, certain species and strains of
zooplankton may be affected by cyanotoxins, whereas others will be unaffected. As such,
cyanotoxins may influence the zooplankton community structure, especially during times
when cyanobacteria are dominant within the phytoplankton.
3.5.3 Effects on fish
If fish are dosed with cyanotoxins by i.p. injections or by force-feeding, they develop
similar symptoms of intoxication as laboratory mammals. The question relevant for field
exposure is whether cyanotoxins enter healthy fish. For example, Tencalla et al. (1994)
showed that gastrointestinal uptake by gavage (force-feeding) caused massive hepatic
necrosis followed by fish deaths, whereas immersion of adults and juveniles in
contaminated water did not cause toxic effects. Other reported evidence suggests that
immersion in toxic cyanobacteria or cyanotoxins may be harmful to fish. Differences in
sensitivity may be pronounced between species: goldfish were found to be nearly 30
times less susceptible to i.p. microcystin than mice (Sugaya et al., 1990). Release of
toxic compounds from mass developments of cyanobacteria was considered to be the
cause of fish kills by Penaloza et al. (1990). Histopathological investigations offish
deaths during cyanobacterial blooms in the UK, indicated that the cause of death was
mostly due to damage of the gills, digestive tract and liver (Rodger et al., 1994). The gill
damage was probably caused by the high pH induced by cyanobacterial photosynthesis
activity prior to the bloom collapse, together with the higher level of ammonia arising
from the decomposition of the cyanobacteria. However, gill damage may have enhanced
microcystin uptake and thus led to liver necrosis. Damage to gills by dissolved
microcystin-LR has been shown experimentally in Tilapia and trout (Garcia, 1989; Gaete
et al., 1994; Bury et al., 1996).
Other pathological symptoms ascribed to toxic cyanobacterial blooms include damage to

the liver, heart, kidney, gills, skin and spleen (Garcia, 1989; Råbergh et al., 1991).
Garcia (1989) and Rodger et al. (1994) carried out experiments on trout, while Råbergh
et al. (1991) experimented on carp. The latter study highlighted degenerative changes in
kidney tubules and glomeruli. The effect of microcystins on European carp, Cyprinus
carpio, under natural field conditions in Australia has been described by Carbis et al.
(1997) as atrophy of hepatocytes, gills with pinpoint necrosis, epithelial ballooning,
folded lamellar tips, exfoliation of the lamellar epithelium, elevated asparate
aminotransferase activity and serum bilirubin concentrations. Laboratory studies indicate
that dissolved microcystins may affect fish embryos (Oberemm et al., 1997) and
behaviour of fish (Baganz et al., 1998).
The most definitive effect of microcystin on fish concerns Atlantic Salmon reared in net
pens in coastal waters of British Columbia and Washington State, USA. As yet
unidentified microcystin-producing organisms produce a progressive degeneration of the
liver in salmon smolts placed into open-water net pens (Anderson et al., 1993). The
disease, referred to as Net Pen Liver Disease (NPLD), has resulted in significant
economic losses for the mariculture industry.
3.6 References
Anderson, D.M. 1994 Red tides. Scientific American, August, 52-58.
Anderson, R.J., Luu, H.A., Chen, D.Z.X., Holmes, C.F.B., Kent, M., LeBlanc, M., Taylor,
F.J.R. and Williams, D.E. 1993 Chemical and biological evidence links microcystins to
salmon "Netpen liver Disease". Toxicon, 31, 1315-1323.
Annila, A., Lehtimäki, J., Manila, K., Eriksson, J.E., Sivonen, K., Rantala, T.T. and
Drakenberg, T. 1996 Solution structure of nodularin an inhibitor of serine/threonine
specific protein phosphatases. J. Bio. Chem., 271(28), 16695-16702.
Arment, A.R. and Carmichael, W.W. 1996 Evidence that microcystin is a thio-template
product. J. Phycol., 32, 591-597.
Azavedo, S.M.F.O., Evans, W.R., Carmichael, W.W. and Namikoshi, M. 1994 First
report of microcystins from a Brazilian isolate of the cyanobacterium Microcystis
aeruginosa. J. App. Phycol., 6, 261-265.
Baganz, D., Staaks, G., Steinberg, C. 1998 Impact of the cyanobacteria toxin,
microcystin-LR, on the behavior of zebrafish, Danio rerio. Wat. Res., 32, 948-952.
Baker, P.D. and Humpage, A.R. 1994 Toxicity associated with commonly occurring
cyanobacteria in surface waters of the Murray-Darling basin, Australia. Aust. J. Mar.
Freshwat. Res., 45, 773-786.
Banker, P.D., Carmeli, S., Hadas, O., Teltsch, B., Porat, R. and Sukenik, A. 1997
Identification of cylindrospermopsin in Aphanizomenon ovalisporum (Cyanophyceae)
isolated from Lake Kinneret, Israel. J. Phycol., 33, 613-616.
Barford, D. and Keller, J. 1994 Co-crystallization of the catalytic subunit of the

serine/threonine specific protein phosphatase 1 from human in complex with
microcystin-LR. J Molecular Biol., 235, 763-766.
Bateman, K. P., Thibault, P., Douglas, D. J. and White, R. L. 1995 Mass spectral
analyses of microcystins from toxic cyanobacteria using on-line chromatographic and
electrophoretic separations. J. Chromatog. A, 712, 253-268.
Beattie, K.A, Kaya, K., Sano, T. and Codd, G.A. 1998 Three dehydrobutyrine (Dhb)-
containing microcystins from the cyanobacterium Nostoc sp. Phytochemistry, 47(7),
1289-1292.
Benndorf, J. and Henning, M. 1989 Daphnia and toxic blooms of Microcystis aeruginosa
in Bautzen Reservoir. Int. Rev. ges. Hydrobiol., 74, 233-248.
Berg, K. Skulberg, O.M. and Skulberg, R. 1987 Effects of decaying toxic blue-green
algae on water quality - a laboratory study. Arch. Hydrobiol., 108, 549-563.
Berg, K., Skulberg, O.M., Skulberg, R., Underdal, B. and Willén, T. 1986 Observations of
toxic blue-green algae (Cyanobacteria) in some Scandinavian lakes. Acta. Vet. Scand.,
27, 440-452.
Blackburn, S.I., Bolch, C.J., Jones, G.J., Negri, A.P. and Orr, P.T. 1997 Cyanobacterial
blooms: why are they toxic? In: J.R.D. Davis [Ed.] Managing Algal Blooms: Outcomes
from the CSIRO Blue-Green Algal Research Program. J.R.D. Davis [Ed.] CSIRO Land
and Water, Canberra, 67-77.
Blackburn, S., McCausland, M.A., Bolch, C.J.S., Newman, S.J. and Jones, G.J. 1996
Effect of salinity on growth and toxin production in cultures of the bloom-forming
cyanobacterium Nodularia spumigena from Australian waters. Phycologia, 35, 511-522.
Boland, M.P., Smillie, M.A., Chen, D.Z.X. and Holmes, C.F.B. 1993 A unified bioscreen
for the detection of diarrhetic shellfish toxins and microcystins in marine and freshwater
environments. Toxicon, 31, 1393-1405.
Bolch, C.J.S., Blackburn, S.I., Jones, G.J., Orr, P.T. and Grewe, P.M. 1997 Plasmid
content and distribution in the toxic cyanobacterial genus Microcystis Kutzing ex
Lemmermann (Cyanobacteria: Chroococcales). Phycologia, 36, 6-11.
Botes, D.P., Tuinman, A.A., Wessels, P.L., Viljoen, C.C. and Kruger, H. 1984 The
structure of cyanoginosin-LA, a cyclic heptapeptide toxin from the cyanobacterium
Microcystis aeruginosa. J. Chem. Soc., Perkin Transactions, I, 2311-2318.
Botes, D.P., Wessels, P.L., Kruger, H., Runnegar, M.T.C., Santikarn, S., Smith, R.J.,
Barna, J.C.J. and Williams, D.H. 1985 Structural studies on cyanoginosins-LR, -YR, -YA,
and -YM, peptide toxins from Microcystis aeruginosa. J. Chem. Soc., Perkin
Transactions, I, 2747-2748.
Bourne, D., Jones, G.J. Blakeley, R.L., Jones, A., Negri, A.P. and Riddles, P. 1996
Enzymatic pathway for the bacterial degradation of the cyanobacterial cyclic peptide
toxin microcystin -LR. Appl. Env. Microbiol., 62,4086-4094.
Bruno, M., Barbini, D.A., Pierdominici, E., Serse, A.P. and Ioppolo, A. 1994 Anatoxin-a
and previously unknown toxin in Anabaena planctonica from blooms found in Lake
Mulargia (Italy). Toxicon, 32, 369-373.
Bumke-Vogt, C. 1998 Anatoxin-a and neurotoxic cyanobacteria in German lakes and

reservoirs. Env. Toxicol. Water Qual., (In Press).
Bumke-Vogt, C., Mailahn, W., Retard, W. and Chorus, I. 1996 A highly sensitive
analytical method for the neurotoxin anatoxin-a, using GC-ECD, and first application to
laboratory cultures. Phycologia, 35(6, Supplement), 51-56.
Bury, N.R., Flik, G., Eddy, F.B. and Codd, G.A 1996 The effects of cyanobacteria and
the cyanobacterial toxin microcystin-LR on Ca 2+ transport and Na+/K+ - ATPase in Tilapia
gills. J. Exp. Biol., 199,1319-1326.
Canter, H., Heaney, S.I. and Lund, J.W.G. 1990 The ecological significance of grazing
on planktonic populations of cyanobacteria by the ciliate Nassula. New Phytol., 114,
247-263.
Carbis, C.R., Rawlin, G.T., Grant, P., Mitchell, G.F., Anderson, J.W. and McCauley, I.
1997 A study of feral carp Cyprinus carpio L., exposed to Microcystis aeruginosa at Lake
Mokoan, Australia, and possible implication on fish health. J. Fish Diseases, 20, 81-91.
Carbis, C.R. Simons, J.A., Mitchell, G.F., Anderson, J.W. and McCauley, I. 1994 A
biochemical profile for predicting the chronic exposure of sheep to Microcystis
aeruginosa, an hepatotoxic species of blue-green alga. Res. Vet. Sci,. 57(3), 310-316.
Cardellina II, J.H., Marner, F.J and Moore, R.E. 1979 Seaweed dermatitis, structure of
lyngbyatoxin A. Science, 204, 193-195.
Carmichael, W.W. 1986 Algal toxins. In: J.A. Callow [Ed.] Advances in Botanical
Research, No. 12. Academic Press, London, 47-101.
Carmichael, W.W. 1992 A Status Report on Planktonic Cyanobacteria (Blue Green

Algae) and their Toxins. EPA/600/R-92/079, Environmental Monitoring Systems
Laboratory, Office of Research and Development, US Environmental Protection Agency,
Cincinnati, Ohio.
Carmichael, W.W. 1997 The Cyanotoxins. Adv. Bot. Res., 27, 211-256.
Carmichael, W.W., Beasley, V.R., Bunner, D.L., Eloff, J.N., Falconer, I., Gorham, P.,
Harada, K.I., Krishnamurthy, T., Yu, M.J., Moore, R.E., Rinehart, K., Runnegar, M.,
Skulberg, O.M. and Watanabe, M. 1988a Naming of cyclic heptapeptide toxins of
cyanobacteria (blue-green algae). Toxicon, 26, 971-973.
Carmichael, W.W., Biggs, D.F. and Gorham, P.R. 1975 Toxicology and pharmacological
action of Anabaena flos-aquae toxin. Science, 187, 542-544.
Carmichael, W.W., Eschedor, J.T, Patterson, G.M.L. and Moore R.E. 1988b Toxicity and
partial structure of a hepatotoxic peptide produced by the cyanobacterium Nodularia
spumigena Mertens. Appl. Env. Microbiol., 54, 2257-2263.
Carmichael, W.W., Evans, W.R., Yin, Q.Q., Bell, P. and Mocauklowski, E. 1997
Evidence for paralytic shellfish poisons in the freshwater cyanobacterium Lyngbya wollei
(Farlow ex Gomont) comb. nov. Appl. Environ. Microbiol., 63, 3104-3110.
Carmichael, W.W. and Gorham, P.R. 1981 The mosaic nature of toxic blooms of
cyanobacteria. In: W.W. Carmichael [Ed.] The Water Environment: Algal Toxins and
Health, Plenum Press, New York, 161-172.
Carmichael, W.W., Mahmood, N.A. and Hyde, E.G. 1990 Natural toxins from
cyanobacteria (blue-green algae). In: S. Hall and G. Strichartz [Eds] Marine Toxins,
Origin, Structure and Molecular Pharmacology, Vol. 418, American Chemical Society,
Washington D.C., 87-106.
Carmichael, W.W., Yu, M. J., He, Z.R, He, J.W. and Yu, J-L. 1988c Occurrence of the
toxic cyanobacterium (blue-green alga) Microcystis aeruginosa in Central China. Arch.
Hydrobiol., 114, 21-30.
Chiswell, R.K., Shaw, G.R., Eaglesham, G.K., Smith, M.J., Norris, R.L., Seawright, A.A.
and Moore, M.R. 1999 Stability of cylindrospermopsin, the toxin from the
cyanobacterium Cylindrospermopsis raciborskii. Effects of pH, temperature and sunlight
on decomposition. Environ. Toxicol. Water Qual. (In Press).
Choi, B.W., Namikoshi, M., Sun, F., Rinehart, K.L., Carmichael, W.W., Kaup, A.M.,
Evans, W.R. and Beasley, V.R. 1993 Isolation of linear peptides related to the
hepatotoxins nodularin and microcystins. Tetrahedron Letters, 34, 7881-7884.
Chorus, I. 1998 Die Blaualgen ("Cyanobakterien") - eine Berliner Besonderheit welcher

Zusammenhang besteht zwischen dem Massenhaften Auftreten von Blaualgen und der
Nutzung von Badegewässern? In: Nachhaltigkeit im Wasserwesen in der Mitte Europas.
Symposium 17-19 June, 1998, Berlin (In Press).
Christoffersen, K. 1996 Ecological implications of cyanobacterial toxins in aquatic food

webs. Phycologia, 35(6 Supplement), 42-50.
Codd, G.A. and Bell, S.G. 1996 The Occurrence and Fate of Blue-green Algae in Fresh-
waters. National Rivers Authority Research and Development Report 29, Her Majesty's
Stationery Office, London, 30 pp.
Codd, G.A. and Poon, G.K. 1988 Cyanobacterial toxins. In: J. G. Gallon, and L. J.
Rogers [Eds] Proc. Phytochem. Soc. Europe, 28, Oxford University Press, Oxford, 283-
296.
Codd, G.A., Edwards, C., Beattie, K.A., Lawton, L.A., Campell, D.L. and Bell, S.G. 1995
Toxins from blue-green algae. The Pringsheim Lecture. In: W. Wiessner, E. Schnepf and
R. C. Starr [Eds] Algae, Environment and Human Affairs. Biopress, Bristol, 1-17.
Costa, M.S. and Azevedo, S.M.F.O. 1994 Implantacao de um banco de culturas de

cianoficeas toxicas. Iheringia, 45, 69-74 (in Portuguese).
Cousins, I.T., Healing, D.J., James, H.A., Sutton, A. 1996 Biodegradation of microcystin-
LR by indigenous mixed bacterial populations. Wat. Res., 30, 481-485.
Craig, M., McCready, L., Luu, H.A., Smillie, M.A., Dubord, P. and Holmes, C.F.B. 1993
Identification and characterization of hydrophobic microcystins in Canadian freshwater
cyanobacteria. Toxicon, 31, 1541-1549.
Cremer, J. and Henning, K. 1991 Application of reversed-phase medium-pressure liquid

chromatography to the isolation, separation and amino acid analysis of two closely
related peptide toxins of the cyanobacterium Microcystis aeruginosa strain PCC 7806. J.
Chromatog., 587, 71-80.
Davidson, F.F. 1959 Poisoning of wild and domestic animals by a toxic waterbloom of
Nostoc rivulare Kuetz. J. Am. Water Works Ass., 51,1277-1287.
DeMott, W.R. Zhang, Q. Carmichael, W.W. 1991 Effects of toxic cyanobacteria and
purified toxins on the survival and feeding of a copepod and three species of Daphnia.
Limnol. Oceanog., 36(7), 1346-1357.
de Silva, E.D., Williams, D.E., Andersen, R.J., Klix, H., Holmes, C.F.B. and Alien, T.M.
1992 Motuporin, a potent protein phosphatase inhibitor isolated from the Papua New
Guinea sponge Theonella swinhoei Gray. Tetrahedron Letters, 33, 1561-1564.
Devlin, J.P., Edwards, O.E., Gorham, P.R., Hunter, M.R., Pike, R.K. and Stavric, B. 1977
Anatoxin-a, a toxic alkaloid from Anabaena flos-aquae NCR-44h. Cana. J. of Chem., 55,
1367-1371.
Dittman, E., Neilan, B.A., Erhard, M., v. Döhren, H. and Börner, T. 1997 Insertional
mutagenesis of a peptide synthetase gene which is responsible for hepatotoxin
production in the cyanobacterium Microcystis aeruginosa PCC 7806. Mol. Microbiol., 26,
779-787.
Edwards, C., Beattie, K.A., Scrimgeour, C.M. and Codd, G.A. 1992 Identification of
anatoxin-a in benthic cyanobacteria (blue-green algae) and in associated dog poisonings
at Loch Insh, Scotland. Toxicon, 30, 1165-1175.
Ekman-Ekebom, M., Kauppi, M., Sivonen, K., Niemi, M. and Lepistö, L. 1992 Toxic
cyanobacteria in some Finnish lakes. Environ. Toxicol. Water Qual., 7, 201-213.
Elleman, T.C., Falconer, I.R., Jackson, A.R.B., and Runnegar, M.T. 1978 Isolation,
characterization and pathology of the toxin from a Microcystis aeruginosa (=Anacystis
cyanea) bloom. Aust. J. Biol. Sci., 31, 209-218.
Endean, R., Monks, S.A., Griffith, J.K. and Llewellyn, L.E. 1993 Apparent relationship
between toxins elaborated by the cyanobacterium Trichodesmium erythraeum and those
present in the flesh of the narrow-barred Spanish mackerel Scomberomorus
commersoni. Toxicon, 31, 1155-1165.
Entzeroth, M., Blackman, A.J., Mynderse, J.S. and Moore, R.E. 1985 Structures and
stereochemistries of oscillatoxin B, 31-noroscillatoxin B, oscillatoxin D, and 30-
methyloscillatoxin D. J. Organic Chem., 50, 1255-1259.
Eriksson, J.E., Meriluoto, J.A., Lindholm, T. 1989 Accumulation of peptide toxin from the
cyanobacterium Oscillatoria agardhii in the freshwater mussel Anadonta cygnea.
Hydrobiologia, 183, 211-216.
Falconer, I.R., Choice, A. and Hosja, W. 1992 Toxicity of the edible mussel (Mytilus
edulis) growing naturally in an estuary during a water-bloom of the blue-green alga
Nodularia spumigena J. Environ. Toxicol. Water Qual., 7,119-123.
Fastner, J. 1994 First evidence on the occurrence of microcystin-LR in Berlin and

Brandenburg lakes. In: G. A. Codd, T. M. Jefferies, C. W. Keevil and E. Potter [Eds]
Detection Methods for Cyanobacterial Toxins. The Royal Society of Chemistry,
Cambridge, 149-151.
Fastner, J. 1998 Microcystins in German fresh-water bodies. Environ. Toxicol. Water
Qual., (In Press).
Fastner, J., Neumann, U. and Erhard, M. 1998 Patterns of different microcystins in field
samples dominated by different species of cyanobacteria. In: I. Chorus, [Ed.]
Cyanotoxins - Occurrence, Effects, Controlling Factors. Springer Verlag, (In Press).
Foxall, T.L. and Sasner J.J. 1988 Effect of a hepatic toxin from the cyanophyte
Microcystis aeruginosa. In: W.W. Carmichael [Ed.] The Water Environment. Algal Toxins
and Health. Plenum Press, New York.
Francis, G. 1878 Poisonous Australian lake. Nature, 18, 11-12.
Fujiki, H., Suganuma, M., Suguri, H., Yoshizawa, S., Takagi, K., Nakayasu, M., Ojika, M.,
Yamada, K., Yasumoto, T., Moore, R.E. and Sugimura, T. 1990 New tumor promoters
from marine natural products. In: S. Hall and G. Strichartz [Eds] Marine Toxins, Origin,
Structure and Molecular Pharmacology, Vol. 418, American Chemical Society,
Washington D.C., 232-240.
Fulton, R.S. 1988 Resistence to blue-green algal toxin by Bosmina longirostris. J.

Plankton Res., 10(4), 771-778.
Gaete, V., Canelo, E., Lagos, N., and Zambrano, F. 1994 Inhibitory effects of Microcystis
aeruginosa toxin on ion pumps of the gill of freshwater fish. Toxicon., 82, 121-127.
Gallon, J.R., Chit, K.N. and Brown, E.G. 1990 Biosynthesis of the tropane-related
cyanobacterial toxin anatoxin-a: role of ornithine decarboxylase. Phytochemistry, 29,
1107-1 111.
Gallon J.R., Kittakoop, O. and Brown, E.G. 1994 Biosynthesis of anatoxin-a by

Anabaena flos-aquae: examination of primary enzymic steps. Phytochemistry, 35, 1195-
1203.
Garcia, G.O. 1989 Toxicity of the cyanobacterium Microcystis aeruginosa strain 7820 to
trout and tilapia: a clinical and histopathological study. MSc Thesis, University of Stirling,
UK.
Gathercole, P.S. and Thiel, P.G. 1987 Liquid chromatographic determination of the
cyanoginosins, toxins produced by the cyanobacterium Microcystis aeruginosa. J.
Chromatog., 408, 435-440.
Gentile, J.H. and Maloney, T.E. 1969 Toxicity and environmental requirements of a
strain of Aphanizomenon flos-aquae (L.) Ralfs. Can. J. Microbiol., 15,165-173.
Goldberg, J., Huang, H.B., Kwon, Y.G., Greengard, P., Nairn, A.C. and Kuriyan, J. 1995
Three-dimensional structure of the catalytic subunit of protein serine/threonine
phosphatase-1. Nature, 376, 745-753.
Gorham, P.R. 1962 Laboratory studies on the toxins produced by waterblooms of blue-
green algae. Am. J. Public Health, 52, 2100-2105.
Hahn, S.T. and Capra, M.F. 1992 The cyanobacterium Oscillatoria erythraea a potential
source of toxin in the ciguatera food-chain. Food Additives and Contaminants, 9, 351-
355.
Hammer, U.T. 1968 Toxic blue-green algae in Saskatchewan. Can. Vet. J., 9, 221-229.
Harada, K-I., Matsuura, K., Suzuki, M., Watanabe, M.F., Oishi, S., Dahlem, A.M.,
Beasley, V.R. and Carmichael, W.W. 1990b Isolation and characterization of the minor
components associated with microcystins-LR and -RR in the cyanobacterium (blue-
green algae). Toxicon, 28, 55-64.
Harada, K.I., Nagai, H., Kimura, Y., Suzuki, M., Park, H.D., Watanabe, M.F., Luukkainen,
R., Sivonen, K. and Carmichael, W.W. 1993 Liquid chromatography/mass spectrometric
detection of anatoxin-a, a neurotoxin from cyanobacteria. Tetrahedron, 49, 9251-9260.
Harada, K.I., Ogawa, K., Kimura, Y., Murata, H., Suzuki, M., Thorn, P.M., Evans, W.R.
and Carmichael, W.W. 199 la Microcystins from Anabaena flos-aquae NRC 525-17.
Chem. Res. Toxicol., 4, 535-540.
Harada, K.I., Ogawa, K., Matsuura, K., Murata, H., Suzuki, M., Watanabe, M.F., Itezono,
Y. and Nakayama, N. 1990a Structural determination of geometrical isomers of
microcystins-LR and -RR from cyanobacteria by two-dimensional NMR spectroscopic
techniques. Chemical Research in Toxicology. 3, 473-481.
Harada, K.-I., Ogawa, K., Matsuura, K., Nagai, H., Murata, H., Suzuki, M., Itezono, Y.,
Nakayma, N., Shirai, M. and Nakano, M. 1991b Isolation of two toxic heptapeptide
microcystins from an axenic strain of Microcystis aeruginosa, K-139. Toxicon, 29, 479-
489.
Harada, K.I., Ohtani, I., Iwamoto, K., Suzuki, M., Watanabe, M.F., Watanabe, M. and
Terao, K. 1994 Isolation of cylindrospermopsin from a cyanobacterium Umezakia natans
and its screening method. Toxicon, 32, 73-84.
Harada, K-I., Tsuji, K, Watanabe, M.F. 1996 Stability of microcystins from cyanobacteria
- III. Effect of pH and temperature. Phycologia, 35(6 Supplement), 83-88.
Hashimoto, Y., Kamiya, H., Yamazato, K. and Nozawa, K. 1976 Occurrence of a toxic
blue-green alga inducing skin dermatitis in Okinawa. In: A. Ohsaka, K. Hayashi and Y.
Sawai [Eds] Animal, Plant, and Microbial Toxins. Vol. 1, Plenum Publishing, New York,
333-338.
Hawkins, P. and Lampert, W. 1989 The effect of Daphnia body size on filtering rate
inhibition in the presence of a filamentous cyanobacterium. Limnol. Oceanog., 34(6),
1084-1089.
Hawkins, P.R., Chandrasena, N.R., Jones, G.J., Humpage, A.R. and Falconer, I.R. 1997
Isolation and toxicity of Cylindrospermopsis raciborskii from an ornamental lake. Toxicon,
35(3), 341-346.
Hawkins, P.R., Runnegar, M.T.C., Jackson, A.R.B. and Falconer, I.R. 1985 Severe
hepatotoxicity caused by the tropical cyanobacterium (blue-green alga)
Cylindrospermopsis raciborskii (Woloszynska) Seenaya and Subba Raju isolated from a
domestic water supply reservoir. Appl. Environ. Microbiol., 50, 1292-1295.
Hawser, S.P., Codd, G.A., Carpenter, E.J. and Capone, D.G. 1991 A neurotoxic factor
associated with the bloom-forming cyanobacterium Trichodesmium. Toxicon, 29, 277-
278.
Hemscheidt, T., Burgoyne, D.L. and Moore, R.E. 1995a Biosynthesis of anatoxin-.
(2S,4S)-4-hydroxyarginine as an intermediate. J. Chem. Soc., Chemical
Communications, 205-206.
Hemscheidt T., Rapala, J., Sivonen, K. and Skulberg, O.M. 1995b Biosynthesis of
anatoxin-a in Anabaena flos-aquae and homoanatoxin-a in Oscillatoria formosa. J.
Chem. Soc., Chemical Communications, 1361-1362.
Henning, M. and Kohl, J.G. 1981 Toxic blue-green algae water blooms found in some
lakes in the German Democratic Republic. Int. Rev. ges. Hydrobiol., 66, 553-561.
Henriksen, P. 1996a Microcystin profiles and contents in Danish populations of

cyanobacteria/blue-green algae as determined by HPLC, Phycologia, (Supplement),
35(6), 102-110.
Henriksen, P. 1996b Toxic cyanobacteria/blue-green algae in Danish fresh waters.

Thesis, Dept. of Phycology, University of Copenhagen.
Henriksen, P., Carmichael, W.W., An, J. and Moestrup, Ø. 1997 Detection of an

anatoxin-a(s)-like anticholinesterase in natural blooms and cultures of
cyanobacteria/blue-green algae from Danish lakes and in the stomach contents of
poisoned birds. Toxicon, 35, 901-913.
Henriksen, P., Kaas, H., Damsø, L. and Moestrup, Ø. 1995 Toxic cyanobacteria (blue-
green algae) in Danish fresh waters. Poster presented in 1st International Congress on
Toxic Cyanobacteria (Blue-green Algae), August, Rnne, Denmark, 20-24.
Henriksen, P., Sivonen, K., Kaas, H. and Moestrup, Ø. 1998 Toxic cyanobacteria/blue-
green algae in Danish fresh waters. J. Phycol. (In Press).
Hietala, J., Laurén-Määttä, C. and Walls, M. 1997 Sensitivity of Daphnia to toxic

cyanobacteria: effects of genotype and temperature. Freshwat. Biol., 37, 299-306.
Humpage, A.R., Rositano, J., Bretag, A.H., Brown, R., Baler, P.D., Nicholson, B.C. and
Steffensen, D.A. 1994. Paralytic shellfish poisons from Australian cyanobacterial blooms.
Aust. J. Mar. Freshwater Res., 45(5), 761-771.
Ikawa, M., Wegener, K., Foxall, T.L. and Sasner, J.J. 1982 Comparison of the toxins of
the blue-green alga Aphanizomenon flos-aquae with the Gonyaulax toxins. Toxicon. 20,
747-752.
Jackim, E. and Gentile, J. 1968 Toxins of a blue-green alga, similarity to saxitoxin.

Science, 162, 915-916.
Jacobs, D., Bass, R.A. and Neilan, B.A. 1995 Isolation and characterization of candidate
genes responsible for toxins biosynthesis in cyanobacteria. Abstract, Presented at the
Vth Cyanobacterial Molecular Biology Workshop, July, 1995, Asilomar, CA, USA, 98, 21-
25.
James, K.J., Sherlock, I.R. and Stack, M.A. 1997 Anatoxin-a in Irish freshwater and
cyanobacteria, determined using a new fluorimetric liquid chromatographic method.
Toxicon, 35(6), 963-971.
Jones, G.J. and Falconer, I.R. 1994 Factors affecting the production of toxins by
cyanobacteria. Final Grant Report to the Land and Water Resources Research and
Development Corporation, Canberra, Australia.
Jones, G.J. and Negri, A.P. 1997 Persistence and degradation of cyanobacterial
paralytic shellfish poisons (PSPs) in freshwaters. Wat. Res., 31, 525-533.
Jones, G.J. and Orr, P.T. 1994 Release and degradation of microcystin following
algicide treatment of a Microcystis aeruginosa bloom in a recreational lake, as
determined by HPLC and protein phosphatase inhibition assay. Wat. Res., 28, 871-876.
Jones, G.J., Blackburn, S.I. and Parker, N.S. 1994 A toxic bloom of Nodularia
spumigena Mertens in Orielton Lagoon, Tasmania. Australian J. Mar. Freshwat. Res., 45,
787-800. 1783
Jones, G.J., Falconer, I.F. and Wilkins, R.M. 1995 Persistence of cyclic peptide toxins in
dried cyanobacterial crusts from Lake Mokoan, Australia. Environ. Toxicol. Water Qual.,
10, 19-24.
Jungmann, D. 1992 Toxic compounds isolated from Microcystis PCC 7806 that are more
active against Daphnia than two microcystins. Limnol. Oceanog., 37, 1777-1783.
Jungmann, D. and Benndorf, J. 1994 Toxicity to Daphnia of a compound extracted from

laboratory and natural Microcystis spp., and the role of microcystins. Freshwat. Biol., 32,
13-20.
Kaya, K. and Watanabe, M.M. 1990 Microcystin composition of an axenic clonal strain of
Microcystis viridis and Microcystis viridis - containing waterblooms in Japanese
freshwaters. J. App. Phycol., 2, 173-178.
Keleti, G. and Sykora, J.L. 1982 Production and properties of cyanobacterial endotoxins.
Appl. Environ. Microbiol., 43, 104 - 109.
Kenefick, S.L., Hrudey, S.E., Peterson, H.G. and Prepas, E.E. 1992 Toxin release from
Microcystis aeruginosa after chemical treatment. Wat. Sci. Tech., 27, 433-440.
Kerr, T., Jones, G.J. and Negri, A. 1995 Hydroxy fatty acid analysis of
lipopolysaccharides in cyanobacteria. CSIRO Division of Water Resources, Australia,
Summer Studentship report, 73 pp.
Kiviranta, J., Sivonen, K., Luukkainen, R., Lahti K. and Niemelä, S. I. 1991 Production
and biodegradation of cyanobacterial toxins; a laboratory study, Arch. Hydrobiol., 121,
281-294.
Kononen, K., Sivonen, K. and Lehtimäki, J. 1993 Toxicity of phytoplankton blooms in the
Gulf of Finland and Gulf of Bothnia, Baltic Sea. In: T. J. Smayda and Y. Shimizu [Eds]
Toxic Phytoplankton Blooms in the Sea, Elsevier Scientific Publications, Amsterdam,
269-274.
Kotak, B.G., Kenefick, S.L., Fritz, D.L., Rousseaux, C.G., Prepas, E.E. and Hrudey, S.E.
1993 Occurrence and toxicological evaluation of cyanobacterial toxins in Alberta lakes
and farm dugouts. Wat. Res., 27, 5-506.
Kotak, B.G., Lam, A.K.Y., Prepas, E.E., Kenefick, S.L. and Hrudey, S.E. 1995 Variability
of the hepatotoxin microcystin-LR in hypereutrophic drinking water lakes. J. Phycol., 31,
248-263.
Krishnamyrthy, T., Szafraniec, L., Hunt, D.F., Shabanowitz, J., Yates, J.R., Hauer, C.R.,
Carmichael, W.W., Skulberg, O., Codd, G.A. and Missler, S. 1989 Structural
characterization of toxic cyclic peptides from blue-green algae by tandem mass
spectrometry. Proceedings of National Academy of Sciences USA., 86, 770-774.
Kusumi, T., Ooi, T., Watanabe, M.M., Takahashi, H. and Kakisawa, H. 1987
Cyanoviridin RR, a toxin from the cyanobacterium (blue-green alga) Microcystis viridis.
Tetrahedron Letters, 28, 4695-4698.
Lagos, N., Liberona, J.L., Andrinolo, D., Zagatto, P.A., Soares, R.M. and Azevedo,
S.M.F.Q. 1997 First evidence of paralytic shellfish toxins in the freshwater
cyanobacterium Cylindrospermopsis raciborskii isolated from Brazil. Abstract, VIII
International Conference on Harmful Algae, June 25-29, 1997, Vigo, Spain, 115.
Lahti, K., Kilponen, J., Kivimäki, A.L., Erkomaa, K. and Sivonen, K. 1996 Removal of
cyanobacteria and their hepatotoxins from raw water in soil and sediment columns. In:
A.L. Kivimäki and T. Suokko [Eds] Artificial Recharge of Groundwater. NHP Report No.
38, Helsinki, 187-195.
Lahti, K., Niemi, M. R., Rapala J. and Sivonen, K. 1997a Biodegradation of

cyanobacterial hepatotoxins - characterization of toxin degrading bacteria. Proceedings
of the VII International Conference on Harmful Algae.
Lahti, K., Rapala, J., Färdig, M., Niemelä, M. and Sivonen, K. 1997b Persistence of
cyanobacterial hepatotoxin, microcystin-LR, in particulate material and dissolved in lake
water. Wat. Res., 31(5), 1005-1012.
Lam, A.K.-Y., Fedorak, P.M. and Prepas, E.E. 1995 Biotransformation of the
cyanobacterial hepatotoxin microcystin-LR, as determined by HPLC and protein
phosphatase bioassay. Environ. Sci. Technol., 29(1), 242-246.
Lampert, W. 1981 Inhibitory and toxic effect of blue-green algae on Daphnia. Rev. Ges.
Hydrobiol., 66, 285-298.
Lanaras, T. and Cook, C.M. 1994 Toxin extraction from an Anabaenopsis milleri -
dominated bloom. Sci. Total Environ., 142, 163-169.
Lanaras, T., Tsitsamis, S., Chlichlia, C. and Cook, C.M. 1989 Toxic cyanobacteria in
Greek freshwaters. J. Applied Phycol., 1, 67-73.
Laurén-Määttä, C., Hietala, J. and Walls, M. 1997 Responses of Daphnia pulex

populations to toxic cyanobacteria. Freshwat. Biol., 37, 635-647.
Lawton, L.A., Beattie, K.A., Hawser, S.P., Campbell, D.L. and Codd, G.A. 1994
Evaluation of assay methods for the determination of cyanobacterial hepatotoxicity. In:
G.A Codd, T.M. Jeffries, C.W. Keevil, E. Potter [Eds] Detection Methods for
Cyanobacterial Toxins. Proceedings of the First International Symposium on Detection
Methods for Cyanobacterial Toxins, September 1993 The Royal Society of Chemistry,
Cambridge, 111-116.
Leeuwangh, P., Kappers, F.I., Dekker, M. and Koerselman, W. 1983 Toxicity of

cyanobacteria in Dutch lakes and reservoirs. Aquat. Toxicol., 4, 63-72.
Lehtimäki, J., Moisander, P., Sivonen, K. and Kononen, K. 1997 Growth, nitrogen
fixation, and nodularin production by two Baltic Sea cyanobacteria. Appl. Environ.
Microbiol., 63, 1647-1656.
Lehtimäki, J., Sivonen, K., Luukkainen, R. and Niemelä, S.I. 1994 The effects of
incubation time, temperature, light, salinity, and phosphorus on growth and hepatotoxin
production by Nodularia strains. Arch Hydrobiol., 130, 269-282.
Lindholm, T. and Meriluoto, J.A.O. 1991 Recurrent depth maxima of the hepatotoxic
cyanobacterium Oscillatoria agardhii. Can. J. Fish. Aquat. Sci., 48, 1629-1634.
Lukac, M. and Aegerter, R. 1993 Influence of trace metals on growth and toxin
production of Microcystis aeruginosa. Toxicon, 31, 293-305.
Luukkainen, R., Namikoshi, M., Sivonen, K., Rinehart, K.L. and Niemelä, S.I. 1994
Isolation and identification of 12 microcystins from four strains and two bloom samples of
Microcystis spp.: structure of a new hepatotoxin. Toxicon., 32, 133-139.
Luukkainen R., Sivonen, K., Namikoshi, M., Färdig, M., Rinehart, K.L. and Niemelä, S.I.
1993 Isolation and identification of eight microcystins from 13 Oscillatoria agardhii
strains: structure of a new microcystin. Appl. Environ. Microbiol., 59, 2204-2209.
Lyck, S., Gjølme, N. and Utkilen, H. 1996 Irom-starvation increases toxicity of
Microcystis aeruginosa CYA 22/1 (Chroococcales, Cyanophyceae). Phycologia, 35(6
Supplement), 120-124.
Mahmood, N.A. and Carmichael, W.W. 1986 Paralytic shellfish poisons produced by the
freshwater cyanobacterium Aphanizomenon flos-aquae NH-5. Toxicon, 24, 175-186.
Mahmood, N.A. and Carmichael, W.W. 1987 Anatoxin-a(S), an anticholinesterase from

the cyanobacterium Anabaena flos-aquae NCR-525-17. Toxicon, 25, 1221-1227.
Maršálek, B., LeDu, A., Turánek, J. and Machala, M. 1995 The selection of bioassays for
cyanotoxin detection. In: Proceedings 1st International Congress on Toxic
Cyanobacteria, 20-24 August 1995, Rønne, Denmark, 45.
Maršálek, B., Turánek J. and Hindák F. 1996 Toxic cyanobacteria in Czech and Slovak
republic. In: Proceedings 1st. European Phycological Congress, Cologne, August 11-18,
1996, Eur. J. Phycol. Suppl., 42.
Matsunaga, S., Moore, R.E., Niemczura, W.P. and Carmichael, W.W. 1989 Anatoxin-
a(s), a potent anticholinesterase from Anabaena flos-aquae. J. Amer. Chem. Soc., 111,
8021-8023.
Mattson, R. and Willén, T. 1986 Toxinbildande blågröna alger i svenska insjöar 1985
(Toxin producing blue-green algae in Swedish lakes in 1985). Naturvårdsverket rapport
3096 (2)/1986. Laboratoriet för miljökontroll, Uppsala, ISSN 0282-7298.26. pp. (In
Swedish, English summary).
Matveev, V., Matveeva, L., and Jones, G.J. 1994a Study of the ability of Daphnia
carinata King to control phytoplankton and resist cyanobacterial toxicity: implications for
biomanipulation in Australia. Aust. J. Mar. Freshwat. Res., 45, 889-904.
Matveev, V., Matveeva, L. and Jones, G.J. 1994b Phytoplankton stimulation by mosquito
fish in the presence of large Daphnia. Ver. Verein. fur Theor. Ang. Limnol., 25, 2193-
2197.
McDermott, C.M., Feola, R. and Plude, J. 1995 Detection of cyanobacterial toxins

(microcystins) in waters of northeastern Wisconsin by a new immunoassay technique.
Toxicon., 33, 1433-1442.
Meissner, K., Dittman, E. and Börner, T. 1996 Toxic and non-toxic strains of the
cyanobacterium Microcystis aeruginosa contain sequences homologous to peptide
synthetase genes. FEMS Microbiology Letters, 135, 295-303.
Meriluoto, J.A.O., Sandström, A., Eriksson, J.E., Remaud, G., Grey Craig, A. and
Chattopadhyaya, J. 1989 Structure and toxicity of a peptide hepatotoxin from the
cyanobacterium Oscillatoria agardhii. Toxicon, 27, 1021-1034.
Mez, K., Hanselmann, K., Naegeli, H. and Preisig, H.R. 1996 Protein phosphatase-
inhibiting activity in cyanobacteria from alpine lakes in Switzerland. Phycologia, 35(6
Moikeha, S.N. and Chu, G.W. 1971 Dermatitis-producing alga Lyngbya majuscula
Gomont in Hawaii. II. Biological properties of the toxic factor. J. Phycol., 7, 8-13.
Moore B.S., Ohtani, I., de Koning, C.B., Moore, R.E. and Carmichael, W.W. 1992
Biosynthesis of Anatoxin-a(s). Origin of the carbons. Tetrahedron Letters. 33, 6595-6598.
Moore, R.E., Corbett, T., Patterson, G.M.L. and Valeriote, F.A. 1996 The search for new
antitumor drugs from blue-green algae. Current Pharm. Design, 2, 317-330.
Moore, R.E., Lu Chen, J., Moore, B.S. and Patterson, G.M.L. 1991 Biosynthesis of
microcystin-LR. Origin of the carbons in the Adda and Masp units. J. Amer. Chem. Soc.,
113, 5083-5084.
Moore, R.E., Ohtani, I., Moore, B.S., de Koning, C.B., Yoshida, W.Y., Runnegar, M.T.
and Carmichael, W.W. 1993 Cyanobacterial toxins. Gazzetta Chimica Italiana, 123, 329-
336.
Mynderse, J.S., Moore, R.E., Kashiwagi, M. and Norton, T.R. 1977 Antileukemia activity
in the Oscillatoriaceae, isolation of debromoaplysiatoxin from Lyngbya. Science, 196,
538-540.
Nagata, S., Tsutsumi, T., Hasegawa, A., Yoshida, F., Ueno, Y. and Watanabe, M.F.
1997 Enzyme immunoassay for direct determination of microcystins in environmental
water. J. AOAC Int., 80, 408-417.
Namikoshi, M., and Rinehart, K.L. 1996 Bioactive compounds produced by

cyanobacteria. J. Ind. Microbiol., 17, 373-384.
Namikoshi, M., Choi, B.W., Sakai, R., Sun, F., Rinehart, K.L., Carmichael, W.W., Evans,
W.R., Cruz, P., Munro, M.H.G. and Blunt, J.W. 1994 New nodularins, A general method
for structure assignment. J. Org. Chem., 59, 2349-2357.
Namikoshi, M., Rinehart, K.L., Sakai, R., Sivonen, K. and Carmichael, W.W. 1990
Structures of three new cyclic hepatotoxins produced by the cyanobacterium (blue-green
alga) Nostoc sp. strain 152. J. Org. Chem., 55, 6135-6139.
Namikoshi, M., Rinehart, K.L., Sakai, R., Stotts, R.R., Dahlem, A.M., Beasley, V.R.,
Carmichael, W.W. and Evans, W.R. 1992 Identification of 12 hepatotoxins from a Homer
lake bloom of the cyanobacteria Microcystis aeruginosa, Microcystis viridis, Microcystis
wesenbergii; nine new microcystins. J. Org. Chem., 57, 866-872.
Namikoshi, M., Sivonen, K., Evans, W.R., Carmichael, W.W., Rouhiainen, L.,
Luukkainen, R. and Rinehart, K.L. 1992b Structures of three new homotyrosine-
containing microcystins and a new homophenylalanine variant from Anabaena sp. strain
66. Chem. Res. Toxicol., 5, 661-666.
Namikoshi, M., Sivonen, K., Evans, W.R., Carmichael, W.W., Sun, F., Rouhiainen, L.,
Luukkainen, R. and Rinehart, K.L. 1992c Two new L-serine variants of microcystins-LR
and -RR from Anabaena sp. strain 202 A1 and 202 A2. Toxicon, 30, 1457-1464.
Namikoshi, M., Sivonen, K., Evans, W.R., Sun, F., Carmichael, W.W. and Rinehart, K.L.
1992d Isolation and structures of microcystins from a cyanobacterial water bloom
(Finland). Toxicon, 30, 1473-1479.
Namikoshi, M., Sun, F., Choi, B.W., Rinehart, K.L., Carmichael, W.W., Evans, W.R. and
Beasley, V.R. 1995 Seven more microcystins from Homer lake cells: application of the
general method for structure assignment of peptides containing,-dehydroamino acid
unit(s). J. Org. Chem., 60, 3671-3679.
Namikoshi, M., Yuan, M., Sivonen, K., Carmichael, W.W., Rinehart, K.L., Rouhiainen, L.,
Sun, F., Brittain, S. and Otsuki, A. 1998 Seven new microcystins possessing two L-
glutamic acit units, isolated from Anabaena sp. strain 186. Chem. Res. Toxicol., 11, 143-
149.
Negri, A.P. and Jones, G.J. 1995 Bioaccumulation of paralytic shellfish poisoning (PSP)
toxins from the cyanobacterium Anabaena circinalis by the freshwater mussel Alathyria
condola. Toxicon, 33(5), 667-678.
Negri, A.P., Jones, G.J., Blackburn, S.I., Oshima, Y. and Onodera, H. 1997 Effect of
culture and bloom development and of sample storage on paralytic shellfish poisons in
the cyanobacterium Anabaena circinalis. J. Phycol., 33, 26-35.
Negri, A.P., Jones, G.J. and Hindmarsh, M. 1995 Sheep mortality associated with
Paralytic Shellfish poisoning toxins from the cyanobacterium Anabaena circinalis.
Toxicon, 33, 1321-1329.
Oberemm, A., Fastner, J. and Steinberg, C. 1997 Effects of microcystin-LR and

cyanobacterial crude extracts on embryo-larval development of zebrafish (Danio rerio).
Wat. Res., 31(11), 2918-2921.
Ohtani, I., Moore, R.E. and Runnegar, M.T.C. 1992 Cylindrospermopsin, a potent
hepatotoxin from the blue-green alga Cylindrospermopsis raciborskii. J. Amer. Chem.
Soc., 114, 7941-7942.
Olson, T.A. 1960. Water poisoning - a study of poisonous algae blooms in Minnesota.
Amer. J. Public Health, 50, 883-884.
Onodera, H., Oshima, Y., Henriksen, P. and Yasumoto, T. 1997a Confirmation of

anatoxin-a(s) in the cyanobacterium Anabaena lemmermannii as the cause of bird kills
in Danish lakes. Toxicon, 35, 1645-1648.
Onodera, H., Satake, M., Oshima, Y., Yasumoto, T. and Carmichael, W.W. 1997b New
saxitoxin analogues from the freshwater filamentous cyanobacterium Lyngbya wollei.
Natural Toxins (In Press).
Orr, P.T. and Jones, G.J. 1998 Relationship between microcystin production and cell
division rates in nitrogen-limited Microcystis aeruginosa cultures. Limnol. Oceanogr.,
43(8), (In Press).
Padisák, J. 1997 Cylindrospermopsis raciborskii (Woloszynska) Seenayya et Subba

Raju, an expanding, highly adaptive cyanobacterium: worldwide distribution and review
of its ecology. Arch. Hydrobiol., Suppl., 107, 563-593.
Painuly, P., Perez, R., Fukai, T. and Shimizu, Y. 1988 The structure of a cyclic peptide
toxin, cyanogenosin-RR from Microcystis aeruginosa. Tetrahedron Letters, 29, 11-14.
Park, H.-D., Watanabe, M.F., Harada, K.-I., Nagai, H., Suzuki, M., Watanabe, M. and
Hayashi, H. 1993a Hepatotoxin (microcystin) and neurotoxin (anatoxin-a) contained in
natural blooms and strains of cyanobacteria from Japanese freshwaters. Natural Toxins,
1, 353-360.
Park, H.-D., Watanabe, M.F., Harada, K.-I., Suzuki, M., Hayashi, H. and Okino, T. 1993b
Seasonal variations of Microcystis species and toxic heptapeptide microcystins in Lake
Suwa. Environ. Toxicol. Water Qual., 8, 425-435.
Penaloza, R., Rojas, M., Vila, I. and Zambrano, F. 1990 Toxicity of a soluble peptide
from Microcystis sp. to zooplankton and fish. Freshwat. Biol., 24(2), 223-240.
Prepas, E.E., Kotak, E.G., Campbell, L.M., Evans, J.C., Hrudey, S.E. and Holmes, C.F.B.
1997 Accumulation and elimination of cyanobacterial hepatotoxins by the freshwater
clam Anodonta grandis simpsoniana. Can. J. Fish. Aquat. Sci., 54, 41-46.
Prinsep, M.R., Caplan, F.R., Moore, R.E., Patterson, G.M.L., Honkanen, R.E. and
Boynton, A.L. 1992 Microcystin-LR from a blue-green alga belonging to the
Stignonematales. Phytochemistry, 31, 1247-1248.
Råbergh, C.M.I., Bylund, G., Eriksson, J.E. 1991 Histopathological effect of microcystin
LR a cyclic polypeptide from the cyanobacterium Microcystis aeruginosa on common
carp (Cyprinus carpio L.). Aquat. Tox., 20,131-146.
Rapala, J. and Sivonen, K. 1998 Assessment of environmental conditions that favour

hepatotoxic and neurotoxic Anabaena spp. strains in cultured under light-limitation at
different temperatures. Microbial. Ecol., 36, 181-192.
Rapala, J., Lahti, K., Sivonen, K. and Niemelä, S. I. 1994. Biodegradability and
adsorption on lake sediments of cyanobacterial hepatotoxins and anatoxin-a. Letters in
Applied Microbiol., 19, 423-428.
Rapala, J., Sivonen, K., Luukkainen, R. and Niemelä, S.I. 1993 Anatoxin-a concentration
in Anabaena and Aphanizomenon at different environmental conditions and comparison
of growth by toxic and non-toxic Anabaena strains, a laboratory study, J. App. Phycol., 5,
581-591.
Rapala, J., Sivonen, K., Lyra, C. and Niemelä, S.I. 1997 Variation of microcystins,
cyanobacterial hepatotoxins, in Anabaena spp. as a function of growth stimuli. App.
Environ. Microbiol., 64, 2206-2212.
Raziuddin, S., Siegelman, H.W. and Tornabene, T.G. 1983 Lipopolysaccharides of the
cyanobacterium Microcystis aeruginosa. Eur. J. Biochem., 137, 333-336.
Reinikainen, M., Kiviranta, J., Ulvi, V. and Niku-Paavola, M.L. 1995 Acute toxic effects of
a novel cyanobacterial toxin on the crustaceans Artemia salina and Daphnia pulex. Arch.
Hydrobiol., 133, 61-69.
Repavich, W.M., Sonzogni, W.C., Standridge, J.H., Wedepohl, R.E. and Meisner L.F.
1990 Cyanobacteria (blue-green algae) in Wisconsin waters: acute and chronic toxicity.
Wat. Res., 24, 225-231.
Ressom, R., Soong, F.S., Fitzgerald, J., Turczynowicz, L., El Saadi, O., Roder, D.,
Maynard, T. and Falconer, I.1994 Health Effects of Toxic Cyanobacteria (Blue-green
Algae). National Health and Medical Council, The Australian Government Publishing
Service, Canberra, Australia, 108 pp.
Rinehart, K.L., Harada, K.-I., Namikoshi, M., Chen, C., Harvis, C.A., Munro, M.H.G.,
Blunt, J.W., Mulligan, P.E., Beasley, V.R., Dahlem, A.M. and Carmichael, W.W. 1988
Nodularin, microcystin, and the configuration of Adda. J. Amer. Chem. Soc., 110, 8557-
8558.
Rinehart, K.L., Namikoshi, M. and Choi, B.W. 1994 Structure and biosynthesis of toxins
from blue-green algae (Cyanobacteria). J. App. Phycol., 6, 159-176.
Rodger, H.D., Turnbull, T., Edwards, C. and Codd, G.A. 1994 Cyanobacterial bloom
associated pathology in brown trout Salmo trutta L. in Loch Leven, Scotland. J. Fish Dis.,
17, 177-181.
Rouhiainen L., Buikema, W., Paulin, L., Fonstein, M., Sivonen, K. and Haselkorn, R.
1994 Peptide synthetase genes from hepatotoxin producing Cyanobacteria. Abstract In:
VIII International Symposium on Phototrophic Prokaryotes, 10-15, September, Urbino,
Italy, 26.
Rudolph-Böhner, S., Mierke, D.F. and Moroder, L. 1994 Molecular structure of

cyanobacterial tumour-promoting microcystins. FEBS Letters, 349, 319-323.
Schwabe W., Weihe, A., Börner, T., Henning, M. and Kohl, J.G. 1988 Plasmids in toxic
and nontoxic strains of the cyanobacterium Microcystis aeruginosa. Current Microbiology,
17, 133-137.
Scott, W.E. 1991 Occurrence and significance of toxic Cyanobacteria in Southern Africa.
Wat. Sci. Techno., 23, 175-180.
Shimizu, Y. 1986 Toxigenesis and biosynthesis of saxitoxin analogues. Pure Appl.

Chem., 58, 257-262.
Shimizu, Y., Norte, M., Hori, A., Genenah, A. and Kobayashi, M. 1984 Biosynthesis of
saxitoxin analogues: the unexpected pathway. J. Amer. Chem. Soc., 106, 6433-6434.
Shirai, M., Ohtake, A., Sano, T., Matsumoto, S., Sakamoto, T., Sato, A., Aida, T.,
Harada, K.I., Shimida, T., Suzuki, M. and Nakano, M. 1991 Toxicity and toxins of natural
blooms and isolated strains of Microcystis spp. (Cyanobacteria) and improved procedure
for purification of cultures. App. Environ. Microbiol., 57, 1241-1245.
Shumway, S.E., van Egmond, H.P., Hurst, J.W. and Bean, L.L. 1995 Management of
shellfish resources. In: G.M. Hallegraeff, D.M. Anderson, A.D. Cembella and H.O
Enevoldsen [Eds] Manual on Harmful Marine Microalgae. UNESCO Intergovernmental
Oceanographic Commission, Manuals and Guides 33, United Nations Educational,
Scientific and Cultural Organization, 433-457.
Sivonen, K. 1990a Toxic Cyanobacteria in Finnish Fresh Waters and the Baltic Sea.
Reports from Department of Microbiology, 39, University of Helsinki, Hakapaino, Helsinki,
87 pp.
Sivonen, K. 1990b Effects of light, temperature, nitrate, orthophosphate, and bacteria on

growth of and hepatotoxin production by Oscillatoria agardhii strains. App. Environ.
Microbiol., 56, 2658-2666.
Sivonen, K. 1996 Cyanobacterial toxins and toxin production. Phycologia, 35(6

Supplement), 12-24.
Sivonen, K., Carmichael, W.W., Namikoshi, M., Rinehart, K.L., Dahlem, A.M. and
Niemelä, S.I. 1990a Isolation and characterization of hepatotoxic microcystin
homologues from the filamentous freshwater cyanobacterium Nostoc sp. strain 152. App.
Env. Microbiol., 56, 2650-2657.
Sivonen, K., Himberg, K., Luukkainen, R., Niemelä, S.I., Poon, G.K. and Codd, G.A.
1989a Preliminary characterization of neurotoxic blooms and strains from Finland. Tax.
Assess., 4, 339-352.
Sivonen, K., Kononen, K., Carmichael, W.W., Dahlem, A.M., Rinehart, K.L., Kiviranta, J.
and Niemelä, S.I. 1989b Occurrence of the hepatotoxic cyanobacterium Nodularia
spumigena in the Baltic Sea and the structure of the toxin. App. Env. Microbiol., 55,
1990-1995.
Sivonen, K., Namikoshi, M., Evans, W.R., Carmichael, W.W., Sun, F., Rouhiainen, L.,
Luukkainen, R. and K.L. Rinehart. 1992a Isolation and characterization of a variety of
microcystins from seven strains of the cyanobacterial genus Anabaena. App. Env.
Microbiol., 58, 2495-2500.
Sivonen, K., Namikoshi, M., Evans, W.R., Färdig, M., Carmichael, W.W. and Rinehart,
K.L. 1992b Three new microcystins, cyclic heptapeptide hepatotoxins, from Nostoc sp.
strain 152. Chem. Res. Toxicol., 5, 464-469.
Sivonen, K., Namikoshi, M., Evans, W.R., Gromov, B.V., Carmichael, W.W. and
Rinehart, K.L. 1992c Isolation and structures of five microcystins from a Russian
Microcystis aeruginosa strain CALU 972. Toxicon., 30,1481-1485.
Sivonen, K., Namikoshi, M., Luukkainen, R., Färdig, M., Rouhiainen, L., Evans, W.R.,
Carmichael, W.W., Rinehart, K.L. and Niemelä, S.I. 1995 Variation of cyanobacterial
hepatotoxins in Finland. In: M. Munawar and M. Luotola [Eds] The Contaminants in the
Nordic Ecosystem, Dynamics, Processes and Fate. Ecovision World Monograph Series,
SPB Academic Publishing, Amsterdam, 163-169.
Sivonen, K., Niemelä, S.I., Niemi, R.M., Lepistö, L., Luoma, T.H. and Räsänen, L.A.
1990b Toxic cyanobacteria (blue-green algae) in Finnish fresh and coastal waters.
Sivonen, K., Skulberg, O.M., Namikoshi, M., Evans, W.R., Carmichael, W.W. and
Rinehart, K.L. 1992d Two methyl ester derivates of microcystins, cyclic heptapeptide
hepatotoxins, isolated from Anabaena flos-aquae strain CYA 83/1. Toxicon., 30, 1465-
1471.
Skulberg, O.M., Carmichael, W.W., Anderson, R.A., Matsunaga, S., Moore, R.E. and
Skulberg, R. 1992 Investigations of a neurotoxic Oscillatorialean strain (cyanophyceae)
and its toxin. Isolation and characterization of homoanatoxin-a. Env. Toxicol. Chem., 11,
321-329.
Skulberg, O.M., Underdal, B. and Utkilen, H. 1994 Toxic waterblooms with cyanophytes
in Norway - current knowledge. Algol. Stud., 75, 279-289
Smith, C. and Sutton, A. 1993 The persistence of anatoxin-a in reservoir water.

Foundation for Water Research, UK Report No. FR0427.
Snell, T.W. 1980 Blue-green algae and selection in rotifer populations. Oecologia Berlin,
46, 343-346.
Stevens, D.K. and Krieger, R.I. 1991 Stability studies on the cyanobacterial nicotinic
alkaloid anatoxin-a. Toxicon, 29, 167-179.
Stoner, R.D., Adams, W.H., Slatkin, D.N. and Siegelman, H.W. 1989 The effects of
single L-amino acid substitutions on the lethal potencies of the microcystins. Toxicon.,
27, 825-828.
Sugaya, Y., Yasuno, M. and Yani, T. 1990 Effects of toxic Microcystis viridis and isolated
toxins on goldfish. Jpn J. Limnol., 51(3), 149-153.
Sugiura, N. Nishimura, O. Inamori, Y. Ouchiyama, T. Sudo, R. 1997 Grazing

characteristics of musty-odour-compound-producing Phormidium tenue by a
microflagellate, Monasguttula. Wat. Res., 31(11), 2792-2796.
Swoboda, U.K., Dow, C.S., Chaivimol, J., Smith, N. and Pound, B.P. 1994 Alternatives
to the Mouse Bioassay for Cyanobacterial Toxicity Assessment. In: G.A. Codd, T.M.
Jeffries, C.W. Keevil and E. Potter [Eds] Detection Methods for Cyanobacterial Toxins.
Proceedings of the First International Symposium on Detection Methods for
Cyanobacterial Toxins, September 1993, The Royal Society of Chemistry, 106-110.
Takenaka, S. and Watanabe, M.F. 1997 Microcystin-LR degradation by Pseudomonas

aeruginosa alkaline phosphatase. Chemosphere, 34, 749-757.
Tencalla, F.G., Dietrich, D.R. and Schlatter, C. 1994 Toxicity of Microcystis aeruginosa
peptide toxin to yearling rainbow trout (Oncorhynchus mykiss). Aqua. Toxicol., 30(3),
215-224.
Todorova, A. and Jttner, F. 1996 Ecotoxicological analysis of nostocyclamide, a modified

cyclic hexapeptide from Nostoc. Phycologia, 35(6 Supplement), 183-188.
Törökné, A.K. 1991 Toxin producing cyanobacteria in Hungarian fresh waters. Doctoral
Dissertation, Budapest, Hungary.
Törökné, A.K. 1997 Interlaboratory trial using Thamnotox kit for detecting cyanobacterial
toxins. In: Proceedings of VIII International Conference on Harmful Algae, Vigo, Spain,
114.
Trimurtulu, G., Ogino, J., Helzel, C.E., Husebo, T.L., Jensen, C.M., Larsen, L.K.,
Patterson, G.M.L., Moore, R.E., Mooberry, S.L., Corbett, T.H. and Valeriote, F.A. 1995
Structure determination, conformational analysis, chemical stability studies, and
antitumor evaluation of the cryptophycins. Isolation of 18 new analogs from Nostoc sp.
strain GSV 224. J. Amer. Chem. Soc., 117, 12030-12049.
Tsuji, K., Naito, S., Kondo, F., Ishikawa, N., Watanabe, M.F., Suzuki, M., Harada, K-I.
1993 Stability of microcystins from cyanobacteria: Effect of light on decomposition and
isomerization. Environ. Sci. Technol. 28, 173-177.
Tsuji, K., Setsuda, S., Watanuki, T., Kondo, F., Nakazawa, H., Suzuki, M. and Harada,
K.I. 1996 Microcystin levels during 1992-95 for lakes Sagami and Tsukui Japan. Natural
Toxins, 4, 189-194.
Utkilen, H. and Gjølme, N. 1992 Toxin production by Microcystis aeruginosa as a

function of light in continuous cultures and its ecological significance. App. Environ.
Microbiol., 58, 1321-1325.
Utkilen, H. and Gjølme, N. 1995 Iron-stimulated toxin production in Microcystis

aeruginosa. App. Environ. Microbiol., 61, 797-800.
Ueno, Y., Nagata, S., Tsutsumi, T., Hasegawa, A., Watanabe, M.F., Park, H.D., Chen,
G.C., Chen, G. and Yu, S.Z. 1996a Detection of microcystins, a blue-green algal
hepatotoxin, in drinking water sampled in Haimen and Fusui, endemic areas of primary
liver cancer in China, by highly sensitive immunoassay. Carcinogenesis, 17, 1317-1321.
Ueno, Y., Nagata, S., Tsutsumi, T., Hasegawa, A., Yoshida, F., Suttajit, M., Putsch M.
and Vasconcelos, V. 1996b Survey of microcystins in environmental water by a highly
sensitive immunoassay based on monoclonal antibody. Natural Toxins, 4, 271-276.
Van der Westhuizen, A.J. and Eloff, J.N. 1983 Effect of culture age and pH of culture
medium on the growth and toxicity of the blue green alga Microcystis aeruginosa. Zeit.
Planzenphysiol., 110, 157-163.
Van der Westhuizen, A.J. and Eloff, J.N. 1985 Effect of temperature and light on the
toxicity and growth of the blue-green alga Microcystis aeruginosa (UV-006). Planta, 163,
55-59.
Van der Westhuizen, A.J., Eloff, J.N. and Krüger, G.H.J. 1986 Effect of temperature and
light (fluence rate) on the composition of the toxin of the cyanobacterium Microcystis
aeruginosa (UV-006). Arch. Hydrobiol., 108, 145-154.
Vasconcelos, V.M. 1994 Toxic cyanobacteria (blue-green algae) in Portuguese fresh

waters. Arch Hydrobiol., 130, 439-451.
Vasconcelos, V.M., Sivonen, K., Evans, W.R., Carmichael, W.W. and Namikoshi, M.
1995 Isolation and characterization of microcystins (heptapetide hepatotoxins) from
Portuguese strains of Microcystis aeruginosa Kutz emend Elekin. Arch. Hydrobiol., 134,
295-305.
Vasconcelos, V.M., Sivonen, K., Evans, W.R., Carmichael, W.W. and Namikoshi, M.
1996 Microcystin (heptapeptide hepatotoxins) diversity in cyanobacterial blooms
collected in Portuguese fresh waters. Wat. Res., 30, 2377-2384.
Vezie, C., Brient, L., Sivonen, K., Betru, G., Lefeuvre, J.C. and Salkinoja-Salonen, M.
1997 Occurrence of microcystins containing cyanobacterial blooms in freshwaters of
Brittany (France). Arch. Hydrobiol., 139, 401-413.
Vezie, C., Brient, L., Sivonen, K., Bertru, G., Lefeuvre, J.C. and Salkinoja-Salonen, M.
1998 Variation of microcystin content of cyanobacterial blooms and isolated strains in
Grand-Lieu lake (France). Microbiol. Ecol., 35, 126-135.
Von Elert, E. and Juttner, F. 1996 Factors influencing the allelopathic activity of the
planktonic cyanobacterium Trichormus doliolum. Phycologia, 35(6 Supplement), 68-73.
Watanabe, M. F. 1996 Production of Microcystins. In: M.F. Watanabe, H. Harada, W. W.

Carmichael and H. Fujiki, [Eds] Toxic Microcystis. CRC Press, London, 262 pp.
Watanabe, M.F. and Oishi, S. 1980 Toxicities of Microcystis aeruginosa collected from
some lakes, reservoirs, ponds and moat in Tokyo and adjacent regions. J. Limnol., 41,
5-9.
Watanabe, M.F. and Oishi, S. 1985 Effects of environmental factors on toxicity of a

cyanobacterium (Microcystis aeruginosa) under culture conditions. App. Environ.
Microbiol., 49, 1342-1344.
Watanabe, M.F., Harada, K.I., Matsuura, K., Oishi, S., Watanabe, Y. and Suzuki, M.
1989 Heptapeptide toxins contained in natural samples of Microcystis species. Tox.
Assess., 4, 487-497.
Watanabe M.F., Oishi, S., Harada, K-.I., Matsuura, K., Kawai, H. and Suzuki, M. 1988
Toxins contained in Microcystis species of cyanobacteria (blue-green algae). Toxicon,
26, 1017-1025.
Watanabe, M.F., Oishi, S., Watanabe, Y. and Watanabe, M. 1986 Strong probability of
lethal toxicity in the blue-green alga Microcystis viridis Lemmermann. J. Phycol., 22,
552-556.
Watanabe, M.F., Park, H-D., Kondo, F., Harada, K-I., Hayashi, H. and Okino, T. 1997
Identification and estimation of microcystins in freshwater mussels. Nat. Toxins, 5, 31-35.
Watanabe, M.M., Kaya, K. and Takamura, N. 1992 Fate of the toxic cyclic heptapeptides,
the microcystins, from blooms of Microcystis (cyanobacteria) in a hypertrophic lake. J.
Phycol., 28, 761-767.
Watanabe, M.M., Zhang, X and Kaya, K. 1996 Fate of toxic cyclic heptapeptides,
microcystins, in toxic cyanobacteria upon grazing by the mixotrophic flagellate
Poterioochromonas malhamensis (Ochromonadales, Chrysophyceae). Phycologia, 35(6
Weckesser, J. and Drews, G. 1979 Lipopolysaccharides of photosynthetic prokaryotes.

Ann. Rev. Microbiol., 33, 215-239.
Weckesser, J., Martin, C. and Jakobi, C. 1996 Cyanopeptolins, depsipeptides from

cyanobacteria. System. App. Microbiol., 19, 133-138.
Weise, G., Drews, G., Jann, B. and Jann, K. 1970 Identification and analysis of
lipopolysaccharide in cell walls of the blue-green algae Anacystis nidulans. Arch.
Microbiol., 71, 89-98.
Welker, M. and Steinberg, C.E.W. 1998 Indirect photolysis of cyanotoxins: one possible
mechanism of their low persistence. Wat. Res. (In Press).
Wicks, R.J. and Thiel, P.G. 1990 Environmental factors affecting the production of
peptide toxins in floating scums of cyanobacterium Microcystis aeruginosa in a
hypertrophic African reservoir. Environ. Sci. Technol., 24, 1413-1418.
Wiedner, C. and Nixdorf, B. 1997 Verbreitung und Steuerung der Entwicklung von
toxischen und nicht-toxischen Cyanobakterien in ostbrandenburgischen Gewässern
unterschiedlicher Trophie und Hydrographie im Zuge der reduzierten Belastung.
WaBoLu-Hefte 4/97, Berlin ISSN 0175-4211.
Willn, T. and Mattsson, R. 1997 Water-blooming and toxin-producing cyanobacteria in

Swedish fresh and brackish waters, 1981-1995. Hydrobiologia, 353, 181-192.
Williams, D.E., Craig, M., Dawe, S.C., Kent, M.C., Holmes, C.F.B. and Anderson, R.J.
1997 Evidence for a covalently bound form of microcystin-LR in salmon larvae and
dungeness crab larvae. Chem. Res. Toxicol., 10, 463-469.
Yanni, Y.G. and Carmichael, W.W. 1997 Screening of cyanobacteria isolated from soil,
rice fields and water resources of the Nile Delta for production of cyanotoxins. Abstract,
In: VIII International Conference on Harmful Algae, 25-29 June, 1997 Vigo, Spain, 216.
Yoo, S., Carmichael, W.W., Hoehn, R.C. and Hrudey, S.E. 1995 Cyanobacterial (Blue-
green Algal) Toxins: A Resource Guide. American Water Works Association Research
Foundation, Denver, Colorado.
Zhang, Q.X., Carmichael, W.W., Yu, M.J. and Li, S.H. 1991 Cyclic peptide hepatotoxins
from freshwater cyanobacterial (blue-green algae) waterblooms collected in central
China. Env. Toxicol. Chem., 10, 313-321.
Zhang, X. Watanabe, M.M., Inouye, I. 1996 Light and electron microscopy of grazing by
flagellate Poterioochromonas malhamensis (Chrysophyceae) on a range of
phytoplankton taxa. J. Phycol., 32, 37-46.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 4. HUMAN HEALTH ASPECTS
This chapter was prepared by Tine Kuiper-Goodman, Ian Falconer and Jim Fitzgerald
Evidence for adverse human health effects from cyanotoxins derives from three principal
sources: epidemiological evidence including human poisonings, animal poisonings, and
toxicological studies.
Epidemiological evidence results from studies of human populations that have shown
symptoms of poisoning or injury attributed to the presence of cyanotoxins in drinking
water or other sources of water. This type of evidence depends upon good case
definition, good characterisation of exposure and a reporting system that enables these
data to be compared. Such evidence, discussed later, includes the fatal intoxication by
microcystin of 50 dialysis patients in Brazil, and the hospitalisation of 140 children
supplied with water from a dam containing Cylindrospermopsis raciborskii in Australia.
Reports of gastro-enteritis after the appearance of cyanobacterial blooms in drinking
water sources come from North and South America, Africa and Europe. Most cases of
human injury attributed to cyanobacterial toxins have been studied retrospectively, and
complete epidemiological data, especially regarding exposure (number of organisms,
type and concentration of cyanotoxins), are rarely available. Nevertheless,
epidemiological evidence is of special importance in directly demonstrating the link
between toxin exposures and human health outcomes, which otherwise cannot be
derived directly from animal experiments.
A large number of animal poisonings have been recorded, some of which have been
reported in detail, while others are anecdotal. These are important in the overall body of
knowledge on cyanobacterial toxicity, because they demonstrate effects under natural
conditions. However, cyanotoxin exposure is rarely quantified and cannot be directly
extrapolated to human populations.
Although animal toxicity tests are not performed under conditions of natural exposure
and are undertaken with different species, they are of particular importance because
they are conducted under controlled laboratory conditions. They provide plausibility for
the role of cyanotoxins in human and animal poisonings, and provide information on their
comparative toxicity. Oral toxicity testing is especially relevant, and has been carried out
on large animals as well as on rodents.
The overall case for the human health relevance of exposure to cyanotoxins can only be
assessed properly by combining all available evidence and understanding its strength
and its weaknesses.
Of all the cyanobacterial toxins currently known, the cyclic peptides represent the
greatest concern to human health because of the potential risk of long-term exposure to
comparatively low concentrations of the toxins in drinking water supplies. As will be
discussed in sections 4.2.1 and 4.2.2, the cyclic peptide microcystins and nodularins are
specific liver poisons in mammals. Following acute exposure to high doses, they cause
death from liver haemorrhage or from liver failure, and they may promote the growth of
liver and other tumours following chronic exposures to low doses.
The alkaloid neurotoxins (anatoxins and saxitoxins/PSP toxins) have shown only acute
effects in mammals. Risk assessment will, therefore, be limited to acute exposure.
However, there are a number of complicating factors. Firstly, while there is a great deal
of published information on the mammalian and human toxicity and toxicology of
saxitoxins/PSP toxins, the animal data for anatoxins is rather sparse, and completely
lacking for human exposure. Secondly, some of the alkaloid toxins, PSP toxins in
particular, are known to accumulate to high concentrations in marine and freshwater
biota (see section 3.4.5). In many areas of the world, the local human population will be
consuming fish, shellfish and animals that have grown in, or nearby, water bodies
contaminated with toxic cyanobacteria. Thus there may be several sources of oral
exposure for neurotoxins in addition to drinking water or recreational water. The potential
for transfer of PSP toxins, and possibly other neurotoxins, to humans via these routes
may be significant in freshwaters.
The toxicity of pure cylindrospermopsin has recently been studied following the
identification of the toxin in 1992 (Ohtani et al., 1992). These studies have helped to
characterise the histopathology of cylindrospermopsin after intraperitoneal (i.p.) injection
(Hawkins et al., 1997; Falconer et al., 1999; Seawright et al., 1999). It is not known
whether cylindrospermopsin will elicit histopathological damage following chronic (long-
term) exposure at low concentrations.
Lipopolysaccharides (LPS) from cyanobacteria can elicit both allergic and toxic
responses in humans, although little is known about their acute or chronic effects. The
lipid part of the molecule is believed to cause the deleterious response in humans, while
the sugar moiety is important in determining LPS solubility and micelle properties, and
hence is important in modulating toxicity. A lack of knowledge regarding the occurrence
and toxicity of cyanobacterial LPS, combined with the diversity of LPS structures within
the cyanobacteria, should be considered by health officials and water managers,
particularly when gastrointestinal and respiratory symptoms are reported from exposed
humans.
4.1 Human and animal poisonings

Water supplies have been associated with gastrointestinal illness throughout human
history with cholera, dysentery, and typhoid responsible for much human misery and
death. The epidemiological evidence for human illness due to cyanobacterial toxins
therefore has to be viewed against a background of alternative causes, with bacterial,
viral or protozoal infections being the first causes to be investigated.
4.1.1 Short-term effects in humans
The recorded cases of gastrointestinal and hepatic illness that can be reliably attributed
to cyanobacterial toxins in water supplies have all been coincident with either the
breakdown of a natural cyanobacterial bloom or with the artificial lysis of a bloom by
application of copper sulphate. Both mechanisms lead to cyanotoxin release from
decomposing cells. Whereas treatment procedures might have removed cyanotoxins
bound in intact cells, they were not effective in removing the dissolved cyanotoxins in
these cases. The earliest reported cases of gastro-enteritis from cyanobacteria were in
the population of a series of towns along the Ohio River in 1931. In these cases low
rainfall had caused the water of a side branch of the river to develop a cyanobacterial
bloom which was then washed into the main river. As this water moved downstream a
series of outbreaks of illness were reported, which could not be attributed to infectious
agents (Tisdale, 1931). In Harare, Zimbabwe, children living in an area of the city
supplied from a particular water reservoir, developed gastro-enteritis each year at the
time when a natural bloom of Microcystis was decaying in the reservoir. Other children in
the city with different water supplies were not affected and no infectious agent was
identified (Zilberg, 1966). The most lethal outbreak attributed to cyanobacterial toxins in
drinking water occurred in Brazil, when a newly flooded dam developed an immense
cyanobacterial bloom. Eighty-eight deaths, mostly children, were reported to have
occurred (Teixera et al., 1993) (Box 4.1).
Examples of illness following the use of copper sulphate to destroy a cyanobacterial

bloom in a water storage reservoir have been described in the USA and in Australia. In
each of these instances the cyanobacterial genera involved were identified, either at the
time or subsequently. In one incident, 62 per cent of the population connected to a
filtered, chlorinated water supply developed symptoms of gastro-enteritis within a period
of five days. No pathogens were found, and it was concluded that abloom of
cyanobacteria in an open storage reservoir which had over 100,000 cells per ml of
Schizothrix calcola, Plectonema, Phormidium and Lyngbya was responsible. The
reservoir had just been treated with copper sulphate (Lippy and Erb, 1976).
Box 4.1 Gastro-enteritis epidemic in the area of the Itaparica Dam, Bahia, Brazil
A severe gastro-enteritis epidemic in the Paulo Afonso region of Bahia State in Brazil followed the
flooding of the newly constructed Itaparica Dam reservoir in 1988. Some 2,000 gastro-enteritis
cases, 88 of which resulted in death, were reported over a 42-day period.
Clinical data and water sample tests were reviewed, blood and faecal specimens from gastro-
enteritis patients were subjected to bacteriological, virological and toxicological testing and
drinking water samples were examined for micro-organisms and heavy metals. The results
demonstrated that the source of the outbreak was water impounded by the dam and pointed to
toxin produced by cyanobacteria present in the water as the responsible agent. No other
infectious agent or toxin was identified, and cases occurred in patients who had been drinking
only boiled water. The cases were restricted to areas supplied with drinking water from the dam.
Cyanobacteria of the Anabaena and Microcystis genera were present in untreated water at 1,104
to 9,755 units per ml (conversion of colony units to cells per ml depends on colony size, but a
minimum of 100 cells per colony is likely in a mixed bloom of these genera).
In Armidale, Australia, the water supply reservoir had been monitored for blooms of toxic
Microcystis for several years, when a particularly dense bloom occurred. Within three
weeks the water supply authority treated the reservoir with 1 ppm (part per million) of
copper sulphate, which killed the bloom. An epidemiological study of the local population
indicated liver damage occurring simultaneously with the termination of the bloom (see
Box 4.2).
A more severe outbreak of cyanobacterial toxicity in a human population occurred on an

island off the north-eastern coast of Australia. Due to complaints of bad taste and odour
in the water supply, which were attributed to a cyanobacterial bloom, the authorities
treated the reservoir with copper sulphate. Within a week numerous children developed
severe hepato-enteritis, and a total of 140 children and 10 adults required hospital
treatment (Byth, 1980). Cylindrospermopsis raciborskii was identified as the
cyanobacterium responsible for this episode (see Box 4.3).
Within human populations, for a variety of reasons, there will be individuals who are at a
much greater risk of injury from cyanotoxins than the population as a whole. Children are
the most obvious example, because they drink a higher volume of water in proportion to
their body weight than adults. Individuals who already have injury to organs susceptible
to cyanobacterial toxins are also at increased risk, such as people with hepatitis, liver
cirrhosis, toxic liver injury from other sources, or kidney damage. Kidney dialysis patients,
if exposed to microcystins in the water used for dialysis, are especially vulnerable
because treatment exposes them intravenously to large volumes of water. In a
disastrous incident in Caruaru, Brazil, 117 patients developed cholestatic liver disease
and at least 47 deaths were attributed to dialysis with water containing cyanobacterial
toxins. Examination of the carbon filter from the dialysis unit demonstrated microcystin-
LR, as did the blood and liver tissue of deceased patients (Jochimsen et al., 1998) (see
Box 4.4).
Box 4.2 Toxic Microcystis in the Armidale water supply reservoir and public health
The city of Armidale, New South Wales, Australia has a drinking water supply from a eutrophic
reservoir which has experienced repeated blooms of Microcystis aeruginosa from the early 1970s
to the present day. The reservoir, Malpas Dam, is on a plateau 150 m higher than the treatment
plant to which it is connected by about 20 km of pipeline. The plant treated water by pre-
chlorination, alum flocculation, sedimentation, rapid sand filtration and post chlorination and
fluoridation. The geography of the water storage results in toxic scums accumulating around the
offtake area through wind-drift. The local water authority use copper sulphate dosing to control
blooms, with aerial distribution of 1 ppm of copper in the top metre of water. The cyanobacterial
blooms have been monitored for toxicity, and the toxin has been characterised as microcystin-YM
(Botes et al., 1985).
In 1981, a particularly extensive toxic bloom of Microcystis was monitored during its development
and subsequent termination with copper sulphate. At the time of this bloom, complaints of bad
taste and odour were received, leading to the copper sulphate treatment of the reservoir. This
event was used as the basis for a retrospective epidemiological study of liver function in the
population consuming the water, compared with a population in the same region supplied from
other reservoirs. The data for the activity of plasma enzymes measuring liver function were
obtained for patients having blood samples at the Regional Pathology Laboratory for the six
weeks prior to the bloom, the six weeks of peak bloom and its termination, and for six weeks after
that. The data were then separated into analyses from patients having the Malpas drinking water
supply, and those using other supplies.
Serum enzymes measuring liver function in patients consuming drinking water from Malpas Dam
or from other supplies during a heavy bloom of Microcystis aeruginosa and its termination with
copper. GGT = gamma glutamyl transferase; ALT = alanine aminotransferase; AST = aspartate
aminotransferase; AP = alkaline phosphatase (From The Medical Journal of Australia, 1983;
1511-1514)
As shown in the figure above, a statistically significant increase in gamma glutamyl transferase in
the blood was seen with the group using the Malpas water supply during the peak of the bloom
and its lysis with copper sulphate, compared with the same population before and after, or the
other population on independent water supplies. The clinical record gave no evidence of an
infectious hepatitis outbreak or disproportionate alcoholism (Falconer et al., 1983). While the
mean increase in enzyme activity was indicative of minor liver toxicity, individuals within the
population studied had highly elevated enzyme activity, indicating substantial liver damage. The
serum enzyme showing this change has also been used as an indicator of Microcystis toxicity in
experimental studies with pigs and rodents, where it is an effective marker for liver injury (Fawell
et al., 1993; Falconer et al., 1994).
4.1.2 Chronic effects in humans
While acute toxicity is the most obvious problem in cyanobacterial poisoning, a long-term
risk may also be present. Short exposures to toxins may result in long-term injury, and
chronic low-level exposure may cause adverse health effects. Animal experiments have
shown chronic liver injury from continuing oral exposure to microcystins. In particular the
possibility of carcinogenesis and tumour growth promotion need careful evaluation,
because both have been shown in animal experimentation.
The incidence of human hepatocellular carcinoma in China is one of the highest in the
world, and studies have explored whether cyanobacterial toxins are part of a complex of
risk factors which increase the incidence of this disease. The distribution of
hepatocellular carcinoma varies geographically. In south-east China, rates of less than
15 incidents per 100,000 people are seen in some districts, compared with over 60
incidents per 100,000 people in adjacent localities. Two proven risk factors are infection
with hepatitis B virus and intake of aflatoxin B1 from food items, such as corn. The third
significant element of association was the source of drinking water. On a village basis,
lower cancer mortality rates were seen when the water was drawn from deep wells,
compared with much higher rates when the water came from ponds and ditches.
Cyanobacteria are abundant in surface waters in south east China where the incidence
of hepatocellular carcinoma is highest, and it has been proposed that microcystins in the
drinking water are responsible for the higher incidences of cancer among drinkers of
pond and ditch water (Yu, 1989, 1995). Very low levels of cyanotoxins were found in one
study with limited sampling of some drinking water sources in China (Ueno et al., 1996),
but a more representative study would be needed to investigate this association. At
present, all three risk factors are being reduced concurrently in China and liver cancer
rates appear to be decreasing (Yu, 1995) (see also Box 5.3).
Box 4.3 Palm Island Mystery Disease
In 1979, there was a major outbreak of hepato-enteritis among the children of an Aboriginal
community living on a tropical island off the coast of Queensland, Australia. Altogether 140
children and 10 adults required treatment, which was provided by the local hospital for less
severe cases and by the regional hospital on the mainland for severe cases possibly requiring
intensive care. Diagnostic information included a detailed clinical examination showing malaise,
anorexia, vomiting, headache, painful liver enlargement, initial constipation followed by bloody
diarrhoea and varying levels of severity of dehydration. Urine analysis showed electrolyte loss
together with glucose, ketones, protein and blood in the urine, demonstrating extensive kidney
damage. This was the major life-threatening element of the poisoning. Blood analysis showed
elevated serum liver enzymes in some children, indicating liver damage. Sixty-nine percent of
patients required intravenous therapy and in the more severe cases the individuals went into
hypovolaemic/acidotic shock. After appropriate treatment all the patients recovered (Byth, 1980).
Examination of faecal samples and foods eliminated a range of infectious organisms and
toxins as possible causes for the outbreak, hence the name "Palm Island Mystery Disease". The
affected population, however, all received their drinking water supply from one source, Solomon
Dam. Families on alternative water supplies on the island were not affected by the disease.
Prior to the outbreak of the illness an algal bloom occurred in Solomon Dam. The bloom
discoloured the water and gave it a disagreeable odour and taste. When the bloom became
dense, the dam reservoir was treated with 1 ppm of copper sulphate (Bourke et al., 1983).
Clinical injury among consumers on that water supply was reported the following week.
The organisms from the dam were cultured and administered to mice, following which the
mice slowly (over several days) developed widespread tissue injury involving the gastrointestinal
tract, the kidney and liver (Hawkins et al., 1985). The widespread tissue damage and delayed
effects are quite different to those following Microcystis aeruginosa administration (Falconer et al.,
1981; see section 4.2.1).
Subsequent monitoring of the algal blooms in the dam identified the cyano-bacterium
Cylindrospermopsis raciborskii as the cause of the blooms, with seasonal cell concentrations of
up to 300,000 cells per ml of water. This organism does not form scums, and has the highest cell
concentrations well below the water surface. In order to reduce bloom formation, the responsible
authorities later introduced destratification of the reservoir (Hawkins and Griffiths, 1993).
Subsequent research on toxins produced by Cylindrospermopsis raciborskii has identified
the cytotoxic alkaloid cylindrospermopsin which is toxic after i.p. and oral administration. It is
possible that other toxins will be isolated from this organism (Hawkins et al., 1997, see also
section 4.2.7).
Box 4.4 Outbreak of severe hepatitis following haemodialysis in Caruaru, Brazil
In February 1996, an outbreak of severe hepatitis occurred at a Brazilian haemodialysis centre in

Caruaru, Brazil, located 134 km from Recife, the state capital of Pernambuco. At this clinic 117 of
136 patients (86 per cent) experienced visual disturbances, nausea, vomiting, muscle weakness
and painful hepatomegaly, following routine haemodialysis treatment. Subsequently, 100 patients
developed acute liver failure and 50 of these died. As of October 1997, 49 of the deaths could be
attributed to a common syndrome now called "Caruaru Syndrome". This syndrome includes:
• Symptoms. Painful huge hepatomegaly, jaundice and a bleeding diathesis manifested

by ecchymosis, epistaxis and metrorrhagia.
• Laboratory picture. Elevated transaminases, variable hyperbilirubinemia, prolonged
prothrombin time and severe hypertriglyceridemia.
• Histopathology. Light microscopy - disruption of liver plates, liver cell deformity,
necrosis, apoptosis, cholestasis, cytoplasmic vacuolisation, mixed leukocyte infiltration
and multinucleated hepatocytes; electron microscopy - intracellular oedema,
mitochondrial changes, rough and smooth endoplasmic reticulum injuries, lipid vacuoles
and residual bodies.
The pattern of liver plate disruption was identical to that found with previous laboratory animal
experiments involving microcystin exposure.
Initial reports, at the beginning of March 1996, from the medical and public health profession in
Pernambuco, implicated several xenobiotic or microbial sources as possibly responsible for a
cluster of human fatalities. However, comparison with previous knowledge about cyanotoxicosis,
from other parts of the world, suggested that cyanobacteria toxins might be responsible. Since
phytoplankton counts were not being made by the city's water utility at the time of the outbreak,
the presence of toxic cyanobacteria or microcystin could not be confirmed or denied. Examination
of previous years' phytoplankton counts showed that cyanobacteria had been dominant in the
reservoir during the summer months since at least 1990, with the most common genera present
being Microcystis, Anabaena and Anabaenopis (Cylindrospermopsis). Samples collected on
March 29, 1996 showed that the most common cyanobacteria present were Aphanizomenon,
Oscillatoria and Spirulina (average number 24,500 cells per ml).
Carbon from the dialysis centre's in-house water treatment system showed chromatogram peaks
that had characteristics of microcystins when analysed using HPLC. One of these peaks had a
retention time corresponding to microcystin-LR, the most commonly found of the microcystin
family. Further examination of carbon, sand and cation/anion exchange filters showed microcystin
in the ppm range.
Blood sera, provided by state health officials in Pernambuco and sent through the Centers for
Disease Control in Atlanta Georgia, USA, from affected and control patients, and liver tissue from
deceased patients, showed a microcystin content of up to 10 ng ml-1 for sera and 0.1 to 0.5 ng
mg-1 in liver tissue. All exposed patient sera and tissue were positive for microcystins. Chemical
analysis of the most abundant microcystins revealed the presence of microcystins -YR, -LR and -
AR. The average microcystin concentration in liver samples from 33 victims compares closely
with that reported in laboratory test animals receiving acute exposure to microcystin-LR. When
the analyses are completed, the resulting information on dose and related effects will be
important in the hazard characterisation and risk assessment of microcystins in humans. A clinic
for survivors set up by Brazilian health authorities will help monitor their progress and serve as a
data base to assess long-term health effects, including cancer rates, because the microcystins
are known to be potent liver tumour promoters in experimental animals.
The available biological and chemical evidence supports microcystins from the reservoir water as
being the major factor in the deaths of patients at the dialysis clinic. The occurrence of
microcystin in the water used for dialysis was due to insufficient treatment of the water obtained
by truck from the city's water treatment system, as well as to insufficient functioning and
maintenance of the clinic's dialysis water treatment system.
4.1.3 Injury from recreational exposure
There have been repeated descriptions of adverse health consequences for swimmers
exposed to cyanobacterial blooms. Even minor contact with cyanobacteria in bathing
water can lead to skin irritation and increased likelihood of gastrointestinal symptoms
(Pilotto et al., 1997), Some severe skin reactions have been reported, particularly from
contact with the marine cyanobacterium, Lyngbya majuscula, which causes deep
blistering when trapped under the bathing suit of swimmers (Grauer, 1961). In this case
the organism contains a powerful dermal toxin which is further described in section 4.2.8.
Individual sensitivity to cyanobacteria in bathing waters varies greatly, because there

can be both allergic reactions and direct responses to toxins. The cyanobacterial
pigments can cause severe allergic reactions in sensitive individuals (Cohen and Reif,
1953). Cyanobacteria have features in common with general airborne allergens and
surveys have shown allergic responses to cyanobacteria in patients with naso-bronchial
allergy (Mittal et al., 1979). More extensive discussion on this aspect can be found in
Ressom et al. (1994). Illnesses from accidental swallowing of cyanobacteria during
swimming are described in Box 4.5.
4.1.4 Animal poisonings
Numerous cases of animal poisoning (often lethal) substantiate the concern of health
hazards for humans exposed to cyanobacteria. Therefore, the most important incidents
are reported below. For impact of cyanotoxins on aquatic biota, see section 3.5.
The first report of cyanobacterial poisoning was of the deaths of cattle, sheep, dogs,
horses and pigs after drinking a scum of Nodularia spumigena in Lake Alexandrina,
Australia (Francis, 1878). Francis confirmed the source of poisoning by dosing a calf
with the scum from the lake, which subsequently caused the death of the animal. Since
that time there have been frequent instances of farm animal poisonings from
cyanobacterial water blooms, and also deaths of pet dogs after swimming in, or eating,
cyanobacteria. Other affected animals range in size from ducks, coots and other
waterfowl, to skunks and mink, and even up to rhinoceros (see Carmichael, 1992).
Besides the consumption of cyanobacteria from the water, it has been suggested that an
additional source of intoxication for terrestrial animals is cyanotoxins that have
bioaccumulated in the food chain. For example, freshwater mussels accumulate both
microcystins (Prepas et al., 1997) and saxitoxins (Negri and Jones, 1995), and mussels
are an important food source for water rats, musk rats and birds.
An extensive list of poisoning incidents, and discussion of them, is included in Ressom

et al. (1994). A selected group is given in Table 4.1. As with cases of suspected human
poisoning by cyanobacterial toxins, in the earlier cases much of the evidence necessary
for proof of poisoning is unavailable. Ideally, for such proof, the toxic organisms or toxins
need to be identified in the body of the victim, together with the evidence of access to
toxic material (see also section 3.5). One of the most convincing mammalian poisonings
has been the recorded deaths of sheep drinking from a farm dam contaminated with the
neurotoxic Anabaena circinalis in Australia (Negri et al., 1995). The authors recovered
high concentrations of saxitoxins from cyanobacteria in the farm dam and from the
rumen fluid from the dead sheep.
Box 4.5 Examples of health effects due to recreational exposure
1959 Saskatchewan, Canada. In spite of recreational use warnings and deaths in livestock,
people swam in a lake infested with cyanobacteria. Thirteen persons became ill (headaches,
nausea, muscular pains, painful diarrhoea). In the excreta of one patient (a medical doctor who
had accidentally ingested 300 ml of water), numerous cells of Microcystis spp. and some
trichomes of Anabaena circinalis were identified (Dillenberg and Dehnel, 1960).
1989 United Kingdom. Ten of 20 army recruits showed symptoms indicating intoxication (e.g.
vomiting, diarrhoea, central abdominal pain, blistering of the lips, sore throats) after swimming
and canoe training in water with a dense bloom of Microcystis spp. Two of the recruits developed
severe pneumonia attributed to the aspiration of a Microcystis toxin and needed hospitalisation
and intensive care (Turner et al., 1990). The severity of illness appeared to be related to the
swimming skills and amount of water ingested.
1995 Australia. Epidemiological evidence of adverse health effects after recreational water
contact was established in a prospective study involving 852 participants. Results showed an
elevated incidence of diarrhoea, vomiting, flu symptoms, skin rashes, mouth ulcers, fevers, eye or
ear irritations within seven days following exposure (Pilotto et al., 1997). Symptoms increased
significantly with duration of water contact and cell density of cyanobacteria.
The post mortem symptoms of toxicity can also be characteristic of the poison involved.
Animal deaths from cyanobacterial toxicity have been reported from North and South
America, Europe, Australia and Africa. The major injury reported is hepatotoxicosis, i.e.
liver poisoning. The cyanobacteria responsible have been Microcystis aeruginosa,
Nodularia spumigena and Oscillatoria (Planktothrix) agardhii. In the most recent cases,
post mortem examination has shown evidence of cyanobacterial ingestion as well as
characteristic tissue injury in the liver.
Table 4.1 Selected examples of animal poisoning associated with cyanobacteria
Country Species killed Pathology Organism Reference

Argentina Cattle Hepatotoxicity Microcystis Odriozola et al., 1984
aeruginosa
Australia Sheep Hepatotoxicity Microcystis Jackson et al., 1984
aeruginosa
Australia Sheep Neurotoxicity, PSPs Anabaena Negri et al., 1995
circinalis
Canada Cattle Neurotoxicity, Anabaena flos- Carmichael and
anatoxin-a aquae Gorham, 1978
Canada Waterfowl Neurotoxicity, Anabaena flos- Pybus and Hobson,
anatoxin-a aquae 1986
Finland Dogs Hepatotoxicity, Nodularia Perrson et al., 1984
nodularin spumigena
Finland Waterfowl, fish, Hepatotoxicity, gill Planktothrix Eriksson et al., 1986
muskrats damage agardhii
Norway Cattle Hepatotoxicity, Microcystis Skulberg, 1979
microcystin aeruginosa
UK
England Shepherd dogs Hepatotoxicity, Microcystis Pearson et al., 1990
microcystin aeruginosa
Scotland Dogs Neurotoxicity, Oscillatoria spp. Gunn et al., 1992
anatoxin-a
Scotland Fish (trout) Gill injury, microcystin Microcystis Bury et al., 1995
aeruginosa
USA Dogs Neurotoxicity, Anabaena flos- Mahmood et al., 1988
anatoxin-a(S) aquae
PSPs Paralytic shellfish poisons
The other main cause of livestock and pet deaths due to cyanobacterial toxins has been
from acute neurotoxicity leading to respiratory failure, with no post mortem indications of
organ injury. In one case (Gunn et al., 1992) the neurotoxin, anatoxin-a, was isolated
from the stomach contents of a dog. Henriksen et al., (1997) demonstrated lethality in
wild ducks due to anatoxin-a(S). The cyanobacteria associated with deaths from
neurotoxicity are Anabaena flos-aquae, Anabaena circinalis, Aphanizomenon flos-aquae
and Oscillatoria spp. The toxins anatoxin-a, anatoxin-a(S) and saxitoxins have been
implicated in different cases.
While the reported deaths have usually occurred shortly after the animals have ingested
cyanobacterial scums, lasting injury with progressive mortality has also been seen in
animal poisoning by Microcystis aeruginosa. The characteristic symptoms are those of
liver failure with secondary photo-sensitisation, i.e. severe sunburn-like reactions
(McBarron and May, 1966; Carbis et al., 1995).
4.2 Toxicological studies
In order to set safe levels of toxicants or contaminants in food or drinking water, it is first
necessary to determine the dose level in humans that is considered to be without
adverse effects when taken daily over a lifetime; this is known as the Tolerable Daily
Intake (TDI). Ideally, this value is derived from human studies, but often such studies are
inadequate or non-existent. Alternatively, when there is an adequate experimental
database, it can be derived from animal studies. One of the major shortcomings of
animal studies is that differences in sensitivity between animals and humans vary, so
that safety factors need to be incorporated to deal with this uncertainty (see below).
A further important extrapolation problem associated with animal studies relates to

routes of exposure. Many toxins are more toxic when given by the intravenous (i.v.) or
i.p. route of administration than by the oral route. This difference is evident when LD50
values (single dose level that will cause death in 50 per cent of the exposed animals
within 7-14 days) are examined for various routes of exposure (see Chapter 3 and Table
4.2). Studies using the i.p. or i.v. route of administration require much less toxin and can
be used to indicate relative acute toxicity and may provide information on the
mechanism of toxicity. Such studies are used in bioassays (see section 13.3.1), but they
have little use in predicting toxicity after oral administration. Another problem with many
natural toxins, including cyanotoxins, is obtaining sufficient pure toxin to conduct
controlled experiments. Several studies, to be described, have used algal extracts rather
than pure toxin. An advantage of this is that the extract more closely mimics the real
environmental situation. A disadvantage is that unless the extract is fully characterised
for all the toxins that may be present and their concentrations, the assay may have
limited predictive value for scenarios involving other conditions.
Several steps are involved in determining the TDI from appropriate human or animal
studies. The highest dose associated with the absence of adverse health effects (the
NOAEL or No Observed Adverse Effect Level) is selected or, when no NOAEL is
available, the lowest dose associated with adverse effects (the LOAEL or Lowest
Observed Adverse Effect Level) is selected. In order to extrapolate from animal data to
the human situation, the next step involves the application of a number of uncertainty
factors. In the absence of reliable data on inter- and intraspecies differences in
sensitivity to a chemical substance, standard default uncertainty factors of 10 are
generally used to give a composite uncertainty factor of 100. Additional uncertainty
factors may be used to deal with inadequacies in the database and the severity of
effects noted (WHO, 1993; IPCS, 1994). When extrapolating from human data, it is also
necessary to use some of these factors. With human data there are, in addition, often
problems in determining exposure accurately and thus the dose causing adverse effects.
The expert review of the pertinent literature in April 1997, which led to the production of
this book, revealed that information currently available is insufficient for calculation of a
TDI for most of the cyanotoxins. For microcystin-LR more data exist and a provisional
value could be derived. The following sections summarise the present health-related
information on each of the cyanotoxins, in order to provide research data to health
officers and others, and in order to provide a basis for the assessment of the health
hazards of cyanotoxins.
4.2.1 Microcystins
In the various reported incidents of poisoning in humans and livestock caused by

cyanobacteria or their toxins, Microcystis is the most frequently cited organism. As a
consequence, extensive experimental studies have been carried out on this
cyanobacterium and its toxins. Although there is qualitative evidence of cause and effect,
indicating human susceptibility, these data have so far been inadequate to establish a
dose response relationship that can be used for assessing human risk in a variety of
exposure scenarios. The animal studies described below have been conducted in an
attempt to address these issues.
Metabolic disposition
There have been no pharmacokinetic studies with orally administered microcystins. After
i.v. or i.p. injection of sublethal doses of variously radiolabelled microcystins in mice and
rats, about 70 per cent of the toxin was rapidly localised in the liver (Falconer et al., 1986;
Runnegar et al., 1986; Brooks and Codd, 1987; Robinson et al., 1989, 1991; Meriluoto et
al., 1990; Lin and Chu, 1994a; Nishiwaki et al., 1994). Plasma half-lives of microcystin-
LR, after i.v. administration, were 0.8 and 6.9 minutes for the first and second phases of
elimination, but the concentration of radioactive (3H-microcystin-LR) label in the liver did
not change throughout the later part of the six-day study period (Robinson et al., 1991).
In this study, about 9 per cent of the dose was excreted early via the urinary route, with
the remainder being excreted slowly (~1 per cent per day) via the faecal route. Based on
the protective effect of microsomal enzyme inducers, it is evident that the liver plays a
large role in the detoxification of microcystins (Brooks and Codd, 1987). Time-dependent
appearance and disappearance of additional chromatograph peaks, thought to represent
detoxification products, were seen in urine, faeces and liver cytosol fractions (Robinson
et al., 1991). Three metabolic products have since been identified, a glutathione
conjugate, a cysteine conjugate and a conjugate with the oxidised ADDA diene (Kondo
et al., 1996).
Microcystin-LR does not readily cross cell membranes, and hence does not enter most
tissues. After oral uptake it is transported across the ileum into the bloodstream through
a bile acid type transporter (the multispecific organic ion transport system) present in
hepatocytes and cells lining the small intestine (Runnegar et al., 1991; Falconer et al.,
1992) and is concentrated in the liver as a result of active uptake by hepatocytes
(Runnegar et al., 1981). It is covalently bound to a 40 kdalton protein (protein
phosphatase 2A and possibly protein phosphatase 1) in the hepatocyte cytosol (Holmes
et al., 1994; Bagu et al., 1997). Some other microcystin congeners are more hydro-
phobic than microcystin-LR and may cross cell membranes by other mechanisms,
including diffusion.
Acute and subacute toxicity
Microcystin-LR is highly toxic. The LD50 by the i.p. route ranges from 25 to 150 µg kg-1
body weight (bw) in mice (a value of 50 or 60 µg kg-1 bw is commonly accepted). The
oral LD50 (administered by gavage, i.e. dosing directly into the stomach through the
mouth) is 5,000 µg kg-1 bw in one strain of mice (Fawell et al., 1994), 10,900 µg kg-1 bw in
a different strain of mice (Yoshida et al., 1997), and higher in rats (Fawell et al., 1994).
This indicates that, even by the oral route, microcystin-LR displays acute toxicity in
rodents. There is no evidence of hydrolysis of microcystins by peptidases in the stomach
and it is apparent that a significant amount of microcystin-LR passes the intestinal
barrier and is absorbed. Similarly, the oral LD50 of Microcystis extracts in mice was 50- to
170-fold higher than the i.p. LD50 of the same extracts (Falconer, 1991; Kotak et al.,
1993).
The i.p. LD50s of several of the commonly occurring microcystins (microcystin-LA, -YR,
and -YM) are similar to that of microcystin-LR, but the i.p. LD50 for microcystin-RR is
about tenfold higher (see Table 3.1).
However, because of differences in lipophilicity and polarity between the different

microcystins, it cannot be presumed that the i.p. LD50 will predict toxicity after oral
administration.
The microcystins are primarily hepatotoxins. After acute exposure by i.v. or i.p. injection
of microcystin, severe liver damage is characterised by a disruption of liver cell structure
(due to damage to the cytoskeleton), a loss of sinusoidal structure, increases in liver
weight due to intrahepatic haemorrhage, haemodynamic shock, heart failure and death.
Other organs affected are the kidneys and lungs (Hooser et al., 1990) and the intestines
(Falconer, 1994; Falconer and Humpage, 1996).
In a recent study, a single oral dose of microcystin-LR at 500 µg kg-1 bw was given to 5
and 32 week old mice, and liver pathology was examined 2-19 hours later. In 62 per cent
of the older mice, there was clear microscopic evidence of liver injury. Furthermore,
microcystin-LR and a metabolite were detected in the livers. On the contrary, none of the
young mice developed liver injury. In untreated control mice, an examination of gastric
mucosa and small intestine revealed age-related disruption of surface cell structure. The
authors suggested that this disruption may markedly influence gastro-intestinal
absorption of microcystin-LR and hence explain the observed age-dependent liver
toxicity in exposed mice (Ito et al., 1997a).
Intranasal installation of microcystin-LR in mice resulted in extensive necrosis of the

epithelium of the nasal mucosa of both the olfactory and respiratory zones, progressing
to destruction of large areas of tissue down to levels of deep blood vessels (Fitzgeorge
et al., 1994). The LD50 by this route of administration was the same as the i.p. LD50, and
dose-dependent liver lesions were observed. The same authors also demonstrated
cumulative liver damage after repeated dosing. While a single dose of 31.3 µg kg-1 bw
did not result in an increase in liver weight, repeated daily administration over a period of
seven days caused a 75 per cent increase in liver weight, which was almost as high as
the effect of a single intranasal dose of 500 µg kg-1 bw.
Repeated oral administration
Pure microcystin-LR. For assessing possible chronic human health effects, studies
involving repeated oral administration of pure microcystins at various dose levels are
most desirable. In a 13-week gavage study in mice, conducted under good laboratory
practice (GLP), pure microcystin-LR was administered orally to groups of 15 male and
female mice at 0, 40, 200, or 1,000 µg kg-1 bw per day for 13 weeks (Fawell et al., 1994).
The NOAEL was 40 µg kg-1 bw per day. At the next highest dose level there were slight
liver tissue changes in some male and female mice. At the highest dose level, all male
and most female mice showed liver changes, including chronic inflammation, focal
degeneration of hepatocytes and haemosiderin deposits. In male mice at the two highest
dose levels, serum transaminases were significantly elevated, serum gamma glutamyl
transferase was significantly reduced, and there were small but significant reductions in
total serum protein and serum albumin. In female mice, at the highest dose level, only
the changes in transaminases were observed (Fawell et al., 1994).
Microcystis extract. In an oral dosing study, Microcystis aeruginosa extract was supplied
to mice at five concentrations (equivalent to 750 to 12,000 µg kg-1 bw per day of
microcystin-YM) in their drinking water for up to one year. At the higher concentrations
there was increased death, increased bronchopneumonia (which was endemic in the
colony), and chronic liver injury. There was no evidence of tumourigenic changes in the
liver, despite the liver injury caused by chronic oral exposure to microcystins. No clear
NOAEL was established (Falconer et al., 1988).
In another study in pigs, Microcystis aeruginosa extract was given to groups of five pigs
in their drinking water for 44 days at dose levels calculated from potency estimates using
the mouse i.p. bioassay to be equivalent to 280, 800 or 1,310 µg kg-1 bw per day of
microcystins (assuming an average i.p. LD50 for microcystins of 100 µg kg-1 bw). The
extract contained at least seven microcystin variants, with microcystin-YR tentatively
identified by high pressure liquid chromatography (HPLC) as the major constituent. A
LOAEL of 280 µg kg-1 bw per day of toxins was identified, with general liver injury
(evident from histopathology and changes in serum enzymes) observed at the two
higher dose levels. At the lowest dose level, one pig was affected. The authors
determined the potency of their extract by the mouse i.p. LD50 bioassay, by HPLC
analysis and by the in vitro phosphatase inhibition assay (Falconer et al., 1994).
Summation of the peak areas from the HPLC identification of microcystin variants,
standardised against microcystin-LR (see Chapter 5), indicated that the LOAEL equated
with 100 µg microcystin-LR equivalents per kg bw per day.
Developmental effects
In an investigation on the effects of microcystin-LR on embryonic and foetal

development of the mouse, groups of 26 time-mated female mice were dosed once daily
by gavage with aqueous solutions of pure microcystin-LR from days 6 to 15 of
pregnancy, at dose levels of 0, 200, 600, or 2,000 µg kg-1 bw per day. Maternal clinical
signs, body weights and food consumption were recorded. Only treatment at 2,000 µg
kg-1 bw per day was associated with maternal toxicity and mortality. Seven of the 26
females died and two were sacrificed because of distress during the dosing period. On
day 18 of pregnancy the remaining females were killed. At the highest dose level, there
was a retardation of foetal weight and skeletal ossification, but no foetal deaths. Apart
from this there was no effect of microcystin-LR on sex ratio, implantation, post-
implantation loss, or on external or visceral or skeletal abnormalities. The NOAEL for
any aspect of developmental toxicity was 600 µg kg-1 bw per day (Fawell et al., 1994).
In a scoping study, eight 20-week old male and female mice which had received an
extract of Microcystis aeruginosa in their drinking water since weaning were mated;
extract exposure continued throughout pregnancy. Examination of litters showed no
effects on weight, number or sex ratio of offspring, compared with control litters.
However, there was some evidence of hippocampal injury and reduced brain size in 7
out of 73 of the five-day old young from parents who had received the Microcystis extract,
and in none of the controls (Falconer et al., 1988).
Carcinogenicity
Microcystin administered alone. In a recent study, microcystin-LR, administered i.p. to

mice, induced neoplastic liver nodules. Animals were given the toxin at 20 µg kg-1 bw,
100 times over 28 weeks. At autopsy, nodules up to 5 mm in diameter were observed in
the livers of all exposed mice. Some mice were kept a further two months after cessation
of dosing, and autopsy showed that liver nodules persisted (Ito et al., 1997b). In the
same study, mice orally administered microcystin-LR, at a dose level of 80 µg kg-1 bw,
100 times over the same time period showed no evidence of liver injury or nodule
formation. Previous work had shown that microcystin-LR given 20 times i.p. to mice, at
25 µg kg-1 bw, over 10 weeks did not induce liver nodules (Ohta et al., 1994).
Microcystin interaction with tumour initiators. It is generally understood that some

chemicals (usually DNA-damaging) can initiate the cancer process while other classes of
chemicals are able to promote the appearance of cancer after initiation has occurred
(Boutwell, 1974; Yamasaki, 1988; Fitzgerald and Yamasaki, 1990). Microcystins have
been tested for tumour promoting activity.
In vivo animal experiments. There has been some evidence of tumour promotion in
animal studies (see also section on mechanism of action). In a modified, two-stage
carcinogenesis mouse skin bioassay, a single dose of 7,12-dimethyl benzanthracene
(DMBA, 500 µg) was applied to the dorsal skin of groups of 20, three-month old, female
mice. After one week, groups of treated and control mice received either water alone,
water with Microcystis extract (80 mg microcystin-YM per litre; equivalent to 50 mg
microcystin-LR per litre), or croton oil (as a tumour-promoting phorbol ester-containing
positive control) applied to the skin (0.5 per cent in 0.1 ml acetone twice a week). At 52
days after DMBA exposure, there was a 1.6-fold increase in the number and a sevenfold
increase in the mean weight of skin papillomas (a type of benign tumour) per mouse in
animals given the cyanobacterial extract compared with those given water alone. It was
concluded by the authors that oral consumption of Microcystis extract from drinking
water may act directly or indirectly as a tumour promoter (Falconer, 1991). The
mechanism of action is not clear because microcystin-LR, at least, has difficulty
penetrating epidermal cells (Matsushima et al., 1990). In this study, there was
considerable liver damage, which could affect the interpretation of these findings
(Falconer, 1991). In a short-term two-stage carcinogenicity bioassay, groups of 10-19
seven-week old male F344 rats were initiated by i.p. injection with 200 mg kg-1 bw
diethylnitrosamine (DEN), followed by partial hepatectomy (performed to stimulate cell
division and thus increase the sensitivity of the assay). In two separate experiments,
twice-weekly i.p. doses of 1 and 10 µg kg-1 microcystin-LR and of 10,25 and 50 µg kg-1
microcystin-LR were then administered during eight weeks and this resulted in a dose-
dependent increase in the number (up to threefold) and area (up to 11-fold) of GST-P-
positive liver foci (GST-P is the placental form of glutathione-S transferase, a biomarker
for preneoplastic changes in liver (Sato et al., 1984)). The i.p. NOAEL in this study was 1
µg kg-1 (Nishiwaki-Matsushima et al., 1992). Microcystin-LR given without DEN initiation
showed no induction of liver foci. Macroscopic nodules, histologically diagnosed as
neoplastic nodules, were seen in this study when DEN was followed by microcystin-LR
at 10 µg kg-1 (before) and 50 µg kg-1 (after) partial hepatectomy (Nishiwaki-Matsushima et
al., 1992).
In another tumour initiation and promotion assay aimed at evaluating possible tumour
promoting effects in the upper small intestine, two doses of 40 mg kg-1 bw of the initiator
N-methyl-N-nitrosourea (MNU) were orally administered to C57 black mice, one week
apart, followed by drinking water containing various levels of Microcystis extracts,
estimated to be equivalent to 0, 1.2 or 4.2 mg kg-1 bw per day of microcystins for up to 22
weeks. Time to 50 per cent survival was about 17 weeks in controls and 15 weeks in
extract-exposed mice. No primary liver tumours were seen in any group and there was
no evidence of microcystin-induced promotion of lymphoid or duodenal tumours
(Falconer and Humpage, 1996).
In vitro studies on interaction of microcystins and tumour initiators are discussed in the
section on in vitro studies below.
Genotoxicity
No mutagenic response has been observed for purified toxins (exact nature not
determined) derived from Microcystis in the Ames Salmonella assay (strains TA98,
TA100 and TA102) with or without S9 activation. The Bacillus subtilis multigene
sporulation test was also negative with regard to mutagenicity using both the 168 and
hcr-9 strains (Repavich et al., 1990). An earlier study with purified Microcystis extract
also elicited a negative response in the Ames test (Runnegar and Falconer, 1982).
The purified toxins from a Microcystis species tested against human lymphocytes
suggested that the toxins may be clastogenic as indicated by dose-related increases in
chromosomal breakage (Repavich et al., 1990).
In vitro studies
Extracts from Microcystis aeruginosa blooms containing mainly microcystins (no

analytical data presented) from a lake in China were tested in a two-stage cell-
transformation assay using Syrian hamster embryo (SHE) cells. In this assay, the
microcystin extract had no initiating activity when followed by the tumour promoter TPA
(12-O-tetradecanoyl phorbol 13-acetate). With methylcholanthrene as the initiator (0.5
µg ml-1), followed by bloom extract, a dose-related (up to sevenfold) increase in
transformation frequency was observed (Wang and Zhu, 1996).
Primary hepatocyte cultures in the presence of picomolar and nanomolar concentrations

of microcystin-LR showed selective cell toxicity and selective cell proliferation depending
on the ploidy (chromosome copy number) of the cells (Humpage and Falconer, 1999).
Mechanism of action
Microcystin-LR was found to be a potent inhibitor of eukaryotic protein serine/threonine

phosphatases 1 and 2A both in vitro (Honkanen et al., 1990; MacKintosh et al., 1990)
and in vivo (Runnegar et al., 1993), and this effect has become the basis of one of the
bioassays to detect its presence. Substances that inhibit these protein phosphatases are
considered to be non-phorbol ester (TPA)-type tumour promoters. Other substances that
act in a similar way to microcystins are okadaic acid, nodularin, tautomycin and calyculin
(for a review see Fujiki and Suganuma, 1993). The protein phosphatases serve an
important regulatory role to maintain homeostasis in the cell (Cohen, 1989). Protein
phosphatase inhibition results in a shift in the balance towards higher phosphorylation of
target proteins, such as tumour suppresser proteins. This is a major post-translational
modification which can result in excessive signalling and may lead towards cell
proliferation, cell transformation and tumour promotion (Fujiki and Suganuma, 1993). In
liver cells, intermediate filaments of the cytoskeleton are hyperphosphorylated leading to
cellular disruption (Falconer and Yeung, 1992). In monolayer cell cultures this leads to
detachment from adjacent cells, and involves actin filament contraction (Hooser et al.,
1991, Ghosh et al., 1995). The inhibition of protein phosphatase 2A by microcystin-LR
can be effectively reversed in the presence of polyclonal antibodies against microcystin-
LR (Lin and Chu, 1994b); such antibodies can also protect in vivo against microcystin-
LR toxicity as shown with i.p. co-administration studies in mice (Nagata et al., 1995).
The implications of protein phosphatase inhibition in humans, due to low level chronic
exposure to microcystins, are not known.
Additional study requirements
Further short-term studies are needed to understand better the genotoxic and tumour
promoting potential of microcystins. Such studies would also be useful as a preliminary
to the design of appropriate chronic or other in vivo studies to assess their carcinogenic
potential. Regarding the possible role of microcystins in tumour promotion, further
studies are especially needed to establish a dose-effect relationship for nodule induction
with microcystin alone using various routes of administration.
Derivation of a tolerable daily intake (TDI)
Most of the relevant data on microcystin toxicity have come from either reported human
injury related to consumption of drinking water containing cyanobacteria (see section 4.2)
or from limited work with experimental animals (see above). Although the cyanotoxins
have not been reviewed by the International Agency for Research on Cancer (IARC),
their standard evaluation procedures (IARC, 1995) lead to the conclusion that, at present,
the human evidence for microcystin carcinogenicity is inadequate and the animal
evidence is limited. In such instances, the current practice for deriving a TDI is to use a
LOAEL or NOAEL divided by appropriate safety or uncertainty factors as described in
the Addendum to the Guidelines for Drinking Water Quality (WHO, 1998).
A 13-week mouse oral (by gavage) study with pure microcystin-LR has been considered
the most suitable for the derivation of a guideline value for microcystin-LR. In a study by
Fawell et al. (1994) a NOAEL of 40 µg kg-1 bw per day was determined, based on liver
histopathology and serum enzyme level changes. By applying a total uncertainty factor
of 1,000 (10 for intra-species variability, 10 for inter-species variability and 10 for
limitations in the database, in particular a lack of data on chronic toxicity and
carcinogenicity), a provisional TDI of 0.04 µg kg-1 bw per day has been determined for
microcystin-LR. This TDI is supported by the results of a 44-day pig study, in which pigs
were given Microcystis extract in their drinking water, resulting in a LOAEL of 100 µg
microcystin-LR equivalents per kg bw per day (Falconer et al., 1994; see above). To this
LOAEL an overall uncertainty factor of 1,500 was applied, arrived at by using 10 for
intra-species variability, 3 rather than 10 for inter-species variability (because pigs
physiologically resemble humans more closely than rodents), 5 for extrapolating from a
LOAEL to a NOAEL (10 was considered inappropriate due to the low incidence of effects
in the lowest dose group and the deduced shape of the dose-response curve) and 10 for
the less-than-lifetime exposure. This resulted in a provisional TDI of 0.067 µg kg-1 bw per
day. The lower of these two values, 0.04 µg kg-1 bw per day, has been used in deriving a
provisional guideline value (see Chapter 5).
4.2.2 Nodularin
Compared with Microcystis and microcystins, there have been fewer reported incidents
of human and livestock disease involving Nodularia spp. and nodularin (see section 4.1).
Experimental animal data
The toxicity and liver pathology induced by nodularin is similar to that caused by
microcystins, with hepatocyte necrosis and haemorrhagic diathesis (Runnegar et al.,
1988). Nodularin inhibits protein phosphatases 1 and 2A with the same potency as
microcystin-LR (Yoshizawa et al., 1990). In a two-stage liver carcinogenesis experiment
in male F344 rats initiated with DEN and without partial hepatectomy, repeated i.p.
administration of 10 µg nodularin per kg bw induced GST-P-positive foci more effectively
than microcystin-LR. Nodularin alone also induced some foci. In addition, nodularin was
capable of activating several proto-oncogenes of the fos and jun family, which are
considered to play a role in tumour promotion (Ohta et al., 1994).
Derivation of a tolerable daily intake
There are no studies available that use oral administration of nodularin, thus a TDI
cannot be set. Nevertheless, several experimental studies cited above indicate that
nodularin has similar toxicity to microcystin-LR. It may be appropriate, therefore, to
consider nodularin in an analogous fashion to microcystins in terms of human health risk
assessment.
4.2.3 Anatoxin-a
The potent neurotoxin, anatoxin-a, from Anabaena flos-aquae has frequently been
involved in animal and wildfowl poisoning (Ressom et al., 1994) (Table 4.1).
Experimental studies
Anatoxin-a is a nicotinic (cholinergic) agonist that binds to neuronal nicotinic

acetylcholine receptors. It has been suggested that the activation of presynaptic nicotinic
acetylcholine receptors by anatoxin-a results in an influx of Na+, producing sufficient local
depolarisation to open voltage sensitive Ca++ and Na+ channels. The latter may then
amplify the response, activating further Ca++ channels (Soliakov et al., 1995). As a result
of this depolarisation there is a block of further electrical transmission, and at sufficiently
high doses this can lead to paralysis, asphyxiation and death (Carmichael et al., 1975;
Carmichael, 1997). Anatoxin-a is more potent than nicotine or acetylcholine in evoking
type 1A or type 2 current responses in rat hippocampal neurones (Alkondon and
Albuquerque, 1995), and it is more potent than nicotine in its ability to evoke the
secretion of endogenous catecholamines from bovine adrenal chromaffin cells through
their neuronal-type nicotinic receptors (Molloy et al., 1995). Similar to nicotine, anatoxin-
a was more potent than noradrenaline in releasing dopamine from striatal nerve
terminals from rat superfused hippocampal synaptosomes (Clarke and Reuben, 1996).
In vivo studies in the rat showed that the toxin stimulates the sympathetic system
through the release of catecholamines from nerve endings (Dube et al., 1996).
Acute toxicity
In mice, the i.p. LD10 (lowest dose causing death) of anatoxin-a was 250 µg kg-1 bw
(Stevens and Krieger, 1989) and the i.p. LD50 of anatoxin-a is 375 µg kg-1 bw (Fitzgeorge
et al., 1994). The i.v. LD50 is less than 100 µg kg-1 bw (Fawell and James, 1994). The oral
LD50 for anatoxin-a is greater than 5,000 µg kg-1 bw, and the intranasal LD50 in mice is
2,000 µg kg-1 bw (Fitzgeorge et al., 1994).
Subacute toxicity
In two studies, anatoxin-a was administered to rats orally in the drinking water for 54
days at 0.51 or 5.1 ppm (equivalent to 51 and 510 µg kg-1 bw per day, respectively) or by
i.p. injection at 16 µg per rat per day for 21 days (equivalent to 80 µg kg-1 bw per day;
Astrachan et al., 1980). No toxicity was observed (no deaths due to treatment; no
significant changes in body weight gain, haematology or clinical chemistry).
In a dose range-finding study for a four-week oral toxicity experiment, groups of 2 male
and 2 female mice were administered anatoxin-a hydro-chloride by gavage at dose
levels of 1,500, 7,500 or 15,000 µg kg-1 per day for five days. All mice at the highest dose,
and one female at the intermediate dose, died within three minutes of their first dose.
Males at the intermediate dose were hyperactive after their third dose. All other animals
survived, and no treatment-related signs of clinical toxicity, or changes in body weight
were observed. No treatment-related changes were observed at necropsy (Fawell and
James, 1994).
In a 4-week study, groups of 10 male and female mice were administered anatoxin-a
hydrochloride by gavage at dose levels of 0, 120, 600 or 3,000 µg kg-1 bw. In the course
of the experiment, one male receiving 600 µg kg-1 bw per day and one female receiving
3,000 µg kg-1 per day died. No signs of clinical toxicity or histopathological abnormalities
were observed in these two animals, and no cause of death could be identified. It was
concluded by the authors that the possibility that these deaths were treatment-related
could not be excluded, although this was considered to be unlikely. There were no other
treatment-related findings. Although the NOAEL for anatoxin-a hydro-chloride in this
study was determined to be 120 µg kg-1 bw per day, equivalent to 100 µg kg-1 bw per day
of anatoxin-a, the toxicological database was considered to be inadequate to use these
results for setting a formal TDI.
Reproductive effects
Doses of 200 or 125 µg kg-1 bw anatoxin-a were given i.p. to hamsters one or three times
per day, respectively, at days 12 to 14 of pregnancy (after organo-genesis), and the
dams were sacrificed at day 15. The treatment given three times per day caused foetal
malformation (hydrocephaly) in all foetuses in one of 10 litters, and stunted growth in
almost all litters; treatment given once per day resulted in stunted growth. No maternal
toxicity was observed (Astrachan et al., 1980).
Groups of 10 and 12 time-mated female mice were given anatoxin-a hydrochloride by

gavage at 0 or 3,000 µg kg-1 bw per day (equivalent to 2,460 µg kg-1 bw per day anatoxin-
a) respectively, on days 6 to 15 of pregnancy. The anatoxin-a dose was considered the
maximum dose that could be tolerated. No treatment-related effects were observed in
the dams or offspring, although there was a slight decrease in foetal weight compared
with controls (Fawell and James, 1994). No data on the genotoxic potential of anatoxin-a
are available.
Anatoxin-a is a potent acute neurotoxin. The available data indicate that significant
concern with regard to chronic toxicity (Fawell and James, 1994) is unlikely, but the data
base is insufficient for derivation of a TDI.
4.2.4 Homoanatoxin-a
Homoanatoxin-a is a potent neuromuscular blocking agent with an i.p. LD50 in mice of

250 µg kg-1 bw. Toxicosis in the lethal dose range leads to severe body paralysis,
convulsions and death by respiratory arrest in 7-12 minutes. Experiments with rat
phrenic nerve hemidiaphragm preparations demonstrated that the physiological effects
of homoanatoxin-a are related to those observed for d-tubocurarine (Skulberg et al.
1992). Recent studies have shown that homoanatoxin-a enhances the influx of Ca++ ions
in the cholinergic nerve terminals (Aas et al., 1996). No TDI can be derived.
4.2.5 Anatoxin-a(S)
Anatoxin-a(S) is an organophosphate produced by the cyanobacteria Anabaena flos-

aquae and A. lemmermannii (see Chapter 3). This toxin blocks acetylcholinesterase
activity in a manner analogous to organophosphate insecticides. The i.p. LD50 in mice is
20 µg kg-1 bw with muscle weakness, respiratory distress (dyspnea) and convulsions
(effect on seizure threshold) preceding death (Mahmood and Carmichael, 1986a;
Matsunaga et al. 1989). There are no oral toxicity studies for this toxin.
A lack of adequate experimental data or human data precludes the derivation of a TDI.
4.2.6 Saxitoxins
Effects in humans
Saxitoxin and some of its analogues are produced by Anabaena circinalis in Australian
freshwaters and Aphanizomenon flos-aquae in the USA (Sawyer et al., 1968; Mahmood
and Carmichael, 1986b; Humpage et al., 1994). The saxitoxin group has been the cause
of paralytic shellfish poisoning (PSP) in people. Several species of dinoflagellates
produce PSP toxins that accumulate in molluscs which filter-feed on these organisms.
People who have consumed shellfish containing high levels of PSP toxins may suffer
from this acute illness. The signs and symptoms of PSP in humans may range from a
slight tingling and numbness about the lips to complete paralysis and death from
respiratory failure (Meyer et al., 1928; Medcof et al., 1947; McFarren et al., 1958). More
than 1,000 cases of PSP have been reported since the early 1900s in North and Central
America and there have been 109 deaths. In a recent epidemic in Guatemala, the
mortality rate in children was 50 per cent, while for adults it was 5 per cent (Rodrigue et
al. 1990; for a review see Kao, 1993). No PSP-like illnesses have been reported in
humans from the consumption of drinking water containing saxitoxins, although there
have been fatalities in sheep after ingesting a PSP toxin-containing bloom of A. circinalis
(Negri et al., 1995).
Experimental studies
Mode of action. Of the various PSP toxins, only saxitoxin has been studied in detail for
pharmacological effects, partly because the other toxins are usually not available in
sufficient quantities for such studies. Nearly all the systemic actions of saxitoxin can be
explained by its pharmacological effect on nerve axon membranes. This involves a wide
spread blockage of sodium ion channels of the excitable membranes of nerves, thereby
affecting (partially or completely, depending on dose) impulse generation in peripheral
nerves and skeletal muscles (Catterall, 1980). This results in generalised nerve
dysfunction as measured by electromyography (Easthaugh and Shepherd, 1989). In
mammals, these effects lead to paralysis, respiratory depression and respiratory failure.
Direct cardiac effects are usually minimal (see review by Kao, 1993).
Metabolism. No data are available on PSP toxin absorption, distribution, metabolism and
excretion.
Table 4.2 LD50 values following a single dose of PSP toxin extract in the mouse in
relation to the route of administration
Route of administration LD501 (µg PSP kg-1 body weight)

Male Female
Intravenous 3.4 (3.2-3.6)
Intraperitoneal 10.0 (9.7-10.5) 8.0 (7.6-8.6)
Oral 263.0 (251-267)
PSP Paralytic shellfish poison
Source: IPCS, 1984; adapted from Wiberg and Stephenson, 1960
1
The 95% confidence limits are given in parentheses
Acute toxicity
The principle of the standardised mouse bioassay developed by Sommer and Meyer
(1937) is measurement of time to death after i.p. injection. In that study, the authors
suggested that signs characteristic of PSP, such as dyspnea, could be observed after
the i.p. administration of toxin. Hypotensive effects have been observed to accompany
the respiratory depression, implicating both central and peripheral nervous system
actions (Watts et al., 1966).
Acute toxicity studies have been conducted in several species with extracts containing
PSP toxins obtained from the Alaskan butter clam (Genenah and Shimizu, 1981). Using
this preparation, Wiberg and Stephenson (1960) determined the LD50 values for male
mice, using three routes of administration (i.v., i.p. and oral). In addition, the i.p. LD50
was determined in female mice (Table 4.2). The PSP toxin extract was much less toxic
when administered by the oral route than by the i.v. or i.p. routes (Table 4.2). Increasing
the pH of the injection medium (> 3.8) or the addition of sodium ions affected i.p. toxicity,
while the addition of sodium ions did not influence oral or i.v. toxicity. Similar
comparative LD50s for different routes of administration were obtained in rats; in addition
new-born rats were about tenfold more susceptible than adults after oral administration
(Watts et al., 1966) (Table 4.3). For a number of animal species tested, the oral LD50 for
PSP toxin ranged from 128 µg kg-1 bw PSP toxin in guinea pigs to 420 µg kg-1 bw in mice
(IPCS, 1984).
Table 4.3 LD50 values following oral or intraperitoneal administration of a single dose of
PSP toxin extract to rats of different ages
Age LD501 (µg PSP kg-1 body weight)

Oral Intraperitoneal
New-born (24 hours) 64 (51-80) 5.5 (4.7-6.5)
Weanling (21 days) 270 (204-356) 8.3 (7.7-9.0)
Adult (60-70 days) 531 (490-576) 10.0 (8.5-11.8)
1
The 95% confidence limits are given in parentheses
Source: IPCS, 1984; adapted from Watts et al., 1966
Prior exposure to non-lethal doses of PSP toxin seems to lower the susceptibility of rats
to lethal doses of PSP toxin. In a study using Sprague-Dawley rats (sex not indicated),
the oral LD50 value for the purified PSP extract was determined (McFarren et al., 1958).
One group of rats was given a non-lethal dose of PSP toxin (about one-third of the LD50)
14 days before the test. The LD50 for the pre-treated rats was about 50 per cent higher
than that for untreated rats. This finding corroborates the fact noted by Prakash et al.
(1971) that fishermen who habitually eat shellfish containing low levels of PSP toxins
appear to be less susceptible to developing PSP.
There is a wide range in i.p. toxicity of the various PSP toxins (i.e. saxitoxin,
neosaxitoxin, the gonyautoxins and C toxins) as tested following the Association of
Official Analytical Chemists (AOAC) mouse bioassay (AOAC, 1984). Potencies of these
toxins are usually expressed in mouse units (MU) per µmol of toxin. Saxitoxin is one of
the most toxic of the PSP toxins (2,483 MU per µmol) and the C toxins are the least toxic
(15-143 MU per µmol) (Oshima, 1995). Potencies may also be expressed as saxitoxin
equivalents. Inter-conversions during storage, cooking or digestion may modify the i.p.
and oral toxicity (Humpage et al., 1994; see also section 3.4.2). No subchronic or
chronic animal studies with PSP toxins are available.
The animal toxicity data for the saxitoxins are inadequate to set a TDI.
4.2.7 Cylindrospermopsin
This cyanotoxin was initially isolated from a culture of Cylindrospermopsis raciborskii

obtained from a water supply reservoir in tropical northern Australia. The organism was
identified as a result of an outbreak of acute hepato-enteritis and renal damage among
an Aboriginal population on Palm Island, off the coast of North Queensland (Hawkins et
al., 1985) (see Box 4.2). Intraperitoneal injection of the lysed organism to mice resulted
in widespread and progressive tissue injury, with cell necrosis in the liver, kidneys,
adrenals, lung, heart, spleen and thymus (Hawkins et al., 1985, 1997). In mice, the i.p.
LD50 at 24 hours was 52 mg dry weight (dw) of cells per kg bw, equivalent to 300 µg kg-1
bw of cylindrospermopsin, whereas the i.p. LD50 at seven days was approximately 32 mg
cells per kg bw, equivalent to 180 µg kg-1 bw of toxin (Hawkins et al., 1997).
Administration of the pure toxin to mice showed this delayed toxicity more clearly, with
the 24-hour i.p. LD50 being 2,100 µg kg-1 bw and the 5-6 day i.p. LD50 being 200 µg kg-1
bw (Ohtani et al., 1992). In vitro studies with pure cylindrospermopsin have shown that it
inhibits glutathione synthesis and protein synthesis in general (Runnegar et al., 1994,
1995; Terao et al., 1994). In mouse liver after i.p. administration, major changes were
seen in hepatocytes, with progressive proliferation of the smooth endoplasmic reticulum
and accumulation of lipid over five days (Terao et al., 1994). No data on the oral toxicity
of pure cylindrospermopsin are available, but studies with aqueous extracts of
Cylindrospermopsis provide a preliminary indication. After administering to mice a single
oral dose of an aqueous extract of freeze-dried Cylindrospermopsis cells, a median
lethal dose in the range of 4.4 to 6.9 mg kg-1 toxin equivalent was determined (Seawright
et al., 1999). Because the i.p. LD50 is 0.2 mg kg-1 and the LD100 is likely to be double that,
the oral toxicity appears to be over tenfold lower than i.p. toxicity.
At present it is not known if cylindrospermopsin is the only toxin in Cylindrospermopsis,

because the major kidney damage seen on i.p. dosing of some toxic extracts is not
similarly observed after administration of pure toxin.

On the basis of present data it is not possible to derive a TDI.
4.2.8 Other cyanotoxins produced by marine cyanobacteria
Swimmers off Hawaii and Okinawa who have come into contact with the marine
cyanobacterium Lyngbya majuscula have contracted acute dermatitis, causing
"swimmers' itch" (Moikeha and Chu, 1971; Hashimoto et al., 1976). The effect is a
cutaneous inflammation with signs of erythema, blisters and desquamation within 12
hours of exposure to the cyanobacterium. The inflammatory activity of Lyngbya is
caused by aplysiatoxins and debromoaplysiatoxin (Mynderse et al., 1977). These toxins
are potent animal skin tumour promoters and protein kinase C activators (Fujiki et al.,
1990). Aplysiatoxins are toxic to mice at a minimum lethal dose of about 0.3 mg kg-1
(Moore, 1977). Debromoaplysiatoxin, along with other toxic compounds, has also been
isolated from other Oscillatoriaceae such as Schizothrix calcicola and Oscillatoria
nigroviridis. The chemically different lyngbyatoxin-a (Cardellina et al., 1979), found in
another shallow water strain of marine Lyngbya majuscula, has caused dermatitis and
severe oral and gastrointestinal inflammation (Moore et al., 1993). Its toxicity to mouse
corresponds to aplysiatoxins (Moore, 1977) and it has skin tumour promoting activity
similar to the well known tumour promoter, TPA (Fujiki et al., 1981, 1984). Lyngbya
majuscula sometimes grows epiphytically on edible algae, such as Acanthophora
spicifera which is eaten in Indonesia and in the Philippines. Cyanobacteria have also
been suspected to be a source of toxins in the ciguatera food chain that may lead to
human poisoning (Hahn and Capra, 1992; Endean et al., 1993).
4.2.9 Cyanobacterial lipopolysaccharides
Lipopolysaccharides, cell wall components of gram-negative bacteria, are pyrogenic

(fever-causing agents) and toxic (Weckesser and Drews, 1979). In Sewickley,
Pennsylvania, an outbreak of gastro-enteritis is suspected to have been caused by
cyanobacterial LPS (Lippy and Erb, 1976; Keleti et al., 1979). The few studies carried
out on cyanobacterial LPS indicate that they are less toxic than the LPS of other bacteria
such as Salmonella (Keleti and Sykora, 1982; Raziuddin et al., 1983). Lack of axenic
cyanobacterial strains has previously hindered detailed studies on structures and
toxicities of cyanobacterial LPS.
4.2.10 Other bioactive compounds in cyanobacteria
Cyanobacteria produce a wide variety of bioactive compounds in addition to the

cyanotoxins described in this chapter. They include anti-tumour (cytotoxic), anti-viral,
and anti-fungal compounds, antibiotics and protease inhibitors (Moore et al., 1996;
Namikoshi and Rinehart, 1996). Further screening of these biomedically interesting
compounds is underway and is likely to lead to the discovery of many new compounds in
the future, some of which may be toxic. Bioassays of cyanobacterial cell extracts have
often revealed a higher toxicity than expected from the content of known toxins in the
extract.
4.3 References
Aas, P., Eriksen, S., Kolderup, J., Lundy, P., Haugen, J-E., Skulberg, O.M., Fonnum, F.
1996 Enhancement of acetylcholine release by homoanatoxin-a from Oscillatoria
formosa. Environ. Toxicol. Pharmacol., 2, 223-232.
Alkondon, M. and Albuquerque, E. X. 1995 Diversity of nicotinic acetylcholine receptors

in rat hippocampal neurons. III. Agonist actions of the novel alkaloid epibatidine and
analysis of type II current. J. Pharmacol. Exp. Ther., 274(2), 771-782.
AOAC 1984 Official Methods of Analysis. S. Williams [Ed.] Association of Official

Analytical Chemists, Arlington, VA.
Astrachan, N.B., Archer, B.G. and Hilbelink, D.R. 1980 Evaluation of the sub-acute
toxicity and teratogenicity of anatoxin-a. Toxicon, 18, 684-688.
Bagu, J.R., Sykes, B., Craig, M., Holmes, C.F.B. 1997 A molecular basis for different
interactions of marine toxins with protein phosphatase 1. J. Biol. Chem., 272, 5087-5097.
Botes, D.P., Wessels, P.L., Kruger, H., Runnegar, M.T.C., Santikarn, S., Smith, R.J.,
Barna, J.C.J. and Williams, D.H. 1985 Structural studies on cyanoginosins-LR, -YR, -YA,
and -YM, peptide toxins of Microcystis aeruginosa. J. Chem. Society Perkin Trans., 1,
2747-2748.
Boutwell, R.K. 1974 The function and mechanism of promoters in carcinogenesis. CRC
Crit. Rev. Toxicol., 2, 419-443.
Bourke, A.T.C., Hawes, R.B., Neilson, A. and Stallman, N.D. 1983 An outbreak of
hepato-enteritis (the Palm Island Mystery Disease) possibly caused by algal intoxication.
Toxicon, 3, Supplement, 45-48.
Brooks, W.P. and Codd, G.A. 1987 Distribution of Microcystis aeruginosa peptide toxin
and interactions with hepatic microsomes in mice. Pharmacol. Toxicol., 60(3), 187-191.
Bury, N.R., Eddy, F.B. and Codd, G.A. 1995 The effects of the cyanobacterium
Microcystis aeruginosa, the cyanobacterial hepatotoxin microcystin-LR, and ammonia on
growth rate and ionic regulation of brown trout. J. Fish Biol., 46(6), 1042-1054.
Byth, S. 1980 Palm Island mystery disease. Med. J. Aust., 2, 40-42.
Carbis, C.R., Waldron, D.L., Mitchell, G.F., Anderson J.W. and McCauley, I. 1995
Recovery of hepatic function and latent mortalities in sheep exposed to the blue-green
alga Microcystis aeruginosa. Vet. Rec., 137(1), 12-15.
Cardellina, II J.H., Marner, F.J. and Moore, R.E. 1979 Seaweed dermatitis: structure of
lyngbyatoxin A. Science, 204, 193-195.
Carmichael, W. 1997 The cyanotoxins. Adv. Botan. Res., 27, 211-256.
Carmichael, W.W. 1992 A Status Report on Planktonic Cyanobacteria (Blue Green

Algae) and their Toxins. EPA/600/R-92/079, Environmental Monitoring Systems
Laboratory, Office of Research and Development, US Environmental Protection Agency,
Cincinnati, Ohio.
Carmichael, W. and Gorham, P. 1978 Anatoxins from clones of Anabaena flos-aquae

isolated from lakes of Western Canada. Mitt. Int. Ver. Limnol, 21, 285-295.
Carmichael, W.W., Biggs, D. and German, P. 1975 Toxicology and pharmacological

action of Anabaena flos-aquae toxin. Science, 187, 542-544.
Catterall, W. 1980 Neurotoxins that act on voltage-sensitive sodium channels in

excitable membranes. Ann. Rev. Pharmacol. Toxicol., 20, 15-43.
Clarke, P.B.S. and Reuben, M. 1996 Release of [3H]-noradrenaline from rat hippocampal
synaptosomes by nicotine: mediation by different nicotinic receptor subtypes from striatal
[3H]-dopamine release. Brit. J. Pharmacol., 117, 595-606.
Cohen, P. 1989 The structure and regulation of protein phosphatases. Ann. Rev.
Biochem. 58, 453-508.
Cohen, S.G. and Reif, C.B. 1953 Cutaneous sensitization to blue-green algae. J. Allergy.
24, 452-457.
Dillenberg, H.O. and Dehnel, M.K. 1960 Toxic water bloom in Saskatchewan 1959. Can.
Med. Assoc. J., 83, 1151-1154.
Dube, S.N., Mazumder, P.K., Kumar, D., Rao, P.V.L. and Bhaskar, A.S.B. 1996
Cardiorespiratory and neuromuscular effects of freshwater cyanophyte Anabena flos
aquae in rats. Def. Sci. J., 46(3), 135-141.
Easthaugh, J. and Shepherd, S. 1989 Infectious and toxic syndromes from fish and
shellfish consumption. Arch. Intern. Med., 149, 1735-1740.
Endean, R., Monks, S.A., Griffith, J K. and Llewellyn, L.E. 1993 Apparent relationship
between toxins elaborated by the cyanobacterium Trichodesmium erythraeum and those
present in the flesh of the narrow-barred Spanish mackerel Scomberomorus
commersoni. Toxicon., 31, 1155-1165.
Eriksson. J.E, Meriluoto, J. and Lindholm, T. 1986 Can cyanobacterial toxins accumulate
in aquatic food chains? In: Proceedings of the 4th International Symposium of Microbiol
Ecology, Ljubljana (Yugoslavia), 658-658.
Falconer, I.R. 1991 Tumor promotion and liver injury caused by oral consumption of
cyanobacteria. Environ. Toxicol. Water Qual., 6(2), 177-184.
Falconer, I.R. 1994 Health problems from exposure to cyanobacteria and proposed
guidelines for drinking and recreational water. In: G.A. Codd, T.M. Jefferies, C.W. Keevil
and E. Potter [Eds] Detection methods for Cyanobacterial Toxins. Royal Society of
Chemistry, Cambridge, 3-10.
Falconer, I.R. and Humpage, A.R. 1996 Tumour promotion by cyanobacterial toxins.
Phycologia, 35(6 supplement), 74-79.
Falconer, I.R. and Yeung, S.K. 1992 Cytoskeletal changes in hepatocytes induced by
Microcystis toxins and their relation to hyperphosphorylation of cell proteins. Chem. Biol.
Interact., 81, 181-196.
Falconer, I.R., Beresford, A.M. and Runnegar, M.T.C. 1983 Evidence of liver damage by
toxin from a bloom of the blue-green alga, Microcystis aeruginosa. Med. J. Aust., 1, 511-
514.
Falconer, I.R., Buckley, T. and Runnegar, M.T. 1986 Biological half-life, organ
distribution and excretion of 125-I-labelled toxic peptide from the blue-green algae
Microcystis aeruginosa. Aust. J. Biol. Sci., 39(1), 17-21.
Falconer, I.R., Burch, M.D., Steffensen, D.A., Choice, M. and Coverdale, O.R. 1994
Toxicity of the blue-green alga (cyanobacterium) Microcystis aeruginosa in drinking
water to growing pigs, as an animal model for human injury and risk assessment. J.
Environ. Toxicol. Water Qual. 9, 131-139.
Falconer, I.R., Choice, A. and Hosja, W. 1992 Toxicity of edible mussels (Mytilus edulis)
growing naturally in an estuary during a waterbloom of the blue-green alga Nodularia
spumigena. J. Environ. Toxicol. Water Qual. 7, 119-123.
Falconer, I.R, Hardy, S.J., Humpage, A.R., Froscio, S M., Tozer, G.J. and Hawkins, P.R.,
1999 Hepatic and renal toxicity of the blue-green alga (cyanobacterium)
Cylindrospermopsis raciborskii in male Swiss Albino mice. J. Environ. Toxicol. Water
Qual., 14(1), In Press.
Falconer, I.R., Jackson A.R.B., Langley, J. and Runnegar, M. T. 1981 Liver pathology in
mice in poisoning by the blue-green alga Microcystis aeruginosa. Aust. J. Biol. Sci,. 34,
174-187.
Falconer, I.R., Smith, J.V., Jackson, A.R.B., Jones, A. and Runnegar, M.T.C. 1988 Oral
toxicity of abloom of the cyanobacterium Microcystis aeruginosa administered to mice
over periods up to 1 year. J. Toxicol. Environ. Health, 24, 291-305.
Fawell, J.K. 1993 Toxins from blue-green algae: toxicological assessment of

microcystin-LR Volume 4. Microcystin-LR: 13 week oral (gavage) toxicity study in the
mouse (final report), Water Research Centre, Medmenham, UK, 1-259.
Fawell, J.K. and James, H.A. 1994 Report No. FR 0434/DoE 3728. Alien House, The
Listons, Listen Road, Marlow, Bucks, SL7 1FD, UK.
Fawell, J.K., Hart, J., James, H.A. and Parr, W. 1993 Blue-green algae and their toxins-
analysis, toxicity, treatment and environmental control. Water Supply, 11(3/4), 109-121.
Fawell, J.K., James, C.P. and James, H.A. 1994 Toxins from Blue-Green Algae:
Toxicological Assessment of Microcystin-LR and a Method for its Determination in Water,
Water Research Centre, Medmenham, UK, 1-46.
Fitzgerald, D.J. and Yamasaki, H. 1990 Tumor promotion: models and assay systems.
Teratogen, Carcinogen. Mutagen., 10, 89-102.
Fitzgeorge, R.B., Clark, S.A., and Keevil, C.W. 1994 Routes of intoxication. In: G.A.
Codd, T. M. Jeffries, C.W. Keevil and E. Potter [Eds] 1st International Symposium on
Detection Methods for Cyanobacterial (Blue-Green Algal) Toxins, Royal Society of
Chemistry, Cambridge, UK, 69-74.
Francis, G. 1978 Poisonous Australian lake. Nature, 18, 11-12.
Fujiki, H., Mori, M., Nakayasu, M., Tereda, M., Sugimura, T. and Moore, R.E. 1981
Indole alkaloids: dihydroteleocidin B, teleocidin, and lyngbyatoxin-A as members of a
new class of tumor promoters. Proc. Natl. Acad. Sci (USA)., 78, 3872-3876.
Fujiki, H., Suganuma, M., Hakii, H., Bartolini, G., Moore, R.E., Takayama, S. and
Sugimura, T. 1984 A 2-stage mouse skin carcinogenesis study of lyngbyatoxin-A. J.
Cancer Res. Clin. Oncol., 108(1), 174-176.
Fujiki, H. and Suganuma, M. 1993 Tumor promotion by inhibitors of protein

phosphatases 1 and 2A: the okadaic acid class of compounds. Adv. Cancer Res., 61,
143-194.
Fujiki, H., Suganuma, M., Suguri, H., Yoshizawa, S., Takagi, K., Nakayasu, M., Ojika, M.,
Yamada, K., Yasumoto, T., Moore, R.E. and Sugimura, T. 1990 New tumor promoters
from marine natural products. In: S. Hall and G. Strichartz [Eds] Marine Toxins: Origin,
Structure and Molecular Pharmacology, American Chemical Society, Washington D.C.,
232-240.
Genenah, A. and Shimizu, Y. 1981 Specific toxicity of paralytic shellfish poisons. J. Agric.
Food Chem., 29(6), 1289-1291.
Ghosh S., Khan, S.A., Wickstrom, M., Beasley, V. 1995 Effects of microcystin-LR on
actin and the actin-associated proteins alpha- actinin and talin in hepatocytes. Nat.
Toxins, 3(6), 405-414.
Grauer, F. 1961 Seaweed dermatitis. Arch. Dermatol., 84, 720-732.
Gunn, G., Rafferty, A., Rafferty, G., Cockburn, N., Edwards, C., Beatty, K. and Codd, G.
1992 Fatal canine neurotoxicosis attributed to blue green algae (cyanobacteria). Vet.
Rec., April 4, 301-302.
Hahn, S.T. and Capra, M. F. 1992 The cyanobacterium Oscillatoria erythraea - a

potential source of toxin in the ciguatera food-chain. Food Add. Contam., 9, 351-355.
Hashimoto, Y., Kamiya, H., Yamazato, K. and Nozawa, K. 1976 Occurrence of a toxic
blue-green alga inducing skin dermatitis in Okinawa. In: A. Ohsaka, K. Hayashi and Y.
Sawai [Eds] Animal. Plant, and Microbial Toxins. Plenum Publishing, New York, 333-338.
Hawkins, P.R. and Griffiths, D. J. 1993 Artificial destratification of a small tropical

reservoir: effects upon the phytoplankton. Hydrobiologia, 254, 169-181.
35, 341-346.
Hawkins, P.R., Runnegar, M.T.C., Jackson, A.R.B. and Falconer, I. R. 1985 Severe
hepatotoxicity caused by the tropical cyanobacterium (blue-green alga)
Cylindrospermopsis raciborskii (Woloszynska) Seenaya and Subba Raju isolated from a
domestic water supply reservoir. Appl. Environ. Microbiol., 50(5), 1292-1295.
Henriksen, P., Carmichael, W.W., An, J. and Moestrup, Ø.1997 Detection of an

anatoxin-a(s)-like anticholinesterase in natural blooms and cultures of
cyanobacteria/blue-green algae from Danish lakes and in the stomach contents of
poisoned birds. Toxicon, 35, 901-913.
Holmes, C.F.B., Craig, M., McCready, T.L., Boland, M.P., Dawson, J.F., Chen, D.Z.X.,
Wang, K., Klix, H., Luu, H. A. et al., 1994 Regulation of signal transduction pathways by
peptide toxins. In: R.S. Hodges,. and J.A. Smith [Eds] Pept. Chem. Struct. Biol., Proc.
Am. Pept. Symp., 13th, ESCOM: Leiden, Netherlands, 577-579.
Honkanen, R.E., Zwiller, J., Moore, R.E., Daily, S L., Khatra, B.S., Dukelow, M. and
Boynton, A. L. 1990 Characterization of microcystin- LR, a potent inhibitor of type 1 and
type 2A protein phosphatases. J. Biol. Chem., 265, 19401-19404.
Hooser, S.B., Beasley, V.R., Basgall, E.J., Carmichael, W.W. and Haschek, W.M. 1990
Microcystin-LR-induced ultrastructural changes in rats. Vet. Pathol., 27(1), 9-15.
Hooser, S.B., Beasley, V.R., Waite, L.L., Kuhlenschmidt, M.S., Carmichael, W.W. and
Haschek, W.M. 1991 Actin filament alteration in rat hepatocytes induced in vivo and in
vitro by microcystin-LR, a hepatotoxin from the blue-green alga, Microcystis aeruginosa.
Vet. Pathol., 28, 259-266.
Humpage, A.R, Falconer, I.R. 1999 Microcystin-LR and liver tumour promotion: effects
on cytokinesis, ploidy, and apoptosis in cultured hepatocytes. Env. Toxicol. Water Qual.,
14(1), In Press.
Humpage, A.R., Rositano, J., Bretag, A.H., Brown, R., Baker, P.D., Nicholson, B. C. and
Steffensen, D.A. 1994 Paralytic shellfish poisons from Australian cyanobacterial blooms.
Aust. J. Mar. Freshwater Res., 45(5), 761-771.
IARC 1995 IARC Monographs on the Evaluation of Carcinogenic Risks to Humans,

Volume 63, Dry cleaning, some chlorinated solvents and other industrial chemicals.
Preamble 12: Evaluation, International Agency for Research on Cancer, Lyon, 22-26.
IPCS 1984 Environmental Health Criteria 37. Aquatic Marine and Freshwater Biotoxins.
International Programme on Chemical Safety, World Health Organization, Geneva.
IPCS 1994 Environmental Health Criteria 170. Assessing Human Health Risks of
Chemicals; Derivation of Guidance Values for Health-based Exposure Limits.
Ito, E., Kondo, F. and Harada, K.-I. 1997a Hepatic necrosis in aged mice by oral
administration of microcystin-LR. Toxicon, 35(2), 231-239.
Ito, E., Kondo, F., Terao, K. and Harada, K.-I. 1997b Neoplastic nodular formation in
mouse liver induced by repeated intraperitoneal injections of microcystin-LR. Toxicon,
35(9), 1453-1457.
Jackson, A., McInnes, A., Falconer, I. and Runnegar, M. 1984 Clinical and pathological
changes in sheep experimentally poisoned by the blue-green algae Microcystis
aeruginosa. Vet. Pathol., 21, 102-113.
Jochimsen, E.M., Carmichael, W.W., An, J., Cardo, D.M., Cookson, S.T., Holmes,
C.E.M., Antunes, M.B. de C., Filho, D.A. de M., Lyra, T.M., Barreto, V.S.T., Azevedo,
S.M.F.O. and Jarvis, W. R. 1998 Liver failure and death after exposure to microcystins at
a haemodialysis center in Brazil. New Engl. J. Med., 338(13), 873-878.
Kao, C.Y. 1993 Paralytic shellfish poisoning. In: I. Falconer [Ed.] Algal Toxins in Seafood
and Drinking Water, Academic Press, London, 75-86.
Keleti, G. and Sykora, J.L. 1982 Production and properties of cyanobacterial endotoxins.
Appl. Environ. Microbiol., 43, 104-109.
Keleti, G., Sykora, J.L., Libby, E.C. and Shapiro, M.A. 1979 Composition and biological
properties of lipopolysaccharides isolated from Schizothrix calcicola (Ag.) Gomont
(cyanobacteria). Appl. Environ. Microbiol., 38, 471-477.
Kondo, F., Matsumoto, H., Yamada, S., Ishikawa, N., Ito, E., Nagata, S., Ueno, Y.,
Suzuki, M. and Harada, K.-I. 1996 Detection and identification of metabolites of
microcystins formed in vivo in mouse and rat livers. Chem. Res. Toxicol., 9(8), 1355-
1359.
Kotak, B.G., Kenefick, S.L., Fritz, D.L., Rousseaux, C.G., Prepas, E.E. and Hrudey, S.E.
1993 Occurrence and toxicological evaluation of cyanobacterial toxins in Alberta lakes
and farm dugouts. Water Res., 27(3), 495-506.
Kuiper-Goodman, T., Gupta, S., Combley, H. and Thomas, B.H. 1994 Microcystins in
drinking water: risk assessment and derivation of a possible guidance value for drinking
water. In: D.A. Steffensen and B.C. Nicholson [Eds] Toxic Cyanobacteria - A Global
Perspective, Australian Centre for Water Quality Research, Salisbury, 17-23.
Lin, J.R. and Chu, F.S. 1994a Kinetics of distribution of microcystin LR in serum and
liver cytosol of mice: An immunochemical analysis. J. Agric. Food Chem., 42(4), 1035-
1040.
Lin, J.R. and Chu, F. S. 1994b In vitro neutralization of the inhibitory effect of
Microcystin-LR to protein phosphatase 2A by antibody against the toxin. Toxicon, 32(5),
605-613.
Lippy, E.C. and Erb, J. 1976 Gastrointestinal illness at Sewickley, PA. J. Am. Water
Works Assoc., 68, 606-610.
MacKintosh, C., Beattie, K.A., Klumpp, S., Cohen, P. and Codd, G.A. 1990
Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases
1 and 2A from both mammals and higher plants. Fed. Eur. Biochem. Soc. Lett., 264(2),
187-192.
Mahmood, N.A. and Carmichael, W.W. 1986a The pharmacology of anatoxin-a(s), a

neurotoxin produced by the freshwater cyanobacterium Anabaena flos-aquae NRC 525-
17. Toxicon, 24(5), 425-434.
Mahmood, N.A. and Carmichael, W.W. 1986b Paralytic shellfish poisons produced by
the freshwater cyanobacterium Aphanizomenon flos-aquae nh-5. Toxicon, 24(2), 175-
186.
Mahmood, N.A., Carmichael, W.W. and Pfahler, D. 1988 Anticholinesterase poisonings

in dogs from a cyanobacterial (blue-green algae) bloom dominated by Anabaena flos-
aquae. Am. J. Vet. Res., 49(4), 500-503.
Matsunaga, S., Moore, R.E., Niemczura, W.P. and Carmichael, W.W. 1989 Anatoxin-a(s)
a potent anticholinesterase from Anabaena flos-aquae. J. Am. Chem. Soc., 111(20),
8021-8023.
Matsushima, R., Yoshizawa, S., Watanabe, M.F., Harada, K., Furusawa, M., Carmichael,
W W. and Fujiki, H. 1990 In vitro and in vivo effects of protein phosphatase inhibitors,
microcystins and nodularin, on mouse skin and fibroblasts. Biochem. Biophys. Res.
Comm., 171(2), 867-874.
McBarron, E. and May, V. 1966 Poisoning of sheep in New South Wales by the blue-
green alga Anacystis cyanea (Kuetz.) Dr. and Dail. Aust. Vet. J., 42, 449-453.
McFarren, E., Schafer, M., Cambell, J. and Lewis, K. 1958 Public health significance of
paralytic shellfish poisons. Adv. Food Res., 10, 135-179.
Medcoff, J., Leim, A., Needler, A., Needler, A., Gibbard, J. and Naubert, J. 1947
Paralytic shellfish poisoning on the Canadian Atlantic coast. Fish Res. Board Can. Bull.,
75, 32.
Meriluoto, J.A., Nygard, S.E., Dahlem, A.M. and Eriksson, J.E. 1990 Synthesis,
organotropism and hepatocellular uptake of two tritium-labeled epimers of
dihydromicrocystin-LR, a cyanobacterial peptide toxin analog. Toxicon, 28(12), 1439-
1446.
Meyer, K., Sommer, H. and Schoenholz, P. 1928 Mussel poisoning. J. Prev. Med., 2,
365-394.
Mittal, A., Agarwal, M. and Schivpuri, D. 1979 Respiratory allergy to algae: clinical
aspects. Ann. Allergy, 42, 253-256.
Moikeha, S.N. and Chu, G.W. 1971 Dermatitis-producing alga Lyngbya majuscula
Gomont in Hawaii. II. Biological properties of the toxic factor. J. Phycol., 7, 8-13.
Molloy, L., Wonnacott, S., Gallagher, T., Brough, P.A. and Livett, B.G. 1995 Anatoxin-a
is a potent agonist of the nicotinic acetylcholine receptor of bovine adrenal chromaffin
cells. Eur. J. Pharmacol., 289(3), 447-453.
Moore, R.E. 1977 Toxins from blue-green algae. BioScience, 27, 797-802.
Moore, R., Corbett, T., Patterson, G.M.L. and Valeriote, F.A. 1996 The search for new
antitumor drugs from blue-green algae. Curr. Pharm. Design, 2, 317-330.
Moore, R.E., Ohtani, I., Moore, B.S., de Koning, C.B., Yoshida, W.Y., Runnegar, M.T.C.
and Carmichael, W. W. 1993 Cyanobacterial toxins. Gazzetta Chimica Italiana, 123,
329-336.
Mynderse, J.S., Moore, R.E., Kashiwagi, M. and Norton, T.R. 1977 Antileukemia activity
in the Oscillatoriaceae: isolation of debromoaplysiatoxin from Lyngbya. Science, 196,
538-540.
Nagata, S., Soutome, H., Tsutsumi, T., Hasegawa, A., Sekijima M., Sugamata, M.,
Harada, K., Suganama, M. and Ueno, Y. 1995 Novel monoclonal antibodies against
microcystin and their protective activity for hepatotoxicity. Natural Toxins, 3, 78-86.
Namikoshi, M. and Rinehart, K L. 1996 Bioactive compounds produced by cyanobacteria.

J. Ind. Microbiol., 17, 373-384.
Negri, A.P. and Jones, G.J. 1995 Bioaccumulation of paralytic shellfish poisoning (PSP)
toxins from the cyanobacterium Anabaena circinalis by the freshwater mussel Alathyria
condola. Toxicon, 33(5), 667-678.
Negri, A P., Jones, G.J. and Hindmarsh, M. 1995 Sheep mortality associated with
paralytic shellfish poisons from the cyanobacterium Anabaena circinalis. Toxicon, 33(10),
1321-1329.
Nishiwaki, R., Ohta, T., Sueoka, E., Suganuma, M., Harada, K., Watanabe, M.F., Fujiki,
H. 1994 Two significant aspects of microcystin-LR: specific binding and liver specificity.
Cancer Lett., 83(1-2), 283-289.
Nishiwaki-Matsushima, R., Ohta, T., Nishiwaki, S., Suganuma, M., Kohyama, K.,
Ishikawa, T., Carmichael, W. W. and Fujiki, H. 1992 Liver tumor promotion by the
cyanobacterial cyclic peptide toxin microcystin LR. J. Cancer Res. din. Oncol., 118(6),
420-424.
Odriozola, E., Ballabene, N. and Salamano, A. 1984 Poisoning of cattle by blue-green

algae (Microcystis aeruginosa). Rev. Argent. Microbiol., 16(4), 219-224.
Ohta, T., Sueoka, E., Iida, N., Komori, A., Suganuma, M., Nishiwaki, R., Tatematsu, M.,
Kim, S.J., Carmichael, W.W., Fujiki, H. 1994 Nodularin, a potent inhibitor of protein
phosphatases 1 and 2A, is a new environmental carcinogen in male F344 rat liver.
Cancer Res., 54(24), 6402-6406.
Ohtani, I., Moore, R.E. and Runnegar, M.T.C. 1992 Cylindrospermopsin - a potent
hepatotoxin from the blue-green alga Cylindrospermopsis raciborskii. J. Am. Chem. Soc.,
114(20), 7941-7942.
Oshima, Y. 1995 Postcolumn derivatization liquid chromatographic method for paralytic

shellfish toxins. J. AOAC Int., 78, 528.
Pearson, M.J., Ferguson, A.J.D., Codd, G.A., Reynolds, C.S., Fawell, J.K., Hamilton,
R.M., Howard, S.R. and Attwood, M.R. 1990 Toxic Blue-Green Algae. A report by the
National Rivers Authority, Water Quality Series No. 2, London, England, 128 pp.
Persson, P., Sivonen, K., Keto, K., Kononen, K., Niemi, M. and Viljamaa, H. 1984
Potentially toxic blue-green algae (cyanobacteria) in Finnish natural waters. Aqua Fenn,
14(2), 147-154.
Pilotto, L.S., Douglas, R.M., Burch, M.D., Cameron, S., Beers, M., Rouch, G.R.,
Robinson, P., Kirk, M., Cowie, C.T., Hardiman, S., Moore, C. and Attewell R.G. 1997
Health effects of recreational exposure to cyanobacteria (blue-green algae) during
recreational water-related activities. Aust. N. Zealand J. Public Health, 21, 562-566.
Prakash, A., Medcof, J. and Tennant, A. 1971 Paralytic Shellfish Poisoning in Eastern
Canada. Bulletin 177, Fisheries Research Board of Canada, Ottawa, Ontario, 87 pp.
Prepas, E.E., Kotak, B.G., Campbell, L.M. Evans, J.C., Hrudey, S.E. Holmes, C.F.B.
1997 Accumulation and elimination of cyanobacterial hepatotoxins by the freshwater
clam Anodonta grandis simpsoniana. Can J. Fish. Aquat. Sci., 54, 41-46.
Pybus, M. and Hobson, D. 1986 Mass mortality of bats due to probable blue-green algae
toxicity. J. Wildl. Dis. 22(3), 449-450.
Raziuddin, S., Siegelman, H.W. and Tornabene, T.G. 1983 Lipopolysaccharides of the
cyanobacterium Microcystis aeruginosa. Eur. J. Biochem., 137, 333-336.
Repavich, W.M., Sonzogni, W.C., Standridge, J.H., Wedepohl, R.E. and Meisner, L.F.
1990 Cyanobacteria (blue-green algae) in Wisconsin waters: acute and chronic toxicity.
Water Res., 24(2), 225-231.
Ressom, R., Soong, F.S., Fitzgerald, J., Turczynowicz, L., El Saadi, O., Roder, D.,
Maynard, T. and Falconer, I. 1994 Health Effects of Toxic Cyanobacteria (Blue-Green
Algae). Australian National Health and Medical Research Council, Looking Glass Press,
108 pp.
Robinson, N.A., Miura, G.A., Matson, C.F., Dinterman, R.E. and Pace, J.G. 1989
Characterization of chemically tritiated microcystin-LR and its distribution in mice.
Toxicon, 27(9), 1035-1042.
Robinson, N.A., Pace, J.G., Matson, C.F., Miura, G.A. and Lawrence, W.B. 1991 Tissue
distribution, excretion and hepatic biotransformation of microcystin-LR in mice. J.
Pharmacol. Exp. Ther., 256(1), 176-182.
Rodrigue, D.C., Etzel, R.A., Hall, S., de Porras, E., Velasquez, O.H., Tauxe, R.V.,
Kilbourne, E. M. and Blake, P.A. 1990 Lethal paralytic shellfish poisoning in Guatemala.
Am. J. Trop. Med. Hyg., 42(3), 267-271.
Runnegar, M.T.C. and Falconer, I.R. 1982 The in vivo and in vitro biological effects of
the reptide hepatotoxin from the blue-green alga Microcystis aeruginosa. South African J.
Sci., 78, 363-366.
Runnegar, M.T.C. and Falconer, I.R. 1986 Effect of toxin from the cyanobacterium
Microcystis aeruginosa on ultrastructural morphology and actin polymerization in isolated
hepatocytes. Toxicon, 24(2), 109-115.
Runnegar, M.T.C., Andrews, J., Gerdes, R.G. and Falconer, I.R. 1987 Injury to
hepatocytes induced by a peptide toxin from the cyanobacterium Microcystis aeruginosa.
Toxicon, 25(11), 1235-1239.
Runnegar, M.T.C., Falconer, I R. and Silver, J. 1981 Deformation of isolated rat

hepatocytes by a peptide hepatotoxin from the blue-green alga Microcystis aeruginosa.
Naunyn-Schmiedebergs Arch Pharmacol., 317, 268-272.
Runnegar, M.T.C., Falconer, I.R., Buckley, T. and Jackson, A.R. 1986 Lethal potency
and tissue distribution of 125I-labelled toxic peptides from the blue-green alga Microcystis
aeruginosa. Toxicon, 24(5), 506 - 509.
Runnegar, M.T.C., Gerdes, R G. and Falconer, I R. 1991 The uptake of the

cyanobacterial hepatotoxin microcystin by isolated rat hepatocytes. Toxicon, 29(1), 43-
51.
Runnegar, M.T.C., Jackson, A.R.B. and Falconer, I.R. 1988 Toxicity of the
cyanobacterium Nodularia spumigena mertens. Toxicon, 26(2), 143-151.
Runnegar, M.T.C., Kong, S.M. and Berndt, N. 1993 Protein phosphatase inhibition and
in vivo hepatotoxicity of microcystins. Am. J. Physiol., 265(2 Part 1), G224-G230.
Runnegar, M.T.C., Kong, S.M., Zhong, Y Z., Ge, J.L. and Lu, S.C. 1994 The role of
glutathione in the toxicity of a novel cyanobacterial alkaloid cylindrospermopsin in
cultured rat hepatocytes. Biochem. Biophys. Res. Commun., 201, 235-241.
Runnegar, M.T.C., Kong, S.M., Zhong, Y.Z. and Lu, S.C. 1995 Inhibition of reduced
glutathione synthesis by cyanobacterial alkaloid cylindrospermopsin in cultured rat
hepatocytes. Biochem. Pharmacol., 49(2), 219-225.
Sato, K., Kitahara, A., Satoh, K., Ichikawa, T., Tatematsu, M. and Ito, N. 1984 The
placental form of glutathione S-transferase as a new marker protein for preneoplasia in
rat chemical carcinogenesis. Gann, 75, 199-202.
Sawyer, P., Gentile, J. and Sasner, J.J. 1968 Demonstration of a toxin from
Aphanizomenon flos-aquae (L)Ralfs. Can. J. Microbiol., 14, 1199-1204.
Scarafia M.E., Agnese A.M., Cabrera J.L. 1995 Microcystis aeruginosa: behaviour and
toxic features in San Roque Dam (Argentina). Nat. Toxins, 3(2), 75-77.
Seawright, A.A., Nolan, C.C., Shaw, G.R., Chiswell, R.K., Norris, R.L., Moore, M.R.,
Smith, M.J. 1999 The oral toxicity for mice of the tropical cyanobacterium
Cylindrospermopsis raciborskii (Woloszynska). Env. Toxicol. Water Qual., 14(1), In
Press.
Shaw, G.R., Seawright, A.A., Chiswell, R.K., Nolan, C.C. 1997 Report to the South East
Queensland Water Board on Cylindrospermopsis research project by NRCET, National
Research Centre for Environmental Toxicology, Queensland, Australia.
Skulberg, O.M. 1979 Giftvirkninger av blågrønnalger - første tilfelle av Microcystis-

forgiftning registrert i Norge (Toxic effects of blue-green algae - first case of Microcystis
poisoning reported from Norway). Tema-rapport 4, Norsk Institutt for Vannforskning,
Oslo, 42 pp. (English summary).
Skulberg, O.M., Carmichael, W.W., Andersen, R.A., Matsunaga, S., Moore, R.E. and
Skulberg, R. 1992 Investigations of a neurotoxic oscillatorialean strain (Cyanophyceae)
and its toxin. Isolation and characterisation of homoanatoxin-a. Environ. Toxicol. Chem.,
11, 321-329.
Soliakov, L., Gallagher, T. and Wonnacott, S. 1995 Anatoxin-a-evoked [3H] dopamine

release from rat striatal synaptosomes. Neuropharmacology, 34(11), 1535-1541.
Sommer, H. and Meyer, K. 1937. Paralytic shellfish poisoning. Arch. Pathol., 24, 560-
598.
Stevens, D.K. and Krieger, R.I. 1989 Application of a sensitive, GC-ECD analysis for
anatoxin-a. Toxicon, 27, 79.
Terao, K., Ohmori, S., Igarashi, K., Ohtani, I., Watanabe, M., Harada, K.-I., Ito, E. and
Watanabe, M. 1994 Electron microscopic studies on experimental poisoning in mice
induced by cylindrospermopsin isolated from blue-green alga Umezakia natans. Toxicon,
32, 833-843.
Teixera, M.G.L.C., Costa, M.C.N., Carvalho, V.L.P., Pereira, M.S. and Hage, E., 1993
Bulletin of the Pan American Health Organization 27, 244-253.
Tisdale, E. 1931 Epidemic of intestinal disorders in Charleston, WVa, occurring

simultaneously with unprecented water supply conditions. Am. J. Public Health, 21, 198-
200.
Toivola, D.M., Eriksson, J.E., Brautigan, D.L. 1994 Identification of protein phosphatase
2A as the primary target for microcystin-LR in rat liver homogenates. FEBS Lett., 344(2-
3), 175-180.
Turner, P.C., Gammie, A.J., Hollinrake, K. and Codd, G.A. 1990 Pneumonia associated
with cyanobacteria. Br. Med. J., 300, 1440-1441.
Ueno, Y., Nagata, S., Tsutsumi, T., Hasegawa, A., Watanabe, M., Park, H.D., Chen, G.-
C., Chen, G. and Yu, S. -Z. 1996 Detection of microcystins, a blue-green algal
liver cancer in China, by highly sensitive immunoassay. Carcinogenesis, 17(6), 1317-
1321.
Wang, H.-B. and Zhu, H -G. 1996 Promoting activity of microcystins extracted from
water blooms in SHE cell transformation assay. Biomed. Environ. Sci., 9, 46-51.
Watts, J.S., Reilly, J., Dacosta, F. and Krop, S. 1966 Acute toxicity of paralytic shellfish
poison in rats of different ages. Toxicol. Appl. Pharmacol., 8, 286-294.
Weckesser, J. and Drews, G. 1979 Lipopolysaccharides of photosynthetic prokaryotes.

Ann. Rev. Microbiol., 33, 215-239.
WHO 1993 Guidelines for Drinking Water Quality, Volume 1. Recommendations, World
Health Organization, Geneva.
WHO 1996 Guidelines for Drinking Water Quality, Volume 2. Health Criteria and other
supporting information. World Health Organization, Geneva.
WHO 1998 Guidelines for Drinking water Quality, Second Edition, Addendum to Volume
2, Health Criteria and other supporting information. World Health Organization, Geneva.
Wiberg, G.S. and Stephenson, N.R. 1960 Toxicologic studies on paralytic shellfish
poison. Toxicol. Appl. Pharmacol., 2, 607-615.
Yamasaki, H. 1988 Multistage carcinogenesis: implications for risk estimation. Cancer

Metastasis Rev., 7, 5-18.
Yoshida, T., Makita, Y., Nagata, S., Tsutsumi, T., Yoshida, F., Sekijima, M., Tamura, S.-I.
and Ueno, Y. 1997 Acute oral toxicity of microcystin-LR, a cyanobacterial hepatotoxin, in
mice. Nat. Toxins, 5, 91-95.
Yoshizawa, S., Matsushima, R., Watanabe, M.F., Harada, K.I., Ichihara, A., Carmichael,
W.W. and Fujiki, H. 1990 Inhibition of protein phosphatases by microcystin and nodularin
associated with hepatotoxicity. J. Cancer Res. din. Oncol., 116, 609-614.
Yu, S. -Z. 1989 Drinking water and primary liver cancer. In: Z.Y. Tang, M.C. Wu and S.S.
Xia [Eds] Primary Liver Cancer. China Academic Publishers, New York, 30-37.
Yu, S.-Z. 1995 Primary prevention of hepatocellular carcinoma. J. Gastroenterol

Hepatol., 10(6), 674-82.
Zilberg, B. 1966 Gastroenteritis in Salisbury European children - a five-year study. Cent.

Afr. J. Med., 12(9), 164-168.
Chapter 5. SAFE LEVELS AND SAFE PRACTICES
This chapter was prepared by Ian Falconer, Jamie Bartram, Ingrid Chorus, Tine Kuiper-
Goodman, Hans Utkilen, Mike Burch and Geoffrey A. Codd
Throughout their lifespan, humans are exposed to many chemical substances, both
beneficial and harmful. It is not possible to eliminate exposure to all toxins in our
environment. Of the harmful chemicals, some are anthropogenic and others occur
naturally. Although cyanotoxins occur naturally, their presence and abundance are, to a
large extent, influenced and increased by human action. The extent to which
cyanotoxins pose a human health risk depends on human exposure to these toxins.
Reducing human exposure to cyanotoxins may be achieved either through preventing
the occurrence of hazardous cell densities of toxic cyanobacteria, or by placing barriers
(such as drinking water treatment or bathing restrictions) that would reduce human
exposure to the cyanobacterial hazard.
The purpose of this chapter is to describe how "safe" levels, such as guideline values,
are derived and how "safe practices" are likely to assist in minimising unsafe human
exposures.
Establishing and controlling safe practices and procedures to reduce or prevent

microbiological problems are common in the food industry, where the process is known
as Hazard Analysis of Critical Control Points (HACCP). There is also a long history of
similar activities in relation to drinking water supply, where the process is referred to as
"sanitary inspection". This type of approach is also starting to be applied in dealing with
some natural toxins, such as fungal toxins (mycotoxins). Safe practice guidelines can
assist in reducing exposure, even for substances for which a guideline value cannot be
set (for example because of inadequate scientific data), or where implementation of a
guideline value is very difficult (for example because of analytical or sampling problems).
Major routes of exposure to cyanotoxins include oral and dermal routes through drinking
water and recreational water use. However, the very limited available information
suggests that inhalation in aerosols (potentially possible while showering, water-skiing or
during certain work practices) may be an equally important route given that the toxicity of
microcystin following intranasal instillation approaches that for intraperitoneal (i.p.)
injection (Fitzgeorge et al., 1994). In some countries, cyanobacterial dietary supplements
may constitute a major route of oral intake for a small sub-population, if the cyanotoxin
levels in the supplements are not controlled. A specific route (intravenous) affecting a
susceptible sub-population may occur in dialysis clinics (see Box 4.4).
The extent or duration of exposure throughout the year is shorter if water resources are
populated by cyanobacteria at specific seasons. In temperate climates, water bodies
dominated by the genus Microcystis usually exhibit a bloom season of 3-5 months,
whereas in milder climates, such as in Australia, some of South America and South
Africa, this genus may bloom for 6-10 months. Other taxa, such as Planktothrix agardhii,
may show perennial mass development even in temperate climates. This is the case in
some lowland regions of north-western Europe, especially in years with mild winters
without ice cover (see Figure 2.7).
5.1 Tolerable exposures

5.1.1 Tolerable daily intake
The repair mechanisms of the body are continuously active and ensure that cells and
tissues are replaced as they are damaged by normal "wear and tear", as well as by
external factors such as toxic chemicals. The tolerable daily intake (TDI) is the amount of
a potentially harmful substance that can be consumed daily over a lifetime with
negligible risk of adverse health effects (see Chapter 4).
Exposure to several harmful chemicals or conditions simultaneously may lead to

potentiation, or to antagonistic interaction. Potentiation results in effects that are larger
than the responses caused by the individual chemicals or conditions by themselves and
this effect has been observed with cyanotoxins, as noted in section 4.2. However, there
is experimental evidence that potentiation is unlikely to occur at low levels of exposure,
such as the TDI. For cyanotoxins, the animal and human toxicity data are incomplete for
the majority of the compounds. The available data have been reviewed in Chapter 4 and
allow only the derivation of a provisional TDI for microcystin-LR (see section 4.2.1).
Clearly, this does not imply that microcystin-LR is the only microcystin that is toxic, or
that other cyanotoxins are less harmful. It merely reflects the lack of toxicological data.
5.1.2 WHO guideline values for drinking water quality
The World Health Organization (WHO) Guidelines for Drinking-water Quality (WHO,
1993, 1996) represent a scientific consensus, based on very broad international
participation, of the health risks presented by microbes and chemicals in drinking water.
This scientific consensus is used to derive "guideline values" which are associated with
guidance on monitoring and management aspects. The guideline values themselves are
based upon a number of assumptions that may be amended locally or nationally
according to specific circumstances (Box 5.1).
A guideline value for lifetime consumption of a chemical contaminant of drinking water is

usually calculated by applying the derived TDI to a typical daily water intake in litres (L)
by an individual of a given body weight (bw). The proportion (P) of total daily intake of
the contaminant which is ingested from the drinking water needs to be considered,
because some intake may come from food or by inhalation from air (WHO, 1996). The
guideline value is then calculated as:
A provisional guideline value of 1.0 µg l-1 has been adopted by WHO for microcystin-LR
(WHO, 1998). In order to derive this, an average adult body weight of 60 kg and an
average water intake for adults of 2 litres per day was used, as is standard practice. The
provisional TDI of 0.04 µg kg-1 bw per day (section 4.2) was used and the proportion of
the TDI allocated to drinking water (P) was assumed to be 0.8. The resulting
concentration (0.96 µg l-1) was rounded to 1.0 µg l-1. In water containing cyanobacterial
cells, this guideline value should be applied to the total cell-bound and extracellular
microcystins.
There were insufficient data to derive a guideline value for cyanotoxins other than
microcystin-LR. For saxitoxins (STX), the guideline of 80 µg STX equivalents per 100 g
(this is the conventional way of expressing this value) shellfish, used in North America
for closing shellfish growing areas for harvesting, may provide preliminary orientation
(IPCS, 1984).
Exceeding the provisional guideline value of 1 µg l-1 for microcystin-LR can be tolerated
(see Box 5.1 for an explanation of handling short-term deviations). This may occur if, for
example, discontinuation of exposure is expected in the near future due to
implementation of measures to eliminate cyanotoxins from drinking water or
cyanobacteria from the water resource. In such instances of guideline exceedances, it
may be appropriate that information is communicated to the public, and especially to
particularly susceptible sub-populations (such as patients with liver disease, parents of
infants, dialysis centres or dialysis patients).
5.1.3 Assessment of microcystins other than microcystin-LR in relation to the

guideline value
There are more than 60 different analogues of microcystin (see section 3.1.1) and, in
many regions, microcystin-LR is not the most commonly occurring microcystin. It may
not even be amongst the microcystins detected. The expression and interpretation of
quantitative results from analysis or assay for these toxins in relation to the WHO
guideline value for microcystin-LR may, therefore, be problematic.
Box 5.1 Derivation of WHO guidelines for drinking water quality
The primary aim of the Guidelines for Drinking-water Quality (WHO, 1993) is the protection of
public health. The Guidelines are intended to be used as the basis for the development of
national standards that, if properly implemented, will ensure the safety of drinking water supplies
through the elimination, or reduction to a minimum concentration, of constituents of water that are
known to be hazardous to health. The guideline values recommended are not mandatory limits,
they are intended to be used in the development of risk management strategies which may
include national or regional standards in the context of local or national environmental, social,
economic and cultural conditions.
The main reason for not promoting the adoption of international standards for drinking water
quality is the advantage provided by the use of a risk-benefit approach (qualitative or quantitative)
to the establishment of national standards or regulations. This approach should lead to standards
and regulations that can be readily implemented and enforced and which ensure the use of
available financial, technical and institutional resources for maximum public health benefit.
The judgement of safety, or what is a tolerable level of risk in certain circumstances, is a matter in
which society as a whole has a role to play. It should be recalled that water is essential to sustain
life and the Guidelines therefore emphasise the importance of securing water supply. They also
indicate that protection of supplies from contamination is almost invariably the best method of
ensuring safe drinking water and is to be preferred to treating a contaminated water supply to
render it suitable for consumption.
A principle of the Guidelines is that the potential consequences of microbial contamination are
such that its control must always be of paramount importance and must never be compromised.
The health risk due to toxic chemicals in drinking water differs from that caused by microbiological
contaminants. There are few chemical constituents of water that can lead to acute health
problems except through massive accidental contamination of a supply. Moreover, experience
shows that, in such incidents, the water usually becomes undrinkable because of unacceptable
taste, odour and appearance. The fact that chemical contaminants are not normally associated
with acute effects places them in a lower priority category than microbial contaminants, the
effects of which may be acute and widespread.
It is important that guideline values are both practical and feasible to implement as well as
protective of public health. Guideline values are not set at concentrations lower than the detection
limits achievable under routine laboratory operating conditions. Moreover, guideline values are
recommended only when control techniques are available to remove or reduce the concentration
of the contaminant to the desired level. In some instances provisional guideline values are set for
constituents for which there is some evidence of a potential hazard but where the available
information on health effects is limited. Provisional guideline values are also set for substances
for which the calculated guideline value would be: (i) below the practical quantification level; or (ii)
below the level that can be achieved through practical treatment methods. Finally, provisional
guideline values are set for substances when it is likely that guideline values will be exceeded as
a result of disinfection procedures.
The first edition of Guidelines for Drinking Water Quality was published by WHO in 1984 and
1985. The second editions of the three volumes of the guidelines were published in 1993, 1996
and 1997; and the Addenda to the second edition were published in 1997 and 1998.
• A guideline value represents the concentration of a constituent that does not result in any
significant risk to the health of the consumer over a lifetime of consumption.
• The quality of water defined by the Guidelines for Drinking-water Quality is such that it is
suitable for human consumption and for all usual domestic purposes, including personal hygiene.
However, water of a higher quality may be required for some special purposes, such as renal
dialysis.
• When a guideline value is exceeded, this should be a signal: (i) to investigate the cause with a
view to taking remedial action; and (ii) to consult with, and seek advice from, the authority
responsible for public health.
• Although the guideline values describe a quality of water that is acceptable for lifelong
consumption, the establishment of these guideline values should not be regarded as implying that
the quality of drinking water may be degraded to the recommended level. Indeed, a continuous
effort should be made to maintain drinking water quality at the highest possible level.
• Short-term deviations above the guideline values do not necessarily mean that the water is
unsuitable for consumption. The amount by which, and the period for which, any guideline value
can be exceeded without affecting public health depends upon the specific substance involved. It
is recommended that when a guideline value is exceeded, the surveillance agency (usually the
authority responsible for public health) should be consulted for advice on suitable action, taking
into account the intake of the substance from sources other than drinking water (for chemical
constituents), the toxicity of the substance, the likelihood and nature of any adverse effects, the
practicability of remedial measures, and similar factors.
• In developing national drinking water standards based on these guideline values, it will be
necessary to take account of a variety of geographical, socio-economic, dietary and other
conditions affecting potential exposure. This may lead to national standards that differ appreciably
from the guideline values.
The most widely used analytical technique for the detection and quantification of
individual microcystin variants for which quantitative reference materials are available is
high pressure liquid chromatography (HPLC) (see section 13.4.1 and Box 13.5). For
toxin variants where reference materials are available, HPLC results can be truly
quantitative. For HPLC peaks that identify microcystin variants for which no standards
are available but the result has been derived from a comparison with the concentration
of the standard for microcystin-LR, the estimates can be expressed as "concentration
equivalents" (CE) of microcystin-LR. It is therefore possible, using HPLC, to derive an
approximate concentration of total microcystins in a sample expressed in terms of
microcystin-LR CE. However, some of the observed HPLC peaks may relate to
toxicologically inactive or weakly toxic variants. In terms of "toxicity equivalents" (TE) of
microcystin-LR the actual toxicity of an unknown sample reported as microcystin-LR
equivalents (CE) is likely to be less than the same concentration of pure microcystin-LR,
because microcystin-LR is one of the most potent microcystins, at least in acute terms
(see Table 3.2). Thus, the microcystin CE approach would result in a "worst case"
estimate of toxic microcystin concentration.
A toxin concentration measured from a water or bloom sample by a bioassay (such as

the mouse bioassay or the phosphatase bioassay) may give a toxin concentration value
which can be expressed as microcystin-LR TE, provided the assay has been calibrated
using microcystin-LR as the quantification standard (which is usually the case). This
measure of total microcystin concentration expressed as microcystin-LR TE will be
closer in actual (acute) toxicity to a solution of the same concentration of pure
microcystin-LR. The position of enzyme assays, such as in vitro protein phosphatase
inhibition assays, in this context is currently unclear. Differences in toxicity of different
microcystins include variation in their ability to enter intact cells and in their capacity to
inhibit protein phosphatases.
In practice, it is important to report quantitative analytical results for samples containing

several microcystins with the above qualifications in mind and to indicate the method
and assumptions used for deriving the quantitative value. If it is necessary to calculate
quantitative values for total microcystins in a sample (and it is certainly important not to
ignore microcystins other than microcystin-LR), then the value should be qualified as
either microcystin-LR CE or TE. This information can be used (at best) as a preliminary
orientation of the hazard presented by the sample in relation to the guideline value (see
section 5.1.2).
Box 5.2 Epidemiological evidence for low-level cyanobacterial hazard
The epidemiological data of Pilotto et al. (1997) can be used as a basis for guideline derivation for
acute, non-cumulative health effects which are more likely to result in discomfort rather than
serious health outcomes. These data encompass the health effects on humans of intact
cyanobacterial cells and colonies and thus include effects of currently unknown substances and
bacteria associated with cyanobacterial colonies. The effects measured were eye irritation, ear
irritation, skin rash, as well as vomiting, diarrhoea, cold/flu symptoms, mouth ulcers and fever. An
elevated "Odds Ratio" for symptoms (3.44) was shown by the people who were in water contact
for more than one hour, at above 5,000 cyanobacterial cells per ml. Similar Odds Ratios were
seen for symptoms in people bathing in water with 5,000-20,000 cells per ml (2.71) and above
80,000 cells per ml (2.90).
5.1.4 Recreational water exposure
Three potential routes of exposure to cyanotoxins can be distinguished: direct contact of

exposed parts of the body, including sensitive areas such as the ears, eyes, mouth and
throat, and the areas covered by a bathing suit (Pilotto et al., 1997); accidental
swallowing (Turner et al., 1990); and inhalation of water. Cases of illness from accidental
swallowing and inhalation of Microcystis have been reported (see section 4.1) and
provide direct evidence of harm to recreational water users from cyanobacterial blooms
in the recreational water bodies.
Health effects observed in the prospective epidemiological study of Pilotto et al. (1997)
occurred at low cyanobacterial cell densities. These related clearly to the cyanobacterial
cell population, but not to the concentration of microcystins (see Box 5.2). Thus, this
hazard appears to be due to additional, or other unidentified, cyanobacterial metabolites
or compounds from associated bacteria, even at moderate levels of exposure.
Intake through oral ingestion or inhalation
Incorporation of toxins through swallowing, contact with nasal mucosa, or by inhalation

are likely to be important routes of exposure to cyanotoxins during water-contact sports.
Well-documented evidence from one animal experiment (Fitzgeorge et al., 1994) and
one case of multiple human illness (Turner et al., 1990) indicates that inhalation and
resorption through nasal and pharyngeal mucous membranes may present a high risk in
water sports involving intensive submersion of the head (jumping from diving boards,
sailboarding, canoe capsizing, competitive swimming) and inhalation of aerosols (water
skiing).
Experimental results indicate a hazard of cumulative liver damage by repeated

microcystin intake (Fitzgeorge et al., 1994, see section 4.2.1), as can occur during a
holiday with daily bathing at a recreational site with a high density of microcystin-
containing cyanobacteria. Sub-acute liver injury is likely to go unnoticed, because signs
of liver injury are only apparent after severe injury. In addition, the dose-response curve
for liver injury from microcystins is relatively steep. There may be little evidence of acute
liver damage when levels are close to those that could lead to severe acute toxicity, and
thus exposure at such levels is likely to be continued by people if they are uninformed of
the risk (e.g. for consecutive days of a holiday or hot spell), thereby increasing the risk of
cumulative liver damage.
Risks of ingestion are particularly high for children playing in shallow near-shore water
where scums tend to accumulate. Because the hazard of microcystin uptake is directly
related to the levels of toxins in the water (cell-bound as well as dissolved) and the
volume of water ingested or inhaled, the range in these levels needs to be recognised in
deriving guidelines for recreational water safety.
Direct contact
Allergic and toxic dermal reactions of varying severity are known from cyanobacteria as
well as from freshwater algae, but have not been documented extensively. Bathing suits,
and particularly diving suits, tend to aggravate such effects by accumulating
cyanobacterial cells, thereby enhancing the disruption of cells and hence the liberation of
cell contents onto the wearer's skin. Reports from the USA have recorded allergic
reactions from recreational exposure, and the cyanobacterial pigment phycocyanin was
shown to be responsible in one case (Cohen and Reif, 1953). Severe dermatitis,
resembling skin bums, has been reported from marine bathing in the presence of
cyanobacteria dislodged from rocks, particularly after storms in tropical seas (see
section 4.2.8).
5.2 Safe practices

The placing of barriers that reduce exposure to a cyanotoxin hazard is an important
measure and involves identifying "critical control points" and implementing measures for
their monitoring and control. In the case of cyanobacteria, critical control points might
include, for example, noting the tendency of a water body to develop blooms, scums or
mats. Monitoring schemes need to be developed that are capable of detecting
proliferation of cyanobacteria (linked to a programme of appropriate actions) and
drinking water treatment technology needs to be in existence that is capable of
preventing human exposure if cyanobacteria occur in source waters.
5.2.1 Drinking water
A drinking water supply safe from cyanotoxins will either draw upon a resource which
does not harbour cyanotoxins (e.g. groundwater or surface water which does not support
cyanobacterial growth), or have treatment in place that is likely to remove cyanobacterial
cells (without causing their disrupture) as well as removing cyanotoxins. However, in
many circumstances a potential cyanotoxin hazard can be managed effectively without
the necessity of advanced treatment processes, through water resource management
techniques (see Chapter 8) and removal of intact cells (see Chapter 9). The critical
control points for safe practices are indicated in Table 5.1.
Most of the reported incidents of human injury that have raised awareness of the
importance of cyanotoxins in drinking water have involved the inappropriate treatment of
water supplies, such as the use of copper sulphate in dealing with an established bloom
of cyanobacteria.
A very effective approach to safe practices may involve changing the drinking water
source. In a number of regions, surface waters are used for reasons of easy access and
tradition, although groundwater of high quality is available. Exploring options of
improving practices of drinking water abstraction with low technological input (such as
drilling wells, or using bankside filtration) may lead to health benefits. In China, a high
prevalence of endemic primary liver cancer was related to several factors: hepatitis B,
aflatoxins in the diet, and drinking surface water polluted with cyanobacteria likely to
contain microcystins. Changing the drinking water source from shallow, eutrophic ponds
and ditches to groundwater was a major element of a package of measures which
showed some success in improving health (Box 5.3).
5.2.2 Recreational waters
Recreational water use is likely to be a major route of exposure to cyanotoxins in some

parts of the world. Whereas similar approaches to resource protection apply as for
drinking water, there are very few further management options available once
cyanobacteria proliferate or accumulate in a recreational water. Because adequate
surveillance is sometimes difficult and management options, except precluding or
discouraging use, may be scarce, a large share of the responsibility for safe practices
lies with the users of a bathing site. The provision of adequate information to the public
becomes, therefore, a major responsibility of public authorities.
The growth of cyanobacteria in lakes and rivers used for recreational purposes has been
well recognised as a public nuisance. Water blooms of cyanobacteria may be associated
with unpleasant odours and the offensive appearance of lake shores, especially when
scums of the organisms accumulate and decay. Areas with extensive cyanobacterial
scums or accumulated detached mats on bathing beaches may be avoided by
swimmers and other water users because of the obviously unpleasant environment,
particularly if locally anaerobic water conditions or cyanobacterial toxins cause fish-kills,
further emphasising the unattractiveness of water contact. In temperate climates,
cyanobacterial dominance is most pronounced during the summer months, when the
demand for recreational water is highest. In some regions, cyanobacteria have been
abundant for more than a generation and visitors have accepted this water quality as
"normal" for their region. Multiple anecdotal observations of children playing with scum
material have been reported.
Table 5.1 Critical control points for assessing the intrinsic safety of a drinking water
supply which may contain cyanobacterial cells and/or toxins
Control point/issue Comments

Source water type The health risk associated with cyanobacterial contamination of
groundwaters is generally negligible. An exception may occur where
infiltration galleries are strongly influenced by eutrophic surface waters
Occurrence of Many surface water sources do not support cyanobacterial growth. In
cyanobacteria in source others, cyanobacteria may occur occasionally at low population
water and tendency for densities. In reservoirs and lakes with very low nutrient concentrations
bloom formation (total phosphorus < 10 µg P l-1) or rivers and reservoirs with a
hydrodynamic regime unfavourable for cyanobacteria (continuous high
flows especially during summer, or deep vertical mixing), other
phytoplankton species may regularly out-compete cyanobacteria. A
water source which does not have a history of cyanobacterial growth or
bloom formation is generally considered to present a low cyanotoxin
risk, regardless of treatment type. Where bloom formation is well
characterised in terms of annual cycles, the health risk may similarly be
low if control measures are in place for times of bloom formation. If
regular monitoring of source phytoplankton is in place, waters
presenting no significant cyanotoxin risk are easily identified (see Table
6.2)
Likelihood of cell lysis in Throughout cyanobacterial growth, most cyanotoxins are cell-bound.
transport or treatment Removal of intact cyanobacterial cells therefore largely removes
cyanotoxins (see section 3.4). Neurotoxins may be an exception under
some circumstances. When cyanobacterial cells die and decay (lyse),
toxins are released. Lysis can occur naturally or be caused by chemical
treatment, hydraulic and pumping regimes in different treatment steps,
and by long transport pipes for raw water. Thus, abstraction and
treatment systems which lead to cell lysis present an increased risk of
cyanotoxin release.
Treatment systems Methods, such as adsorption to some types of granular activated
capable of toxin carbon, and oxidation, can be effective in cyanotoxin removal. However,
removal conditions of operation are critical for success. If processes are
operated only periodically during cyanobacterial growth or reservoir
treatment, monitoring of plant functioning must be adequate to ensure
cyanotoxin removal. Substantially less is known about removal of
neurotoxins and cylindrospermopsin than about microcystins, thus toxin
monitoring of treatment steps and finished water is especially important
if potentially neurotoxic or cylindrospermopsin-producing cyanobacteria
proliferate
Box 5.3 Primary liver cancer and cyanotoxins in China
Primary Liver Cancer is one of the most common cancers in China. In 1994 and 1995, it
accounted for 24 mortalities per 100,000 population in some rural counties and cities; in these
areas it was ranked with stomach cancer as the two most important causes of cancer death.
The uneven geographic distribution of liver cancer was conspicuous, and "hot spots" could be
related to drinking water supply, e.g. in some clearly delineated areas of Nandong District, in
Jiangsu Province (particularly in Rudong, but also in Haimen and Qidong), in Nanhui (suburb of
Shanghai) and Fusui (Guangxi).
• In Rudong, Nanhui and Fusui people had blocked the drainage system, causing stagnation of
the water used for the drinking supply.
• In areas of Qidong-Haimen, with mortality rates 20 per 100,000, people drank water from the
Yangtze River, but in areas with mortalities of 100 per 100 000, pond and ditch water was used.
• Primary liver cancer mortalities 10 per 100,000 were found in areas where water from deep
wells were used for drinking.
Epidemiological study of the mortality showed strongest correlation with hepatitis B incidence, a
lesser correlation with aflatoxins in the diet, and a third correlation with drinking of pond and ditch
water. No correlations were found with insecticides. Samples of pond and ditch water showed
microcystin present in both endemic liver cancer areas and in areas with lower liver cancer rates.
Children in some endemic areas were fed corn paste and drank pond or ditch water from infancy.
Further, up to 43 per cent carry the hepatitis-B virus from infection by their HBsAg positive
mothers. The evidence suggests that aflatoxins from corn and microcystins from drinking water
act together with the hepatitis B virus in causing and promoting primary liver cancer.
In order to alleviate this situation, attempts have been launched over the past 20 years to change
the staple food and drinking habits of the people. Efforts began with the methods of harvest,
following the motto "quick to reap, quick to store, at a moisture content 12.5 per cent", aimed at
the reduction of fungal contamination. For some time, the government bought corn and
exchanged it for rice to reduce aflatoxin exposure, but this function has now been transferred to a
private initiative in the market economy. Recently, it has been estimated that more than 95 per
cent of the population eats rice rather than corn.
Even prior to the recognition of microcystins as possible promoters of endemic primary liver
cancer, the connection to poor quality surface water for drinking was observed and programmes
for construction of deep wells were begun. At present, 80 per cent of the population in some of
the afflicted regions have changed their water source to deep well water, and the incidence of
liver cancer has dropped consistently.
The mottoes for prevention of primary liver cancer now are:
"control of water - control of crops - prevention of hepatitis"
For additional discussion, see section 4.1.2.

Health impairments from cyanobacteria in recreational waters must be differentiated
between the chiefly irritative symptoms caused by unknown cyanobacterial substances
(as described in Box 5.2), and the more severe hazard of exposure to high
concentrations of known cyanotoxins, particularly microcystins. A single guideline
therefore, is not appropriate. Rather, a series of guidelines associated with incremental
severity and probability of adverse effects has been defined at three levels as described
below.
1. Relatively mild and/or low probabilities of adverse health effects
For protection from health outcomes not due to cyanotoxin toxicity, but due to the
irritative or allergenic effects of other cyanobacterial compounds, a guideline level of
20,000 cyanobacterial cells per ml (corresponding to 10 µg l-1 of chlorophyll a under
conditions of cyanobacterial dominance) can be derived from the prospective
epidemiological study by Pilotto et al. (1997) (see Box 5.2). Whereas the health
outcomes reported in this study were related to cyanobacterial density and duration of
exposure, they affected less than 30 per cent of the individuals exposed. At this
cyanobacterial density, 2-4 µg l-1 of microcystins may be expected if microcystin-
producing cyanobacteria are dominant, with 10 µg l-1 being possible with highly toxic
blooms (regional differences in microcystin content of the cells may be substantial). This
level is close to the WHO provisional drinking water guideline value of 1 µg l-1 for
microcystin-LR (WHO, 1998) which is intended to be safe for life-long consumption.
Thus, health outcomes due to microcystin are unlikely and providing information for
visitors to bathing sites with this low-level risk is considered to be sufficient. Additionally,
it is recommended that the authorities are informed in order to initiate further surveillance
of the site.
2. Moderate probability of adverse health effects
At higher concentrations of cyanobacterial cells, the probability of irritative symptoms is

elevated. Additionally, cyanotoxins (usually cell-bound) may reach concentrations with
potential health impact. To assess risk under these circumstances the data used for the
drinking water provisional guideline value may be applied. Swimmers involuntarily
swallow some water while bathing and the harm from ingestion of bathing water will be
comparable with that from a drinking water supply with the same toxin content. A
swimmer can expect to ingest up to 100-200 ml of water in one session, sail-board riders
and water skiers would probably ingest more.
A density of 100,000 cyanobacterial cells per ml (which is equivalent to approximately 50

µg l-1 of chlorophyll a if cyanobacteria dominate) is a guideline for a moderate health alert
in recreational waters. At this density, 20 µg l-1 of microcystins are likely, if the bloom
consists of Microcystis and has an average toxin content per cell of 0.2 pg, or 0.4 µg
microcystin per µg chlorophyll a (up to 50 µg l-1 of microcystin are possible) but toxin
levels may approximately double if Planktothrix agardhii is dominant. This toxin
concentration is equivalent to 20 times the WHO provisional guideline value for
microcystin-LR in drinking water, but would result in consumption of an amount close to
the TDI for an adult of 60 kg consuming 100 ml of water while swimming (rather than 2
litres of drinking water). However, a child of 15 kg consuming 250 ml of water during
extensive playing could be exposed to 10 times the TDI. The health risk will be
increased if the person exposed is particularly susceptible (e.g. because of chronic
hepatitis B). Therefore, cyanobacterial densities likely to cause microcystin
concentrations of 20 µg l-1 should trigger further action.
Non-scum-forming species of cyanobacteria, such as Planktothrix agardhii, have been

observed to reach cell densities corresponding to 200 µg l-1 of chlorophyll a or even more
in shallow water bodies. Transparency in such situations will be less than 0.5 m when
measured with a Secchi disk (see Chapter 11). Planktothrix agardhii has been shown to
contain a very high cell content of microcystin (1-2 µg per µg chlorophyll a) (see Figure
3.5) and therefore toxin concentrations of 200-400 µg l-1 can occur without scum
formation.
An additional reason for increased alert at 100,000 cells per ml is the potential of some
frequently occurring cyanobacterial species (particularly Microcystis spp. and Anabaena
spp.) to form scums (see Figure 5.1). These scums may increase local cell density and
thus toxin concentration by a factor of 1,000 or more in a few hours, thus rapidly
changing the risk from moderate to high (see next subsection) for bathers and others
involved in body-contact water sports.
Cyanobacterial scum formation presents a unique problem for routine monitoring carried
out at the usual time intervals of one or two weeks, because such monitoring intervals
are unlikely to detect hazardous maxima. Because of the potential for rapid scum
formation at a cyanobacterial density of 100,000 cells per ml or 50 µg l-1 chlorophyll a
(from scum-forming cyanobacterial taxa), intensification of surveillance and protective
measures are appropriate at these levels. Daily inspection for scum formation (if scum-
forming taxa are present) and measures to prevent exposure in areas prone to scum
formation are the two main options.
Intervention is recommended to trigger effective public information campaigns educating

people on avoidance of scum contact. Furthermore, in some cases (e.g. with frequent
scum formation), restriction of bathing may be judged to be appropriate. An intensified
monitoring programme should be implemented, particularly looking for scum
accumulations. Health authorities should be notified immediately.
Figure 5.1 Schematic illustration of scum-forming potential changing the cyanotoxin risk
from moderate to very high
Lake profile
Moderate risk level:
• 50 µg l-1 chlorophyll a
• or 100,000 cells l-1
• possibly 20 µg l-1 of microcystin in top 4 m of water body
100-fold accumulation to high risk level scum:
• 5,000 µg l-1 chlorophyll a

• or 10,000,000 cells l-1
• possibly 2,000 µg l-1 of microcystin in top 4 cm of water body
1,000-fold accumulation to very high risk level shore scum if wind sweeps scums from
100 m into 10 m:
• 50,000 µg l-1 chlorophyll a

• or 100,000,000 cells l-1
• possibly 20,000 µg l-1 of microcystin concentrated in one bay of the water body
Lake plan
3. High risk of adverse health effects
Abundant evidence exists for potentially severe health hazards associated with scums
caused by toxic cyanobacteria (see section 4.1). No human fatalities have been
unequivocally associated with oral ingestion of scum, even though numerous animals
have been killed by consuming water containing cyanobacterial scum material (see
section 4.1). This discrepancy can be explained by the fact that animals would drink
higher volumes of scum-containing water, compared with the small amounts of scum
accidentally ingested by humans during bathing (resulting in a lower dose).
Cyanobacterial scums can represent a thousand-fold to million-fold concentration of

cyanobacterial cell populations. It has been calculated that a child playing in a
Microcystis scum for a protracted period and ingesting a significant volume could receive
a lethal exposure, although there are no reports that this has actually occurred. Based
on evidence that the oral LD50 of microcystin-LR in mice is 5,000-11,600 µg kg-1 bw (see
section 4.2), for a child of 10 kg the ingestion of 2 mg of microcystin or less could be
expected to cause liver injury, because concentrations of up to 24 mg l-1 of microcystins
have been published from scum material (see section 3.2). Substantially higher
enrichment of scums (up to gelatinous consistency) is occasionally observed, and
accidental ingestion of smaller volumes of these could cause serious harm. Anecdotal
evidence indicates that children, and even adults, may be attracted to play in scums.
The presence of scums caused by cyanobacteria is a readily detected indicator of a high
risk of adverse health effects for those bathers who come into contact with the scum.
Immediate action to control scum contact is recommended for such situations.
The approach outlined in this section, however, does not cover all conceivable situations.
Swimmers may be in contact with benthic cyanobacteria after a storm breaks off clumps
of filaments, or cyanobacterial mats naturally detach from the sediment and are
accumulated on the shore (Edwards et al., 1992). Some marine beaches have been
reported to have widespread problems due to a benthic marine cyanobacterium,
Lyngbya majuscula, growing on rocks in tropical seas and causing severe blistering
when trapped under the bathing suits of people swimming following a storm (Grauer,
1961). This response may be due to acute toxicity; Lyngbya can produce irritant toxins.
Measures of cyanobacterial population cell density as outlined in Table 5.2, will not
detect these hazards. Instead, this type of hazard calls for critical and well-informed
observation of bathing sites, coupled with a flexible response.
It is difficult to define "safe" concentrations of cyanobacteria in recreational water in
relation to allergenic effects or skin reactions, because individual sensitivities vary
greatly. Aggravation of dermal reactions due to accumulation of cyanobacterial material
and enhanced disruption of cells under bathing suits and wet suits may be a problem,
even at all densities below the guidelines described above. Further information related to
monitoring and management of recreational waters is available in Bartram and Rees
(1999).
5.3 Other exposure routes

5.3.1 Renal dialysis
Renal dialysis patients are at great risk when water used for dialysis contains
contaminants such as cyanotoxins. For these patients large volumes of water (120 litres)
are used and the route of exposure is similar to the i.v. route, which allows for a much
greater uptake of toxin than following oral ingestion. One serious incident, including a
number of deaths arising from exposure through this route, has already been described
in section 4.1.
The WHO Guidelines for Drinking-water Quality (WHO, 1993) do not consider the
especially high quality of water needed for dialysis treatment, intravenous therapy or
other clinical uses. The treatment processes used at conventional surface water
treatment plants (such as coagulation, clarification and sand filtration) are normally
effective in removing cyanobacterial cells, but are not effective in removing or destroying
dissolved cyanotoxins, especially from water supplies with a high organic content and
cyanobacterial dominance (see Chapter 9). Consequently, clinics and hospitals with
special water needs, such as for dialysis treatment or for transfusions (intravenous
administration), may need to provide additional water treatment to remove the
cyanotoxins. Such treatment ranges from granular activated carbon filtration, followed by
reverse osmosis, to more elaborate treatment including membrane filtration (e.g. 25 µm
pore size filter). The extent of treatment necessary depends on the quality of the
municipal water supply.
Continuous monitoring of performance and equipment is essential to ensure adequate

quality of the water. On-site water treatment systems in clinics and hospitals require
rigorous monitoring and regular maintenance, including back-flushing of filters and
recharge of activated carbon, according to manufacturers' specifications. It is important
that manufacturers' specifications should be assessed for their adequacy for maintaining
performance under local conditions. Activated carbon, for example, may be exhausted
for its ability to remove cyanotoxins long before it reaches saturation for removal of other
organic compounds, and some manufacturers may be unaware of this.
Table 5.2 Guidelines for safe practice in managing bathing waters which may produce or
contain cyanobacterial cells and/or toxins
Guidance level or How guidance level

Health risks Recommended action
situation derived
Cyanobacterial scum Inference from oral Potential for acute Immediate action to
formation in bathing animal lethal poisoning prevent contact with
areas poisonings scums; possible
Potential for long-term prohibition of swimming
Actual human illness illness with some and other water-contact
case histories cyanobacterial species activities
Short-term adverse Public health follow-up

health outcomes, e.g. investigation
skin irritations,
gastrointestinal illness Inform relevant
authorities
100,000 cells From provisional Potential for long-term Watch for scums
cyanobacteria per ml drinking water illness with some
or 50 µg chlorophyll a guideline for cyanobacterial species Restrict bathing and
per litre with microcystin-LR, and further investigate
dominance of data concerning other Short-term adverse hazard
cyanobacteria cyanotoxins health outcomes, e.g.
skin irritations, Post on-site risk advisory
gastrointestinal illness signs Inform relevant
health authorities
20,000 cells From human bathing Short-term adverse Post on-site risk advisory
cyanobacteria per ml epidemiological study health outcomes, e.g. signs Inform relevant
or 10 µg chlorophyll a skin irritations, authorities
per litre with gastrointestinal illness,
dominance of probably at low
cyanobacteria frequency
5.3.2 Irrigation water
The use of water from sources containing cyanobacterial blooms and toxins for spray
irrigation of crops presents potential health hazards through several exposure routes,
including uptake into the food chain. Workers or passers-by may inhale toxins with spray
drift, and skin contact with cyanobacteria and dissolved toxins may also occur.
Questions therefore arise about the health significance of spray irrigation with water
containing cyanobacterial toxins. As shown in section 4.2, animal experimentation has
indicated that microcystin uptake through nasal mucosa may be considerable. When
considered together with the skin irritations, respiratory distress and nasal mucosal
irritations observed after recreational exposure (see sections 4.1 and 5.2.2), the
indicators are that occupational exposure to spray irrigation water should be avoided (by
appropriate work practices) if the water contains cyanobacterial toxins. Incidental
exposure of humans and animals to such spray irrigation water, for example by
downwind drift, should also be avoided.
There are several indications that terrestrial plants, including food crop plants, can take
up microcystins. Mustard seedling development is inhibited if microcystin-LR is
presented to the roots in aqueous solution (Kos et al., 1995). Microcystins have several
perturbatory effects on plant physiology and metabolism, when sufficient levels of toxin
enter the plant cells. Plant protein phosphatases show high susceptibility to inhibition by
microcystin-LR in vitro (MacKintosh et al., 1990). Plant sucrose metabolism is inhibited if
microcystin-LR is administered in solution by injection into the transpiration stream (Siegl
et al., 1990). Inhibition of whole leaf photosynthesis by French Bean plants occurred
after topical exposure of the leaves to an aqueous solution of microcystin-LR during
greenhouse studies (Abe et al., 1996). The degree of whole leaf photosynthesis
inhibition increased with subsequent brief exposures to the toxin in solution at 48-hour
intervals, eventually becoming irreversible. These results were observed at dissolved
microcystin-LR concentrations which can be found in untreated waters containing
cyanobacterial blooms (20 µg l-1 of toxin); leaf necrosis occurred at higher exposure
levels (Abe et al., 1996). The relevance of these findings for field situations is currently
unclear.
In addition to the possibility of internal accumulation of microcystins, irrigation may lead

to accumulation of toxins on the external surfaces of edible plant material. The toxins are
deposited when the water dries on the plant surface between irrigation periods or when
the water becomes trapped in the centres of, for example, salad plants. Further research
is needed into the uptake and fate of microcystins and other cyanobacterial toxins by
food plants and the persistence of the toxins on plant surfaces.
5.3.3 Cyanobacteria sold as dietary supplements
In some countries cyanobacteria are sold as dietary supplements, with the number of
users of these products estimated to be well over a million in North America alone.
Large-scale production of cyanobacteria and microalgae started some 50 years ago.
Much of the early research work dealt with the basic photosynthetic properties of
microalgae, their possible therapeutic, antibiotic and toxicological properties and their
potential as an agricultural commodity. The microalgae biomass industry now provides
significant biomass for pigments and speciality chemicals used primarily in the food
industry. The bulk of this microalgal biomass comes from two filamentous genera of
cyanobacteria: Spirulina including S. platensis and S. maxima (Belay et al., 1994) and
Aphanizomenon flos-aquae. While Spirulina is grown in artificial outdoor ponds, mainly
in southern California, Hawaii, Thailand, Taiwan and Japan, Aphanizomenon is at
present harvested from a natural lake. Production of food-grade Spirulina exceeds 1 ×
106 kg a-1 (Belay et al., 1993). Aphanizomenon production is also substantial.
As a consequence of the consumption of these products many quality control issues

arise. One such issue concerns the possible production of cyanotoxins by cyanobacterial
genera used for dietary supplements. In particular, Aphanizomenon flos-aquae has been
shown to be capable of producing saxitoxins (Mahmood and Carmichael, 1986) and the
neurotoxin anatoxin-a (Bumke-Vogt et al., 1999). While no saxitoxins have been
detected in Aphanizomenon flos-aquae marketed as a dietary supplement, it is
appropriate to monitor all supplements in order to ensure safety. In natural lakes,
mixtures of species often occur in cyanobacterial blooms. In particular Microcystis and
Anabaena, which usually contain microcystins, can both occur in association with
Aphanizomenon. The microcystins may then become part of the biomass harvested for
human consumption. Failure to monitor and regulate these toxins in cyanobacterial
biomass used as part of a human diet could lead to an increased risk for the consumer
(see Box 5.4).
5.3.4 Cyanobacteria and Vibrio cholerae
Islam (1991) described detection of Vibrio cholerae inside the mucilaginous sheath of
Anabaena variabilis for up to 15 months after artificial exposure and also noted that V.
cholerae 01 did not lose toxigenicity during the association (Islam, 1991). Field studies
have also detected V. cholerae in the mucilaginous sheath of Anabaena sp. from a pond
in Dhaka when it could not be detected in association with microalgae collected from the
same environment (Euglena sp. and Phacus sp. (Islam et al., 1994)).
The association between V. cholerae and cyanobacteria remains poorly understood, but
it has been noted that vibrios may produce mucinase (Schneider and Parker, 1982) and
it has been suggested that exchange of oxygen (from photosynthesis for aerobic
respiration) and carbon dioxide may permit a symbiotic relationship (Islam, 1987; Islam
et al., 1994). Other workers have noted that motile bacteria can easily discriminate
heterocysts from vegetative cells and attach to the heterocyst vegetative cell junction,
following which both host and epiphytes start growing (Paerl and Gallucci, 1985; Islam et
al., 1990). The bacteria rarely penetrate cyanobacterial cell walls (Islam et al., 1990).
Evidence is accumulating that association with mucilaginous cyanobacteria may be an
important factor in inter-epidemic survival of V. cholerae. The implications of this for the
control of cholera in humans remain unclear and it should be noted that the evidence
relates to mucilaginous cyanobacteria, and Anabaena in particular. No studies have
suggested or apparently investigated a relationship with cyanobacterial toxigenicity.
Box 5.4 Calculation of risk associated with consumption of cyanobacterial products

contaminated with microcystins
The State of Oregon, USA, has adopted 1 µg g-1 (1 ppm) of microcystins as a standard for
cyanobacterial products. The consumption of 2 g (as suggested by some producers) by a 60 kg
person of product containing 1 µg g-1 would result in a microcystin intake of 0.033 µg kg-1 bw. This
intake is slightly below the tolerable daily intake (TDI) of 0.04 µg kg-1 bw per day used for
derivation of the WHO guideline for microcystin-LR in drinking water (WHO, 1998) (see section
4.2). However, an intake exceeding 2 g per day of a product containing microcystins at a
concentration near the State of Oregon standard may exceed the TDI, and a consumption of 2 g
per day by children may also exceed the TDI because of their lower body weight. In deriving its
drinking water guideline value, WHO apportioned 20 per cent of intake to other sources. For
persons consuming cyanobacterial products, this apportionment may be inappropriate.
5.4 Tastes and odours

Cyanobacteria have, for a long time, been recognised as a nuisance in the drinking
water industry because of the ability of several taxa to produce earthy and musty
smelling compounds, notably geosmin and 2-methyl isoborneol (2-MIB), for which the
odour detection thresholds of less than 10 ng l-1 are remarkably low among sensitive
individuals.
The cyanobacterial genera that are known to produce geosmin are Anabaena,
Aphanizomenon, Lyngbya, Microcystis, Oscillatoria, Phormidium, Schizothrix and
Symploca (Perrson, 1983). All these (except Symploca) are also known to include toxin-
forming species and strains.
Because of this, the possibility of using odour compounds as an early warning for the
development of toxin-producing cyanobacteria blooms has been considered. However,
there is no evidence of a correlation between toxin production and the production of
taste- and odour-producing compounds that would provide a warning of toxicity.
Evidence from the literature on the capability of various cyanobacterial species to
produce both toxins and taste and odour compounds has been summarised by Kenefick
et al. (1992) and does not indicate that cyanobacterial species which produce toxins
invariably also produce taste and odour. Nevertheless, some characteristic tastes and
odours may indicate the presence of cyanobacteria, and toxic cyanobacteria frequently
occur without noticeable tastes or odours. In Alberta, Canada, 89 bloom samples from
10 lakes were analysed for the presence of microcystin-LR, and the taste and odour
compounds geosmin, 2-MIB and β-cyclocitral. The latter compound, which is only mildly
odorous compared with geosmin and 2-MIB, was reported to be produced in large
quantity by Microcystis spp. (Jüttner, 1988). The results showed that all but three of the
bloom samples had detectable levels of microcystin while none had detectable levels of
2-MIB. Several samples had detectable levels of geosmin, but there was no clear
relationship between the presence of geosmin and the presence of microcystin-LR. In
the case of β-cyclocitral, there was a significant correlation, at the 1 per cent level.
However, this relationship is of no practical use for providing early warning of the
presence of microcystin-LR, because the relationship is not consistent and β-cyclocitral
is not odorous enough to act as a sensitive surrogate for microcystin-LR (Hrudey et al.,
1993). Although there have been some rare occasions when cyanobacterial isolates
have been found to produce geosmin, microcystin and anatoxin-a simultaneously, as for
Anabaena lemmermannii from Lake Hallevann in Norway (Haneberg et al., 1994), such
reports are exceptions. The biochemical pathways to the biosynthesis of microcystin
(Dittmann et al., 1996), anatoxin (Gallon et al., 1989) and saxitoxin (Skimizu et al., 1984)
are becoming understood. These show no connection between toxin production and the
production of the alcohols geosmin and 2-MIB. It is therefore very unlikely that the
production of the taste and odour compounds are biochemically connected to the
production of the cyanotoxins.
Just as the presence of earthy or musty odours in water indicates the presence of
cyanobacteria and/or actinomycetes, taste and odour problems can be used as a
warning of the need for further investigation in the event that the occurrence of
cyanobacteria could result in the presence of cyanotoxins. It is, however, important to
recognise that the converse does not apply: lack of taste and odour by no means implies
the absence of cyanobacteria.
5.5 References
Abe, T., Lawson, T., Weyers, J.D.B. and Codd, G.A. 1996 Microcystin-LR inhibits
photosynthesis of Phaseolus vulgaris primary leaves: implications for current spray
irrigation practice. New Phytol., 133, 651-658.
Bartram, J. and Rees, G. 1999 Recreational Water Monitoring. E & FN Spon, London (In
Press).
Belay, A., Ota, Y., Miyakawa, K. and Shimamatsu, H. 1993 Current knowledge on
potential health benefits of Spirulina. J. Appl. Phycol., 5, 235-241.
Belay, A., Ota, Y., Miyakawa K. and Shimamatsu H. 1994 Production of high quality
Spirulina at Earthrise Farms. In: Phang et al. [Ed.] Algal Biotechnology in the Asia-
Pacific Region, 92-102.
Bumke-Vogt, C., Mailahn, W., Chorus, C. 1999 Anatoxin-a and neurotoxic cyanobacteria
in German lakes and reservoirs. Env. Toxicol. Water Qual., 14, In Press.
Cohen, S.G. and Reif, C.B. 1953 Cutaneous sensitisation to blue-green algae. J. Allergy.
24, 452-457.
Dittmann, E., Meissner, K. and Borner, T. 1996 Conserved sequences of peptide

synthetase genes in the cyanobacterium Microcystis aeruginosa. Phycologia, 35(6), 62-
67.
Edwards, C., Beattie, K.A., Scrimgeour, C. and Codd, G.A. 1992 Identification of
at Loch Insh, Scotland. Toxicon, 30, 1165-1175.
Fitzgeorge, R.B., Clark, S.A. and Keevil, C.W. 1994 Routes of intoxication. In: G.A. Codd,
T.M. Jefferies, C.W. Keevil and E. Potter. [Eds] Detection Methods for Cyanobacterial
Toxins. The Royal Society of Chemistry, Cambridge, UK, 69-74.
Gallon, J.R., Khadije, C.N. and Brown, E.G. 1989 Biosynthesis of the tropane-related
cyanobacterial toxin anatoxin-a: role of ornithine decarboxylase. Phytochemistry, 29,
1107-1111.
Grauer, F. 1961 Seaweed dermatitis. Arch. Dermatol., 84, 720-732.
Haneberg, B., Hanssen, T.R., Ringdal, E. 1994 Hallevannet. Toksisk algeoppblomstring i

en oligotrof innsjø. Telemark distrikshøgskole, Bø, Thesis.
Hrudey, S.E., Kenefick, S.L., Best, N., Gillespie, T., Kotak, B.G., Prepas, E.E. and
Peterson, H.G. 1993 Liver toxins and odour agents in cyanobacterial blooms in Alberta
surface water supplies. In: Disinfection Dilemma: Microbiological Control Versus By-
Products. Proc. 5th National Conference on Drinking Water. Winnipeg, American Water
Works Association, 383-390.
IPCS 1984 Aquatic Marine and Freshwater Biotoxins. Environmental Health Criteria 37,
Islam, M.S. 1987 Studies of aquatic flora as possible reservoirs of toxigenic Vibrio
cholerae 01. PhD Thesis, Faculty of Medicine, University of London.
Islam, M.S. 1991 Toxigenicity and Toxin genes of Vibrio cholerae 01 Isolated from an
Artificial Aquatic Environment. World Microbiol. Biotechnol., 7, 269-271.
Islam, M.S., Drasar, B.S. and Bradley, D.J. 1990 Long-term persistence of toxigenic
Vibrio cholerae 01 in the mucilagenous sheath of a blue-green alga Anabeana variabilis.
J. Trop. Med. Hygiene, 93, 396-401.
Islam, M.S., Miah, M.A., Hasan, M.K., Sack, R.B. and Albert, M.J. 1994 Detection of
non-culturable Vibrio cholerae 01 in a blue-green alga from aquatic environment in
Bangladesh. Trans. Royal Soc. Tropical Med. Hyg., 88, 298-299.
Jüttner, F. 1988 Biochemistry of biogenic off-flavour compounds in surface water. Wat.

Sci. Technol., 20, 107-116.
Kos, P., Gorzo, G., Suranyi, G. and Borbely, G. 1995 Simple and efficient method for
isolation and measurement of cyanobacterial hepatotoxins by plant tests (Sinapis alba
L.). Analyt. Biochem., 225, 49-53.
Kenefick, S.L., Hrudey, S.E., Prepas, E.E., Motkosky, N. and Peterson, H.G. 1992
Odorous substances and cyanobacterial toxins in prairie drinking water sources. Wat.
Sci. Tech., 25(2), 147-154.
MacKintosh, C., Beattie, K.A., Klumpp, S., Cohen, P. and Codd, G.A. 1990
Cyanobacterial microcystin-LR is a potent and specific inhibitor of protein phosphatases
1 and 2A from both mammals and higher plants. FEBS Letters, 264, 187-192.
Mahmood, N.A. and Carmichael, W.W. 1986 Paralytic shellfish poisons produced by the
freshwater cyanobacterium Aphanizomenon flos-aquae NH-5. Toxicon, 24(2), 175-186.
Paerl, H.W. and Gallucci, K.K. 1985 Role of chemotaxis in establishing a specific
nitrogen-fixing cyanobacterial-bacterial association. Science, 227, 647-649.
Persson, P.E. 1983 Off-flavors in aquatic ecosystems - An introduction. Wat. Sci.

Technol., 15, 1-11.
Pilotto, L.S., Douglas, R.M., Burch, M.D., Cameron, S., Beers, M., Rouch, G.R.,
Robinson, P., Kirk, M., Cowie, C.T., Hardiman, S., Moore, C. and Attewell, R.G. 1997
Health effects of recreational exposure to cyanobacteria (blue-green) during recreational
water-related activities. Aust. N. Zealand J. Public Health, 21, 562-566.
Shimizu, Y., Norte, M., Hori, A., Genenah, A., Kobayashi, M. 1984 Biosynthesis of
saxitoxin analogues: The unexpected pathway. Am. Chem. Soc., 106, 6433-6434.
Siegl, G., MacKintosh, C. and Stitt, M. 1990 Sucrose-phosphate synthetase is

dephosphorylated by protein phosphatase 2A in spinach leaves. FEBS Letters, 270,
198-202.
Turner, P.C., Gammie, A.J., Hollinrake, K. and Codd, G.A. 1990 Pneumonia associated
with contact with cyanobacteria. Brit. Med. J., 300, 1440-1441.
WHO 1993 Guidelines for Drinking-water Quality, Second edition. Volume 1,
Recommendations. World Health Organization, Geneva.
WHO 1996 Guidelines for Drinking-water Quality, Second edition. Volume 2, Health
Criteria and Other Supporting Information. World Health Organization, Geneva.
WHO 1998 Guidelines for Drinking-water Quality. Second edition, Addendum to Volume
2, Health Criteria and Other Supporting Information. World Health Organization, Geneva.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 6. SITUATION ASSESSMENT, PLANNING AND

MANAGEMENT
This chapter was prepared by Jamie Bartram, Mike Burch, Ian R. Falconer, Gary Jones
and Tine Kuiper-Goodman
Chapters 3 and 4 of this volume present evidence that cyanobacterial toxins can and do
cause significant adverse health effects. These effects are associated with the
occurrence of cyanobacterial blooms. As described in Chapter 2, such blooms are a
natural occurrence, but occur more frequently in waters which have been subject to
certain forms of human interference. The most important types of interference are
enrichment of waters with nutrients (eutrophication) from point sources such as
municipal wastewater outlets and non-point sources such as agriculture, and the
damming of rivers (which increases retention time and exposure to sunlight). Chapter 2
also describes how, although blooms are more frequent and severe in eutrophic waters,
they may occur in waters which would be considered in many world regions to be of
good or acceptable quality. Of more concern is the fact that the available evidence
concerning trends in eutrophication indicates that the situation is severe world-wide (see
Table 6.1), although it is improving in some regions whilst deteriorating in others.
The purpose of this chapter is to assist those concerned with assessing or managing the
potential risks to human health arising from toxic cyanobacteria. It is intended to serve
as a guide to readers who are confronted with immediate questions and issues related to
risk management, whether arising from an event or because of the suspicion of a
potential hazard. It also provides an axis to assist the reader in using other parts of this
book and deals with four major areas:
• The overall risk management framework.
• Situation assessment (Is there a problem? Would a problem be detected if it existed?

How can the severity of the problem be interpreted in relation to other demands on
resources?).
• Management options (What types of management actions are available? What are
their basic characteristics?).
• Planning for management (How should a management, contingency or emergency
response and investigation plan be put together?).
Table 6.1 Perceived eutrophication problem in different continents and countries

Reservoirs, rivers Marine
Natural Estuaries, lagoons
Region/country and irrigation coastal
lakes and closed areas
systems waters
AFRICA
Central + ++ +
North + ++ +
South ++
CENTRAL AMERICA
Caribbean + + +
Guatemala/Nicaragua +
Mexico + ++
NORTH AMERICA
Canada ++ + +
USA ++ ++ ++ +
SOUTH AMERICA
Argentina/Chile + ++ + +
Brazil + ++ ++ +
Columbia/Ecuador/Peru + ++ + +
Venezuela/Suriname + + ++
ASIA
China ++ + +
India/Pakistan + ++ +
Indochina + +
Indonesia/Philippines ++ +
Japan ++ + ++ +
OCEANIA
Australia/New Zealand ++ ++ ++ +
EUROPE (EU countries)
Belgium +
Denmark ++ +
France ++ + ++
F. Germany, Fed. Rep. ++ ++ + +
Greece + + +
Ireland ++ +
Italy ++ ++ + ++
Netherlands ++ ++
Portugal ++ +
Spain ++ +
UK ++ ++
EUROPE (other countries)
Austria ++
Former Czechoslovakia +
Finland ++ + +
Former German Dem. + ++ + +
Rep.
Hungary + +
Norway ++ + ++ +
Poland ++ +
Romania +
Sweden ++ + ++
Switzerland ++
Former USSR + ++ + +
Former Yugoslavia +
+ Identified problems
++ Serious problems
Source: Adapted from Earthwatch, 1992
6.1 The risk-management framework

Risk management is a relatively recent discipline, in which developments are still
occurring very rapidly. Various model schemes for risk management have been
portrayed, most of which have some common elements. These include the need for an
information base upon which to make decisions; the need to make decisions based on
often inadequate information; the need to compare and "value" different and often very
diverse outcomes with one another; and the need for broad participation in the process
in all its stages. One schematic example is presented in Figure 6.1. This representation
is different from many others because it completes a circle, showing the feedback of
policy evaluation into hazard identification and prioritisation, and thereby into improved
risk assessment and eventually into implementation of revised (improved) policy. It also
places communication as a central and two-way process, indicating that it has an
important role throughout risk management.
The implementation of risk management will vary greatly according to the political, social
and economic context in which it takes place. Whilst it is often seen as a highly rational
process, it should be recognised that the scientific basis for many of its elements is
actually often extremely weak (the status of knowledge regarding some of the
cyanobacterial toxins as outlined in Chapters 3 and 4 is a good example of this) and
procedures for valuation of health effects are generally poorly developed. More
importantly, no scientific assessment will support effective risk management if it fails to
address the perceptions and priorities of the society concerned. Public acceptance of
cyanobacterial-related turbidity and discoloration in recreational waters in some
countries illustrates this point. Thus, a commitment to communication and participation is
an essential element of any effective risk management plan.
Figure 6.1 The risk management cycle (Adapted from Soby et al., 1993)
The movement towards a more comprehensive approach to risk management is

inhibited by four key impediments, sometimes referred to as "institutional failures"
(adapted from Gerrard, 1995):
• Data limitations are of two kinds: a lack of historical record on which to base current
change and to recognise trends, and the inadequacy of scientific understanding on
which to base judgements about cause and consequence. Such data limitations are
common in the management of cyanobacterial and eutrophication problems. A common
question that faces water managers is whether a cyanobacterial problem is getting
worse, due for example to cultural eutrophication, or whether the problem has always
been present but has remained unnoticed or, perhaps, unreported. Following
widespread dissemination of information concerning the potential toxicity of
cyanobacteria, blooms may start to be reported more frequently by, or to, relevant
authorities.
• Poor frameworks for analysis impede comprehensiveness in assessment and decision-

making necessary to make informed judgements. Cyanobacteria are good examples.
Until recently, a general lack of knowledge and poor communication by scientists were a
serious impediment to sound management action - a constraint which is now being
overcome in some countries.
• Inadequate regulatory principles which lay too much discretion at the door of the safety
official and which lead to discrepancies in levels of safety provided for different groups in
society (according to their environmental circumstances and income levels). In some
countries, cyanobacteria-infested drinking water sources are the only ones available to
the poor or geographically isolated.
• Insufficient consultation procedures restrict participation of different interests that have

a legitimate role to play in the determination of risk and its management (the important
role of multi-agency and vested interest committees or "task forces" in the effective
management of toxic cyanobacterial blooms is outlined in section 7.1). Lack of
communication between government and industry sectors (including, for example, water
suppliers) is especially detrimental in managing cyanotoxin issues. For example, health
impact assessment generally requires health sector participation, resource management
is generally under the control of environmental and/or local authorities, and abatements
such as water treatment may be undertaken by local authorities or the private sector.
6.1.1 Communication and participation
In many cultures, certain characteristics lead to some risks being perceived as high
priorities (that is they are little accepted or tolerated). These include:
• Risks which are "new" or unfamiliar to the population.
• Risks which are perceived to have been caused by a (especially distant and
unaccountable) third party.
• Risks for which the consequences are poorly characterised and understood.
• Risks with potentially widespread or catastrophic and irreversible consequences.
• Risks outside the control of the individual.
• Risks in which the population or individual has had little possibility of influencing official
responses.
Risks with these characteristics are perceived as high priorities when compared with
other risks which are perhaps of equal or greater importance to health but which are
voluntarily undertaken and familiar. The example of the social acceptance of smoking
across much of the globe, despite widespread knowledge of the adverse health effects
associated with it, is a good example of this.
A brief review of the public perception of cyanobacteria as a health hazard indicates that
they will often be perceived as a high priority, perhaps higher than rational analysis of
the (limited) available data would otherwise indicate. Newspaper headlines have been
printed proclaiming "Water More Toxic Than Cyanide" after cyanobacteria were found in
a particular water body. While cyanotoxins are, on a per-unit-mass basis, more toxic
than cyanide, this does not mean that drinking water containing a low concentration of
cyanotoxins is anywhere near as dangerous as drinking a lethal solution of cyanide, as
the newspaper headline implied. Being a "new" issue, with which the general population
is not familiar, the blame may be assigned to municipalities, public service utilities, or
farmers. Individual choice does not enable risk avoidance (at least for drinking water
exposure) and entire communities may be affected through their drinking water supply.
Experience indicates that, in some countries, public concern regarding cyanobacteria
has often been greater than the corresponding concern from the professional community.
In other countries exactly the opposite perception exists - the public has become
accustomed to "green" or "smelly" water and may disbelieve scientific warnings of risk.
Improved understanding of risk assessment and management, and an improving

understanding of effective intervention, has demonstrated that differences in public and
professional perceptions are not an error by either party that needs to be corrected (for
example through attempts at altering public perception or discrediting scientific attempts
to assess objectively and to describe risks), but are legitimate viewpoints to be taken into
account through open channels of communication and especially through enabling
public participation in risk management. It should be recalled that the factors which lead
to a low public tolerance of risk often include the lack of familiarity (i.e. the "newness")
and the inability to exert influence. Effective communication and provision of
mechanisms for participation will, therefore, often address these directly.
Several sections of this book provide guidance concerning public participation and
communication in relation to cyanobacterial hazards. These actions are an important
aspect of all types of planning, whether for preventive management (section 6.4.1), for
contingency planning (section 6.4.2) or planning in response to an incident (section
6.4.3). They are also discussed in more detail in Chapter 7.
6.1.2 Hazard identification and hazard characterisation
In the context of risk management, a hazard is usually understood to mean the property
of a substance (or activity) to cause harm. Many substances are hazardous but will not
necessarily lead to harm unless circumstances lead to human exposure. Even after
exposure, an adverse health outcome is not necessarily certain, but rather a probability.
A hazard is therefore defined as an intrinsic property of a biological, chemical or physical
agent to cause adverse health effects under specific conditions. This definition implies
some certainty that under similar conditions the agent will cause similar adverse health
effects.
The term "risk" refers to a probability that exposure to a hazard will lead to a specific
(adverse) health outcome and is usually expressed as a frequency in a given time. Thus,
for example, the WHO Guidelines for Drinking-water Quality (WHO, 1993) define
guideline values as concentrations of specific chemicals estimated to lead to a negligible
additional risk for the consuming population. Where such a point of reference is adopted,
it is generally referred to as the "acceptable risk", although the term "tolerable risk" is
preferred by some people (because the risk is never really acceptable, but it is tolerated).
Hazard identification involves the identification of known or potential adverse health

effects associated with a particular agent, based on studies conducted under specific
conditions, such as the species tested and the experimental conditions. Epidemiological
studies and animal toxicity studies are ranked as providing the greatest predictive
information, and this is followed by in vitro (test tube) assays and qualitative structure
activity relationships (QSAR) predictions.
Hazard characterisation is the extrapolation phase of risk assessment aimed at making a
predictive characterisation of the hazard to humans based on animal studies (species
extrapolation) under low exposure conditions (extrapolation from high to low dose). The
endpoint of hazard characterisation is the estimation of a "safe dose" such as a tolerable
daily intake (TDI) or equivalent. In general, TDIs are only determined when there is likely
to be a threshold in the relationship between dose and effect, based in part on
theoretical knowledge of the mechanism of action. For genotoxic carcinogens it has
been accepted that there is no threshold dose below which effects, such as initiation of
the carcinogenic process, will not occur. When such chemicals cannot be completely
avoided (such as some natural toxicants and contaminants), mathematical models (most
of which presume linearity at low doses) have been used to estimate, through
extrapolation, the possibility of adverse effects at low doses. The dose corresponding to
a risk level of 1 in 1 × 105 or 1 in 1 × 106 has been considered as posing a negligible or
tolerable risk.
With cyanotoxins, risk assessment is made more difficult by the paucity of scientifically-
sound toxicological and epidemiological studies (see Chapter 4). The available animal
data are limited, particularly in the case of chronic or long-term effects of cyanotoxins.
The lack of data is reflected in the fact that a WHO guideline has been agreed only for
one group of cyanotoxins (i.e. the microcystins, see Chapter 5) and only as a provisional
guideline value for the toxin most commonly studied (i.e. microcystin-LR). Uncertainties
about the environmental fate of cyanotoxins (for example, to what extent the toxins are
accumulated in fish and shellfish that may be consumed by humans, see Chapter 3) add
to the difficulty.
6.1.3 Exposure assessment and risk characterisation
Exposure assessment can relate to exposure to cyanobacterial toxins during an

outbreak of disease, or it can be an assessment of the likely exposure of people through
consuming drinking water or swimming in lakes known to contain certain levels of
cyanotoxins or cyanobacteria.
Risk characterisation is the qualitative and/or quantitative estimation, including the

attendant uncertainties, of the severity and probable occurrence or absence of known or
potential adverse health effects in an exposed population. This estimation is based on
hazard identification, hazard characterisation and exposure assessment. If it is
calculated as the "probable risk" (such as the number of persons in a population that are
expected to get cancer from exposure to a toxic chemical), the estimated risk takes on
more meaning than it deserves, because of uncertainties in the process. Alternatively,
risk characterisation can be taken as establishing levels of daily exposure over a lifetime
at which the risk is "negligible" (see section 4.2).
6.1.4 Policy development
The processes of hazard identification, hazard characterisation, exposure assessment

and risk characterisation may be readily viewed as rational, scientific activities. In
contrast, policy development takes account of both rational assessment and societal
values. It therefore requires the valuation of a specific health outcome (such as skin rash,
gastro-enteritis, cancer or death). Most definitions of risk assessment therefore combine
a frequency estimation with some valuation of the seriousness of the consequences.
Whilst several approaches have been proposed for the rational comparison of the
"value" of different adverse health outcomes it should be recalled that such valuation is
principally driven by public perception and societal concerns, and rational analysis may
be of very limited relevance. Experience with the enforcement of recreational water
quality standards leading to restrictions on bathing has shown that public reaction may
vary between the extremes of proclaiming that "there is nothing wrong with the water, we
have been swimming in it for years without any illness, the authorities do not know what
they are talking about" to "this water has killed a pet dog and must be doing me harm too
- something must be done about it immediately".
Knowledge of the characteristics of a hazard, the local occurrence of the hazardous

conditions and an assessment of the seriousness of the outcomes of exposure, provides
the basis for development of policy. However, other factors should be considered. These
may include: the seriousness of other hazards and associated health outcomes that
might compete for limited resources; the cost and effectiveness of remedial and
preventative actions; and the availability of technical solutions and of experience in their
application in the country or region. For effective policy development the above-
mentioned factors must be brought together. In most circumstances this is best done in
a forum which enables participation of all concerned parties including, for example,
water utilities, professional associations, public representatives and experts in the topic
under consideration (see section 7.1).
6.1.5 Policy implementation
One of the most frequent failures of policy development is the formulation of legislation
in the absence of consideration of its sound and sustainable implementation. In the field
of water supply, there has been an increasing recognition by governments of the general
need to ensure availability of water supplies, rather than costly treatment only for
favoured localities.
A number of actions are available to governments with which to support policy

implementation, the most obvious being regulatory enforcement. This implies a capacity
to monitor the implementation of the regulations and a will to enforce compliance when
the regulations are not met. Other mechanisms for implementation include voluntary
codes, conflict resolution, economic instruments and public information and participation
(see Chapter 7). Most frequently, multiple actions will be used and will interact with one
another to encourage safer behaviours and practices.
6.1.6 Evaluation of management plans and actions
The types of actions described above should, ideally, be combined in appropriate

preventative, remedial and contingency plans according to local circumstances and
should constitute part of a declared policy for control of adverse human health impacts
from cyanobacteria. Thus, they would be integrated into water sector policy more
generally, established at national level and operated at local levels supported by
legislative frameworks, trained staff and effective institutions with proactive strategies for
awareness raising and information dissemination (see Chapter 7).
These plans must be periodically reviewed. In particular, after an incident, it is useful to

reflect on the parts of the contingency plan that worked well and those parts that did not
function effectively. With long-term, preventative or remedial management actions it is
crucial to put in place a system of monitoring that will enable the efficacy of the actions
to be evaluated. There may be long delays before any benefit of a management action is
perceived in lakes and reservoirs - in some cases 5-10 years or longer.
It is important to consider the tenets and principles of Adaptive Environmental

Management (AEM), particularly with long-term management actions. Decisions and
actions are often made on the best available advice, recognising that with any complex
ecological problem, such as with a toxic cyanobacterial bloom, the information available
to guide a specific management action will always be limited and inadequate. It is
important, therefore, to monitor the outcome of the management action, and then to
modify or revise the action depending upon the response, or as new scientific
information and techniques become available.
6.2 Situation assessment

Assessing the risk posed by toxic cyanobacteria, or the potential for development of
cyanobacterial blooms, and linking this to effective measures for the protection of public
health within available resources, is complex. Situation assessment may be proactive,
for example to determine whether contingency planning is required or to inform long-
term action, such as pollution control to minimise bloom formation; or it may be reactive,
for example to assist in interpretation of specific local events or conditions to inform
emergency or incident response.
An important factor in situation assessment is understanding the adequacy of available

information with which to make the assessment. In many, if not all, cases
epidemiological evidence of cyanobacteria-related health effects would not be available
because of poor or non-standardised, or poorly differentiated, diagnoses; lack of
awareness of cyanotoxins as potential causes of symptoms; and inadequate reporting
systems, research programmes, or information analysis. In most situations a limited
range of information is available to assist in identifying whether a problem or potential
problem exists. The types of information possibly available to aid in assessment are
summarised in Table 6.2.
6.2.1 Drinking water supply information
The monitoring of water bodies and supply systems for cyanobacteria and cyanotoxins is
not yet common practice in most countries in the world. There are a number of critical
control points in the potable water supply system where testing for cyanotoxins and
intact cyanobacterial cells should be carried out if significant cyanobacterial populations
occur in the source water. These may include the water storage reservoir or river; the
treatment plant raw water intake; key points in the treatment process; and in the final
treated drinking water, depending on local circumstances. Details of such critical points
are given in Table 5.1 with more detailed monitoring information being provided in
Chapters 10-13.
One requirement is to be aware of which members of the community receive drinking

water from which water supply. This information is usually readily available in cities and
large towns but this may not be the case in rural areas. In addition, the existence of
water treatment systems and their effectiveness in cyanotoxin removal (see Chapter 9)
should be ascertained. For small community supplies there may be little or no water
treatment, and this must be taken into account by health authorities when assessing any
potential risk situation.
To aid in making a rapid situation assessment based on available water supply

information, and critical control points as outlined in Table 5.1, the protection categories
outlined in Figure 6.2 should be consulted. This schematic flow diagram may be
particularly useful in those countries and regions where little or no consideration has
been given previously to the potential risk posed by toxic cyanobacteria in drinking water
supplies. The flow diagram focuses on the raw water supply and treatment stream, with
attention being paid to the likelihood of toxin release from intact cells (either in the
storage reservoir or during the transport network to the treatment plant), removal of
intact cells (and their toxins), and the capacity for removal or destruction of dissolved
cyanotoxins.
Table 6.2 Types of information of assistance to assess whether a health problem from a
cyanobacterial bloom exists or is likely in a particular water body
Observation Sources of information Management options

Potential for bloom Water quality monitoring data (nutrients, Basis for proactive
formation temperature, etc.) management
History of bloom formation Cyanobacterial blooms may follow marked Can inform proactive
seasonal and annual patterns management
Monitoring of Turbidity, discolouration, cell microscopic Possible only during
cyanobacteria and/or identification, cell counts and toxin analysis event and enables only
cyanotoxins provide increasingly reliable information reactive management
"Scum scouting" In areas of high public interest the general Possible only during
public and untrained agency staff may play event and enables only
a role in identifying and reporting obvious reactive management
hazards such as scums
Reporting of animal Requires both volition and a mechanism Possible only during
deaths and human illness for data collection which may not exist event and informs only
reactive management
Epidemiological detection Requires both effective reporting and Normally well after an
of disease patterns in the large-scale effects before detection likely event; can inform future
human population management strategies
6.2.2 Recreational water information
Assessing the potential hazard at recreational water bodies is complicated by the

numerous points at which people may enter or move around the water and by the
heterogeneous and often rapidly changing distributions of cyanobacterial populations,
particularly scums which may be blown around by the wind. As noted in section 5.2.2,
concentrated scums pose the greatest risk to bathers. However, monitoring of their
potentially rapid formation and dispersal, as well the monitoring of all water bodies used
for recreational purposes, is rarely likely to be achievable, nor is it appropriate unless
conditions indicate otherwise. Therefore, additional approaches to assessing
cyanobacterial risks at recreational sites are needed.
Figure 6.2 Simple and rapid assessment of the degree of protection afforded by
drinking water supply systems. Note that this figure should be used with, rather
than instead of, more detailed health investigations
Visual monitoring for scums around bathing areas considered to be at-risk is crucial for
protecting human health and is quite easy to perform (see Chapters 11 and 12).
Operators of recreational sites can be involved in assessing changes of situation.
Furthermore, participation of the public in monitoring for cyanobacteria, and particularly
for scums, is a very effective approach (see Table 6.2 under "scum scouting"). This
involves education on appearance and toxicity of scums, as well as on recognition of
cyanobacteria at high cell densities when they do not form scums (see Chapter 7). An
example of a message to convey this may be:
"If you walk into the water up to your knees, carefully, without stirring up sediment, and
cannot see your feet because of a greenish discoloration, don't swim and inform the
local authority using the following telephone number".
In many countries and regions, the current situation is that public awareness is negligible
and knowledge of the risk posed by bathing in cyanobacterial scums is lacking. Eye
witness accounts in Australia, Germany and Japan report swimmers deliberately
covering themselves in cyanobacterial scum for the sake of an "exciting" photograph,
and one well-documented poisoning case in the UK resulted from army cadets
undertaking kayak "Eskimo rolls" in cyanobacterial scum. If public education not only
addresses personal health risk but also stimulates a sense of responsibility in reporting
scums, valuable support for risk assessment at recreational sites can be gained (see
Chapter 7).
6.2.3 Environmental information
Environmental information, including physical, chemical and biological data can aid in
the prediction and assessment of the likelihood, or presence, of cyanobacterial mass
developments, including blooms, scums and mats. It can also help predict and assess
types of cyanobacterial toxins and their location.
Historical records and local knowledge
Consultation of historical records, if available, can indicate whether a water body has
been prone to cyanobacterial bloom development. Useful information is sometimes
available from the local community, including descriptions of the water body and
examples of human health incidents, livestock mortalities and fish-kills associated with
blooms and scums. However, a lack of historical and local evidence of blooms cannot be
taken as assurance that cyanobacterial blooms have not occurred, or will not occur,
because data may be lacking and recognition of cyanobacterial blooms and associated
problems may have been inadequate (Skulberg et al., 1984) and because increases in
cultural eutrophication may be ongoing.
Physical data
Environments in which various species of cyanobacteria can flourish, together with the
physical conditions that promote such developments are outlined in Chapter 2. In the
case of the many species of planktonic cyanobacteria whose growth is favoured in warm,
thermally-stratified environments, the onset of favourable growth conditions is indicated
by a rise in surface water temperature above about 18 °C and the establishment of
persistent thermal stratification (Reynolds, 1984; NRA, 1990). For example, the
Queensland Department of Natural Resources in Australia undertook a two-year survey
of thermal stratification in its major water supply reservoirs to aid in its assessment of
reservoirs potentially at risk of developing cyanobacterial blooms. The study enabled
reservoirs to be categorised as seasonally strongly stratified, weakly stratified or well
mixed, with the strongly stratified bodies being considered most at risk. In addition, the
seasonal timing and persistence of stratification was used as a trigger to increase
monitoring effort (i.e. as an indicator of when to switch from monthly to fortnightly or
weekly sampling) (Chudek et al., 1998).
Hydraulic mixing and transport processes
The ratio between the depth of the mixed layer and the depth to which sufficient light for
photosynthesis penetrates, strongly influences cyanobacterial mass development and
the extent to which the populations may be dominated by particular cyanobacterial types.
Data on flushing rates in lakes as well as river flow rates are useful because planktonic
cyanobacteria do not usually attain high population densities in highly flushed
environments with retention times (i.e. the time it takes for the water volume to be
exchanged once) of less than 5-10 days, or in the open channels of flowing rivers. If river
flows are reduced due to drought and/or excessive abstraction of water, cyanobacterial
bloom development can be anticipated provided nutrient concentrations and light
penetration are adequate (e.g. Bowling and Baker, 1996). Section 2.3 on cyanobacterial
"ecostrategists" provides further details of this, together with section 8.5 on the
hydrophysical control of cyanobacteria.
Chemical data
The mass development of cyanobacteria is dependent on the nutrient concentrations

(especially phosphorus and nitrogen) in a water body. The relationship between mean
chlorophyll a concentrations (as a simple measure of cyanobacterial and planktonic algal
biomass) and annual mean phosphorus concentrations provides a valuable (but easily
misused) basis for assessing the likelihood of planktonic biomass development; this is
discussed critically in Chapter 8 (Vollenweider, 1968; Vollenweider and Kerekes, 1980).
Inputs and concentrations of nitrate and ammonia should also be considered because
these can influence growth rates, maximum biomass and phytoplankton species
composition. The ability of several toxigenic cyanobacterial genera to fix dissolved
atmospheric nitrogen under aerobic conditions (e.g. Anabaena, Aphanizomenon,
Nodularia), but not others (e.g. Microcystis, Oscillatoria, emphasises the need to take
physical, chemical and biological factors into account when attempting to predict the
likelihood of cyanobacterial mass development.
Biological data
Long-term and within-year monitoring records are useful in contributing to the

assessment of the likelihood, onset and persistence of cyanobacterial mass
developments. Such long-term data sets are not widely available, and often their value
may not be apparent to managers who may see long-term monitoring as difficult to
justify. This may be the case particularly in water bodies that have no history of
cyanobacterial problems. Health authorities responsible for the quality of recreational
waters and drinking water resources in many countries may not be sufficiently informed
of data available in environmental authorities or local research institutions. Establishing
such contacts is strongly recommended for assessing potential cyanobacterial risks.
In addition to monitoring for total phytoplankton biomass (measured by cell counts or as

chlorophyll a concentration) and cyanobacterial genera or species distribution and
numbers (see Chapter 12), information on other biota in the reservoir, lake or river can
be useful. This could include the types and abundance of phytoplankton grazers
(zooplankton) and of zooplanktivorous fish (see section 8.5 for details).
6.2.4 Health information
Information on the health of the population is collected in nearly all countries for the
purpose of providing assistance in the prevention and control of disease. Often included
in this information are records of outbreaks of gastro-enteritis and, where possible, their
causes. Most sources of gastro-enteritis are infectious organisms, although in most
cases of gastro-enteritis a specific cause is not identified. When a substantial outbreak
of gastro-enteritis occurs it will generally be investigated in order to determine the source
and the causative organisms. Faecal, food and water samples may be screened for a
variety of possible pathogens, and only if no pathogens have been identified will the
possibility of toxicity in the water or food be investigated. As awareness of
cyanobacterial toxicity increases, the likelihood of these toxins being considered as a
possible cause for clinical illness increases.
Routine monitoring for the presence of cyanobacterial cells or cyanotoxins in drinking

water is undertaken in only a few countries at present, and then only by some water
supply companies or authorities. In those countries where there is an awareness of the
problem, monitoring of reservoirs after the onset of a cyanobacterial bloom has been
detected is more common. As a result of the absence of routine monitoring, reports of
gastro-enteritis outbreaks that have been later attributed to cyanobacterial poisoning
have been made in the absence of cell counts or toxin measurement at the time of the
event (for examples see section 4.1).
It is unlikely that an outbreak of illness will be related to cyanobacteria in the drinking or

bathing water unless a specific local investigation is conducted. A link between data
gathered by health authorities and cyanobacterial data obtained from water monitoring
will be required if acute gastro-enteritis caused by cyanotoxins is to be understood and
avoided. This may be established by the reporting of monitoring data collected by water
supply agencies to health authorities. In outbreaks of gastro-enteritis in which no
pathogen has been detected, it may be useful to look at the geographical distribution of
cases to see whether the drinking water distribution system is the likely source.
Investigation of the presence of cyanotoxins can follow (if still present), or proactive
investigation for a subsequent bloom can be introduced.
Until cyanobacterial monitoring and cyanotoxin analysis are more widely established, it
will remain difficult to correlate clinical findings with the toxic effects of cyanobacteria.
6.2.5 Other data
In addition to the drinking water, recreational water environment and health information
that may assist in developing a situation assessment, additional information may be
gleaned from veterinary records of animal deaths and post mortem examinations (see
Chapter 4 for likely symptoms and pathologies). While in many cases it may not be
possible to attribute unequivocally animal deaths, or even poor water quality, to toxic
cyanobacteria, consistent relationships between these observations and particular water
bodies at certain times of the year may be indicative of water bodies with potential
cyanobacterial problems.
6.3 Management actions, the Alert Levels Framework

An Alert Levels Framework is a monitoring and management action sequence that water
treatment plant operators and managers can use to provide a graduated response to the
onset and progress of a cyanobacterial bloom. Circumstances and operational
alternatives will vary depending upon the source of the water supply and the analytical
and water treatment facilities available. The managerial response model, presented as a
"decision tree" in Figure 6.3, is based upon the critical control points identified in Table
5.1, the drinking water supply protection categories defined in Figure 6.2, and on an alert
levels framework developed earlier in Australia. The decision tree should be seen as a
general framework, recognising that it may be appropriate to adapt specific alert levels
and actions to suit local conditions in different countries.
The decision tree provides for the assessment of a potentially toxic cyanobacterial bloom,
with appropriate actions and responses, through three "threshold" stages. The sequence
of response levels is based upon the initial detection of cyanobacteria at the Vigilance
Level, progressing to moderate to high cyanobacterial numbers and possible detection
of toxins above guideline concentrations at Alert Level 1. Alert Level 1 conditions require
decisions to be made about the suitability of treated drinking water based on the efficacy
of water treatment and the concentrations of toxins detected (if such measurements can
be made). At very high cyanobacterial biomass levels in raw water, the potential health
risks associated with treatment system failure, or the inability to implement effective
treatment systems at all, are significantly increased. This justifies progression to a
heightened risk situation denoted by Alert Level 2 conditions.
The framework has been developed largely from the perspective of the drinking water
supply operator but is also important for the manager of the raw water supply. The
actions accompanying each level cover categories such as additional sampling and
testing, operational options, consultation with health authorities and media releases. An
important part of the framework is consultation at various stages with other agencies,
particularly health authorities that generally have responsibility to oversee the safety of
water for potable supply.
6.3.1 Vigilance Level
The Vigilance Level encompasses the possible early stages of bloom development,
when cyanobacteria are first detected in unconcentrated raw water samples (see
Chapter 12). The indicative value for the Vigilance Level is the detection of one colony,
or five filaments, of a cyanobacterium in a 1 ml water sample, although this threshold
may be adapted according to local knowledge and prior monitoring history. Taste and
odours may become detectable in the supply as the population develops above the
Vigilance Level, but their absence does not indicate absence of toxic cyanobacteria. For
example, recognition of the highly odorous earthy/muddy smelling compound geosmin,
produced by Anabaena, may occur at concentrations less than 1,000 cells per ml (Jones
and Korth, 1995). In contrast, Microcystis produces weakly odorous compounds that will
be detected only at cell concentrations 100-1,000 times higher than this, and are
therefore unlikely to be noticed in raw water.
Figure 6.3 Decision tree incorporating a model Alert Levels Framework for monitoring and
management of cyanobacteria in drinking water supplies. Note that this framework should
be adapted according to local conditions
The presence of cyanobacteria in low numbers (and sometimes detection of
characteristic tastes and odours) constitutes an early warning for potential bloom
formation; therefore, when the Vigilance Level is exceeded, it is generally appropriate to
increase the sampling frequency to at least once a week, so that potentially rapid
changes in cyanobacterial biomass can be monitored.
6.3.2 Alert Level 1
The Alert level 1 threshold (cyanobacterial biomass 2,000 cells per ml or 0.2 mm3 l-1
biovolume or 1 µg l-1 chlorophyll a) is derived from the WHO guideline for microcystin-LR
(see Chapter 5) and the highest recorded microcystin content for cyanobacterial cells
(Chapter 3). In other words, threshold is the cyanobacterial biomass level, assuming the
species is a potential producer of microcystin, where raw water microcystin
concentration could exceed the WHO guideline. Consequently, Alert Level 1 conditions
require an assessment to be made (preferably based on an analysis of total toxin
concentration in the raw water by a quantitative technique such as high pressure liquid
chromatography (HPLC) or enzyme linked immuno sorbent assay (ELISA), see Chapter
13) as to whether the concentration of potentially toxic cyanobacteria in the raw water
supply can be reduced (e.g. by offtake management) or whether the water treatment
system(s) available are effective in reducing toxin concentrations to acceptable levels
(see Figure 6.2 and Chapter 9).
Alert Level 1 conditions require consultation with health authorities for ongoing
assessment of the status of the bloom and of the suitability of treated water for human
consumption. This consultation should be initiated as early as possible and continue
after the results of toxin analysis on drinking water become available. Clearly, as the
biomass of potentially toxic cyanobacteria increases in the raw water, so does the risk of
adverse human health effects in the absence of effective water treatment systems.
Therefore, on-going monitoring for cyanobacterial biomass and toxin concentrations is
essential. The monitoring programme, which should be at least weekly in frequency,
may be extended throughout the source water body to establish the spatial variability of
the cyanobacterial population and of toxin concentration. It may also be appropriate at
this time to issue advisory notices to the public through the media or other means.
Government departments and authorities with possible interests or legal responsibilities
should also be contacted (see Table 7.1), as should organisations or facilities that treat
or care for special "at risk" members of the public (such as kidney dialysis patients, see
Chapter 4).
For toxic cyanobacteria other than those that produce microcystin (see Chapter 3, or as
indicated by bioassay results) it will be necessary for local health authorities to
undertake a detailed risk assessment based on the inherent hazard of the toxin detected
(Chapter 4), its concentration in treated drinking water, and the exposure expected. If
Alert Level 1 biomass levels are maintained, but toxins or toxicity are not detected in
cyanobacterial or raw water samples, regular monitoring should continue to ensure that
toxic strains or species do not develop over ensuing weeks or months (see Chapter 3).
6.3.3 Alert Level 2
The threshold for Alert Level 2 (cyanobacterial biomass 100,000 cells per ml or 10 mm3 l-
1
biovolume or 50 µg l-1 chlorophyll a, with the presence of toxins confirmed by chemical
or bioassay techniques) describes an established and toxic bloom with high biomass
and possibly also localised scums (although scums may also form under Alert Level 1
conditions). The sampling programme will have indicated that the bloom is widespread
with no indication of a cyanobacterial population decline in the short term. Conditions in
Alert level 2 are indicative of a significant increase in the risk of adverse human health
effects from the supply of water that is untreated, or treated by an ineffective system
(see Chapter 9), even for short-term exposure. The need for effective water treatment
systems and on-going assessment of the performance of the system thus becomes of
heightened importance. Whereas hydrophysical measures to reduce cyanobacterial
growth may still be attempted in this phase (Chapter 8), application of algicides can
enhance the problem by causing high concentrations of dissolved toxins as a
consequence of cell lysis. Whereas filtration systems (possibly combined with
flocculation-coagulation) may have removed cell-bound toxins, dissolved toxin is more
likely to break through treatment systems.
If effective activated carbon or other advanced treatment is not available, Alert Level 2
conditions should result in the activation of a contingency water supply plan which is
appropriate for the operator and the user or community. This may involve switching to an
alternative supply for human consumption, the implementation of contingent treatment
systems, or in some circumstances the delivery of safe drinking water to consumers by
tanker or in bottles. More extensive media releases, and even direct contact with
consumers via letterbox delivery of leaflets with appropriate advice to householders, may
be necessary (see Chapter 7). Where advice is provided to the public because of a
cyanobacterial hazard to human health, it will generally be advisable to emphasise that
the water would still be suitable for purposes such as washing, laundry, toilet flushing,
etc. Withdrawal of a piped drinking water supply because of a cyanobacterial toxin
hazard is usually not justifiable because the adverse health effects resulting from
disruption of supply (e.g. lack of toilet flushing, lack of fire fighting) are likely to outweigh
substantially the cyanobacterial toxin risk itself.
Monitoring of the bloom should continue in order to determine when the bloom starts to
decline so that normal supply can be resumed. The sequence at Level 2 may follow
through to deactivation of alert level conditions with media releases, and advice to
government departments and health authorities to confirm this. It is possible that the
collapse of a bloom, or a management action such as flushing, could lead to a rapid
decline from Level 2 back to Level 1 or beyond. Likewise the sequence might escalate
rapidly, bypassing Level 1 to Level 2, if adequate monitoring and early warning
information is not available. Cyanobacterial populations in natural water bodies,
especially in hot climates, may double in size in less than two days (growth rate, µ, 0.35
d-1). Therefore, monitoring frequency needs to take such potentially rapid population
growth rates into account. Figure 6.4 gives an indication of the rate of change of an
exponentially dividing population at two growth rates observed in field studies of
cyanobacteria.
Figure 6.4 Predicted development of cyanobacterial population from initial

concentration of 100 or 1,000 cells per ml and at growth rates (µ) of 0.1 and 0.3 d-1
(After Jones, 1997)
6.3.4 Overall risk assessment and summary of action-response thresholds
The information presented in Table 6.3 will enable water resource managers, treatment
plant operators and health authorities to make a rapid assessment of the potential risks
associated with increasing levels of potentially toxic cyanobacteria in water (based on
the Alert Levels Framework presented above and in Figure 6.3) and the protection
afforded by the monitoring and water treatment systems in place (based on the
protection categories given in Figure 6.2). The risk category outcomes in Table 6.3 take
into account the possibility of toxin persistence after a bloom has collapsed and the
possibility, therefore, that dissolved toxins are present in the absence of viable
cyanobacterial cells (hence the different risk outcomes for long term compared with short
term, low biomass), and the increased risk associated with treatment system failure at
very high cyanobacterial biomass (Alert Level 2).
Table 6.3 Relative risk categories for adverse human health outcomes due to toxic
cyanobacteria in drinking waters
Protection Cyanobacterial biomass category1

category2 Absent or low Absent or low Moderate - high Very high (Alert
(long term) (short term) (Alert Level 1) Level 2)
I - Low High Very high
II Minimal Low High Very high
III Minimal Low Medium- high High
IV Minimal Low Low - medium Medium
V Minimal Low Low Medium
VI Minimal Minimal Minimal - low Low
VII Minimal Minimal Minimal Minimal - low3
1
The cyanobacterial biomass categories are defined as follows (see Figures 6.3 and
6.5):
Low: Cyanobacterial biomass < 2,000 cells per ml or 0.2 mm3 l-1 biovolume or 1 µg l-1
chlorophyll a; the category is "long term" when based on data compiled over at least a
two-month period, and "short term" when based on data compiled over less than two
months (including analysis of a single sample only and taking into account the risk of
toxin persistence after a bloom collapses)
Moderate - high: Whether for a single measurement or for repeated measurements over
several weeks, cyanobacterial biomass greater than in low category but, < 100,000 cells
per ml or 10 mm3 l-1 biovolume or 50 µg l-1 chlorophyll a
Very high: Cyanobacterial biomass > 100,000 cells per ml or 10 mm3 l-1 biovolume or 50
µg l-1 chlorophyll a, and presence of toxins confirmed by chemical analysis or bioassay
2
Protection categories from Figure 6.2
3
Risk category is greater than minimal because of the increased risk associated with
treatment system breakthrough or failure at high biomass loads
Finally, Figure 6.5 provides an "at a glance" summary of the threshold managerial action
levels for drinking water (Vigilance and Alert Levels) and the health guidance levels for
recreational waters outlined in Chapter 5, Table 5.2.
6.4 Planning and response
The intensity and scope of management actions to control human health hazards
associated with cyanobacteria may vary widely. In the simplest case, a one-off study
may indicate little or no cause for concern or it may lead to an information dissemination
and public awareness and education campaign. An initial situation assessment that
indicates significant risk calls for the establishment of a monitoring programme,
formulation of contingency plans, changes in pollution control and water supply
management, and a more substantial public information and participation strategy.
Which type of response is appropriate in a given situation will depend on a number of
complex and interconnected factors, of which rational assessment of human health risk
is one part. Other important aspects include, for example, technical and institutional
capacities, economic considerations and social values. The principal management
actions that may be taken fall into three main groups:
• Water resource and catchment management. Most long-term actions are intended to
minimise nutrient inputs to water bodies, both from external (watershed) and internal
(sediment) sources, as well as altering the hydro-logic conditions in such a way as to
prevent or reduce conditions favourable for cyanobacterial bloom formation (see Chapter
8 for guidance on long-term measures). Other actions may include manipulation of the
aquatic food web to increase grazing pressure on cyanobacteria (see Chapter 8).
• Remedial measures in drinking water supply. Some of these actions may be seen to
be applicable in short time frames and may therefore be deployed in response to a
specific situation; other are also medium- or long-term actions, such as installing
safeguards (such as treatment steps) in drinking water supply that will assist in
controlling risks while such safeguards remain installed and operated. Other actions are
associated with contingency plans to be activated in case of need (see Chapter 9).
• Public information and awareness-raising. A well-informed public (see Chapter 10),

aware of the potential and real risks of toxic cyanobacteria, is important for many
reasons. These reasons include an improved surveillance capacity through public
participation; for example swimmers and other users can contact local authorities if they
see cyanobacterial scums, and householders can report unusual odours in their drinking
water supply. Several Australian States have community-based "WaterWatch" or
"Stream-Watch" monitoring programmes that are undertaken by high school students
and community groups. Experience has shown that more aware communities, which are
adequately informed by appropriate authorities, will be less likely to react inappropriately
in the event of a bloom.
Figure 6.5 Summary of managerial action levels for drinking waters (see Alert
Levels Framework decision tree in Figure 6.3) and for bathing waters (see
Guidance Levels in Table 5.2). Note that for bathing waters, the special Guidance
Level 3 (scum formation) can be achieved during calm weather conditions at open
water biomass levels similar to Guidance Level 1
Few remedial measures specific to recreational water sites are available in addition to
those for minimising bloom formation, increasing public awareness, and informing the
public. In some instances, fencing-off or the placing of a boom to prevent accumulation
of scums may be considered for densely used and highly impacted areas, in addition to
public information and awareness-raising.
Proactive management, that is taking action before significant cyanobacterial

proliferation has occurred, is generally preferable to reactive (or crisis) management,
that is placing controls in place once significant cyanobacterial proliferation has begun.
Few countries presently operate monitoring programmes capable of identifying
potentially hazardous conditions and early warning systems have not as yet been
developed (for example giving several weeks notice of impending toxic cyanobacterial
hazards), even though previous monitoring data may indicate annual "high risk" months
or periods. In practice, therefore, management of cyanobacterial hazards typically
occurs in three ways, preventative management plans, contingency plans and planning
in emergency situations.
6.4.1 Preventative management plans
Management to prevent human health effects will typically involve prevention and
minimisation of cyanobacterial blooms, deployment of appropriate drinking water
treatment where required, and controls on recreational water use in order that human
populations are not exposed to significant risks. It will also involve general contingency
planning, which is considered in section 6.4.2.
History has shown that single-action management plans tend to have poor success rates;
one example of this was the ban during the 1970s in several European countries on the
use of detergents containing phosphate. Where only this action was applied, little
success was seen in the abatement of cyanobacterial and algal blooms because other
sources of phosphorus pollution were largely ignored (including the remaining 50-70 per
cent of phosphorus in sewage that does not arise from detergents).
Reduction of nutrient pollution below threshold values which control cyanobacterial

bloom formation is highly effective and sustainable. However, many water bodies require
large reductions in nutrient loads, and the implementation of the necessary measures is
likely to take a number of years. Furthermore, the high nutrient concentrations within
such water bodies may decline only slowly in response to reduced external loading,
particularly if water exchange rates are low or release from sediments is high. Thus, total
prevention of cyanobacterial bloom development may require extended recovery time
spans, and may even be unachievable in naturally eutrophic systems. In such situations,
bloom minimisation will generally be accompanied by contingency planning and, if
blooms are expected to occur, by application of appropriate drinking water treatment
either continuously or at times of cyanobacterial occurrence in the source water.
Preventative management may effectively address the hydrophysical conditions of

cyanobacterial growth through the use of hydraulic management (flushing, artificial
mixing), or reduce drinking water intake concentrations by offtake management.
An effective approach to preventative management may be changing the drinking water

source (where this is feasible). This approach was illustrated by the change from using
water in shallow eutrophic ponds and ditches to using groundwater in China (see Box
5.3).
There is no single formula that can be followed to compile a good preventative

management plan. However, key elements may include:
• The convening of a multi-agency and multidisciplinary committee to develop the plan

and to co-ordinate its implementation.
• Development of comprehensive policy relating to point and diffuse source pollution

control, and for the regulation of river flows and reservoir management for the prevention
of cyanobacterial problems.
• Compilation or review of relevant technical information (such as for bloom prevention

and management, drinking water treatment and recreational water management), and
the involvement of key technical (e.g. scientific and engineering) personnel to provide
expert advice.
• Development of procedures (including financial and institutional procedures) for

implementing key actions arising from the plan.
• Development of monitoring systems to determine the effectiveness of management
actions.
• Establishment of response mechanisms for modification of action in the light of

feedback on management plans.
• Development of means for effective communication between agencies and with the
public and media.
Two examples of how preventative plans may be co-ordinated are provided in Boxes 7.2
and 7.4.
6.4.2 Contingency plans
Plans and actions for prevention of health hazards arising from cyanobacterial blooms
should aim to prevent and curtail blooms and ensure that plans are in place to respond
to blooms when they do occur. Planning for such events is an important part of the
overall strategy for managing health hazards associated with toxic cyanobacterial
blooms.
Contingency plans are normally developed and managed at a local or regional level.
National and regional authorities may, nevertheless, have important roles to play in
supporting and facilitating plan formulation and in providing expertise, should an event
occur. Key elements of a contingency plan overlap with those for preventative
management and include:
• The convening of a multi-agency and multidisciplinary committee to develop, maintain

and, if necessary, modify the plan and co-ordinate its implementation if required. The
members of such a committee should be aware of their authority and responsibility as
committee members in advance of an occurrence (see Chapter 7).
• Development of a comprehensive response plan including specific actions at different

alert levels and the responsibilities of different agencies.
• Compilation of a manual or guide for incident response addressing the major areas of
activity and including management, drinking water treatment and recreational water
management, and communication with the public and media.
• Plans for effective communication between key government agencies, health

authorities, water supply agencies, hospitals and the public need periodic testing.
• Ensuring the availability of technical capacity (especially analytical capacity and access
to expertise) to deal with the demands of the contingency plan. The specified experts or
institutions should be able to respond to specific questions in time horizons relevant to
incident response, such as:
• What is the size, extent and toxicity of the bloom?

• If toxic, what types and concentrations of cyanotoxins are present in the drinking water
supply and how are they partitioned between cell-bound and dissolved phases?
• Is an adequate water treatment system in place (see Chapter 9 and Table 6.2 for
details), and if not will the general public be exposed to "unsafe" concentrations of
cyanotoxins?
• Special precautions (e.g. portable water treatment systems, or transported safe water
supplies) may be advisable for "at risk" groups especially susceptible to cyanotoxins,
such as patients with previous acute liver damage.
• Special precautions (usually additional treatment facilities with careful monitoring of

performance) are of crucial importance for hospitals treating patients with kidney dialysis
or intravenous therapy.
• Identification of potential alternative water supplies, preferably from uncontaminated

groundwater (there is the possibility that other local surface water storage facilities may
be suffering simultaneously from cyanobacterial problems) to be exploited in the case of
severe health hazards. This may include plans for transporting clean, treated water from
other areas or deploying portable water treatment systems.
• Establishment of awareness amongst local health practitioners where significant

hazards are believed to exist and development of systems for communicating with them
in the event of an outbreak, including for example advice regarding the possible
symptoms of cyanobacterial intoxication and what treatments are advised.
• Prior agreements regarding standardised press releases and the conditions under
which their release would be justified. Release of information to the media should be co-
ordinated through the main organising committee or task force.
6.4.3 Emergency responses and incident investigation
The actions to be taken in responding to an incident are similar to the elements listed
under "contingency planning" (above). However, time constraints will be greater and,
because of lack of prior preparation, resources may be less available. The risk of
contradictory "messages" from concerned authorities is proportionately greater.
Experience has shown that initiating interagency co-operation, especially between the
drinking water supplier and the health authority, securing an expert opinion on the real
risk to human health and initiating communication with the media and public, are crucial
elements in the earliest stages of responding to an incident. It should be recalled that
whilst true emergencies can arise from cyanobacterial blooms, as was the case with the
Caruaru dialysis tragedy in Brazil (see Box 4.4), an event may be perceived as an
emergency or "environmental crisis" by the public and the media even if this is not the
case from a health viewpoint.
If the incident is deemed to be severe (as was the case in the Palm Island Mystery
Disease, see Box 4.3), a health investigation should be instigated without delay. A
follow-up investigation of an incident will often provide valuable information for both
preventative and contingency planning. It may also lead to substantial improvement of
the regional assessment of hazard due to cyanotoxin exposure, as in the case of one
such investigation from Canada (Box 6.1).
Box 6.1 The Manitoba incident
Deacon Reservoir is the City of Winnipeg's main storage facility for water from Shoal Lake. The
lake is generally considered to be of high quality and its water is only disinfected with chlorine
prior to distribution for drinking. In late August 1993, a large cyanobacterial bloom developed in
Deacon Reservoir. In an attempt to control both cyanobacterial density and taste and odour
problems, municipal officials isolated the reservoir and treated it with copper sulphate. This action
raised concerns that if the bloom contained toxin-producing cyanobacteria, significant quantities
of the toxins may have been released into the reservoir.
Sampling determined that toxin-producing cyanobacteria were not present in the Deacon
Reservoir, but they were present in Shoal lake, the dominant species being Microcystis
aeruginosa. Analysis of water samples indicated that microcystin-LR produced by M. aeruginosa
was present in samples collected from Shoal Lake and from the distribution system, but it was not
present at detectable levels (< 0.05 µg l-1) in samples from Deacon Reservoir. Maximum
microcystin-LR concentrations measured in the raw water of Shoal Lake and in treated tap water
were 0.45 µg l-1 and 0.55 µg l-1, respectively. Subsequent monitoring showed a steep decline in
concentrations, suggesting that higher microcystin-LR levels may have been present earlier in
August 1993.
As the weather during the summer of 1993 was characterised by below-normal temperatures and
above-normal precipitation (conditions that are usually not supportive of cyanobacterial bloom
formation) there was concern that higher levels of microcystin-LR could develop in Shoal Lake
during the more usual relatively hot, dry summers. As a result, Manitoba Environment, in co-
operation with the City of Winnipeg, continued to monitor for microcystin-LR in Winnipeg's water
supply. On six occasions between 1994 and the end of 1996, microcystin-LR was detected at
concentrations ranging from 0.1 to 0.5 µg l-1.
Because Shoal Lake (a relatively nutrient-poor water body) had supported a toxic cyanobacterial
bloom, Manitoba Environment became concerned that toxic blooms might also occur in rural
surface water supplies in southern Manitoba, which are generally more nutrient-rich. A
comprehensive two-year study was conducted on water quality in rural south-western Manitoba
surface water supplies in 1995 and 1996. In the first year of the study, microcystin-LR was found
to be widely distributed in all water supply categories. Rural municipal water supplies had a
higher detection frequency (93 per cent) than on-farm domestic/livestock dugouts (57 per cent),
suggesting that conventional treatment methods were only partially successful in removing the
toxin. Mean concentrations ranged from 0.23 µg l-1 in recreational sites to 0.35 µg l-1 in dugouts
used exclusively for livestock. In the second year of the study, seven rural surface water supplies
were intensively sampled for microcystin-LR. The hepatotoxin was found throughout the entire
sampling period (June to December 1996), sometimes at levels greater than 0.5 µg l-1, which was
the "Emergency Health Advisory Guideline" formulated by Health Canada in response to the
1993 incident.
6.5 References
Bowling, L.C. and Baker, P.D. 1996 Major cyanobacterial bloom in the Barwon-Darling
River, Australia in 1991, and underlying limnological conditions. Aust. J. Mar. Freshwat.
Res., 47, 643-657.
Chudek, E., Horn, A., Joo, M., McLaren, S. 1998 Stratification Patterns in Queensland
Dams. Report by Water Monitoring Group, Queensland Department of Natural
Resources, Australia, 76 pp.
Earthwatch 1992 Chemical Pollution: a Global Overview. United Nations Environment

Programme, Nairobi.
Gerrard, S. 1995 Environmental risk management. In: T. O'Riordan [Ed.] Environmental

Science for Environmental Management. Longman Scientific and Technical, Harlow,
296-316.
Jones, G.J. 1997 Limnological study of cyanobacterial dominance in three South-East

Queensland reservoirs. In: R.R.D. Davis [Ed.] Managing Algal Blooms: Outcomes from
the CSIRO Blue-Green Algal Research Program. CSIRO Land and Water, Canberra, 51-
66.
Jones, G.J. and Korth, W. 1995 In situ production of volatile odour compounds by river
and reservoir phytoplankton populations in Australia. Wat. Sci. Tech., 31, 145-151.
NRA 1990 Toxic Blue-Green Algae. Water Quality Series No. 2. The National Rivers
Authority, London, 125 pp.
Reynolds, C.S. 1984 The Ecology of Freshwater Phytoplankton. Cambridge University

Press, Cambridge.
Skulberg, O.M., Codd, G.A. and Carmichael, W.W. 1984 Toxic blue-green algal blooms
in Europe: a growing problem. Ambio, 13, 244 - 247.
Soby, B.A., Simpson, A.C.D. and Ives, D.P. 1993 Integrating Public and Scientific
Judgements into a Toolkit for Managing Food-Related Risks, Stage I. Literature Review
and Feasibility Study. University of East Anglia, Norwich.
Vollenweider, R.A. 1968 Scientific Fundamentals of the Eutrophication of Lakes and

Flowing Waters with Particular Reference to Nitrogen and Phosphorus as Factors in
Eutrophication. Organisation for Economic Cooperation and Development, Paris.
Vollenweider, R.A. and Kerekes, J. 1980 The loading concept as basis for controlling
eutrophication philosophy and preliminary results of the OECD programme on
eutrophication. Prog. Wat. Technol., 12, 5-38.
WHO 1993 Guidelines for Drinking-water Quality. Second edition. Volume 1

© 1999 WHO
ISBN 0-419-23930-8
Chapter 7. IMPLEMENTATION OF MANAGEMENT

PLANS
* This chapter was prepared by Jamie Bartram, Jessica C. Vapnek, Gary Jones, Lee
Bowling, Ian Falconer and Geoffrey A. Codd
Control of hazards to human health from cyanobacteria is part of a wider picture of water
quality and quantity management, environmental protection, policy formulation and
general development concerns. Implementation of programmes to address
cyanobacteria-related health hazards therefore takes place within the larger framework
of other local, national, and international plans and activities relating to water. That
framework is the subject of this chapter, which examines the principal elements required
to respond effectively to cyanobacteria-related health risks. First the organisations,
agencies and groups that may become involved where there is a cyanobacterial bloom
are identified, then the tools governments can use to implement policy are examined,
together with legislation as an implementation tool. The remainder of the chapter
outlines how professionals and the public may be educated, informed and mobilised.
7.1 Organisations, agencies and groups

7.1.1 Vested interests
Every water body has associated with it a set of persons and organisations with a vested
interest and an involvement. Vested interests can be associated with the level of water
(flooding, hydroelectricity, transportation, recreation); with the contents of the water
(fishing, recreation); with the quality of the water (drinking water/domestic use,
recreation); with the quantity of water (irrigation, industry, transportation); with its motility
(wastewater, transport); with the preservation of the water body itself (ecosystem
maintenance, wildlife conservation); or with some combination of these.
Few rivers, lakes, estuaries and seas can support all vested interests equally and so
compromises amongst conflicting activities are inevitable. Diverting too much water for
irrigation lowers water levels for transportation and may increase the salinity and mineral
content of the water, reducing the numbers offish and causing stagnation. It may also
increase the likelihood of algal and cyanobacterial bloom formation in river reaches
downstream of irrigation reaches, because many cyanobacteria species prefer slow-
flowing river conditions for bloom development. Allowing unrestricted transportation of
dangerous cargoes may damage the ecosystem and wildlife of the surrounding area.
Limiting the construction of dams to preserve a riverine environment may mean
insufficient energy production and water inputs for industry, domestic needs and mining.
Activities particularly likely to increase cyanobacterial hazards in water resources include:
• Land-use, land-use changes and abstraction practices, which may have an impact
through diffuse and point sources of nutrients (such as agriculture) and through
changing run-off regimes (e.g. caused by urbanisation and forest cutting).
• Pollutant discharges and especially nutrient-rich municipal wastewaters and sewage

discharges, which are likely to increase eutrophication.
• Flow control and river flow reduction through, for example, damming, which may
increase opportunities for cyanobacterial proliferation by increasing hydraulic retention
times and the propensity for water column stratification.
• Abstraction, whether constant (such as for drinking water supply), seasonal (such as
for irrigation) or intermediate (such as for power generation), which may tend to
decrease flow rates and also increase opportunities for cyanobacterial proliferation by
increasing retention times.
As with any other activity in relation to water resources, taking action in response to a
cyanobacteria risk requires the recognition and balancing of vested interests. For
example, increasing stream flows to decrease hydraulic residence time will have
beneficial effects on drinking water and animal health, but may have detrimental effects
on recreational use and agricultural (irrigation) interests.
Responding to an environmentally-related health issue, such as a toxic cyanobacterial

blooms, implicates and involves many with vested interests who can be defined broadly
as representing the public, the media, the government (or its agents) and the private or
corporate sector. The largest and most vociferous vested interest group will often be the
general public, who may demand that the quality of their drinking or recreational water
be returned to a level that they deem acceptable, as soon as possible. Sometimes the
water quality standard that the public expects may seem unrealistic (e.g. "The water
used to be crystal clear when I was a child, so why isn't it now?"). Farmers may also
represent a vociferous lobby who may insist that the water necessary for farms and
livestock be again made safe so that productive activity may resume. However, they
may also be a strong lobby against reduction of diffuse pollution input by arguing that
changes in agricultural practices, e.g. fertiliser application, would inevitably reduce crop
yields. Water suppliers often have a strong vested interest in a better resource quality
because that increases the safety of their product and reduces treatment costs, but
agencies may tend to avoid publicity on this issue in order to avoid a negative image and
adverse public comment. Thus, the greatest driving force for action will nearly always
come from the user sector.
The media can play a key role in acting as vehicles for public awareness, thereby
triggering public outcry. They can also be vital for advising the public of potential health
risks (or their absence), for informing the public whether appropriate actions have been
taken and of the efficacy of these actions.
Notwithstanding governments' primary public safety role, where they exist, private
companies involved in water supply and wastewater treatment processes will also have
a key role in any response. Collaboration with such companies will be necessary
whether planning is of an emergency or long-term strategic nature. However, private
companies may view the monitoring data they generate as proprietary or commercial
and their active participation needs to be encouraged, whereas governments may view
information they hold as being in the public domain.
Other important sectors with a vested interest in many countries are tourism and
recreation. Hotels, holiday resorts, water skiing and fishing-based tourism operations, for
example, can be affected profoundly by adverse publicity and recreational restrictions
brought about by toxic cyanobacterial blooms. Involvement of these sectors will be
extremely important if local acceptance of management plans (short- and long-term) is to
be obtained. The fact that private companies operate on a profit motive does not
necessarily place them at odds with governments' public safety goals, because safe
water is also a prerequisite for most, if not all, kinds of tourism and recreation activities,
and is seen increasingly as a priority amongst tourists and other recreational water users.
Where a cyanobacterial bloom occurs, private companies, governments, the media, the
public and others with a vested interest will express and seek support for their particular
interests, some of which may conflict with one another. As a result, multi-sectoral co-
operation is essential.
7.1.2 Multi-sectoral involvement, agencies and their roles
Governments have ultimate responsibility for safeguarding public health and their role
will be to minimise the damage and to assess the circumstances that led to the
cyanobacterial problem (s) in the first place. They can also assist with public awareness
activities, as outlined in section 7.4. Although governments usually take the leading role
in co-ordinating policy amongst vested interests, they may later turn over leadership to a
multi-sectoral committee or group to continue developing longer-term plans or strategies
(see Box 7.1).
Human health issues affect many different aspects of environmental policy and
management and therefore programme co-ordination is both desirable and necessary
for effective action in environment and health. Yet in many countries, environmental
health programmes are assigned to different specialised agencies, making co-ordination
difficult to achieve. Chapter 18 of Agenda 21, the plan of action arising from the United
Nations Conference on Environment and Development outlines the problem as follows:
"The fragmentation of responsibilities for water resources development among sectoral
agencies is proving, however, to be an even greater impediment [to effective action]...
than has been anticipated. Effective implementation and co-ordination mechanisms are
required" (Anon, 1992).
Box 7.1 Development of the New South Wales Blue-Green Algal Task Force and New South
Wales Algal Coordinating Committee
In 1991, an extensive toxic cyanobacterial bloom along the Barwon/Darling River system in
Australia triggered the implementation of policy for reactive and planned management. The
Barwon arises in the Great Dividing Range of Eastern Australia and flows southwest for 2,735 km
until it joins the Murray River. The Barwon/Darling River runs through an arid and seasonally very
hot environment where temperatures of 40-45 °C occur during the summer; flow is regulated for
irrigation and the supply of drinking water.
Flow in the Barwon/Darling River is controlled by the release of water from large reservoirs in the
mountain catchment area, by numerous weirs along the length of the river, and by an extensive
off-river storage towards the downstream end of the system. The release is determined largely by
irrigation requirements. The river is subject to floods at irregular intervals, usually through
subtropical storms. Rainfall in the catchment was low during 1991, leading to a high demand for
water for irrigation and very low river flows. In late spring (October/November 1991), a massive
bloom of Anabaena circinalis occurred in over 1,000 km of the river. As an example, 245,000
cells per ml were recorded in the weir pool from which drinking water is supplied to the town of
Bourke, New South Wales (NSW). The drinking water in this town was, at the time, chlorinated
river water with no additional treatment. Such supplies are common world-wide. Widespread
scums occurred along the riverbanks where livestock drank.
The bloom came to public attention when the local newspaper printed a picture of several dead
cows in a weir pool, with comments on the water quality. Water samples were collected and sent
to the laboratory for toxicity testing. Intraperitoneal injection of lysed extracts of the Anabaena
cells showed considerable neurotoxicity, with evidence of some neurotoxicity in drinking water
samples. The toxins concerned have subsequently been identified as saxitoxin-type neurotoxins.
The NSW State Government declared a State of Emergency, which enabled rapid action
including the deployment by the Army of portable water treatment plants capable of using highly
contaminated water to produce a safe supply. These units provided dissolved air flotation with
flocculation for cyanobacterial cell removal, followed by filtration and granular activated carbon
adsorption of any toxic organic materials. The water supplied to the affected towns from these
plants had no detectable toxin content. A major public information campaign was rapidly put in
place with local meetings, radio and press coverage. Health officers recommended that
swimming and water sports were discontinued, with variable success. Farms drawing their own
water supply directly from the river were particularly at risk. Livestock deaths along the river were
difficult to quantify, but about 2,000 sheep and cattle deaths were reported. The State of
Emergency ended when heavy rain in the catchment flushed the river and increased turbidity
greatly. No further blooms occurred in the river that summer.
As a consequence of the Barwon/Darling River bloom the State Government established the
NSW Blue-Green Algal Task Force (BGATF). The task force was chaired by the State Water
Resources Department and made up of representatives from key state government departments
(environment protection authority, departments of agriculture, conservation and land
management, health and public works, state emergency services board and state "total
catchment management committee" (itself a co-ordinated multi-sector public group)), local water
boards and corporations, research scientists and the Murray Darling Basin Commission, a multi-
state water management agency. Later, when the state of Emergency passed, the BGATF
became the NSW State Algal Coordinating Committee (SACC). Members were added to the
committee from the adjoining states of Queensland, Victoria and South Australia (these states
had, in the interim, set up their own algal task forces). Their membership "helped speed
information transfer, and reflected the principle that management measures must be implemented
as part of an integrated resource management approach which cuts across established agency
and geographical boundaries" (NSWBGATF, 1993).
During the emergency phase of the bloom, the BGATF co-ordinated cyano-bacterial monitoring
throughout the river, the provision of alternative water supplies, the drilling of new boreholes (for
groundwater), the installation of water boom curtains to minimise cyanobacterial access to pump
inlet valves, the installation of emergency water treatment systems that could remove (or were
believed to remove) cyanobacterial toxins from contaminated drinking water supplies and, most
importantly, the transfer of information on a regular basis to the media. After the emergency
phase, the BGATF went on to develop the State's comprehensive integrated "Algal Management
Strategy" which aimed to minimise the future occurrence and impact of algal (mostly
cyanobacterial) blooms. The strategy included contingency planning, improving water system
management, reducing nutrient levels in waterways, education and awareness raising (including
media interactions) and research. Specific activities of SACC under the Algal Management
Strategy were: development of a nutrient control strategy (which included establishment of
nutrient management plans with actions such as plans for the upgrade of sewage treatment and
disposal systems); reduction in the phosphorus content of washing powders and detergents;
upgrades to septic tank systems in "high risk" catchments; reduction of soil erosion; fertiliser and
stormwater control strategies; and a public and local government "Phosphorus Awareness and
Reduction Campaign". The Committee also subsequently co-ordinated or implemented water
allocation and water system management plans, some of which were governed by multi-state
agreements (Box 7.3).
To improve the transfer of information to and from the State's many regions (some of which are
over 1,000 km from its headquarters in Sydney), regional algal co-ordinating committees
(RACCs) were also established. The RACCs were charged with the responsibility for developing
local contingency and management plans (based largely on guidelines put forward by SACC), co-
ordinating local media and public education, co-ordinating local monitoring and training in
cyanobacterial and algal sampling and identification, and identifying when cyanobacterial alerts
should be issued for regional waterways.
Table 7.1 outlines some of the functions necessary for cyanobacterial bloom
management and indicates the different government departments or other organisations
that may have responsibility for each function.
In addition to single-issue government or quasi-governmental agencies in a particular

country, there may be national or regional co-ordinating groups which have as their
mandate to bring together, or to attempt to bring together, the various sectors, user
groups and agencies involved in water management. There may also be water users
associations or co-operative societies with interests in water. Any one or several of these
agencies and co-ordinating groups may be involved when a cyanobacteria problem
arises.
The co-ordination necessary for effective action on environment and health issues is
difficult to achieve, not only because it takes place amongst those with vested interests,
agencies and groups representing (at times) divergent agendas, but also because the
co-ordination itself is complex. That is, it does not simply consist of linking different
institutions towards a common goal. It also requires combining inputs from different
areas of expertise (inter-disciplinarity); facilitating collaboration between policy-orientated
institutions and regulatory institutions; and fostering co-operation among sectors that
positively and negatively influence environmental health quality, including industry,
agriculture, local government and transport. Experience in the UK and Australia has
shown that an effective strategy to achieve this complex co-ordination is the formation of
interdepartmental task forces, with representatives from all relevant government
agencies and other key groups with a vested interest (Box 7.1). Such task forces have
been crucial to the success of emergency and long-term management of toxic
cyanobacterial blooms.
7.2 Policy tools

A number of tools may be deployed by governments in the pursuit of water management
and health protection policy. Although sometimes discussed in isolation, these tools are
generally inter-related, and policy implementation is most effective when a range of tools
is employed (see Box 7.2).
The principal interventions and tools relevant to the control of cyanobacterial hazards
include:
• Consensus through conflict resolution. Conflict resolution involves negotiation, through

formal and informal processes, guided or unguided by an expert, which may be binding
or non-binding, depending on the agreement of the parties before entering into the
negotiation. Each party's interests are aired, then jointly evaluated, ranked and then
balanced. Ideally, this process precludes, or at least places in abeyance, any
subsequent conflict, because it is through consensus that a plan of action (balancing all
relevant interests) is developed and agreed upon.
• Guidelines and recommendations. An enhanced role for experts informs the

development of guidelines and recommendations. Such guidelines and
recommendations suggest methods of complying with desired policies. They are
generally not binding, but rather serve an educational and advisory purpose, in that
those studying them learn what actions to take. To be effective (because adherence is
voluntary), each guideline or recommendation should be explained and justified.
• Non-binding agreements, including industry codes. Non-binding agreements, such as

industry codes, are similar to guidelines and recommendations, except that they may be
more formal. Although not binding, such codes have a high success rate, especially
where industry members are involved in their formulation. Such involvement is desirable
because industry members know their capacities and, furthermore, once they have
signed up to a code there is "peer pressure" by other members to adhere to it. In
addition, there might have been reluctance to follow codes developed by third parties,
such as environmental and advocacy groups. Such concerns are assuaged where the
codes have been developed following wide consultation. As with consensus building
through negotiation, non-binding agreements involve all parties in creating acceptable
compromises, thus ensuring greater success.
• Economic measures. A set of measures, both supportive and penalising, can be

implemented to encourage the reduction of water pollution. Generally, such economic
incentives and disincentives are contained in legislation, although they may also appear
in non-binding agreements such as industry codes. Supportive measures can be direct
subsidies for water protection technologies or practices (such as setting aside
agricultural areas), or various forms of tax reductions (such as for introduction of
treatment steps). Penalties involve fees for discharge, which can be calculated
according to the load of individual hazardous substances, with each substance tagged
with a sum per unit of loading.
• National legislation. Because of its formality and inflexibility, law may not necessarily
be the best avenue for implementing policy. Nevertheless, it should be recalled that in
situations where there are competing interests, legislation offers certain benefits. It
provides stability in a complex and changing environment and those with vested
interests know where to go to seek the rules that apply to them. In addition, at least at
the national level, legislation provides a coercive effect (the force of the State) upon
those that have neglected to take action despite having agreed to do so (this has been
known to occur particularly where there are financial interests at stake).
Table 7.1 Involvement of government agencies and other organisations in short-term

(including emergency) and long-term management of cyanobacteria-related health
hazards
Function Agency
Water pollution control, water resources Ministry of Environment, Ministry of Water Resources,
monitoring, enforcing water quality Environmental Protection Agency
standards
Protection of habitats and fisheries Ministry of Environment, Environmental Protection
Agency, Ministry of Wildlife, Ministry of Fisheries
Independent surveillance of drinking water Ministry of Health, Ministry of Public Health, local
supplies and recreational water quality, government
enforcement of corresponding standards
Drinking water supply Local government, private companies
Provision of recreational facilities, Ministry of Tourism, local and state government,
promotion of recreational water use private companies
Allocation of water rights Government agencies, parastatals, water users'
associations, courts adjudicating disputes
Water storage and river management, Ministry of Water Resources, Ministry of Environment,
monitoring, water allocation policy Ministry of Water, Ministry of Public Works,
Environmental Protection Agency
Catchment (watershed) management, Ministry of Natural Resources, Ministry of Agriculture
agricultural policy
Upgrades to public infrastructure, e.g. Ministry of Public Works, private and semiprivate
drinking water and sewage treatment sector utilities and service organisations
plants
Shipment of emergency water supplies Local government, private companies
Public education, research into causes Ministry of Education, Ministry of Health, local
and control of toxic blooms government, universities and research institutions,
recreational or other users' groups
Implementation of emergency response Local defence organisations, local government,
plans emergency preparedness agencies, Environmental
Protection Agency, health ministries/departments
Box 7.2 A combination of tools successfully reduced pollutant loads in Germany
To reduce hazardous substances and nutrients in wastewater, the federal government of the
former Federal Republic of Germany ("West Germany") introduced pollution fees per unit loading
in the 1970s. The combination of continuously rising fees with options of exemption from their
payment if new investments were made in wastewater treatment, resulted in great improvements
in treatment technology and water quality. At the time, many critics expected the fees to have little
effect because the fine per unit of loading appeared to be substantially lower than the costs for
improving treatment. However, the imposition of fees was only one of several tools used by the
government and society to achieve the goal of reducing loads. At the same time a strong "green
movement" produced public pressure against industrial pollution, industry developed new
production procedures which either avoided producing wastewater or included treatment, an
international treaty in 1987 targeted the reduction of pollution loads to the North Sea by 50 per
cent, and new legislation was discussed and implemented (such as European Union and national
legislation to introduce nutrient stripping in all treatment plants handling more than 10,000
population equivalents). In conjunction, these measures caused a very clear downward trend for
many pollutants, largely because industrial effluents have become substantially "cleaner", and
many sewage treatment facilities have introduced steps for nutrient removal. The major source of
nutrient loading into surface waters now is agriculture, and measures to reduce this source of
input require implementation.
All legislation must be underpinned with a firm policy background so as to ensure that it
will be effective once enacted. As one expert explains:
"Simply to promulgate laws and designate responsible officers will not suffice. It is first
necessary to build an effective political base of opinion and power and then to design
adequate systems, assign authority, provide resources and translate plans into action;
finally the development of the systems must be monitored, so that they can be adapted
as required. In most countries sectoral bureaucracies have to be brought under a
considerable degree of control and given clear, consistent and persistent policy direction,
if co-ordination schemes are not to flounder." (Schaefer, 1981, pp 42-43)
In most countries, before new legislation is introduced by a sector ministry, policy
approval must be sought and obtained from the Cabinet or Council of Ministers. This
ensures that the process of formulation of legislation is not an academic exercise,
undertaken with no thought for the realistic chances for enactment and implementation.
The existence of laws which are either unachievable or unenforceable means that
financial or human resources have been diverted from other, perhaps more important
(e.g. health), priorities. Moreover, the existence of such legislation reflects negatively on
the entire legal system.
• International and interstate plans and agreements. As noted in section 7.1.1, actions
taken to further one interest may have effects on other vested interests. Where a body of
water crosses international or state lines, such actions have international and, at times,
diplomatic implications. Few rivers and lakes remain within national or state boundaries,
adding a diplomatic or more complex legislative dimension to many water resources
management issues. There is a clear trend towards development of international plans
and agreements which organise water management around particular water bodies. Box
7.3 gives one example of an inter-state response to cyanobacterial bloom management
in Australia.
Box 7.3 The Murray-Darling Basin Algal Management Strategy
The Murray-Darling Basin (MDB) is the largest riverine catchment in Australia, crossing four
states, Queensland, New South Wales, Victoria and South Australia, and the Australian Capital
Territory. The MDB covers an area similar in size to Western Europe. The Murray-Darling River
system is actually a network of more than a dozen major rivers that ultimately join with the Murray
River and discharge to the sea at Lake Alexandrina in South Australia (the site of the world's first
scientifically documented toxic cyanobacterial bloom in 1878, see Chapter 3). The catchment is
the agricultural "bread basket" of Australia, with a heavy reliance on water abstraction from the
major rivers and headwater storages for irrigation.
Australia is a Commonwealth of states and legislative responsibility for the control and
management of water resources (indeed, all natural resources) is vested in the individual states.
In 1990, it was recognised that the four states covering the MDB needed to co-ordinate their
actions in controlling cyanobacterial blooms because actions, or lack of action, by one state could
have a deleterious effect on downstream states. Thus the MDB Commission (a non-government,
multi-state organisation that manages the water resources of the MDB) set up an Algal
Management Working Group. The working group was made up of members from the four states,
the commonwealth government and CSIRO (the national scientific research organisation). Its
charter was to "reduce the frequency and intensity of algal (actually cyanobacterial) blooms and
other water quality problems associated with nutrient pollution in the MDB through a framework of
coordinated planning and management actions". Later the cyanobacterial management strategy
was also to focus on the impact of the altered river flow regime on the occurrence of
cyanobacterial blooms, as well as on problems associated directly with nutrient inputs.
The first project commissioned by the working group was the production of a major report
identifying and quantifying (by modelling estimates only) all sources of nitrogen and phosphorus
inputs in the MDB. Soon after that report was released in 1992, the MDB Commission convened
14 Technical Advisory Groups (TAGs) to provide expert advice on issues ranging from nutrient
input control to cyanobacterial physiology and toxicology to water treatment. The 14 TAG reports
were released in one major report (MDBC, 1993). Following the release of the TAG Report, the
working group released a draft algal management strategy that was circulated widely among
government departments, community organisations and interested members of the public. Public
meetings and workshops were held to discuss key issues. The importance of gaining broad
community support and "ownership" of the draft report were seen to be crucial to the success of
the project. Comments obtained during this consultation phase were summarised and
incorporated in the final report (Murray Darling Basin Ministerial Council, 1994).
Unlike the state task forces that were set up during a "crisis" period (see Box 7.1), there was little
focus on short-term or emergency management. Rather, the MDB algal management strategy
focused on longer-term, strategic initiatives to reduce the frequency and severity of
cyanobacterial blooms. Its key elements were nutrient reduction, river flow management,
education and awareness, research and monitoring, and funding. The published strategy also
contains algal management case studies (on nutrient and flow management), techniques for
determining nutrient targets for individual catchments (including the use of catchment-nutrient
export models), and modelling studies on the impact of river regulation on natural flows in the
MDB.
In addition to plans and strategies, many water bodies are governed by bi- or multilateral
agreements which embody the customary rules and policy priorities of all the countries
with an interest in the river, lake or sea. Such agreements may address issues such as
reducing pollution, exchanging information on the level and volume of water, carrying out
flood management, early warning systems, monitoring changes in the path of
watercourses, sharing costs and managing dispute resolution mechanisms (Nanni,
1996). Few of these are of direct relevance to the control of cyanobacterial hazards
beyond policies relevant to the control of eutrophication in general. Nevertheless, some
international agreements specifically include provisions for notification of downstream
countries of pollution events in order to enable them to take corrective action (Box 7.4).
Including bloom formation as an issue requiring notification of downstream countries
might assist in control of cyanobacterial hazards.
Box 7.4 The UN ECE Convention on the Protection and Use of Transboundary Waters and
International Lakes
The Convention was drawn up under the auspices of the United Nations Economic Commission
for Europe (UN ECE) and adopted at Helsinki on 17 March 1992 (UN ECE, 1992). It entered into
force on 6 October 1996. The convention is intended to strengthen local, national and regional
measures to protect and use transboundary surface waters and groundwaters in an ecologically
sound way. The parties will prevent, control and reduce the pollution of transboundary waters by
hazardous substances, nutrients, bacteria and viruses. The precautionary principle and the
polluter-pays principle have been recognised as guiding principles in the implementation of such
measures, together with the requirement that water management should meet the needs of the
present generation without compromising the ability of future generations to meet their own
needs. This will protect and conserve not only water resources but also soil, flora, fauna, air,
climate, landscape and cultural heritage.
In order to comply with the Convention, emission limits for discharges from point sources shall be
based on the best available technology. The Parties will also issue authorisations for wastewater
discharges and monitor compliance therewith, adopt water quality objectives, apply at least
biological or equivalent processes to treat municipal wastewater, and develop and implement
best environmental practices to reduce the input of nutrients and hazardous substances from
agriculture and other diffuse sources. Parties bordering the same transboundary waters will
conclude specific bilateral or multilateral agreements which will provide for the establishment of
joint bodies (e.g. river or lake commissions). They will consult each other on any measures to be
carried out under the Convention, jointly elaborate water quality objectives, develop concerted
action programmes, jointly monitor and assess transboundary waters, set up joint warning and
alarm systems, and provide mutual assistance in critical situations.
Governments will also undertake any additional action that may be required to protect human
health and safety. One of these measures is the preparation, under the auspices of UN ECE and
the European Regional Office of WHO, of a protocol on water and health expected to be adopted
at the 1999 London Ministerial Conference on Environment and Health. Its objective is to promote
the protection of human health and well being and sustainable development through improving
water management and preventing, controlling and reducing water-related disease.
At the international level there is no "supranational" enforcement mechanism for nations

that decline to follow even the agreements they have signed. Nonetheless, international
agreements do offer certain benefits, namely that they are formal, written, and embody
two or more countries' policy compromises (Box 7.5). They are also usually vetted and
approved by a majority of the national legislature before being signed.
Box 7.5 The Baltic Sea
The Baltic Sea, the largest body of brackish water on Earth, has notorious phenomena of blooms
of cyanobacteria. The understanding of the biology of the species of cyanobacteria involved and
their production of cyanotoxins is necessary for an effective water management of the Baltic Sea.
The problems related to such mass development of cyanobacteria illustrate the need to combine
efforts at the local level with national and international policies in order to improve the
environmental situation in this geographical area.
The Baltic Sea has a surface area of 374,000 km2 and a mean depth of approximately 60 m and
is subdivided into a number of areas: the Gulf of Bothnia in the northernmost part; the Finnish
Gulf, bordered by Finland, Russia and Estonia; the Gulf of Riga; the Baltic proper; the Gotland
Sea and the Bornholm Sea. The Baltic Sea is connected by narrow channels to the Kattegat and
Skagerrak that lead to the North Sea. The salinity gradient varies from 1-4 ‰ in the most eastern
and northern parts, to 7-8 ‰ in the south.
The Helsinki Commission has estimated the pollution loads entering the Baltic Sea (Helcom,
1993, 1997). Nutrient input from the atmosphere (through rainfall and particulate deposition) and
from the surrounding land by rivers and run-off is high. Elevated levels of nutrients in the water,
especially of phosphate and nitrogen compounds, stimulate the growth of cyanobacteria and
algae. In the Baltic Sea several cyanobacteria, such as the genera Nodularia and
Aphanizomennon, are capable of biological nitrogen fixation.
The cyanobacteria blooms that are typical of the Baltic Sea may cause a number of serious
problems: they discolour the water and may produce bad smells; the decomposition of their
organic matter may result in depletion of oxygen, which leads to the death of fish and other
animals; and cyanotoxins are produced with consequences to other living organisms, including
people.
Co-operation between the countries in the Baltic Sea region occupies a central position in the
field of environmental protection. A priority action plan that also addresses problems of blooms
has been made and is being operated by the governments in order to determine the major
problems and the main measures to reduce them (Ministry of the Environment of Finland, 1991).
7.3 Legislation, regulations, and standards

Making generalised observations about legislative frameworks is difficult, not least
because of the diverse forms they may take (common law, civil law, Islamic law) but also
because of the varied environments in which they operate. Nonetheless, it is possible to
identify certain principal characteristics of the (national) legislative framework within
which actions related to the control of cyanobacterial health risks may be effected.
7.3.1 Forms of legislation
The term "legislation" or "laws" refers broadly to all legal texts which are promulgated by
the legislative arm of the state exercising its legislative powers under the Constitution, or
by the executive arm of the state exercising either its own executive powers under the
Constitution or powers delegated to it from the legislative branch. A basic law is usually
introduced by a sector ministry and enacted by Parliament or other equivalent national
lawmaking body.
The term "subordinate legislation" (most commonly known as regulations but in some
jurisdictions denominated orders, notices, rules, schedules, bye-laws, ordinances,
instruments, directives, or proclamations) refers to all legal instruments promulgated by
the executive branch exercising delegated legislative power. Subsidiary regulations to a
particular law enacted by the national legislature are elaborated and issued by the
Minister with responsibility for the subject matter (with or without the approval of all other
Ministers, depending upon the country). There may also be schedules or annexes to the
regulations, consisting of even more detailed subject matter.
Under the principle of supremacy, laws adopted by the national legislature nearly always
take precedence over subsidiary regulations as well as over legal instruments adopted
by the legislative branches of the country's political subdivisions. In some countries,
however, a variant of this provides that legislation promulgated by the national
legislature has primacy for certain subject matters, whereas legislation by the local
legislatures takes precedence for others. In other constitutional systems, such as the
USA or Australia, any powers not expressly granted to the federal government (or
Commonwealth) in the Constitution are the responsibility of the states.
In addition to laws and regulations, there may be other administrative legal texts such as
rulings, circulars, guidelines, standards, administrative notes and decisions. Such legal
instruments, although non-binding, assist the governments in achieving the aims of the
umbrella legislation while at the same time providing guidance for private groups as to
how government authorities will go about enforcing and implementing the law
(Chiaradia-Bousquet, 1995).
7.3.2 Water laws and standards
Enactment of a basic water law is important in countries where there are laws covering
different subject matters and enacted at different times containing conflicting provisions
relating to water. This might occur, for example, amongst a country's environment
protection law, electricity law and public health regulations. However, during the process
of updating or replacing legislation on water, it should be kept in mind that the trend in
many countries is towards considering water issues in conjunction with other issues,
such as the conservation and use of land, rather than viewing them in isolation (Burchi,
1991).
In order to be effective in addressing cyanobacterial hazards, the law should refer to the
jurisdictions, responsibilities, and authority of specified competent agencies in relation to
water and to their relationships to one another. Thus it becomes the instrument which
brings together the diverse parts of the water medium from which problems arise, as well
as the diverse groups associated with their control. These groups include water users,
those constructing and administering water supply systems, those regulating different
aspects of water and having responsibilities in related domains, and agencies involved in
data and information management and sector planning (Table 7.1).
It is important for the water law to reflect the policy priorities and political realities of the
particular country. For example, in the case of drinking water supply, legislation in the
established market economies is orientated principally towards the regulation of
established water supplies administered by recognisable entities where extensive water
supply infrastructure is in place. This may have little relevance to countries where the
goal is principally to support the best use of available resources in the provision of
incrementally improved service standards to the population as a whole. Such differences
argue strongly for developing the policy for the particular country in light of its particular
legislative and policy framework, rather than relying on some sort of "model" legislation
which would not reflect the realities at hand.
Water laws generally follow the basic principle of national legislation, i.e., that a basic
law is indeed basic, whereas any details, which may have to be changed in response to
scientific advancements or other exigencies, are contained in the regulations or other
subsidiary legislation. The rationale is that regulations are more easily changed because
they are elaborated by the responsible Ministry and can be easily revoked, amended or
replaced, rather than having to go through the Parliamentary process.
Among the details which will normally form part of the subsidiary legislation of a water
law are standards of various kinds. For example, there may be standards on minimum
treatment requirements linked to source type and quality. Chapter 5 describes the levels
of safety provided by treatment regimes in relation to the nature and degree of
cyanobacteria-related hazards in the source water. However, standard setting should not
be restricted to water quality, but should also extend to water resource protection and
water supply service quality more generally. It should also extend beyond the simple
assessment of quality and include the adequacy of structures and systems, such as the
definition of safe facilities and practices, minimum standard specifications, and minimum
standard operating systems.
In some countries, service quality standards are contained in a "sanitary code" or "code
of good practice". As mentioned above in the discussion of industry codes, such
standards are most likely to receive support in their implementation if consultation with
the affected entities (e.g. supply agencies and professional bodies) has occurred. Other
standards consist of quality targets, or what are known as water quality standards.
Comprehensive guidance on the setting of drinking water quality standards is available
in the Guidelines for Drinking-water Quality (WHO, 1993). The Guidelines for the Safe
Use of Wastewater and Excreta in Agriculture and Aquaculture (Mara and Cairncross,
1989) are also available and further guidelines for safe recreational-water environments
are in preparation.
Table 7.2 outlines some of the many issues that may be addressed through water laws,
regulations and standards. Not all of these should be seen as essential components of
water legislation for all countries, because the contents of a particular country's laws,
regulations and standards depend on the policy priorities of that country. This table does
not attempt to indicate which matters "belong" in the basic law and which should be
contained in subsidiary legislation. Although the general principle (that the more detailed
material and the matters that may need to be changed should not be contained in the
basic law) applies, the precise dividing line depends to a large extent on the particular
legislative scheme. Other water management issues, which are not directly tied to a
particular water use, but which might also be addressed through legislation, are given in
Table 7.3.
The above discussion should make clear that legislation may play a positive, supportive
role in target setting, particularly by virtue of the inclusion of details (such as standards)
in subsidiary regulations. Nevertheless, penalties and other sanctions to assist in
enforcement are also included in most laws and will occasionally be used. In order to
moderate their use, it is appropriate that explicit schemes are provided for enforcement
implementation and that they are pursued (Jensen, 1967). The imposition of other more
innovative solutions, such as mandatory participation in working groups or monitoring
programmes, may obviate the need for some of the harsher penalties. Alternatively, any
of these may be used in combination (Box 7.6).
7.4 Awareness raising, communication and public participation

The understanding that cyanobacterial blooms and toxins present hazards to human and
animal health is a prerequisite for anticipating, avoiding or reducing their adverse effects.
Raising awareness on the causes and effects of cyanobacterial blooms serves two goals:
protection from the health hazards presented by the toxins, and long-term reduction of
toxic bloom development through public participation. Early recognition of blooms and
scums facilitates better management of the blooms and associated problems and helps
to reduce their impact on the community as a whole (NRA, 1990; NSWBGATF, 1992).
Table 7.2 Types of water use and features of laws, regulations and standards
Water use Area of

Main features of laws, regulations and standards
type regulation
Drinking Drinking water Place obligation to supply continuously safe water upon the
water supply quality supply agency
Require supply agency to exercise due care
Define what is "safe" in terms of acute or long-term exposure (for
cyanobacteria this would normally relate to concentrations of
some specific toxins (see Chapter 4) as well as toxins not
recognised at the time of standard formulation)
Monitoring Divide monitoring obligations amongst a number of agencies or
requirements sectors (public sector oversight of water resource management,
public health authorities, drinking water suppliers, and
recreational facility providers including local government)
Require information sharing amongst these agencies
Define the extent of monitoring required of each agency and
define the conditions under which monitoring should be
undertaken
Define the technical requirements against which monitoring
results would be compared
Treatment Require supply agency to treat water to achieve the required
requirements standards (may be implicit in the requirement of the basic law to
supply safe water)
Require newly constructed or rehabilitated supply systems to
reach minimum treatment capabilities (these may be formulated
in terms of performance criteria or processes required, generally
in relation to the characteristics of both the source water and the
supplied population)
Require the upgrading of existing systems to reach the standards
required of newly constructed or rehabilitated systems
Define technical details of construction or operating practice
Contingency Oblige government agencies to establish contingency plans
plans Define the conditions under which contingency planning is
required, and the type of contingency to be planned for (for Alert
Levels Framework, see Chapter 6)
Describe the basic components of an "adequate" plan (this may
include, for example, the need to have alternative supply capacity
in very high risk areas)
Recreational Recreational Impose a "duty of care" (the standard of reasonableness) on
waters water quality those promoting or offering the use of facilities for recreational
water use (both private facility owners or operators and local
governments may be bound by these duties)
Relate use types and water types or qualities to alert levels, and
associate alert levels with actions
Define a methodology for defining alert levels with consideration
for local conditions
Table 7.3 Water resources management: features of laws, regulations and standards
Management Main features of laws, regulations and standards

types
Pollution control Require companies to control pollution (this could be interpreted as imposing
a duty to control eutrophication and, by implication, cyanobacteria)
Impose duty on industry to address nutrient discharge (especially sewage)
and run-off
Require agencies to manage river flows to prevent cyanobacterial blooms
Integrated Define alert levels and associate specific actions with each level (see
management Chapter 6)
Include methodology for defining alert levels according to local conditions
Information and the Oblige government agencies to disclose information relating to public health
public risks
Define the levels of information to be disseminated according to the alert
level
Impose a duty on public health authorities to inform the public and other
government agencies of relevant risks
Require public education and define its contents (this is different from
disclosure)
Intersectoral co- Define roles of agencies and place responsibilities upon them
ordination Require minimum frequency of co-ordination at national/local level and wider
consultation
Establish local procedures for co-ordination and consultation
Case histories (e.g. Bell and Codd, 1994; Ressom et al., 1994; Falconer, 1996; Codd et
al., 1997) and risk assessments (Chapter 4) point to the need for greater awareness of
cyanobacterial bloom and toxin hazards amongst not only water users (such as the
general public, water recreational groups and water resource managers), but also
professional groups. Awareness raising can serve several functions:
• To alert professionals and water users to the presence of hazardous accumulations of

cyanobacteria and toxins and to the risks involved in drinking and using water containing
blooms and toxins.
• To inform health-care professionals of the circumstances and exposure routes that

lead to intoxications.
• To inform water users of the causes of cyanobacterial bloom development and of ways
in which they can assist in reducing bloom formation.
• To inform environmental regulators, water user and water supply organisations about
cyanobacterial toxin hazards, and to help identify and implement appropriate policies on
water access, use, consumption and treatment.
Box 7.6 Imposition of a monitoring programme after prosecution of a local council for
serious environmental harm
In May 1996, the waste management officers responsible for the safe operation of sewage
treatment works in Tasmania, Australia, noted that a discharge was occurring from an almost
empty sewage treatment pond into a nearby dry watercourse. This was not allowed under the
terms of the operating licence. There was a dense bloom of Microcystis aeruginosa in the next
pond, and the officers observed pools containing Microcystis in the watercourse and scums of
blue-green slime along the banks and down into a recreational lake called Wrinklers Lagoon.
Water samples contained 511,000 cells per ml of Microcystis in the sewage pond and 144,000
cells per ml in a pool in the watercourse. Microcystin content in the pond was measured at 27 µg
ml-1, and the scum in the lake at 120 µg g-1 dry weight. Total nitrogen in the pond was 25.9 mg I-1,
and total phosphorus was 7.3 mg l-1. About 14 × 106 litres had been discharged from the sewage
pond into the lake, which had a volume of about 200 × 106 litres. The reason for the discharge
was a subsidence in the sewage pond retaining wall, which caused the operators to employ
repeated discharges from the pond to avoid collapse of the wall. Because the lake is a well-used
swimming and fishing location, the discharge caused an immediate health hazard and a long-
term risk of repeated blooms of toxic Microcystis. On receipt of the officers' report, the local
Council excavated the retaining sand bank of the lake and established a channel to the sea,
partly draining the lake and allowing sea water flushing.
Environment Tasmania, the state environmental protection agency, prosecuted the Council for
serious environmental harm on eight counts of discharge of sewage pond contents containing
toxic Microcystis into a recreational lake.
In discussions between the parties and an expert advisor, the Council proposed a monitoring
programme for the lake and a reduction of the severity of the charges. The magistrate hearing the
case found the Council guilty on a lesser charge of "material environmental harm", imposed a fine
of A$ 30,000 (approximately US$ 20,000) and directed that a monitoring programme be carried
out. The proposed monitoring programme was intended to make the lagoon safer with respect to
blooms of Microcystis (or Nodularia which can be handled in a similar way to blooms of
Microcystis); to improve the amenity value of the lagoon; and to minimise any risks to people,
pets, livestock, or fishermen eating fish from the lagoon. The monitoring programme was to be
reviewed after three years.
Source: FCR Cattell v. Break O'Day Council; Court of Petty Sessions, Launceston, Tasmania,
Australia. Complaint No. 9745521.
7.4.1 Professional group awareness
Professional groups are often no more cognisant of the circumstances leading to

cyanobacteria-associated health problems than are the general public (Skulberg et al.,
1984). The kinds of professional groups that require up-to-date information on
cyanobacterial blooms and toxins in order to manage effectively the associated
problems include water treatment and supply authorities and companies, medical and
veterinary practitioners, public and environmental health authorities, and national/state
environmental agencies. Initiatives to increase the awareness of these groups and
organisations have centred around conferences, workshops, correspondence in widely-
read professional journals, review-type papers, and the large-scale publication of
handbooks and reports. Such educational materials have been developed for the water
supply, treatment and environmental sectors (NRA, 1990; Lawton and Codd, 1991;
Carmichael, 1992; NSWBGATF, 1992; Yoo et al., 1995); the human health-care sector
(Codd and Roberts, 1991; Elder et al., 1993; Ressom et al., 1994), and the veterinary
sector (Codd, 1983; Beasley et al., 1989). In countries intending to address
cyanobacterial risks, these communication routes should be established and developed
further.
7.4.2 Public awareness
In some parts of the world, such as Australia, the UK and Scandinavia, cyanobacterial
awareness programmes have operated for several years. However, in other countries
with water resources and supplies containing toxic cyanobacterial blooms, awareness
and the availability of information are more limited. These deficiencies can be combated
in several ways. When cyanobacterial populations in water bodies exceed threshold
levels, media warnings should generally be issued to the public. Such warnings and
notifications may be issued through the newspapers, radio and television, and may
require media targeting and press conferences. The content of any announcement
varies according to individual bloom situations and the purposes for which the affected
water body is used. For example, if a cyanobacterial bloom occurs in a municipal water
supply source, media releases may be appropriate at lower cell populations than if the
waters are used only for recreational purposes.
The information supplied in the warnings should be prompt, concise and should include
details of expected changes in the quality of supply, such as tastes, odours,
discolouration and the actions being taken to alleviate the problems. Where recreational
activities, livestock watering, and further water uses other than human supply, are
involved, media warnings may also contain a brief description of the bloom and how it
can be recognised (such as a noticeable discolouration of the water or the presence of
scums). Other information which may be supplied in the warnings and notices, if
appropriate to the bloom situation, includes:
• Possible health problems including gastrointestinal upsets, skin rashes, and eye
irritations.
• Warnings not to use the water for swimming, bathing, or showering.
• Advice on finding alternate sources of water for animals and preventing pets from
swimming (and subsequently grooming scum material out of their fur) in affected areas.
• Notification that boiling the water will not destroy the toxins.
The community often has several concerns when a cyanobacterial bloom occurs in its
drinking water supply or recreational amenity. Commonly-asked questions concern the
use of water in evaporative air coolers; its use for drinking, laundry, dish washing,
washing fruits and vegetables and cooking; whether fish, shellfish and crustaceans
harvested from cyanobacterial bloom-containing water are suitable for human
consumption; and where to obtain alternative water supplies. Information or fact sheets
prepared in advance can be circulated upon request (see below).
It may be appropriate to display warning signs and notices at major points of public
access to affected water bodies. For eutrophic waters with occasional bloom problems,
provision of basic information to visitors is particularly likely to contribute to protecting
them from health hazards, because the rapid appearance and disappearance of scums
cannot always be detected in time for the responsible authorities to implement specific
measures (such as temporarily restricting use). In such situations, health protection must,
to some extent, rely on the users' own informed judgement.
Warning signs should be prominent and simple, with symbols or minimum text to
indicate no swimming, paddling, drinking, livestock watering, pets in the water, or other
activities deemed by the responsible agencies to present unacceptable exposure risks.
The signs may be removed when the blooms decrease to acceptable levels, or they may
be kept in place throughout the year in areas with year-round problems. If the latter
option is taken, the gradual decrease in visual impact of warning signs and notices
needs to be borne in mind.
Brochures and fact sheets are other tools which are increasingly used to convey to the
general public and specialist water user groups information on the appearance and
hazards presented by cyanobacterial blooms and scums. Examples are available from
several countries which have experienced bloom-related problems, such as the UK,
Australia, Denmark, Finland, South Africa, and the USA (NRA, 1990; NSWBGATF, 1992;
Yoo et al., 1995).
Brochures and fact sheets intended for the public should be simple and free of scientific
jargon, and should include brief descriptions with the following kinds of information:
• The nature and occurrence of cyanobacteria.

• The health hazards presented by cyanobacteria and their toxins.
• The recognition of cyanobacterial blooms and scums.
• The propensity of cyanobacteria to produce toxins.
• Precautions and steps to be taken to avoid potential health problems.
• Whom to contact if the presence of a cyanobacterial bloom or scum is suspected.
Special fact sheets for farmers, regarding protection of farm water supplies for animals,
and also for water sports organisations, have also been used. Professional associations
and user-group networks, such as national water sports bodies, provide a useful means
of spreading information. Posters and displays in public places and association premises
are additional tools for conveying information to the public on cyanobacteria and for
increasing community awareness. Videos are another useful method of educating the
public about cyanobacterial problems. They can be used to present the causes and
consequences of cyanobacterial blooms, as well as to provide good visual examples of
waters containing blooms and scums. They may also outline the strategies proposed
and used by authorities to manage the problems, and to inform the public about what, as
individuals, they can do to help. For example, to inform the public and to enable them to
initiate action at a community level, Australia's Murray-Darling Basin Commission (Box
7.3) initiated and supported the development of a "study circle" kit for adults entitled
"Blooming Blue-Green Algae". The kit contains a video, an audiotape, posters and
printed information on cyanobacteria which enables people to discuss and find out more
about cyanobacteria.
Education on the causes of blooms can begin through programmes run in schools.
Students may be encouraged to participate in simple water quality investigations, such
as in the Australian "Streamwatch" programme. Easy-to-use kits are employed to enable
the early detection of impending cyanobacterial blooms.
The effectiveness of the tools discussed above is only as good as the efforts made to
publicise and to distribute them. The need for contact points may be especially important
in remote locations where monitoring waters by government agencies or water body
owners is not possible and thus the assistance of the community is vital.
7.5 References
Anon 1992 Agenda 21: Programme of Action for Sustainable Development. United
Nations, New York.
Beasley, V.R., Cook, W.O., Dahlem, A.M., Hooser, S.B., Lovell, R.A. and Valentine,
W.M. 1989 Algae intoxication in livestock and water fowl. din. Toxicol., 5, 345-336.
Beasley, V.R., Cook, W.O., Dahlem, A.M., Hooser, S.B., Lovell, R.A. and Valentine,
W.M. 1994 Algae intoxication in livestock and water fowl. Microbiology, 5, 256-264.
Burchi, S. 1991 Current developments and trends in the law and administration of water
resources - A comparative state-of-the-art appraisal. J. Environ. Law, 3(1).
Carmichael, W.W. 1992 A Status Report on Planktonic Cyanobacteria (Blue-Green

Algae) and their Toxins. EPA/600/R-92/079, United States Environmental Protection
Agency, Cincinnati, Ohio.
Chiaradia-Bousquet, J.-P. 1995 Legislation. Legal Office Information Brief, Food and
Agricultural Organization of the United Nations, Rome.
Codd, G.A. 1983 Cyanobacterial poisoning hazard in British freshwaters. Vet. Rec., 113,
223-224.
Codd, G.A. and Roberts, C. [Eds] 1991 Public Health Aspects of Cyanobacteria (Blue-
Green Algae). Proceedings of a Seminar, London, 14 June 1991. Pub. Hlth Lab. Service
Microbiol. Digest, 8, 78-100.
Codd, G.A, Ward, C.J. and Bell, S.G. 1997 Cyanobacterial toxins: occurrence, modes of
action, health effects and exposure routes. In: J.P. Seiler and E. Vilanova [Eds] Applied
Toxicology: Approaches through Basic Science. Archives of Toxicology Supplement 19,
Springer-Verlag, Berlin, 399-410.
Elder, G.H., Hunter, P.R. and Codd, G.A. 1993 Hazardous freshwater Cyanobacteria
(blue-green algae). Lancet, 341, 1519-1520.
Falconer, I.R. 1996 Potential impact on human health of toxic Cyanobacteria. Phycologia,
35 (Supplement 6), 6-11.
Helcom 1993 Second Baltic Sea Pollution Load Compilation. Baltic Sea Environ. Proc.
No 45, ISSN 0357-2994, Hamburg, 161 pp.
Helcom 1997 Baltic Marine Environment Bibliography. Baltic Sea Environ. Proc. No. 66,
ISSN 0357-2994, Espoo, 172 pp.
Jensen, P. 1967 Examination of water supply and drinking-water. Dan. Med. Bull., 14(1),
273-280.
Lawton, L.A. and Codd, G.A. 1991 Cyanobacterial (blue-green algal) toxins and their
significance in UK and European waters. J. Inst. Wat. Env. Man., 5, 460-465.
Mara, D. and Cairncross, S. 1989 Guidelines for the Safe Use of Wastewater and
Excreta in Agriculture and Aquaculture. Published in collaboration with the United
Nations Environment Programme by World Health Organization, Geneva.
MDBC 1993 Algal Management Strategy Technical Advisory Group Report. Murray-
Darling Basin Commission, Canberra.
Ministry of the Environment of Finland 1991 Environmental Priority Action Programme

for Leningrad, Leningrad Region, Karelia and Estonia. Synthesis report, Plancenter Ltd.,
Helsinki, 193 pp.
Murray Darling Basin Ministerial Council 1994 The Algal Management Strategy. Murray
Darling Basin Ministerial Council, Canberra.
Nanni, M. 1996 The Aral Sea Basin: Legal and institutional issues. RECIEL, 5(2).
NRA 1990 Toxic Blue-Green Algae. Water Quality Series No. 2., The National Rivers
Authority, London, 125 pp.
NSWBGATF 1992 Blue-Green Algae. Final Report of the New South Wales Blue-green
Algal Task Force, New South Wales Department of Water Resources, Parramatta,
Australia, 159 pp.
NSWBGATF 1993 Blue-Green Algae. First Annual Report of the New South Wales Blue-
green Algal Task Force, New South Wales Department of Water Resources, Parramatta,
Australia.
Ressom., R., Soong, F.S., Fitzgerald, J., Turczynowicz, L., El Saadi, O., Roder, D.,
Maynard, T. and Falconer, I.R. 1994 Health Effects of Cyanobacteria (Blue-Green
Algae). National Health and Medical Research Council, Canberra, 108 pp.
Schaefer, M. 1981 Intersectoral Coordination and Health in Environmental Management.

An Examination of National Experience. Public Health Papers No. 74, World Health
Organization, Geneva.
Skulberg, O.M., Codd, G.A. and Carmichael, W.W. 1984 Toxic blue-green algal blooms
in Europe: a growing problem. Ambio, 13, 244-247.
UN ECE 1992 Convention on the Protection and Use of Transboundary Watercourses

and International Lakes, Helsinki, 17 March 1992. United Nations Economic Commission
for Europe, United Nations, New York.
WHO 1993 Guidelines for Drinking-water Quality, Second edition. Volume 1,

Yoo, R.S., Carmichael, W.W., Hoehn, R.C. and Hrudey, S.E. 1995 Cyanobacterial (Blue-
Green Algal) Toxins: A Resource Guide. American Waterworks Association Research
Foundation, Denver, Colorado, 222 pp.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 8. PREVENTATIVE MEASURES
This chapter was prepared by Ingrid Chorus and Luuc Mur
Cyanobacterial bloom formation can be avoided by measures which address their

growth requirements, i.e. plant nutrients and light. The basis for excessive growth of
cyanobacteria and other phytoplankton organisms (planktonic algae) is enrichment of
aquatic ecosystems with plant nutrients. This process is termed eutrophication. The key
nutrient in many cases is phosphate. In some systems, not all of the phosphate available
is actually used for phytoplankton growth because other resources limit the maximum
possible biomass. These may be light intensity or availability of nitrogen. Furthermore,
other biota can affect the growth of cyanobacteria and phytoplankton organisms:
submerged aquatic plants may compete for nutrients, and grazing by zooplankton may
reduce the stock of many phytoplankton organisms and (to a lesser extent) also of some
cyanobacteria.
The key management action for abatement of Cyanobacterial blooms is to address the
source of the problem by control and reduction of external nutrient loading to the water
body, and thus of the concentrations within it. Measures addressing light availability
directly (e.g. artificial mixing) or targeting the community structure of the biocoenosis (e.g.
biomanipulation) have been successful chiefly in less eutrophic situations. For highly
eutrophic water bodies under restoration by reduction of nutrient loading, such measures
may accelerate and enhance success.
Control and reduction of nutrient loading usually focuses on phosphorus (for the reasons
discussed in section 8.1), but measures addressing phosphorus may be designed to
reduce nitrogen input simultaneously. Targets for nutrient concentrations can be
achieved by following basic principles of good catchment management with respect to
agriculture and sewage treatment. This chapter aims to assist decision making by giving
information on:
• Target values for phosphorus concentrations likely to control cyanobacterial blooms in

a given water body.
• The potential impact of hydrophysical and biological methods for control of specific
Cyanobacterial ecostrategists (see section 2.3) in a given water body.
• The likelihood of success of poorly substantiated methods, sometimes propagated with
remarkably effective marketing strategies.
A decision tree for application of different management approaches in order to control

phytoplankton growth is given in Reynolds (1997). It leads from restriction of all sources
of phosphorus enrichment to enhancing flushing or sediment removal, biomanipulation
and artificial destratification. Reynolds (1997) points out that decision points for
application of measures are still being quantified by current research.
8.1 Carrying capacity

The concept of the carrying capacity of the resources in a given ecosystem to sustain a
population has proved very helpful in planning measures to control the size of that
population; Applied to cyanobacteria, this means asking questions such as:
• How much biomass can be sustained on the basis of the amount of nitrogen available?
• How much biomass can be sustained on the basis of the amount of phosphorus
available?
• How much biomass can be sustained with the amount of light that penetrates into the
water?
At any one point in time, it is likely that one of these three resources will limit the
possible amount of biomass at a lower biomass level than the others. However, the
limiting resource may change seasonally, for example at higher latitudes it changes in
relation to the angular height of the sun and day length, and in tropical climates it
frequently changes in relation to turbidity changes caused by seasonality of the flow
regime. During winter (even in clear water) or in turbid situations, light is usually the
limiting factor, whereas the available nitrogen and phosphorus could have allowed a
higher level of biomass. As light intensity increases in spring or as water becomes
clearer, phytoplankton organisms begin to multiply and incorporate available nitrogen
and phosphorus into their biomass, often up to the point where either of these resources
is depleted in the environment and further growth is not possible. If nutrient
concentrations are excessively high, phytoplankton may reach a density that causes
such a high level of turbidity that light availability limits any further growth, and in these
situations populations will be light- rather than nutrient-limited.
For planning and management, it is important to be able to estimate which of the key
resources (light, nitrogen or phosphorus) is likely to control phytoplankton biomass in
any given system. In other words, the questions are:
• Which resource determines the carrying capacity for phytoplankton?

• How high is the carrying capacity?
For a first step in answering this, it is not important to differentiate between

cyanobacteria and other phytoplankton, because the maximum amount of phytoplankton
possible can be equal to the maximum amount of cyanobacteria possible, once
cyanobacteria have become dominant. One approach at estimating the carrying capacity
in relation to nitrogen and phosphorus is to look at the relative amounts of these
nutrients in phytoplankton biomass, known as Redfield Ratio by mass (Round, 1965).
These ratios are:
C H O NP
42 8.5 57 7 1
Among these components, hydrogen and oxygen are never limiting in aquatic
environments. Carbon is available as carbon dioxide (CO2) and is consumed by
photosynthesis. Consumption of dissolved CO2 enhances diffusion of atmospheric CO2
into the water, a process which takes time. Carbon limitation has been extensively
investigated and has been summarised by Reynolds (1997). He concluded that CO2
limitation can occasionally have an impact, particularly in soft-water lakes with low
bicarbonate alkalinity, but that these situations are generally brief and do not
substantially limit the maximal amount of biomass possible.
8.1.1 Nitrogen
Nitrogen may enter water bodies as leachate from soils, as run-off from animal feedlots,
and from untreated or biologically-treated sewage, unless treatment includes nitrification
and denitrification. Phytoplankton can take up inorganic dissolved nitrogen in the form of
nitrate, nitrite and ammonia. In some arid continental regions, nitrogen is found to be the
chief factor limiting phytoplankton growth (Reynolds, 1997). The relevance of nitrogen to
limitation of cyanobacterial biomass is under debate, because a number of
cyanobacterial taxa can compensate for its lack by fixing atmospheric nitrogen at rates
of up to 175 kg ha-1a-1 (Rönicke, 1986). Thus, lack of dissolved inorganic nitrogen may
actually support the dominance of such species as Anabaena and Aphanizomenon.
However, these taxa also occur under conditions of surplus inorganic nitrogen. More
importantly, nitrogen fixation is a process requiring high amounts of light energy and will
not be effective in very turbid waters (as is the case during dense algal blooms). Thus, in
a given water body the maximum amount of biomass that can grow, in addition to the
biomass already present, can be estimated from the Redfield Ratio (see above) on the
basis of the concentrations of dissolved inorganic nitrogen.
8.1.2 Phosphorus
Phosphorus, like nitrogen, enters water bodies from untreated and from biologically
treated sewage, and further treatment steps are required to eliminate it. Phosphorus is
biologically available as phosphate, which binds to soil particles more effectively than
nitrate. Thus, the main entry route into water bodies from land areas is as surface run-off
and with erosion. Although biomass needs only about one seventh of the amount of
phosphorus as it needs of nitrogen, phosphorus is the resource which most frequently
limits phytoplankton growth in aquatic environments. Cyanobacteria and many other
phytoplankton organisms have developed storage mechanisms for phosphate (known as
luxury uptake). These enable them to store enough phosphate for 3-4 cell divisions. As a
consequence, one cell can multiply into 8-16 cells without requiring any further
phosphate uptake, and biomass can increase by a factor of 10 or more even when
dissolved phosphate is entirely depleted. For this reason, the amount of biomass that
can grow in addition to the biomass present cannot be predicted from the concentrations
of dissolved phosphate (see Box 8.1).
Phosphorus is naturally abundant only in very few aquatic ecosystems (such as some
lowland estuaries, some volcanic lakes and some ground-water-fed lakes). Furthermore,
its inputs to aquatic environments are often easier to control than nitrogen inputs.
Methods for elimination of phosphorus from domestic sewage are well developed and
currently more cost-effective than nitrification and denitrification (although current
developments may provide better approaches to combined elimination of both nutrients).
Measures to protect soils from erosion can also be very effective against loss of
phosphorus, whereas control of nitrate leachate from over-fertilised soils may be more
difficult. Nitrogen limitation may be to some extent compensated by fixation of
atmospheric nitrogen by cyanobacteria, whereas there is no comparable compensation
mechanism for phosphorus.
8.1.3 Available light energy
Light energy is a critical resource indirectly affected by nutrient concentrations. Light

arriving at a water surface is partly reflected, and the remainder is very quickly absorbed
by the water itself as well as by the dissolved substances and by the suspended
particles in the water. An upper limit of phytoplankton cell density is reached when the
cells shade each other to such an extent that further growth is no longer possible
because the individual cells do not receive enough light. This level can be estimated
following Lambert-Beer's basic law of exponential extinction with increasing thickness of
the water layer. This law can be expressed as:
where IZ is the intensity at depth Z

I'o is the surface intensity
ε is the vertical extinction coefficient, which in turn is the sum of extinction by the water
itself and the substances dissolved in it such as humic acids εw, the algae suspended in
the water εa, and other particles suspended in the water εp.
The average amount of light I* available to a phytoplankton organism entrained in
vertical mixing of the entire water body or (under conditions of thermal stratification)
within the upper, warm water layer (the epilimnion) is the square root of the intensity at
the surface Io and at the bottom of the mixed layer or of the water body Im (Reynolds,
1997). This relationship illustrates the decisive influence of depth on light availability and
enables estimation of the carrying capacity for phytoplankton biomass. For otherwise
clear water with the sum of εw and εp being only 0.2, Reynolds (1997) uses chlorophyll as
a measure of phytoplankton biomass and demonstrates that at 1 m depth and a daily
insulation of 10-3 mol photons m-1 s-1 a maximum of 670 µg l-1 of chlorophyll may be
sustained, whereas at a mixed depth of 10 m, only 49 µg l-1 are possible, and if mixing
occurs down to 30 m, carrying capacity declines to only 3 µg l-1 chlorophyll. At these
phytoplankton biomass levels, turbidity has also increased (expressed as increase of the
term εa), and not enough light can penetrate to enable further growth.
Nutrient availability often influences light limitation. If nutrients are limiting, phytoplankton
cannot grow to density levels that reach the light-determined carrying capacity. If nutrient
concentrations are excessive, phytoplankton will reach the biomass limit determined by
light (unless other factors such as hydrological flushing prevent growth). Further
increase of nutrient concentrations will then have no further effect on phytoplankton
biomass. This is often the case in hypertrophic water bodies. Turbid situations where the
light-determined carrying capacity has been reached are often dominated by
cyanobacteria, because at low light intensity these have a higher growth rate than many
other phytoplankton organisms (see section 2.2).
8.2 Target values for total phosphorus within water bodies

In determining target values of phosphorus within water bodies to control cyanobacterial
blooms, two questions are important:
• What phytoplankton biomass density can be expected at a given concentration of total

phosphorus?
• At what threshold concentration of total phosphorus does phytoplankton density create

a turbidity level high enough to reach the light-determined carrying capacity, and thus
switch a water body from total phosphorus control of biomass to control by light limitation?
Finding answers to these questions requires a clear definition of total phosphorus. In the
past, the soluble phosphate fractions have frequently been addressed when dealing with
eutrophication issues. This has some predictive value if it can be measured in seasons
where very little phytoplankton is present to consume dissolved phosphate (e.g. during
severe light limitation in winter), and if inputs are fairly constant throughout the year.
However, the carrying capacity for phytoplankton biomass is more reliably analysed in
terms of the total amount of phosphate, i.e. the sum of phosphate bound in biomass and
phosphate dissolved in the water, known as total phosphorus (Box 8.1).
Box 8.1 Monitoring total phosphorus as opposed to soluble phosphate fractions
Considerable confusion prevails in the use of the term "phosphate". Historically, soluble reactive
phosphate (SRP) or orthophosphate has been measured and addressed when dealing with
phytoplankton growth, because this is the fraction of total phosphate which is directly available for
uptake by cyanobacteria and algae. However, recycling of phosphate molecules within the
plankton communities has proved to be extremely rapid (within 5-100 minutes) (Wetzel, 1983),
and phosphate liberated by degradation of organic material will be taken up by bacteria and algae
faster than scientists can sample and measure it. Furthermore, cyanobacteria and algae can
store enough phosphate for up to four cell divisions and increase 16-fold, even if no soluble
reactive phosphate could be measured. If SRP is found above detection limits, this means that it
is surplus to the requirements of the cyanobacteria and algae. The only informational value of
such a finding is that growth is limited by some factor other than phosphate. The upper limit of the
biomass of cyanobacteria and/or algae that can develop in a given water body is, therefore, often
largely determined by the amount of phosphate bound within the cells, and total phosphate
phosphorus is the variable that should be studied for biomass management. This variable is not
equivalent to total phosphorus, which includes the mineral form (such as apatite) unavailable for
biological uptake. However, mineral forms are of quantitative importance only in some water
bodies (e.g. with high silt loading) and, for the sake of simplification, total phosphorus has
become widely used to represent total phosphate phosphorus.
The term total phosphorus is preferable to the term total phosphate, because results are reported
in terms of phosphorus rather than phosphate. This is important because the weight of the PO4
molecule is about three times that of its central P atom, and lack of specification in reporting
results as to whether they refer to µg PO4 or µg P has caused considerable confusion in the
literature.
For predicting phytoplankton density from total phosphorus concentrations several
models have been developed. The most comprehensive statistical model was
established through an international (largely European and North American) co-
operative study organised by the Organisation for Economic Co-operation and
Development (OECD) (Vollenweider and Kerekes, 1982). The concentration of
chlorophyll a was used as an easy-to-analyse measure for phytoplankton density. Data
for annual mean values of chlorophyll a and for maxima of chlorophyll a were compiled
from a wide variety of phosphorus-limited lakes (77 for annual means and 50 for maxima)
and related to annual mean concentrations of total phosphorus. The resulting
regressions were almost linear and highly significant (Figure 8.1).
The result roughly means that per microgram of total phosphorus, an annual mean
phytoplankton biomass corresponding to 0.25 µg of chlorophyll a, and a maximum of up
to 1 µg of chlorophyll a, may be expected. These results, together with results on the
occurrence of cyanotoxins given in Chapter 3, can be used as a basis for rough guide
values for estimation of maximum cyanobacterial bloom biomass and toxin
concentrations. In natural ecosystems, 1 µg of total phosphorus can support a biomass
up to 100 µg of organic substance (corresponding to approximately 1 µg of chlorophyll a),
which in turn may contain up to 1 µg microcystin. Substantially higher biomass and
higher microcystin content are possible, but occur only through accumulation of cells in
surface scums, or in some laboratory cultures.
Regressions, such as the OECD model illustrated in Figure 8.1, can be applied as
management tools to predict the average and the maximum phytoplankton biomass
range likely at a given concentration of total phosphorus. However, this approach has
been criticised because these models integrate the behaviour of a number of lakes
rather than the response of any one lake to changes in phosphorus concentrations. It
must be emphasised that regression B is useful to estimate the maximum phytoplankton
biomass at a given phosphorus concentration. However, the estimate given by the
double logarithmic regression is only rough, the scatter of points within the 95 per cent
confidence limits covers a factor of 10. Maximum ratios of chlorophyll to total
phosphorus of 2 are still within this limit. This scatter reflects the effects of other
environmental factors controlling phytoplankton biomass, particularly depth and mixing
conditions, and losses due to grazing of algae and some cyanobacteria by zooplankton.
The carrying capacity for phosphorus will not have been reached in all of the lakes
among the wide variety used for regression B. Further management actions in addition
to phosphorus control may be useful to avoid this carrying capacity being reached, i.e. to
move the vertical position of a water body downwards in Figure 8.1.
Predictability of maximum biomass levels is enhanced by combining this model of

phosphorus-determined carrying capacity with a measure for light-determined carrying
capacity by introducing the threshold concentrations where phosphorus limitation
switches to light limitation. This requires knowledge of the depth of the water body and, if
it is thermally stratified, knowledge of the depth of the warm upper mixed layer
(epilimnion), also termed "mixing depth" Zm. If data on light extinction are available as
outlined above, average light intensity through the mixing depth I* can be calculated. In
absence of such data, light availability can be estimated by regarding Zm in relation to
the depth of light penetration (euphotic depth, Zeu). If mixing is deep in relation to light
penetration, cell or colony densities cannot become very high, because the deeply
entrained cells would be spending too much time in the dark. If mixing is shallow, cells
are frequently moved near the surface, where enough light can penetrate, even through
a dense suspension, to enable extensive proliferation.
Figure 8.1 Vollenweider/OECD regressions for phytoplankton biomass (as chlorophyll a).
A. Annual means;
B. Maxima in relation to total phosphorus, amended with threshold levels for
different mixing depths (Zm) taken from Reynolds (1997) at which carrying
capacity is limited by light (εw+ εp m-1 = 0.2). Dotted lines represent 95% confidence
intervals of the regressions
The maximum biomass density which can be reached is higher in lakes with shallow
mixing depths than in deeply mixed lakes, because the latter reach their light-determined
carrying capacity at a much lower level of biomass. Figure 8.1B gives the chlorophyll
levels at which light limitation truncates phosphorus limitation for different mixed depths
in otherwise clear water (εw+ εp = 0.2 m-1), and further phosphorus input will not result in
a further biomass increase. In turbid waters (e.g. due to silt loads), this level will be
reached at much lower phosphorus concentrations (see Reynolds (1997) for chlorophyll
capacities at higher light extinction coefficients).
When designing programmes to reduce cyanobacterial proliferation in highly eutrophic

lakes, total phosphorus concentrations must be reduced below the threshold value for
phosphate limitation of biomass in a given lake in order to have an effect; this threshold
value will depend upon the depth of the lake. However, once phosphorus levels have
been brought below the threshold value the OECD model can be applied. For example,
if a shallow drinking water resource has total phosphorus levels of 600 µg l-1 and
cyanobacterial densities corresponding to 200 µg l-1 chlorophyll a, and if restoration
measures are applied that achieve total phosphorus levels of 200 µg l-1, the annual mean
chlorophyll a concentration is likely to decline to 70 µg l-1. Although this is a step in the
right direction, cyanobacterial biomass is still very high and problems of bloom formation
are not yet resolved.
Experience collected during the past two decades with restoration of water ecosystems
shows that phosphorus control for abatement of cyanobacterial blooms should target
concentrations at least as low as 30-50 µg l-1 total phosphorus (Cooke et al., 1993). In
many water bodies, substantial reduction of cyanobacterial and algal population density
can be expected at these concentrations when compared with higher concentrations.
However, significantly lower total phosphorus concentrations (less than 10 µg l-1) may be
required, particularly in deep lakes, in order to prevent blooms of some stratifying
ecostrategists in the long term. Information on the prevalent ecostrategists within the
cyanobacterial population will help to predict the success of management measures (see
section 2.3 and Box 8.2).
Box 8.2 Thresholds for phosphorus control of different cyanobacterial ecostrategists
Knowledge of the prevalent ecotypes in a given water body leads to the following consequences
for total phosphorus management:
• If scum-forming ecostrategists prevail (such as Microcystis spp. or Anabaena spp.) cell numbers
and biomass are likely to decline if total phosphorus concentrations can be brought well below 50
µg I-1 P. This will also reduce scum formation, because less cells and colonies will be available to
concentrate into scums. Nonetheless, some scums will probably continue to occur until
phosphorus limitation becomes so severe that cell density (and therefore turbidity) decreases to
the point where the depth of light penetration is as deep as the depth of large areas of the water
body (Zeu = Zm). Under these conditions, vertical migration of these taxa is less effective because
their buoyancy regulating mechanism requires some time in the dark (see section 2.2). Therefore,
they lose their competitive advantage over other phytoplankton.
• If dispersed ecostrategists prevail (such as the filamentous species Planktothrix agardhii,

formerly named Oscillatoria agardhii) very pronounced "switches" may be expected. As
phosphorus limitation reduces filament density, and thus turbidity, to the point where the
relationship of the depth of light penetration to the depth of the mixed layer (Zeu/Zm) is greater
than 0.4, these species are likely to disappear quite abruptly, and turbidity will increase even
further, thus stabilising the result.
• If metalimnetic ecostrategists prevail (such as Planktothrix rubescens), the water layer above
these cells is usually quite clear. Very low concentrations of total phosphorus (often below 10 µg
I-1 P) are necessary to decrease turbidity further and thus increase light intensity down to the
depth inhabited by these species. If this can be achieved, metalimnetic ecotypes may disappear.
If not, hydrophysical measures may be more successful in controlling their density.
• If nitrogen fixing ecotypes prevail (such as Anabaena spp.), reduction of total phosphorus down
to concentrations effectively limiting biomass will cause dissolved nitrogen concentrations in
excess of uptake by phytoplankton. Nitrogen fixation is then no longer an advantage in
competition over other cyanobacteria and algae. This may induce disappearance of the nitrogen
fixing species.
8.3 Target values for total phosphorus inputs to water bodies

The OECD study, which provided the regression shown in Figure 8.1, evaluated data
from 87 lakes with respect to the relationship between total phosphorus concentrations
in the lake (annual means) and the external load (input) (Vollenweider and Kerekes,
1982). The regression showed a close correlation between annual means of in-lake
concentrations and the annual means of inlet concentrations in relation to the residence
time of the water:
Table 8.1 Permissible and dangerous inputs for phosphorus and nitrogen for different
depths and for a renewal time of 2 m3 m-2 a-1
Mean depth (m) Permissible inputs1 Dangerous inputs
P (g m a ) N (g m a ) P (g m-2 a-1) N (g m-2 a-1)
-2 -1 -2 -1
<5 <0.07 <1.0 >0.13 >2.0

<10 <0.1 <1.5 >0.2 >3.0
<50 <0.25 <4.0 >0.5 >8.0
<100 <0.4 <6.0 >0.8 >12.0
<150 <0.5 <7.5 >1.0 >15.0
<200 <0.6 <9.0 >1.2 >18.0
1
Permissible inputs increases with residence time; a doubling of the residence time
increases the permissible inputs by a factor of 1.6
Source: Harper, 1992
The "safe loadings" given in Table 8.1 were derived from this model. The model may
serve for preliminary scaling of measures to reduce inputs of phosphorus. For prediction
of the total phosphorus concentrations in a given water body, estimates of inputs from
inlets, from surface run-off (especially from agricultural areas with tillage, fertilisation and
erosion), from urban storm-water outfalls and from atmospheric precipitation are needed.
Acquiring all of the necessary data may be difficult, and approaches to estimates are
discussed in the context of lake and reservoir diagnosis by Cooke et al. (1993).
Currently, very little information on relationships between inputs and in-lake

concentrations is available from tropical and subtropical aquatic ecosystems. Future
research must investigate whether relationships established for water bodies in
temperate climates apply, or whether changes are necessary. Differences may be
expected, especially with respect to sediment-water interactions and mineralisation rates,
because these depend strongly upon temperature and upon thermal stratification.
If inputs exceed critical values for a given system, increase of the concentrations of total
phosphorus within that system are likely. In turn, cyanobacterial proliferation is likely,
and management measures are then needed to reduce phosphorous inputs.
8.4 Sources and reduction of external nutrient inputs

In most cases, eutrophication is enhanced by anthropogenic activities. The three major
sources of external nutrient inputs are run-off and erosion from fertilised agricultural
areas, erosion resulting from deforestation, and sewage. Exceptions may occur and are
illustrated by the example of Lake Victoria. This large lake has an area of 68, 000 km2.
Tributaries supply about 15 per cent of its water and 85 per cent originates from
precipitation. Burning of field stubble is widely practised and leads to substantial air
pollution. Thus, 60 per cent of the phosphorus load is estimated to enter Lake Victoria
through precipitation. Only 40 per cent originates from sewage and run-off into the
tributaries (Lindenschmidt et al., 1998). Replacement of the practice of burning stubble
would substantially reduce this load.
Sustainable approaches aim at reducing nutrient loads at their source or as close to the
source as possible (sections 8.4.1 and 8.4.2). If this is not feasible, approaches to
reducing inputs from a main tributary (i.e. treating a main tributary as if it were a sewage
channel) may be considered (sections 8.4.3). Reduction of external inputs beneath the
threshold expected to be effective is an important basis for the success of further in-lake
restoration measures (which may address internal nutrient loads or ecosystem structure,
see section 8.5).
The first questions to ask in designing programmes for abatement of eutrophication by

improving agricultural practices and/or by introducing or improving sewage treatment are:
• How high is the phosphorus input from wastewater (sewerage outfalls or diverse small
sewage inlets) to a given water body (excretion of 2-4 g P per person per day may be
assumed) (Siegrist and Boiler, 1996)?
• How high is the input from agriculture and run-off from other surfaces?
• Down to what concentrations must phosphorus input be reduced in order to reduce

concentrations in the receiving water body beneath a total P threshold likely to be
effective in the given water body (see section 8.2)?
Answering these questions requires specific evaluation of the resources to be protected,

of their catchment land use, and of the water and effluent drainage network. Ideally,
nutrient inputs and the relative share of different nutrient sources should be estimated
and catchment characteristics, such as soil type, run-off potential and vegetation cover
should be considered. In a region with nutrient-rich, erodible soils and reduced
vegetation cover or natural eutrophication (e.g. river deltas or some tropical areas),
reductions in inputs will not be possible to the same extent as in a region with sandy
soils, flat relief and dense tree cover (Cooke et al., 1993). Thus, the same measures and
similar investments into reducing inputs are likely to be more successful in reducing
eutrophication in a potentially oligotrophic ecosystem than in a naturally eutrophic one.
In many cases, quantitative assessments of inputs will not readily be available because
this requires detailed analysis of hydrological conditions (e.g. assessment of stream flow
rates and water retention times), as well as nutrient concentrations and their variations
over time in all of the main tributaries. Such investigations require time and resources.
Whereas inputs from point sources like sewage effluents are relatively easy to assess,
diffuse inputs from agriculture are often very difficult to quantify. Managers are likely to
be confronted with water bodies for which almost no limnological data are available, and
perhaps not even the depth contours of the water body are known. Planning and
implementation of resource protection measures may be delayed for several years
before reliable data become available. The dilemma for managers is whether to begin
with measures to reduce obviously substantial inputs, without having the data basis to
predict whether the measures taken will reduce nutrient concentrations below the
threshold effective for controlling cyanobacteria, or whether to delay planning and
decision-making until data become available.
In some countries, user friendly "decision support" software programs (see Box 8.3)
have been compiled for use by managers and community groups in identifying the main
sources of nutrients from a catchment, and for identifying possible actions e.g. off-river
disposal of wastewaters or (re) construction of riparian buffer strips to protect against
inputs from erosion. There is an array of computer models that may be used to simulate
the hydrodynamic and transport conditions in a catchment system. Trudgill (1995) gives
examples of available models and profound discussions of various processes that may
be included in the compilation. Since numerous parameters are required for calibration
of these models, the complexity of the model should be tailored to the extent of the data
base available. For applying these models, the study should include the entire
watershed (and, in some cases, the airshed as well) and not just the lake water body.
No general recommendation can be given to resolve the dilemma between the necessity
of adequate planning data, and the need to implement obvious measures without delay.
Although numerous restoration measures around the world have proved to be ineffective
due to insufficient diagnosis and evaluation by scientists and managers, most of these
were measures addressing ecological balances within the water body (see section 8.5).
In contrast, measures addressing reduction of external nutrient inputs are not likely to be
applied mistakenly. In the worst case they may prove to be insufficient and require
further action - either further reduction of inputs, or in-lake action.
An effective alternative to quantitative assessment of loading, is a "common-sense" or

qualitative approach. This begins with studying maps and geographical information to
identify main tributaries, slopes critical for erosion, precipitation patterns and land use.
Detailed and critical inspection of the catchment area may provide an excellent basis for
recognising priority actions, some of which may be implemented at low cost. Such
inspection is generally of underestimated value. Qualitative assessments should include
identification of sewage outfalls (possibly illegal or unregistered), land use, vegetation
cover, agricultural practices (e.g. soil tilling supportive of erosion, lack of protective
riparian buffer strips with dense vegetation cover as a barrier between surface runoff and
water body, and stubble burning).
Box 8.3 Testing the Catchment Management Support System (CMSS) in the Murrumbidgee
River Catchment, New South Wales, Australia
The Catchment Management Support System (CMSS) is a simple computerised decision support
system developed by CSIRO Land and Water, Australia. It combines land use, nutrient
generation and land management information into a single model which can predict the impacts
and costs of different land management practices in a catchment. As the set-up and use of CMSS
does not require a technical or computing background, it is highly suited to use by catchment
management committees and other community or management groups (Davis and Farley, 1997).
In a recent application (Cuddy et al., 1997), CMSS was developed for the Murrumbidgee River
Catchment, New South Wales (NSW) with the intention that it should be provided to local
catchment committees as a tool in the development of nutrient management plans. This pilot
study was used to demonstrate the application of CMSS to a specific area, as an initial phase of
its application to major catchments in NSW. In applying CMSS to the Murrumbidgee Catchment,
significant inputs were required from relevant management organisations and individuals. The
necessary data, ranging from soil types and rainfall distribution to the initial and ongoing costs of
changed sewage treatment practices, were available from these organisations. The
Murrumbidgee CMSS was developed within the timeframe of the pilot study and handed over to
the local agencies. Although the pilot study successfully met its objectives and the program has
now been widely adopted, CMSS has yet to be influential in the production of nutrient
management plans. This reflects both the prolonged nature of developing plans through
community consultation and the lack of involvement by local catchment committee members in
the pilot study.
While investments in internal or "in-lake" control measures can be wasted without

sufficient pre-restoration evaluation, reduction of external nutrient inputs will be at least a
first step in the right direction. Managers are encouraged to implement nutrient control
measures, even if the data base is not sufficient to predict the quantitative impact on
concentrations within the water body.
8.4.1 Domestic wastewater
Wastewater emissions have frequently been managed by a philosophy of using water as

a medium for transporting wastes out of the locality without considering the impairment
of the function of the water as a resource further downstream, or of the possible
enrichment of lakes with pollutants and nutrients.
Many developing and least developed countries are extending their coverage of drinking
water supplies in order to improve human health. Evaluation of experience shows that
this usually leads to a several-fold per capita increase in water consumption and thus
also in the amount of waste-water generated. Improvement of human health therefore
also requires development of wastewater collection and treatment, as highlighted by the
World Bank:
"For urban water supply, experience indicates that the collection and proper treatment of
sewage must be an integral part of water supply projects. Bringing water into a city
without taking sewage out exposes the population - and particularly the poor -to
increased pollution." (World Bank, 1993)
Possible exposure to pathogens is a major reason why collection and treatment of
wastewater is important. Another reason is that cyanobacteria proliferate in eutrophic
lakes and rivers fertilised by wastewaters.
Replacement of phosphorus in laundry detergents can typically reduce phosphate loads

in sewage by 50 per cent at best, the remaining 50 per cent being inevitable because it
originates from human excreta. Wherever sewage outfalls are considered to be a
significant input of phosphorus to a water resource, phosphate elimination, alternative
treatment approaches (see Box 8.4), or sewage diversion is necessary. Criteria for
adequate technology largely depend upon population structure and on geographic
conditions.
If population density is low, and the flushing rates of water bodies or phosphorus
absorption capacity of the soils is high, nutrient elimination from sewage may not be
necessary - high nutrient concentrations in water resources in such situations are more
likely to originate from agriculture. Treatment methods adequate for protection from
infectious agents may also be sufficient in these circumstances (on-site treatment such
as properly constructed latrines, septic tanks and sewage lagoons). If domestic waste-
water is used in agriculture, health risks should be avoided by following the WHO
guidelines for the use of wastewater in agriculture and aquaculture (Mara and
Cairncross, 1989). A special aspect of such situations is tourism. Low population areas
affected by tourism may need special consideration because the population may
increase temporarily several-fold and overload sewage treatment capacities. In
temperate regions, the tourism season may coincide with the cyanobacterial growth
season.
Box 8.4 Nutrient retention using low- and medium-technology approaches
Alternatives to "high-tech" nutrient stripping methods in sewage treatment (involving an array of

methods using lagoons or land treatment) have been in use for more than a century, and their
advantages as well as their drawbacks are well established. Many land treatment systems require
large areas (up to 10 m2 per population equivalent), and infiltration of wastewater into the ground
below has often occurred without control for hazardous substances or pathogens. Modern
artificial wetland systems are being developed to overcome these shortcomings. These systems
combine mineralisation processes in the water body with filtration through the soil substrate; they
are sealed towards the bottom and they have controlled effluents. As with "high-tech" treatment
plants, artificial wetland systems require careful maintenance operation, and control. They are by
no means a solution for "letting nature do the job alone". Poor maintenance and overloading
rapidly lead to malfunctioning and to poor hygienic conditions in the system. In temperate regions,
performance during the cold season may be less effective, but as a means for handling additional
sewage loads caused by tourism during the warm season, such systems may be excellent
supplements to year-round treatment technologies. Other approaches comprise hygienic
methods of collection of excreta, re-use of waste in agriculture and regular emptying of septic
tanks. Care must be taken particularly with respect to occupational hazards, and with the
designing and siting of alternative treatment methods in order to avoid relocation of the
wastewater problem to another site (e.g. causing groundwater contamination).
Sparsely populated regions with water bodies highly susceptible to eutrophication, e.g.
because of long water retention times (such as many lakes in Sweden), will require more
carefully designed sanitation systems to protect these resources. Supplementary
treatment techniques may be necessary, particularly during the tourist season (see Box
8.5).
In many densely populated areas, municipal sewage is the source of at least half of the
total phosphorus inputs to rivers and lakes. In large urban areas, treatment of
wastewater collected in sewerage systems requires industrial-scale plants for the
protection of surface waters. Untreated municipal sewage contains more than 10 mg 1-1
of phosphorus. Biological (secondary) treatment oxidises organic matter, but does not
substantially reduce phosphorus content. Where municipal sewage constitutes a
significant source of phosphorus pollution, removal of phosphorus at treatment plants is
necessary (see Box 8.5). Heavy seasonal tourism may also be a problem in such
situations because it increases the demand on the capacity of treatment plants and
sewerage and also causes substantial fluctuations in the sewage load. Sewerage and
treatment for fluctuating amounts of sewage present specific technical difficulties. Lake
Balaton is an example of such a situation, where the number of tourists during July and
August is twice that of the local population (Somlyody and van Straten, 1986).
Box 8.5 Two well-established and widely used technologies for phosphorus removal in
treatment plants
Chemical precipitation with ferric or aluminium salts, often performed simultaneously with
biological treatment, can reduce phosphorus concentrations by an order of magnitude to values
around 1 mg I-1.
Biological phosphorus removal ("bio-P") can be achieved by alternating aerobic and anaerobic
steps in biological treatment and thus substantially enhancing P-uptake by bacteria. This method
saves flocculation chemicals and produces less sludge, but requires an adequate design of
basins and careful operation of the process. In large treatment plants (> 100, 000 population
equivalents) it is more economic than chemical flocculation, because operation costs are lower
(Gleisberg et al., 1995). It can reduce treatment plant effluent concentrations down to 0.2-0.5 mg
l-1 P. Although the method has been known for several decades, experience has only recently
accumulated to a level of understanding which allows stable and reliable performance, and it is
advised to maintain chemical precipitation facilities as a back-up (Harremoes, 1997), especially
for treatment plants discharging into water bodies with critical phosphorus concentrations.
In lowland regions, discharge of domestic and industrial wastewater may amount to 50

per cent and in dry seasons to almost 100 per cent of the total flow of the recipient river,
e.g. River Thames in London (Gray, 1994) and Havel River in Berlin (Köhler and Klein,
1997). Such slow flowing rivers are suitable habitats for cyanobacterial growth but the
river water may also be needed for production of drinking water. In such situations,
phosphorus elimination by simultaneous chemical precipitation or "bio-P" is not
adequate, because concentrations in the recipient water body will almost equal the
outfall concentrations. Treatment objectives therefore are 0.03-0.05 mg l-1 P and this can
be attained by adding a further treatment step (filtration) for removal of phosphorus in
small, slowly settling floes. The removal of pathogens is usually also a further important
objective of filtration treatment in regions with such intensive water use. Various
methods of filtration over sand or gravel and pumice beds are available, and new
methods of membrane filtration are also being developed.
The costs of such treatment technology may appear intimidating at first glance because
of the necessary investments. However, even sophisticated procedures involving
filtration need not cost more than US$ 0.15-0.30 per m3 of treated water (Heinzmann and
Chorus, 1994), and this is usually only a small fraction of the costs of drinking water.
Effective resource protection in such densely populated areas will save the cost of
drinking water treatment, for example by saving the necessity of activated carbon
filtration.
Eutrophication due to sewage outfalls has been recognised as a widespread problem

since the 1970s. Abatement was begun in that decade with several isolated projects,
such as diversion of sewage around lakes with specific touristic value (e.g. at Lake
Constance in the 1970s and at major parts of Lake Balaton in the 1980s) and the
introduction of phosphorus precipitation in a few treatment plants. Comprehensive
programmes began in the mid 1980s. In Europe, an international convention for the
protection of the North Sea triggered introduction of phosphorus and nitrogen elimination
in larger sewage treatment plants (those treating more than 10,000 population
equivalents). Recently, the implementation of measures to eliminate nutrients in sewage
has started to show substantial success:
• Denmark achieved a 79 per cent reduction of phosphorus inputs from sewage between
1985 and 1995 and further reduction is expected in 1997 when two plants in
Copenhagen go into full operation (Harremoes, 1997).
• Switzerland achieved a 60 per cent decline in total phosphorus inputs from municipal
wastewater (Siegrist and Boiler, 1996).
• The USA and Canada together achieved a 50 per cent reduction in phosphorus
concentrations in Lake Ontario and the west basin of Lake Erie, with significant impact
upon "algal blooms" and accumulations of filamentous cyanobacteria on shorelines, by
the construction and upgrading of sewage treatment plants along the Great Lakes
(Charlton, 1997). However, in order to maintain the quality now achieved despite the
expected increase of population density by the year 2011, and in order to meet the
quality targets for Hamilton Harbour, tertiary sewage treatment (effluent filtration) is
considered necessary.
Charlton (1997) emphasises the importance of reliable performance of sewage

treatment plants: "Part of the difficulty in understanding sewage problems is the
mistaken belief that sewage treatment plants, once built, will perform as planned, with no
operational problems" (Charlton, 1997). Steady degradation of performance during
critical summer months has led to considerable phosphorus inputs, the highest being 2.7
times the target. Internal assessment, optimisation of performance, and identification of
staff with the job and the achievement of effluent target concentrations, are crucial for
reducing effluent loads.
8.4.2 Agriculture and erosion
Losses of phosphorus and nitrogen from deforested, agriculturally-used areas into

surface waters are the other major factor enhancing eutrophication. As with phosphorus
pollution from wastewater, this problem has increased exponentially in some parts of the
world since the 1950s. The causes are structural changes in agriculture involving
intensification by tillage of larger plots, extensive application of mineral fertilisers, and
the establishment of large-scale animal husbandry. Simultaneously, other major parts of
the world suffer substantial lack of phosphorus in topsoils. Zehnder (1996) points out a
striking global imbalance between surplus phosphorus in most of the industrialised
regions and a lack of phosphorus in most developing regions. On a global scale,
phosphorus must be perceived as a limited resource. In regions with phosphorus
deficiency, controlled fertilisation in combination with adequate protection from erosion,
would help maintain fertility of tropical and subtropical soils, so that further deforestation
would no longer be necessary in order to gain new (only transiently fertile) farmland.
Further, development of sewage treatment methods that allow reclamation of
phosphorus for reuse as fertiliser should be a long-term target for sustainable handling
of this resource.
Excessive use of fertilisers and manure has created eutrophication problems in lakes
and rivers. In regions with intensive agriculture and slowly flowing rivers with little
discharge, e.g. in much of north-western Europe, water bodies without cyanobacterial
problems have become scarce. Heavy surplus fertilisation has been enhanced by the
widespread trust in phosphorus retention in soils (soils as "savings banks" for
phosphorus), but soil erosion and surface runoff have proved to be major pathways into
surface waters, particularly through storms and intensive rainfall shortly after application
of fertiliser or manure. The extent of these losses is site-specific and largely depends
upon geographic and hydrological conditions: "It has been suggested that up to 90 per
cent of the annual phosphorus losses occur from only 5 per cent of the land during only
one or two storms, especially in areas where surface runoff and erosion are the
dominant routes for phosphorus losses" (Oenema and Roest, 1997). The importance of
leaching, first recognised as a pathway for nitrate inputs, is being recognised
increasingly for phosphate in some types of soils, e.g. sandy, acidic soils with a high
degree of saturation of their phosphorus adsorption capacity (Oenema and Roest, 1997).
For the Netherlands, Oenema and Roest (1977) estimate 300,000-400,000 ha of
phosphorus-leaking sandy soils to be pollution "hot spots" requiring high priority in
identification and remediation.
In tropical and subtropical regions, eutrophication of lakes and reservoirs due to inputs
associated with erosion is greatly enhanced.
Closing cycles by reuse of manure as a nutrient resource in agriculture, especially if

combined with changing practices of land tillage and deforestation, can contribute to
reduction of nutrient pollution. Such approaches are sustainable alternatives to
considering animal slurry as waste, and reduce the costs of purchase of fertilisers. In this
context, performance of large-scale livestock farming in industrial dimensions requires
regulation just as for industry:
"The output of waste from many cattle and pig units measures up to that of a large town
and needs to be managed with at least the same care as that accorded to human waste.
At one time the only environmental requirement placed on farmers was that they should
observe good agricultural practice. This is no longer a satisfactory basis for
environmental protection. Agriculture must be put on the same basis as other major
industries with strict controls on the quality and quantity of effluent discharges."
(Packham, 1994)
Box 8.6 Good agricultural practices - best management practices
For fertilisation
• Planning land use, choice of crops and crop rotation to minimise erosion losses.
• Structuring the farmland to minimise erosion by measures such as terracing, interruption of

large areas with shrub hedges, and buffer strips planted with shrubs along river banks and lake
shores.
• Planning nutrient management in order to avoid losses from the farm by closing nutrient cycles,
using manure as fertiliser and avoiding phosphorus import.
• Measuring current fertiliser content of soils and dosing according to the demand calculated for
the crop.
• Timing application of fertilisers according to the growth of the crop.

• Cover crops to reduce erosion from bare soil.
• Managing irrigation and groundwater levels.
• Using animal slurry as fertiliser according to the demand of the crop, rather than misusing crops
or grasslands as a deposit site for animal slurry.
For animal husbandry

• Reduction of livestock density to 1.5-2 cattle units per hectare.
• Closing nutrient cycles by limiting stock numbers to the fertilisation requirements of the area
used for growing crops.
• Placing feedlots and watering sites away from surface waters.
• Protecting river banks and lake shores with fences to keep out livestock in order to reduce both
direct pollution by excreta and increased erosion by treading.
The alternative to Packham's request is to redefine "good agricultural practice" to include

sustainable resource use. Criteria for "best management practices" or "good agricultural
practices" are listed in Box 8.6.
Apart from closing nutrient cycles, sustainable biological methods of production have
been developed in many pilot projects and are becoming increasingly popular. Model
projects have demonstrated that productivity of biological or "organic" farming methods
is not substantially lower than that of conventional farming, provided methods are
adapted adequately to the given geographic conditions. The economic balance of
"biological" or "organic" farms is frequently equally good because of reduced
expenditure for agrochemicals and, in some cases, better prices for the product. Co-
operation between water supply agencies and farmers has supported this development
in Germany (see Box 8.7) and has shown success in improving resource quality.
In drinking water catchments, it is particularly important that agriculture follows "best

agricultural practices". This can be encouraged by designating protection zones around
the drinking water source and regulating practices allowed or prohibited within these
protection zones. Sophisticated models distinguish two to three degrees of protection,
depending upon the relative impact of the respective part of the catchment upon the
water quality. In addition to agriculture, other activities which impact water quality, such
as forestry, fisheries and tourism, may be regulated in drinking water protection zones.
Countries with a traditional rural society may have options for reclaiming historical
agricultural experience and combining it with modern approaches to sustainable "best
agricultural practice" in order to attain high outputs of high quality products at low
environmental and health impact. Such an integrated approach requires continuous
development, evaluation of experience and training.
Countries with large-scale industrialised agriculture may have problems in implementing

change. In contrast to the success achieved in phosphorus elimination by wastewater
treatment, progress in abatement of agricultural phosphorus pollution has at best been
modest. For Switzerland, Wehrli et al. (1996) estimate that, while phosphorus emissions
from sewage have been reduced by 60 per cent during the past 15 years, losses from
agricultural areas into water bodies have rather increased and the need for a new
agricultural policy is just beginning to be widely perceived. Some of the reasons for this
delay in awareness and action are:
• Losses of phosphorus from farmland are rarely perceived as economic losses.
• Effective measures will differ regionally or even from farm to farm, and management
practices must be optimised locally rather than administered generally. This requires
shifts of attitudes.
• Even where programmes and regulations exist, their implementation and control may
be difficult.
A basic change in attitude is required. In some countries this appears to be developing

slowly as a new generation of farmers with better training (including education on
sustainable farm management and ecological impacts) takes over, and as consumer
awareness for quality criteria and the ecological impact of products is growing. A wide
array of measures can be used by government authorities to support such developments.
Examples are training and advice to farmers, eco-audits on products, subsidies for
setting land aside from use, subsidies or tax redemptions during periods of transition to
organic farming methods, pollution taxes and legislation to enforce water protection.
Box 8.7 Co-operation between water suppliers and agriculture for sustainable provision of
healthy drinking water
Legislation should include the principle that use of land and water must occur in such a
sustainable way that subsequent use by others Is not hampered. This provides a legal basis for
requiring co-operation of agriculture with water supply agencies. Furthermore, protection zones
above aquifers or around reservoirs must be staked out so that they cover the actual "intake" of
the respective resource. Where protection zones already exist, new hydrological understanding
often shows that in many cases these areas are much larger than previously presumed, and that
protection zones must be expanded.
Such areas or zones are especially suitable for developing models of co-operation. One such
model is the foundation of a voluntary association of those concerned, namely farmers, water and
health administrations, representatives of agricultural associations and, amongst others, the
water supply organisation. A steering committee or executive board should be elected, in which
the water supply organisation should not take the lead. The guiding principle is "co-operation
instead of confrontation". An alternative may be to have direct contracts between the farmers and
the water supply agency. Such contracts bind the farmers to certain methods of production in
return for some financial support, especially during the years of transition from intensive farming
to sustainable methods. Tasks for such associations and their leadership are:
• Issuing regular advice, e.g. for suitable situations for applying manure, fertiliser or well-targeted
pesticides, and issuing prohibited periods (e.g. "no liquid manure on frozen ground"), or
computer-supported fertilisation schedules.
• Regulating the maximum density of livestock tolerable without risking pollution of the aquifer.
• Organising advice and training for farmers, e.g. for measuring soil content of fertilisers.
• Establishing time schedules for changes in land use (e.g. four years transition time for
extensivation, eight years of use as pasture, but then use only for forestry). Often, lenient time
schedules will be fulfilled much more quickly than required.
• Provide seeds for intercropping or keeping the ground covered to protect against erosion.
• Purchase suitable machines for demonstration or for communal use.
The general emphasis is on advice; on using, developing, publishing and making available local
experience and expertise; and on expressing concern and requesting responsible co-operation
rather than on issuing prohibitions. A very successful tool has been the installation of counsellors
for sustainable farming whose salaries are paid by the water supply agency, but who work within
the agricultural authorities and organise courses as well as giving individual advice and training.
The economic aim is to market products which have been audited in relation to health and
sustainable land use. Farmers can join an organisation for organic production methods and sell
their products under the name of the association; this should be encouraged by the association.
Such organisations then take over the responsibility for checking that members comply with the
rules and the methods of production; this helps to enforce sustainable production methods. Such
organisations can be encouraged to advertise for membership in the region.
The costs of such models of co-operation vary considerably, depending on whether farmers must
be supported during phases of transition and on the services provided. Nevertheless, experience
shows that these measures increase the price of water only by a few cents per cubic metre
(Such, 1996).
Establishing such co-operations is easiest if the land in the protection zone belongs to the water
supply company or agency and is only leased to the farmers. Furthermore, employing an expert
on agriculture may help a water supply agency considerably in negotiating with farmers. Success
with this approach has been reported by Such (1996), Höllein (1996) and Fleischer (1996).
8.4.3 Treatment of drinking water reservoir inlets
Where drinking water reservoirs with one major inflow have a large share of diffuse, non-
point source inputs and a strong need for rapid remediation, reduction of nutrients in the
inflow may be the most effective option. Pre-reservoirs with retention times of at least
several days can reduce total phosphorus inputs by 50-65 per cent (Klapper, 1992).
Retention times should allow incorporation of phosphorus into algal biomass and
sedimentation of that biomass, but should not be large enough for slow growing taxa,
such as cyanobacteria, to establish dominance. Sediment dredging may be necessary at
intervals of several years in order to counteract the re-release of phosphorus.
The Kis-Balaton reservoir in Hungary is an example of a special wetland and shallow

reservoir system of 60 km2 designed to retain phosphorus and reduce inputs to the
tenfold larger Lake Balaton. Water is retained for one month (mean value) in an intricate
system between coffer dams and reed zones. Phosphorus retention has been
successful, but flooding with stormwater from the Zala River has caused pulsed
phosphorus inputs to the lake (Padisák, Pers. Comm.).
If the largest nutrient share originates from a single major inflow, phosphorus stripping
facilities can be very effective in reducing inputs. Successful examples, with different
degrees of technological sophistication, are the Wahnbach Reservoir, Lake
Schlachtensee, Lake Tegel (Sas, 1989), and the Haltern Reservoir (Paetsch and Kötter,
1980).
8.5 Internal measures for nutrient and cyanobacterial control

In planning restoration measures for lakes and reservoirs, it is important to realise that
substantial time lags may occur between measures to reduce external inputs and the
results achieved in the water body. Feedback mechanisms within the ecosystem (e.g.
sediment-water interactions or the establishment of new dominant species) require time
to reach a new equilibrium. Hypertrophic aquatic ecosystems have specific positive
feedback mechanisms which stabilise trophic state and cyanobacterial dominance and
therefore resilience effects are not uncommon, even after substantial reduction of inputs
below thresholds calculated to be effective. Sas (1989) pointed out that resilience
patterns occur on two levels:
• Delayed response of in-lake total phosphorus concentrations to a reduction of input,

due to the time required for flushing phosphorus out of the water body, and the time
required for establishment of new sediment-water equilibria.
• Delayed response of phytoplankton biomass and species composition to reductions of

in-lake total phosphorus concentrations, due to stability of prevailing biocoenosis
structures and/or biotic enhancement of internal load.
Experience shows that several years (up to 10) may be necessary between the
implementation of a restoration measure that substantially decreases inputs, and visible
success in terms of reduction of phytoplankton biomass and cyanobacterial blooms.
Monitoring of phosphorus inputs and phosphorus concentrations in the recipient water
body during this time is recommended. Usually, a declining trend in total phosphorus
concentration will be the first detectable response to a reduction in inputs and will
indicate whether a particular measure can be expected to be successful, but it may take
years for phosphorus concentrations to decline below the threshold effective for
controlling phytoplankton biomass. Often, such time lags are due to the (sometimes
substantial) phosphate storage capacity of anoxic sediments typical in hypertrophic
waters. Flushing rates (i.e. the inverse of retention times) strongly influence the time
necessary to reach a new equilibrium. In some cases, particularly in water bodies with
low water exchange rates, supplementary "internal" measures may be advisable in order
to accelerate a response. The following sections briefly introduce and evaluate a number
of such measures for which experience is available.
With very few exceptions, internal measures are appropriate only after, or in combination
with, an effective reduction of external inputs. In principle, reduction of inputs should be
the actual restoration or resource protection measure, and internal measures should
serve as a further boost to switch the ecosystem out of resilience and into a new balance.
If possible, a few years of patience and observation of nutrient concentration trends
within the lake will show whether internal measures are necessary. Only rarely are
internal measures without adequate reductions of inputs justified as an emergency
approach; usually such measures require continuous operation (such as aeration, see
section 8.5.2) or repeated application (such as in-lake phosphate precipitation, see
section 8.5.1). Medium- to long-term success of restoration investments is at stake if this
principle is not considered.
8.5.1 In-lake phosphorus precipitation
In lakes and reservoirs with high water retention times, decline of phosphorus
concentrations may be very slow, even after external inputs have been reduced to levels
which should ensure a mesotrophic or oligotrophic state. If the water body has a high
phosphorus content that is flushed out only slowly, only some of the phosphorus within
the biomass will settle to the sediments. Much of it is released from decaying organic
material, entrained back into the water body by water circulation, taken up by
cyanobacteria or algae and, in part, passed on to higher levels of the food web.
Degradation of organic material at the sediment surface has often led to anoxic
conditions which may accelerate phosphorus release rates dramatically. Thus,
phosphorus within a lake can be recycled many times, and no decline of cyanobacterial
biomass can be achieved without reducing this in-lake phosphorus pool. Sometimes, this
situation also applies to lakes which are naturally eutrophic, such as lakes in western
Canada situated on phosphorus-rich glacial till (Prepas et al., 1997). Precipitation of
phosphorus from the water body to the sediment can be a successful measure, if it is
undertaken so that phosphorus remains permanently bound in the sediment.
Prerequisites for lasting success are low external loading, sufficient depth to prevent
sediment resuspension due to wind events, and adequate choice of flocculants.
Experiments with precipitation of phosphorus have been undertaken with aluminium
sulphate, ferric salts (chlorides, sulphates), ferric aluminium sulphate, clay particles and
lime (as Ca(OH)2 and as CaCO3).
Ferric salts are effective in precipitating phosphorus, but difficult to handle because of
their aggressive acidity. Furthermore, the iron-phosphorus complex is stable only under
oxic conditions. Thus application of ferric salts usually requires subsequent continuous
aeration to avoid re-dissolution of phosphorus under anoxic conditions. Due to the high
mobility of iron ions, addition of iron frequently often has to be repeated at regular
intervals. In addition, Prepas et al. (1997) point out that iron may be a limiting micro-
nutrient in some systems and, in such situations, treatment with ferric salts may actually
stimulate growth of cyanobacteria and algae.
Aluminium sulphate is poorly soluble under neutral and high pH conditions, but may
decrease pH in waters with low buffering capacity, which leads to solubilisation and
problems of alum toxicity.
Lime (both Ca(OH)2 and CaCO3) has been used as an algicide to coagulate and
precipitate phytoplankton cells out of the water column (Murphy et al., 1990; Zhang and
Prepas, 1996). It is non-toxic, usually fairly inexpensive, and the pH-shock for the
aquatic biota can be minimised by careful dosing over an extended time span. Unlike
treatment with copper sulphate, the precipitation of cyanobacterial cells with Ca(OH)2
does not appear to cause cell lysis and toxin release into the water (Kenefick et al., 1993;
Lam et al., 1995). Lime also functions, to some extent, as a longer-term algal inhibitor,
reducing eutrophication by precipitating phosphorus from the water (Murphy et al., 1990).
It appears that Ca(OH)2 is more effective than CaCO3 in precipitating phosphorus
(Murphy et al., 1990). Many of the studies of both the mechanism and effects of liming
for algal control have been carried out in eutrophic, hard water lakes or farm dugouts
(dams) in Alberta, Canada (Murphy et al., 1990; Zhang and Prepas, 1996). It is possible
that the technique may be more effective in these conditions than in soft water. The dose
rates used are also quite high (e.g. 50-250 mg l-1 Ca(OH)2) (Zhang and Prepas, 1996)
which would make the technique prohibitive for large lakes. Techniques for the
application of lime, which involve pumping or spraying of a slurry, are described by
Prepas et al. (1990b).
Experience with in-lake precipitation of phosphorus is increasingly being compiled. A

number of documented case studies show success either in terms of reducing
phytoplankton biomass or in terms of shifting species dominance away from
cyanobacteria. Nevertheless, numerous unsuccessful cases have also been
documented, and further development of these techniques is ongoing (see compilation
in Klapper (1992) and in Cooke et al. (1981, 1993)). Furthermore, in some water bodies,
the concentrations of iron or calcium compounds in the inflow are naturally high and
regularly provide sufficient binding sites for phosphate to induce natural phosphorus
precipitation. Measures in the catchment area or changes in inflow regime may have
considerable impact in either increasing or decreasing this input, and thus may have a
significant impact on the trophic state of the water body.
8.5.2 Sediment dredging and phosphorus binding
Release from sediments may be a substantial source of phosphorus (sometimes

referred to as internal loading) for many years after external inputs have been minimised.
Water exchange rates, sediment chemistry, temperatures, mixing conditions, and
bioturbation govern phosphorus release rates. Iron-bound phosphorus is highly sensitive
to redox conditions; when sediment surfaces turn anoxic during summer stratification,
phosphorus concentrations may increase dramatically, fertilising cyanobacteria in their
optimum growing season. Under oxic conditions in shallow, unstratified systems, high
pH (> 9.8) may strongly enhance oxic phosphorus release (Ryding, 1979). Because high
pH values are a result of intensive photosynthetic activity, this phosphorus release
pathway is a positive feedback mechanism in favour of cyanobacterial blooms. Other
aerobic phosphorus release mechanisms may also be significant, especially bioturbation
by feeding fish and invertebrates (Gardner et al., 1981).
Options for measures to counteract sediment release are removal of sediment (dredging)
or treatment to bind phosphorus. Dredging is costly and will reduce release rates only if:
• It is carried out down to sediment layers with a lower or less mobile phosphorus
content.
• Phosphorus-rich interstitial water is handled in such a fashion that it does not reach the
water body and cause additional inputs.
• Dredged sludge can be deposited where it does not create a new external input with
erosion and stormwater runoff into the lake.
In some urban and industrial regions, dredging is precluded or complicated by high

concentrations of heavy metals and organic contaminants in the sediments which would
then require disposal as hazardous waste. Dredging is particularly recommended for
smaller water bodies where the trophic state can be further improved by gaining depth,
or which also need to be cleared of dumped rubbish.
Sediment treatments aim at trapping phosphorus in the sediment, either by oxidation to

insoluble iron compounds, or by adsorption onto calcium carbonate or clay particles.
During the past two decades, broad experience collected with numerous failures (see
Box 8.8) and a few successful cases has shown that effective treatment requires careful
design on the basis of profound understanding of the sediment chemistry and hydrology
of the water body to be treated. Oxidisation may be achieved by aeration, artificial mixing
(see also section 8.5.5), or the introduction of pure oxygen. It appears to be most
effective if achieved with nitrate, which transports more oxygen and penetrates more
readily into sediments. Well-treated sewage effluent (not contaminated with harmful
substances, fully nitrified and after phosphorus removal) may be suitable for this purpose,
if the process is controlled so that nitrate concentrations are not elevated in drinking
water.
Box 8.8 Is aeration effective in binding phosphate in sediments?
The frequent failures in the use of aeration to meet the objective of reducing phosphorus efflux
from sediments require critical highlighting. Many aeration projects had several objectives, often
not carefully distinguished and planned, such as (i) providing sufficiently high oxygen
concentrations for survival offish and fish eggs in deep waters and on the sediment surface, (ii)
destratification in order to entrain buoyant cyanobacteria, and (iii) oxidising sediment surfaces.
Some of these objectives may be conflicting, e.g. destratification will increase sediment surface
temperatures, thus potentially enhancing phosphorus release, and it will transport nutrient-rich,
near-surface water into upper strata where these nutrients can be used for growth of
cyanobacteria or other phytoplankton. Often, aeration has proved to be insufficient for achieving
the aim of reducing phosphorus release. Even the prominent, carefully designed, experiment at
the Swiss Baldegger See did not succeed in increasing phosphorus retention of the sediments
after 10 years of operation (Wuest and Wehrli, 1996). Energy costs of aeration may be
considerable. At the present state of the art, it can be recommended only for increasing the
oxygen content of the water (e.g. as a fish habitat), or if artificial mixing is desired - success in
increasing phosphorus retention in lake sediments appears doubtful. Injection of pure oxygen
appears to be more successful in some cases (Gemza, 1997; Prepas et al., 1997).
8.5.3 Withdrawal of bottom water from the hypolimnion
In thermally stratified eutrophic lakes, phosphorus accumulates in the hypolimnion (cold

bottom water layer) during summer stagnation, partly from settled organic material
originating in the upper water layers and, in many lakes, largely from the release of
sediment-bound phosphorus under anoxic conditions. Although most natural outflows
drain surface water, it is often possible to dam the natural outflow and to abstract
hypolimnetic water instead (Olzewski, 1961). This is especially easy to apply to
reservoirs and can reduce in-lake concentrations significantly. In the Swiss Mauensee
the biomass of Planktothrix rubescens was reduced from 152 g m-3 to 42 g m-3 using this
approach (Gächter, 1976).
Nürnberg (1997) compiled the advantages of hypolimnetic withdrawal during summer
stratification as a method based solely on selective output of total-P rich water. The
advantages of the method are:
• It addresses the cause of eutrophication.

• It does not add chemicals.
• It does not necessarily change the water budget.
• It can break the cycle of enhanced sediment accumulation of total phosphorus.
• It can flush more phosphorus out of the system than the sediments accumulate each
year.
Hypolimnetic withdrawal is effective only if enough water flows into the lake. Furthermore,
some lowering of the water level may be tolerable, but complete destratification by
removal of most of the hypolimnion should be avoided, because increasing the contact
area between warm surface water and sediments will enhance phosphorus release due
to elevated temperatures. In addition, impairment of water quality downstream will
require attention if the amount of phosphorus released is high in relation to the total flow.
Downstream phosphorus pollution may be avoided by treatment of the hypolimnion
outlet with chemical phosphorus precipitation. Nevertheless, the low temperatures of the
hypolimnion water may have a substantial impact on downstream biological processes,
such as fish breeding.
8.5.4 Reduction by flushing
Flushing with water of low phosphorus concentrations can greatly reduce external inputs
and will also accelerate recovery from internal loading by removing in-lake phosphorus
which would otherwise be recycled for a number of growing seasons. If suitable water is
available in sufficient quantity, flushing can be a very effective tool for reduction of
cyanobacterial proliferation. Successful examples are Veluwemeer in the Netherlands
(Sas, 1989) and Moses Lake in the USA (Welch et al., 1972). However, this measure
also implies a relocation of the phosphorus to another water body, and this impact must
also be evaluated.
8.5.5 Hydrophysical measures
Cyanobacteria show different "strategies" of survival in competition against other

phytoplankton organisms (section 2.3). Many of these strategies are adapted to specific
hydrophysical conditions. Changing these conditions may therefore substantially reduce
the success of these cyanobacterial "ecostrategists" and allow other phytoplankton
species to become dominant. This approach can be an effective temporary,
supplementary measure alongside reduction of external inputs of nutrients, particularly if
in-lake nutrient concentrations have declined to values around the threshold where
success may be expected. In some cases, where eutrophication levels cannot be
decreased, permanent installation of hydrophysical measures can be a solution (see
Visser et al. (1996) for the example of Nieuwe Meer in Amsterdam).
The mass development of scum-forming species is highly dependant on the stability of

the water column. In water without vertical mixing, the colonies of Microcystis or other
colony-forming taxa can migrate up and down by changing their specific weight (see
section 2.3). Interrupting this vertical migration of the colonies by artificial mixing of an
otherwise stably stratified water body, can prevent rapid development of surface scums.
Furthermore, disrupting the possibility for these organisms to move into strata with
optimum light conditions is likely to reduce their growth rate and thus their efficiency in
competing against other phytoplankton. In contrast, mixing improves growth conditions
for taxa such as diatoms, which depend on mixing to remain in suspension. Thus,
increased mixing may shift species composition from cyanobacteria to, for example,
diatoms.
Thermally stratified water bodies naturally have an upper mixed layer known as the
epilimnion. If artificial mixing substantially increases the depth of this layer, it reduces the
light-determined carrying capacity, or the concentration of phytoplankton biomass
possible (see also section 8.1). To be successful, artificial mixing measures must satisfy
three conditions (see Visser et al., 1996):
• At least 80 per cent of the water volume should be mixed.
• The artificial mixing rate must be higher than the rate of vertical movement of the
colonies of cyanobacteria. Rates of colony movement depend on colony size and thus
are somewhat variable (see section 2.3), but as a general rule, a mixing rate of 1 m h-1 is
sufficient to prevent cyanobacterial blooms.
• A large part of the water body must be sufficiently deep. In most cases artificial mixing
has been caused by installing aeration tubes which are connected to a compressor on
the shore. The aeration tubes are situated in the deeper regions of the water body.
Waters with extensive shallow areas have a low circulation rate which can negatively
influence the results of artificial mixing. Furthermore, if the water body is too shallow,
mixing cannot reduce the light-determined carrying capacity strongly enough to prevent
cyanobacterial growth. The example of Nieuwe Meer (Visser et al., 1996) shows that
more than 20 m depth may be required.
A number of mixing projects have been unsuccessful because these principles were
neglected. Many systems are now on the market, provided by different engineering
companies. Engineering expertise is sufficiently developed to design systems that can
meet the hydrophysical requirements. Care must be taken, however, to select competent
companies, and to plan the measure to meet the ecological targets set in combating
cyanobacterial blooms. Furthermore, in tropical and subtropical countries with high and
prolonged insulation, the costs of systems are enough for mixing to become prohibitive.
8.5.6 Biomanipulation
Biomanipulation includes a range of techniques that influence algal growth by

manipulation of parts of the food web of a lake. Examples are removal of planktivorous
and benthivorous fish populations, providing refuges for zooplankton and introducing
predatory fish such as pike (Esox lucius) in order to decimate planktivorous fish
populations, and introducing submerged aquatic plants to compete with phytoplankton in
consuming nutrients (Kitchell, 1992). These techniques aim at stimulating the growth or
presence of phytoplankton-grazing organisms or of phytoplankton competitors.
Increasing grazing pressure
In shallow lakes, the removal of a large proportion of benthic and planktivorous fish can
be helpful to diminish algal growth. Without this predation pressure, zooplankton and
benthic fauna can develop and feed on algae and some species of cyanobacteria (e.g.
early stages of Microcystis population growth when colonies are still very small).
Selective removal of benthic fish reduces resuspension of sediments and thus
mobilisation rates of sediment phosphorus (in deep lakes this is difficult). To stimulate
these effects, predatory fish fingerlings can be introduced to diminish the population
growth of the planktonic and benthic fish. Artificial refuges can be placed to provide
habitats for zooplankton and pike. The artificial refuges are important when the
development of submerged aquatic plants (macrophytes) is insufficient to serve this
purpose.
The introduction of predatory fish can be effective. Interventions into established

hypertrophic ecosystem structures by fish stock management techniques have proved
successful in smaller ponds and lakes over shorter periods of time (Hrbácek et al., 1978).
If successful the water may become clearer due to a reduction of algal and/or
cyanobacterial turbidity, the zooplankton populations increase, and fields of macrophytes
may develop which compete for phosphate with the phytoplankton (thus reducing their
capacity for growth) (see below). However, the breeding success of the remaining
planktivorous fish stock in the lake will be high if insufficient predatory fish are present.
Continued control of the development of the fish stocks is required, and the removal of
planktivorous fish must be repeated regularly. Depending upon local salary levels, this
may be expensive in terms of personnel. Biomanipulation is by no means a cheap
method because of the continuous monitoring and management requirements. It is also
unlikely that the technique will work naturally and unaided once the change in biological
structure has been introduced.
A disadvantage of biomanipulation is that not all phytoplankton species are eaten

efficiently by zooplankton. Stimulating the zooplankton without reducing concentrations
of phytoplankton nutrients may stimulate dominance of inedible phytoplankton species,
such as colony-forming (Microcystis, Aphanizomenon) or filamentous cyanobacteria
(Planktothrix agardhii), or the green alga, Enteromorpha. High nutrient levels may also
stimulate the growth of epiphytic algal species which grow on the surfaces of
macrophytes and suppress their development.
Box 8.9 Ecosystem theory to explain how biomanipulation works
As indicated in section 2.2, eutrophic systems with blooms of Planktothrix (formerly Oscillatoria)
can show enormous stability, with uninterrupted dominance over years. May (1977) indicated that
multispecies assemblages of plants and animals can have several different equilibrium states.
Scheffer (1990) highlighted this for shallow eutrophic lakes. He concluded that during the process
of lake restoration, two different equilibria are possible at a state of moderate eutrophication - one
with large populations of phytoplankton and planktivorous and benthic fish, and another in which
nutrients are incorporated more evenly distributed among zooplankton, fish and macrophytes.
The first system is turbid, the second system is clear. Resilience of the ecosystem during
restoration maintains the turbid state over long periods, particularly if phosphate concentrations
oscillate around the threshold effective for reducing phytoplankton biomass, but do not
substantially decline below this level. In such situations, biomanipulation can help to switch the
ecosystem from a turbid phytoplankton community to a clear macrophyte community.
Enhancing competition by introducing macrophytes
The introduction of macrophytes has the best chance of success in water bodies with a
relatively large shallow littoral area (< 2-3 m deep) and at moderate concentrations of
total phosphorus. Reynolds (1997) points out that if the areal nutrient input rate leads to
phosphorus concentrations above 0.15 mg 1-1, phytoplankton density can readily reach
0.15 mg 1-1 of chlorophyll a, and then submerged macrophytes will simply be "shaded
out". However, at lower concentrations of phosphorus, particularly in spring,
macrophytes have the chance to begin to grow and to incorporate enough of the
available phosphorus to achieve substantial phosphorus limitation of phytoplankton
biomass. Particularly if combined with the management of fish stocks, measures to
support macrophytes may switch an aquatic ecosystem into a different, sometimes
rather stable, biological structure resulting in clear water and low cyanobacterial biomass
(Box 8.9).
General assessment
Reynolds (1997) summarises current knowledge on mechanisms of biomanipulation

under the heading "bottom-up or top-down control"; "bottom up" implies control by
nutrients and "top-down" implies control by the top end of the food chain, i.e. by
consumers. Reynolds (1997) lists arguments against biomanipulation, such as self-
starvation of consumers by outstripping the availability of phytoplankton or
consequences for other components of the food web, and points out that the conceptual
flaw is thinking in terms of "steady states". He comments:
"the state that generally attains is a lurching alteration between responses to plenty and
responses to over-consumption.... When tropically-related organisms, with lives
measured in hours to years, inhabit environments fluctuating with periods of days to
months, their interactions are certainly likely sometimes to result in strong top-down
pressures on producer biomass. At other times, however, the trophic cascade slows
down to barely a trickle" (Reynolds, 1997).
These dynamic fluctuations are most likely to occur, such that they affect phytoplankton
populations, in situations which are not extremely biased by total lack of nutrient
limitation. Biomanipulation as a management tool to reduce algal or cyanobacterial
growth is most likely to be successful in situations of moderate nutrient concentrations
and in combination with reductions in inputs. Experience shows that, as long as the
trophic level of the systems remains high, the risk that the ecosystem switches back into
its original composition is also higher. For reviews of biomanipulation see Demerol et al.
(1992), Carpenter and Kitchell (1992) and Moss et al. (1994).
8.5.7 Algicides
Algicides, especially copper sulphate, have been used rather widely in some regions to
kill prevailing cyanobacterial blooms. As a result of the Palm Island catastrophe (see
section 4.1) it was established that lysine a bloom may exacerbate problems because
toxins previously contained within the cells are liberated and pass through drinking water
filters far more readily than toxins within intact cells (see also Lam et al., 1995).
Nevertheless, preventative treatment at the beginning of bloom development has been
widely used (Cameron, 1989) and may be necessary (see section 9.2).
Algicide treatment of water bodies is best considered as an emergency measure and
may involve ecological risks. Toxic copper deposits may accumulate in the sediments
(Prepas and Murphy, 1988). Repeated treatment may induce shifts in species
composition towards more copper-resistant, but not necessarily more pleasant, species.
This was the case in Lake Matthews, a drinking water reservoir for California, where
taste-and-odour problems caused by Oscillatoria spp. were handled by copper sulphate
application. Within very few years, the dosage needed to combat these species had to
be increased from 27 to 400 t. The treatment led to the replacement of Oscillatoria by a
more copper-resistant cyanobacterium, Phormidium sp., which prevailed for longer time
spans and caused almost all year-round off-flavour problems (Izaguirre, 1992). Other
undesirable ecosystem impacts of algicide treatment cannot be excluded. Wherever
possible, it is preferable to choose abatement measures which address the source of the
problem (i.e. growth conditions for cyanobacteria) because such solutions may be
effective in the long term and actually safeguard human health by improving
environmental quality.
8.5.8 Barley straw
The use of decomposing barley straw for the control of cyanobacteria and microalgae
has been investigated recently (Welch et al., 1990; Jelbart, 1993; Newman and Barrett,
1993; Everall and Lees, 1996). The effect of rotting barley straw in reducing filamentous
green algal growth was reported by Welch et al. (1990) and algistatic effects were shown
in laboratory cultures of the cyanobacterium Microcystis aeruginosa by Newman and
Barrett (1993). The inhibitory effects were suggested to be due to antibiotic production
by the fungal flora or to the release of phenolic compounds such as ferulic acid and f-
coumaric acid from the decomposition of the straw cell walls.
A reduction in cyanobacterial populations has also been reported in reservoir trials after
applying barley straw (Everall and Lees, 1996). These authors suggested that phytotoxic
compounds released from decomposing straw inhibited the cyanobacterial populations,
but that further chemical identification, and risk and environmental assessment were
required prior to use in water supply reservoirs. However, a recent full-scale field trial
has been carried out in a potable supply reservoir and was credited with reducing
regular summer cyanobacterial populations (Barrett et al., 1996). Although these trials
gave apparently favourable algistatic results, they were carried out without replication or
control trials and, as such, the study design cannot account for the influence of other
factors (such as impact of weather conditions) on phytoplankton development and
succession. An earlier trial which did include the dosing with barley straw of one of a pair
of closely adjacent lakes (with similar hydrology and biology) resulted in a decrease in
the cyanobacterial population in the straw-dosed lake as compared with the non-dosed
lake, throughout the two years of post-dose monitoring (Harriman et al., 1997).
There are conflicting data from Australia on the effects of barley straw. Jelbart (1993)
failed to find any inhibitory effects with extracts of rotting straw on Microcystis
aeruginosa isolates. Cheng et al. (1995) also found no algicidal or algistatic effects from
barley straw over a six month period in a comprehensive field trial in six experimental
ponds. The ponds were fertilised to encourage cyanobacterial growth and there were no
differences in species composition or final standing crop between control and straw-
dosed ponds.
These contradictory findings and the unknown identity of the inhibitory factors in rotting
barley straw indicate that straw-dosing is still too poorly understood to recommend for
reliable use as a cyanobacterial control measure, particularly in potable water supply
reservoirs. Whether barley-straw dosing influences the composition and size of toxin
pools in cyanobacterial populations still needs to be determined. Dosing with barley
straw has gained unwarranted popularity and notoriety because it is an apparently
simple procedure which is relatively inexpensive and highly visible. It is being used in
freshwaters for cyanobacterial control in some countries even though the benefits are
dubious (e.g. the introduction of rotting, oxygen-consuming organic matter needs
consideration).
8.5.9 Other approaches
A market for rapid and cheap water resource protection and restoration methods is
evolving. In Europe, poorly validated methods for reduction of cyanobacterial and
phytoplankton growth are being advertised, some of which are questionable. In some
cases, transient success was actually due to natural seasonal "clear water" phenomena.
Although new approaches require field testing as part of development, independent
verification of their success can reasonably be requested of their promoters prior to
marketing. Public health officers and other public authorities are trained in other fields
than environmental sciences and rarely have the expertise to judge restoration
proposals. Review by environmental authorities or experts is therefore desirable prior to
investment.
8.6 References
Barica, J. and Allan, R.J. 1997 Aquatic ecosystem restoration. Wat. Qual. Res. J.
Canada, 32, 452 pp.
Barrett, P.R.F., Curnow, J.C. and Littlejohn, J.W. 1996 The control of diatom and
cyanobacterial blooms in reservoirs using barley straw. Hydrobiologia, 340(1-3), 307-311.
Bartram, J. and Ballance, R. 1996 Water Quality Monitoring. A Practical Guide to the
Design and Implementation of Freshwater Quality Studies and Monitoring Programmes.
E & FN Spon, London, 383 pp.
Bossard, P. and Gächter, R. 1996 Controversial hypothesis related to the ban on

phosphates. EAWAG News, 42E, 18-20.
Cameron, C.D. 1989 Is this a way to run a reservoir? In: Practical Lake Management for
Water Quality Control. Proceedings of a Seminar, Los Angeles, CA, American
Waterworks Association, Denver, 63-83.
Carpenter, S.R. and Kitchell, J.F. 1992 Trophic cascade and biomanipulation: interface
of research and management. A reply to the comment by De Melo et al. Limnol.
Oceanog., 37, 208-213.
Charlton, M. 1997 The sewage issue in Hamilton Harbour: Implications of population

growth for the remedial action plan. Wat. Qual. Res. J. Canada, 32, 407-420.
Cheng, D., Jose, S. and Mitrovic, S. 1995 Assessment of the possible algicidal and
algistatic properties of barley straw in experimental ponds - confirmatory trial. Report
prepared for the State Algal Coordinating Committee, University of Technology, Sydney,
21 pp.
Cooke, G.D. and Kennedy, R.H. 1981 Precipitation and Inactivation of Phosphorus as a
Lake Restoration Technique. Technical Report EPA-600/8-81/012, Environment
Protection Agency.
Cooke, G.D., Welch, E.B., Peterson S.A. and Newroth P.R. (Eds) 1993 Restoration and
Management of Lakes and reservoirs. Second edition, Lewis Publishers, CRC Press Inc.,
Boca Raton, Fla., 548 pp.
Cuddy, S., Young B., Davis R. and Farley T. 1997 Trialing the Catchment Management
Support System in the Murrumbidgee catchment, New South Wales. In: J.R.D. Davis
[Ed.] Managing Algal Blooms: Outcomes from the CSIRO Blue-Green Algal Research
Program. CSIRO Land and Water, Canberra, 103-113.
Davis, J.R. and Farley, T.F.N. 1997 CMSS: Policy analysis software for catchment
managers. J. Env. Modelling and Software, 12, 197-210.
DeMelo, R., France, R. and McQueen, D.J. 1992 Biomanipulation - hit or myth? Limnol.
Oceanog., 37, 192-207.
Everall, N.C. and Lees, D.R. 1996 The use of barley-straw to control general and blue-
green algal growth in a Derbyshire reservoir. Wat. Res. 30(2), 269-276.
Fleischer, H. 1996 (Kommunale Wasserwerke Leipzig GmbH, Johannesgasse 7-9, D-

04103 Leipzig) Biologischer Landbau im Einzugsgebiet von Waserversorgungsanlagen -
Ergebnisse und Perspektiven. 3. Wasserhygienetage Bad Elster, 5-7 February 1996.
Gächter, R. 1976 Die Tiefenwasserableitung, ein Weg zur Sanierung von Seen. Schweiz.
Z. Hydrolog,. 38, 1-28.
Gardner, W.S., Nalepa, T.F., Quigley, M. and Malczyk, J. 1981 Release of phosphorus
by benthic invertebrates. Can. J. Fish. Aquat. Sci., 38, 978-981.
Gemza, A.F. 1977 Water quality improvements during hypolimnetic oxygenation in two
Ontario lakes. Wat. Qual. Res. J. Canada, 32, 365-390.
Gleisberg, D., Erftstadt, H. and Hahn, H. 1995 Zur Entwicklung der Phospherentfernung
aus Abwässern der Bundesrepublik Deutschland. Korrespondenz Abwasser, 42, 958-
969.
Gray, N.F. 1994 Drinking Water Quality. John Wiley & Sons, Chichester, 315 pp.
Harremoes, P. 1997 The challenge of managing water and material balances in relation
to eutrophication. In: R. Roijackers, R.H. Aalderink and G. Blorn [Eds] Eutrophication
Research, State-of-the-Art. Department of Water Quality Management and Aquatic
Ecology, Wageningen Agricultural University, 3-12.
Harriman, R., Adamson, E.A., Shelton, R.G.J. and Moffett, G. 1997 An assessment of
the effectiveness of straw as an algal inhibitor in an upland Scottish loch. Biocon. Sci.
Technol., 7(2), 287-296.
Heinzmann, B. and Chorus, I.1994 Restoration concept for Lake Tegel, a major drinking
and bathing water resource in a densely populated area. Environ. Sci. Technol., 28,
1410-1416.
Höllein, K. 1996 (Hauptabteilung Wasserversorgung, Unterer Anger 3, D-80287

München) Biologischer Landbau 1996 - Lösungsweg für die Koexistenz von
Wasserwirtschaft und Landwirtschaft. (biological farming - solutions for coexistence of
water supplies and agriculture) 3. Wasserhygienetage Bad Elster, 5-7 February 1996.
Hrbácek, J., Desortová, B. and Popovský, J. 1978 Influence offish stock on the
phosphorus-chlorophyll-ration. Verh. Verein. Theor. Ang. Limnol., 20, 1624-1628.
Izzaguire, G. 1992 A copper-tolerant phormidium species from Lake Mathiews,

California, that produces 2-methylisoborneol and geosmin. Wat. Sci. Tech.. 25, 217-223.
Jelbart, J. 1993 Effect of rotting barley straw on cyanobacteria: a laboratory investigation.

Water, 5, 31-32.
Kenefick, S.L., Hrudey, S.E., Peterson, H.G. and Prepas E.E. 1993 Toxin release from
Microcystis aeruginosa after chemical treatment. Wat. Sci. Technol., 27, 433-440.
Kitchell, J.F. [Ed.] 1992 Food Web Management - A Case Study of Lake Mendota.
Springer Verlag, New York, 553 pp.
Klapper, H. 1992 Eutrophierung und Gewässerschutz. Gustav Fischer Vlg. Jena,

Stuttgart, 277 pp.
Köhler, A. and Klein, M. 1997 Cyanobakterien und die Nutzung der Berliner Gewässer.
In: I. Chorus [Ed.] Toxische Cyanobakterien in deutschen Gewässern. WaBoLu Hefte
4/97, 58-66.
Lam, A.K.Y., Prepas, E.E., Spink, D. and Hrudey, S.E. 1995 Chemical control of
hepatotoxic phytoplankton blooms: implications for human health. Wat. Res., 29, 1845-
1854.
Lindenschmidt, K.E., Suhr, M., Magumba, M.K., Hecky, R.E. and Bugenyi, F.W.B. 1998
Loading of solute and suspended solids from rural catchment areas flowing unto Lake
Victoria in Uganda. Wat. Res., 32, 2776-2786.
May, R. 1977 Thresholds and breakpoints in ecosystems with a multiplicity of stable

states. Nature, 269, 471-477.
Moss, B., McGowan, S. and Carvalho, L. 1994 Determination of phytoplankton crops by
top-down and bottom-up mechanisms in a group of English lakes, the West Midland
meres. Limnol. Oceanog., 39, 1020-1029.
Murphy, T.P, Prepas E.E., Lim, J.T., Crosby, J.M. and Walty, D.T. 1990 Evaluation of
calcium carbonate and calcium hydroxide treatments of prairie drinking water dugouts.
Lake Reserv. Manage. 6, 101-108.
Newman, J. and Barrett, P.R.F. 1993 Control of Microcystis aeruginosa by decomposing

barley straw. J. Aq. Plant Manage., 31, 203-206.
Nürnberg, G.K. 1988 The prediction of phosphorus release rates from total and
reductant-soluble phosphorus in anoxic lake sediments. Can. J. Fish. Aquat. Sci. 45,
453-462.
Nürnberg, G.K. 1997 Coping with water quality problems due to hypolimnetic anoxia in
Central Ontaria Lakes. Wat. Qual. Res. J. Canada, 32, 391-405.
Oenema, O. and Roest, C.W.J. 1997 Nitrogen and phosphorus losses from agriculture
into surface waters. In: R. Roijackers, R.H Aalderink and G. Blorn [Eds] Eutrophication
Research, State-of-the-Art. Department of Water Quality Management and Aquatic
Ecology, Wageningen Agricultural University, 13-15.
Olszewski, P. 1961 Versuch einer Ableitung des hypolimnischen Wassers an einem See.
Ergebnisse des ersten Versuchsjahres. Verh. Int. Ver. Limnol., 18, 1792-1797.
Packham, R.F. 1994 The contamination of water from agriculture. In: A.M.B Golding, N.
Noah and R. Stanwell-Smith Water and Public Health. Sith Gordon and Co. Limited and
Nishimura Co. Limited, 145-154.
Paetsch, B. and Kötter, K. 1980 Verminderung der Algenentwicklung in der Talsperre

Haltern durch Phosphat-Faellung. Gwf-wasser/abwasser, 212, 496-498.
Prepas, E.E. and Murphy, T.P. 1988 Sediment-water interactions in farm dugouts
previously treated with copper sulfate. Lake Reserv. Manage., 4, 161-168.
Prepas, E.E., Murphy, T.P., Crosby, J.M., Walty, D.T., Lim, J.T., Babin J.M. and
Chambers, P.A. 1990 Reduction of phosphorus and chlorophyll a concentrations
following CaCO3 and Ca(OH)2 additions in hypereutrophic Figure Eight Lake, Alberta.
Environ. Sci. Technol., 24, 1252-1258.
Prepas, E.E., Murphy, T.P., Dinsmore, W.P., Burke, J.M., Chambers, P.A. and Reedyk,
S. 1997 Lake management based on lime application and hypolimnetic oxygenation: the
experience in eutrophic hardwater lakes in Alberta. Wat. Qual. Res. J. Canada, 32, 273-
293.
Reynolds, C.S. 1997 Vegetation Processes in the Pelagic: A Model for Ecosystem
Theory. Excellence in Ecology, Ecology Institute, Oldendorf-Luhe, 371 pp.
Rönicke, H. 1986 Beitrag zur Fixation des molekularen Stickstoffs durch planktische
Cyanophyceen in einem dimiktischen, schwach durchfolssenen Standgewässer. Diss. A.
Humboldt-Univ. Berlin, 129 pp.
Round, F.E. 1965 The Biology of the Algae. Edward Arnold, London.
Ryding, S.O. 1979 Reversibility of man-induced eutrophication. Experiences of a lake

recovery study in Sweden. Int. Revue ges. Hydrobiol., 66, 449-503.
Sas, H. 1989 Lake Restoration by Reduction of Nutrient Loading: Expectations,

Experiences, Extrapolations. Academia Vlg. Richarz, 479 pp.
Scheffer, M. 1990 Multiplicity of stable states in freshwater systems. Hydrobiologia,

200/201, 475-486.
Siegrist, H. and Boller, M. 1997 Effects of the phosphate ban on sewage treatment.
EAWAG News, 42 E, 9-11.
Somlyody, L. and van Straten, G. 1986 Background to the Lake Balaton Eutrophication
Problem. In: L. Somlyody and van Straten [Eds] Modeling and Managing Shallow Lake
Eutrophication. Springer Verlag, Berlin, 3-18.
Such, W. 1996 (Wahnbachtalsperrenverband, Kronprinzenstr. 13, D-53721 Siegburg)

1995 Kooperation Wasserwirtschaft - Landwirtschaft. 3. Wasserhygienetage Bad Elster,
5-7 February 1996.
Trudgill, S.T. [Ed.] 1995 Solute Modelling in Catchment Systems. John Wiley and Sons,
Inc. N.Y., 460 pp.
P.M. Visser, Ibelings, B.W., van der Veer, B., Koedood, J. and Mur, L.R. 1996 Artificial
mixing prevents nuisance blooms of the cyanobacterium Microcystis in lake Nieuw Meer,
the Netherlands. Freshwat. Biol., 36, 435-450.
Vollenweider, R. and Kerekes, J. 1982 Eutrophication of Waters, Monitoring,

Assessment, Control. Organisation for Economic Co-operation and Development, Paris.
Wehrli, B., Wüest, A., Bührer, H., Gächter, R. and Zobrist, J. 1996 Überdüngung der
Schweizer Seen - erfreulicher Trend nach unten. EAWAG News, 42D, 12-14.
Welch, E.B., Buckley, J.A. and Bush, R.M. 1972 Dilution as an algal bloom control. J.
Water Poll. Contr. Fed., 44, 2245-2265.
Welch, I.M., Barren, P.R.F., Gibson M.T. and Ridge, I.1990 Barley straw as an inhibitor
of algal growth I: Studies in the Chesterfield Canal. J. App. Phycol., 2, 231-239.
Wetzel, R.G. 1983 Limnology. Second edition. Saunders College Publishing,

Philadelphia, 766 pp.
World Bank 1993 Water Resources Management. A World Bank Policy Paper. The
World Bank, Washington D.C., 140 pp.
Wüest, A. and Wehrli, B. 1996 Zehn Jahre Seenbelüftung - Erfahrungen und Optionen.
EAWAG News, 42D, 28-29.
Zhang, Y. and Prepas, E.E. 1996 Short-term effects of Ca(OH)2 additions on

phytoplankton biomass: a comparison of laboratory and in situ experiments. Wat. Res.,
30, 1285-1294.
Zehnder, A. 1996 Blick über die Grenzen. EAWAG News, 42D, 24-26.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 9. REMEDIAL MEASURES
This chapter was prepared by Steve Hrudey, Mike Burch, Mary Drikas and Ross
Gregory
The management and control of cyanobacteria in water supply storage facilities and of
cyanotoxins in drinking water may be approached at a number of points and levels in the
hierarchy of the total supply system. A detailed assessment of water supply systems
with respect to the potential impact of blooms and cyanotoxins on water quality and
public health has been presented in Chapter 6. The first preference for control is the
prevention of eutrophication, which is discussed in Chapter 8. The next level of
management response is reservoir and water body management which can include
some engineering techniques to alter hydrophysical conditions in the water body in order
to reduce cyanobacterial growth (section 8.5). The more immediate and short-term
control techniques which can be used in the management of raw water abstraction
include the avoidance of contamination by positioning of offtakes, selection of intake
depth, offtake by bank filtration, and the use of barriers to restrict scum movement.
Another intervention technique is chemical treatment with algicides. Algaecides have
been, and will continue to be, used as emergency measures for the control of
cyanobacteria, and their role in management strategies needs to be assessed from
practical and environmental viewpoints. The final option for management of
cyanobacterial problems and cyanotoxins in water supplies is within the treatment
system. Research on removal of algal and cyanobacterial cells has been widely
published (see review by Mouchet and Bonnélye, 1998) and recent work has generated
quite detailed knowledge on cyanotoxin removal during drinking water treatment.
Much of the work on cyanotoxin removal has focused on single treatment steps, and a
few studies have investigated the common combinations of coagulation, clarification and
filtration. As a research approach, this is useful because assessments of the
performance of individual treatment steps may be generalised more readily than
observations gained in complete, but individual, supply systems with their respective
site-specific characteristics. However, management approaches aimed at providing safe
drinking water from cyanobacteria-infested surface waters require considering the
system as a whole, and using different combinations of resource management tailored to
the specific locality and different treatment steps. In such an overall approach, steps that
individually may be unsatisfactory can have their place in combination with others and
thus contribute to a multi-barrier approach. Furthermore, the individual aspects of each
drinking water supply necessitate local assessment of performance as well as local
optimisation of resource management and treatment strategies.
9.1 Management of abstraction
Management of raw water abstraction is effective in reducing the amount of
cyanobacteria in the raw water, often by orders of magnitude. This can be achieved by
choosing an optimum position for the offtake, or by abstracting surface water through
bank filtration.
9.1.1 Direct abstraction from surface water bodies
General resource management, including hydraulic intervention measures, are

described in Chapter 8. The horizontal and vertical distribution of cyanobacterial
populations can vary enormously throughout a water body, whether in a lake, reservoir
or river. This has obvious implications for both the siting of offtakes and the choice of
offtake depth. Considerable contamination of raw water can be avoided by locating
offtakes away from sheltered bays where scums may accumulate (usually downwind of
the prevailing winds during the critical summer growth period). If this is not practical, it
may be possible to employ temporary extensions to pipe intake points.
Selection of offtake depth can also be important in reducing contamination by avoiding

surface or subsurface maxima of cell numbers. Many modern reservoir offtake structures
(towers) have the provision for multiple offtake depths. If multiple offtakes are not
available it may be possible to install siphon offtakes, at least as a temporary measure in
small systems. In relation to cyanobacterial contamination, the choice of intake depth
must take into account the time of daily maximum surface accumulation of cells and the
amplitude of passive diurnal vertical sinking and rising of cells due to light-and
photosynthetic-driven changes in cell buoyancy. In thermally stratified, mesotrophic
reservoirs, attention must be given to the possibility of meta-limnetic maxima (i.e.
maxima between warm upper and cold, lower water layers) of Planktothrix rubescens
(syn. Oscillatoria rubescens).
Operators need to become familiar with the amplitude of vertical movement of

cyanobacterial populations, and also with the potential for the formation of metalimnetic
peaks, in order to avoid high cell densities as much as possible. This requires multiple
depth sampling to determine vertical profiles of cyanobacterial cell density. Collecting
information and building up knowledge and understanding of local ecology and
conditions can increase flexibility in the management of blooms (see Chapter 10).
Another option to avoid contamination is to employ physical barriers or booms at the

surface to prevent surface scums accumulating near the offtake site. Surface booms or
curtains, similar to oil-spill containment booms, have been used successfully in Australia,
the UK and North America to keep surface scums away from offtake structures (see
Figure 9.1 in the colour plate section). These physical barriers often only extend to a
depth of 0.5-1.0 m, and do not affect bulk horizontal flow significantly. This technique is a
worthwhile emergency measure for transient blooms and its use will depend upon the
practical aspects of installation.
9.1.2 Bank filtration and groundwater recharge
An abstraction method that has proved to be very effective in removing particles and
many dissolved compounds in localities with suitable underground conditions is bank
filtration or abstraction of groundwater artificially recharged with surface water. The
process uses bore holes or infiltration galleries which are located near to the banks of a
surface water supply (river or reservoir). These wells fill with water which has infiltrated
through the intervening porous soil materials. Depending upon the underground
characteristics, water may travel for several hours, or even for weeks before it reaches
the well. Longer retention times may enhance purification, but even retention times of
between a few hours and days have substantially improved water quality. Many types of
soils may be suitable, provided they allow water flow, are not too coarsely structured to
achieve a filtering effect, or are not in contact with saline or otherwise unsuitable
groundwater. Planning bank filtration requires local assessment of sites for their
suitability.
Evaluation of bank filtration with respect to cyanotoxin removal is currently only

beginning, and no published results are available. A study of elimination of algal and
cyanobacterial taste and odour compounds has shown very effective removal by bank
filtration at three study sites (Chorus et al., 1993). Because of the generally positive
experience with respect to removal of suspended materials, micro-organisms and a
variety of chemical contaminants (Laszlo, 1984; UNDP/WHO, 1992) it may be expected
that bank filtration will be a highly promising abstraction method to avoid contamination
with cyanobacterial cells as well as dissolved toxins. This expectation is supported by
the favourable results of a laboratory study from Finland which demonstrated good
performance of experimental soil and sediment columns for both cell and toxin removal
(Lahti et al., 1996). In this case, lake water was inoculated with both toxic and non-toxic
cultures of cyanobacterial cells and pure microcystin-LR and filtered through soil and
lake sediment columns. It was found that during the experimental period of one week,
both cells and dissolved toxins were removed very efficiently, although there was some
breakthrough in sediment columns at high loadings. The mean rates of removal for cells
were 93.7-99.7 per cent and 97.5-99.5 per cent for extracellular toxins for both soil and
sediment columns. It was suggested that the removal of microcystins in this filtration
process was the result of both adsorption and biodegradation (Lahti et al., 1996).
However, the relative performance of the two processes would be very site specific and
dependent upon local soil characteristics and microbial activity.
The performance of bank filtration in relation to adsorption capacity, overloading and the
potential for release over time of toxins from trapped cells would require monitoring (see
Chapter 13 for methods).
9.2 Use of algicides

Algaecides are used in reservoirs to control cyanobacterial growth and to prevent or
reduce to some extent the problems of toxins in the associated drinking water supply.
Their role in the management scheme may be to provide effective short-term control of
growth of cyanobacteria, at one point in time, particularly in circumstances where
alternative drinking water sources are not available and preventive measures (as
outlined in Chapter 8) are not feasible or not yet effective. Algicide treatment has been
proposed as being more cost-effective than toxin removal in drinking water treatment
plants, as has been suggested for the control of off-flavour problems (McGuire and
Gaston, 1988), because an extended period of persistent blooms greatly enhances the
need for additional treatment for toxin removal. However, experience with abatement of
off-flavours caused by cyanobacteria through algicide treatment has also demonstrated
that this treatment may actually enhance the problem by supporting the development of
species resistant to the treatment (Izaguirre, 1992).
Environmental concerns have been raised because the most commonly used algicide,
copper sulphate, has broad ecological impact. It should be used only in dedicated water
supply reservoirs in special circumstances, but is nevertheless an unsatisfactory long-
term solution. In many countries, national or local environmental regulations prohibit or
limit the use of algicides due to their adverse environmental impact. This needs to be
established prior to considering the use of algicides.
Algicides, like all management techniques, must be applied correctly to work effectively.
If algicides are used they must be applied at the early stages of bloom development
when cell densities are low, in order to reduce the potential for liberation of the high
concentrations of intracellular toxin that may be associated with dense blooms. Early
application will further enhance the effectiveness of treatment because cyanobacterial
cells can form a major part of the "copper demand" along with other organic matter in
natural water.
A major limitation of any agent which disrupts cyanobacterial cells is the release of
toxins and of taste and odour compounds from the cells. A range of studies have
indicated that cyanotoxins are predominantly intracellular in healthy cells, and are only
released into the water at an advanced stage of bloom senescence, or following
treatment with chemicals such as algicides (Lahti et al., 1996). This release can be quite
rapid and has been shown to occur within 3-24 hours in different studies (Jones and Orr,
1994; Kenefick et al., 1993). These dissolved toxins will then disperse and be diluted
throughout the water body, but will not be removed by conventional flocculation and
filtration procedures. Installation of additional treatment for removing cyanotoxins may be
costly. The dangers of treating dense blooms with algicides was demonstrated in an
incident which occurred on tropical Palm Island, Australia, where members of the
community became ill with hepato-enteritis following treatment of the water supply
reservoir with copper sulphate for a cyanobacterial bloom problem (Bourke et al., 1983)
(see Box 4.3).
If algicides are used to control toxic cyanobacteria, the reservoir should be isolated for a
period to allow the toxins and odours to degrade (see section 3.4). Unfortunately, very
little data exist on the withholding period in relation to toxin loss, but it could be in excess
of 14 days (Jones and Orr, 1994).
In some cases algicide treatment may be unsuccessful or only partially successful. This
can be due to inadequate dispersal and contact with the target organisms, variable
sensitivity of cyanobacteria, and reduced toxicity due to complexation of the copper
(Burch et al., 1998). The form of copper most toxic to aquatic organisms is the free
cupric ion (Cu2+) and this can be reduced by complexation with both inorganic ligands
under alkaline conditions, and organic ligands present in natural waters (McKnight et al.,
1983).
9.2.1 Copper sulphate
Chemical control of algae in water supply storage has been a widespread water quality
management practice for over 100 years. Records of the use of copper sulphate date
from 1890 in Europe (Sawyer, 1962), from 1904 in the USA (Moore and Kellerman,
1905), and at least since the mid 1940s in Australia (Burch et al., 1998). Copper
sulphate has been regarded as the algicide of choice because it is economical, effective,
relatively safe and easy to apply. It is also considered to be of limited significance to
human health at the doses commonly used (WHO, 1996) and has been considered not
to cause extensive environmental damage (McKnight et al., 1983; Elder and Home,
1978). The latter point has been an issue of debate for some time (see Mackenthun and
Cooley, 1952) because copper tends to accumulate in lake sediments (Sanchez and Lee,
1978; Hanson and Stefan, 1984). In some cases it appears not to be remobilised and is
bound permanently to the bottom sediments (Elder and Home, 1978; Sanchez and Lee,
1978). However, in a study of 10 drinking water dugouts (small reservoirs) in Canada,
sediment copper (previously accumulated from copper sulphate treatments) was
released back into the open water under low dissolved oxygen conditions in the
hypolimnion in summer (Prepas and Murphy, 1988). It has also been suggested that
sediment-bound copper could have an impact on the benthic macroinvertebrate
community (Hanson and Stefan, 1984). It is important to remember that copper and
other heavy metals differ from some other toxic contaminants in that they are not
biodegradable, and once they have entered the environment their potential toxicity is
controlled largely by their speciation or physicochemical form (Florence, 1982). Copper
sulphate treatment has been shown to cause short-term changes in phytoplankton
abundance and species succession (Effler et al., 1980; McKnight, 1981). Fish kills may
also occur following copper sulphate treatment, although it is not clear whether this is as
a result of copper toxicity or oxygen depletion (Hanson and Stefan, 1984).
A recent extensive survey of water utilities in the USA and Canada indicated that copper
sulphate is by far the most widely used algicide, although other alternatives are used
under some circumstances (Casitas Municipal Water District, 1987). Some of the
compounds that have been used and evaluated for potential as algicides over the years
are summarised in Table 9.1.
McKnight et al. (1983) give an assessment of the use of copper sulphate for the control
of nuisance algae and cyanobacteria. They also indicate that there are wide differences
in copper sensitivity among species. The relative growth inhibiting concentrations for a
range of phytoplankton are given in terms of cupric ion activity (i.e. [Cu2+]), derived from
laboratory toxicity studies. The toxic cupric ion activities range from greater than 10-6-10-11
M (0.063-6.3 × 10-7 mg l-1 Cu2+) for species of diatoms, dinoflagellates, green algae and
cyanobacteria - a difference of over four orders or magnitude (McKnight et al., 1983).
These toxic Cu2+ concentrations are very much less than the usual doses applied as total
copper in copper sulphate treatments. The relative toxicity is given in terms of ionic
copper because it is believed that phytoplankton react principally to the concentration of
Cu2+ or loosely complexed copper rather than the total dissolved metal in the water.
McKnight et al. (1983) have used these findings to develop an experimental procedure
to determine the required dose rates for target species in individual reservoirs, taking
account of the particular water chemistry. This experimental procedure to determine
dose rates is suggested as more effective than simple empirical formulae, based on pH
and alkalinity, which were not very useful (McKnight et al., 1983). The experimental
approach requires access by the water supply operators to a good level of biological and
chemical expertise and analytical capacity. The approach is based on first determining
the cupric ion activity as a function of added copper, and thereby the complexing
capacity of the reservoir water by a copper ion selective electrode. This is followed by a
culture assay to determine the sensitivity of the particular nuisance algae to copper. The
local nuisance species preferably need to have been isolated into laboratory culture. The
required copper sulphate dose rates can be derived from a simple formula relating
growth inhibition, in terms of cupric ion concentration, to the Cu2+ concentrations in the
reservoir after complexation (McKnight et al., 1983).
Table 9.1 Compounds that have been used as algicides, their formulation and key
references
Compound Formulation Reference(s)

Copper sulphate CuSO4.5H2O McKnight et al., 1983;
Holden, 1970; Palmer, 1962;
Casitas Municipal Water
District, 1987
Cutrine® - plus Cu alkanolamine. 3H2O++1 Humburg et al., 1989
Copper - triethanolamine complex Cu N(CH2CH2OH)3.H2O Humburg et al., 1989
Copper citrate Cu3 [(COOCH2)2C(OH)COO]2 Casitas Municipal Water
District, 1987; Raman, 1988;
McKnight et al., 1983;
Fitzgerald and Faust, 1963
Potassium permanganate KMnO4 Fitzgerald, 1966; Holden, 1970
Chlorine Cl2 Holden, 1970
1
Copper II alkanolamine complex
9.2.2 Copper chelates
The problem of the reduced effectiveness of copper sulphate treatment in hard alkaline
water has long been recognised (Palmer, 1962; Casitas Municipal Water District, 1987).
Chelated copper algicides were developed to overcome the problems of the
complexation and precipitation loss of toxic copper under these circumstances.
Examples of copper chelate algicides include copper ethanolamine complexes and
copper citrate (Table 9.1). Copper citrate has been used as an algicide in the USA
(Casitas Municipal Water District, 1987; Raman, 1988). It is available either as a
commercial preparation (Hoffman et al., 1982) or by simultaneously dosing copper
sulphate and citric acid (Raman, 1988). It is claimed that the use of citric acid as a
chelating agent enhances the solubility of copper allowing it to remain in solution longer
under alkaline conditions (Raman, 1985; 1988). Raman (1988) recommends applying
copper sulphate: citric acid in the weight ratio 2:1 in high-alkalinity waters (> 40 mg l-1
CaCO3). A study which examined equilibrium speciation of copper in water to determine
the changes in distribution of Cu(II) in relation to pH, dissolved organic carbon (DOC)
and citrate was carried out by Casitas Municipal Water District (1987). This study
demonstrated that citrate greatly enhances the solubility of copper even in the presence
of appreciable alkalinity (100 mg 1-1). McKnight et al. (1983) suggested that the
advantage of using synthetic copper chelating agents in hard, alkaline waters probably
results from decreasing the supersaturation of malachite (Cu(OH)2CO3) and tenorite
(CuO) and thereby the rate at which equilibrium with these insoluble forms (precipitates)
is approached. It is possible that a longer time taken to reach equilibrium would result in
the maintenance of toxic ionic Cu2+ activities and the inhibition of algal growth for longer
periods after dosing (McKnight et al., 1983). It is acknowledged that, despite their
relatively widespread use in the USA, the efficacy of chelated copper algicides in relation
to water chemistry is poorly understood (Casitas Municipal Water District, 1987).
9.2.3 Use of oxidants
Potassium permanganate has been used as an algicide from as early as 1935 (Holden,
1970). A survey of North American utilities indicated that a small number use potassium
permanganate relative to those who use copper sulphate (Casitas Municipal Water
District, 1987). Commercial formulations of potassium permanganate marketed
specifically as algicides are available in the USA (Casitas Municipal Water District, 1987).
Fitzgerald (1966) investigated the relative toxicity of potassium permanganate to eight
species of algae and cyanobacteria and found the algicidal dose was in the range 1-5
mg l-1, except for one green algae where up to 8 mg l-1 was required.
Chlorine is used mainly for control of algae in water treatment works but is also known to
have been employed in reservoir situations (Holden, 1970). The effective dose rates are
dependent on the chlorine demand of the water, but most algae are reported to be
controlled by residues of free chlorine between 0.25 and 2.0 mg 1-1 (Holden, 1970).
9.2.4 When to use algicides
Because cyanobacterial toxins are primarily intracellular, algicides must be used with
particular caution to avoid release of intracellular toxins. Algaecides should be used
when cell numbers are low to avoid excessive toxins or taints following rupture of the
cells. This should be checked by post-dosing monitoring. Algaecides may be used at
higher cell numbers only if the reservoir can be taken out of supply until the toxins and
taints degrade, or if treatment for removal of the toxins and taints is available. In the
latter cases the use of algicide should be assessed against the capability for whole cell
removal offered by treatment processes, because cell removal may be safer. It is
important to know how effective the chosen algicide is in the specific waters. For
example, copper may be less effective in waters with high dissolved carbonate or at
alkaline pH.
Table 9.2 Distribution of microcystins during laboratory culture of Microcystis aeruginosa
Age of culture Distribution of toxins (%)

Cells Water
Young
Slowly-growing cells 100 0
Rapidly-growing cells 75-90 10-25
Old
Slowly-growing intact cells 70-80 20-30
Decaying cells (leaking cell contents) 30-40 60-70
Source: National Rivers Authority, 1990
Algaecides should only be used in waters where the environmental impacts are
acceptable and this should be checked with the local environmental agency.
9.3 Efficiency of drinking water treatment in cyanotoxin removal
Cyanobacterial toxins represent a challenge to drinking water treatment which involves
removal of organic substances in both soluble and insoluble form. Water treatment
processes may remove target substances by either separation or conversion. Separation
processes are those which remove the target substance from the treated water, usually
to a treatment residual which becomes a waste stream for disposal. Conversion
processes involve transforming the target substance into a different chemical form,
thereby reducing the water quality problem. Although conversion processes are
sometime characterised as though they achieve destruction, there will always be
reaction products and thus transformation is a more accurate description than
destruction. Ideal conversion processes are those which yield innocuous reaction
products.
A major factor in assessing water treatment for cyanobacterial toxin removal involves
consideration of soluble and suspended substance removal. The primary toxins which
have been studied, microcystins, nodularins and anatoxins, are all water soluble.
However, laboratory observations for microcystins have shown that these toxins are
produced within the cyanobacterial cells and are expected to be predominantly found
within slow growing, healthy cells (Table 9.2).
Recent work (Mole et al., 1997) has shown that microcystin release from cultured
Microcystis aeruginosa began to occur late in the exponential growth phase and
increased significantly during the stationary phase. This release was linked to a
decrease in the integrity of the cells as determined by staining with fluorescein diacetate.
The amount of toxin release was influenced by the culture medium and reached as
much as 50 per cent in most commonly used media late in the stationary phase of
population growth.
Until a bloom collapses or is otherwise affected by some treatment practice, the majority
of toxins will be retained within the cells, making removal of intact cells a high treatment
priority. However, under bloom conditions, a substantial proportion of toxin would also be
expected to be released to the water column, making removal of soluble toxin an
unavoidable concern.
Physicochemical treatment has been shown to cause cell lysis and toxin release (James
and Fawell, 1991). Operational investigations in Africa demonstrated significant cell lysis
during extended transport in pipelines (Dickens and Graham, 1995). However, other
experiments conducted with cultured Microcystis showed that the flow and mixing
conditions associated with water treatment did not cause cell lysis or toxin release. In
addition, changes in pH from 5 to 9, which can occur in the treatment of some waters,
did not cause any release of the intracellular toxins (WRc, 1996). Effects of physical and
chemical stress on toxin release from cyanobacterial cells should therefore be assessed
in treatment and conveyance systems.
The following sections review the capacity of established and novel treatment processes
for the removal of cells and dissolved toxins.
9.3.1 Screening and prefiltration
Water treatment facilities usually employ coarse screens to remove debris from the
water intake. These screens have no effect on the removal of either cyanobacterial cells
or soluble toxins. However, microstrainers or fine screens may be used to remove larger
algae, cyanobacterial cells and aggregated cells. Mouchet and Bonnélye (1998) reported
removal rates of 40-70 per cent for two cyanobacterial species but pointed out that
smaller species (e.g. single cells and small colonies of Microcystis) are poorly retained
(to sometimes less than 10 per cent). Concerns regarding possible cell rupture, lysis and
toxin release resulting from pressure on the filter screen have not been sufficiently
addressed.
9.3.2 Aeration and air stripping
There are a number of methods for contacting air with water in drinking water treatment
that may be required for various purposes, such as to oxidise iron and manganese from
soluble to insoluble forms, to prevent reducing conditions which may yield odorous
compounds, and to remove dissolved gases such as carbon dioxide, hydrogen sulphide,
other reduced sulphur compounds and other volatile organic compounds (Hamann et al.,
1990).
Neither aeration nor air stripping will be effective for removing soluble toxins because
they are non-volatile compounds. Nor would they be effective for removal of
cyanobacterial cells (for aeration techniques applied in reservoirs to reduce growth of
cyanobacteria see section 8.5.5).
9.3.3 Coagulation and clarification
Coagulation promotes the aggregation of small, dispersed particles into larger particles
which can be separated by sedimentation, filtration or flotation (Grohman et al., 1985;
Hamann et al., 1990). Coagulation differs from precipitation because the latter involves
converting soluble substances into insoluble particles, whereas coagulation deals with
pre-existing dispersed particles such as mineral turbidity (clay, silt), larger molecular
weight natural organic matter, micro-organisms including cyanobacteria, and oxidised,
insoluble forms of iron and manganese.
Common chemicals used for drinking water coagulation include various aluminium and
ferric iron salts. More recently synthetic organic polymers have gained some acceptance.
Coagulation with multivalent metal salts can also be aided by adding various organic
polymers to promote floe growth. Leuschner (1984) reported substantially improved
flocculation of Planktothrix agardhii after addition of a cationic polymer. Efficient removal
of algae is dependent on optimisation of chemical doses and coagulation pH. Mouchet
and Bonnélye (1998) have shown that the coagulant dose necessary for algal removal is
proportional to the sum of alkalinity and the logarithm of cell number. They emphasise
that minimising turbidity in ajar test is not a sufficient criterion for adjusting treatment to
remove algae and cyanobacteria, and recommend measuring the electrophoretic
mobility of the cells (zeta potential) for optimising dosage (particularly because at
insufficient coagulant dose, cyanobacteria will be the last phytoplankton cells to be
removed). Bernhardt and Clasen (1991) have reported that coagulation of algal cells that
are smooth and more or less spherical occurs largely by charge neutralisation. In
contrast, filamentous algae, large algae or species with bristles on their cell surface can
be dealt with effectively only by sweep coagulation, by encountering the algae with large
amounts of metal hydroxide floe.
Coagulation, by its nature, offers some promise for removal of intact cyanobacterial cells.
For neurotoxins, Falconer (1989) reported that alum dosed at 120 mg l-1 alone and in
combination with a number of polyelectrolytes removed about 20 per cent of the toxicity
from a neurotoxic bloom of Anabaena circinalis. For microcystins, a number of published
studies have shown that coagulation has a negligible capability for removal of any
soluble toxins present in water. This has been demonstrated with aluminium sulphate
coagulation jar tests in which total toxin concentration was reduced as a result of the
removal of algal cells rather than the extracellular toxin (Figure 9.2) (WRc, 1996).
Rositano and Nicholson (1994) also demonstrated this expectation by evaluating
removal of purified, soluble microcystins by three coagulants: ferric sulphate, alum and
polyaluminium chloride. In all cases they found essentially no toxin removal. Lambert et
al. (1996) found inconsistent and low levels of microcystin removal (0-39 per cent)
across the coagulation-sedimentation stage of a small, full-scale water treatment plant
using an alum dose of over 60 mg l-1.
Figure 9.2 The effect of coagulation with alum on the concentration of intra- and
extracellular microcystin-LR (After Hart et al., 1997. Reproduced courtesy of
Blackwell Science)
By contrast, it must be emphasised that a study on raw water treatment with high doses
of alum (200 mg l-1) found over 23 per cent of the cell-bound microcystin-LR was
released, mostly within two days of treatment (Lam et al., 1995). However, at
concentrations and conditions that would occur in water treatment plants, Velzeboer et al.
(1995) found that aluminium sulphate did not appear to cause lysis of cells of cultured
Anabaena circinalis or Microcystis aeruginosa. Flocculation under laboratory conditions,
which simulated operating water treatment plants, resulted in removal of cells in a
healthy state, with no additional release of geosmin or microcystin-LR. Further work by
Chow et al. (1997a) using ferric chloride as the coagulant showed similar results with
some stimulation of growth of both algal species. There was no increase in
concentration of microcystin in the water following treatment of Microcystis aeruginosa,
although it appeared that Anabaena circinalis may be more susceptible to damage from
chemicals. Later work using alum in a pilot plant with cultured Microcystis aeruginosa
harvested at the late exponential phase of growth confirmed that the cells were not
damaged through the treatment process and that no additional toxin was released during
treatment (Drikas et al., 1997). However, this study also confirmed that the low
concentrations of extracellular microcystin present in the feed water (2-6 µg l-1) are not
removed during the treatment process. It was further found that the total cell number in
sludge collected from the pilot plant decreased to half its initial value after two days, and
that toxin release began virtually immediately, reaching almost 100 per cent after two
days. After five days the toxin concentration began to decrease and was reduced by
approximately 80 per cent after eight days and completely removed after 13 days. This
corresponds to findings of Jones and Orr (1994) who observed that bacterial
degradation of microcystin-LR occurred after nine days in a lake after chemical
treatment of a Microcystis aeruginosa bloom. The importance of toxin release from
sludge depends on the time that sludge is retained in sedimentation tanks and it could
have implications for sludge management, particularly if supernatant is returned from
sludge treatment processes to the head of the plant.
Selection of clarifier type will also affect cell removal rates. Mouchet and Bonnélye (1998)
have summarised experience largely from warm climates and have shown that sludge
blanket-type clarifiers are substantially more effective than static settlers (largely
because of longer flocculation time), particularly if upflow pulsed systems are used. This
achieved consistent reduction of total phytoplankton by 95-99 per cent at a plant treating
Seine River water, 95-98 per cent elimination of cyanobacteria at a Philippine plant (as
compared with 90-95 per cent removal by static settling), and 96.7-99.5 per cent removal
of Anabaena and Microcystis at an industrial-scale plant in Harare (Zimbabwe). In Cairo
and Alexandria, Egypt, older settling tanks were successfully upgraded to upflow pulsed
sludge blanket clarifiers, thus not only improving performance for algal and
cyanobacterial removal, but also efficiency per unit area and a reduction in coagulant
consumption by 15-45 per cent and chlorine consumption by 15-35 per cent.
9.3.4 Dissolved air flotation
Although coagulation is normally followed by a sedimentation step, in some waters

where the content of the suspended matter is low it is often easier to float the floe rather
than attempting to settle a light floe. Recycled water saturated with air under pressure is
introduced following the flocculation stage. Following the release of pressure the air
comes out of solution and forms tiny bubbles which attach to the floe and cause it to float
to the surface. The floated sludge is then collected and removed. This process is called
dissolved air flotation (DAF) and is more effective than sedimentation, particularly for
water with low turbidity and high colour, because the resultant floe is lighter and floats
easily.
Dissolved air flotation is also generally more effective than sedimentation processes for
treating algal-rich waters; for example floe blanket clarification has been shown to
remove 76.5 per cent of Microcystis cells whilst DAF removed 98 per cent in the
presence of other algae (Gregory and Zabel, 1990). A Belgian DAF plant achieved 40-80
per cent removal of Microcystis, 90-100 per cent removal of Anabaena but only 30 per
cent removal of Planktothrix (syn Oscillatoria) (Steffensen and Nicholson, 1994).
Markham et al., 1997) have reported on the efficiency of algae removal at eight DAF
plants. Like Bernhardt and Clasen (1991), they observed that the characteristics of algae
influence their removal by any clarification process. They found that most of the
treatment plants produced more than 80 per cent removal and they expected this would
be improved by optimisation. Vlaski et al. (1997) found that, in a pilot plant, DAF
achieved high particle (algae) removal during a cyanobacteria bloom (mainly Microcystis
aeruginosa).
Dissolved air flotation is unlikely to be more effective than conventional sedimentation

processes for removing extracellular toxins. It may, however, remove more intact cells
because the floating sludge tends to be removed more frequently than settled sludge in
horizontal flow tanks, where the algae may die and then lyse. This assumption needs to
be evaluated further.
Periods of high turbidity often cause problems for DAF, and any interruption in the
process leads to an interruption in the treatment process. Thus a stock of spare parts
and regular maintenance by qualified personnel are critical issues when using this
approach (Mouchet and Bonnélye, 1998).
9.3.5 Precipitation for hardness reduction
Conversion of soluble compounds into insoluble particulates for separation by

sedimentation or filtration is commonly used for water softening (calcium and
magnesium removal) and for iron and manganese removal (Hamann et al., 1990). Some
concurrent removal of soluble metals and dissolved natural organic matter may also be
achieved. Lime is commonly used for adjusting hardness or for precipitation of soluble
metals. This process typically uses rapid mixing followed by flocculation and
sedimentation.
No studies evaluating lime precipitation as a separate process in a water treatment plant

sequence are available. However, some insight into the expected removal of intracellular
toxins has been provided by two studies looking at treatment of raw water blooms with
lime. Kenefick et al. (1993) found that lime doses from 100 mg l-1 as Ca(OH)2
precipitated the cells in cyanobacterial bloom material containing microcystin-LR without
releasing toxin compared with control batches over 14 days, while Lam et al. (1995)
found only 4 per cent release of microcystin-LR for the same lime dosage. These studies
suggest that lime softening would be effective at removing intracellular toxin by removing
the cyanobacterial cells without causing cell lysis, but that there is no evidence to
suggest that lime softening can reduce extracellular toxins.
9.3.6 Direct rapid filtration
Filtration is a process for the removal of suspended particulate matter, typically including
clay, silt, natural organic matter, coagulated floes, lime softening precipitates, iron and
manganese precipitates, and microorganisms (Hamann et al., 1990). Filters most
commonly use granular media such as coarse sand, crushed anthracite coal, garnet and
granular activated carbon (GAC). Direct filtration is applied for low turbidity waters by
filtering directly after coagulation/destabilisation without an intervening clarification stage
to remove the bulk of the floe. Conventional water treatment uses rapid filtration rates
which require regular backwashing to maintain performance.
Mouchet and Bonnélye (1998) reported poor removal rates of 10-75 per cent, depending
upon phytoplankton species, by direct rapid filtration without prior chemical treatment.
Drikas et al. (1997) found that removal of Microcystis aeruginosa cells in the filtration
stage of a pilot plant varied between 14 and 30 per cent following alum
coagulation/sedimentation. Lepisto et al. (1996) evaluated full scale water treatment
plants for their ability to remove cyanobacterial cells and found rapid sand filtration
achieved only a 14 per cent reduction in cells. Rapid sand filtration, including GAC was
somewhat better achieving 42 per cent reduction of cyanobacterial cells. These
researchers expressed concern over the possible fate of intracellular toxins which may
be released from degrading cells trapped in the filtration stage. Lambert et al. (1996)
found inconsistent incremental removal of microcystins from 14-60 per cent across a
dual media sand-anthracite filtration stage, following an alum coagulation-sedimentation
stage, at a small, full-scale water treatment plant.
As an overall assessment of direct rapid filtration for elimination of algae and

cyanobacteria, Mouchet and Bonnélye (1998) have indicated that direct filtration is
generally not satisfactory, unless more sophisticated multimedia filters and adequate
initial treatment are applied. They particularly emphasised the excellent results in algal
removal after pre-ozonation (explicitly with the aim of enhancing cell removal through
further steps, rather than for oxidation of cyanotoxins, see section 9.4.1).
A potential issue of concern, which currently has been inadequately investigated, is the
effect of long filter runs between backwashing. Death and lysis of cyanobacteria retained
on filters could lead to substantial toxin release.
9.3.7 Combined coagulation, sedimentation and rapid filtration
Conventional water treatment commonly involves the combination of coagulation,

clarification (sedimentation or dissolved air flotation) and filtration. Consequently, much
of the limited research that has been published on water treatment performance for the
removal of cyanotoxins has looked at overall removal across the common combinations
of coagulation-filtration and coagulation-clarification-filtration, rather than looking at each
stage individually.
Himberg et al. (1989) evaluated hepatotoxic fractions from Microcystis wesenbergii, M.

viridis and Planktothrix agardhii (syn. Oscillatoria agardhii) in bench-scale treatment
processes consisting of alum or ferric chloride coagulation combined with sand filtration
and chlorination. Alum coagulation, at doses from 36 to 71 mg l-1, with filtration achieved
toxin removals from 11 to 32 per cent, while ferric chloride at 55 mg l-1 achieved from 9 to
16 per cent. The removal contribution of the low chlorination dosage in this case was
apparently negligible. They also studied a similar conventional process at pilot scale
using freeze dried Microcystis bloom material and found negligible toxin removal (Keijola
et al., 1988). Nonetheless, Lambert et al. (1996) found combined microcystin removal
was 50-60 per cent across coagulation, sedimentation and dual media filtration in a
small full scale plant.
Similar studies at bench scale with anatoxin-a have indicated no removal for either alum
or ferric chloride process combinations at a toxin concentration of 20 µg l-1, but at 10
times higher toxin concentrations, the alum process achieved a 14 per cent anatoxin-a
removal and the ferric chloride process achieved a 49 per cent anatoxin-a removal
(Keijola et al., 1988).
Leuschner (1984) studied phytoplankton retention by flocculation, sedimentation and

rapid filtration in a plant treating highly eutrophic river water. Whereas Microcystis spp.
(occurring as large colonies) were rarely observed in the finished water, Planktothrix
agardhii was poorly retained, showing an average breakthrough of 27 per cent of the
filaments. As also reported by Mouchet and Bonnélye (1998), addition of a cationic
polymer during flocculation substantially improved retention.
The removal of whole, intact cells presents the best opportunity to remove toxins in
separation processes, whereas the literature indicates removal efficiencies are low with
extracellular toxins. Some unsatisfactory results reported with lysis of entire cells may
have been due to an excessive time delay between flocculation and analysis. In
summary, currently available results indicate that conventional coagulation and rapid
filtration processes assist in toxin removal, particularly if cyanobacterial cells are kept
intact, but cannot be generally relied upon as the main removal process. Mouchet and
Bonnélye (1998) have emphasised the need for:
"... pilot scale investigation in order to estimate the technical and economical advantages
of this choice in each case. Generally, a conventional treatment line, including
coagulation, flocculation, settling or flotation, and filtration, is preferred to treat algae-rich
waters. However, algae removal is somewhat more delicate than turbidity removal and,
consequently, greater attention is required when selecting technology and adjusting the
chemical treatment." (Mouchet and Bonnélye, 1998)
9.3.8 Slow sand filtration
In contrast to rapid filtration, slow sand filters operate at lower rates and develop a
surface filter cake which performs most of the filtration together with (often high)
biological treatment activity. These biofilms establish after some time of operation and
contribute significantly to degradation of dissolved substances. Mouchet and Bonnélye
(1998) reported a likely removal of 99 per cent of algal cells by slow sand filtration.
Operation of these filters in the dark can prevent intensive algal growth on the filter.
However, overloading of filters with algae or cyanobacteria from the raw water may lead
to rapid blocking, requiring removal of the bioactive surface layer, thus temporarily
reducing the efficiency for retention of dissolved substances. For removal of toxic
cyanobacteria, this constitutes a dilemma because bloom-containing waters are likely to
lead to rapid blocking and thus undermine the practicability of slow sand filtration.
However, experiments have shown that before blocking, slow sand filters may be quite
effective in the removal of toxic cyanobacteria and dissolved toxins.
Keijola et al. (1988) evaluated laboratory-scale slow sand filters and reported over 80
per cent removal of toxins from Microcystis, 30-65 per cent removal of toxins from
Planktothrix (syn. Oscillatoria) and about 70 per cent removal of anatoxin-a. Because
filtration itself would not be expected to achieve any removal of extracellular toxin, these
results suggest that the mechanisms were at least biosorption, and perhaps some
biotransformation. Australian studies (Sherman et al., 1995) with roughing filters followed
by slow sand filters showed that M. aeruginosa and some Planktothrix (syn. Oscillatoria)
cells from toxic bloom material could be removed by physical means and biological
processes. Superior microcystin removal, in one of two river water sources being treated
with GAC filters, was attributed to biological activity (Drikas, 1994). Freeze dried bloom
material was used in this study.
Work on microcystin-LR degradation using an isolated bacterium for use in water

treatment has been undertaken by Bourne et al. (1996). A pseudo-monad has been
isolated which possesses an enzyme system capable of degrading microcystin, but the
work is currently only at the laboratory scale. Pilot plant studies using a solid phase
support for this bacterium to investigate this process are to be undertaken.
Developments have occurred in the exploitation of slow sand filters at large treatment
works, notably in the UK and Netherlands. These have included use of various
pretreatments, such as conventional treatment by coagulation and filtration and pre-
ozonation to control the rate of blocking by algae and cyanobacteria. Whilst these
processes will assist with removal of cells containing toxins, they have not been
adequately assessed for their reliability in degradation of extracellular toxins. A notable
development has been the sandwiching of a layer of GAC within the bed of sand in slow
sand filters in order to assist in removal of dissolved toxins.
General experience with slow sand filters suggests that they are potentially very useful
for removal of particles and dissolved substances, particularly if further developed or
combined with other treatment steps to avoid blocking when loaded with waters rich in
algae and cyanobacteria (or other particles). New approaches to slow sand filtration are
experimenting with horizontal rather than vertical water flow (as used in cross-flow
membrane techniques). This requires larger amounts of water but will remove most of
the potentially filter-blocking particles and, in particular, would keep cyanobacteria
suspended. Such systems may be developed locally, particularly to serve small
communities. For large treatment facilities, bulk cell removal by coagulation and
clarification before slow sand filtration may be an effective approach for obtaining the
benefits while avoiding rapid blocking.
9.3.9 Activated carbon adsorption
The use of activated carbon adsorption has expanded greatly in Europe and North
America during the past two to three decades because most other water treatment
processes are ineffective in removing soluble organic matter. This approach uses either
powdered activated carbon (PAC) which can be added intermittently whenever the need
arises or GAC adsorbers which are used continuously. Accordingly, GAC may be more
expensive than PAC when used only intermittently, but it is also generally more effective
and more reliable for consistent removal of soluble organic compounds (Hamann et al.,
1990). Given the nature of cyanobacterial toxins, activated carbon adsorption would be
expected to offer some promise for toxin removal.
Powdered activated carbon
Keijola et al. (1988) found that 20 mg l-1 of PAC was able to achieve a 90 per cent
removal of hepatotoxins following conventional treatment combined with pre-ozonation.
Hart and Stott (1993) and Croll and Hart (1996) have reported the evaluation of several
PACs for the removal of microcystin-LR at an initial concentration of 40 µg l-1. With the
most effective PAC tested (wood based), doses greater than 20 mg l-1 were required to
achieve toxin removal of greater than 85 per cent.
Donati et al. (1993) also evaluated several different PACs for the removal of dissolved
microcystin-LR at an initial concentration of 50 µg l-1. For the best PAC they studied, a
dose of 25 mg l-1 with 30 minutes contact time was able to achieve 98 per cent removal,
while for the poorest a dose of 50 mg l-1 only achieved a 60 per cent removal. They
suggested that the mesopore volume of the various carbons was the best predictor of
carbon performance (Donati et al., 1994a). Nodularin was also removed with PAC
(Donati et al., 1994b). Likewise, Bernazeau (1994) found that 12 mg l-1 of PAC could
achieve a 95 per cent reduction of dissolved microcystin-LR from an initial concentration
of 50 µg l-1. Monitoring of a full scale conventional water treatment plant which was using
a PAC dose of 30 mg l-1 showed the combined treatment processes removed an average
of 82 per cent when microcystin levels in raw water were above 0.5 µg l-1 (Lambert et al.,
1996).
There is general agreement that to achieve high removal efficiencies, very high doses of
PAC are required for toxin removal and that contact time is very important. Lower doses
of PAC are required with pure water compared with natural water containing organic
matter and when using actual plant mixing conditions and contact times. Alum
coagulation in conjunction with PAC was also found to affect adversely toxin removal
(Jones et al., 1993).
Granular activated carbon
As might be expected, research into the performance of GAC has shown effective
removal of toxins, provided the adsorption capacity of the GAC has not been
compromised. Pilot scale tests treating microcystins at 30-50 µg l-1 showed greater than
90 per cent toxin removal for water treatment volumes up to 7,000-10,000 activated
carbon bed volumes before efficiency dropped to less than 63 per cent (probably
because of saturation of the GAC with dissolved organic carbon (DOC)) (Bernezeau,
1994). In these trials, the raw water had DOC levels at 5-6.5 mg l-1, more than 100-fold
greater concentration than the microcystins. The DOC: toxin ratio would be at least this
high under any realistic bloom conditions.
Studies by Hart and Stott (1993), using rapid column tests to simulate the performance
of GAC under dynamic conditions predicted bedlives to be fairly short for continuous
exposure to microcystin concentrations of 5-20 µg l-1. For example, Figure 9.3 shows
predicted bedlife for four different carbons, based on rapid column tests. The bedlife is
the time taken to reach 1 µg l-1 in the treated water with a constant concentration of 10
µg l-1 in the feed water, for a range of empty bed contact times (EBCTs). For EBCTs
typically used in water treatment of 10-15 minutes, the best performing carbon for this
water gave a bedlife of only 30-45 days. These results were confirmed in Australian
studies by Jones et al. (1993) and Craig and Bailey (1995) in both laboratory and pilot
plant studies, using air dried bloom material. The results showed that while various
GACs were effective for microcystin-LR removal, the life of the GAC was limited.
Saturation conditions probably explain the observations that a full-scale GAC adsorber
was achieving only between 40 and 60 per cent microcystin removal down to 0.6-1.2 µg
l-1 for raw water which typically had DOC levels of 20 mg l-1, 2,000 fold greater than the
toxin levels (Lambert et al., 1996).
Figure 9.3 Predicted GAC bedlives for 10 µg l-1 microcystin-LR input and 1 µg l-1
limit in filtrate from rapid column test results for four different carbons (After
Carlisle, 1994. Reproduced courtesy of the Foundation for Water Research, UK)
No published studies have been found using GAC specifically for removal of nodularin,
cylindrospermopsin or PSP toxins. However, a report by Falconer et al. (1989) showed
removal of Anabaena neurotoxicity on GAC. Given the later identification of the toxin of
the population studied, their study almost certainly assessed PSP toxin removal. Carlile
(1994) repeated the tests by Hart and Stott (1993), using the GAC they found to be most
effective, and found anatoxin-a to be adsorbed better than microcystin-LR.
Biologically active carbon
Granular activated carbon is not only an effective adsorption process but it is also an
effective medium for biological treatment. Because microcystin-LR has been shown to
be biodegradable (Fawell et al., 1993), it is therefore possible that the toxin could be
degraded on a biologically active GAC. Carlile (1994) undertook pilot scale tests using
two GACs, one that had been previously used on a pilot plant for total organic carbon
(TOC) removal and an unused GAC. For the tests, for each GAC, two different contact
times of 7.5 and 15 minutes were used. The pilot plant results showed that there was no
significant difference between the performance of the unused GAC and the used GAC at
both contact times. However, the comparison of pilot plant results with results of
modelling assuming removal only by adsorption, shows poorer removal by adsorption
only (without any biological activity) at both contact times (Figure 9.4). The implications
from this are that the better removal on the pilot plant resulted from biological activity on
the GAC, and that this biological activity developed very quickly also on the unused GAC.
Figure 9.4 Comparison of pilot scale test results for microcystin-LR removal by
used GAC (probably biologically active) with performance predicted from models
for adsorption only (After Hart et al., 1997. Reproduced courtesy of Blackwell
Science)
Pilot plant trials investigating anatoxin-a removal by GAC showed no breakthrough,

whereas modelled results for the same operating conditions predicted breakthrough (UK
WIR, 1995). This suggested that biological activity was also important for anatoxin-a
removal by GAC.
In practice, it is difficult to exclude biological activity from GAC adsorbers and therefore
better removal of both toxins than indicated by rapid column tests would be expected.
The pilot-scale experiments discussed above suggest that when biological activity is
established, GAC at 15 minutes effective bed contact time provides a high degree of
security for both microcystin-LR and anatoxin-a removal. However, as these results
currently are poorly confirmed in full scale application, careful surveillance of treatment
performance is essential for treatment plants removing cyanotoxins in the raw water with
GAC. This particularly pertains to monitoring of breakthrough when saturation with DOC
is approached.
9.4 Chemical oxidation and disinfection

Drinking water is treated with chemical oxidants to fulfil a wide variety of objectives
including: control of biofilm growth, colour removal, odour control, enhancement of
coagulation and flocculation, and iron or manganese oxidation. The most critical
application of chemical oxidants is for disinfection. The chemicals used most commonly
in municipal water treatment are chlorine, chloramines, ozone, chlorine dioxide and
potassium permanganate.
9.4.1 Oxidation combined with disinfection
Once cyanobacterial cells have been removed from water, dissolved cyanotoxins are
potentially susceptible to oxidation by disinfectants. Several substances have been
tested for this purpose in drinking water treatment.
Chlorine
Early work reported that substantial doses (5 mg l-1) of chlorine were ineffective in
destroying toxicity from algal extracts, as measured in mouse bioassays (Hoffman,
1976). Likewise, combined treatment processes which included chlorination at 0.5 mg l-1
were also found ineffective, suggesting little contribution from the chlorination stage
(Keijola et al., 1988; Himberg et al., 1989). Similarly, Lambert et al. (1996) found that
chlorination achieved negligible reduction in microcystin levels of 0.3-0.5 µg l-1 in treated
water. In these studies, chlorine may have been consumed rapidly by the high
concentrations of organic matter reported to be present, leaving insufficient available for
removal of microcystins. However, Nicholson et al. (1994) showed that chlorination could
be very effective at destroying microcystin-LR and nodularin under the correct treatment
conditions, i.e. free chlorine residual of 0.5 mg l-1 after 30 minutes contact time with pH <
8. In contrast, they found that chloramination was completely ineffective at destroying
microcystin-LR and nodularin, and this creates a problem for treating natural waters with
any substantial nitrogenous chlorine demand.
Carlile (1994), Croll and Hart (1996) and Hart et al. (1997) have reported tests with a
variety of oxidants using water spiked with dissolved microcystin-LR or anatoxin-a in the
range 5-10 µg l-1. The tests with chlorine used an applied dose of 1.7 mg l-1, which was
found to give a free residual of approximately 0.7 mg l-1 after 30 minutes. The
effectiveness of the chlorine in reducing microcystin-LR concentration was very
dependent on pH and time. At pH 5, removal was more than 93 per cent within 30
minutes whilst at pH 7 removal reached only 88 per cent after 22 hours. Tests with a
water containing Microcystis cells indicated that chlorination could be similarly effective.
Chlorination during treatment at a pH sufficiently low to show maximum effect might not
be feasible in practice. However, in conjunction with extended contact times with a
residual free chlorine concentration, microcystin is likely to be degraded. Monitoring of
this effect is important.
Chlorination tests have also been undertaken with water containing dissolved anatoxin-a.
Nicholson et al. (1994), as well as Carlile (1994), reported no discernible removal of
anatoxin-a by chlorination. Rositano and Nicholson (1994) also showed that chlorination
of anatoxin-a was ineffective with a dose of 15 mg l-1 at pH 7 for 30 minutes contact time,
providing only a 16 per cent removal. Recent Australian studies (unpublished results)
have shown that removal of cylindrospermopsin can be achieved with chlorine doses of
1-2 mg l-1 at pH levels between 6 and 7.5 and a chlorine residual of 0.5 mg l-1.
Care must be taken with chlorination procedures to avoid occupational exposure to toxic
levels of chlorine in the air, or the formation of excess levels of trihalomethanes.
Ozone
The most consistently efficient process for destruction of both ultra- and extracellular
microcystins appears to be ozonation, which can rapidly achieve essentially complete
destruction of microcystins, nodularin and anatoxin-a (Keijola et al., 1988; Himberg et al.,
1989; Rositano and Nicholson, 1994; Croll and Hart, 1996; Rositano et al., 1996; Hart et
al., 1997). The major consideration in the application of ozonation is the ozone
demanded by background DOC concentrations because, at a DOC level of 8.5 mg l-1,
ozone doses above 1 mg l-1 were necessary to achieve complete microcystin-LR
destruction (Rositano and Nicholson, 1994). The results of Hart et al. (1997)
demonstrate the importance of sufficiently high ozone doses (Figure 9.5). At low doses
up to 0.6 mg l-1, ozone degraded DOC and had little effect on microcystin-LR. Only after
the DOC demand was satisfied, did the ozone show an effect on microcystin-LR.
However, between 0.6 and 1.3 mg l-1, this effect consisted almost entirely of cellular lysis,
and only at 2 mg l-1 was extracellular toxin subsequently converted. These results
highlight the crucial importance of sufficiently high ozone doses as well as of careful
monitoring of performance, particularly with variable DOC concentrations in the water
source as occur during cyanobacterial blooms. As discussed in section 9.4.2, the
performance of ozone may be improved substantially if it is applied in several steps, e.g.
before destabilisation/flocculation as well as after filtration.
Recent work in Australia (unpublished results) has shown that the ozone dose
necessary to achieve removal of a range of PSP toxins in the concentration range 10-
100 µg l-1 was less than the ozone demand of the water. Other recent studies in Australia
(unpublished results) have shown that ozone is also effective for the removal of
cylindrospermopsin.
Figure 9.5 Effect of ozonation on the distribution of both intra- and extracellular
microcystin-LR from Microcystis dosed into a raw lowland water (After Hart et al.,
1997. Reproduced courtesy of Blackwell Science)
Care must be taken with ozone procedures to avoid occupational exposure to toxic
levels in the air.
Potassium permanganate
Potassium permanganate at 1 mg l-1 was found to achieve 95 per cent removal of

microcystin-LR in 30 minutes. However, in the presence of live intact cells removal was
much poorer, suggesting that permanganate was unable to penetrate or lyse the cells
effectively and was therefore unable to come into contact with the toxin (Rositano, 1996).
Hart and Stott (1993), Carlile (1994), Croll and Hart (1996) and WRc (1996) have all
reported similar observations for the removal of dissolved microcystin-LR and anatoxin-a
and the same limitation in treating Microcystis cells. Lam et al. (1995) reported that
potassium permanganate caused some cell lysis and liberation of microcystin-LR. This
finding may be influenced by longer contact times than those used by Rositano (1996).
Hydrogen peroxide and UV radiation
Hydrogen peroxide was found ineffective in toxin removal, whereas either UV alone or
UV with hydrogen peroxide achieved about a 50 per cent removal of microcystin-LR
after 30 minutes (Rositano and Nicholson, 1994).
In contrast, Croll and Hart (1996) and WRc (1996) found UV radiation was capable of
efficiently degrading both microcystin-LR and anatoxin-a, but only at very high doses of
about 20,000 mWs/cm2. A typical water disinfection dose is about 30 mWs/cm2, and
therefore UV on its own cannot be regarded as a practical method of toxin reduction. A
recent finding has shown that very high concentrations of microcystin-LR (50-200 mg l-1)
were rapidly (10-40 minutes) destroyed using UV light in the presence of a titanium
dioxide catalyst (Robertson et al., 1997). The potential applications of this finding in
water treatment remain to be explored.
Chlorine dioxide
Chlorine dioxide has strong oxidising ability, although only limited studies have been
conducted with this oxidant. Hart and Stott (1993) found that whilst a dose of 6 mg l-1 was
required to reduce 4.6 µg l-1 of dissolved microcystin-LR to less than 1 µg l-1, a dose as
great as 10 mg l-1 had no effect on about 4 µg l-1 of intracellular microcystin.
9.4.2 Pre-oxidation (before cell removal)
Pre-oxidation has been widely reported to assist coagulation, especially in the removal
of some algae and cyanobacteria. Oxidants have been shown to breakdown some
cyanotoxins effectively under certain conditions (see section 9.4.1) but may also lead to
cell lysis and toxin release. Thus pre-oxidation of toxic cyanobacteria is a highly critical
issue in treatment design.
Ozone has been most effective in oxidation of cell-bound microcystin, if applied at a

sufficiently high dose and contact time (see section 9.4.1). Dissolved air flotation has
been proposed in which the recycled water is saturated with ozone-rich air (Baron et al.,
1997). Ozone-rich air has also been proposed to be used in dispersed air flotation.
These approaches might result in reduction of extracellular toxin as well as enhanced
removal of cells.
Chlorine has been applied to destroy cell-bound microcystins before further treatment.
However, Lam et al. (1995) showed that chlorination of bloom material using a high dose
of 44 mg l-1, resulted in release of 64 per cent of the intracellular microcystin. Thus, pre-
chlorination of raw waters containing cyanobacterial cells risks the release of toxin from
otherwise intact cells.
Mouchet and Bonnélye (1998) have compared pre-ozonation and pre-chlorination with
respect to their effect in elimination of algae and cyanobacteria, as well as toxin release
and formation of by-products. They concluded that pre-chlorination is slightly more
effective than pre-ozonation in enhancing coagulation (96.9 per cent removal as
compared with 94.1 per cent in one treatment plant in France). However, this advantage
is offset by the problems of cell damage resulting in release of DOC and metabolites
which either may be toxic or may impart offensive taste and odour, as well as leading to
formation of by-products (particularly highly unpleasant chlorophenols). In contrast, for
pre-ozonation (usually dosed at 1 mg l-1) these authors found little, if any, cell lysis at
doses up to 3 mg l-1. They recommend pre-ozonation as the better choice, especially in
conjunction with a main ozonation step further in the treatment line, e.g. between
clarification and filtration. It is however acknowledged that pre-chlorination is still very
common, particularly in developing countries. While the advantages for improving
clarification, keeping filters clean, eliminating ammonia and enhancing post-chlorination
are well established, pre-chlorination in plants without subsequent adsorption onto
activated carbon is not recommended.
Prior to cell removal, the total and dissolved organic carbon load of water with
cyanobacterial blooms will vary by orders of magnitude, and consumption of the oxidant
will therefore also vary widely. Continuous control of the oxidising step and very high
doses may be necessary to ensure complete oxidation of cyanotoxins in one pre-
treatment step. This is likely to be difficult in practice, and is associated with a risk of
toxin liberation. Removing cyanobacterial cells before application of oxidant is safer. In
contrast, pre-oxidation with a low ozone dose may be useful because it substantially
enhances cell removal by subsequent steps. Safe and effective operation is possible if
further cyanotoxin barriers (such as a further ozone step or GAC) are available.
Consequently, pre-oxidation may be regarded as a step for enhancement of cell removal
rather than cyanotoxin degradation, and requires either monitoring for breakthrough of
dissolved toxins during cyanobacterial blooms or for the use of further multiple barriers in
the treatment system.
9.5 Membrane processes and reverse osmosis

Membrane processes, particularly microfiltration (MF) and ultrafiltration (UF) are
increasingly seen, under some circumstances, as economically viable treatment
alternatives to conventional treatment for small and large communities. They should be
effective in the removal of cyanobacteria and intracellular toxins.
Experimental studies at laboratory scale with flat-sheet UF and MF membranes, in both

dead-end and crossflow modes, have shown high efficiency of removal (> 98 per cent) of
whole cells of toxic M. aeruginosa (Chow et al., 1997b). This study also examined the
effect of the filtration process on cell integrity by fluorescence microscopy and assessed
cell damage by measuring the leakage of cell chlorophyll and toxin (microcystin-LR) into
the permeate. There was evidence of damage to a small proportion of cells following
filtration, but no significant increase in toxin in the permeate with all modes of filtration. In
experiments with the ultrafiltration membrane, the amount of microcystin was
significantly lower in the permeate than in the feed, which suggested that the particular
UF membrane employed may have rejection properties or adsorption ability for
microcystin. This would not be expected for UF membranes although removal of soluble
toxin may be achieved with a very low molecular weight cut-off pore size, such as those
offered by nanofiltration membranes. Hart and Stott (1993) evaluated the effect of
nanofiltration for the removal of microcystin spiked into natural water at concentrations
between 5 µg l-1 and 30 µg l-1 and found removal to below 1 µg l-1. Australian studies with
membranes (Muntisov and Trimboli, 1996) also showed that using nanofiltration
microcystin-LR and nodularin at 8 µg l-1 were removed from water from the River Murray
that had been spiked with toxin.
Neumann and Weckesser (1998) have tested three reverse osmosis membranes at 25-
35 bar for elimination of microcystin-LR and microcystin-RR from tap and salt (3,000 mg
l-1 NaCl) water. Initial toxin concentrations in the retentate were in the range 70-130 µg l-1.
With a detection limit of 0.2 µg l-1, average retention levels were 96.7-99.6 per cent.
There was no statistical difference in retention of the microcystins between the two
waters.
9.6 Microcystins other than microcystin-LR

Most of the published research relates to microcystin-LR, even though its concentration
can be exceeded by those of other variants, or by the sum of the concentrations of other
variants (Codd and Bell, 1996).
Computer models are available which can be used to predict the properties of chemical
compounds, based on their chemical structure, in order to provide information in relation
to toxicology and environmental impact. Such models have been used to predict the
properties of the microcystin variants which would be important in relation to removal by
water treatment processes (WRc, 1997).
A physical measure of solubility and interaction with water molecules that gives an
indication of the potential adsorption by activated carbon is the octanol-water partition
coefficient, Kow. This is defined as the ratio of the concentration in the octanol phase to
the concentration in the water phase in a two-phase octanol-water system at equilibrium,
and is usually expressed as a logarithm. Readily adsorbed, hydrophobic compounds
have high values and poorly adsorbed hydrophilic compound have low values (often
negative, indicating a higher concentration of the compound in the water phase). Kow
values have been estimated from molecular structure to provide an indication of the
relative hydrophobicity compared with microcystin-LR (for which some information on
activated carbon adsorption is available). The calculated Kow values suggest that the
majority of variants would be adsorbed by activated carbon similarly to, or better than,
microcystin-LR. Hence any strategy for using activated carbon, based on the data
available for microcystin-LR, would probably be suitable for the majority of the other
microcystins.
Attempts to model the reactivity of microcystin variants with oxidants have been
unsuccessful because of the complexity of the molecular structure. A principal
mechanism of action of oxidants, particularly ozone and chlorine, on organic compounds
is by the breakdown of double bonds. Any modifications to the basic microcystin
structure which increases the degree of double bonding in the molecule would therefore
be expected to enhance its reaction with ozone or chlorine. It has been concluded, from
consideration of the amino acid functional groups in the variants, that some variants
would be expected to be more reactive with oxidants than microcystin-LR, although the
effect may not be important in practical terms because the basic molecular structure is
not changed radically. For the same reason, the other variants would not be expected to
be much less reactive with oxidants than microcystin-LR. Hence, any strategy for oxidant
application based on microcystin-LR data would probably be just as effective for the
other microcystins.
The modelling approach available for biodegradability can only class compounds as
biodegradable or non-biodegradable, and cannot provide any further quantification to the
degree of biodegradability. Modelling has classed microcystin-LR as biodegradable, and
changes to the amino acids have not changed this classification. Hence all the variants
would be expected to show similar biodegradability to microcystin-LR. This would be of
significance in relation to the performance of biological GAC and slow sand filtration
processes.
The lack of experimental data on the elimination of microcystins other than microcystin-
LR emphasises the need to monitor performance of any treatment system that is applied
for cyanotoxin removal.
9.7 Effective drinking water treatment at treatment works

There are a number of messages that arise from the published work with respect to
good practice as well as effective design and operation of water treatment works. These
include:
• Resources and abstraction should be managed to minimise the presence of algal

concentrations in the raw water delivered for treatment.
• Chemical preparation and dosing facilities must be of adequate size, process control
should ensure rapid dispersion and appropriate retention times, and chemical doses
should be optimised at the appropriate pH.
• Some oxidants, e.g. ozone, can be dosed before coagulation and clarification but
require particular care, not only to avoid lysis of cells but also to limit problems with
disinfection by-product formation. Separation of steps into a low pre-oxidation dose to
enhance flocculation and a higher dose after cell removal to oxidise dissolved toxins is a
safer approach.
• Granular activated carbon plants with a high EBCT and ozone-GAC facilities may
remove toxins effectively, especially if the GAC supports substantial biological activity.
• The effectiveness of treatment plants without ozone but with GAC will depend on the
GAC EBCT value, on the degree of biological activity on the GAC, on the extent of
exhaustion of the GAC and of the magnitude and duration of toxin occurrence.
• Conventional treatment plants without ozone and GAC might remove cyanobacterial
cells and dissolved toxins satisfactorily if coagulation, clarification, filtration and
superchlorination-dechlorination (with a contact time of>15 mg min l-1) or ozonation are
carried out effectively.
• Slow sand filter plants remove algal cells effectively, although pre-treatment steps are
generally applied to maximise filter runs and efficiency. Because of the biological activity
in slow sand filters and long contact times, some removal of dissolved toxin should be
expected but this capability is unclear. Slow sand filter plants with pre-ozonation and/or
sand-GAC sandwiching would be expected to be effective for dissolved toxins (although
confirmation of this expectation is needed).
• Frequent monitoring of treatment performance is crucial to ensure safety, particularly

with respect to cyanotoxin removal, because available information on the performance of
different treatment steps is specific to the conditions of the experiments reported, and
performance under other conditions is unclear. Variable and often high loads of DOC
during cyanobacterial blooms may rapidly compromise treatment procedures that were
initially successful.
• Most procedures have been studied for cyanotoxin removal as isolated treatment steps,
rather than as a combination following the multi-barrier principle. Planning of treatment
will lead to best results if combinations are considered, and if cell and dissolved toxin
removal are separately evaluated (e.g. combinations of pre-oxidation to enhance cell
removal with effective post-oxidation to ensure destruction of liberated toxin, or
combinations of cell removal and slow sand filtration).
Perhaps because of the intermittent nature of cyanobacterial blooms, very little

information has been reported from full-scale treatment plants treating water at naturally
occurring toxin levels. Laboratory and pilot-scale investigations have shown that
dissolved toxins can be removed effectively to less than 1 µg l-1 under conditions
normally used in water treatment by biologically active GAC, ozone, potassium
permanganate and chlorine (microcystin only). The information which has been reported
to date is summarised in Table 9.3.
Table 9.3 Summary of water treatment performance on microcystins
Expected removal1
Treatment technique Comments
Cell bound Extracellular
Coagulation/sedimentation/dissolved > 80% <10% Removal only achievable
air flotation for toxins in cells,
provided cells are not
damaged
Precipitation/sedimentation > 90% <10% Removal only achievable
for toxins in cells,
damaged
Rapid filtration > 60% <10% Removal only achievable
for toxins in cells,
damaged
Slow sand filtration ~ 99% Probably Removal effective for
significant toxins in cells; efficiency
for dissolved microcystin
is likely to depend on
biofilm formation and thus
on filter run length
Combined > 90% < 10% Removal only achievable
coagulation/sedimentation/filtration for toxins in cells,
damaged
Dissolved air flotation > 90% Not assessed, Removal only achievable
probably low for toxins in cells,
damaged
Adsorption - Powdered activated Negligible > 85% For adequate PAC doses
carbon (PAC) (> 20 mg l-1) with a PAC
shown to be effective,
DOC competition will
reduce capacity
Adsorption - Granular activated See rapid > 80% For practical EBCTs, DOC
carbon (GAC) filtration competition will reduce
capacity and hasten
breakthrough, filtration
also removes algal cells
Biological granular activated carbon See rapid > 90% See GAC, biological
filtration activity enhances removal
efficiency and bed life
Pre-ozonation Very effective Potential Useful in low doses to
in enhancing increase assist coagulation of cells;
coagulation risk of toxin release
requires careful
monitoring and possibly
subsequent treatment
steps
Pre-chlorination Very effective Causes lysis Useful to assist
in enhancing and release of coagulation of cells but
coagulation dissolved applicable for toxic
metabolites cyanobacteria only if
subsequent treatment
steps will remove
dissolved toxins and other
released metabolites
Ozonation (post clarification) - > 98% Rapid and efficient on
soluble toxin provided that
DOC demand is satisfied
Free chlorine (postfiltration) - > 80% Effective when free
chlorine is > 0.5 mg I-1
after > 30 minutes at pH <
8 and low DOC; effect
negligible when dose low
or pH > 8
Chloramine - Negligible Ineffective. Free chlorine
application will yield
ineffective chloramines in
waters enriched with
nitrogenous organic
matter
Chlorine dioxide - Negligible Not effective with doses
used in drinking water
treatment
Potassium permanganate - 95% Effective on soluble toxin
but only in absence of
whole cells
Hydrogen peroxide - Negligible Not effective on its own
UV radiation - Negligible Capable of degrading
microcystin-LR and
anatoxin-a, but only at
impractically high doses
Membrane processes Likely to be Uncertain Depends on membrane
very high (> type, further research
99%) required to characterise
performance
DOC Dissolved organic carbon
Source: Adapted from Yoo et al., 1995
1
Likely efficiency of removal when continuously applied at optimal doses and pH and
under proper operating conditions
9.8 Drinking water treatment for households and small
community supplies
Domestic upgrading of piped drinking water supplies has been a recent issue of concern
in some countries. Many central supplies provide excellent quality drinking water and
additional household treatment may actually cause deterioration rather than
improvement. However, domestic treatment may have a role in regions supplied with
poor quality drinking water, or for especially sensitive sub-populations. Furthermore, in
many parts of the world, simple and easily maintained treatment for households and
small communities may improve the quality of water otherwise used for drinking without
any treatment. Boiling water will not remove or degrade cyanotoxins (Chen et al., 1998).
Lawton et al. (1998) tested three different domestic jug filtration units for their capacity to
remove extracellular microcystins (LR, LY, LW and LF) and one unit for removal of intact
cells (Microcystis aeruginosa as single cells, spiral filaments of Anabaena flos-aquae
and straight filaments of Planktothrix (syn. Oscillatoria) agardhii). Treatment in the jug
units is based on activated carbon and ion exchange resins. Whereas approximately 60
per cent of the filamentous cyanobacteria were removed, 90 per cent of the single cells
of Microcystis passed through the filter (removal of large Microcystis colonies was not
tested but may be more effective). Cell morphology was thus considered crucial for
elimination performance. Removal of microcystin variants ranged from 32 to 57 per cent
(using new cartridges) and could be augmented to a cumulative removal of 88 per cent
by three repeated passages of the same water through the filter. On filter cartridges
which have reached the half-life recommended by the manufacturer, performance for
extracellular microcystin-LR dropped for two of the three brands tested, in one case
down to 15 per cent elimination. The study draws attention to the possibility of lysis of
cells retained on the filter. It also highlights the need for further development of domestic
jug filters if they are to be suitable for microcystin removal. In addition, the study
emphasises the need for evaluation of performance of treatment processes in specific
situations, particularly if scaled down for domestic use.
Other approaches for individual households and small communities involve methods of
filtration, activated carbon and oxidation. As for large-scale plants, slow sand filtration
will be effective in removal of cells, and will probably contribute to removal of dissolved
cyanotoxins. Rapid blocking can be avoided by pre-treatment to control turbidity or by
management of flow regime. Bankside filtration could also be effective and applicable to
small community supplies. Addition of chlorine to filtered water at a dose high enough to
oxidise microcystins has already been discussed with respect to the benefits for
microcystin removal in relation to the problems of by-product production at high DOC
levels.
Household treatment approaches have the problem of assessment of performance and

quality control. Furthermore, they may enhance social differences, if they are available
only to those who can afford them rather than providing "health for all".
9.9 References
Baron J., Ionesco, N.M. and Bacquet, G. 1997 Combining flotation and ozonation the
Flottazone process. In: Dissolved Air Flotation. Proceedings of an International
Conference, London, Chartered Institution of Water and Environment Management,
London.
Bernezeau, F. 1994 Can microcystins enter drinking water distribution systems, In: D.A.
Steffensen and B.C. Nicholson [Eds] Toxic Cyanobacteria, Current Status of Research
and Management. Proceedings of an International Workshop, Adelaide, Australia,
American Water Works Association Research Foundation, Australian Centre for Water
Quality Research, Centre for Water Research, Belgium, 115-118.
Bernhardt, H and Clasen, J. 1991 Flocculation of micro-organisms, J. Water SRT Aqua.

40(22), 76-87.
Bourke, A.T.C., Hawes, R.B., Neilson, A. and Stallman, N.D. 1983 An outbreak of
hepato-enteritis (the Palm Island Mystery Disease) possibly caused by algal intoxication.
Toxicon, 3, Supplement, 45-48.
Bourne, D.G., Jones, G.J., Blakeley, R.L., Jones, A., Negri, A.P. and Riddles, P. 1996
Enzymatic pathway for the bacterial degradation of the cyanobacterial cyclic peptide
toxin, microcystin-LR. App. Environ. Microbiol., 62, 4086-4094.
Burch, M.D., Velzeboer, R.M.A., Chow, C.W.K., Stevens, H.C., Bee, C.M. and House, J.
1998 Evaluation of Copper Algaecides for the Control of Algae and Cyanobacteria.
Research Report No. 130. Urban Water Research Association of Australia, Melbourne,
Australia.
Carlile, P.R. 1994 Further Studies to Investigate Microcystin-LR and Anatoxin-a

Removal from Water. Report No. 0458, Foundation for Water Research, Marlow, UK.
Casitas Municipal Water District 1987 Current Methodology for the Control of Algae in
Surface Reservoirs. American Water Works Association Research Foundation, Denver.
Chen, G., Chen, C.W., Yu, S.Z., Ueno, Y. 1998 Evaluation of different water treatment to
remove microcystins by using a highly sensitive ELISA. In: M. Miraglia, H. van Egmond,
C. Bera, J. Gibert [Eds] Mycotoxins and Phycotoxins-Developments in Chemistry,
Toxicology and Food Safety.
Chorus I., Klein, G., Fastner, J. and Rotard, W. 1993 Off-flavors in surface waters, how
efficient is bank filtration for their abatement in drinking water? Wat. Sci. Technol., 25/2,
251-258.
Chow, C.W.K., House, J., Velzeboer, R.M.A., Drikas, M., Burch, M.D. and Steffensen
D.A. 1997a The effect of ferric chloride flocculation on cyanobacterial cells. J. Water
SRT Aqua, 46, 324-334.
Chow, C.W.K., Panglisch, S., Mole, J., Drikas, M., Burch, M.D. and Gimbel, R. 1997b A
study of membrane filtration for removal of cyanobacterial cells. AQUA (In Press).
Codd, G.A and Bell, S.G. 1996 The Occurrence and Fate of Blue-green Algal Toxins in
Freshwaters. R&D Report 29, Dundee University, Her Majesties Stationary Office,
London.
Collins, M.R. 1998 Experiences introducing "new" technology: slow sand filtration.
Lecture and abstract of the NSF International PAHO/WHO first International Symposium
on Safe Drinking Water in Small Systems, Technology, Operation and Economics,
Washington, DC.
Craig, K. and Bailey, D. 1995 Cyanobacterial toxin microcystin 'LR' removal using
activated carbon, Hunter Water Corporation experience. In: Proceedings of the
Australian Water and Wastewater Association 16th Federal Convention, Sydney.
Croll, B and Hart, J. 1996 Algal toxins and customers. Paper presented at the UKWIR-
AWWARF Technology Transfer Conference, Philadelphia.
Dickens, C.W.S. and Graham, P.M. 1995 The rupture of algae during abstraction from a
reservoir and the effects on water quality. J. Water SRT, 44(1), 29-37.
Donati, C.D., Drikas, M., Hayes, R. and Newcombe, G. 1993 Adsorption of microcystin-
LR by powdered activated carbon. Wat. J. AWWA, 20(3), 25-28.
Donati, C, Drikas, M, Hayes, R and Newcombe, G. 1994a Activated carbon in drinking

water treatment: II Adsorption of nodularin. Poster presented at the 8th International
Conference on Surface and Colloid Science, Adelaide, February, 1994.
Donati, C.D., Drikas, M., Hayes, R. and G. Newcombe. 1994b Microcystin-LR adsorption
by powdered activated carbon. Wat. Res., 28, 1735-1742.
Drikas, M. 1994 Session IV: Control and or Removal of Toxic Cyanobacteria. In: D.A.,
Steffensen and B.C. Nicholson [Eds] Toxic Cyanobacteria, Current Status of Research
and Management. Proceedings of International Workshop, Adelaide, Australia. American
Water Works Association Research Foundation, Australian Centre for Water Quality
Research, Centre for Water Research, Belgium.
Drikas, M., Chow, C.W.K., House, J. and Burch, M.D. 1997 A pilot study of the removal
of intact cyanobacterial cells. J. AWWA (In Preparation).
Effler, S.W., Linen, S., Field, S.D., Tong-Ngork, T., Hale, F., Meyer, M., and Quirk, M.
1980 Whole lake responses to low level copper sulphate treatment. Wat. Res. 14, 1489-
1499.
Elder, J.F. and Home, A.J. 1978 Copper cycles and CuSO4 algicidal capacity in two
California lakes. Environ. Manage. 2, 17-30.
Falconer, I.R., Runnegar, M., Buckley, T., Huyn, V. and P. Bradshaw, 1989 Using
activated carbon to remove toxicity from drinking water containing cyanobacterial blooms.
J. Am. Water Works Assoc., 81(2), 102-105.
Fawell, J.K., Hart, J, James, H.A. and Parr. W. 1993 Blue-green algae and their toxins
analysis, toxicity, treatment and environmental control. Wat. Supply, 11(3/4), 109-121.
Fitzgerald, G.P. 1966 Use of potassium permanganate for control of problem algae. J.
AWWA, 58, 609-614.
Fitzgerald, G.P. and Faust, S.L. 1963 Factors affecting the algicidal and algistatic
properties of copper. Appl. Microbiol., 11, 3345-351.
Florence, T.M. 1982 The speciation of trace elements in waters. Talanta, 29, 345-364.
Gregory, R and Zabel, T.F. 1990 Sedimentation and flotation. In: F. W. Pontius [Ed.]
Water Quality and Treatment, A Handbook of Community Water Supplies. 4th edition.
American Water Works Association, McGraw Hill, Inc., New York, 443-445.
Grohmann, A., Hahn, H.H., Klute, R. [Eds] 1985 Chemical Water and Wastewater
Treatment. Gustav Fischer Verlag, Stuttgart, New York, 311 pp.
Hamann, C.L., McEwen, J.B. and Myers A.G. 1990 Guide to selection of water treatment
processes. In: F. W. Pontius [Ed.] Water Quality and Treatment - A Handbook of
Community Water Supplies. 4th edition. American Water Works Association, McGraw
Hill Inc., New York, 157-187.
Hanson, M.J. and Stefan, H.G. 1984 Side effects of 58 years of copper sulphate
treatment of the Fairmont Lakes, Minnesota. Wat. Res. Bull., 20, 889-900.
Hart, J. and Stott, P. 1993 Microcystin-LR Removal from Water. Report FR 0367,
Foundation for Water Research, Marlow, UK.
Hart, J, Fawell, J.K and Croll, B. 1997 The fate of both intra and extracellular toxins
during drinking water treatment. Special subject No. 18, SS18-1-6, IWSA World
Congress, Blackwell Science, Oxford.
Himberg, K., Keijola, A.M., Hiisvirta, L., Pyysalo, H. and Sivonen, K. 1989 The effect of
water treatment processes on the removal of Microcystis and Oscillatoria cyanobacteria:
a laboratory study. Wat. Res., 23, 979-984.
Hoffman, J.R. 1976 Removal of Microcystis toxins in water purification processes. Water
SA. 2(2), 58-60.
Hoffman, R.W., Bills, G., and Rae, J. 1982 An in situ comparison of the effectiveness of
four algicides. Wat. Res. Bull., 18, 921-927.
Holden, W.S. 1970 The control of organisms associated with water supplies. In: Water
Treatment and Examination. J.&A. Churchill, London, 453-460.
Humberg, N.E., Colby, S.R., Hill, E.R., Kitchen, L.M., Lym, R.G., McAvoy, W.J. and
Prasad, R. 1989 Herbicide Handbook of the Weed Science Society of America. 6th
edition. Weed Science Society of America, Illinois.
Izaguirre, G. 1992 A copper-tolerant Phormidium species from Lake Matthews,
California, that produced 2-methylisoborneol and geosmin. Wat. Sci. Tech., 25, 217-223.
James, H. and Fawell, J.K. 1991 Detection and Removal of Cyanobacterial Toxins from
Freshwaters. Report No. FR0211 Foundation for Water Research, Marlow, UK.
Jones, G., Minatol, W., Craig, K. and Naylor, R. 1993 Removal of low level
cyanobacterial peptide toxins from drinking water using powdered and granular activated
carbon and chlorination. Result of laboratory and pilot plant studies. In: Proceedings of
the Australian Water and Wastewater Association 15th Federal Convention, Gold Coast,
Australia.
Jones, G. and Orr, P.T. 1994 Release and degradation of microcystin following algicide
treatment of a Microcystis aeruginosa bloom in a recreational lake, as determined by
HPLC and protein phosphatase inhibition assay. Wat. Res., 28(4), 871-876.
Keijola, A.M., Himberg, K., Esala, A.L., Sivonen, K. and L. Hiisvirta, 1988 Removal of
cyanobacterial toxins in water treatment processes: laboratory and pilot-scale
experiments. Tox. Assess., 3, 643-656.
Kenefick, S.L., Hrudey, S.E., Peterson, H.G. and E.E. Prepas, 1993 Toxin release from
Microcystis aeruginosa after chemical treatment. Wat. Sci. Tech., 27(3-4), 433-440.
Lahti, K., Kilponen, J., Kivimäki, A.-L., Erkomaa, K. and Sivonen, K. 1996 Removal of
cyanobacteria and their hepatotoxins from raw water in soil and sediment columns. In:
A.-L. Kivimäki and T. Suokko [Eds] Artificial Recharge of Groundwater. NHP/Report No.
38, Helsinki, 187-195.
Lam, A., Prepas, E., Spink, D. and Hrudey, S.E. 1995 Control of hepatotoxic
phytoplankton blooms; implications for human health. Water Res. 29, 1845-1854.
Lambert, T.W., Holmes, C.F. and Hrudey, S.E. 1996 Adsorption of microcystin-LR by
activated carbon in full scale water treatment. Wat. Res. 30,1411-1422.
Laszlo, F. 1984 Potential release of pollutants in riverbank filtration systems along the
River Danube, Hungary. In: Proceedings of the 3rd International Symposium on
Interactions Between Sediments and Water. CEP Consultants Ltd., Edinburgh, 264-267.
Lawton, L.A., Cornish, B.J.P.A. and MacDonald, A.W.R. 1998 Removal of

cyanobacterial toxins (microcystins) and cyanobacterial cells from drinking water using
domestic water filters. Wat. Res. 32, 633-638.
Lepistö, L., Lahti, K., Niemi, J. and Färdig. M. 1994 Removal of cyanobacteria and other
phytoplankton in four Finnish waterworks. Arch. Hydrobiol. Algological Studies, 75, 167-
181.
Leuschner, C. 1984 Auswirkungen der Phosphateliminierungsanlage Beelitzhof auf die

qualitative und quantitative Zusammensetzung der Phytoplanktonpopulation im
Wannseewasser bei der Überleitung in den Schlachtensee. Report for the Institute of
Water, Air and Soil Hygiene, Federal Environmental Agency, Berlin, 62 pp.
Mackenthun, K.M. and Cooley, H.L. 1950 The biological effect of copper sulphate
treatment on lake ecology. Trans. Wisconsin Acad. Sci. Arts Lett., 40, 177-187.
Markham. L, Porter, M. and Schofield, T. 1997 Algal and zooplankton removal by

dissolved air flotation at Severn Trent Ltd. surface water treatment works. In: Dissolved
Air Flotation. Proceedings of an International Conference, Charted Institution of Water
and Environmental Management, London.
McGuire, M.J. and Gaston, J.M. 1988 Overview of technology for controlling off-flavours
in drinking water. Wat. Sci. Tech., 20(8/9), 215-228.
McKnight, D. 1981 Chemical and biological processes controlling the response of a

freshwater ecosystem to copper stress: a field study of the CuSO4 treatment of Mill Pond
Reservoir, Burlington, Massachusetts. Limnol. Oceanogr., 26, 518-531.
McKnight, D.M., Chisholm, S.W., and Harleman, D.R.F. 1983 CuSO4 treatment of
nuisance algal blooms in drinking water reservoirs. Environ. Manage., 7, 311-320.
Mole J., Chow, C., Drikas, M., Burch, M. 1997 The influence of culture media on growth
and toxin production of the cyanobacterium Microcystis aeruginosa Kütz Emend Elenkin.
Paper presented at the 13th Annual Conference of the Australian Society for Phycology
and Aquatic Botany, Hobart, January 1997.
Moore, G.T. and Kellerman, K.F. 1905 Copper as an algicide and disinfectant in water
supplies. Bull. Bureau Plant. Indus. USDA, 76, 19-55.
Mouchet P. and Bonnélye V. 1998 Solving algae problems: French expertise and world-
wide applications. J. Water SRT, Aqua., 47, 125-141.
Muntisov, M. and Trimboli, P. 1996 Removal of algal toxins using membrane technology.
Water, 23(3), 34.
National Rivers Authority 1990 Toxic Blue-Green Algae. Water Quality Series No. 2.,
National Rivers Authority, London, 128 pp.
Neumann, U. and Weckesser, J. 1998 Elimination of microcystin peptide toxins from

water by reverse osmosis. Environ. Toxicol. Water Qual., 13.
Nicholson, B.C., Rositano, J. and Burch, M.D. 1994 Destruction of cyanobacterial

peptide hepatotoxins by chlorine and chloramine. Wat. Res. 28,1297-1303.
Palmer, C.M. 1962 Control of algae. In: Algae in Water Supplies. An Illustrated Manual
on the Identification, Significance, and Control of Algae in Water Supplies. U.S.
Department of Health, Education, and Welfare Public Health Service, Washington D.C.,
63-66.
Prepas, E.E. and Murphy, T.P. 1988 Sediment-water interactions in farm dugouts
previously treated with copper sulphate. Lake Reserv. Manage., 4, 161-168.
Raman, R.K. 1985 Controlling algae in water supply impoundments. J. AWWA, August,
41-43.
Raman, R.K. 1988 Integration of laboratory and field monitoring of copper sulphate
applications to water supply impoundments. In: Advances in Water Analysis and
Treatment. AWWA Technology Conference Proceedings, St. Louis, Missouri, 203-224.
Robertson, P.K., Lawton, L.J., Munch, B. and Rouzade, J. 1997 Destruction of

cyanobacterial toxins by semiconductor photocatalysis. Chem. Commun., Issue No. 4,
393-394.
Rositano J. 1996 The Destruction of Cyanobacterial Peptide Toxins by Oxidants used in

Water Treatment. Report 110, Urban Water Research Association of Australia.
Rositano, J. and Nicholson, B.C. 1994 Water Treatment Techniques for Removal of
Cyanobacterial Toxins from Water. Australian Centre for Water Quality Research.
Salisbury, South Australia, 55 pp.
Rositano, J., Nicholson, B.C. and Pieronne, P. 1996 Destruction of cyanobacterial toxins
by ozone. In: Proceedings of the First Australasian Conference of the International
Ozone Association, Sydney, Australia.
Sanchez, I. and Lee, G.F. 1978 Environmental chemistry of copper in Lake Monona,
Wisconsin. Wat. Res., 12, 899-903.
Sawyer, C.N. 1962 Causes, effects and control of aquatic growths. J. Wat. Pollut.
Control Fed., 34, 279-288.
Sherman, P., Tully, I. and Gibson, H. 1995 Removal of cyanobacterial cells and toxins
from drinking water with biologically active filters. In: Proceedings of the Australian Water
and Wastewater Association 16th Federal Convention, Sydney.
Steffensen, D.A. and Nicholson, B.C. [Eds] 1994 Toxic Cyanobacteria -Current Status of
Research and Management. Proceedings of an International Workshop, Adelaide,
Australia, American Water Works Association Research Foundation, Australian Centre
for Water Quality Research, Centre for Water Research, Belgium, 172 pp.
UNDP/WHO, 1992 Prediction of Pollution Hazards Affecting Subsurface Drinking Water

Resources, Hungary, Project findings and recommendations. UNDP/HUN/90/001-
WHO/HUN/CWS/010, World Health Organization Regional Office for Europe,
Copenhagen, 33 pp.
UK WIR, 1995 GAC Tests to Evaluate Algal Toxin Removal. Report DW-07/C, UK Water
Industry Research Ltd., London.
Velzeboer, R., Drikas, M., Donati, C., Burch, M. and Steffensen, D. 1995 The removal of
cyanobacterial cells by alum flocculation. In: Proceedings of the Australian Water and
Wastewater Association 16th Federal Convention, Sydney.
Vlaski, A., van Breemen, A.N. and Alaerts, G.J. 1997 Algae laden water treatment by
dissolved air flotation. In: Dissolved Air Flotation. Proceedings of an International
Conference, London, Chartered Institution of Water and Environmental Management,
London.
WHO 1996 Guidelines for Drinking Water Quality. Volume 2, Health Criteria and other
Supporting Information. World Health Organization, Geneva, 973 pp.
WRc, 1996 The Fate of Intracellular Microcystin-LR During Water Treatment. Report Ref.
96/DW/07/4, UK Water Industry Research Ltd., London.
WRc, 1997 Algal Toxins: Occurrence and Treatability of Anatoxin and Microcystins.
Report 97/DW/07/E, UK Water Industry Research Ltd., London.
Yoo, R.S., Carmichael, W.W., Hoehn, R.C. and Hrudey, S.E. 1995 Cyanobacterial (Blue-
Green Algal) Toxins: A Resource Guide. American Water Works Association Research
Foundation, Denver, 229 pp.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 10. DESIGN OF MONITORING PROGRAMMES
This chapter was prepared by Geoffrey A. Codd, Ingrid Chorus and Mike Burch
Cyanobacterial blooms and cyanotoxins present a special challenge to monitoring

programmes because the requirements are different from well-recognised monitoring
designs both for pathogenic bacteria and for toxic chemicals. Pathogen concentrations
are highest close to sewage outfalls or the inflow of agricultural runoff polluted by
livestock faeces and they are diluted with increasing distance from such sources. In
contrast, cyanobacteria multiply in the open water environment and scum-forming
species are often dramatically concentrated by wind action. Furthermore, formation and
dispersion of scums may change within days or even hours, making assessment of the
associated hazard difficult. Toxic chemicals are dissolved in the water or bound to
sediments. In contrast, at least while the producer cells remain intact, cyanotoxins are
chiefly contained within the cells. Therefore, they may shift position in the water body
with the cells and accumulate to hazardous concentrations. Consequently, monitoring
strategies must encompass cell-bound toxins in addition to extracellular toxin pools.
Cyanobacterial distributions and their changes in space and time depend on the
morphological, hydrological, meteorological and geographic characteristics of a given
water body. Because the distribution of cyanobacteria is central to hazard assessment,
the design of monitoring programmes should be specifically tailored for each water body
to optimise the relation of information output to work input. Monitoring approaches also
need to be more flexible than for many other parameters. Local knowledge of bloom
history and a good understanding of the local growth conditions for cyanobacteria will
greatly enhance the capacity to anticipate bloom formation. As knowledge and
understanding of a given water body accumulate, regular patterns of Cyanobacterial
growth may be noticed, so that in the long-term, monitoring may be focused upon critical
periods and locations. To ensure efficiency of Cyanobacterial monitoring programmes,
they should be reviewed regularly to provide the most cost effective use of resources
and in order to continue to satisfy the primary needs for which they were established.
Rapid evaluation and interpretation of results is important in order to achieve feedback
into ongoing programmes and their adaptation to current needs.
Analytical quality assurance, as well as data analysis, interpretation and presentation are
important aspects of monitoring programmes. These topics are covered in two of the
companion volumes in this series: Water Quality Assessments (Chapman, 1996) and
Water Quality Monitoring (Bartram and Ballance, 1996).
10.1 Approaches to monitoring programme development
10.1.1 Objectives of monitoring programmes
The objectives of a monitoring programme determine the approach, the design and the
resources required. The aims and further applications of monitoring programmes
focused on cyanobacterial populations and toxins may include, for example:
• Assessment of health hazards caused by cyanobacteria and their toxins.
• Identification of contaminated areas (e.g. in relation to drinking water intakes and

recreational sites).
• Development of regulations concerning the development and use of recreational sites.
• Public education and information.
• Assessment of the causes of cyanobacterial problems (nutrient concentrations and

other limnological data for understanding cyanobacterial growth).
• Development of a nutrient pollution control programme.
• Checking whether compliance with cyanobacterial cell (or biomass) and toxin level
standards for the respective water use is being achieved.
• Prediction of levels and changes in cyanobacterial populations and toxins resulting

from natural phenomena and human influence.
• Information of the effect of interventions, including lake and reservoir management and
water treatment methods, on cyanobacterial cell and toxin levels.
• Wider contribution to the knowledge of cyanobacterial ecology, hydrobiology and the

state of the environment.
The approach to monitoring programme development will differ for each of these aims.
These examples would each require a programme with combinations of monitoring for
cyanobacterial cells, cyanotoxins and growth conditions. Frequently, cyanobacterial
monitoring will be connected to, or included in, other general purpose water quality
monitoring programmes.
10.1.2 Monitoring strategies
Monitoring water bodies can be facilitated by using a structured approach which may
significantly enhance efficiency of laboratory resource use, especially where resources
are limited (see Figure 10.1). Because many commonly occurring cyanobacteria are
more often toxic than non-toxic, the simplest approach is to assume toxicity and to
monitor cyanobacteria rather than their toxins in the water body. Such an approach
begins with simple visual inspection. If this indicates a possible cyanobacterial problem,
the approach moves on to assessing which level of cyanobacterial development can be
sustained by the nutrient concentrations available (i.e. the carrying capacity, which is
most frequently determined by total phosphorus, see section 8.1.) If nutrient
concentrations are high enough for cyanobacterial proliferation to be likely,
cyanobacteria must be monitored at time intervals adequate to identify hazards (see
section 7.5 for time intervals of monitoring).
Monitoring may be supplemented with valuable information by collection of historical

data on bloom occurrence and of health information, including veterinary records of
animal poisonings (see Chapter 6).
10.1.3 Variable selection
Easy-to-assess visual indicators (Step 1 in Figure 10.1) may provide valuable

information concerning cyanobacterial proliferation at low cost and often enable a high
frequency of observation, especially if they are periodically supported by microscopy in
order to ascertain that observed phenomena are due to cyanobacteria. If performed by
local staff with alert and flexible observation skills as well as increasing experience,
regular site inspection (including monitoring and recording of transparency,
discolouration and scum formation) can provide much information in relation to the effort
required. Including an assessment of land-based pollution in critical areas of the
catchment may also provide information on nutrient pollution sources.
Monitoring of variables which enhance cyanobacterial growth and/or accumulation is

valuable in recognising which water resources are at risk of bloom development and
scum formation. In many regions, this may also assist substantially in ruling out
resources unlikely to sustain major cyanobacterial populations. Monitoring of total
phosphorus as a key factor for mass developments may be particularly relevant for
setting priorities in monitoring recreational waters where short-lived minor surface scums
are less of a problem. Total phosphorus data can further provide the basis for planning
and assessing the success of measures to counter the causes of the problem (see
Chapter 8). The collection of further environmental data on hydrological conditions (such
as retention times and thermal stratification), light availability (as assessed by the
relation of depth of light penetration to mixed depth, see section 2.2) as well as dissolved
nitrogen (nitrate and ammonia) provides a basis for understanding why certain
cyanobacterial genera or species dominate.
Figure 10.1 Example of structured, quantitative investigation approach to
monitoring and surveillance for toxic cyanobacteria and their growth potential
Box 10.1 Limitations of assessing toxin risk from monitoring cyanobacterial populations
Although cyanobacterial identification and quantification provide a basis for estimating the toxic
risk, the monitoring of cyanobacterial populations alone cannot be used as a surrogate for
monitoring cyanobacterial toxins because:
• Toxin types and levels per unit cyanobacterial biomass can vary widely, ranging from
undetectable to present at acutely toxic levels.
• Toxin levels per cell vary widely between individual strains and blooms of the same species.
• Individual cyanobacterial strains of the same species can contain more than one type of toxin.
• The toxins can persist in water bodies and water treatment plants in extracellular (soluble) form
after release from the producer-cells upon cell lysis caused by biological, physical or chemical
agents.
It thus follows that cyanobacterial toxins can be present in a waterbody or treatment facility in the
absence of intact cyanobacterial cells or cell debris if the waterbody had recently contained
cyanobacterial cells.
Monitoring cyanobacterial taxa and population densities (cell numbers or biomass) can
provide an excellent basis for assessing risk, particularly if supported periodically with
toxicity tests or toxin analysis. If data on toxin content of prevalent cyanobacteria are
available for specific water bodies or regions, and have been found to remain fairly
constant for the predominant taxa, a tentative prediction of toxin risks can be inferred
from quantitative assessments of cyanobacterial taxa present using samples taken
between the occasions of toxin analysis (see Box 10.1 for the limitations of this
approach). Microscopy may also be used for monitoring finished drinking water prior to
distribution, as well as different treatment steps, for breakthrough of potentially toxic
cyanobacterial cells. The necessary laboratory equipment for monitoring cyanobacteria
is limited to a microscope and some accessories (see Chapter 12). The training
requirements for staff are lower than is frequently assumed (see Box 10.2).
In situations where the phytoplankton is largely dominated by cyanobacteria (e.g. they

constitute more than half of the biomass seen through a microscope), measurement of
the concentration of chlorophyll a can be used as an estimate of cyanobacteria present
(see Chapter 12). A simple photometer is adequate for this approach.
Monitoring for cyanobacterial taxa, cell numbers and/or biomass may provide data which
are of value for a variety of assessments, including:
• The occurrence, types, distribution and abundance of cyanobacteria, including

potential toxin-forming types, in natural and controlled water bodies, water treatment and
distribution systems and in potable water supplies.
• Spatial and temporal changes in cell populations, their composition, abundance and
integrity.
• Relations between cyanobacterial populations and the types and levels of

cyanobacterial toxins (if accompanied by toxin analysis), and associated water quality
problems and health incidents.
• Warning systems to trigger contingency action plans in the event of cyanobacterial

mass development in waters required for human or animal use.
• Responses of water bodies and water supplies to eutrophication control and strategies
to destroy and/or remove cyanobacterial cells in water treatment.
Box 10.2 Adequate training for identifying important cyanobacterial taxa
A common misunderstanding is the assumption that the sophisticated taxonomic training, on a

level adequate for ecological research, is necessary for practical monitoring of cyanobacterial
hazards. This may be intimidating for beginners and practitioners. In practice, it may only be
necessary to focus training on determining the taxa relevant in the region or waterbody to be
monitored, frequently only to the level of genera (e.g. Microcystis), and down to the level of
species only if these are easy to identify and if species differentiation is particularly important for
indication of toxin content (e.g. Planktothrix agardhii and Planktothrix rubescens). Basic
identification of cyanobacteria by local personnel should be supplemented by periodic quality
control by experts (see the Alert Levels Framework Level 2 in section 6.3.3) to ensure adequate
recognition of the important groups, especially after conditions in a water resource have been
changed and other taxa may have proliferated.
Monitoring cyanotoxin concentrations and/or assessment of toxicity may be warranted to

characterise the hazard presented by a given cyanobacterial population. The role of
such monitoring is different in hazard assessment for drinking water and for recreational
water use.
The increasing evidence of health outcomes due to unknown irritative cyanobacterial

components may discourage the use of recreational sites with high cyanobacterial
population densities for water contact-intensive activities, regardless of the
concentrations of known toxins (e.g. microcystins, neurotoxins, cylindrospermopsin).
Analysis of these toxins will not characterise the hazard of irritative effects (see section
5.2.2). However, monitoring of microcystins or other known toxins can assess whether
specific blooms present hazards from the cyanotoxins which are of greater concern for
public health. In particular, it may identify some blooms as not presenting high risk levels.
This information may be particularly relevant for heavily used sites, such as those
associated with holiday facilities and for which temporary closure might have a
substantial economic and social impact.
Monitoring of recreational sites should emphasise bloom prevalence and the potential of
bloom formation. Step 1 of the structured approach presented in Figure 10.1 is
particularly important here. This approach can easily be communicated to the general
public and to bathing site users, and can involve them in assisting with hazard
assessment (in addition to providing a basis for their own decision-making on water
contact activities) by encouraging reporting of scums or strong discolouration and
turbidity to authorities.
Monitoring toxin concentrations is especially important in drinking water supply systems

in order to detect toxins released from the cells into the water during treatment and to
determine the level of risk associated with a specific bloom in the water resource.
Monitoring for cyanobacterial toxins is necessary to provide data:
• On the occurrence, types, abundance and distribution of cyanobacterial toxins in

aquatic environments, water treatment and distribution systems, potable supplies and
food products which contain, or have been exposed to, cyanobacteria and their toxins.
• On relations between environmental conditions, cyanobacterial populations and

cyanobacterial toxins.
• For use in alert levels schemes and in the activation of contingency action plans.
• On relations between cyanobacterial toxins and water quality and health, with
reference to human and animal exposure levels and health effects.
• For the derivation of standards for drinking water quality, to enable compliance with
these values to be achieved and to determine, in the longer-term, whether such
standards remain appropriate or need to be changed.
10.2 Laboratory capacities and staff training

Monitoring for cyanobacterial health hazards makes a range of demands upon analytical
resources, some of which are different from those required by water quality monitoring
for other types of variables. An overview of the requirements for the monitoring
approaches discussed in section 10.1 and their respective information return is given in
Table 10.1. Proper interpretation of information concerning cyanobacteria and their
toxins requires expertise from the health and water resource sectors. However, these
areas of expertise occur within a single authority in a few countries. Multisectoral co-
operation is therefore important. Planning of monitoring programmes should generally
involve co-operation between the environmental and the health sectors, bringing in
further agencies or organisations where appropriate (e.g. drinking-water suppliers and
authorities responsible for tourism, for public education or for water management).
Table 10.1 Approaches to monitoring for cyanobacteria and analysis for cyanotoxins:
requirements and options for their organisation
Monitoring Demands on
Parameters/variables Who Where
type equipment and skills
Basic Minimal
Site inspection Transparency, Secchi disc, regular site Environmental or Local
for indicators of discolouration, scum inspection by trained health officers,
toxic formation, detached mat staff; skill requirement trained health
cyanobacteria accumulation basic, training easily staff or
in waterbody provided supervised local
Background Low to moderate
Potential for Total phosphorus, nitrate Photometer, boat, depth Environmental Local,
cyanotoxin and ammonia, flow sampler, Secchi disc, officers or regional
problems in regime, thermal submersible experts with
waterbody stratification, transparency temperature/oxygen limnological
probe; skills basic but expertise
require specific training
and supervision
Cyanobacteria Low to moderate
In waterbody Dominant taxa (quantity): Microscope, photometer Environmental or Local,
and drinking often determination to is useful; specific health officers regional
water genus level only is training and supervision (with occasional
sufficiently precise; is required, but quite quality control by
quantification only as easily achieved experts);
precise as needed for consultants with
management limnological
expertise
Toxicity Moderate
assessment
In waterbody Toxicity Demands on equipment Toxicologists Central
and drinking are low, but rather high
water on skills
Toxin Moderate to high
concentration
In waterbody Toxin concentration New methods with lower Skilled analysts Central
and drinking financial demands
water presently in
development for some
cyanotoxins (e.g.
immuno-assay); skill
requirements vary
widely from moderate to
very high
Monitoring for visual indicators of cyanobacteria focuses on critical site inspection and
requires almost no facilities. Training of staff is necessary, but not difficult and
experience leads to improved performance. However, much time in the field is required
and this can be reduced substantially by involving local people who have been given
specific training for visual inspection. Nevertheless, professional staff should exercise
periodic quality control over their work.
Environmental monitoring of chemical and physical variables indicating bloom-forming

potential, such as nutrient concentrations, hydrophysical conditions and transparency,
make limited demands upon analytical resources (a submersible temperature and
oxygen probe and a photometer with optical filters to provide the necessary wavelengths
are the most complicated instruments required). Such analytical capacities may be
readily decentralised. While laboratory analysis can be carried out by any capable
chemical laboratory, some limnological expertise is necessary for the planning of field
work, quality control of data and interpretation of results. The staff time required can be
reduced once seasonal patterns of variation are known and sampling regimes can be
adjusted to be most effective.
Health authority staff with experience in microscopy can learn to recognise the most
important toxin-producing cyanobacteria in the water bodies under their responsibility, if
training by experts can be provided (attention may need to be given to dampening the
general taxonomic enthusiasm of some experts, in order to concentrate on the skills
really needed for monitoring the cyanobacterial taxa relevant for the water resources in
question, see Box 10.2). Cyanobacterial identification and quantification can be
centralised or subcontracted because preserved samples can readily be transported
(see Chapter 11). However, the development of local skills is recommended because
this should enable more rapid identification of, and response to, current cyanobacterial
problems.
Cyanotoxin analysis with customised immuno- or enzyme assays, or toxicity tests with
simple bioassays (see Chapter 13), may make only moderate demands on equipment
and can be performed potentially by local health or environmental authorities. However,
these techniques require specific staff training and periodic quality control by comparing
results with those of more elaborate methods. More advanced programmes addressing
toxicity assessment and toxin analysis require developed analytical capacities and
exacting quality control. Even in countries with extensive advanced analytical facilities it
is unlikely that the demand for toxin analysis would justify establishment of widespread
or local facilities and some form of co-operation on a broader scale of centralisation is
therefore advisable. Methods for cyanobacterial quantification and toxin analysis should
be standardised and a system of official accreditation for analytical laboratories should
be implemented.
Several options may be available for conducting analyses of water samples. The agency
responsible for the monitoring programme may have its own laboratory or laboratories,
the facilities of another agency or of a government ministry may be available, or some or
all of the analytical work may be done under contract by a private laboratory. Some
analytical work may be done in the field using either field kits or a mobile laboratory.
Regardless of the options chosen, the analytical services must be adequate for the
volume of work expected. Furthermore, good communication between those planning
and performing field work and the analytical laboratory is crucial for ensuring appropriate
sample collection, preservation and transportation (see Chapter 11). Periodic quality
control is highly recommended particularly with respect to the handling of samples from
field to final analysis.
10.3 Reactive versus programmed monitoring strategies
Monitoring strategies can be regarded as either reactive or programmed, although these
are not necessarily mutually exclusive. A reactive strategy is needed when an
unexpected cyanobacterial bloom develops and affects, or has the potential to affect,
water supplies, recreational water and human health. It can be triggered by an
unanticipated bloom event, by health impairments reported to authorities and related to
cyanobacterial proliferation, by results of routine visual site inspection, or if routine
analysis in drinking water treatment facilities detects cyanobacteria or toxins in the raw
water intake or in recreational areas. This response strategy can include a range of ad
hoc assessments of cyanobacterial numbers, toxicity assessment and toxin analyses.
Programmed monitoring strategies are being applied increasingly to the investigation of
cyanobacterial population and toxin problems where cyanobacterial problems are
ongoing, occur regularly, are anticipated or have occurred in the past. These structured
programmes can provide additional preventative benefits by warning of necessary
actions before a developing cyanobacterial population presents an operational,
environmental, or health problem, or by triggering the implementation of preventive
measures at the source of the problem (see Chapter 8).
The benefit and information obtained from reactive as well as from programmed
monitoring strategies can be greatly enhanced by ensuring that samples of
cyanobacteria or water from natural or controlled environments are supplemented by
clinical observations and clinical samples in the event of associated human and animal
health incidents. This can be assisted by heightening the awareness of medical
practitioners, public health authorities and veterinarians through training and information
programmes (see section 7.4).
The potential for community involvement in monitoring strategies is high, provided that
adequate information is supplied using leaflets, publicity and educational campaigns
(section 7.4.2). "Algae Watch" programmes, or "scum scouting" to report on the
appearance of blooms and scums. Schools programmes and water sports associations
in Australia, the USA and the UK have provided useful information to monitoring
agencies, as well as having helped to promote community action and joint responsibility
for the causes and cures of cyanobacterial bloom problems.
Programmed monitoring strategies have the potential to detect and anticipate changes in
cyanobacterial populations and potential levels of toxins. Such strategies can thus
provide information to trigger appropriate contingency plans. Alert Levels Framework
(ALF) systems are finding application in some countries; these are systems of
programmed monitoring which incorporate action sequences in the event of warning
thresholds being exceeded. Alert Levels Frameworks may be used for the monitoring of
cyanobacterial populations only, or cyanobacterial populations plus toxins, depending on
monitoring objectives and resources (see section 6.3).
10.4 Sample site selection

The selection of sampling sites is a key factor in determining the value of the data to be
sought from the subsequent sample examination and analysis procedures. Sample site
selection should be tailored to meet the overall aims and objectives of the monitoring
programme (or even a single sampling visit). Thus site selection must consider and take
account of the following:
• The uses made of the water body must be considered (e.g. potable supply, recreation,
animal watering). For recreational use, sampling will include shoreline areas particularly
frequented by visitors and may focus on public bathing sites. It may also include offshore
sites where immersion sports take place. For drinking water resources, sites at or close
to the raw water intake are important, and sampling within a treatment plant might
include sampling at different treatment steps.
• If sampling aims at assessing the total population size of the cyanobacteria and their
scum-forming potential, or the nutrient concentrations which influence the maximum
possible population size, it should cover a central reference site in open, mixed water
(experience may indicate if this can be used as a representative site for the main water
mass). Selection of adequate depths must consider stratification of organisms and
nutrients using depth-differentiating or depth-integrating sampling techniques as
described in Chapter 11.
• Morphometric and hydrophysical characteristics of the water body (e.g. exposure to

wind or thermal stratification) may help identify sites which are prone to scum
accumulation. These factors are likely to influence the development and fate of
cyanobacterial populations and their subsequent location in parts of the water body.
• Current weather conditions, particularly wind direction, which lead to scum

accumulation along certain shorelines may require flexible choices of sites, particularly if
the aim is assessing the highest cyanotoxin concentrations by sampling maximum scum
densities.
• Specific incidents, such as animal deaths or human illness, if these are suspected to
be associated with exposure to cyanobacteria and toxins at a specific location in the
water body concerned.
• The history, if available, of cyanobacterial population development and occurrence of

toxins in the water body, because this information may indicate sites particularly likely to
harbour scums.
• Local logistical resources, accessibility and safety factors (e.g. Secchi transparency
should be measured from a pier or boat, offshore sampling requires access to a boat
and to a site for launching it, and sampling from steep shores or reservoir dams might be
dangerous).
• Potential local sources of nutrient pollution (e.g. inlets or slopes affected by erosion).
For adequate sampling site selection it is of critical importance to consider the location
and potential concentration of cyanobacteria in a water body, as described in Chapter 2
and in section 11.3.2. This site selection should account for accumulations of
cyanobacteria as scums (usually quite unevenly distributed over the area of a water
body), as subsurface maxima at some metres depth, as homogeneous distributions
throughout the mixed strata of the water body, or growing on the sediment surface (from
which they may become detached and driven onshore to present acutely toxic
accumulations). Possible scenarios are shown in Figure 10.2.
The heterogeneous and dynamic nature of many cyanobacterial populations presents

difficult problems for sample site selection. A flexible response to the current situation
when choosing the sampling sites may, at times, be more appropriate than following a
rigid programme. Alternatively, fixed sites always sampled within a broader monitoring
programme may be supplemented with the sampling of sites currently harbouring
cyanobacterial scums.
The horizontal and vertical heterogeneities in cell distribution are compounded by further
variability in cyanobacterial toxin levels and distribution. Although the toxins are largely
retained within the producer-cells during growth, they are released into the water during
cell lysis due to natural agents, some algicides and pressure-induced disruption.
Circumstances therefore arise where cyanobacterial toxins may be present in the
absence of intact cyanobacterial cells. Sampling site selection for dissolved toxins may
thus include locations such as the water close to decaying scums and in water treatment
works and distribution systems, when it is suspected that cyanobacterial breakdown may
have occurred.
Figure 10.2 Some locations of cyanobacteria in thermally stratified lakes or

reservoirs. R, recreational area; AP, water abstraction point; 1, shoreline scum of
planktonic cyanobacteria (often decaying); 2, planktonic cyanobacterial scum on
open water during calm conditions; 3, dispersed cyanobacteria in epilimnion; 4,
planktonic cyanobacteria on sediment; 5, upper mixed layer during autumn
overturn; 6, spring and autumn conditions of complete mixing and in summer in
shallow lakes in windy conditions; 7, scum under ice; 8, subsurface maximum of
planktonic cyanobacteria (not apparent at surface); 9, mats of benthic
cyanobacteria on sediment in shallow water; 10, shoreline accumulation of
detached benthic cyanobacteria (Modified from Lindholm et al., 1989 with
additions)
10.5 Monitoring frequency

Cyanobacteria generally have fairly slow growth rates compared with many other micro-
organisms. This helps to simplify monitoring frequency requirements. For those taxa
which do not form scums but are dispersed in the water, weekly or even two-weekly
monitoring intervals are often sufficient, even during their growing season, in order to
monitor population development and to assess the cyanotoxin hazard. However, the
ability of scum-forming cyanobacteria to change their concentration and position in the
water body within very short time spans of only a few hours poses a specific challenge to
the design of monitoring programmes.
Monitoring frequencies for cyanobacteria and cyanotoxins are suggested in the Alert
Levels Framework given in Chapter 6. For example, monitoring may begin on a
fortnightly or weekly basis and be increased to twice-weekly, or to daily, whilst alert
levels are exceeded, and then be reduced again after values decline below alert levels
and guideline values for cyanobacterial cells and toxins. The same principles should be
applied to the monitoring of recreational waters, whether they are used on a year-round
or seasonal basis (section 5.2.2). If a water body prone to cyanobacterial mass
developments is used for water-contact sports on a seasonal basis, or for a single event,
monitoring should begin not less than two weeks before the start of the season or the
event. As monitoring is continued, frequency may be adjusted to enable decisions to be
made on access to the facility throughout the season, or on whether to proceed with a
special event.
Structured approaches to monitoring (e.g. Figure 10.1) introduce nutrients as a further

variable for analysis. This may improve the information return for effort expended,
particularly if each step in the structure is monitored at an appropriate frequency.
Patterns of investigation may begin with longer time intervals and may be intensified as
cyanobacteria begin to proliferate. As knowledge and understanding of a given
ecosystem and the behaviour of its cyanobacterial populations accumulates, monitoring
frequencies can be optimised to meet the demands of the specific situation. For this
reason involving limnological expertise is particularly important in the planning of
monitoring programmes, in the evaluation of the data, and in periodic reassessment of
the adequacy of ongoing programmes. The following monitoring frequencies are
suggested for the structured approach given in Figure 10.1:
• Visual site inspection may begin at weekly or two-weekly intervals, which can be
increased to weekly or even more frequent intervals once cyanobacteria begin to
proliferate.
• Assessment of the carrying capacity for cyanobacteria in terms of nutrients (phosphate

and in nitrogen-limited systems, possibly also dissolved inorganic nitrogen) may be
undertaken less frequently in many situations. This depends on prevalent nutrient levels
and their rate of change. For example, in some water bodies, total phosphorus
concentrations may show little seasonal change, or they may always be far too high to
limit cyanobacterial biomass. In either case, occasional monitoring (in temperate
climates once in spring and once in summer) may be sufficient. On the contrary, if the
hydrological regime shows pronounced fluctuations, or if the total phosphorus
concentration oscillates around levels critical for limiting cyanobacterial biomass (0.03-
0.05 mg l-1 P), monthly or biweekly measurements may be necessary in order to assess
the carrying capacity for cyanobacteria. Often it will be advisable to begin a programme
with monthly sampling for one or two years. Evaluation of patterns may then enable a
reduction of sampling frequency to be justified.
• Assessment of cyanobacterial cell numbers or biomass can be affected by the rapid
changes discussed above, particularly if scum-forming taxa prevail. Knowledge of a
water body's carrying capacity for cyanobacteria, of the taxa typically occurring, and of
seasonal time patterns of their occurrence will help anticipate critical situations which
require increased monitoring frequency. If the aim of monitoring is to check compliance
with standards for drinking water or recreational use, and toxin levels in a given water
body are in the borderline range or above, sampling and analysis may be necessary
several times a week. Time patterns of water body use, particularly for recreation, may
be a useful further criterion for determining the time patterns for sampling.
• Toxin analysis may be necessary less frequently than assessment of cyanobacterial

cell numbers or biomass. Although toxicity of populations is variable, it does not appear
to change within a few days. Assessment of toxicity is particularly recommended when
situations in the water body change, e.g. when other taxa appear or if bloom lysis occurs.
Ideally, "real-time" information on the state of the cyanobacterial population and their
toxins is desirable for scum-forming taxa. Some approaches to meeting this demand are
currently being developed. The simplest approach is semiquantitative assessment of
cyanobacterial cell numbers through a microscope. This can be performed within an
hour or less of sampling, provided a microscope is available locally, and can be repeated
frequently during problem phases provided the laboratory is close to the water body
(which is often the case at drinking water supply reservoirs). Continuous automatic
provision of fluorescence data indicating pigment concentrations has become possible
by means of submersible flow-cell fluorescence spectrophotometers or by continuous
water flow through laboratory fluorimeters and cytometers. These procedures are
currently beginning to differentiate successfully between cyanobacteria and other
components of the phytoplankton. Installation of such devices may be especially
attractive for drinking water supply abstraction points, in order to adapt offtake levels to
the current location of cyanobacteria (if offtake depths are flexible), or in order to have
an immediate indication of the need to apply further treatment steps. Another approach
currently investigated for acquisition of real-time data for chlorophyll distribution and
levels, and potentially for cyanobacterial phycobiliprotein pigments in freshwaters, is
remote sensing of the optical properties of the water body by high resolution airborne
scanners (Cracknell et al., 1990; Jupp et al., 1994). However, flight times may be
infrequent and data collection depends on factors beyond human control, e.g. cloud
cover. Nevertheless, the remote sensing of cyanobacterial populations as a contribution
to water body management has excellent potential particularly for the monitoring of scum
locations. Remote sensing may become cost-effective for areas that have to monitor
many recreational sites.
Further aspects to consider in determining monitoring programme frequencies are, as

with site selection (section 10.4), the monitoring programme objectives, water-use,
specific incidents associated with exposure to cyanobacteria and their toxins, historical
evidence of blooms in the water body, local water body and catchment characteristics,
and the wider knowledge of cyanobacterial ecology. Available resources, sampling and
analytical logistics also need to be taken into account, for example, the time needed for
sample transportation, processing and analysis, and for the interpretation and reporting
of results.
Monitoring intervals should be timed to provide information for the following situations:
• To give warnings of developing cyanobacterial populations and toxin levels.
• On the duration of cyanobacterial populations and toxin levels which exceed guideline
values.
• On the decline of cyanobacterial populations and toxins due to natural processes or

the persistence or reduction in cyanobacterial populations and toxin levels due to
intervention, such as eutrophication control and water treatment.
In subtropical and tropical latitudes where appreciable cyanobacterial populations can

occur all year round, it may not be feasible or necessary to maintain high frequency
sampling programmes throughout the year, particularly if the population density is
subject to less change because there is little seasonal change in growth conditions. In
this event, sampling of the natural or untreated water may need to be at less frequent
intervals, e.g. monthly, with laboratory resources being directed towards higher
frequency monitoring of treated drinking water.
10.6 References
Bartram, J. and Ballance, R. [Eds] 1996 Water Quality Monitoring. A Practical Guide to
the Design and Implementation of Freshwater Quality Studies and Monitoring
Programmes. E & FN Spon, London, 383 pp.
Chapman, D. [Ed.] 1996 Water Quality Assessments. A Guide to the Use of Biota,
Sediments and Water in Environmental Monitoring. E & FN Spon, London, 626 pp.
Cracknell, A.P., Wilson, C.C., Omar, D.N., Mort, A. and Codd, G.A. 1990 Toxic algal
blooms in lochs and reservoirs in 1988 and 1989. In: Proceedings of the NERC
Symposium on Airborne Remote Sensing. Natural Environment Research Council,
Swindon, 203-210.
Jupp, D.L.B., Kirk, J.T.O. and Harris, G.P. 1994 Detection, identification and mapping of
cyanobacteria - using remote sensing to measure the optical quality of turbid inland
waters. Aust. J. Mar. Freshwat. Res., 45, 801-828.
Lindholm, T., Eriksson, J.E. and Meriluoto, J.A.O. 1989 Toxic cyanobacteria and water
quality problems - examples from a eutrophic lake on Åland, Southwest Finland. Wat.
Res., 23, 481-486.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 11. FIELDWORK: SITE INSPECTION AND

SAMPLING
This chapter was prepared by Hems Utkilen, Jutta Fastner and Jamie Bartram
Fieldwork, including site inspection, sample collection and, in most programmes, some
on-site analysis, determines to a large extent the quality of information obtained and
represents a significant proportion of the total cost of a cyanobacteria or cyanotoxin
monitoring programme. A well-designed and implemented fieldwork programme
enhances the quality of the data obtained and may assist greatly in containing overall
costs. The importance of careful programme design has been emphasised in Chapter 10,
together with the need for pilot testing to refine programmes and to assess logistics.
Properly trained field workers are the backbone of effective sampling and inspection
programmes; aspects of their professional development and training are described in
Chapter 7. Inclusion of members of the public as active participants in monitoring
programmes is rapidly gaining acceptance and can contribute significantly to the quantity
and quality of information obtained from a monitoring programme. Special interest
groups (such as non-governmental organisations and user associations) and also
concerned local persons in sensitive or affected areas can provide useful information.
Proper orientation and training, for example in site inspection, in flexible responses to
the results of inspections, and in conducting basic tests such as transparency, can assist
in providing valuable additional data.
Further information concerning planning and performing fieldwork is presented in Water

Quality Monitoring (Bartram and Ballance, 1996), a companion volume in this series.
11.1 Planning for fieldwork

The principal components of effective planning for fieldwork comprise:
• Timing and preparation (in the context of the monitoring programme plan and
information needs for management).
• Prior liaison with the laboratory that will receive and process samples.
• Logistic preparation.
• Local co-ordination as required.
• Prior liaison with other information recipients (local or central).
Preparation of the monitoring programme principally concerns the timing of visits and
careful cataloguing of sampling sites. The timing of sampling and inspection visits is
described in Chapter 6.
Before routine field visits are performed, a period of pilot testing should be implemented.
This will help ensure that time requirements for inspection and sampling are reconciled,
and that activities are planned to make the best use of staff time and other resources
(e.g. vehicles). Realistic estimation of travelling time is important to avoid exceeding
allowable sample storage times prior to analysis. Pilot testing should lead to the
development of a detailed inventory and description of sampling sites. If changes in
water quality with time are to be interpreted with confidence, samples must be taken
consistently from the same locations and/or from precisely identified locations. Pilot
testing also provides an opportunity for training personnel and allows their familiarisation
with particular aspects of the monitoring programme itself.
Co-ordination with the laboratory is an important aspect in determining the final value of
the sampling expedition. In some cases, the laboratory will be responsible for
preparation of sample containers and chemical additives for sample preservation, and
may also be responsible for the provision and maintenance of equipment for on-site
testing.
Laboratory capacity is an important area of concern which should be addressed in

programme design and in pilot testing. It is essential that the workload generated by a
sampling expedition is properly managed within the laboratory; it is therefore vital that
analysts know how many samples will be arriving, the approximate time of arrival and
the analyses that are to be carried out. Excessive delays before sample processing and
analysis may render the sample results invalid (and thereby useless) for the
management purposes for which they had been collected. Therefore, the timing of
sample delivery to the laboratory and the workload management within the laboratory
should be co-ordinated prior to fieldwork.
Good logistical preparation prior to fieldwork requires that equipment is checked to

ensure that it is functioning properly (e.g. electrodes tested and calibrated, batteries
charged). The correct number, size and type of sample containers must be prepared and
transport must be arranged, ensuring permission and local access to any restricted sites.
It is essential to prepare a sampling checklist which includes maps of sampling site
locations, a list of equipment required and a detailed explanation of the methods for
sample collection. The checklist should also include lists of the types and numbers of
samples to be taken at each site, as well as of the required volumes. Good preparation
for sampling involves previous labelling of sampling containers with at least the site, date
and depth from which the sample is to be taken.
11.1.2 Frequency of field visits
The frequency of site inspection and sampling must be adapted to the local situation as
described in Chapter 10. Key criteria are:
• Potable water supply reservoirs may need to be monitored regularly throughout the
year if perennial persistence of cyanobacteria cannot be ruled out. This applies
particularly to warm climates and in temperate zones to water bodies populated by
certain taxa such as Planktothrix.
• Monitoring of recreational lakes in tropical countries may cover the whole year, while in
temperate zones it can be focused on the warm season from early summer to autumn.
• The frequency of site inspection and sampling should be increased during

development of cyanobacterial populations or when persistent blooms occur.
11.1.3 Safety
Caution and attention are appropriate while working with cyanobacteria, particularly
when they are highly concentrated in scums. It is wise to treat all blooms as highly toxic.
Contact with water should be minimised during sampling and gloves and rubber boots
should be worn because cyanobacteria might contain toxins and could also have a high
allergenic potential.
In some areas of the world other water-based hazards such as schistosomes (the cause
of schistosomiasis or bilharzia) may also be present. In such circumstances water
contact should be minimised and following contact the skin should be vigorously dried.
Caution during fieldwork should also apply during the use of boats or other vessels and
whilst wading, especially in waters with low transparency where underwater hazards
may not be readily visible.
11.2 Site inspection

Sites used for drinking water abstraction or recreation should be subject to programmed
inspection by trained professional staff, and preferably in conjunction with sampling
expeditions. Careful inspection can assist interpretation of results from sample analysis.
Moreover the development of personal expertise in relation to specific water bodies can
provide the best form of early warning system.
A protocol for site inspection should be established (see Box 11.1) that includes the
ambient data of temperature, wind and other weather conditions as well as an estimate
of the situation during the previous 24 hours.
High nutrient input from, for example sewage outlets or run-off from excessively fertilised
areas favours the development of cyanobacteria (Chapter 8). Site inspection protocols
should also address the identification of sources of nutrient input and significant land
uses, as well as land use changes that may assist in the interpretation of findings. Such
assessment is not necessary during every sampling tour but is particularly
recommended during the pilot phase, and at moderate intervals, such as annually, when
changes in the catchment area are suspected.
Box 11.1 Example protocol for site inspection and follow-up
1. Note ambient and weather conditions and, if possible, also those conditions during the past 24
hours, especially wind direction and velocity.
2. Assess the areas most likely to be affected by cyanobacterial blooms first, i.e. the downwind
shores.
3. Determine if:
• The bottom of the lake is clearly visible at approximately 30 cm depth along the shore line.
• Note any distinct green or blue-green discolouration of the water; if a Secchi disc is available,
note the transparency.
• Note if cyanobacteria can be seen as green or blue-green streaks on the surface, or as

accumulations in bays and along shorelines.
• Note whether green or blue-green scums affect large parts of the water surface.
4. If cyanobacteria are present (according to Item 3 above):

• Initiate monitoring, if not already in place.
• Initiate an inspection of the catchment area for sources of nutrient inputs.
• Initiate temporary intensification of monitoring, if necessary, for safeguarding healthy use for
drinking water or recreation (see Chapter 10).
5. If heavy blooms or scums are observed, immediately:

• Inform other parties concerned (water suppliers, health authorities, operators of recreational
sites).
• Inform public and consider posting warning notices at bathing sites, and intervening against use
for water contact sports (see sections 5.2.2 and 6.2.2).
6. If cyanobacteria or dense algal growth is a problem, check whether nutrient pollution sources
are apparent, or whether a specific catchment inspection tour should be initiated (see Chapter 8).
When scums appear on the water surface, cyanobacteria may be present in densities
hazardous to human health, and thus appropriate responses should be initiated quickly
(see sections 11.3 and 6.4) and samples for further analysis should be taken. Sampling
of scums outside designated or habitual bathing sites is also of great value for
determining and predicting further risk, e.g. for when wind directions change during the
following days.
11.3 Sampling
Sampling may address both cyanobacterial population development and hazardous
accumulations. Informed on-site decision-making is necessary to refine sampling
programmes. Samples addressing population development as a basis for assessing the
potential for hazardous concentrations are usually taken at one or several points in the
water body. These points should be representative for the whole water body and are
often where it is deepest. For assessing hazards, samples must also be taken in areas
where accumulations of cyanobacteria can affect both humans and livestock, or they
should be taken at the raw water intake of drinking water reservoirs. The potential for
spatial heterogeneity demonstrated by some species (see section 2.2), i.e. horizontal
and vertical variations in both cell numbers and toxin content, must also be considered
when selecting the number and location of sampling sites.
Sample collection should always be accompanied by a site inspection (see section 11.2)
because the data obtained during the inspection will be important in the interpretation of
the results of sampling.
Sample collection and storage procedures differ depending on the type of analysis which
will be carried out. The three principal categories of analysis usually performed are:
• Nutrient analyses (phosphorus and nitrogen).

• Cyanobacterial identification and quantification.
• Cyanotoxin analysis, e.g. toxicity testing and analysis of cell-bound and dissolved toxin.
11.3.1 Sample containers
Containers and bottles for the transport of samples should ideally be provided by the
laboratory that will conduct the analyses. This helps to ensure that they are of a suitable
volume, are properly prepared and that due consideration has been given to the need for
pre-treatment and chemical addition. Field work is easier if bottles are pre-labelled and
well-arranged in suitable containers (in climates where storage in insulated containers is
not necessary, soft-drinks crates with subdivisions for each bottle are cheap and
practical). For routine sampling of the same sites, it is advisable to always use the same
bottle for each site and each parameter. This avoids cross-contamination, which is a
particular concern for phosphorus analyses. For most samples, glass bottles are
appropriate but often plastic containers can be used that are considerably lighter and
unbreakable. It should be decided in advance whether it is more practical to subdivide a
water sample into aliquots for each analysis prior to transportation, or whether a single
sample can be divided on receipt in the laboratory. The following containers are
recommended for the transport of cyanobacteria and related samples:
• Phosphorus analysis. Use 100 ml glass bottles pre-washed with and stored containing
sulphuric acid (4.5 mol l-1) or hydrochloric acid until usage. Phosphate is indicative of the
potential for cyanobacterial growth even when it is at very low concentrations (µg l-1) and,
therefore, special care must be taken to avoid contamination of samples. Contamination
may arise from phosphate-containing detergents or from previous storage of samples
with very high phosphate concentrations. Phosphates are easily adsorbed to glass
surfaces and may be released later when the bottle is filled with a new sample with low
concentration (see Chapter 12).
• Nitrate, ammonia and total nitrogen. Use clean 100 ml glass or polyethylene bottles.
• Microscopic identification of cyanobacteria. Wide-mouth polyethylene bottles are
appropriate for studying living material in a fresh grab or net sample (see below).
• Microscopic identification and quantification of cyanobacteria. Brown glass bottles (100

ml) are preferable. These should already have about 1 ml of Lugol's iodine or
formaldehyde solution added, or the preservative may be added immediately after filling
the bottles with the sample (see below). Clear glass bottles may be used but must be
stored in the dark because sunlight destroys iodine, thereby allowing the sample to
degrade.
• Cell material for analysis of cell-bound toxin concentrations. Use 1 litre (minimum
volume) containers, preferably made of glass although good results have also been
obtained with plastic. Containers should be pre-cleaned to prevent contamination.
• Cell material in large amounts (usually enriched with a plankton net) for structural
identification or some toxicity assays. Use plastic containers with wide necks to facilitate
filling. If samples are to be freeze-dried, the sample must be frozen in a layer not thicker
than 1-2 cm. Specimen containers (100 ml) for urine samples (easily obtained from
medical suppliers) are particularly suitable. Well-sealed, heavy-duty household plastic
bags can also be used, but care must be taken to avoid puncture because highly
concentrated cyanobacterial material can present a safety hazard. Bags should be
placed in watertight ice boxes for cool storage and as a precaution against leakage.
• Dissolved cyanotoxins. Use 1 litre (minimum volume) containers, preferably made of

glass although good results have also been obtained with plastic. Containers should be
pre-cleaned to prevent contamination.
• Chlorophyll a analysis. Brown glass bottles of 1 litre capacity are recommended to

avoid degradation of chlorophyll by sunlight. Clear bottles may be used if the samples
can be stored in the dark.
Preparation of Lugol's iodine solution for preserving phytoplankton samples Dissolve 20

g of potassium iodide in 200 ml of distilled water, mix thoroughly and add 10 g of
sublimated iodine. The solution must not be supersaturated with iodine because this can
result in crystal formation which interferes with cell counting. Supersaturation can be
tested by diluting 1 ml of stock solution to 100 ml with distilled water to give
concentrations similar to those used for preserving samples. If iodine crystals appear
after standing, more potassium iodide (approximately 5 g) should be added and the test
repeated. If no crystals appear, 20 ml of glacial acetic acid is added. Use about 1 ml of
Lugol's iodine to preserve 100 ml of phytoplankton sample. The resultant sample should
be the colour of whisky).
Dense scum samples may rapidly consume the iodine from the Lugol's solution. If
samples lose the whisky colour, and particularly if they no longer smell of iodine, they
are likely to decay. Samples stored for more than 10 weeks should be periodically
checked for sufficient preservation; further drops of Lugol's solution may be added if
necessary. If substantially more than 1 ml per 100 ml are required, the volume of Lugol's
solution added to the sample should be determined because the sample volume will
have changed appreciably and a corrective calculation is required for the cell counts.
11.3.2 Types of sample
Two different types of sample can be taken from lakes or similar surface waters. The
simplest, a "grab" sample, is a discrete volume of water taken at a selected location,
depth and time. Grab samples are also known as "spot" or "snap" samples (see Bartram
and Ballance (1996) for different sampling devices and a description of available
methods). In contrast, composite or integrated samples are made up of several sub-
samples from different parts of the water body. These are aimed at representative
sampling of a water body. Whereas grab samples are suitable for analysing situations at
specific sites (e.g. maximum density of cyanobacteria or cyanotoxins at a bathing beach),
composite samples are preferable for assessing the water body's total content of a
substance (e.g. total phosphorus potentially available for phytoplankton growth) or the
population of an organism (e.g. the size of a cyanobacterial population). Composite
samples are particularly important if the variables to be assessed are unevenly
distributed. If knowledge of the precise distribution is required, each sample can be
evaluated individually. However, the integration of samples prior to analysis is often far
more cost-effective.
Depth-integrated samples may be obtained either by continuously sampling the total

column of water from the surface to just above the sediment, or by discontinuously
taking grab samples from representative depths and then mixing them together. The
latter is particularly appropriate for deep lakes. However, in order to choose
representative depths and to achieve meaningful integration, knowledge of thermal
stratification of the water body is necessary.
Figure 11.1 Vertical distribution of chlorophyll a and a toxin (desmethyl-

microcystin-RR) from the species Planktothrix agardhii in Lake Östra Kyrksundet
in relation to hydrophysical conditions on 6 July 1988. Light intensity is given as a
percentage of surface intensity (Reproduced from Lindholm and Meriluoto, 1991,
Can. J. Fish. Aquat. Sci., 48, with permission)
The thermal stratification of a water body is influenced by the morphology of lakes and
reservoirs, the latitude, weather conditions and the physical nature of the water. It can be
determined by measuring vertical profiles of temperature within the water body. Where
thermal stratification occurs, it results in a water body functioning as two separate
masses of water (the epilimnion and the hypolimnion) with different physicochemical
characteristics and cyanobacterial populations, and with a transitional layer (metalimnion)
sandwiched between (Figures 11.1 and 11.2). In temperate climates, thermal
stratification generally occurs seasonally in water bodies of appropriate depth, whereas
in tropical climates it often follows diurnal time patterns. Thermal stratification has
important implications for the depth at which cyanobacteria are likely to be found (some
species may accumulate on the surface, some in the metalimnion, see also section 2.2),
as well as for concentrations of nutrients and interpretation of phosphorus and nitrogen
data. Usually, shallow (2-3 m), wind-exposed lakes are unstratified, whereas in
temperate climates deeper lakes usually exhibit a stable stratification from spring to
autumn. Lakes of intermediate depth (e.g. 5-7 m) may develop transient thermal
stratification for a few calm and sunny days, and which is then disrupted by the next
event of rain or wind.
Figure 11.2 Derivation of representative depths for taking discontinuous samples

and derivation of the volume of each sample
Thermal stratification usually results in inhomogeneous distributions of oxygen, nutrients,

and populations of algae, cyanobacteria and other organisms. However, even when
temperature is uniform throughout depth, stratification of organisms may develop on
calm days. Depth gradients of oxygen concentration and pH are good indicators of this.
Depth-integrated samples are more adequate than surface grab samples for the
assessment of population size and nutrient concentration in such homothermous
situations.
In deeper lakes or reservoirs with thermal stratification, depth-integrated sampling of
representatively chosen depths is necessary - taking samples evenly spaced over depth
may not yield fully representative results. When background information on the typical
stratification characteristics of a given lake is available (e.g. from a temperature profile
previously taken with a probe), sample numbers can be reduced by selecting adequate
depths to represent specific strata. If depth intervals are unequal and samples are to be
integrated, the volume of each sub-sample must be chosen to represent the actual
fraction of the vertical gradient it represents (see Figure 11.2).
Figure 11.3 Simple tube devices for taking depth-integrated samples.
A. From shallow lakes or water layers (< 5 m);
B. From deeper layers (< 20 m)

Continuous depth-integrated samples are often quite adequate for shallow and
moderately deep water bodies. They can be obtained using a water pump (submersible
pumps are available which allow sampling at depth) attached to a garden hose which is
operated at a steady pumping rate while the water inlet is drawn upwards between the
desired depths at a uniform speed. The sample obtained is therefore representative of
the water column. A very simple hose-pipe sampler may be used to depths of 30-35 m.
This sampler is made of a piece of flexible plastic piping (a garden hose will do) of about
2 cm diameter and appropriate length; the pipe is weighted at one end and open at both.
The weighted end, to which a cord is attached, is lowered slowly into the water so that
when the tube is fully extended it encloses a columnar section of the water. Before
hauling in the lower end with the attached line, the upper end should be closed in order
to avoid loss of water once the lower end emerges from the surface. Hoses need to be
stored clean and dry between sampling trips to avoid microbial growth inside the tube.
A simple depth-integrating pipe or tube sampler for shallow water columns (5 m deep) or
for the surface layers of deeper water bodies is shown in Figure 11.3A. By combining
tube segments as described by Sutherland et al. (1992) the sample may be extended to
a total length of up to 20 m (Figure 11.3B).
Sampling scums
Scums of cyanobacteria often occur near shorelines at low water depths, and therefore
working with a grab-sampler or a plankton-net may be difficult. Sampling scums is
carried out more easily with a wide-necked plastic or glass container. When sampling
scums, a suitable approach must be developed which allows for their observed
heterogeneous density. Two samples can be useful for different purposes. One sample
can be aimed at assessment of the maximum density of cyanobacteria or highest toxin
levels by taking a sample from where the scum is thickest (move the bottle mouth along
the surface to collect the dense mats of buoyant cyanobacteria). The other sample can
aim to simulate conditions where shallow waters are mixed by bathers and playing
children (agitate the scum before submerging the bottle). Both types of sample can be
used for comparison.
Plankton net sampling
Sampling of cyanobacteria with a plankton net is mainly carried out when large
quantities of cell material are required (for example for toxicity testing or for extended
chemical analysis) or when only a qualitative analysis of the phytoplankton is necessary.
In contrast to the use of depth samplers which quantitatively trap all of the particles in a
defined volume, the water volume filtered through a plankton net cannot be determined
precisely. Calculations based on the area of the net opening and on the length and the
distance it has been hauled are not recommended because they strongly overestimate
the amount of water actually filtered (due to clogging of the pores in the net, only a
fraction of the water volume actually passes through the net).
The depth at which the plankton net is deployed is dependent on the taxa of algae
and/or cyanobacteria present. Floating cells (e.g. Microcystis, Anabaena) are harvested
within the upper metres of the water column, while the sampling of well mixed or
stratified water bodies with distinct depth distributions of cyanobacteria (e.g. Planktothrix,
see Figure 11.1) may include deeper water layers.
11.4 Nutrients, cyanobacteria and toxins

11.4.1 Nutrient analysis (phosphorus and nitrogen)
Mass developments of cyanobacteria are associated with high nutrient concentrations.

Phosphorus is usually the key nutrient controlling proliferation, although the availability of
nitrogen may be an important variable to assess because it can influence whether or not
nitrogen-fixing species dominate (see Chapters 2 and 8).
Total phosphorus (from unfiltered samples) determines the capacity of a water body to
form cyanobacterial blooms. If dissolved phosphate (soluble reactive phosphate (SRP)
determined from filtered samples) is detected at concentrations of only a few
micrograms per litre, cyanobacterial growth and biomass are not limited by phosphate
(see Chapter 8). Because such low concentrations are critical for interpreting the
situation, sample contamination (e.g. through a few micrograms of dissolved phosphate
leaching from contaminated sample bottles, or through the release of phosphorus by
degradation of organic material during sample transport) may very easily lead to
misinterpretation. In contrast, samples for total phosphate analysis are less sensitive. If
SRP is important in a sampling programme, rapid filtration in the field or even on the
boat may be desirable, particularly in warm climates.
Transformations between nitrate and ammonia may occur, if samples are not stored
properly (cooled) and analysed rapidly (within 24 hours).
11.4.2 Identification and quantification of cyanobacteria
Cyanobacteria tend to accumulate in layers (at the surface or bottom) of sampling

containers. When samples are integrated or when sub-samples are taken, the sample
must be mixed immediately before sub-sampling.
In addition to the samples preserved with Lugol's solution, fresh unpreserved samples
are useful to aid microscopic identification, because Lugol's solution masks the
characteristic colours of the cyanobacteria, making it more difficult to recognise some
species. Unpreserved samples can be qualitative and are easily obtained by hauling a
plankton net (10 µm mesh) through the water column, or by taking a grab sample if the
density of cyanobacteria is high. Unpreserved samples for identification may be stored
for up to 24 hours, provided ambient temperature and light conditions are kept similar to
those in the field. If longer storage is necessary, preservation with formaldehyde may be
useful.
11.4.3 Samples for chlorophyll a analysis
When collecting samples for chlorophyll a determination it should be born in mind that
this pigment will start to degrade rapidly after collection; therefore it is essential to keep
storage times to a minimum. Filtration at the sampling site and storage of the filters in an
ice box is recommended for warm climates. If samples are to be filtered and analysed in
the laboratory, they should be transported as soon as practically possible (e.g. within 4
hours) and processed immediately on arrival.
Apparatus
• Measuring cylinder
• On-site filtration equipment, including vacuum hand pump and filter cup
• Glass fibre filters, GF/C, 47 mm
• Plastic Petri dishes and/or aluminium foil
Procedure
1. Thoroughly mix the sample and pour immediately into a measuring cylinder. Pass the
measured sample through the glass fibre filter with the aid of a hand vacuum pump. If
samples are dense, begin by filtering a small volume. If this is easily filtered, add more
sample, mixing every time a sub-sample is poured from the bottle. Use as much sample
as will easily pass through the filter leaving a distinctly visible greenish layer on the filter.
Record the volume of water filtered.
2. Once filtration is complete, carefully remove the filter from the filtration apparatus and
either place it in a Petri dish or wrap it in aluminium foil.
3. Fold the filter with the cells innermost because this both protects the pigment from
light and also prevents the sample from being dislodged from the filter surface.
4. Keep filters at near freezing temperatures and in the dark prior to their delivery to the
analytical laboratory. Deep-freeze filters if storage of several hours or more is necessary.
11.4.4 Bulk samples of cyanobacteria for cyanotoxin detection

For some purposes, it is important to collect a large amount of cell material using a
plankton net as described in section 11.3; toxicity tests for example require large
quantities because bioassays (e.g. mouse test) are less sensitive than chemical analysis.
Large amounts of cell material may also be needed for some detailed chemical analyses
(see Chapter 13). If the results are to be related to the dry weight of cellular material, a
quantitative sample (i.e. sampling all particles of a defined water volume) is not
necessary.
For later expert identification of the cyanobacterial taxa, and particularly for relating toxin
concentrations to cell numbers or biomass of specific taxa, a subsample for cell counts
may be taken directly from a well-mixed net sample before it is frozen. It is important to
realise that plankton net hauls accumulate particles selectively (enriching large cells and
colonies more than small or thin ones) and are therefore likely to contain a different
quantitative distribution of taxa than the sample used for assessing the composition of
cyanobacteria (as described in 11.3.2). Such subsamples for microscopy should be
preserved in 10 per cent formaldehyde or Lugol's solution in a separate container; small
volumes (e.g. 10 ml) are sufficient.
High biomass densities in such concentrated samples may rapidly consume the
preservation capacity of Lugol's solution. If not analysed within several days, samples
should be periodically checked for their preservation capacity; if they no longer show a
brown colouration, more Lugol's solution should be added. Dilution of the sample may be
useful to avoid rapid consumption of preservative (i.e. 1 ml net plankton sample, 8 ml
water, 1 ml Lugol's iodine solution or formaldehyde).
11.4.5 Quantitative analysis of cell-bound and dissolved toxins
For recreational sites and for assessment of the total toxin concentration, cell-bound
toxins are of primary importance. The concentration of dissolved toxin in the water is of
special interest for drinking water reservoirs and where large amounts of cyanotoxin
such as anatoxin-a are released into the water (Bumke-Vogt et al., 1996). Large
volumes may be required for dissolved toxin analysis (e.g. 1 litre) and this should be
checked with the analytical laboratory. If the concentration of cell-bound and dissolved
cyanotoxin is to be related to the water volume from which the cells were collected, a
defined volume must be filtered. Several methods have been proposed (Coyle and
Lawton, 1996; Gjølme and Utkilen, 1994) and are described in detail in section 13.1.
Essentially samples are handled in a similar manner as those prepared for chlorophyll a
analysis. The filtration step can be readily performed at the sampling site but with the
following modifications:
• Glass fibre or membrane filters may be pre-weighed so that the amount of material
retained (i.e. dry weight of cells) can be determined. Thus, toxin concentration can be
related both to dry weight and to water volume.
• The filtrate can be collected, placed in a glass bottle and processed as described in
Chapter 13 to enable the dissolved cyanotoxin concentration to be determined.
• Filters and filtrate should be kept cool.

Laboratory studies have shown that the microcystin content of material collected and
dried on filters is unaltered for several months, when the filters are stored in the dark at
room temperature and low humidity. So far this method has been shown to be useful for
microcystin as well as for anatoxin-a.
11.5 On-site analysis

Analyses for many variables can be carried out in the field. A significant advantage of
on-site testing is that tests are carried out on fresh samples that have not been
contaminated or otherwise changed as a result of storage. Variables generally analysed
on-site are:
• Transparency (with a Secchi disc). Cyanobacteria may occur as floating streaks or

mats on the water surface (see Chapter 2) making it difficult to obtain representative
transparency data (depending on the measuring site values can vary from 0.2 to 2 m). It
may be useful to determine the transparency in areas without floating cells as well as
within scums. The Secchi disc has to be lowered very carefully so as not to destroy the
formation of accumulated cyanobacterial cells, and before taking the measurement the
surface scums should be given time to return to their original position (i.e. their position
prior to disturbance by the Secchi disc).
Bathers can easily perform an improvised transparency determination while standing
knee-deep in the water and being careful to avoid suspending the sediment. If greenish
turbidity obscures their view of their feet, a significant cyanobacterial bloom could be
present corresponding to a level at which some authorities advise against bathing (see
Chapter 5).
• Temperature (with a probe lowered into the water, or a thermometer reading taken
immediately from the sampling device).
• Dissolved oxygen (electrometrically with a probe or chemically with the Winkler

method).
Further details of the methods mentioned above are available in Bartram and Ballance
(1996).
11.6 Field records

As with every sampling programme, careful field records are of critical importance for
interpretation of results (see Bartram and Ballance, 1996). Field records involve pre-
prepared lists and questionnaires, but which also leave plenty of space for additional
comment and for the recording of observations. Such observation should include:
• Presence of scums.
• Weather conditions on the day of sampling and, if available, an indication of conditions

over the previous 2-3 days, because this affects the accumulation of cyanobacteria.
• Any additional or unusual observations, such as scum distribution, estimation of

numbers of people bathing despite the presence of scums, defect warning or prohibition
signs regarding bathing in scum areas, suggestions for improvement of the sampling
programme, previously undetected inlets, changes in agricultural practices (such as
previously unknown distribution of manure) or other use, etc.
11.7 Sample preservation and transport

As has been emphasised in several sections earlier in this chapter, samples must be
clearly labelled with sampling site (station), date and time of sampling and depth of
sampling. In general, samples should be stored cool and dark in a storage box, taking
into consideration the following criteria:
• Cool storage is most important for all samples to be filtered because biological activity
may enhance shifts between particulate and dissolved fractions of the variable to be
analysed (biological activity is strongly enhanced by high temperatures). This is
particularly important for nutrient samples, chlorophyll a samples and samples for cell-
bound cyanotoxin analysis (whether collected on filters or with a plankton net). If storage
time exceeds 2-3 hours, particularly in warm climates, cooling during sampling and
transportation is recommended.
• Unpreserved samples for identification of cyanobacteria, fixed samples for determining

cyanobacterial cell numbers and biomass, and samples fixed with Winkler's reagents for
oxygen determination should be stored at ambient temperatures but protected from
direct sunlight.
• Storage times should be kept as short as possible, generally in the range of a few
hours. Storage times over 24 hours should be avoided. Specific descriptions of
preservation and transportation of samples for toxicity testing or toxin analysis of
cyanobacterial cells are given together with the sampling methods and in Table 11.1.
Table 11.1 Overview of sampling methods for the determination of cyanobacteria,

toxicity testing and toxin analysis
Purpose Equipment Preservation Transport Advantages Disadvantages
Determination of cyanobacteria
Qualitative Plankton net Unpreserved, Cool and Formaldehyde Caution necessary
(10 µm); or dark does not when using
depth formaldehyde discolour the formaldehyde
sampler1 solution sample
Quantitative Depth Lugol's Cool and Lugol's solution Colour of algae
sampler1 solution dark enhances changed by Lugol's
settling in solution
counting (identification
chambers difficulties for less
experienced staff)
Toxicity tests and toxin analysis
Qualitative, Plankton net Must be frozen Cool up to Provides large No relation to water
(large (25-50 µm within 24 24 hours, amount of cell volume; sampling
amounts) mesh) hours, freeze- otherwise material for efficiency less than
drying frozen or toxicity testing 100%; selective
freeze- and toxin
dried analyses
Quantitative
1. Cell- Depth Directly frozen Cool up to Cheap, quick, No large amounts of
bound toxin sampler1; until analysis 24 hours, easy to send; cell material for
Filter hand vacuum or air dried to or send direct relation to further analysis;
method pump or send to dried filters volume; biomass amount of toxin on
filtering laboratory estimate filters may be below
device; glass possible with detection limit;
fibre or pre-weighed caution required in
membrane filters humid climates with
filters dried filters
Plankton net Plankton net Must be frozen Cool up to Provides large Selective sampling
with defined (10 µm mesh) within 24 24 hours, amount of cell with a net
water graduated hours; freeze- otherwise material for toxin
volume bucket drying frozen or analysis and
freeze- toxicity testing:
dried relation to water
volume
2. Dissolved Depth Frozen Cool up to Filters and
toxin sampler1; 24 hours, filtrates can be
hand vacuum otherwise obtained in one
pump or frozen step
filtering
device; glass
fibre or
membrane
filters
1
Depth samplers include Ruttner, Van Doom, hose sampler or water pump
11.8 References
Bartram, J. and Ballance, R. 1996 Water Quality Monitoring. A Practical Guide to the
Design and Implementation of Freshwater Quality Studies and Monitoring Programmes.
E & FN Spon, London, 383 pp.
Bumke-Vogt, C., Mailahn, W., Rotard, W., Chorus, I. 1996 A highly sensitive analytical
method for the neurotoxin Anatoxin-a and first application to laboratory cultures.
Phycologia, 35 (6, Supplement), 51-56.
Coyle, S.M. and Lawton, L.A. 1996 Development of an extraction procedure for the
quantitative analysis of microcystins in cyanobacterial cells. Phycologia, 35(6,
Supplement), 57-61.
Gjølme, N. and Utkilen, H. 1994 A simple and rapid method for extraction of toxic
peptides from cyanobacteria. In: G.A. Codd, T.M. Jeffries, C.W. Keevil and E. Potter
[Eds] Detection Methods for Cyanobacterial Toxins. Royal Society of Chemistry,
Academic Press, London, 168-171.
Lindholm, T. and Meriluoto, J.A.O. 1991 Recurrent depth maxima of the hepatotoxic
cyanobacterium Oscillatoria agardhii. Can. J. Fish. Aquat. Sci., 48, 1629-1634.
Sutherland, T.F, Leonard, C. and Taylor, F.J.R. 1992 A segmented pipe sampler for
integrated profiling of the upper water column. J. Plankton Res., 14, 915-923.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 12. DETERMINATION OF CYANOBACTERIA IN

THE LABORATORY
This chapter was prepared by Linda Lawton, Blahoslav Marsalek, Judit Padisák, Ingrid
Chorus
Identification and quantification of cyanobacteria in water resources is the principal

component of cyanotoxin monitoring programmes and can provide an effective early
warning system for the development of potentially toxic blooms. Data on concentrations
of total phosphate, nitrate and ammonia are valuable for assessing the potential for
cyanobacteria to develop and whether or not nitrogen-fixing species are likely to occur.
Whereas methods for these nutrients have been extensively reviewed and internationally
harmonised by the International Organization for Standardization (ISO), approaches to
the species determination and quantification of cyanobacteria are very variable and can
be undertaken at different levels of sophistication.
Rapid and simple methods can be employed to analyse the composition of a sample at
the level of differentiation by genera (rather than species), which is often sufficient for a
preliminary assessment of potential hazard and for initial management decisions.
Further investigation may be necessary in order to address quantitative questions, such
as whether cyanobacteria are present above a threshold density. Rapid quantitative
counting methods can give useful estimates of cell numbers with a counting effort of less
than one hour per sample (sometimes within minutes), and the bulk method of biomass
estimation by chlorophyll a determination can be very time-effective with only moderate
equipment demands. More detailed taxonomic resolution and biomass analysis is
required if population development or toxin content needs to be predicted. Distinction
between these approaches is important because management must decide how
available staff hours are most effectively invested. In many cases, the priority will be
evaluation of a larger number of samples at a lower level of precision.
The choice of methods further requires informed consideration of sources of variability

and error at each stage of the monitoring process, particularly with respect to sampling
(see Chapter 11). Water bodies with substantial temporal and spatial variation of
cyanobacterial cell density may show several-fold deviation in cell numbers between
samples taken within a few minutes or within 100 m, and precise determination of
biomass in one sample per week therefore will not produce a basis for assessment of
population size. Much better information can be gained by investing the same effort into
a less precise evaluation of a larger number of samples (e.g. 10 samples taken at
intervals of 100 m, or every day). Information return on working time investment can
further be optimised by regular intralaboratory calibrations of methods and their quality
control by comparing results with the rapid methods to results of elaborate and precise
methods.
This chapter describes methods for cyanobacterial determination and quantification at

different levels of accuracy. For determination of the concentrations of key nutrients
which control cyanobacterial biomass and species composition, the standard
international methods developed by ISO are also reviewed.

Consideration of the type of information required and decisions regarding the type of
analysis required should be made prior to sample collection (see Chapters 10 and 11).
However, this is not always possible, particularly when a routine monitoring programme
is not in place. Samples may therefore require immediate evaluation on arrival in the
laboratory to determine if pretreatment is needed prior to appropriate sample storage.
Samples that have been taken for microscopic enumeration should ideally be preserved
with Lugol's iodine solution at the time of collection (Chapter 11). These samples will be
relatively stable and no special storage is required, although they should be protected
from extreme temperatures and strong light. However, samples should be examined and
counted as soon as practically possible because some types of phytoplankton are
sensitive to prolonged storage, and Lugol's iodine solution disintegrates over extended
storage periods (usually in the range of months, but in a shorter period in very dense
samples).
Unpreserved samples for quantitative microscopic analysis require immediate attention

either by addition of preservative or by following alternative counting methods which do
not use preserved cells. Where unpreserved samples cannot be analysed immediately
they should be stored in the dark with the temperature kept close to ambient field
temperatures. Unpreserved samples are preferable for species identification because
some characteristics cannot be recognised in preserved samples. For example, colonies
of Aphanizomenon have a characteristic bundle structure which facilitates identification,
but preservatives tend to disintegrate the colonies, and the single filaments are more
difficult to distinguish from other genera. While samples for quantification must be
preserved immediately or counted, samples for identification may be analysed within 24
hours because changes in numbers are less important.
Samples for the analysis of chlorophyll a, total and dissolved phosphate as well as
nitrate and ammonia, should be filtered as soon after sampling as possible. Storage for a
few hours in the dark in glass bottles is usually acceptable if temperatures do not exceed
20 °C. Filtration at the sampling site is recommended, particularly in warm climates, or
filtration should occur immediately upon arrival in the laboratory (see section 11.4.3).
Filtered samples for nutrient analysis may be stored in the refrigerator for a few hours
prior to analysis, or deep-frozen at -18 °C for several weeks. Although the suitability of
storage of filters for chlorophyll a analysis at -18 °C is currently under debate, the
method is employed by many laboratories if immediate extraction cannot be organised.
12.2 Cyanobacterial identification
Microscopic examination of a bloom sample is very useful even when accurate
enumeration is not being carried out. The information obtained regarding the
cyanobacteria detected can provide an instant alert that harmful cyanotoxins may be
present. This information can determine the choice of bioassay or analytical technique
appropriate for determining toxin levels (see Chapter 13).
Most cyanobacteria can be readily distinguished from other phytoplankton and particles
under the microscope by their morphological features at a magnification of 200-1,000
times. Figure 12.1 shows the most frequently occurring of the species known to produce
toxins. Cyanobacterial taxonomy, following the established botanical code, differentiates
by genera and species. However, this differentiation is subject to some uncertainty, and
organisms classified as belonging to the same species may nonetheless have
substantial genetic differences, e.g. with respect to microcystin production (see Chapter
3). Genetically identical cells, obtained by isolation of one colony and cultivation of its
daughter cells, are termed strains or genotypes, and field populations of one species (or
morphotype, i.e. identified as species on the grounds of morphological similarity) consist
of a number of genotypes which cannot be differentiated microscopically. Current
understanding of the regulation of cyanotoxin production indicates that distinction of
genera is very important for assessing potential toxicity (see Chapter 3), but that
microcystin content varies extremely at the level of genotypes or strains, rather than at
the level of species. This is one reason why identification to the taxonomic level of
genera (e.g. Microcystis, Planktothrix, Aphanizomenon, Anabaena) is frequently
sufficient. It is preferable to give only the genus name, especially if differentiation
between species by microscopy is uncertain on the basis of current general taxonomic
knowledge, a lack of locally available expertise, or lack of characteristic features of the
specimens to be identified. This must be emphasised because "good identification
practice" has frequently been misunderstood to require determination down to the
species level, and this has lead to numerous published misidentifications of species.
Figure 12.1 The most frequently occurring species of cyanobacteria known to produce
toxins
Coelosphaerium
Gomphosphaeria
Microcystis
Synechococcus
Synechocystis
Pseudanabaena
Oscillatoria
Trichodesmium
Schizothrix
Lyngbya
Phormidium
Cylindrospermopsis
Aphanizomenon
Nostoc
Anabaena
Hormothamnion
Nodularia
Gloeotrichia
Practitioners in health authorities with some experience in using a microscope can easily
learn to recognise the major cyanobacterial genera and some prominent species which
occur in the region they are monitoring. They should not be deterred by the pitfalls of
current scientific work in cyanobacterial taxonomy which targets differentiation to the
species level. Nevertheless, more precise identification of the dominant organisms down
to species level may be useful for a more accurate estimate of toxin content. For
example, Planktothrix agardhii and Planktothrix rubescens have both been shown to
contain microcystins, but each species contains different analogues with different toxicity.
For establishing cyanobacterial identification in a laboratory, initial consultation and later

occasional co-operation with experts on cyanobacterial identification (as outlined in
Chapter 10) is helpful. Training courses for beginners should focus on the genera and
species relevant in the region to be monitored. Experts can assist in initially deriving a
list of these taxa and the criteria for their identification. In the course of further monitoring,
experienced experts should be consulted periodically for quality control and for updating
such a list.
Publications assisting in the determination of cyanobacterial genera and species include

Komárek and Anagnostidis (1986) and Anagnostidis and Komárek (1988).
12.3 Quantification
Rapid methods for frequent monitoring of large numbers of water bodies or sampling
sites have been developed in some countries. These cannot be readily standardised and
evaluated internationally, but require adaptation to regional or local conditions. Deciding
on the appropriate classification of units to count depends upon variations such as
whether prevalent taxa are filamentous or coccoid, whether they are colony-forming or
occur as single filaments or cells, and whether populations are very diverse or largely
mono-specific. One rapid method which is highly standardised and simplified involves
counting a 0.5 ml sample of water for 2 minutes and determining the number of
cyanobacterial units present. A table has been prepared which provides adjustment
factors to take into account differences in the number of cells present in filaments or
colonies.
Cyanobacterial biomass can also be determined using indirect methods, the most
common being chlorophyll a quantification. This is a rapid and simple method but it is
open to interference by chlorophyll a from other phytoplankton, hence it is best used
when cyanobacteria are the main or dominant organisms present.
12.3.1 Enumeration
Microscopic enumeration of cyanobacterial cells, filaments or colonies has the

advantage of directly assessing the presence of potentially toxic organisms. Little
equipment in addition to a microscope is required. The method may be rather time
consuming, ranging from a few minutes to several hours per sample, depending upon
the accuracy required and the number of species to be differentiated. The remaining of
this section begins by outlining precise and widely accepted counting procedures which
are more time consuming and require a moderate level of expertise, but serve as a
bench mark to assess the performance of simplified methods developed to suit the
expertise and requirements of a sampling programme.
Sample concentration by sedimentation or centrifugation
Direct counting of preserved cells is typically carried out by Utermöhl's counting

techniques using a counting chamber and inverted microscope (Utermöhl, 1958). This
method is well suited for assessment of a large variation in cell type and is widely
accepted as one of the most reliable. Counting chambers and sedimentation tubes are
commercially available or can be constructed by the investigator. The most commonly
used chambers have a diameter of 2.5 cm and a height of about 0.5-2 cm and thus can
contain 2-10 ml of sample. These chambers can fit easily on the stage of an inverted
microscope. If larger volumes of water need to be analysed, as is the case when cell
density is low (e.g. in drinking water or at the beginning of population development), then
the height of the tube has to be increased. Such extended tubes, however, are too tall to
fit on the inverted microscope stage and the light would have to pass through a
considerable thickness of liquid before reaching the sedimented specimens. This
problem can be overcome by using a tube in two sections which allows the supernatant
to be removed (after settling the sample) without disturbing the sedimented cells on the
bottom glass (Figure 12.2). The amount of sedimented water required depends on the
density of cells, on the counting technique (fields or transects, see later) and on the
magnification. If concentrations are high (such as from bloom material), even 2 ml may
contain too many cells for enumeration, and the sample will need to be diluted.
Apparatus
• Inverted microscope with 10x and 40x objectives

• Counting chamber with sedimentation tube
• Cyanobacterial identification key
• Sample preserved in Lugol's iodine solution (see section 11.3.1)
Figure 12.2 Counting chambers for use with inverted microscopes
Step 1
Cylinder is pressed firmly onto the counting chamber, sample is poured in, cylinder is sealed with
heavy round glass slide. Sedimentation requires 4 hours per cm of cylinder height.
Step 2
Thin, square cover slide is used to slide cylinder and supernatant off the counting chamber.
Procedure
1. Allow the sample to equilibrate to room temperature. If cold samples are placed
directly in the counting chamber, air-bubbles develop and prevent sedimentation.
2. Gently invert the bottle containing the sample several time to ensure even mixing of
the phytoplankton.
3. Pour the sample into the sedimentation tube in place over the counting chamber.
4. Place the counting chamber on a horizontal surface where it will not be disturbed or
exposed to direct sunlight.
5. Allow the sample to settle. Sedimentation times will vary depending on the height of
the sedimentation tube. Allow at least 3-4 hours per cm height of liquid. Where
neutralised formalin has been used as a preservative, double the time allowed for
sedimentation. Note that buoyant cells (i.e. those with gas vesicles) may not settle and
may require disruption of the gas vacuoles (see below). However, this problem is
frequently overcome by several days of storage with Lugol's solution, because uptake of
iodine increases the specific weight of the cells.
6. Phytoplankton density can now be determined by counting either the total number of
organisms on the base of the chamber or by counting subsections (transects, fields).
If an inverted microscope is not available, and samples with low cyanobacterial density
need to be counted, other techniques may be applied in order to concentrate samples
sufficiently (e.g. sedimentation in a measuring cylinder, followed by careful removal of
the supernatant).
Apparatus
• Glass measuring cylinder, 100 ml

• Glass pipette with pipette bulb or filler
• Standard laboratory microscope with 10x and 40x objectives
• Sample preserved in Lugol's iodine solution (section 11.3.1)
Procedure
1. Allow the sample to equilibrate to room temperature.
2. Gently invert the bottle containing the sample several times to ensure even mixing of
the phytoplankton.
3. Pour 100 ml of the sample into the measuring cylinder.

4. Allow the sample to sediment (3-4 hours per cm height of liquid) in a location where it
will be out of direct sunlight and it will not be disturbed.
5. Using the glass pipette with pipette bulb or filler attached, carefully remove the
supernatant, leaving only the last 5 ml undisturbed.
6. The sample has now been concentrated by a factor of 20 and can be counted using a
counting chamber (e.g. Sedgewick-Rafter or haemocytometer).
Where sedimentation is not possible, centrifugation can offer a rapid and convenient
method of concentrating a sample (Ballantine, 1953). Fixation with Lugol's iodine
solution enhances the susceptibility of cells to separation by centrifugation. However,
buoyant cells (i.e. those with gas vesicles) may still be difficult to pellet and may require
disruption of vacuoles prior to centrifugation by applying sudden hydrostatic pressure
(see below) (Walsby, 1992). Once concentrated, a known volume can be quantified
using a counting chamber or by counting a defined volume using a micropipette to place
a drop on a microscope slide. Observation and counting can be done with a standard
microscope.
Apparatus
• Centrifuge
• Centrifuge tube, 10-20 ml
• Syringe or bottle with cork, or plastic bottle with screw cap
Reagents
• Aluminium potassium sulphate, 1.0 g AIK(SO4)2.12H2O in 100 ml distilled water
Procedure
1. Place 10-20 ml of sample in a centrifuge tube, seal with cap, and centrifuge at 360 × g
for 15 minutes.
2. When pelleting needs to be enhanced, add 0.05 ml of aluminium potassium sulphate

solution per 10 ml of sample. Mix and centrifuge as described.
3. Where problems occur with the pelleting of buoyant cells, try one of the following:
i) Place sample in a plastic syringe, ensure the end is tightly sealed, then apply pressure
to the plunger.
ii) Place sample in a bottle with a tightly fitting cork then bang the cork suddenly.
iii) Place sample in a well sealed plastic bottle and bring it down sharply onto a hard
surface.
These three approaches should be carried out with extreme care to avoid accidental
exposure to toxic cyanobacteria. Once they have been subjected to this pressure shock,
the gas vesicles should have been disrupted and cells should pellet when centrifuged.
4. Once cells have been centrifuged, carefully remove the supernatant and resuspend
the pellet in a small known volume (e.g. 0.5 ml).
5. Samples concentrated by centrifugation can be counted using a counting grid or

haemocytometer.
Counting cyanobacteria
Counting cyanobacteria involves defining the units to be counted. The majority of

planktonic cyanobacteria are present as filamentous or colonial forms consisting of a
large number of cells which are often difficult to distinguish. The accuracy of quantitative
determination depends on the number of counted objects (e.g. cells or colonies); the
relative error is approximately indirectly proportional to the square root of the number of
objects counted. The number of colonies, not the number of cells, is decisive for
accurate enumeration. However, the number of colonies is often not very high, even in
water containing a dense bloom where only 20-40 colonies may be present in a 100 ml
sample. Both filaments and colonies can differ greatly in the number of cells present,
hence results given as number of colonies (e.g. stating that 1 ml of sample contains an
average of 2.43 colonies of Microcystis aeruginosa) give little information on the quantity
of cyanobacteria present).
Typically, unicellular species are counted as cells per ml and filamentous species can
either be counted as number of filaments, and quoting an average number of cells per
filament. The cells per filament in the first 30 filaments encountered are often counted
and averaged. Alternatively, the total filament length per ml may be assessed as the
sum of the extension of each filament within a counting grid placed in the ocular of the
microscope. The latter approach is more precise when filament length is highly variable.
For colonial species, disintegration of the colonies and subsequent counting of the
individual cells is preferable to counting colonies and estimating colony size (Box, 1981).
Disintegration of colonies sometimes occurs several days after fixation with Lugol's
iodine solution. For more stable colonies, it can be achieved by alkaline hydrolysis (80-
90 °C for 15 min, followed by intensive mixing) or gentle ultrasonication. These methods
often separate cells very effectively, and even where colonies are not totally broken
down into single cells the colony size may be reduced sufficiently to allow single cells to
be counted. If this approach is impossible, the geometric volume of individual colonies
may be estimated. If colonies are relatively uniform in size, the average number of cells
per colony may be determined and then the colonies can be counted. Generally, the use
of published values for numbers per colony is not recommended because the size of
colonies varies greatly.
There are several systematic methods for counting cyanobacteria. Most approaches aim
at counting only a defined part of the sample and then calculating back to the volume of
the entire sample. The most common methods are:
• Total surface counting which requires the counting of all cells settled within the
chamber and may be very time consuming. This method is usually only appropriate for
counting very large units at low magnification.
• Counting cells in transects from one edge of the chamber to the other (Figure 12.3).
Transects should pass through the central point of the chamber. Some inverted
microscopes are equipped with special oculars so that the transect width can be
adjusted as required. However, in many cases, the horizontal or vertical sides of a
simple counting grid can be used to indicate the margin of the transect. Back-calculating
to a millilitre of sample requires measuring the area of the transects and of the chamber
bottom as well as the volume of the counting chamber.
• Counting cyanobacteria occurring in randomly selected fields ("Sichtfeld") (Figure 12.3).

It is recommended that the position of the chamber to find the next field should be
changed without looking through the microscope in order to prevent a bias in the
selection of fields. The Sichtfeld area covered by a counting grid is usually considered as
one field. However, if no counting grid is available the total spherical Sichtfeld can be
considered as a single field. Back-calculating to 1 ml of sample requires registration of
the number of Sichtfelds counted, measuring the area of the Sichtfeld and of the
chamber bottom, as well as knowing the volume of the counting chamber.
The density of different species in one sample can vary and there can also be several
orders of magnitude difference between the size of different species; hence it is
necessary to select the counting method to suit the sample. Total chamber surface
counting with low magnification (100x) is required for large species whereas transect or
field counting with higher magnification (200x, 400x) is used for smaller or unicellular
cyanobacteria. Accurate enumeration using transects or fields assumes on even
distribution of cyanobacteria on the bottom of chamber surface after sedimentation. Due
to inevitable convection currents, cells very rarely settle randomly on the surface of the
bottom glass and are, almost always, more dense either in the middle or around the
circumference of the chamber. Sometimes density also varies between opposite edges.
The inaccurate estimate that arises from uneven distribution can be minimised by
transect counting. Consequently, transect counting is the preferred method and counting
four perpendicular diameters minimises the error. The relation of precision to counting
time is very effective if about 100 counting units (cells, colonies, filaments) are settled in
one transect (for simplification, see Box 12.1). Samples are best diluted or concentrated
so that the number of units of the important taxa lies within this range.
Figure 12.3 Counting techniques used to enumerate cyanobacteria in water
samples
A note on filaments:
• Either count them just as cells (inaccurate if length is highly variable). In that case, the
one extending out of the grid to the right would be counted, but not the one extending
out to the left.
• Or estimate the length of filament within the boundaries of the grid. In this case: 3
boxes (bottom left) + 5 boxes (mid-right) = 8 boxes
Specimens occurring exactly on the margin of the counting area (transect or field)
present the common problem of whether to count them or not. When counting transects,
those specimen that lie across the left margin are ignored while those that cross the right
margin are included. When counting fields two predetermined sides of the grid are
included and the other two are ignored (Figure 12.3).
Box 12.1 Simplification for biomass estimates
With some experience and a flexible approach, the time needed for enumeration and
measurement of cell dimensions can be considerably reduced (down to 1 hour or less, if only one
or two species require counting) without substantial loss of accuracy. The procedure is as follows:
• If the deviation of numbers of dominant species counted in two perpendicular transects is less
than 20 per cent between both transects, do not count further transects.
• If the standard deviation of cell dimensions measured on 10 cells is less than 20 per cent, do
not measure further cells.
• If a set of samples from the same water body and only slightly differing sites (e.g. vertical or
horizontal profiles) is to be analysed, enumerate all samples, but measure cell dimensions only
from one. Check others by visual estimate for deviations of cell dimensions and conduct
measurements only if deviations are suspected.
There are different recommendations regarding the number of units per species that
must be counted to obtain reliable data. It is particularly difficult to count each species
with an acceptable error (20-30 per cent if 400 individual units are counted) in each
sample. Mass developments of cyanobacteria are characterised by dominance of one to
three species. Even if total phytoplankton is to be counted (for example in order to
assess the relative share of cyanobacteria), it is rare for more than six to eight species to
contribute to the majority of the biomass. Therefore, for total phytoplankton counts, it is
suggested that 400-800 specimens in each sample are counted, giving a maximum error
for the total count as 7-10 per cent. In this situation there will be a 10-20 per cent error
for the few dominant species, 20-60 per cent for the subdominant species and the rest of
the species can be considered as insufficiently counted. If only cyanobacteria are to be
counted, and only one or two species are present, counting up to the precision level of
20 per cent, by counting 400 individual units per species, can be accomplished within
less than one hour.
The use of mechanical or electronic counters for recording cell counts can shorten
counting time considerably, especially if only a few species are counted. Computer
keyboards can also be used together with suitable programmes for recording cell counts.
The use of an inverted microscope with counting chambers is generally the best
approach for estimating cyanobacterial numbers. However, a standard microscope is
sufficient for preconcentrated samples or for naturally dense samples from mass
developments, provided the size of the water drop enumerated can be defined (e.g. by
using a micropipette). Other counting chambers (e.g. Sedgewick-Rafter or
haemocytometer) are available for use with a standard microscope. It can also be useful
to monitor samples under high magnification with oil-immersion (1,000x) to check the
sample for the presence of very small species which may be overlooked during normal
counting.
An alternative counting method which has been found to be useful is syringe filtration.
This method is considerably less time consuming because it does not depend on lengthy
sedimentation times and uses a standard laboratory microscope.
Apparatus
• Syringe, 10 ml
• Membrane filters, 13 mm diameter with 0.45 µm pore
• Membrane filter holder
• Glass microscope slides plus coverslips
Reagents
• Immersion oil
Procedure
1. Mix water sample by inverting several times.
2. Take up 10 ml of the sample into the syringe.
3. Place filter holder with filter in place, on the end of the syringe.
4. Gently filter the sample through the filter by applying pressure to the syringe piston.
5. Once all the sample has passed through the filter, remove the filter from the holder
and place it on a glass microscope slide with the captured cells uppermost.
6. Allow the filter to dry at room temperature then carefully add one or two drops of
immersion oil to the filter. The will make the filter appear transparent and permit
observation of the cyanobacterial cells trapped on its surface.
7. Finally, cover the filter surface with a glass coverslip and examine under the
microscope.
8. The density of cyanobacteria can be easily calculated by counting the number of cells
on the filter and dividing this by the volume of water filtered (i.e. number of cells per ml).
12.3.2 Determination of cyanobacterial biomass microscopically
Cell size can vary considerably within and between species, and toxin concentration
relates more closely to the amount of dry matter in a sample than to the number of cells.
Hence, cell numbers are often not an ideal measure of population size or potential
toxicity. This can be overcome by determining biomass. Two approaches are available,
either estimation from cell counts and average cell volumes, or from chemical analysis of
pigment content.
Cyanobacterial counts and cell volumes
Biovolume can be obtained from cell counts by determining the average cell volume for
each species or unit counted and then multiplying this value by the cell number present
in the sample. The result is the total volume of each species. Given a specific weight of
almost 1 mg mm-3 for plankton cells, this biovolume corresponds quite closely to
biomass. Average volumes are determined by assuming idealised geometric bodies for
each species (e.g. spheres for Microcystis cells, cylinders for filaments), measuring the
relevant geometric dimensions of 10 to 30 cells (depending upon variability) of each
species, and calculating the corresponding mean volume of the respective geometric
body.
Example 1
By measuring 20 Microcystis cells, an average diameter of 5 µm was established.

Assuming spherical-shaped cells the average cell volume is 4/3 πr3 = 65.4 µm3.
Enumeration resulted in 1 million cells per ml and thus the total biovolume is 65.4 × 106
µm3 ml-1.
Example 2
Measuring 30 Planktothrix filaments resulted in an average length of 225 µm and an

average diameter of 6 µm. Assuming cylindrical shaped filaments, the average filament
volume is 2 πr2 × L = 6,359 µm3. Enumeration resulted in 10,000 filaments per ml. Thus
the biovolume of Planktothrix was 63.6 × 106 µm3 ml-1.
Thus, although the number of Planktothrix was 100-fold less than that of Microcystis,
biovolumes were similar because the volume (and biomass) of a single Planktothrix
filament is about 100 times as large as that of a single Microcystis cell. Both species
often contain microcystins, and it is possible to compare the relative toxin content per
biovolume or biomass whereas there is little point in comparing toxin content in relation
to the respective cell numbers.
12.4 Determination of biomass using chlorophyll a analysis

The pigment chlorophyll a generally contributes 0.5-1 per cent of the ash-free dry weight
of phytoplankton organisms. Although the pigment content may vary according to the
physiological state of the organisms (e.g. it increases if light availability is low),
chlorophyll a is a widely used and accepted measure of biomass. It is an especially
useful measure during cyanobacterial blooms, when the phytoplankton chiefly consists
of cyanobacteria, often of only one species. However, when chlorophyll a determination
is used with mixed phytoplankton populations (cyanobacteria and other species), it gives
an overestimation of cyanobacterial biomass. Rough microscopic estimations of the
relative share of cyanobacterial cells among the total phytoplankton may be used to
correct the overestimate.
Analysis of chlorophyll a requires relatively simple laboratory equipment, principally

filtration apparatus, centrifuge and spectrophotometer. It is considerably less time-
consuming than microscopic biomass determination (but also less specific and less
precise). Standard protocols have been described (e.g. ISO, 1992) but preferred
methods vary somewhat between laboratories. However, the main procedural steps in
most methods are essentially the same: solvent extraction of chlorophyll a,
determination of the concentration of the pigment by spectrophotometry, and
adjustments to the result to reduce the interference by phaeophytin a which is a
degradation product of chlorophyll a. A simple method following the ISO procedure for
the determination of chlorophyll a in a lake water sample is outlined below.
Apparatus
• Spectrophotometer suitable for readings up to 750 nm, or photometer with discrete

wavelengths at 665 and 750 nm
• Glass cuvettes, typically of 1 cm path length, or 5 cm for very low concentrations (e.g.
from drinking water reservoirs at the beginning of population development)
• Centrifuge
• 15 ml centrifuge tubes, graduated and with screw caps
• Water bath at 75 °C or other heating device for boiling ethanol
• Glass fibre filters (GF/C), 47 mm diameter
• Filtration apparatus and vacuum pump
• Tissue grinder or ultrasonication device
• Pipette or similar for addition of acid
Reagents
• Ethanol (90 % aqueous)
• Hydrochloric acid, 1 mol l-1
Procedure
Perform the following steps in low intensity of indirect light because light induces rapid
degradation of chlorophyll.
1. After recording the initial volume of water, separate the cells from the water by
filtration. Filter continuously and do not allow the filter to dry during filtration of a single
sample. If extraction cannot be performed immediately, filters should be placed in
individual, labelled bags (filters folded in half with cells innermost) or Petri dishes and
stored at -20 °C in the dark (this may cause some pigment degradation and is not
recommended by ISO). This step can be carried out at the sampling site and the
samples are readily transported in this form. In preference to freezing, samples may be
stored in the extraction medium (see below) for up to 4 days in the refrigerator.
2. Place the filter in a tissue grinder, add 2-3 ml of boiling ethanol, and grind until the
filter fibres are separated. Ultrasonication can also be used. Pour the ethanol and
ground filter into a centrifuge tube, rinse out the grinding tube with another 2 ml ethanol
and add this to the centrifuge tube. Make up to a total of 10 ml in the centrifuge tube with
ethanol. Place cap on the tube, label and store in darkness at approximately 20 °C for
24-48 hours.
3. Centrifuge for 15 minutes at 3,000-5,000 g to clarify samples. Decant the clear

supernatant into a clean vessel and record the volume.
4. Blank Spectrophotometer with 90 per cent ethanol solution at each wavelength.
5. Place centrifuged sample in the cuvette and record absorbance at 750 nm and 665
nm (750a and 665a; absorbance at 750 is for turbidity correction and should be very low).
Readings at 665 nm should range between 0.1 and 0.8 units.
6. Add 0.01 ml of 1 mol I-1 HCl to sample in cuvette (adjust volume to suit the volume of
cuvette being used, calculating approximately 0.003 ml of 1 mol l-1 HCl per ml of ethanol
solution) and agitate gently for 1 minute. Record absorbance at 750 nm and 665 nm
(750b and 665b).
Calculation
1. Correct for turbidity by subtracting absorbance 665a-750a = corrected 665a absorbance
665b-750b = corrected 665b absorbance
2. Use the corrected 665a and 665b absorbance to calculate:
where: Ve = Volume of ethanol extract (ml)

Vs = Volume of water sample (litres)
l = Path length of cuvette (cm)
Note, the ratio of chlorophyll a to phaeophytin a should give an indication of the condition
of the cyanobacterial (and algal) population, but may also reflect the effectiveness of
sample handling and preservation, because high levels of phaeophytin a indicate
degradation of chlorophyll a either in scenescent field populations or during analysis.
When samples are concentrated by filtration for the purposes of analysis, the cells die.
Consequently, the chlorophyll immediately starts to degrade to phaeopigments. If filters
are not rapidly extracted or frozen, chlorophyll a concentrations are thus reduced.
Occasionally, other factors affect this method, resulting in very low or even negative
values for chlorophyll a. This can be checked by calculating:
The result of this calculation should give a similar value to the sum of the concentrations
of both pigments determined separately, as above. Note also:
• If no centrifuge is available, filtration may be used instead.
• If no tissue grinder or ultrasonication device is available, proceed without this step.

Slight underestimations may occur. For cyanobacteria, these are not likely to be too
serious.
12.5 Determination of nutrient concentrations
The capacity for development of a cyanobacterial bloom depends on the available
concentrations of elements that the cells are composed of (chiefly carbon, hydrogen,
oxygen, phosphorus, nitrogen and sulphur). These elements are needed in the ratio in
which they occur in living cells (in weight units: 42 C, 8.5 H, 57 O, 7 N, 1 P and 0.7 S).
Hydrogen and oxygen are available in unlimited supply in an aqueous environment, and
sulphur is usually also present in surplus concentrations. Carbon has been investigated
as a potentially limiting factor, but has rarely been found to be relevant. Most often,
phosphorus concentrations limit the amount of biomass that can form in a given water
body but sometimes, nitrogen is limiting. The chief sources of nitrogen are nitrate and
ammonia, but to some extent their lack can be compensated by some cyanobacteria
through fixation of atmospheric nitrogen. Thus, even if phosphate is clearly the factor
limiting carrying capacity, knowledge of nitrogen availability helps to predict whether
nitrogen-fixing species are likely to grow.
Cyanobacterial cells appear to have little means of storing excess nitrogen, but can store
enough phosphate for up to four cell divisions, which implies that one cell can grow into
16 without needing to take up dissolved phosphate. Information on dissolved phosphate
concentrations, therefore, only demonstrates that if it can be detected, the phytoplankton
population is not currently limited by phosphate. In order to assess the capacity of the
water body to carry a cyanobacterial population, total phosphate must be determined,
which can then be compared with the total concentration of nitrogen salts and organic
nitrogen. However, in order to assess whether nitrogen may be limiting, analysis of
dissolved components (chiefly nitrate and ammonia) is sufficient.
Among the methods available, the procedure of Koroleff (1983) for determining total
phosphate has proved to be most reliable and is the basis of an ISO protocol. For nitrate
and ammonia, several methods are available, but the ISO method with the least
demands on equipment is described below. Details of ISO methods can be obtained
directly from ISO at Case Postale 56, CH-1211, Geneva 20, or requested through the
Internet on central@iso.ch.
12.5.1 Analysis of phosphorus according to ISO 6878
Phosphorus in various types of waters can be determined spectrometrically by digestion

of organic phosphorus compounds to orthophosphate and reaction under acidic
conditions to an antimony-phosphormolybdate complex which is then reduced to a
strongly coloured blue molybdenum complex. The internationally harmonised method
described by ISO/FDIS 6878 (ISO, 1998a) is applicable to many types of waters
(surface-, ground-, sea- and wastewater) in a concentration range of 0.005 to 0.8 mg l-1
(or higher if samples are diluted). Differentiation by the following fractions is possible
through filtration procedures:
Option Fraction Filtration/procedure

1 Soluble reactive phosphorus (SRP or orthophosphate) Filtered sample
2 Dissolved organic phosphate Digested filtered sample
3 Particulate phosphorus Option 4 minus option 2
4 Total phosphorus Digested unfiltered sample
Digestion or mineralisation of organophosphorus compounds to orthophosphate is

performed in tightly sealed screw-cap vessels with persulphate, under pressure and heat
in an autoclave (in the absence of which good results have also been obtained with
household pressure cookers), or simply by gentle boiling. Polyphosphates and some
organophosphorus compounds may also be hydrolysed with sulphuric acid to
molybdate-reactive orthophosphate. The following gives an overview of the procedure,
necessary equipment and chemicals, see ISO (1998) for details and specific problems.
Apparatus
• Photometer measuring absorbance in the visible and near infrared spectrum above
700 nm; sensitivity is optimal at 880 nm (and reduced by 30 per cent at 700 nm);
sensitivity is increased if optical cells of 50 mm are used (if 100 mm cells are available,
determination down to 0.001 mg I-1 may be possible)
• Filter assembly and membrane filters, 45 mm diameter with 0.45 µm pore
• For digestion of samples, an autoclave (or pressure cooker) suitable for 115-120 °C
• For digestion of samples, borosilicate vessels with heat-resistant caps that can be
tightly sealed
• Bottles for samples as described in Chapter 11
• Pre-cleaned glass bottles for filtered samples
Reagents
All reagents should be of a recognised analytical grade and the distilled water used must
have a negligible phosphorus concentration when compared with the samples
• Sulphuric acid (H2SO4): 9 mol l-1
• Sulphuric acid (H2SO4): 4.5 mol l-1
• Sulphuric acid (H2SO4): 2 mol l-1
• Sodium hydroxide (NaOH): 2 mol l-1
• Ascorbic acid (C6H8O6): 100 g l-1 (stable for 2 weeks in amber glass bottle, refrigerated)
• Acid molybdate solution I: ammonium heptamolybdate tetrahydrate [(NH4)6Mo7O24. 4

H2O] 13 g per 100 ml and antimony potassium tartrate hemihydrate [K(SbO)C4H4O6. ½
H2O] 0.35 g per 100 ml (stable for 2 months in amber glass bottle)
• Orthophosphate standard stock solution: sodium thiosulphate pentahydrate (Na2S2O3.
5H2O) 1.2 g in 100 ml water, stabilised with 0.05 g of anhydrous sodium carbonate
(Na2CO3) as preservative
• Potassium peroxodisulphate: (K2S2O8) 5 g per 100 ml (stable for 2 weeks in amber

glass borosilicate bottle)
Procedure
All glassware (including sampling bottles) must be washed with hydrochloric acid (1.12 g
ml-1) at 40-50 °C and thoroughly rinsed. Do not use detergents containing phosphates
and preferably use the glassware only for the determination of phosphorus.
For measuring orthophosphate:
1. Filter samples with pre-washed filters; discard the first 10 ml of filtrate, collect 5-40 ml
(depending on concentrations expected).
2. Carry out a blank test with distilled water, using all of the reagents and performing the
same procedure as for the samples.
3. Prepare orthophosphate calibration solutions in the concentration range of the

samples (e.g. from 0.05 to 0.5 mg l-1) with a volumetric pipette in 50 ml volumetric flasks
(filling them only up to about 40 ml).
4. Transfer samples into 50 ml volumetric flasks with volumetric pipettes. Depending on

expected concentrations, use 5-40 ml of sample, fill up to about 40 ml with distilled water.
5. Add, while swirling, first 1 ml ascorbic acid solution and then 2 ml acid molybdate
solution, fill flask up to the 50 ml mark with distilled water and mix well.
6. After 10-30 minutes, measure absorbance at 880 nm using distilled water in the
reference cell.
7. Plot absorbance of calibration solutions against their concentration and determine

slope; check for linearity. Run an independently-prepared calibration solution with each
series of samples, but especially when new batches of reagents are used.
8. Occasionally dean the glassware used for developing the colour complex with sodium
hydroxide solution to remove colour deposits.
For measuring total, particulate and dissolved organic phosphorus:

1. Clean digestion vessels with about 50 ml of water and 2 ml of sulphuric acid (1.84 g
ml-1) in autoclave for 30 minutes at 115-120 °C, cool and rinse, repeat procedure several
times, store covered.
2. Carry out a blank test with distilled water, using all of the reagents and performing the
same procedure as for the samples.
3. Add 1 ml of sulphuric acid (4.5 mol l-1) to 100 ml of sample to adjust pH to about 1
(further adjustment with sulphuric acid or sodium hydroxide solution (2 mol l-1).
4. Pipette 5-40 ml of sample into digestion vessel, add 4 ml of potassium

peroxodisulphate solution, mineralise in autoclave (or pressure cooker), or boil gently for
30 minutes.
5. Cool, adjust pH to between 3 and 10 with sodium hydroxide solution or sulphuric acid
(2 mol l-1), transfer to 50 ml flask and proceed as above for orthophosphate.
If large quantities of organic matter are present, oxidation with nitric acid-sulphuric acid
may be necessary. Furthermore, arsenate may cause interference (see ISO, 1998a).
The test report should contain complete sample identification, reference to the method
used, the results obtained and any further details likely to have influence on the results.
12.5.2 Analysis of nitrate
Several methods for determination of nitrate have been provided by the ISO, the
simplest being a spectrometric measurement of the yellow compound formed by reaction
of sulphosalicylic acid with nitrate and subsequent treatment with alkali (ISO, 1988). The
equipment required is a spectrometer operating at a wavelength of 415 nm and optical
path length of 40-50 mm, evaporating dishes, a water bath capable of accepting six or
more dishes, and a water bath capable of thermostatic regulation to 25 °C. This method
is suitable for surface and potable water samples and has a detection limit of 0.003 to
0.013 mg l-1 (depending on optical equipment). Interference from a range of substances,
particularly chloride, orthophosphate, magnesium and manganese (III) is possible.
Interference problems can be avoided with other spectrometric methods ISO (1986a,b).
12.5.3 Analysis of ammonia
A manual spectrometric method is given by ISO (1984a) which analyses a blue

compound formed by the reaction of ammonium with salicylate and hypochlorite ions in
the presence of sodium nitrosopentacyanoferrate (III) at a limit of detection of 0.003-
0.008 mg l-1. An automated procedure is given by ISO (1986c). A distillation and titration
method is given by ISO (1984b).
12.6 References
Anagnostidis, K. and Komárek, J. 1988 Modern approach to the classification system of
cyanophytes. Archiv Hydrobiol., Supplement 80, Algological Studies, 50-53, 327-472.
Ballantine, D. 1953 Comparison of different methods of estimating nanoplankton. J. Mar.

Biol. Ass. UK, 32, 129-147.
Box J.D. 1981 Enumeration of cell concentrations in suspensions of colonial freshwater

microalgae, with particular reference to Microcystis aeruginosa. Brit. Phycol. J., 16, 153-
164.
ISO 1984a Water Quality - Determination of Ammonium - Part 1: Manual spectrometric
method. ISO 7150-1, International Organization for Standardization, Geneva.
ISO 1984b Water Quality - Determination of Ammonium - Distillation and titration method.
ISO 5664, International Organization for Standardization, Geneva.
ISO 1986a Water Quality - Determination of Nitrate - Part 1: 2,6-Dimethylphenol

spectrometric method. ISO 7890-1, International Organization for Standardization,
Geneva.
ISO 1986b Water Quality - Determination of Nitrate - Part 2: 4-Fluorophenol

spectrometric method after distillation. ISO 7890-2, International Organization for
Standardization, Geneva.
ISO 1986c Water Quality - Determination of Ammonium - Part 2: Automated

spectrometric method. ISO 7150-2, International Organization for Standardization,
Geneva.
ISO 1988 Water Quality - Determination of Nitrate - Part 3: Spectrometric method using
sulfosalicylic acid. ISO 7890-3, International Organization for Standardization, Geneva.
ISO 1998a Water Quality - Spectrometric determination of phosphorus using ammonium

molybdate. ISO/FDIS 6878, International Organization for Standardization, Geneva.
ISO 1992 Water Quality - Measurement of biochemical parameters. Spectrometric

determination of the chlorophyll-a concentrations. ISO 10260, International Organization
for Standardization, Geneva.
Komárek, J. and Anagnostidis, K. 1986 Modern approach to the classification system of

cyanophytes. Archiv Hydrobiol., Supplement 73, Algological Studies, 43, 157-164.
Koroleff, F. 1983 Determination of phosphorus. In: K. Grasshoff, M. Ehrhardt and K.

Kremling [Eds] Methods of Seawater Analysis. Verlag Chemie, Weinheim, Deerfield
Beach, FL, Basel, 125-139.
Utermöhl, H. 1958 Zur Vervollkommnung der quantitative Phytoplankton-Methodik. Mitt.

int. Verein. theor. angew. Limnol. 5, 567-596.
Walsby, A.E. 1992 The control of gas-vacuolated Cyanobacteria. In: D.W. Sutcliffe and
G. Jones [Eds] Eutrophication: Research and Application to Water Supply. Freshwater
Biological Association, Windermere, 143-162.
© 1999 WHO
ISBN 0-419-23930-8
Chapter 13. LABORATORY ANALYSIS OF

CYANOTOXINS
This chapter was prepared by Ken-ichi Harada, Fumio Kondo and Linda Lawton
There is a diverse range of laboratory methods used to detect and identify cyanotoxins
in water and cyanobacterial cells. These methods can vary greatly in their degree of
sophistication and the information they provide. Relatively simple low cost methods can
be employed which evaluate rapidly the potential hazard and allow management
decisions to be taken. In contrast, highly sophisticated analytical techniques can be
employed which determine precisely the identity and quantity of cyanotoxins.
Techniques can be selected depending on the facilities and expertise available, coupled
with the type of information required. For example, analysis of water for recreational
purposes may only require detection of potentially harmful cyanobacteria, whereas
ensuring potable water does not exceed guideline levels for cyanotoxins may require
highly specialised equipment and expertise. Information obtained from simple, rapid
screening methods such as microscopic examination can be used to make an informed
decision on the type of bioassay or physicochemical technique which should be
employed. It is important to remember that, currently, there is no single method which
can be adopted which will provide adequate monitoring for all cyanotoxins in the
increasing range of sample types which have to be evaluated. Selectivity and sensitivity
are important criteria for the selection of methods. Figure 13.1 compares three biological
and six chemical methods with respect to these criteria.

When samples arrive in the laboratory, the type of analysis that will be carried out should
have been anticipated previously. Full consideration should be given to the type of
information required prior to sample collection (see Chapters 10 and 11). However, this
is not always possible, particularly when a routine monitoring programme is not in place.
Information obtained from the microscopic examination of the cyanobacterial taxa
(Chapter 12) may also influence the choice of analysis but this information is not usually
available until samples have reached the laboratory. Many samples will therefore require
immediate evaluation on arrival in the laboratory to determine if any pre-treatment is
needed prior to appropriate sample storage.
Figure 13.1 Relationship between sensitivity and selectivity of analytical methods
for microcystins (see text for explanation of methods)
Samples for cyanotoxin analysis should be refrigerated in the dark to prevent toxin
degradation but it is essential that storage be kept to a minimum (preferably less than 24
hours). Where prolonged storage is required, samples can be frozen, although this will
release toxins from the cells and only the total amount of toxin in the sample can then be
determined. It is often desirable, especially for water treatment purposes, to estimate the
amount of toxin contained within cells as well as that which is dissolved in the water.
Where this information is required, samples should be gently filtered as soon as
practically possible and if necessary the filters and filtrate can be refrigerated or frozen.
A concentration step for cyanobacterial cells can be useful especially for less sensitive
cyanotoxin assay methods. This can be achieved at the sampling location with the use
of a plankton net or by using lakeside filtration apparatus (Chapter 12). In the laboratory,
concentration can be achieved by allowing buoyant cells to accumulate in the upper part
of a separating funnel which enables excess water to be removed. This method can
concentrate cells by at least ten fold but is dependent on cells being vacuolate (i.e.
floating). It may also lead to a bias if most of the cells do not accumulate at the surface
because floating cells may have different toxin quotas than cells which do not.
Centrifugation is useful but is often limited by the relatively small volume of laboratory
centrifuge tubes and, furthermore, problems can be encountered when trying to pellet
vacuolate cells. Filtration is increasingly popular because it can allow concentration of
cells by several orders of magnitude and enables the weight of the cell mass to be
determined. The following filtration method can be employed (Figure 13.2).
Apparatus
• Oven/incubator set at 45 °C or freeze-drier

• Vacuum dessicator
• Balance accurate to 0.0001 g
• Glass fibre filters 70 mm, GF/C, 1.2 µm will retain most cyanobacteria but the smaller
pore size of GF/F (0.7 µm) is required for picoplankton cyanobacterial cells.
• Filtration cup and vacuum pump
• Measuring cylinder
• Plastic Petri dishes
Procedure
1. Place filters in dessicator under vacuum, then remove at intervals and weigh. Filters
are ready to use once a constant weight is obtained. It is useful to place each weighed
filter in a separate Petri dish with the weight recorded on the dish. This makes samples
easy to handle and prevents confusion.
2. Mix the water sample by inverting the bottle several times, then measure a known
volume using the measuring cylinder. This volume will differ depending on the
concentration of cells present, because only small volumes of water with a high
concentration of cells can be filtered before the filter becomes blocked.
3. Using one of the pre-weighed filters, filter the water sample and then return the filter to
the labelled Petri dish.
4. Keeping the filter in the Petri dish place it either in the drying oven or freeze-drier. If
using on oven, the temperature must be kept below 50 °C.
5. Once dry, return filters to the dessicator and weigh to constant weight. Calculate the
dry weight of cells collected by subtraction of the initial weight of the filter.
6. Cyanotoxins can either be extracted immediately, or filters may be stored until

required, preferably in a freezer.
By following this procedure it is possible to relate cyanotoxin content to the dry weight of
particulate matter extracted as well as to the volume of water filtered. Sometimes dry
weights are affected by the presence of large particles in the sample (e.g. zooplankton
or feathers from waterfowl). This can easily be overcome by carrying out pre-screening
using a 1-2 mm sieve if necessary. Furthermore, particulate matter (i.e. seston) may
consist of plankton organisms in addition to cyanobacteria (e.g. algae, rotifers, bacteria,
detritus). A brief microscopic check of the fresh sample or a preserved sub-sample (see
Chapter 11) will reveal whether or not the dry weight can be attributed mainly to
cyanobacteria. In dense blooms and scums this is usually the case but, in samples from
more homogenously dispersed situations, other components may dominate and the
relating of the toxins to dry weight will therefore underestimate the toxin content of the
cells. If an accurate balance is not available, the filtration method can be used but the
cyanotoxin content can be related only to the volume of water filtered.
Figure 13.2 Toxin extraction by filtration
Dense cyanobacterial scums or samples concentrated using a plankton net are often
freeze-dried. This provides a dry powder which can easily be weighed prior to extraction.
However, great care must be taken with such dry powders because they can easily
become airborne and may present a health hazard through inhalation.
Most sample handling and storage methods for cyanotoxin analysis have been
evaluated primarily for microcystins, hence the stability of other cyano-toxins may not be
fully understood. It is therefore important to evaluate the chosen method if other
cyanotoxins are being monitored.
13.2 Sample preparation for cyanotoxin determination and
bioassays
13.2.1 Extraction
For cyanotoxin detection, samples may be extracted from cells or biological matrices
such as animal tissue, or water samples. Cells have successfully been extracted in a
number of different liquid phases, and some of the most popular are 5 per cent acetic
acid, methanol, acidified methanol (Trifluoroacetic acid (TFA) added), aqueous methanol,
and butanol:methanol:water (1:4:15) (Harada, 1996). The efficiency of these methods
depends on the sample and the cyanotoxins present. It has been found that the more
polar extraction media, such as 5 per cent acetic acid, provide reasonable extraction
efficiency of the more polar microcystins but give very poor recovery of hydrophobic
microcystins (Lawton et al., 1994a). Methanol has been advocated as the most suitable
solvent because it gives good extraction efficiency and has the added advantage of
allowing rapid sample concentration through evaporation. It has since been suggested
that 100 per cent methanol may give poor recovery of more polar microcystins; but the
addition of a small percentage of water overcomes this. Fastner et al. (1998) showed
that, especially for lyophilised field samples dominated by Microcystis spp., extraction
with 75 per cent methanol and 25 per cent water (by weight) was most effective. One
approach which has been used routinely, with success, for microcystins is as follows:
Apparatus
• Rotary evaporation equipment

• Glass beakers, 50 ml
• Measuring cylinder, 20 ml
• Rotary evaporation flasks, 50 ml, pear-shaped
• Pipette, 0.25 ml
• Glass vials or microcentrifuge tubes, ~1 ml.
Reagents
• Methanol, high purity if possible.
• Aqueous methanol (75 per cent, v/v)
Procedure
1. Place each filter containing cells into a glass beaker with 20 ml of either pure
methanol or 75 per cent aqueous methanol. Filters may be cut into small pieces with
scissors but gloves must be worn and care must be taken not to cross contaminate
samples or to lose material sticking to the scissors.
2. Allow filters to extract for 1 hour. If time is limiting, extract for 30 minutes because the
shorter time has been found to reduce only slightly the recovery of microcystins.
3. Decant extract into rotary evaporation flask and dry in vacuo at 45 °C.
4. Add a further 20 ml of extraction solvent to the filter in the beaker and allow to extract
as before.
5. Repeat this process a total of three times, each time decanting the extract into the
same rotary evaporation flask.
6. Add 0.25 ml of methanol to the dry extract in the rotary evaporation flask, mix and
remove the resuspended extract to a glass vial or microcentrifuge tube. Repeat this with
a second 0.25 ml of methanol, placing both aliquots in the same vial or tube.
7. Consideration must be given to the type of analysis which is going to follow extraction
because organic solvents are toxic to bioassay organisms. This can often be overcome
by using a volatile medium, followed by evaporation and resuspension in a medium
which is compatible with the assay method. With very toxic samples and/or sensitive
bioassays, dilution with the assay medium (using a control containing the same per cent
of solvent) is sufficient to obtain non-toxic solvent concentrations.
Two other rapid extraction procedures have also been successfully applied:
• After freezing and thawing to disrupt cells, toxins may be extracted in the filtration
device by passing aqueous methanol or water, followed by methanol, through the filters
by suction (use a total of 5-25 ml of solvent). Extracts can be used directly for analysis or
further concentrated if enhanced sensitivity is needed (Figure 13.2) (Utkilen and Gjølme,
1994). This method requires little equipment other than a filtration device and deep-
freezer.
• Membrane filters with approximately 20 mg of freeze-dried material may be extracted

in 2 ml microcentrifuge tubes by adding 1.5 ml solvent (preferably 75 per cent aqueous
methanol), sonicating, shaking for 30 minutes and centrifuging. This step extracts a large
share of the microcystins, and extraction is more complete if the pellet is re-extracted
twice and the supernatants are pooled (Fastner et al., 1998).
Saxitoxins are often extracted in acidified media including acetic acid, hydrochloric acid
(HCl) and acidified methanol (Fernandez and Cembella, 1995). A recent study found
methanol acidified with TFA was the most efficient solvent when extracting saxitoxin and
neosaxitoxin from cyanobacterial cells (McElhiney et al., 1998). Anatoxin-a has been
successfully extracted with water, acidified water, acidified methanol (Edwards et al.,
1992), chloroform followed by hydrochloric acid (Harada et al., 1989) or by
dichloromethane after an acidification and neutralisation step (Bumke-Vogt et al., 1996).
Box 13.1 When is sample cleanup needed?
Try analysis or bioassays without performing cleanup if:
• Samples are dominated by cyanobacteria; or
• Precise quantification of toxins and complete identification of minor toxin components is less
important than rapid analysis of a larger number of samples.
Introduce cleanup if:

• Cyanobacteria are a minor component of the sample.
• Low concentrations of cyanotoxins are anticipated.
• Identification and quantification is important.
• Chromatograms obtained without cleanup show considerable baseline problems or poor peak
separation.
• Bioassay results suggest influence from further substances.
To store extracts prior to analysis or bioassays, blow to dryness and deepfreeze (-18 °C).
Samples may then be re-dissolved in the solvent at a concentration adequate for the
subsequent chemical analysis or bioassay.
13.2.2 Sample cleanup
The purpose of cleanup is to eliminate impurities by a simple operation without loss of

analyte and where the concentration of cyanotoxins is low, it also enables enrichment of
the analyte. Whether or not cleanup is necessary depends largely on the precision of the
toxin determination and quantification required. Without cleanup, small toxin peaks in
chromatograms may be missed due to masking by other matrix substances eluting
simultaneously, and toxin concentrations may be overestimated if peaks are not clearly
separated from matrix substances. However, for many screening and monitoring
purposes, particularly of bloom samples containing little material other than
cyanobacteria, the carrying out of bioassays or toxin analysis without sample cleanup
has given satisfactory results (see Box 13.1 for criteria when to perform cleanup).
The establishment of a versatile cleanup method would give an additional advantage to

the analysis of microcystins in other matrixes such as biological samples, fish and
shellfish. Furthermore, it would give useful information for preparative separation of
microcystins and their degradation products. Octadecyl silanised (ODS) silica gel has
been employed extensively to facilitate sample cleanup and trace enrichment of
microcystins and nodularin because it retains the toxins and allows interference
compounds to pass through (Lawton et al., 1994a; Harada, 1996).
Concentration and cleanup method for microcystins and nodularins in water samples
Apparatus
• Porcelain filter funnel, 110 mm diameter and Buchner flask, 1 litre or similar
• Filter disks, GP/C, 110 mm
• Measuring cylinder, 500 ml
• Pipettes, various
• Water vacuum pump
• Glass bottles, 500 ml
• Vacuum manifold
• Solid phase extraction cartridges, 1 g trifunctional, end-capped C18
• PTFE tubing and cartridge adapters
• Glass sample tubes
• Drying hot block (45 °C) with blow-down nitrogen
• Microcentrifuge tube (1.5 ml)
Reagents
• Sodium thiosulphate solution, 1 g Na2O3.5H2O made up to 100 ml with distilled water
• Aqueous methanol, solutions containing 10, 20 and 30 per cent methanol in water
• Trifluoroacetic acid solution, 10 per cent TFA v/v in water
• Trifluoroacetic acid solution, 0.1 per cent TFA v/v in methanol
• Methanol
• Water
Note: all reagents should be analytical quality
Procedure
1. Mix the water sample by inverting the container several times, then measure a 500 ml
portion of the sample and filter it gently through a GF/C filter disc. The filter can be
retained and extracted as described above to determine the particulate microcystin
concentration.
2. Add 0.1 ml sodium thiosulphate solution to eliminate free residual chlorine. Shake the
water sample vigorously and let it stand for a few minutes, then add 5 ml of the 10 per
cent TFA and mix before passing the sample through a GF/C filter disc.
3. Place the sample in a 500 ml glass bottle, add 5 ml methanol and mix. The sample is
now ready for solid phase extraction (SPE).
4. Solid phase extraction cartridges are prepared by attaching them to the vacuum
manifold system and then conditioning them with 10 ml methanol followed by 10 ml
water, ensuring that the cartridge does not become dry at any time. The methanol and
water eluates are discarded.
5. Using PTFE tubing and adapter, attach a tube from the bottle containing the water
sample to the top of the SPE cartridge. The vacuum draws the sample through the
tubing and through the cartridge. The water is not collected but allowed to run to waste
through the water pump.
6. Once all of the water sample has passed through it, the cartridge is washed with 10 ml
of the 10 per cent methanol followed by 10 ml of the 20 per cent methanol and then
finally washed with 10 ml of the 30 per cent methanol. The eluate from the three washes
is discarded.
7. The cartridge is eluted with 3 ml of 0.1 per cent TFA in methanol. This is collected in a
sample tube and dried on a hot block (45 °C) under a gentle stream of nitrogen gas.
8. Samples are resuspended in 0.1 ml methanol and placed in a microcentrifuge tube. A

further 0.1 ml of methanol is used to rinse the sample tube and this is combined with the
first aliquot.
9. This sample can now be analysed or it can be dried and stored in the freezer until
required.
10. It is recommended that this procedure is carried out in duplicate for each water
sample analysed.
Figure 13.3 HPLC profiles of water samples containing microcystins.
A. A toxic fraction from a bloom sample and of fractions from lake water (1 µg
each of microcystin-RR, -YR, -LR added);
B. Before cleanup with silica gel cartridge;
C. After cleanup with silica gel cartridge

Notes
• Different systems can be used to pass the water sample through the SPE cartridge but
a specially designed vacuum manifold apparatus is available for use with SPE cartridges.
Most of these systems are designed so that a number of cartridges can be attached at
one time (e.g. 10 or 20). Automated systems are also available, and peristaltic pumps
have also been used successfully, but are not as practical.
• It is vitally important that all apparatus used is thoroughly cleaned to ensure no

external contamination. Equipment used for trace enrichment should be dedicated to this
procedure only, in order to reduce the opportunities for contamination.
Some modifications to the single SPE method have been proposed with promising
results. Figure 13.3A shows the High Performance Liquid Chromatography (HPLC)
chromatogram of a toxic fraction from a cyanobacterial bloom sample; the microcystins
have been detected with limited interference. However, this method is not always
effective for analysis of the toxins in raw water samples because of the occurrence of
serious background peaks as shown in Figure 13.3B. To eliminate effectively the
impurities, Tsuji et al. (1994) established a tandem cleanup method using ODS silica gel
and silica gel cartridges as follows. The water sample was first applied to an ODS silica
gel cartridge and the desired fraction was eluted with 90 per cent aqueous methanol.
Next, the resultant eluate, in 100 per cent methanol, was applied to a silica gel cartridge.
After washing with methanol, the toxin-containing fraction was obtained by elution with
10 per cent water-0.1 per cent TFA in methanol. Figure 13.3C shows the chromatogram
after the silica gel cleanup, indicating that most of the impurities can be removed. Tsuji
et al. (1996) used HPLC with ultra violet (UV) detection and Liquid
Chromatography/Mass Spectrometry (LC/MS) combined with this tandem cleanup
system to report the intracellular and extracellular microcystin levels between 1992 and
1995 for Japanese lakes. Microcystins-LR, -YR and -RR were detected at 0.02-2.64 µg l-
1
in cell-free water and at 0.02-378 µg l-1 in the cells during this period.
Recently, an immunoaffinity purification method has been developed using an anti-

microcystin-LR monoclonal antibody (Kondo et al., 1996). This cleanup method was
found to be remarkably effective in the removal of contaminants in the hepatic cytosol
and enabled the analysis of microcystins and their metabolites, formed in vivo in mouse
and rat livers, by HPLC and Frit-FAB LC/MS. Figure 13.4 shows the HPLC profiles of a
cytosolic extract from mouse liver spiked with 5 µg each of microcystins-RR and -LR.
Before the immunoaffinity purification, the spiked microcystins could not be accurately
quantified due to the many impurities (Figure 13.4A), whereas after the immunoaffinity
purification, the impurities were effectively eliminated and the peaks of the microcystins
were clearly detected (Figure 13.4B). Although this immunoaffinity column is not
commercially available, it has the potential to enable significant progress in a number of
important areas of research; for example in metabolism studies and in elucidation of the
fate of microcystins in the environment.
Research into appropriate cleanup methods for other cyanotoxins has still to be carried
out, although a similar approach to that which has been developed for microcystins may
prove useful.
13.3 Toxicity tests and bioassays
There have been many biological detection methods developed for cyanotoxins that use
the bioactivity of the toxins (Table 13.1), such as potent hepatotoxicity, neurotoxicity,
cytotoxicity, enzymatic activity and immunological interactions. However, for many years,
the mouse bioassay alone has been used to determine bloom toxicity. Although this
bioassay provides a measure of the total toxicity (response) within a few hours, it is
generally not very sensitive or specific. Considerable research efforts have been made
to find suitable alternative methods to the mouse bioassay as a routine monitoring assay
for cyanotoxins and many novel and sensitive methods have become available in recent
years. However, no single method is currently available to replace the mouse for the
detection of all cyanotoxins using a single assay, and further validation and comparison
of methods is needed before general recommendations on their application can be given
(see also Box 13.2).
Figure 13.4 HPLC profiles of a cytosolic extract from mouse liver spiked with 5 µg each
of microcystin-RR and -LR.
A. Extract after heat-denaturation, pronase digestion, and ODS silica gel cleanup;
B. Extract after further immunoaffinity purification
Different bioassays are described below. Users are advised to test their suitability for
locally-prevalent cyanotoxins with other methods, such as chemical toxin analysis (see
section 13.4). Sensitivity and selectivity are important criteria for the selection of
methods. Figure 13.1 shows the sensitivity of three different biological methods in
relation to six methods of chemical analysis.
13.3.1 Mouse bioassay
Male Swiss Albino mice are the most used animals for toxicity testing for cyanotoxins.
Toxicity is tested by intraperitoneal injection (i.p.) of 0.1-1.0 ml of a lysate of
cyanobacteria prepared either by sonication or by freeze-thawing of a cell suspension
which has been sterilized by membrane ultra-filtration. Samples can be suspended in
water or physiological saline solution which is preferred if the volume to be injected is 0.5
ml or greater.
Table 13.1 Bioassays for the detection of cyanotoxins
Cost2
Method Toxins1 Comments Reference(s)
Cap. Con. Pers.
Vertebrate
Mouse M,N,A, Requires licence, not permitted Falconer, 1993
A(s),C,S in some countries
Invertebrate
Brine shrimp M,N L L(H) M Commercial kit available but Kiviranta et al.,
increases cost 1991;
Campbell et al.,
1994
Daphnia sp. M,N L L H Culturing is labour intensive Lawton et al.,
1994b
Thamnotox M,A,C(?) L H M Commercial kit available but Kozma, 1997
requires full evaluation for
cyanotoxins
Mosquito M L L M Difficult to handle Kiviranta et al.,
1993
Fruitfly M,N L L L Easy to culture Swoboda et al.,
1994
Locust S L M L Easy to handle McElhiney et al.,
1998
House fly S L L L Difficult to administer toxin Ross et al., 1985
Bacterial
Microtox M,N H H L No correlation Lawton et al.,
1994b
Serratia sp. M,S L L M Poor correlation Lawton et al.,
1994b
Biochemical
PPase M,N Very sensitive
inhibition
Radioactive M M L Requires special facilities Holmes, 1991
Colorimetric L/M M L Requires purified enzyme An and
Carmichael, 1994
AChE A(s) L/M L L Only alternative bioassay for Mahmood and
A(s) may react with OP3 Carmichael, 1987
pesticides
ELISA Very sensitive
Polyclonal M,N M H L Reactivity for variants may vary Chu et al., 1989
Monoclonal M,N M H L Reactivity for variants may vary Ueno et al., 1996
Polyclonal S M H L Variable cross-reactivity, does Cembella et al.,
not detect C-toxins 1995
Mammalian cells
Hepatocytes M,N M M M Sensitive and rapid bioassay Heinze, 1996
V79 fibroblasts M H H H Some false negatives observed Lawton et al.,
1994b
Neuroblastoma S H H H Requires careful Cembella et al.,
standardisation 1995;
Gallacher and
Birkbeck, 1992
ELISA Enzyme linked immuno sorbent assay
AChE Acetylcholinesterase
1
M Microcystins; N Nodularins; A Anatoxin-a A(s) Anatoxin-a(S); C Cylindrospermosin;
S Saxitoxin
2
Cap. Capital; Con. Consumable; Pers. Personnel; L Low; M Medium; H High
3
OP Organophosphorus
Box 13.2 When are bioassays needed?
Bioassays are recommended if any of these conditions are fulfilled:
• A laboratory can easily establish them, but has little or no access to adequate equipment and
expertise for establishing physicochemical analysis, or little means of subcontracting analysis.
• There is indication of cyanotoxins other than (or in addition to) the known toxicants.
• Cyanobacterial taxa dominate, the toxins of which have not yet been well studied and which
may therefore contain unknown toxic metabolites.
• Confirmation of results from physicochemical analysis is required, especially to confirm

bioactivity.
• Validation of physicochemical methods by an alternative method is desired.
Mice should be observed for 24 h and then killed by an approved method (Falconer,
1993). The observation period must be extended to seven days where
cylindrospermopsin is suspected and animals should be injected with sterile samples.
This toxin demonstrates protracted symptoms which result from progressive organ
failure, specifically liver and kidneys. At the end of the observation period post-mortem
examination of tissue injury is performed. The observed symptoms and the results of the
post-mortem are used to determine which cyanotoxin is present (see Chapter 3 and 4 for
toxicology). However, where more than one type of cyanotoxin is present, the more
rapid-acting toxin may mask other symptoms.
Toxicity is expressed as LD50 mg cell dry weight per kg mouse body weight. Values are
classified as follows (Lawton et al. 1994b):
> 1,000 non toxic
500-1,000 low toxicity
100-500 medium toxicity
< 100 high toxicity
Note that in some countries non-toxic limits may be at least 2,000 mg cell dry weight per
kg mouse body weight.
13.3.2 Invertebrate bioassays
A number of invertebrates have been investigated for use in routine bioassays for
cyanotoxins. Of these, the brine shrimp (Artemia salina) has been the most popular,
because no culture maintenance or specialist equipment is required. Brine shrimps have
been exploited for many years for the detection of toxic secondary metabolites and are
now commercially available as standardised test kits (although use of the kits increases
the cost of performing the assay). Brine shrimp eggs are readily available from biological
supply companies and can be stored for several years at -20 °C without loss of viability.
Brine shrimp bioassay
Apparatus
• Conical flasks, glass, 250 ml

• Microtitre plates, 96-well
• Incubator, 25 °C
• Desk lamp
• Pasteur pipette and bulb
• Pipette, 0.1 ml
• Universal bottle, 25 ml
• Dissecting microscope with low power objective
Reagents
• Brine shrimp eggs, stored in freezer
• Brine shrimp medium (BSM) stock solution, composed of:
Sodium chloride (NaCl) 300 g
Calcium chloride dihydrate (CaCl.2H2O) 3 g
Magnesium chloride hexahydrate (MgCl2.6H2O) 15 g
Magnesium sulphate heptahydrate (MgSO4.7H2O) 5 g
Potassium chloride (KCl) 8 g
Glycine 60 g
Disodium glycerophosphate 30 g
• Methanol or formalin
The stock solution chemicals are dissolved in 1.25 litres of distilled water, dissolving
each chemical separately in the order shown. It is important to add the disodium
glycerophosphate last to prevent an insoluble precipitate occurring. The stock solution
should be stored in a brown glass bottle in the refrigerator (~4 °C).
Procedure
1. Mix 20 ml of BSM stock solution with 140 ml distilled water in a 250 ml conical flask.
2. Add 100 mg of brine shrimp eggs and incubate at 25 °C until they hatch (usually 36-
48 hours). Once a regular hatching time has been established try to keep it constant.
3. Separate the hatched larvae from unhatched eggs and egg cases by transferring
them into a shallow dish, e.g. Petri dish or slightly deeper.
4. Position a desk lamp to one side of the dish and allow the larvae to concentrate
towards the light.
5. Using a Pasteur pipette carefully, using slow steady suction on the pipette bulb,
collect the hatched larvae.
6. Place larvae in a universal bottle and repeat the collection process until most of the
hatched larvae have been collected.
7. If a substantial number of unhatched eggs have been collected in error, the separation
can be repeated by placing the collected larvae in a clean dish and illuminating.
8. Using a mechanical pipette with plastic tips (first use a sharp blade to remove the first
few mm of the tip) pipette 0.1 ml of the larvae suspension into a microtitre well. Make
sure the larvae suspension is regularly mixed because they tend to cluster rapidly in the
bottle. Mixing is easier if the bottle is only half full.
9. Examine the well containing larvae under the microscope to determine the
approximate number of larvae in 0.1 ml. Ideally this should be between 15 and 25 larvae.
Dilute the sample with fresh media if the larvae are too concentrated. Repeat the
pipetting a few times to ensure reasonably reproducible numbers of larvae are being
deposited in each well and that they are alive and look healthy.
10. Fill as many wells with larvae suspension as will be required to carry out the
bioassay.
11. To determine the toxicity of cyanobacterial cells, the sample must first be extracted.
This can be done as described above (section 13.2.1) from filter discs, but if the final
extract is in methanol it must be diluted in BSM so that the final concentration of
methanol does not exceed 5 per cent in the test well.
12. A dilution series of each sample should be prepared and 0.1 ml of the test solution
added to the larvae in the wells in triplicate. The microtitre plate is then incubated for 18
hours at 25 °C after which the percentage mortality is calculated.
13. First, with the help of a microscope, count the number of dead or immobilised larvae
in each well, then add a few drops of formalin or methanol and wait until all the larvae
are dead. Now count the total number of larvae in each well and calculate the
percentage mortality using a mean of the three replicates.
14. By plotting the concentration of cyanobacteria against the percentage mortality the
LC50 value (i.e. the concentration of cell extract which caused 50 per cent mortality) for
each sample can be determined.
Notes
• Care must be taken to exclude particulate material from the assay and therefore
extracts must be filtered or centrifuged.
• High concentrations of cell extract cause false positives, probably due to oxygen
depletion caused by the activity of bacteria, therefore sample cleanup (see section
13.2.2) is necessary to detect low concentrations of toxin successfully. Sample cleanup,
e.g. the use of SPE, can also enhance the specificity of the assay because it selectively
concentrates microcystins and nodularins, reducing interference by other compounds.
When 21 hepatotoxic bloom samples were assessed by this assay the results compared
very favourably with both mouse bioassay and HPLC (Lawton et al., 1994b). This assay
has not been fully evaluated for toxins other than microcystins, although there does
appear to be a correlation between anatoxin-a content and toxicity.
Daphnia bioassays can successfully detect microcystins, although standardised culturing

is extremely labour intensive (Baird et al., 1989). A commercially prepared test kit,
similar to the brine shrimp assay in that it uses the resting stage of an aquatic
invertebrate, Thamnocephalus platyurus has been found to be sensitive to a number of
cyanotoxins (Kozma, 1997). These kits are relatively expensive and have a limited shelf-
life (6 months), although the standardised format of this bioassay leads to highly
reproducible results and low inter-laboratory variability (Kozma, 1997). The use of
mosquito adults and larvae have both been investigated as potential bioassays (Turell
and Middlebrook, 1988; Kiviranta et al., 1993). Adults were injected and larvae
immersed in aqueous extracts. Both methods were relatively sensitive but have not been
widely adopted due to the difficulties of handling this organism. The other insect which
has been shown to detect microcystins successfully in bloom samples is the fruit fly
(Drosophila melanogaster) (Swoboda et al., 1994). These organisms are easy to
maintain in the laboratory, with no special equipment required. Toxin is administered
orally by adding filter discs spotted with sample plus sucrose to tubes containing pre-
starved (24 h) flies. The flies were not, however, sensitive to neurotoxic Aphanizomenon
(Swoboda et al., 1994).
Two invertebrate assays have been investigated for the detection of saxitoxins. Firstly,
adult house flies injected with purified toxins and natural samples (shellfish extracts)
gave results which compared well with toxicity determined by mouse. However, the flies
are difficult to handle and require microinjection (1.5 µl) which is difficult to administer
(Ross et al., 1985). A locust bioassay has recently been found to detect saxitoxins
successfully in a range of samples, namely cyanobacteria and shellfish (McElhiney et al.,
1998). Locusts are easy to handle because they can be readily immobilised by holding
their rear legs. Samples are administered by injection (10 µl) and results are obtained
within 90 minutes (McElhiney et al., 1998). The LD50 for pure saxitoxin was 8 µg g-1 but
the bioassay was not sensitive to microcystin-LR or anatoxin-a.
Locust bioassay for saxitoxins
Apparatus
• 10-25 pi syringe, e.g. type used for GC analysis

• Desert locust (Schistocerca gregaria), male early fifth instar
• Plastic container, 500 ml, foil covered with a number of small air holes
Reagents
• Simple saline, 7.5 g sodium chloride (NaCl) and 0.37 g potassium chloride (KCl)
dissolved in 1 litre distilled water
Procedure
1. Prepare a dilution series in simple saline of the test sample. It has been found that
saxitoxins can be extracted well from cyanobacterial cells with acidified methanol (TFA
added). Samples can be dried then resuspended in saline for use in the bioassay.
2. Pick up locust by hind legs and inject 10 pi along the abdomen, (parallel to the body)
between the second and third segment. Inject three locusts for each concentration and
inject controls with saline only.
3. Place each locust in a clear plastic container and observe for 90 minutes.
4. Death or the inability to self-right when placed on their back is recorded as a positive
result.
5. The time of death is an indication of saxitoxin concentration.
13.3.3 Bacterial bioassays
Bacterial bioassays have been investigated to determine if they can provide simple
routine methods for cyanotoxin detection. The one that has received the most attention
is the Microtox bioluminescence assay which indicates toxicity by a reduction in the light
emitted by the test bacterium (Photobacterium phosphoreum). Initial investigations
suggested that this system may be suitable for the rapid detection of microcystins in
bloom samples (Lawton et al., 1990) although more detailed analysis revealed that the
assay responded to unknown components of cyanobacterial extracts other than
microcystins (Campbell et al., 1994). Several studies have now been published that
clearly indicate there is no correlation between response in the Microtox assay and
cellular content of the known cyanotoxins (Lawton et al., 1994b; Vezie et al., 1996).
A second bacterial bioassay which used the inhibition of pigment (prodigiosin) formation
in Serratia marcescens as an indication of toxicity has been proposed by Dierstein et al.
(1989). This bioassay was thought to be useful for saxitoxins and microcystins. However,
like the Microtox system, little correlation was found between actual content of known
cyanotoxins and inhibition of pigment formation (Lawton et al., 1994b).
13.3.4 Biochemical assays
The protein phosphatase inhibition assay is a sensitive screening method for

microcystins and nodularins which uses the biochemical activity of these toxins. One
version is based on the quantitation of 32P-phosphate released from a radiolabelled
substrate (Holmes, 1991; Lambert et al., 1994) by the activity of the protein phosphatase
enzyme (PP1 and PP2A). It is sensitive to sub-nanogram levels of microcystin and is a
rapid assay allowing the analysis of many samples in a few hours. The method has been
applied to detect microcystin class compounds in the marine environment, to extracts of
liver tissue taken from Atlantic salmon afflicted with netpen liver disease (Andersen et al.,
1993) and to hydrophobic microcystins from freshwater cyanobacteria (Craig et al.,
1993). The method has also been successfully used for quantitation of microcystins in
drinking water before and after water treatment (Lambert et al., 1994). The detected
amounts in raw and treated waters were estimated to be 0.12-0.87 and 0.09-0.18 µg l-1,
respectively. Although this method has been widely used in research, there is a
reluctance in adopting it for the routine monitoring of microcystins because of the
requirement to use radioactivity which necessitates specialised laboratory equipment
and regulations.
An and Carmichael (1994) have used a colorimetric protein phosphatase inhibition assay
which avoids the complications of using radioactive materials. Isobe et al. (1995)
reported a firefly bioluminescence system for the detection of protein phosphatase 2A
inhibitors, in which luciferin phosphate is hydrolysed to luciferin and inorganic phosphate
by protein phosphatase 2A. The use of the protein phosphatase inhibition assay is
extremely helpful to confirm biological activity, and hence toxicity, of microcystins in
environmental samples. The non-radioactive bioassay may therefore be Used
increasingly for the routine screening of water samples, as shown recently by Ward et al.
(1997).
The biochemical activity of anatoxin-a(S) can be exploited in an enzyme-based assay to

detect the inhibition of acetylcholinesterase (AChE), thereby providing an indication of
the presence of this toxin (Mahmood and Carmichael, 1987). This is a sensitive method
and is the only alternative to the mouse bioassay currently available for this toxin. The
assay is not selective because it will also detect other toxicants, such as
organophosphorus-based pesticides.
13.3.5 Immunological detection
The Enzyme-Linked Immuno Sorbent Assay (ELISA) technique is currently the most
promising method for rapid sample screening for microcystins because of its sensitivity,
specificity and ease of operation. Monoclonal antibodies raised against microcystin-LA
were initially developed by Kfir et al. (1986) and offered a simple approach to a general
immunoassay for microcystins. An ELISA technique was subsequently developed by
Chu et al. (1989) as a practical method. This assay is based on polyclonal antisera
raised in rabbits against bovine serum albumin conjugated to microcystin-LR. The
antisera showed good cross-reactivity with microcystins-LR, -RR, -YR and nodularin, but
less with -LY and -LA. The sensitivity of the assay showed approximately 50 per cent
binding at a toxin concentration of 1 ng ml-1 which is appropriate for normal water quality
testing. In fact, this method has been successfully employed for quantitation of
cyanobacterial hepatotoxins in domestic water supplies and biomass extracts with
detection limits of 0.2 µg l-1 and 0.25 µg g-1 for water and biomass samples, respectively
(Chu et al., 1990).
Recently, Nagata et al. (1995) produced six monoclonal antibodies against microcystin-
LR. Among them, M8H5 antibody showed cross-reactivity with microcystin-RR (106 per
cent), microcystin-YR (44 per cent), microcystin-LA (26 per cent), [D-Asp3]microcystin-LR
(51 per cent), [Dha7] microcystin-LR (48 per cent), glutathione conjugate of microcystin-
LR (47 per cent), monomethyl ester of microcystin-LR (30 per cent), nodularin (46 per
cent) and 6(Z)-ADDA microcystin-LR (< 4 per cent). Although the epitope of this antibody
is not clear, the importance of the Adda moiety for antibody binding has been indicated.
It should be noted that this antibody also reacts to the non-toxic monomethyl ester of
microcystin-LR giving a false positive from the toxicological point of view. Using this
monoclonal antibody, a more sensitive competitive ELISA method has been developed
by Ueno et al. (1996) with detection limits of 0.05 µg l-1 for water samples. This method
has been successfully applied to detect microcystins in drinking water in China (Box
13.3).
Box 13.3 Application of the ELISA method to the detection of microcystins in drinking
water in China
Using an ELISA method, Ueno et al. (1996) analysed microcystin concentrations in drinking water
collected in Haimen and Fusui in China, where the rates of primary liver cancer (PLC) do not
correlate with PLC-causing agents such as aflatoxin and hepatitis-B virus. It had also been
observed that people who drank pond and ditch water had a higher risk of PLC than people who
drank well water. The authors investigated the levels of microcystins in four types of water
(pond/ditch, river, shallow well and deep well) collected from 989 different sampling sites in
Haimen. The results showed a positive detection of microcystin for 17,32,4 and 0 per cent of the
total samples of pond/ditch, river, shallow well and deep well water, respectively. The average
microcystin concentration in the pond/ditch and river water were 101 and 160 pg ml-1,
respectively, which were significantly higher than those of the shallow and deep well water.
Among the samples examined, two samples from the river showed microcystin levels over 1,000
pg ml-1. These data suggested that microcystin in drinking water from ponds/ditches and rivers, or
both, is one of the risk factors for the high incidence of PLC in China. Furthermore, the results
indicate that ELISA can be applied successfully to the monitoring of microcystins in environmental
samples.
Commercially, a polyclonal ELISA kit is available for microcystins. The antibodies are
fixed to the walls of the wells of a microtitre plate. The first step involves binding of the
calibrators (a non-toxic microcystin-LR surrogate at 0.1, 0.4 and 1.6 µg l-1) a negative
control and the samples to the antibodies in the wells. This is followed by addition of a
microcystin-enzyme conjugate which binds to the remaining antibodies. After thorough
rinsing, the concentration of bound enzyme is measured colourimetrically in an ELISA
plate reader. The microcystin concentration is inversely proportional to the colour
intensity.
ELISA method for microcystins
Apparatus
• Filtration equipment for samples containing particles

• Automatic 100 µl pipette
• Timer
• Multichannel pipette for washing the microtitre plate
• Apparatus for shaking the microtitre plate
• ELISA reader with filter at 450 nm
Reagents
• ELISA test kit
• Distilled water
Procedure
1. Water samples to be analysed are treated twice by freeze-thawing followed by
filtration through membrane or glass fibre filters.
2. Samples or standards are first mixed with antibody (M8H5) solution and then added to
a 96-well microtitre plate that is pre-coated with a microcystin-LR bovine serum albumin
conjugate.
3. After washing, bound monoclonal antibody is detected with horseradish peroxidase-

labelled goat anti-mouse IgG and its substrate (0.1 mg ml-1 of 3,3',5,5'-
tetramethlybenzidine, 0.005 per cent H2O2 in citrate buffer).
4. The optical density is measured at 450 nm and the microcystin concentration

determined from a standard competitive curve of microcystin-LR.
Development of immunodiagnostic systems for the detection of saxitoxins have primarily

been aimed at replacing the mouse bioassay for the routine monitoring of shellfish from
the marine environment and are discussed in more detail elsewhere (Cembella et al.,
1995). Both polyclonal and monoclonal antibodies have been produced, although none
have shown cross-reactivity with all the known variants. Antibodies tend to be raised to
saxitoxin mainly because it has been the most extensively studied and because it is also
the most readily available; hence methods reliably detect this variant but most notably
fail to cross-react with neosaxitoxin which is of similar toxicity. Methods continue to be
developed (e.g. Kralovec et al., 1996) and may provide a suitable routine monitoring
system in the future.
13.3.6 Mammalian cells
Bioassays using mammalian cells have received attention as suitable replacements for
mouse toxicity tests. The well documented fact that microcystins have caused acute liver
damage has prompted studies using hepatocytes (liver cells). Freshly isolated rat
hepatocytes were first investigated by Aune and Berg (1986) who reported good
correlation between toxicity measured by leakage of the enzyme lactate dehydrogenase
(LDH) from hepatocytes and results from mouse bioassay. More recently this method
has received renewed interest due to legislative restrictions in the use of the mouse LD50
(Heinze, 1996). Isolated rat hepatocytes have been incubated with pure toxin or bloom
extracts for 4 h and 20 h and then the viability has been assessed using the MTT ((3,4,5-
dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide) test. Toxicity was found to be time-
dependent with the LC50 for microcystin-LR being 0.20 µg ml-1 at 4 h reduced to 0.05 µg
ml-1 after 20 h. Different LC50 values were found for microcystin variants, most notably
microcystin-RR was found to be at least an order of magnitude less toxic in this assay.
This is consistent with in vivo toxicity data.
An in vitro fibroblast cytotoxicity assay for microcystins has been investigated by Codd et
al. (1989) as a replacement for the mouse bioassay. The assay, using V79 hamster
fibroblast cells, responded to microcystins with the results correlating reasonably well
with those of the mouse bioassay (Lawton et al., 1994b). However, the assay
demonstrated a number of false positives and, of more concern, false negatives were
also recorded.
Box 13.4 When is physicochemical analysis needed?
Physicochemical analysis of cyanotoxins is recommended if:
• Cyanobacterial species composition or bioassay results indicate which toxins to look for.
• Unambiguous identification of toxins is required.
• Quantification of toxins is required.
• Rapid screening of a large number of samples is required, especially for regular monitoring of
sites where the toxin patterns are well established.
• Low toxin concentrations which may not be detected by a bioassay are expected (e.g. in
drinking water).
• New toxic cyanobacterial metabolites are to be identified.
Another cell-based assay proposed in the past has used blood cells, with agglutination
of the cells being reported as an indicator of microcystin level (Carmichael and Bent,
1981). Although this bioassay did appear promising, it has since been found to be a poor
indicator for microcystins.
Two in vitro cell bioassays have been found to be successful in detecting saxitoxins and,
like the immunoassay, they were developed primarily for monitoring toxins in shellfish.
First, a neuroreceptor binding assay was developed that uses radiolabelled saxitoxin
and works on the basis of competitive displacement (Davio and Fontelo, 1984). The
initial protocol has subsequently been refined (Doucette et al., 1994) and the data
obtained correlate well with the mouse bioassay (Cembella et al., 1995). A
neuroblastoma cell line technique for sodium channel blocking activity has also been
developed (Gallacher and Birkbeck, 1992; Jellett et al., 1992). This assay, which is
currently undergoing evaluation for its suitability as a replacement for the mouse in
shellfish monitoring, is now available as a commercial test kit.
13.4 Analytical methods for cyanotoxins

Analytical methods use the physicochemical properties of cyanotoxins such as
molecular weight, chromophores and reactivities due to the functional groups in the
molecules. Physicochemical methods used for cyanotoxin detection are summarised in
Table 13.2, which also indicates that the initial capital expenditure to establish most of
these methods is high. Appropriate use of such methods is discussed in Box 13.4.
Table 13.2 Physicochemical methods for the detection of cyanotoxins
Cost2
Method1 Comments Reference(s)
Cap. Con. Pers.
Microcystins and nodularins
HPLC- H M L UV spectra can give tentative id Lawton et al.,
PDA 1994b
LC/MS VH M M/L A number of different interfaces; mass Kondo et al.,
confirmation; can have PDA 1992;
Edwards et al.,
1992
TLC L L M Qualitative; requires standards and further Harada, 1996
confirmation of toxins
MMPB H/VH M M Detection by GC-MS or LC-MS detects total Sano et al., 1992;
microcystin/nodularin Harada et al.,
1996
MALDI VH L M/L Initially poor but recent developments have Erhardt et al,
improved accuracy 1997
CE-MS H L M Requires further development but has future
promise
NMR VH M M/H Can characterise cyanotoxins; needs mg Botes et al., 1984;
quantities and expert interpretation Harada, 1996
Anatoxin-a and homoanatoxin-a
HPLC- H M L Characteristic UV spectra Edwards et al.,
PDA 1992
GC-MS H M L Characteristic ion spectra Smith et al., 1987
GC-ECD H M L Requires sample cleanup Stevens et al.,
1988
LC/MS VH M M/L Sensitive and specific Harada et al.,
1993
Anatoxin-a(S)
HPLC H M L Very poor chromophore, not suitable for routine Matsunaga et al.,
detection 1989
Cylindrospermopsin
HPLC- H M L Lack of available standards; give characteristic Harada et al.,
PDA UV spectra 1994;
Hawkins et al.,
1997
Saxitoxins
HPLC- H M H Precolumn derivatisation; poor stability of Lawrence et al.,
pre derivative 1995
HPLC- H M M Three solvent systems required to analyse for Oshima et al.,
post all variants 1995
LC/MS VH M M/L Best method for all variants but equipment cost Quilliam et al.,
can be prohibitive 1989;
Hines et al., 1993
CE-MS H L M Poor detection limits; needs further Pleasance et al.,
development 1992;
Lock et al., 1994
1
See text for an explanation of the different methods listed
2
Cap. Capital; Con. Consumables; Pers. Personnel; L Low; M Medium; H High; VH Very
high
Another approach for further confirmation and identification of microcystins has been
proposed using a LC-linked protein phosphatase assay (Boland et al., 1993; Chen et al.,
1993). Essentially, this method makes use of HPLC to separate and identify tentatively
the cyanotoxins present, and then monitors the bioactivity of the detected peaks using
the protein phosphatase inhibition assay. This provides excellent confirmation of protein
phosphatase inhibiting cyanotoxins, especially in complex samples.
13.4.1 Detection methods for microcystins and nodularins
The majority of analytical methods have been developed primarily for microcystins
although because both nodularins and microcystins show similar physicochemical
properties, nodularins can easily be analysed by the same methods. The most
commonly-used analytical system for this class of cyanotoxins is HPLC (see Box 13.5).
Combined with UV detection, HPLC has been used extensively for the detection of
microcystins, but because this method relies on retention time for identification,
microcystin standards are required (Harada, 1996). Detection by UV can be made more
specific by using a photodiode array (PDA) UV detector (Lawton et al., 1994a) but it has
very limited ability to identify individual microcystins because almost all microcystins
show a similar UV spectrum. Recent advances in detector hardware can now provide
high resolution spectra that detect very slight variations in chemical composition and can
be used in conjunction with advanced spectral matching software. These developments
may assist in the identification of microcystins by spectral match data in conjunction with
retention times. However, a fundamental problem still exists in the availability of
standards. With over 60 microcystins known, it is currently impossible to create a
definitive spectral library, which is a limiting factor when using this method to identify
unknown microcystins.
Typical HPLC analysis uses a reverse-phase C18 silica column with separation
achieved over a gradient of water and acetonitrile, both containing 0.05 per cent
triflouroacetic acid (TFA). The gradient has to cover a sufficient range of polarities (e.g.
30-70 per cent acetonitrile) to allow the analysis of all microcystins which are known to
vary considerably in their polarities. Data is gathered at 238 nm and where PDA is used
spectral information is collected between 200 and 300 nm. Use of HPLC-PDA can allow
tentative identification of microcystins and this method was found to perform very well
when over 20 samples were assessed by HPLC and compared with mouse bioassay
data. No false negatives were observed and only one false positive was reported, the
latter being attributed to a relatively low level of microcystin which failed to cause death
in the mouse bioassay (Lawton et al., 1994b). Other solvent systems have also been
successfully used including methanol/water and ammonium acetate/acetonitrile.
Box 13.5 Possibilities and limitations of HPLC with UV spectra for microcystin analysis
High pressure liquid chromatography can be used routinely to identify and quantify microcystins,
but not to differentiate between structural variants of most microcystins. For example, in a sample
from Radeburg Reservoir (1 July 1996, monospecific population of Microcystis spp.), UV-spectra
indicated nine microcystins, three of which could be identified with commercially available
standards by their retention times to be microcystin-RR (peak 1), microcystin-YR (peak 3), and
microcystin LR (peak 6). However, six other minor microcystins could not be further specified by
this method alone. For a tentative assessment of the toxicity of this sample, a "worst case"
approach was chosen by calculating the sum of all of the nine microcystins and assuming them to
be as toxic as -LR and -YR, the most toxic variants currently known.
For preliminary microcystin screening and for routine monitoring, HPLC with photodiode array
detection is an excellent approach because it efficiently provides an overview of toxin content,
and a worst-case toxicity estimate can be derived. Therefore, if a local or regional authority must
deal regularly with microcystin monitoring, establishment of HPLC techniques with photodiode
array detection of UV spectra is recommended. Further identification of microcystins can then be
performed with selected samples. In many cases, it will be advisable to subcontract this to
laboratories with specialised expertise.
Results of HPLC analysis of a sample from Radeburg Reservoir (1 July 1996, monospecific
population of Microcystis spp.) showing nine microcystins identified by their characteristic UV-
spectra (Fastner, unpublished data)
Analysis of microcystins and nodularins by HPLC-PDA
Apparatus
• Gradient HPLC system with photodiode array detection

• Data acquisition system
• Auto-sampler, recommended for high sample throughput
• C18 column e.g. 4.6 × 250 mm Symmetry (Waters)
• Column oven, 40 °C
• Pipette, 0.1-0.5 ml
Laboratory conditions
Effective ventilation, taking into account that acetonitrile is heavier than air and
accumulates at ground level.
Reagents
• Eluent A, water plus TFA, 0.5 ml TFA added to 1,000 ml water

• Eluent B, acetonitrile plus TFA, 0.5 ml TFA added to 1,000 ml water
• Helium gas
• Methanol
Note: all reagents must be high purity, HPLC grade.
Procedure
1. Prepare solvents and degas in a stream of helium gas
2. Program a linear gradient (1 ml min-1) as follows:
Time (mins) 0 10 40 42 44 46 55
Eluent A% 70 65 30 0 0 70 70
Eluent B% 30 35 70 100 100 30 30
3. Equilibrate the column at the desired temperature and gradient starting conditions.
4. Set photodiode array detector to monitor between 200 and 300 nm.
5. Running a blank sample first, i.e. injecting only methanol, helps the system settle and
ensures reproducible retention times.
6. Samples and standards are usually prepared in methanol. However, some

microcystins are less soluble in methanol (e.g. microcystin-YR). Check product
information where available. Samples should be centrifuged or filtered to remove
particulates before carrying out HPLC analysis.
7. It is advisable always to run a standard at the beginning and end of a set of analyses,
because this helps to confirm correct operation of the system and indicates the degree
of retention time drift.
8. A calibration curve should be performed when establishing the method and at regular
intervals, especially after changing a column or lamp.
9. Chromatograms are best viewed, and integration carried out, at 238 nm because this
is the absorption maximum of most microcystins and nodularins.
10. Microcystin congeners can be identified where they have the same retention time
and spectrum as a standard, but for many microcystins no standards exist. These can
be identified as microcystins, but the respective congener may only be tentatively
inferred from published retention times in relation to identified congeners. However,
advances in spectral matching software provide increased confidence in microcystin
identification by providing a numeric indication of how similar an unknown is to a range
of microcystins in a spectral library.
When further confirmation and identification of microcystins is required, more advanced

methodology must be used. Liquid chromatography/mass spectrometry (LC/MS) is a
very promising method because it enables the simultaneous separation and
identification of microcystins in a mixture (Kondo et al., 1992; Edwards et al., 1993; Poon
et al., 1993). Figure 13.6 shows the Frit-FAB (fast atom bombardment) LC/MS analysis
data of a toxin from a bloom sample collected in Japan. The toxic fraction contains
mainly two microcystins as shown by the mass chromatogram monitored at the
characteristic ion m/z 135 derived from Adda, which has proved to be useful for the
discrimination of microcystins from other types of compounds (Kondo et al., 1992). The
two peaks were readily identified as microcystins-RR and -LR according to the mass
spectra and mass chromatograms at their [M+H]+. An advanced Frit-FAB LC/MS method
using a microbore column (0.3 mm internal diameter) enabled the identification of
nanogram levels of microcystins in water and biological samples (Kondo et al., 1995,
1996). This increase in sensitivity was achieved by splitting the mobile phase between
the pump and the injector so that the total effluent could be introduced into the mass
spectrometer.
A physicochemical screening method that is based on the detection of 2-methyl-3-

methoxy-4-phenylbutyric acid (MMPB) as an oxidation product of microcystins has been
reported. The MMPB was initially prepared by Lemieux oxidation, followed by analysis
by gas chromatography (GC) with a flame ionisation detector or HPLC with fluorescence
detection (Sano et al., 1992). However, it required tedious procedures such as extraction,
cleanup, oxidation and post-treatment in order to eliminate the reagents used, and
derivatisation for GC and HPLC analysis. An improved method using ozonolysis made it
possible to reduce significantly the formation times of MMPB because the previously
required extraction, cleanup and other procedures could be entirely eliminated (Harada
et al., 1996). The resulting intact MMPB was directly analysed by thermospray (TSP)
interface LC/MS and EI-GC/MS using selected ion monitoring. This new procedure, from
the ozonolysis of the microcystins to analysis of MMPB at picomol levels, took only 30
minutes to perform. The quantification of bloom samples achieved by this method were
consistent with those obtained by HPLC analysis, showing that the method provided a
means of screening for microcystins, as well as for their accurate quantification.
Additionally, the most remarkable feature of this method is the applicability to complex
sample matrixes, including solid material such as animal tissue, without the requirement
for any complicated processing.
The structural determination of microcystins and nodularins has been carried out by
nuclear magnetic resonance (NMR) spectroscopy and recent advances in 2D NMR
techniques have proved to be essential for the structural determination of known and
unknown microcystins. Fast atom bombardment MS and liquid secondary ion (LSI) MS
give a protonated molecule [M+H]+, providing information about molecular weight with
further structural information obtained by Tandem FAB MS (FAB-MS/MS) as used in
recent studies (Namikoshi et al., 1995). However, NMR and MS (except LC/MS) usually
require relatively large amounts of sample (milligram quantities) and completely purified
microcystins, therefore they are not used in routine monitoring (Botes et al., 1984;
Namikoshi et al., 1995; Harada, 1996).
Figure 13.6 Frit-FAB LC/MS analysis of a toxic fraction from a bloom sample
collected in Japan (Data supplied by Ken-ichi Harada and Fumio Kondo)
Microcystin-RR (MW 1037)
Microcystin-LR (MW 994)
For identification of microcystins with very small sample volumes (less than 1 mm3
freeze-dried material). Matrix Assisted Laser Desorption Ionisation Time of Flight
(MALDI-TOF) has recently been developed (Erhardt et al., 1997). This method provides
the molecular mass of all of the peptides in a sample and thus gives strong indications of
the microcystin variants present. Post Source Decay (PDS) spectra may be obtained,
which are characteristic for different microcystins. A library is currently being established.
For rapid qualitative assessment of microcystins and other cyanobacterial peptides, this
method is highly promising. Quantitative assessment is not yet possible.
Different methods provide different and often complementary information, therefore

combined use of suitable methods is recommended according to the purpose and type
of data required. This is necessary because none of the methods currently available
provides all the information which may be required. Furthermore, individual laboratories
must identify the techniques that are both suitable for their analytical requirements and
use their own expertise and available technology (see Meriluoto (1997) for a recent
review of chromatographic methods for microcystins). Initial screening of samples can
check rapidly for the presence of microcystins in a small amount of sample using
sensitive and simple methods. Figure 13.1 summarised the relationship between
sensitivity and selectivity of analytical methods for microcystins, showing that different
methods provide different and complementary information.
For studies requiring enhanced precision, accuracy and sensitivity in detection of
individual toxins, a multistage procedure is required. This may comprise initial screening
of samples to check for the presence of microcystins in a small amount of sample using
sensitive and simple methods, such as bioassays (e.g. ELISA, protein phosphatase
inhibition assay) and MMPB method. The use of screening helps to reduce the number
of samples which require full analytical investigation and thus reduces the laboratory
commitment. Furthermore, rapid results from an initial screen may aid prompt regulatory
responses regarding the suitability of water for human use. If a sample is positive in a
screening test, it will be necessary to follow through with identification and quantitative
analysis. Prior to this, cleanup and sample concentration is often very important for
many samples because relatively low levels of microcystins, around 1-2 µg l-1, are
usually present in water samples. A tandem cleanup system using ODS silica gel and
silica gel cartridges, facilitates the accurate analysis of trace amounts of microcystins in
water. Finally, quantitation of microcystins is usually essential, and although the
screening methods can provide an indication of total microcystin concentrations in
samples, a separation and spectroscopic method such as HPLC with UV detection
should be applied at this final step. Availability of authentic standards would strengthen
the capability of HPLC with UV detection (preferably with photodiode array detection) but
otherwise LC/MS should be the method of choice despite the requirement of a more
specialised technique.
13.4.2 Anatoxin-a
Three methods for analysis of anatoxin-a have been reported to date: GC/MS, GC with
electron capture detection (ECD) and HPLC. One GC/MS method has been published
as a confirmation of anatoxin-a in material taken from a toxic bloom that resulted in the
deaths of 16 cows (Smith and Lewis, 1987). In the procedure, following liquid-liquid
extraction with N-acetylation, the resulting acetylated toxin was analysed by a capillary
GC/MS. Although confirmation was based on interpretation of the mass spectrum, the
detection limit was unclear. Another GC/MS method has also been established for N-
acetylated anatoxin-a (Himberg, 1989) and it was successfully employed for preliminary
characterisation of neurotoxic cyanobacteria from Finland (Sivonen et al., 1989). The
GC/ECD method provided a higher sensitivity than other analytical methods, it used an
internal standard for accurate quantification, and it could be applied to analysis of field
samples (Stevens and Krieger, 1988). Although GC/ECD requires a considerably more
complicated cleanup operation and derivatisation prior to analysis, it has been used
successfully for stability studies on anatoxin-a (Stevens and Krieger, 1991).
Derivatisation with N-pentafluorobenzylbromide was shown to achieve a sensitivity of 2.5
pg (Bumke-Vogt et al., 1996). It is possible to use HPLC with UV detection to analyse
intact anatoxin-a which has a strong absorption at 227 nm. After extraction of a
cyanobacterial suspension with chloroform, followed by re-extraction with 0.01N
hydrochloric acid, the resulting extract is separated under the following conditions:
column, ODS silica gel; mobile phase, methanol-0.01M perchloric acid (7:3); detection,
UV (227 nm). Harada et al. (1989) reported an alternative approach using a reversed
phase HPLC method with a methanol-10.01M ammonium chloride (1:9) solvent system.
Mass spectrometry is a very effective method for identification of anatoxin-a and its
derivatives. Conventional electron ionisation (El) and chemical ionisation (CI) can yield
molecular ions and protonated molecules, respectively. Ross et al. (1989) evaluated
secondary ion mass spectrometry (SIMS), GC/MS, desorption CI and TSP-MS for
detection of the toxin and its derivatives. The use of HPLC coupled with mass
spectrometry using thermospray interface (TSP-LC/MS) has also been investigated
(Harada et al., 1993). The latter method made possible a sensitive, specific and
reproducible analysis of anatoxin-a and its non-toxic oxidation product when used in
combination with a cleanup method including a solid phase extraction with a reversed
phase carboxylic acid cartridge. Using this method, trace amounts of anatoxin-a were
detected in three strains and two bloom samples collected in Japan.
13.4.3 Anatoxin-a(S)
Matsunaga et al. (1989) purified anatoxin-a(s) and determined its structure, although no
analytical method has been developed. Traditional HPLC plus UV detection is not
suitable because this cyanotoxin lacks a strong chromophore. Liquid
chromatography/mass spectrometry may prove useful, but this analytical technique has
not been evaluated yet for anatoxin-a(S).
13.4.4 Cylindrospermopsin
The first analytical method reported for this toxin consisted of a combination of a cleanup
step using HP-20 and ODS silica gel cartridges followed by HPLC with photodiode array
detection. This method was applied to a bloom sample collected in Japan which was
thought to contain Cylindrospermopsin (Harada et al., 1994). However, the authors
suggested that the mobile phase used in this study was not satisfactory because
Cylindrospermopsin showed poor retention power on the ODS column and slight tailing,
and their method requires further development to allow precise analysis.
A recently published method (Hawkins et al., 1997) describes the extraction of cells in 5
per cent aqueous acetic acid and analysis using Spherisorb ODS-2 with a 10 minute
linear gradient from 0 to 5 per cent methanol followed by a further 10 minutes at 5 per
cent methanol. This gave good retention, separation and peak shape.
Cylindrospermopsin was found to have a characteristic UV absorbance spectrum
between 200 and 300 nm, with a maximum absorbance at 262 nm.
13.4.5 Saxitoxins
The methods mentioned here have been developed primarily for the analysis of
saxitoxins in the marine environment, particularly in shellfish. However, they have been
found to be equally suitable for cyanobacterial samples. The most commonly used
analytical method for the saxitoxins is HPLC with on-line post-column oxidation and
fluorescence detection (Oshima et al., 1995). Although this method has been found to be
the most satisfactory to-date, it requires three different mobile phase systems to allow
analysis of all the saxitoxins. Furthermore, there is limited availability of analytical
standards for all saxitoxin variants, which are necessary for peak confirmation. Mass
spectrometry with FAB (Mirocha et al., 1992), TSP (Wils and Hulst, 1993), electrospray
(Hines et al., 1993) and ion-spray ionisation (Quilliam et al., 1989) has been reported for
the paralytic shellfish poison (PSP) toxins. The application of an LC/MS method for
saxitoxins has been reported, although it appears that it will be difficult to achieve
analysis of all PSP toxins within one LC/MS run. The combination of capillary
electrophoresis (CE) with ion-spray ionisation has proved to be successful for the
saxitoxins (Pleasance et al., 1992; Locke and Thibault, 1994). Analysis using CE was
found to give poor detection limits due to the very small injection volumes (< 10 nl), but it
may be possible to overcome this problem with an on-column sample preconcentration
system. Furthermore, developments in CE technology are anticipated in the near future
which will increase flow cell volume, hence enhancing detection limits.
13.5 References
An, J. and Carmichael, W.W. 1994 Use of a colorimetric protein phosphatase inhibition
assay and enzyme linked immuno sorbent assay for the study of microcystins and
nodularins. Toxicon, 32, 1495-1507.
Andersen, R.J., Luu, H.A., Chen, D.Z.X., Holmes, C.F.B., Kent, M., LeBlanc, M., Taylor,
F.J.R. and Williams, D.E. 1993 Chemical and biological evidence links microcystins to
salmon 'Netpen Liver Disease'. Toxicon, 31, 1315-1323.
Aune, T. and Berg, K. 1986 Use of freshly prepared rat hepatocytes to study toxicity of
blooms of blue-green algae Microcystis aeruginosa and Oscillatoria agardhii. J. Toxic.
Environ. Health, 19, 325-336.
Baird, D.J., Soares, A.M.V.M., Girling, A.E., Barber, I., Bradley, M.C. and Calow, P. 1989
The long-term maintenance of Daphnia magna for use in ecotoxicity tests: Problems and
prospects. In: H. Lokke, H. Tyle and F. Bro-Rasmussen [Eds] Proceedings of the First
Conference on Ecotoxicology, Lyngby, Denmark, 144-148.
Boland, M.P., Smillie, M.A., Chen, D.Z.X. and Holmes, C.F.B. 1993 A unified bioscreen
for the detection of diarrhetic shellfish toxins and microcystins in marine and freshwater
environments. Toxicon, 31, 1393-1405.
Botes, D.P., Tuinman, A.A., Wessels, P.L., Viljoen, C.C., Kruger, H., Williams, D.H. and
Santikarn, S. 1984. The structure of cyanoginosin-LA, a cyclic heptapeptide toxin from
the cyanobacterium Microcystis aeruginosa. J. Chem. Soc., Perkin Transactions, 1,
2311-2318.
Bumke-Vogt, C., Mailahn, W., Rotard, W. and Chorus, I. 1996: A highly sensitive
analytical method for the neurotoxin anatoxin-a using GC-ECD, and first application to
laboratory cultures. Phycologia, 35 (6, Supplement), 51-56.
Campbell, D.L., Lawton, L.A., Beattie, K.A. and Codd, G.A. 1994 Comparative
assessment of the specificity of the brine shrimp and Microtox assay to hepatotoxic
(microcystin-LR-containing) cyanobacteria. Environ. Toxicol. Water Qual., 9, 71-77.
Carmichael, W.W. and Bent, P.E. 1981 Hemagglutination method for detection of
freshwater cyanobacteria (blue-green algae) toxins. Appl. Environ. Microbiol., 41(6),
1383-1388.
Cembella, A.D., Milenkovic, L., Doucette, G. and Fernandez, M. 1995 In Vitro

biochemical methods and mammalian bioassays for phycotoxins. In: G.M. Hallegraeff,
D.M. Anderson and A.D. Cembella [Eds] Manual on Harmful Marine Microalgae, IOC
Manuals and Guides No. 33, United Nations Educational, Scientific and Cultural
Organization, Paris, 177-211.
Chen, D.Z.X., Boland, M.P., Smillie, M.A., Klix, H., Ptak, C., Andersen, R.J. and Holmes,
C.F.B. 1993 Identification of protein phosphatase inhibitors of the microcystin class in
the marine environment. Toxicon, 31, 1407-1414.
Chu, F.S., Huang, X. and Wei, R.D. 1990 Enzyme-linked immunosorbent assay for
microcystins in blue-green algal blooms. J. Assoc. Analyt. Chem., 73(3), 451-456.
Chu, F.S., Huang, X., Wei, R.D. and Carmichael, W.W. 1989 Production and
characterisation of antibodies against microcystins. Appl. Environ. Microbiol., 55, 1928-
1933.
Codd, G.A., Brooks, W.P., Priestley, I.M., Poon, G.K. and Bell, S.G. 1989 Production,
detection and quantification of cyanobacterial toxins. Tox. Assess., 4, 499-511.
Craig, M., McCready, T.L., Luu, H.A., Smillie, M.A., Dubord, P. and Holmes, C.F.B. 1993
Identification and characterisation of hydrophobic microcystins in Canadian freshwater
cyanobacteria. Toxicon, 31, 1541-1549.
Davio, S.R. and Fontelo, P.A. 1984 A competitive displacement assay to detect saxitoxin
and tetrodotoxin. Analyt. Biochem., 141, 199-204.
Dierstein, R., Kaiser, I. and Weckesser, J. 1989 Inhibition of prodigiosin formation in

Serratia marcescens by extracts of toxic cyanobacteria. System. Appl. Microbiol., 12,
244-248.
Doucette, G.J., Logan, M.M., Van Dolah, F.M. and Ramsdell, J.S. 1994 A sensitive,
microtiter plate-based receptor assay for paralytic shellfish poisoning (PSP) toxins. In:
J.R. Forbes [Ed.] Proceedings of the Fourth Canadian Workshop on Harmful Marine
Algae, Can. Tech. Rep. Fish. Aquatic Sci., 2016, 9.
Edwards, C., Beattie, K.A., Scrimgeour, C.M. and Codd, G.A. 1992 Identification of
at Loch Insh, Scotland. Toxicon, 30(10), 1165-1175.
Edwards, C., Lawton, L.A., Beattie, K.A., Codd, G.A., Pleasance, S. and Dear, G.J. 1993
Analysis of microcystins from cyanobacteria by liquid chromatography with mass
spectrometry using atmospheric-pressure ionization. Rapid Commun. Mass Spectrom.,
7, 714-721.
Erhardt, M., von Döhren, H., Jungblut, P. 1997 Rapid typing and elucidation of new
secondary metabolites of intact cyanobacteria using MALDI-TOF mass spectrometry.
Nature Biotech., 15, 906-909.
Falconer, I.R. 1993 Measurement of toxins from blue-green algae in water and
foodstuffs. In: I.R. Falconer [Ed.] Algal Toxins in Seafood and Drinking Water. Academic
Press, London, 165-175.
Fastner, J., Flieger, I. and Neumann, U. 1998 Optimised extraction of microcystins from
field samples - a comparison of different solvents and procedures. Wat. Res., 32, 3177-
3181.
Fernandez, M.L. and Cembella, A.D. 1995 Mammalian bioassays. In: G.M. Hallegraeff,
D.M. Anderson and A.D. Cembella [Eds] Manual on Harmful Marine Microalgae, IOC
Manuals and Guides No. 33, 213-228.
Gallacher, S. and Birkbeck, T.H. 1992 A tissue culture assay for direct detection of
sodium channel blocking toxins in bacterial culture supernates. FEMS Microbiol. Lett., 92,
101-108.
Harada, K.-I. 1996 Chemistry and detection of microcystins. In: M.F. Watanabe, K.-I.
Harada, W.W. Carmichael and H. Fujiki [Eds] Toxic Microcystis, Chemical Rubber
Company (CRC) Press, Boca Raton, Florida, 103-148.
Harada, K.-I., Kimura, Y., Ogawa, K., Suzuki, M., Dahlem, A.M., Beasley, V.R. and
Carmichael, W.W. 1989 A new procedure for the analysis and purification of naturally
occurring anatoxin-a from the blue-green algae Anabaena flos-aquae. Toxicon, 27(12),
1289-1296.
Harada, K.-I, Murata, H., Qiang, Z., Suzuki, M. and Kondo, F. 1996 Mass spectrometric
screening method for microcystins in cyanobacteria. Toxicon, 34, 701-710.
Harada, K.-I., Nagai, H., Kimura, Y., Suzuki, M., Park, H., Watanabe, M.F., Luukkainen,
R., Sivonen, K. and Carmichael, W.W. 1993 Liquid chromatography/mass spectrometric
detection of anatoxin-a, a neurotoxin from cyanobacteria. Tetrahedron, 49, 9251-9260.
Harada, H.-I., Ohtani, I., Iwamoto, K., Suzuki, M., Watanabe, M.F., Watanabe, M. and
Terao, K. 1994 Isolation of cylindrospermopsin from a cyanobacterium Umezakia natans
and its screening methods. Toxicon, 32(1), 73-84.
Harwig, J. and Scott, P.M. 1971 Brine shrimp (Artemia salina L.) larvae as a screening
system for fungal toxins. Appl. Microbiol., 21(6), 1011-1016.
35, 341-346.
Heinze, R. 1996 A biotest for hepatotoxins using primary rat hepatocytes. Phycologia, 35
(Supplement), 89-93.
Himberg, K. 1989 Determination of anatoxin-a, the neurotoxin of Anabaena flos-aquae

cyanobacerium, in algae and water by gas chromatography-mass spectrometry. J.
Chromatog., 481, 358-362.
Hines, H.B. 1993 Electrospray ionization of selected low-molecular-weight natural

biotoxins Biol. Mass Specrom., 22, 243-246.
Holmes, C.F.B. 1991 Liquid chromatography-linked protein phospatase bioassay; a
highly sensitive marine bioscreen for okadaic acid and related diarrhetic shellfish toxins.
Toxicon, 29(4/5), 469-477.
Isobe, M., Sugiyama, Y., Ito, T., Ohtani, I.I., Toya, Y., Nishigohri, Y. and Takai, A. 1995
New analysis method for protein phosphatase type 2A inhibitors using the firefly
bioluminescence system. Biosci., Biotechnol. Biochem., 59, 2235-2238.
Jellet, J.F., Marks, L.J., Stewart, J.E., Dorey, M.L., Watson-Wright, W. and Lawrence,
J.F. 1992 Paralytic shellfish poison (saxitoxin family) bioassays: automated endpoint
determination and standardisation of the in vitro tissue culture bioassay, and comparison
with the standard mouse bioassay. Toxicon, 30, 1143-1156.
Kfir, R., Johannsen, E. and Botes, D.P. 1986 Monoclonal antibodies specific for
cyanoginosin-LA: Preparation and characterisation. Toxicon, 24, 543-552.
Kiviranta, J., Abdel-Hameed, A., Sivonen, K., Niemela, S.I. and Carlberg, G. 1993
Toxicity of cyanobacteria to mosquito larvae: screening of active compounds. Environ.
Toxicol. Water Qual., 8, 63-71.
Kiviranta, J., Sivonen, K. and Niemelä, S.I. 1991 Detection and toxicity of cyanobacteria
by Artemia salina bioassay. Environ. Toxicol. Water Qual., 6, 423-426.
Komárek, J. and Anagnostidis, K. 1986 Modern approach to the classification system of

cyanophytes. Archiv Hydrobiol., Suppl. 73, Algological Studies, 43, 157-164.
Kondo, F., Ikai, Y., Oka, H., Ishikawa, N., Watanabe, M.F., Watanabe, M., Harada, K.-I.
and Suzuki, M. 1992 Separation and identification of microcystins in cyanobacteria by
frit-fast atom bombardment liquid chromatography/mass spectrometry. Toxicon, 30(3),
227-237.
Kondo, F., Ikai, Y., Oka, H., Matsumoto, H., Yamada, S., Ishikawa, N., Tsuji, K., Harada,
K.-I., Shimada, T., Oshikata, M. and Suzuki, M. 1995 Reliable and sensitive method for
determination of microcystins in complicated matrices by frit-fast atom bombardment
liquid chromatography/mass spectrometry. Natural Toxins, 3, 41-49.
Kondo, F., Matsumoto, H., Yamada, S., Ishikawa, N., Ito, E., Nagata, S., Ueno, Y.,
Suzuki, M. and Harada, K.-I. 1996 Detection and identification of metabolites of
Microcystins formed in vivo in mouse and rat livers. Chem. Res. Toxicol.. 9(8), 1355-
1359.
Kozma, A. 1997 Interlaboratory trial using Thamnotox kit for detecting cyanobacterial
toxins. Abstract, VIII International Conference on Harmful Algae, Vigo, Spain, 114.
Kralovec, J.A., Laycock, M.V., Richards, R. and Usleber, E. 1996 Immobilisation of small
molecules on solid matrices: A novel approach to enzyme-linked immunosorbent assay
screening for saxitoxins and evaluation of anti-saxitoxin antibodies. Toxicon, 34(10),
1127-1140.
Lambert, T.W., Boland, M.P., Holmes, C.F.B. and Hrudey, S.E. 1994 Quantitation of the
microcystin hepatotoxins in water at environmentally relevant concentrations with the
protein phosphatase bioassay. Environ. Sci. Technol., 28(4), 753-755.
Lawrence, J.F., Menard, C. and Cleroux, C. 1995 Evaluation of prechromatographic

oxidation for liquid chromatographic determination of paralytic shellfish poisons in
shellfish. J. AOAC Int., 78, 514-520.
Lawton, L.A., Beattie, K.A., Hawser, S.P., Campbell, D.L. and Codd, G.A. 1994b
Evaluation of assay methods for the determination of cyanobacterial hepatotoxicity. In:
G.A. Codd, T.M. Jefferies, C.W. Keevil and E. Potter [Eds] Detection Methods for
Cyanobacterial Toxins, Special Publication No. 149, The Royal Society of Chemistry,
Cambridge, 111-116.
Lawton, L.A., Campbell, D.L., Beattie, K.A. and Codd, G.A. 1990 Use of a rapid
bioluminescence assay for detecting cyanobacterial microcystin toxicity. Lett. Appl.
Microbiol., 11, 205-207.
Lawton, L.A., Edwards, C. and Codd, G.A. 1994a Extraction and high-performance liquid
chromatographic method for the determination of microcystins in raw and treated waters.
Analyst, 119, 1525-1530.
Locke, S.J. and Thibault, P. 1994 Improvement in detection limits for the determination
of paralytic shellfish poisoning toxins in shellfish tissues using capillary
electophoresis/electrospray mass spectrometry and discontinuous buffer systems.
Analyt. Chem., 66, 3436-3446.
Lund, J.W.G., Kipling, C. and LeCren, E.D. 1958 The inverted microscope method of
estimating algal numbers and the statistical basis of enumeration by counting.
Mahmood, N.A. and Carmichael, W.W. 1987 Anatoxin-a(s), an anticholinesterase from

the cyanobacterium Anabaena flos-aquae NRC-525-17. Toxicon, 25(11), 1221-1227.
Matsunaga, S., Moore, R.E. and Niemszura, W.P. 1989 Anatoxin-a(S), a potent
anticholinesterase from Anabeana flos-aquae. J. Am. Chem. Soc., 111, 8021-8023.
McElhiney, J., Lawton, L.A., Edwards, C. and Gallacher, S. 1998 Development of a

bioassay employing the desert locust (Schistocerca gregaria) for the detection of
saxitoxin and related compounds in cyanobacteria and shellfish. Toxicon, 36, 417-420.
Meriluoto, J. 1997 Chromatography of microcystins. Analyt. Chim. Acta, 352, 277-298.
Mirocha, C.J., Cheong, W., Mirza, U. and Kim, Y.B. 1992 Analysis of saxitoxin in urine
by continuous-flow fast atom bombardment mass spectrometry. Rapid Commun. Mass
Spectrom., 6, 128-134.
Nagata, S., Soutome, H., Tsutsumi, T., Hasegawa, A., Sekijima, M., Sugamata, M.,
Harada, K.-I, Suganuma, M. and Ueno, Y. 1995 Novel monoclonal antibodies against
microcystin and their protective activity for hepatotoxicity. Natural Toxins, 3, 78-86.
Namikoshi, M., Sun, F., Choi, B.W., Rinehart, K.L., Carmichael, W.W., Evans, W.R. and
Beasley, V.R. 1995 Seven more microcystins from Homer Lake cells: application of the
general method for structure assignment of peptides containing, -dehydroamino acid
unit(s) J. Org. Chem., 60, 3671-3679.
Oshima, Y. 1995 Post-column derivatization HPLC methods for paralytic shellfish

poisons. In: G.M. Hallegraeff, D.M. Anderson and A.D. Cembella [Eds] Manual on
Harmful Marine Microalgae, IOC Manuals and Guides No. 33, 81-94.
Pleasance, S., Thibault, P. and Kelly, J. 1992 Comparison of liquid-junction and coaxial
interfaces for capillary electrophoresis-mass spectrometry with application to compounds
of concern to the aquaculture industry. J. Chromatog., 591, 325-339.
Poon, G.K., Griggs, L.J., Edwards, C., Beattie, K.A. and Codd, G.A. 1993 Liquid
chromatography-electrospray ionization-mass spectrometry of cyanobacterial toxins. J.
Chromatog., 628, 215-233.
Quilliam, M.A, Thomson, B.A., Scott, G.J. and Siu, K.W.M. 1989 Ion-spray mass
spectrometry of marine neurotoxins. Rapid Commun. Mass Spectrom., 3, 145.
Ross, M.M., Kidwell, D.A. and Callahan, J.H. 1989 Mass-spectrometric analysis of
anatoxin-a. J. Analyt. Toxicol., 13(6), 317-312.
Ross, M.R., Siger, A. and Abbott, C. 1985 The house fly: An acceptable subject for
paralytic shellfish toxin bioassay. In: D.M. Anderson, J.A.W. White and D.G. Baden [Eds]
Toxic Dinoflagellates: Proceedings of the Third International Conference on Toxic
Dinoflagellates. Elsevier, Amsterdam, 433-438.
Sano, T., Nohara, K., Shirai, F. and Kaya, K. 1992 A method for microdetection of total
microcystin content in waterbloom of cyanobacteria (blue-green algae). Int. J. Environ.
Analyt. Chem., 49, 163-170.
Sivonen, K., Himberg, K., Luukkainen, R., Niemela, S.I., Poon, G.K. and Codd, G.A.
1989 Preliminary characterization of neurotoxic cyanobacteria blooms and strains from
Finland. Tox. Assess., 4, 339-352.
Smith, R.A. and Lewis, D. 1987 A rapid analysis of water for anatoxin-a, the unstable
toxic alkaloid from Anabaena flos-aquae, the stable non-toxic alkaloids left after
bioreduction and related a amine which may be nature's precursor to anatoxin-a. Vet.
Human Toxicol., 29, 153-154.
Stevens, D.K. and Krieger, R.I. 1988 Analysis of anatoxin-a by GC/ECD. J. Analyt.
Toxicol.. 12, 126-131.
Stevens, D.K. and Krieger, R.I. 1991 Stability studies on the cyanobacterial nicotinic
alkaloid anatoxin-a. Toxicon, 29(2), 167-179.
Swoboda, U.K., Dow, C.S., Chaivimol, J., Smith, N. and Pound, B.P. 1994 Alternatives
to the mouse bioassay for cyanobacterial toxicity assessment. In: G.A. Codd, T.M.
Jefferies, C.W. Keevil and E. Potter [Eds] Detection Methods for Cyanobacterial Toxins,
Special Publication No. 149, The Royal Society of Chemistry, Cambridge, 106-110.
Tsuji, K, Naito, S., Kondo, F., Watanabe, M.F., Suzuki, S., Nakazawa, H., Suzuki, M.,
Shimada, T. and Harada, K.-I. 1994 A clean-up method for analysis of trace amounts of
microcystins in lake water. Toxicon, 32, 1251-1259.
Tsuji, K., Setsuda, S., Watanuki, T., Konda, F., Nakazawa, H., Suzuki, M. and Harada,
K.-I. 1996 Microcystin levels during 1992-95 for lakes Sagami and Tsukui-Japan. Natural
Toxins, 4, 189-194.
Turell, M.J. and Middlebrook, J.L. 1988 Mosquito inoculation: an alternative bioassay for
toxins. Toxicon, 26 (11), 1089-1094.
Ueno, Y., Nagata, S., Tsutsumi, T., Hasegawa, A., Watanabe, M.F., Park, H.D., Chen,
G.-C., Chen, G. and Yu, S.-Z. 1996 Detection of microcystins, a blue-green algal
liver cancer in China, by highly sensitive immunoassay. Carcinogenesis, 17, 1317-1321.
Utkilen, H. and Gjølme, N. 1994 A simple and rapid method for extraction of toxic
peptides from cyanobacteria. In: G.A. Codd, T.M. Jefferies, C.W. Keevil and E. Potter
[Eds] Detection Methods for Cyanobacterial Toxins, Special Publication No. 149, The
Royal Society of Chemistry, Cambridge, 168-171.
Vezie, C., Benoufella, F., Sivonen, K., Bertru, G. and Laplanche, A. 1996 Detection of
toxicity of cyanobacterial strains using Artemia salina and Microtox assays compared
with mouse bioassay results. Phycologia, 35 (6, Supplement), 198-202.
Ward, C.J., Beattie, A.A., Lee, E.Y.C. and Codd, G.A. 1997 Colorimetric protein
phosphatase inhibition assay of laboratory strains and natural blooms of cyanobacteria:
comparisons with high-performance liquid chromatographic analysis. FEMS Microbiol.
Lett., 153, 465-473.
Wils, E.R.J. and Hulst, A.G. 1993 Determination of saxitoxin by liquid-chromatography

thermospray mass-spectrometry. Rapid Commun. Mass Spectrom., 7(6), 413-415.
APPENDIX H:
EPA DOCUMENTS
Algal Toxin Risk Assessment
and Management Strategic Plan
for Drinking Water
Strategy Submitted to Congress to Meet the

Requirements of P.L. 114-45
Product of the
United States Environmental Protection Agency
810R04003
November 2015
Table of Contents
Table of Contents ........................................................................................................................................... i
List of Sections Responsive to P.L. 114-45 .................................................................................................... ii
List of Abbreviations and Acronyms ............................................................................................................ iii
I. Executive Summary .............................................................................................................................. 1
II. Introduction .......................................................................................................................................... 3
III. Strategic Plan ........................................................................................................................................ 5
a. Algal Toxins and Their Human Health Effects ................................................................................... 5
b. Health Advisories .............................................................................................................................. 7
c. Factors Likely to Cause Harmful Algal Blooms .................................................................................. 8
d. Analytical Methods ......................................................................................................................... 12
e. Frequency of Monitoring ................................................................................................................ 13
f. Treatment Options .......................................................................................................................... 14
g. Source Water Protection Practices ................................................................................................. 16
h. Cooperative Agreements and Technical Assistance ....................................................................... 22
IV. Information Coordination ................................................................................................................... 27
a. Information Gaps ............................................................................................................................ 27
b. Information from Other Federal Agencies...................................................................................... 30
c. Stakeholder Involvement ................................................................................................................ 30
V. References .......................................................................................................................................... 32
VI. Appendix 1. Text of Public Law No: 114-45 ........................................................................................ 37
VII. Appendix 2. EPA’s Current Activities Directly Related to Freshwater HABs....................................... 40
VIII. Appendix 3. EPA’s Intended Future Activities Directly Related to Freshwater HABs ......................... 49
IX. Appendix 4. Federal Agencies’ Current and Proposed Activities Directly Related to HABs ............... 53
X. Appendix 5. Summary of Stakeholder Input ....................................................................................... 66
i
List of Sections Responsive to P.L. 114-45
Drinking Water Protection Act – Section of EPA’s Strategic Plan
Elements Necessary for the Strategic Plan
§1459(a)(1)(A)- “evaluate the risk to human health from drinking Algal Toxins and Their Human Health
water provided by public water systems contaminated with algal Effects (Section II, A)
toxins;”
§1459(a)(1)(B)- “establish, publish, and update a comprehensive list
of algal toxins which the Administrator determines may have an
adverse effect on human health when present in drinking water
provided by public water systems, taking into account likely
exposure levels;” and
§1459(a)(1)(C)(i)- “summarize - the known adverse human health
effects of algal toxins included on the list published [by EPA] when
present in drinking water provided by public water systems.”
§1459(a)(1)(D)(i)- “publish health advisories pursuant to section Health Advisories (Section II, B)
1412(b)(1)(F) for such algal toxins in drinking water provided by
public water systems.”
§1459(a)(1)(C)(ii)- “factors that cause toxin-producing Factors Likely to Cause Harmful Algal
cyanobacteria and algae to proliferate and express toxins.” Blooms (Section II, C)
§1459(a)(1)(D)(ii)- “establish guidance regarding feasible analytical Analytical Methods (Section II, D)
methods to quantify the presence of algal toxins.”
§1459 (a)(1)(D)(iii)- “establish guidance regarding the frequency of Frequency of Monitoring (Section II, E)
monitoring necessary to determine if such algal toxins are present in
drinking water provided by public water systems.”
§1459(a)(1)(E)- “recommend feasible treatment options, including Treatment Options (Section II, F)
procedures, equipment, and source water protection practices, to Source Water Protection Practices
mitigate any adverse public health effects of algal toxins included (Section II, G)
on the list published [by EPA].”
§1459(a)(1)(F)- “enter into cooperative agreements with, and Cooperative Agreements and Technical
provide technical assistance to, affected States and public water Assistance (Section II, H)
systems, as identified by the Administrator, for the purpose of
managing risks associated with algal toxins included on the list
published [by EPA].”
§1459(b)(1)- “identify gaps in the Agency’s understanding of algal Information Gaps (Section III, A)
toxins, including—(A) the human health effects of algal toxins
included on the list published [by the EPA]; and (B) methods and
means of testing and monitoring for the presence of harmful algal
toxins in source water of, or drinking water provided by, public
water systems”.
§1459(b)(2)- “consult, as appropriate, (A) other Federal agencies Stakeholder Involvement (Section III, C)
that—(i) examine or analyze cyanobacteria or algal toxins; or (ii)
address public health concerns related to harmful algal blooms; (B)
States; (C) operators of public water systems; (D) multinational
agencies; (E) foreign governments; (F) research and academic
institutions; and (G) companies that provide relevant drinking water
treatment options.”
§1459(b)(3)-“assemble and publish information from each Federal Information from Other Federal Agencies
agency that has—(A) examined or analyzed cyanobacteria or algal (Section III, B)
toxins; or (B) addressed public health concerns related to harmful
algal blooms.”
ii
List of Abbreviations and Acronyms
ARS Agricultural Research Service
ART Analytical Response Team (NOAA’s)
ASDWA Association of State Drinking Water Administrators
AWWA American Water Works Association
BMAA Beta-methylamino-L-alanine
BMP Best Management Practices
CCL Contaminant Candidate List
CDC Centers for Disease Control and Prevention
CEAP Conservation Effects Assessment Project
CHRP Coastal Hypoxia Research Program
CRMs Certified Reference Materials
CWA Clean Water Act
CyAN Cyanobacteria Assessment Network
CYN Cylindrospermopsin
DBP Disinfection byproducts
DHHS Department of Health and Human Services
DNA Deoxyribonucleic acid
DOC Department of Commerce
DOD Department of Defense
DOE Department of Education
DOI Department of Interior
DWMAPs Drinking Water Mapping System for Protecting Source Water
DWPA Drinking Water Protection Act
DWSRF Drinking Water State Revolving Fund
DWTP Drinking water treatment plant
ELISA Enzyme-linked immunosorbent assay
EPA United States Environmental Protection Agency
ESA European Space Agency
FDA Food and Drug Administration
GAC Granulated Activated Carbon
GAO Government Accountability Office
GAP General Environmental Assistance Program
GIS Geographic Information Systems
GLRI Great Lakes Restoration Initiative
HA Health Advisory
HABHRCA Harmful Algal Bloom and Hypoxia Research and Control Act
HAB Harmful algal bloom
HESD Health Effects Support Document
HHWQC Human Health Water Quality Criteria
IWG Interagency Working Group
LC/MS/MS Liquid chromatography tandem mass spectrometry
LPS Lipopolysaccharides
MMPB 2-methyl-3-methoxy-4-phenylbutyric acid
NARS National Aquatic Resource Surveys
NASA National Aeronautics and Space Administration
NCCOS National Centers for Coastal Ocean Science
iii
NCER National Center for Environmental Research
NDWAC National Drinking Water Advisory Council
NERL National Exposure Research Laboratory
NGO Non-Governmental Organization
NHC National HABs Committee
NIEHS National Institute of Environmental Health Sciences
NIFA National Institute of Food and Agriculture
NLA National Lakes Assessment
NOAA National Oceanic and Atmospheric Administration
NORS National Outbreak Reporting System
NRC National Research Council
NRCS Natural Resources Conservation Service
NRWQC Nationally Recommended Water Quality Criteria
NSF National Science Foundation
NWFSC Northwest Fisheries Science Center
NWIS National Water Information System
NWP National Water Program
OCE Division of Ocean Sciences (NSF)
Ohio EPA Ohio Environmental Protection Agency
OKDEQ Oklahoma Department of Environmental Quality
ORD Office of Research and Development (EPA)
OST Office of Science and Technology (EPA)
OW Office of Water (EPA)
PAC Powdered Activated Carbon
PMN Phytoplankton Monitoring Network
PSP Paralytic Shellfish Poisonings
PWS Public Water System
qPCR Quantitative Polymerase Chain Reaction
RARE Regional Applied Research Effort
RPS Recovery Potential Screening
SDWA Safe Drinking Water Act
SEATT South East Alaska Tribal Toxins network
SWP Source Water Protection
TMDLs Total Maximum Daily Loads
UC University of Cincinnati
UCMR Unregulated Contaminant Monitoring Rule
USACE United States Army Corp of Engineers
USDA United States Department of Agriculture
USGS United States Geological Survey
WBPs Watershed-Based Plans
WHO World Health Organization
WQT Water Quality Trading
WRF Water Research Foundation
iv
I. Executive Summary
The prevalence and duration of harmful algal blooms (HABs) in freshwater is rapidly expanding in the
United States and worldwide. The water quality, human health and socioeconomic impacts of HABs can
be significant. Some HABs can produce toxins that are toxic to liver, kidney and nervous system
functions in humans and animals. These toxins, when found in source waters, can contaminate drinking
water supplies if that water is not adequately treated. The challenges that HABs pose to public drinking
water systems include an incomplete understanding of how to prevent, predict, analyze, monitor and
treat toxins in drinking water; determining how to effectively communicate risk to stakeholders; and
developing and implementing resource-efficient methods to reduce the risks posed by HABs in source
waters.
The United States Environmental Protection Agency (EPA) developed this document in accordance with
Section 1459 of the Safe Drinking Water Act, as amended by the Drinking Water Protection Act, which
requires that the Administrator of the EPA develop a strategic plan for assessing and managing risks
associated with algal toxins in drinking water provided by public water systems. This plan presents
examples of recently completed and ongoing HAB-related activities and provides steps and timelines for
intended future EPA activities. These ongoing and future activities outline EPA’s plan for the next few
months through the next five years and beyond. This plan addresses:
Algal Toxins and Their Human Health Effects

Evaluating the risk to human health from drinking water contaminated with algal toxins provided by
public water systems; establishing, publishing and updating a comprehensive list of algal toxins that may
have an adverse effect on human health when found in drinking water provided by public water systems;
and summarizing those health effects.
Steps include: 1) Building on the existing work of compiling information on mechanisms of toxicity in
human and animals for the toxins microcystins, cylindrospermopsin and anatoxin-a; 2) evaluating
information gaps and analyzing the human health risk posed by other toxins of human health concern;
and 3) determining whether sufficient information is available to develop health advisories for
additional toxins.
Health Advisories
Determining whether to publish additional health advisories for the algal toxins represented on the
comprehensive list of algal toxins that may have an adverse effect on human health when found in
drinking water provided by public water systems.
Steps include: 1) Determining if adequate occurrence, toxicology and epidemiology data are available to
develop health advisories for the listed toxins other than those established in June 2015 for the
cyanotoxins microcystins and cylindrospermopsin; 2) evaluating the toxicity of these listed toxins
including the toxico-dynamics and toxicokinetics of microcystin congeners; and 3) analyzing the adverse
effects to the reproductive system from exposure to microcystins.
Factors Likely To Cause Harmful Algal Blooms

Summarizing the factors that cause toxin-producing cyanobacteria and algae to proliferate and express
toxins.
Steps include: 1) Building on research to better understand HAB ecology; 2) developing tools to quantify
HABs in U.S. freshwater lakes and reservoirs using satellite color data; 3) evaluating, interpreting and
linking existing data on algal toxins and the factors that impact their occurrence, including nutrient
1
loading and climate change; and 4) identifying areas where more monitoring is necessary to support
scientific understanding.
Analytical Methods
Establishing additional guidance regarding feasible analytical methods to quantify the presence of algal
toxins.
Steps include: 1) Building on efforts to evaluate the comparability of rapid screening methods and more
specific analytical methods; 2) evaluating methods to fill knowledge gaps and provide improved
analytical methods for algal toxins in drinking water; and 3) providing standardized and validated
detection and analysis methods, as needed, for emerging algal toxins of concern.
Frequency of Monitoring
Evaluating the frequency of monitoring necessary to determine if such algal toxins are present in drinking
water provided by public water systems.
Steps include: 1) Engaging with states and public water systems to update and refine the existing
guidance on monitoring frequency as more information becomes available; and 2) using emerging
science on factors affecting HABs and algal toxins to inform monitoring frequencies.
Treatment Options
Evaluating feasible treatment options, including procedures and equipment to mitigate any adverse
public health effects of algal toxins included on the published algal toxin list.
Steps include: 1) Summarizing the state of knowledge regarding water treatment optimization and
identifying approaches to assist with treatment challenges related to HAB events; 2) researching the
removal effectiveness of unit operations for various toxins and developing better predictive
tools/models; and 3) investigating how to implement treatment process and operational changes for
maximum protection and cost-effectiveness under a variety of site-specific constraints.
Source Water Protection Practices

Evaluating and recommending feasible source water protection practices to mitigate any adverse public
health effects of algal toxins included on the published list.
Steps include: 1) Expanding computerized mapping and water quality modeling for HAB detection and
prediction at the watershed scale; 2) monitoring nutrients across watersheds to both target and assess
protection activities; 3) working with states to prioritize nutrient-impacted waterbodies for water quality
improvements and developing targets for clean-up; and 4) collaboratively working across the EPA’s
regional offices to promote awareness amongst the public drinking water systems on the monitoring,
screening techniques and source water protection practices.
Additionally, this plan outlines a strategy for continuing to utilize cooperative agreements and provide
technical assistance to states and public water systems to address HABs.
2
II. Introduction
On August 7, 2015, Public Law 114-45, titled the Drinking Water Protection Act, amended the Safe
Drinking Water Act (SDWA) by adding Section 1459, Algal Toxin Risk Assessment and Management (see
Appendix 1 for the text of P.L. 114-45). Section 1459 directs the Administrator of the United States
Environmental Protection Agency (EPA) to submit to Congress, no later than 90 days after the date of
enactment, a strategic plan for assessing and managing risks associated with algal toxins in drinking
water provided by public water systems (PWSs). The plan must include steps and timelines to:
 Evaluate the risk to human health from drinking water contaminated with algal toxins provided
by PWSs;
 Establish, publish and update a comprehensive list of algal toxins that may have an adverse
effect on human health when found in drinking water provided by PWSs, accounting for the
levels of likely exposure;
 Summarize known adverse human health effects of the listed algal toxins when present in
drinking water provided by PWSs and summarize factors that cause toxin-producing
cyanobacteria and algae to proliferate and cause cells to express toxins (i.e., produce and
release toxins);
 For the listed algal toxins, determine whether to publish health advisories, establish guidance on
feasible analytical methods to quantify the presence of algal toxins, recommend the frequency
of monitoring necessary to determine if algal toxins are present and recommend feasible
treatment options including source water protection practices; enter into cooperative
agreements with, and provide technical assistance to, affected states and PWSs, as identified by
the Administrator, for the purpose of managing risks associated with algal toxins included on the
algal toxin list developed by the EPA; and update the strategic plan as appropriate.
Section 1459 also directs the EPA to identify information gaps in the understanding of algal toxins,
including the human health effects and the methods and monitoring for algal toxins in source water or
in drinking water provided by PWSs. The new amendment directs the EPA, as appropriate, to consult
with other federal agencies (that evaluate cyanobacteria or algal toxins or that address public health
concerns related to cyanobacteria and algal toxins), states, PWS operators, multinational agencies,
foreign governments, research and academic institutions and companies providing treatment options. In
addition, Section 1459 also directs the EPA to assemble and publish information from each federal
agency that has evaluated cyanobacteria or algal toxins or addressed public health concerns related to
HABs.
This document presents to Congress a strategic plan for the assessment and management of the risk
associated with algal toxins in drinking water provided by PWSs. This strategic plan outlines steps and
timelines for currently planned EPA activities and the activities that could occur in the future, contingent
upon available resources and funding, to address specific items in Section 1459. Nothing in this
document, in and of itself, obligates EPA to expend appropriations or incur other financial obligations
that would be inconsistent with the Agency’s statutory authority, its budget priorities, or the availability
of appropriated funds. This document also does not create any right or benefit, substantive or
procedural, enforceable by law or equity against EPA, its officers or employees, or any other person.
The strategic plan also includes ongoing activities of the Interagency Working Group (IWG) that was
established as part of the Harmful Algal Bloom and Hypoxia Research and Control Act (HABHRCA)
Amendments of 2014 (HABHRCA 2014; P.L. 113-124). The IWG on HABHRCA is chaired by the EPA and
National Oceanic and Atmospheric Administration (NOAA), and includes representatives of the U.S.
3
Centers for Disease Control and Prevention (CDC), Food and Drug Administration (FDA), U.S.
Department of Agriculture (USDA), U.S. Army Corps of Engineers (USACE), United States Geological
Survey (USGS), National Aeronautics and Space Administration (NASA), National Institutes of Health
(NIH) and the National Science Foundation (NSF). The IWG develops action plans, reports and
assessments in coordination with federal agencies to advance the scientific understanding of and ability
to predict, detect, mitigate, control and respond to HABs and hypoxia events. This strategic plan
specifically focuses on toxins associated with cyanobacteria (cyanotoxins).1
1
There are toxins associated with some other algae as well. While cyanotoxins are technically not produced by
algae, this document describes cyanotoxins as algal toxins to be consistent with the common, synonymous usage
of these terms. Similarly, the document at times uses the terms cyanobacterial blooms and harmful algal blooms
synonymously.
4
III. Strategic Plan
a. Algal Toxins and Their Human Health Effects
This section is responsive to §1459(a)(1)(A), §1459(a)(1)(B) and §1459(a)(1)(C)(i) of the SDWA directing
the EPA to develop a strategic plan to “evaluate the risk to human health from drinking water provided
by public water systems contaminated with algal toxins;” “establish, publish, and update a
comprehensive list of algal toxins which the Administrator determines may have an adverse effect on
human health when present in drinking water provided by public water systems, taking into account
likely exposure levels;” and “summarize - the known adverse human health effects of algal toxins
included on the list published [by the EPA] when present in drinking water provided by public water
systems.”
Cyanobacteria can produce a wide range of bioactive compounds, some of which may have beneficial or
therapeutic effects (Jensen et al., 2001). Other cyanobacteria can produce bioactive compounds that
may be harmful, called cyanotoxins. The most commonly recognized bioactive compounds produced by
cyanobacteria fall into four broad groupings: cyclic peptides, alkaloids, amino acids and
lipopolysaccharides (LPSs).
Cyanotoxins present a unique challenge. The same cyanotoxins can be produced by more than one
species of cyanobacteria and some cyanobacteria can produce more than one toxin at a time, resulting
in blooms with multiple cyanotoxins (Funari and Testai, 2008). The toxicity of a particular bloom is
complex, determined by the mixture of cyanobacteria species present and the variation in strains with
toxic and nontoxic genotypes involved (WHO, 1999). Toxin production can vary between blooms and
within an individual bloom over time (Duy et al., 2000).
Drinking water is a source of potential exposure to cyanotoxins. The occurrence of cyanotoxins in

drinking water depends on their levels in the raw source water and the effectiveness of treatment
methods for removing cyanobacteria and cyanotoxins during the production of drinking water. The
SDWA, as amended in 1996, requires the EPA to publish a list of contaminants every five years that are
known or anticipated to occur in PWSs and which may require regulation under the SDWA. This list is
called the Contaminant Candidate List (CCL). Cyanobacteria and their toxins were included in the CCL 1
and the CCL 2. The CCL 3 and the draft CCL 4 also identify cyanotoxins as a priority, highlighting three
particular toxins of interest: microcystin-LR, cylindrospermopsin and anatoxin-a (U.S. EPA, 2015a). As
part of the CCL processes, health and occurrence information were evaluated in establishing the lists.
Under the SDWA, the EPA uses the Regulatory Determination process to evaluate available data to
determine whether contaminants require regulation or if additional information is needed. The EPA has
not addressed cyanobacteria or cyanotoxins in any of the previous Regulatory Determination cycles due
to the limited occurrence and health effects information. The contaminants listed on the CCL generally
represent priorities for the Unregulated Contaminant Monitoring Rule (UCMR) program. Under UCMR,
occurrence data are collected to allow the Agency to evaluate contaminants that currently do not have
drinking water standards and to support subsequent Regulatory Determinations (U.S. EPA, 2012b).
Cyanotoxins have not been included on previous UCMRs due to a need for improvements in cyanotoxin
analytical methods.
Cyanobacteria are the primary harmful algal group in freshwater environments and have been
documented throughout the country. Many species of cyanobacteria are able to produce toxins.
Microcystins, cylindrospermopsin and nodularins are known to impact the liver (hepatotoxins);
anatoxin-a, anatoxin-a(s) and homoanatoxin-a are known to impact the nervous system (neurotoxins).
5
These toxins pose potential risk to human health via exposure to contaminated water. Other common
toxins produced by cyanobacterial species are LPS endotoxins, saxitoxin and beta–methylamino–L–
alanine (BMAA). Saxitoxins, a large toxin family also known as paralytic shellfish poisoning (PSP) toxins,
are common in marine waters but have also been reported in freshwater systems in the United States.
The data on freshwater saxitoxins occurrence are limited, and toxicity data from exposure in drinking
water is not available. BMAA, a recently discovered neurotoxin, also has limited data on toxicity and
environmental fate and transport.
In 2012, the EPA developed an online resource, the EPA Cyanobacterial HABs Website
(http://www.epa.gov/cyanohabs) to provide information to stakeholders on cyanotoxins. The website
also includes available health effects information on these toxins.
Completed Activities
The EPA has compiled information on mechanisms of toxicity including acute, short-term, subchronic,
chronic and cancer in humans and animals, as well as toxicokinetic information for microcystins,
cylindrospermopsin and anatoxin-a. To view the Health Effects Support Documents (HESDs) for these
cyanotoxins in drinking water, visit the EPA’s Health Advisory Web page:
http://water.epa.gov/drink/standards/hascience.cfm (U.S. EPA, 2015b, c, d). The toxicity of cyanotoxins
can vary, even within a specific type of toxin (for example, the microcystins). Symptoms reported after
acute recreational exposure to cyanobacterial blooms (including microcystin-producing genera) include
skin irritations, allergic reactions or gastrointestinal illnesses. Effects reported in humans following acute
or short-term exposure to cyanotoxins in drinking water include gastroenteritis, and liver and kidney
damage. Animal studies have shown that long-term adverse effects from cyanotoxins include liver and
kidney effects. A few available epidemiological studies suggest an association between liver and
colorectal cancers and some cyanotoxins. However, the epidemiology studies are limited by their study
design, including poor measures of exposure, potential co-exposure to microbial and/or chemical
contaminants and, in most cases, failure to control for known liver and colorectal risk factors. More
information is needed to determine the carcinogenicity of these toxins.
Ongoing Activities
The EPA anticipates gathering additional information going forward to determine whether additional
HABs or toxins should be included on the list required by §1459(a)(1)(B). The EPA also expects to
determine what human health effects information is available for these HABs and toxins, particularly
with regard to drinking water exposures. This information, coupled with available information on
occurrence in freshwater, would be used to refine the list of HABs and toxins to consider for future
development of Health Advisories (HAs). This list of HABs and toxins would be included on the EPA
Cyanobacterial HABs website (http://www.epa.gov/cyanohabs).
The EPA will determine if adequate occurrence, toxicology and epidemiology data are available to
develop HAs for additional listed cyanotoxins. The EPA also will continue assessing toxicity data on
microcystins, cylindrospermopsin and anatoxin-a as appropriate. Research activities to assess the
human health effects of cyanotoxins in drinking water include monitoring of cyanotoxins in U.S. waters
and conducting toxicological and epidemiology studies to further understand the effects of cyanotoxins.
Additional ongoing activities can be found in Appendix 2.
Intended Future Activities

The EPA plans to assess new information as it becomes available on current and emerging cyanotoxins
to determine if future HAs are needed. Additional information on EPA’s proposed activities can be found
6
in Appendix 3. Furthermore, as it evaluates human health effects of HAB exposures, EPA intends to
continue collaborating with public health partners.
Timelines for Ongoing and Future Activities

EPA has recently completed activities to evaluate the risk to human health from drinking water
contaminated with algal toxins (§1459(a)(1)(A)) for three algal toxins: anatoxin-a, microcystins and
cylindrospermopsin. Going forward, EPA will continue to evaluate additional toxicity data that may
become available for these three algal toxins, as appropriate. During FY 2016, EPA plans to evaluate the
available information on human health risk associated with other cyanotoxins to determine whether
sufficient information is available to develop HAs for additional cyanotoxins. These efforts will also
continue in the years ahead.
b. Health Advisories
This section is responsive to §1459(a)(1)(D)(i) of the SDWA directing the EPA to develop a strategic plan
to “publish health advisories pursuant to section 1412(b)(1)(F) for such algal toxins in drinking water
provided by public water systems.”
SDWA provides the authority for the EPA to publish HAs for contaminants not subject to any national
primary drinking water regulation. HAs describe non-regulatory concentrations of drinking water
contaminants at which adverse health effects are not anticipated to occur over specific exposure
durations (e.g., one day, ten days, several years and a lifetime). They serve as informal technical
guidance to assist federal, state and local officials, as well as managers of public or community water
systems in protecting public health when emergency spills or contamination situations occur. They are
not legally enforceable federal standards.
There are currently no U.S. federal guidelines, water quality criteria, standards or regulations for
cyanobacteria or cyanotoxins in drinking water under SDWA, or in surface waters under the Clean Water
Act. However, the EPA identifies cyanobacteria and cyanotoxins as a priority, highlighting three
particular toxins of interest—microcystin-LR, cylindrospermopsin, and anatoxin-a—on previous CCLs and
the current draft CCL, which identify contaminants that may need regulation under SDWA. The EPA
found there are adequate health effects data to develop HAs for microcystins and cylindrospermopsin
but found the data inadequate to develop an HA for the cyanobacterial toxin anatoxin-a.
On June 17, 2015, the EPA published two HAs in drinking water for the cyanotoxins, microcystins and
cylindrospermopsin (U.S. EPA, 2015e, f). The HAs for microcystins and cylindrospermopsin provide
states, drinking water utilities and the public with information on health effects of microcystins and
cylindrospermopsin, analytical methods to test for cyanotoxins in water samples, and treatment
technologies to remove cyanobacterial toxins in drinking water. These documents are available at:
http://water.epa.gov/drink/standards/hascience.cfm.
Ongoing Activities
The EPA intends to determine whether adequate occurrence, toxicology and epidemiology data are
available to develop HAs for the cyanotoxins to be included in the list developed under §1459(a)(1)(B).
The EPA also plans to continue assessing toxicity data on microcystins, cylindrospermopsin and
anatoxin-a to determine whether the existing health advisories should be updated. Additional ongoing
activities can be found in Appendix 2.
7
EPA is evaluating whether additional studies to evaluate the toxicity, including the toxico-dynamics and
toxicokinetics, of microcystin congeners are feasible within existing resource constraints. In addition, the
NIH National Toxicology Program intends to conduct toxicity studies to address the adverse effects to
the reproductive system from exposure to microcystins. The EPA also plans to continue evaluating the
health effects of cyanobacteria, including determining the toxicity and allergenic roles of purified
cyanobacteria lipopolysaccharide. Additional information on EPA’s proposed activities can be found in
Appendix 3.
Timeline for Ongoing and Future Activities

The EPA published HAs for microcystins and cylindrospermopsin in June of 2015. If the EPA finds that
additional information is sufficient to develop HAs for cyanotoxins other than microcystins or
cylindrospermopsin, completion of HAs for additional cyanotoxins is expected to take approximately one
to two years per assessment.
c. Factors Likely to Cause Harmful Algal Blooms
This section of the strategic plan is responsive to §1459(a)(1)(C)(ii) of the SDWA, which directs the EPA
to develop a strategic plan to summarize the “factors that cause toxin-producing cyanobacteria and
algae to proliferate and express toxins”.
Cyanobacteria, also known as blue-green algae, naturally occur in marine and fresh waters. Under
certain conditions cyanobacteria can grow rapidly, producing cyanobacterial blooms (AWWA and WRF,
2015). Some cyanobacteria are capable of producing toxins, called algal toxins or cyanotoxins, which can
pose health risks to humans and animals (U.S. EPA, 2014). Blooms producing toxins are often referred to
as HABs. The conditions that cause cyanobacteria to produce cyanotoxins are not well understood. For
example, even when cyanobacteria capable of producing toxins are present, they may not actually
produce toxins under all environmental conditions (U.S. EPA, 2012b). Also cyanotoxins can occur in the
absence of a visual bloom as not all blooms are visual. It is also not possible to determine solely upon
visual observation if a bloom is producing toxins. When blooms occur, the risk of cyanotoxin
contamination of the surface water increases, thus increasing potential risk to drinking water sources
(U.S. EPA, 2014).
Excess nutrient (nitrogen and phosphorus) loadings and concentrations are a leading cause of increased
occurrence of cyanobacterial bloom formation in water bodies (Yuan and Pollard, 2015). These excess
nutrients can originate from agricultural, industrial and urban sources as well as from atmospheric
deposition (Paerl and Otten, 2013; Conley et al., 2009; Glibert et al., 2014). Factors influencing the
occurrence of cyanobacterial blooms can include:
 excess nutrient (nitrogen and phosphorus) loadings and concentrations,
 slow-moving surface water,
 high water temperature,
 high intensity and duration of sunlight,
 water column stratification,
 changes in water pH, and
 occurrence of trace metals.
Many of these factors play a greater role during shifts in wind and/or precipitation patterns (Izydorczyk
et al., 2005; Ohio EPA, 2010). Rapid swings between drought and flooding can increase levels of
nutrients in adjacent and downstream water bodies that have accumulated on the land during the
8
drought. Increased temperatures and changes in frequency and intensity of rainfall associated with
climate change can also favor bloom formation (Paerl and Huisman, 2009). In addition, Doblin et al.
(2007) demonstrated that cyanobacteria can be transported in ballast water from ships at a port where
active blooms occur to other locations when ballast water is discharged.
The EPA has taken several steps to better understand cyanobacterial HAB ecology. In particular, the EPA
is conducting research to evaluate and summarize contributors to cyanobacterial HAB development and
toxin production. This research program also includes the use of molecular methods to characterize risk
in reservoirs due to algal blooms and toxin production.
The EPA provides nationally consistent and scientifically robust assessments of aquatic resources
through the National Aquatic Resource Surveys (NARS), with a variety of indicators including cyanotoxins
and cyanobacteria abundance (U.S. EPA, 2010). Other physical, chemical and biological indicators such
as chlorophyll-a, pathogens, nutrients and sediments are also surveyed. Information from these surveys
is available at: http://water.epa.gov/type/watersheds/monitoring/aquaticsurvey_index.cfm. The NARS
dataset provides information useful for vulnerability assessments for risks of cyanotoxin exposure for
drinking water sources. The EPA evaluated data from the 2007 National Lakes Assessment (NLA) on
cyanotoxin co-occurrence with other environmental variables. A model was developed associating
concentrations of microcystins with concentrations of chlorophyll-a and total nitrogen. This model can
be used for predicting the occurrence of high concentrations of microcystins, and to identify watershed
management thresholds for total nitrogen and chlorophyll-a to reduce the risks of increased cyanotoxin
concentrations in source water (Yuan et al., 2014). The 2012 NLA data were used to describe a statistical
approach for deriving numeric targets for concentrations of total phosphorus and total nitrogen in lakes
and reservoirs that reduce the probability of excess growth of cyanobacteria in source water (Yuan and
Pollard, 2015). This analysis classified different lakes into groups in which the relationships between
cyanobacterial biovolume and nutrient concentrations were similar, improving the strength of
association between nutrient concentrations and cyanobacterial biovolume over the entire dataset.
Then relationships between total nitrogen, total phosphorus and cyanobacterial abundance were
estimated within different lake classes using hierarchical Bayesian statistical models.
In June 2015, the EPA released a step-by-step assessment guidance to help drinking water systems
conduct a system-specific evaluation to determine if and when their source water is vulnerable to
cyanotoxin occurrence (part of a recommendations document released to assist PWSs in managing risks
from cyanotoxins (U.S. EPA, 2015g)). The EPA also released HESDs for the cyanobacterial toxins
microcystins, anatoxin-a, and cylindrospermopsin
(http://water.epa.gov/drink/standards/hascience.cfm) that contain information on the factors likely to
cause cyanobacterial blooms and toxin production, in addition to health effects information.
Ongoing Activities
The EPA is working collaboratively with NASA, NOAA and the USGS on the Cyanobacteria Assessment
Network (CyAN) to detect and quantify cyanobacterial blooms in U.S. freshwater lakes and reservoirs
using satellite color data. These efforts will allow for more frequent observations over broader areas
than can be achieved by taking traditional water samples. Researchers are developing a mobile
application (app) to inform water quality managers of changes in water quality using satellite data on
cyanobacteria algal blooms (Schaeffer et al., Accepted). This network can assist freshwater systems in
incorporating satellite ocean color technologies into U.S. fresh and brackish water quality management
decisions. The overarching project goal is to support the environmental management and public use of
9
U.S. lakes, reservoirs and estuaries by providing the capability to detect and quantify cyanobacterial
blooms using satellite data (Lunetta et al., 2015; U.S. EPA, 2015h). This tool can help states, PWSs and
others obtain efficient and timely information about source water conditions.
The EPA is working on monitoring projects to improve identification and removal of cyanotoxins in
drinking water and is also evaluating the impact of increasing water temperatures and nutrient loads on
bloom development and toxin production. The EPA is currently conducting research on HABs ecology
and the development of watershed and source water management techniques, including the
development of models for nutrient loadings, increasing efficiency of watershed placement of
phosphorus and sediment best management practices (BMPs) to reduce nutrient loadings, and the use
of water quality trading (WQT) to cost-effectively reduce nutrient loadings delivered to a watershed. The
EPA is also assessing the impact of land use and infrastructure on watershed changes and evaluating
ecological contributors to HAB development and toxin production. EPA-led monitoring projects are also
underway to improve identification and removal of cyanobacterial toxins in drinking water and to
identify and characterize the development of blooms in Lake Erie.
The EPA continues to analyze NARS datasets to determine if national recommendations can be made on
the concentrations of total nitrogen and total phosphorus in source waters that would most likely not
lead to formation of HABs. For additional ongoing activities see Appendix 2: EPA’s Current Activities
Directly Related to Freshwater HABs.

The ongoing efforts detailed above are expected to continue to completion. In addition, EPA released its
Strategic Research Action Plans for the 2016-2019 timeframe in October 2015. The EPA’s Office of
Research and Development’s Safe and Sustainable Water Resources Research Program has included a
project focused on HABs, with multiple tasks that are described in EPA’s Intended Future Activities
Directly Related to Freshwater HABs (Appendix 3), including further development of satellite remote
sensing capabilities for freshwater HABs that can be utilized in monitoring and management programs.
The EPA plans to work with its federal, state and local partners to make full use of existing cyanobacteria
and cyanotoxin information from a variety of sources. State and regional investigators have conducted a
number of surveys of cyanotoxins that could be explored for inclusion into a central database. This
information could come from both field monitoring stations and supporting laboratory experiments.
Information on factors affecting bloom occurrence should also be compiled.
The EPA plans to work with state and federal partners to incorporate cyanotoxin monitoring into routine
source water and ambient monitoring programs to better understand the conditions that trigger bloom
occurrence and toxin production. The EPA also intends to develop HAB indicators, sampling designs and
protocols for use in national scale assessments. The EPA intends to also work with state and federal
partners to develop a domestic action plan for meeting the updated nutrient load targets for Lake Erie
established under Annex 4 of the Great Lakes Water Quality Agreement. The Great Lakes Water Quality
Agreement provides a case study of a framework the EPA could use to develop and implement nutrient
load targets on a large scale.
The EPA intends to evaluate existing data from case studies and modeling efforts to identify the factors
relating bloom occurrence and toxin production. EPA also intends to develop improved approaches to
understanding the interactive effects of increasing water temperatures and nutrient loads on HAB
development and toxin production as well as improved models to predict risks of HABs under climate
change scenarios. A summary of findings is anticipated to be shared broadly and incorporated into
10
predictive tools. Also of interest is an improved understanding of the temporal dynamics of the
relationship between increased nutrients and HABs. Intensive sampling, over time, of a sub-set of the
sites included in the NARS will provide the data that could be analyzed in combination with existing
national datasets. This would allow the EPA to more accurately characterize the contributions of
temporal changes to observed relationships between nutrients and ecological effects in order to make
scientifically sound numeric nutrient criteria recommendations that are protective of the nation’s
drinking water sources.
Additionally, the EPA intends to evaluate the links between changing temperatures and changing risk of
blooms on a national scale. Potential studies include relating air temperature to photic-zone
temperature with the intention of modelling this at a broader scale, evaluating how cyanobacteria or
indicators respond to changes in photic-zone temperature, including how this can be predicted over
large spatial extents, and studying how forecasted changes in air temperature impact the likelihood and
extent of bloom events. Additional information on EPA’s proposed activities can be found in Appendix 3.
The contaminants listed on the CCL (including cyanotoxins) generally represent priorities for the UCMR
program. Under UCMR, occurrence data are collected to allow the EPA to evaluate contaminants that
currently do not have drinking water standards and to support subsequent regulatory determinations
(U.S. EPA, 2012b). EPA is currently evaluating whether to include certain cyanotoxins in UCMR 4, which
is scheduled for proposal by early 2016.
The EPA’s goal is an improved understanding of the factors that are responsible for cyanobacterial
growth and bloom formation as well as an improved ability to predict when cyanobacteria are likely to
produce toxins. Additionally, the EPA hopes to develop an improved understanding of the relationship
between nutrient loading and cyanotoxin concentrations across a range of temporal and spatial scales.
This information is expected to be summarized and incorporated into tools that would help predict and
prevent algal toxin occurrence in drinking water sources.
An important approach to reducing potentially toxic cyanobacterial blooms is to develop and implement
cost-effective and scientifically sound nutrient reduction strategies to achieve healthy water quality in
drinking water sources. The EPA’s research will inform tools that can predict downstream water quality
impacts, including cyanotoxin concentration, associated with various nutrient management decisions in
watersheds. This information will be useful for drinking water managers and others in comparing cost-
effective source water control practices to treatment activities at the utility. These tools are anticipated
to help predict source water responses to nutrients in the context of other drivers (e.g., climate change,
coastal acidification and hydrologic changes).

Efforts to better understand the factors that cause toxin-producing cyanobacteria and other algae to
proliferate and express toxins are ongoing, and planned research activities are expected to require one
to five years to complete, dependent upon the project and contingent upon available resources. For
example, the capability of satellite detection of algal blooms is estimated to require up to three years to
complete. As another example, the evaluation of NARS data is an ongoing process, with the most recent
NARS lake assessment data with an anticipated publication date of spring 2016.
The EPA plans to publish the final UCMR 4 by late 2016 or early 2017. If cyanotoxins are monitored as
part of that effort, cyanotoxin national occurrence information in raw and finished drinking water will be
collected from 2018 to 2020.
11
d. Analytical Methods
This section of the strategic plan is responsive to §1459(a)(1)(D)(ii) of the SDWA directing the EPA to
develop a strategic plan to “establish guidance regarding feasible analytical methods to quantify the
presence of algal toxins.”
Accurate and scientifically validated methods to detect algal toxins are critical to assessing and
managing risks associated with algal toxins in drinking water. The EPA is actively collaborating with
states, utilities, and commercial laboratories to develop and validate analytical methods for algal toxins
in drinking water.
With the CCL and UCMR in mind, EPA scientists developed and recently published two liquid
chromatography tandem mass spectrometry (LC/MS/MS) methods for cyanotoxin analysis in drinking
water: EPA Method 544 for determination of select microcystins and nodularin-R (U.S. EPA, 2015i) and
EPA Method 545 for determination of anatoxin-a and cylindrospermopsin (U.S. EPA, 2015j). In
developing the HAs for total microcystins and cylindrospermopsin, the EPA reviewed a variety of
additional analytical methods available for measuring these toxins in drinking water (U.S. EPA, 2015e, r).
Based on the Agency’s understanding as of June 2015, the EPA provided recommendations for water
utilities on the use of enzyme-linked immunosorbent assays (ELISA) as a rapid, cost-effective screening
and monitoring tool and the LC/MS/MS methods to determine the concentration of a number of specific
toxins (U.S. EPA, 2015g).
Ongoing Activities
The EPA is continuing to develop and validate improved analytical methods for algal toxins in drinking
water, and evaluating other methods to fill knowledge gaps. Current efforts include further evaluating
the comparability of results from rapid screening methods and more specific analytical methods. The
EPA is also investigating a new LC/MS/MS method for microcystins based on analysis of an oxidative
product (2-methyl-3-methoxy-4-phenylbutyric acid or “MMPB”) which may serve as a surrogate for the
total concentration of microcystins present in a sample. This method is an alternative to the ELISA
method, providing confirmatory data utilizing a more sophisticated, albeit more time-consuming,
complex and expensive method. Another concurrent effort involves adapting drinking water analytical
methods EPA Methods 544 and 545 for use in ambient water. Analysis of ambient water tends to be
more complex due to additional constituents in the water (e.g., organic matter, particulates) that can
potentially interfere with the analysis. The EPA plans to standardize and validate the ambient water
methods for use by states, utilities and commercial laboratories to measure cyanotoxins in source
waters. Additionally, the EPA is evaluating analytical tools such as real-time sensors, qPCR, and
fluorescence-based technologies of microspectrophotometry and flow cytometry to detect
cyanobacteria in source water. EPA is also planning to develop methods for analyzing toxins in
fish/animal tissues. The EPA is also increasing its laboratory capacity for analyzing cyanotoxins; for
example, EPA Region 7 enhanced its EPA lab capabilities and conducted microcystin analysis in
September 2015 for the Kickapoo Nation of Kansas for the Delaware River (source of water) and finished
water at the treatment plant. Additional ongoing activities can be found in Appendix 2.

In addition to the ongoing efforts discussed above, as the EPA continues to evaluate the human health
risk from cyanotoxins in drinking water and establishes a list of algal toxins under §1459(a)(1)(B), there
will be continued interest in standardized and validated detection and analysis methods for additional
12
algal toxins. Since it may prove impractical to develop methods to quantify each and every algal toxin in
water (as there are multiple classes of algal toxins and potentially more than one hundred variants
within one class, e.g., microcystins), it is important to develop effective and targeted methods for algal
toxins of concern. A significant challenge to analyzing algal toxins is the limited availability of certified
reference materials (CRMs) for many of the toxins that may impact U.S. waters. There is a need for
affordable methods for toxin analysis that can be implemented by a variety of user groups, quality-
assured with CRMs, and validated through inter-laboratory trials. The EPA plans to continue to
collaborate with other federal agencies and stakeholders to develop intra- and interagency methods and
approaches. The EPA intends to further investigate alternative ways to assess the impact of the many
toxins should quantitation of each and every toxin of interest prove impractical (e.g., due to lack of
available CRMs). Additional information on EPA’s proposed activities can be found in Appendix 3.

The EPA completed the development of EPA Methods 544 and 545 in 2015, as discussed above.
Evaluating, and as appropriate developing, an MMPB method is expected to take approximately two
years. Developing methods for ambient water is also anticipated to take around two years. As additional
algal toxins are identified or prioritized, the EPA plans to continue developing methods as needed, with
development of a method typically requiring two to four years.
e. Frequency of Monitoring
This section is responsive to §1459 (a)(1)(D)(iii) of the SDWA directing the EPA to develop a strategy to
“establish guidance regarding the frequency of monitoring necessary to determine if such algal toxins
are present in drinking water provided by public water systems.”
Monitoring of algal bloom indicators and toxins in raw water and drinking water can provide early
warnings of HAB events and allow water managers to take actions when the HAB events threaten their
source water. Toxin concentrations are highly variable with season and time of the day and are
impacted by many factors (e.g., bloom dynamics, characteristics of water body, weather, etc.).
Currently, no national database on the occurrence of freshwater cyanotoxins is available, and no federal
program is in place to monitor for cyanotoxins at U.S. drinking water treatment plants. Therefore, data
on the presence or absence of cyanotoxins in finished drinking water are limited. Understanding the
factors and conditions that cause bloom formation could lead to better informed and cost-effective
monitoring activities, as discussed in Section III, c above. Cyanotoxins can be held within the cell
(intracellular) or outside the cell (extracellular). Toxins are released from the cell due to multiple factors
and during the normal bloom cycle die off. The relationship between the environmental conditions that
trigger the cyanobacteria to produce toxins is poorly understood. This variability and unpredictability of
the presence of toxins can make monitoring challenging.
After consulting with states and other stakeholders, the EPA developed its recommendations on
monitoring frequency for microcystins and cylindrospermopsin in raw and finished drinking water
based, in part, on conditions in source water and at the treatment plant (U.S. EPA, 2015g). The EPA
advised that it is important for PWSs to establish their own monitoring frequency based on their site-
specific conditions, available resources, treatment capabilities and other factors (U.S. EPA, 2015g).
Additional information regarding monitoring procedures is available on the EPA Cyanobacterial HABs
website (http://www2.epa.gov/nutrient-policy-data/cyanohabs), where there are recommended
procedures for sampling, preservation, handling and transportation of samples collected to identify the
presence of algal toxins in drinking water.
13
Ongoing Activities
Many of the inter- and intra-agency monitoring programs described in Section III, c , such as the CyAN
Project, will provide a better understanding of the appropriate monitoring frequencies of drinking water
in addition to helping understand the factors likely to cause HABs. For recreational waters and drinking
source waters, continuous, real-time monitoring offers some advantages over traditional water
sampling. These efforts can help inform PWS operators and states as to when they should sample their
raw and finished waters. The EPA is working with an interagency task force led by NOAA to develop
sensitive, quantitative, field deployable assays and sensors for HAB cells, toxins and relevant toxin
metabolites; develop remote sensing capabilities for HABs; and integrate HAB and toxin sensors into
emerging U.S. and global ocean observation systems.
Other examples of ongoing EPA efforts include a pilot-scale study that EPA Region 8 is conducting with
PWSs in Wyoming to monitor and collect samples for cyanotoxin analysis. As another example, with EPA
Region 10 support, through the Indian General Environmental Assistance Program, the Sitka Tribe of
Alaska formed the Southeast Alaska Tribal Toxins network (SEATT) with seven other tribes to gather HAB
baseline information. Additional ongoing activities can be found in Appendix 2.

The EPA plans to continue the ongoing efforts detailed above, engaging with states and PWSs to update
and refine the existing guidance on monitoring frequency as more information becomes available. The
EPA anticipates that the development of HAB forecasts under the CyAN program will continue during
fiscal year 2016 and EPA intends to continue research on determining temporal and spatial variability of
blooms (see Section III, c for further discussion). Furthermore, the EPA plans to continue working with
NOAA on a systematic approach to provide warnings to states on water quality and occurrence of
cyanobacterial blooms, which will allow the states to evaluate patterns and trends in lakes and estuaries
that are at risk based on region-specific information. Additional information on EPA’s proposed activities
is described in Appendix 3.
Additionally, as the EPA continues to evaluate monitoring frequency for UCMR 4, the EPA may suggest
possible cyanotoxin monitoring schedules and approaches as part of that effort. The toxins identified as
priority in CCL 4 that could be considered in UCMR 4 are: microcystin-LR, cylindrospermopsin, and
anatoxin-a.

The EPA anticipates it will take four to six months to seek public input and analyze available information
from the 2015 HAB season (including PWS experience with the current recommendations in the 2015
HAB season). An additional three months is anticipated to update the current monitoring
recommendations as appropriate, based on this evaluation. Building of the Cyanobacteria Assessment
Network is currently underway and is expected to take an estimated three to five years to complete.
The EPA plans to publish the final UCMR 4 by late 2016 or early 2017. If cyanotoxin monitoring is
finalized as part of that rulemaking, any amended or new cyanotoxin-related monitoring schedules or
approaches to consider will be included as appropriate.
f. Treatment Options
This section is responsive to §1459(a)(1)(E) of the SDWA directing the EPA to develop a strategic plan to
“recommend feasible treatment options, including procedures, equipment, and source water protection
14
practices, to mitigate any adverse public health effects of algal toxins included on the list published [by
the EPA].”
Controlling and managing cyanobacteria in source water and treating cyanobacteria and cyanotoxins in
drinking water are critical to protecting public health. If operated properly, conventional water
treatment designed to reduce turbidity can generally remove intact algal cells and low levels of toxins
(AWWARF, 2001; Haddix et al., 2007). More recently, a study conducted in the United States from 2008
to 2010 in five conventional drinking water treatment plants found microcystins and cylindrospermopsin
at low concentrations in raw water, but found toxins were removed to levels below detection in any of
the finished drinking water samples (Szlag et al., 2015). However, PWSs may face challenges in providing
safe drinking water during a severe bloom event, which can increase the cyanobacteria and cyanotoxin
levels in source waters. There are various prevention and treatment strategies and approaches at the
source, throughout the treatment train, and in the finished water storage and distribution system for a
PWS. As with other contaminants, a multiple-barrier approach is useful.
The EPA has been working collaboratively with regional offices, states and PWSs to characterize the
effectiveness of drinking water treatment technologies in reducing algal toxins. In developing the HAs
for microcystins and cylindrospermopsin, the EPA reviewed available treatment technologies for
treating these toxins in drinking water. These available treatment technologies were published in the
HAs for microcystins and cylindrospermopsin (U.S. EPA, 2015 e, f). During the 2013 and 2014 blooms
seasons in Lake Erie, EPA researchers conducted sampling at seven drinking water treatment plants. In
addition, bench-scale studies on the impact of oxidation and powdered activated carbon (PAC) addition
early in the treatment process on toxin removal has been evaluated. Based on this and other
experiences, the EPA developed four basic treatment strategies that PWSs can implement to provide
immediate response to any cyanotoxins detected in drinking water intakes and included these strategies
within the recommendations support document released concurrently with the HAs (U.S. EPA, 2015g).
Ongoing Activities
In order to provide further assistance to utilities, the EPA is developing a document to summarize the
state of knowledge regarding water treatment optimization and identify approaches to assist with
treatment challenges related to HAB events. The EPA is also undertaking research to better understand
the removal effectiveness of unit operations for various toxins and develop better predictive
tools/models. For example, it is not known to what capacity a granulated activated carbon (GAC)
contactor unit is able to mitigate such compounds in the event of a severe bloom event. Additional
ongoing activities are described in Appendix 2.

The EPA plans to continue the ongoing efforts detailed above as well as to undertake a systematic study
to evaluate the capacity of GAC to remove cyanotoxins from source water. In addition, the EPA plans to
investigate how to implement process and operational changes for maximum protection and cost-
effectiveness under a variety of site-specific constraints. Ideally, these changes would minimize capital,
maintenance and operational expenses and be scalable across treatment facility size and resource level.
In order to address these questions, EPA intends to perform pilot-scale studies at field locations and at
in-house facilities. The EPA also plans to continue to engage with water managers and other private and
public sector stakeholders to help ensure treatment goals are met, to streamline transfer and adoption
of viable management strategies and technologies and to utilize the available treatment research
information that is currently available and directly applicable to cyanobacteria and cyanotoxin removal.
15
Additional intended future activities are listed in Appendix 3, EPA’s Intended Future Activities Directly
Related to Freshwater HABs.

The EPA anticipates the field studies will take approximately four years and the in-house pilot studies
will take about two years to complete. The optimization guidance document is estimated to take one
year to complete with additional research to be completed in four years. The evaluation of cyanotoxin
removal by GAC is estimated to take three years to complete.
g. Source Water Protection Practices
This section is responsive to §1459(a)(1)(E) of the SDWA directing the EPA to develop a strategic plan to
“recommend feasible treatment options, including procedures, equipment, and source water protection
practices, to mitigate any adverse public health effects of algal toxins included on the list published [by
the EPA].”
Source water protection (SWP) refers to watershed protection measures intended to prevent
contaminants such as cyanotoxins from entering or forming in a source of drinking water. SWP serves as
an early-stage barrier against drinking water contamination and is a proactive, often cost-effective
option to reduce contamination that would otherwise need to be addressed by drinking water
treatment technologies. PWSs can effectively reduce cyanobacteria and related contamination by
addressing factors likely to cause toxic blooms (hereafter “risk factors”). Numerous risk factors for toxic
blooms are discussed in Section III, c. Notably, high loadings of nutrients, like phosphorus and nitrogen,
under certain ambient water and climate conditions are drivers of HABs. While other factors like vertical
stratification and water temperature may impact HABs, recommended SWP practices address nutrient
loading as the most immediate, controllable risk factor.
Nitrogen and phosphorus in source waters can come from point sources of pollution like wastewater
treatment plants and/or nonpoint sources of pollution like agricultural or stormwater runoff. Air
deposition of nitrogen and legacy from in-stream sediments are also contributors. Effective SWP options
to reduce pollution from point vs. nonpoint sources differ; for example, point sources may be addressed
through facility-specific actions like Clean Water Act (CWA) permitted effluent limits, while pollution
from nonpoint sources may be reduced through broader measures like landscape-scale fertilizer
management (by agricultural producers and homeowners) and BMPs such as buffer strips and cover
crops on agricultural lands. SWP methods must also be attuned to drainage conditions, soil
characteristics and other hydrologic and geologic factors impacting nutrient discharge (Ohio EPA, 2015).
Steps toward recommending the most effective SWP practices to reduce incidents of HABs include:
 Identify source waters vulnerable to cyanotoxins, accounting for present, future and seasonal
conditions, in order to target early monitoring for cyanotoxins and SWP activities.
 Develop new and apply existing tools to inventory point and nonpoint discharges of nutrients
in each vulnerable source water to develop the most appropriate matrix of SWP options.
 Evaluate nutrient contributions
o Compare the relative contribution of potential sources of nutrients to in-stream nutrient
levels (what are the “root causes” of nutrients in the source water?).
o Estimate magnitude of HAB risk factors, considering time lags between nutrient load and
bloom response.
16
o Establish the baseline ambient data to measure the impact of SWP activities, once
implemented.
 Assess any institutional factors (e.g., financing options, policy frameworks and partnership
opportunities) that will help stakeholders implement SWP.
Implementing nutrient input control requires the cooperation of many programs and stakeholders.
Examples of EPA collaborations to advance the strategies above include:
 A partnership of the EPA, the Association of Clean Water Administrators (ACWA), the
Association of State Drinking Water Administrators (ASDWA), the Ground Water Protection
Council (GWPC) and their networks worked together to produce Opportunities to Protect
Drinking Water Sources and Advance Watershed Goals through the Clean Water Act, a toolkit
that describes ways PWSs can use the strengths of SDWA and CWA programs to protect drinking
water. In the Toolkit, partners describe how programs like point source permitting, water quality
standards, listings, Total Maximum Daily Loads (TMDLs) and Section 319 watershed project
funding can protect source water, thus alleviating public health risks and treatment costs for
downstream PWSs (ASDWA et al., 2014; GWPC, 2012). Several elements of this Toolkit are
described below.
 The EPA is working alongside state and utility associations, non-governmental organizations
(NGOs), federal agencies like the USDA and other partners in the national Source Water
Collaborative (SWC), a group of 26 organizations dedicated to protecting sources of drinking
water. The SWC provides planning resources and technical support for local, state and regional
source water partnerships with a focus on reducing nutrient pollution. For example, the SWC
offers online guides to networking across sectors, accessing funding for SWP and designing
specific projects like manure storage systems and GIS scenario analysis for conservation
practices (SWC, 2015a). SWC members including USDA and the National Association of
Conservation Districts (NACD) also created a Conservation Partners toolkit, which offers a step-
by-step guide for understanding conservation programs through Soil and Water Conservation
Districts and USDA State Conservationists (SWC, 2015b).
 The EPA is working with states to develop and implement nutrient reduction frameworks to
identify their specific sources of nutrient pollution and prioritize watersheds and actions they
will take to reduce these sources, as well as measures to track progress in meeting their Clean
Water Act goals. These goals include meeting water quality standards for nutrients and
preventing HABs. The EPA builds state capacity to reduce nutrient pollution by providing grants
for state water pollution control programs and programs for controlling nonpoint sources of
pollution. The EPA also makes capitalization grants for state loan programs for municipal
wastewater infrastructure and stormwater best management practices. The EPA also provides
technical assistance and oversees regulatory programs that states use to reduce nutrient
pollution (e.g., National Pollutant Discharge Elimination System (NPDES) permits for point
source dischargers, TMDLs that set “pollution budgets” that are the basis for permit limits for
point sources and inform financial and technical assistance to nonpoint sources).
 The EPA works with states and other partners in geographically targeted programs to reduce
nutrient pollution contributing to harmful algal blooms in the Great Lakes, Chesapeake Bay and
its tributaries, and other places. The EPA co-leads the Gulf of Mexico Hypoxia Task Force, a
voluntary partnership of five federal agencies and 12 states, that seeks to reduce one of the
largest hypoxic zones in the world. Actions by Task Force members to reduce nutrient pollution
in the Gulf also have benefits in more local waters, including reduced HABs. The EPA’s recently
released Report to Congress on the Hypoxia Task Force (U.S. EPA, 2015k) includes numerous
examples of collaborative work to control nutrients.
17
 The EPA and USDA continue to collaborate in multiple geographic programs and in a National
Water Quality Initiative to demonstrate the benefits of using systems of conservation practices
on vulnerable lands to avoid, control and trap nutrients and maximize the effectiveness of
conservation investments.
SWP and nutrient management planning: The 1996 amendments to the SDWA Section 1453 required
state drinking water agencies to complete Source Water Assessments no later than 3.5 years following
the Agency’s approval of the state’s program. Source Water Assessments can help stakeholders identify
whether a source water is vulnerable to cyanotoxins. The assessment delineates the Source Water
Protection Area of every public water supply, inventories significant potential sources of contamination
within the Protection Area, and evaluates the susceptibility of each system to contamination (U.S. EPA,
1997). All states completed Source Water Assessments by 2003. States and local stakeholders often use
the assessment as a baseline for proactive source water protection plans and activities. However, since
most of these assessments are more than 15 years old and the data used to develop them have
improved dramatically, the information may not be accurate today. In addition, many were not made
available to the public due to concerns about security. In some cases, because the assessment data were
not available to the public, it prevented the data from being used to make planning decisions at the
watershed scale. The advent of new sources of contamination (e.g., new urban development) and new,
open data sources provide strong incentives to update past assessments to reflect more current
information and HABs-specific vulnerabilities.
Additionally, the EPA has worked with states to create and update Nonpoint Source Management Plans
and Watershed-Based Plans (WBPs). The CWA requires states to develop Nonpoint Source Management
Plans which outline objectives to restore impaired waters and protect healthy waters against nonpoint
source pollution. Nonpoint Source Management Plans often form the basis for state regulatory and
voluntary initiatives (e.g., conservation programs) to curb nutrient pollution. WBPs, which target specific
waterbodies within a state, provide a roadmap to guide cost-effective, well-informed restoration and
protection efforts. WBPs serve as the planning framework for CWA §319 watershed projects (ASDWA et
al., 2014; GWPC, 2012). At the state level, watershed-specific source water assessments can be
compared to Nonpoint Source Management Plans and WBPs to inform SWP planning and activities.
Nutrient monitoring: State water quality agencies monitor and assess waters for nutrients as well as, in
some cases, cyanobacteria or microcystins, and share these data through the EPA’s Water Quality Data
Portal. Another source of monitoring data for HAB information is satellite imaging, such as that used in
the Lake Erie HABs Bulletins by NOAA (NOAA, 2015). The United States Geological Survey (USGS) also
collects data on nutrients and cyanotoxins through the National Water Information System (NWIS). In
addition, cyanobacteria and microcystins are a part of the National Lakes Assessment included in NARS
(U.S. EPA, 2013b). However, monitoring for nutrients and cyanotoxins varies in frequency and quality
across watersheds and states. Current methods for measuring nutrient loading are expensive and do not
fully capture nutrient flux within ecosystems, limiting data availability (see “Ongoing Activities” for
additional information).
EPA Region 1 developed a GIS-based approach to identify potential risks from nutrient-related
impairments, including cyanobacteria blooms in New Hampshire’s drinking water sources. The same
analysis and mapping is expected to be conducted for the other five New England states. This effort is
helping the region and states to gain a better understanding of the connection between drinking water
18
source waters, CWA 303(d) impaired waters and algal blooms. This is a fundamental step to aligning
CWA and SDWA priorities.
Source water standards: The EPA’S Office of Water/Office of Science and Technology (OST) has
developed Nationally Recommended Water Quality Criteria (NRWQC) for Total Nitrogen and Total
Phosphorus (aquatic life ecoregional criteria) and nitrates (human health criteria) to help states and
tribes to develop Water Quality Standards under Section 304(a) of the CWA (U.S. EPA, 1986) (U.S. EPA,
2015l). Additionally, the EPA continues to collaborate with states and tribes to develop and implement
region-specific Water Quality Standards that account for site-specific information, current science and
implementation flexibilities under the CWA. These standards form the first step toward controlling
nutrient discharge from point sources in drinking water. Water Quality Criteria inform Total Maximum
Daily Loads (TMDLs), which states can use to define nutrient permit limits for point sources.
To help restore waters that do not meet Water Quality Standards, the EPA has developed the Recovery
Potential Screening (RPS) tool, which outlines ecological, geographic and social factors that lead to
effective watershed protection. RPS helps watershed programs make decisions on where to invest in
protections for the highest chances of success (U.S. EPA, 2012c).
Ongoing Activities
New tools for HAB detection and tracking: The EPA is developing new HAB tracking tools and approaches
to help states and drinking water utilities identify vulnerable source waters and plan SWP activities that
are most suitable to those watersheds. The EPA is in the initial stages of developing mobile apps to help
citizen scientists report and analyze new blooms. As discussed in Section III, c, the EPA’s ORD, NASA,
NOAA and USGS are also developing an early warning indicator system using historical and current
satellite data to detect algal blooms. Given additional support for these initial efforts, these tools can
help the EPA, states and utilities track and swiftly respond to HAB events nationwide. The EPA is also
coordinating with states and water systems to share information about protecting source waters,
monitoring for cyanotoxins, and managing cyanotoxins in drinking water.
Nutrient monitoring: Additional monitoring information across watersheds is necessary to both target
and assess SWP activities by measuring the most significant sources of contamination. For nonpoint
source discharges, real time water quality monitoring sensors for nitrogen and phosphorus could be
expanded in strategic locations such as downstream of point sources (see “Intended Future Activities”
below).
A coalition of federal agencies, including the EPA, NOAA, National Institute of Standards and Technology
(NIST), and USGS, has launched the Nutrient Sensor Challenge—an open-innovation competition to
accelerate the development and deployment of affordable sensors that can measure nutrients in
aquatic environments. The Challenge aims to spur development of inexpensive sensors that can be
commercially available by 2017. Sensors can be used by federal and state agencies, researchers, utilities
and watershed managers across the United States to gain a better understanding of nutrient levels and
how nutrients move through the environment—improving watershed management decisions (ACT,
2015).
The EPA is partnering with the dairy and swine industries to develop a Nutrient Recycling Challenge to
accelerate development and use of technologies that can recover nitrogen and phosphorus from animal
manure and generate value-added products. Environmental and economic benefits can become
substantial as more efficient ways to manage and transport nutrients are developed. The call for
concepts will launch November 16, 2015.
19
Source water standards: For waters experiencing high nutrient and cyanotoxin levels, states, often with
assistance from the EPA, work to prioritize waterbodies for TMDL development and establish waste load
allocations and permitted effluent limitations under Section 402 of the Clean Water Act. By lowering
nutrient loads from upstream sources, states can reduce the burden on PWSs to remove nutrients and
cyanotoxins from raw water. The EPA’s mapping tools like the Drinking Water Mapping System for
Protecting Source Water (DWMAPS) can identify watersheds critical to drinking water and the
impairment status of those waters so that states can easily locate impaired source waters and take
protective action (e.g., TMDL development). In addition, the Source Water Collaborative is currently
creating an online shared library for states to exchange technical information and Water Quality Criteria
for contaminants like nutrients and cyanotoxins, which can help states efficiently establish nutrient
criteria. The EPA is also providing NPDES permit writer training for state permit writers to help them
translate narrative nutrient criteria into permit limits to control nutrient inputs from point sources.
The EPA is co-leading a binational workgroup to develop and implement the Nutrients Annex (“Annex
4”) of the 2012 Great Lakes Water Quality Agreement. Under Annex 4, the United States and Canada are
charged with establishing binational phosphorus targets for the nearshore and offshore waters of Lake
Erie, needed to meet several ecosystem objectives, including minimizing the extent of hypoxic zones
associated with excessive phosphorus loading and maintaining cyanobacteria biomass at levels that do
not produce concentrations of toxins that pose a threat to human or ecosystem health.
The EPA is also working closely with states and encouraging them to develop numeric nutrient criteria
for causal (nitrogen and phosphorus) and response (chlorophyll-a; water clarity) variables for multiple
water body categories (streams/rivers, lakes/reservoirs and estuaries/coastal waters). The increasing
frequency of HABs and cyanotoxins in drinking water supplies further underscores the need for the EPA
regions and states to strengthen their efforts. This could include developing these criteria or translators
of narrative nutrient criteria in a timely fashion and at levels protective of all uses, including the drinking
water use.
Regional HABs workshops and information-sharing: Where data on sources of drinking water exist,
partnerships between watershed stakeholders can allow pooling and sharing of information to ensure
that all stakeholders benefit. The EPA and the national Source Water Collaborative work to promote
information-sharing partnerships at the watershed scale through online guides like the Source Water
Collaborative’s “How to Collaborate” toolkit and site-specific pilot programs. The EPA also encourages or
sponsors regional workshops designed to bring together state environmental agencies, health
departments, drinking water utility managers, public water supply operators, State Conservationists
(USDA-Natural Resources Conservation Service) and other agriculture partners to discuss HAB issues. For
example, the EPA hosted a HAB workshop on September 30, 2015 – October 1, 2015 in Rapid City, South
Dakota. The EPA plans to support at least two additional workshops of this kind in 2016. Additional
ongoing activities are described in Appendix 2.

The EPA, along with other federal partners, plans to continue the ongoing efforts detailed above as well
as to expand computerized mapping and water quality modeling in order to estimate cyanotoxin risk at
the watershed scale. Current tools used by the EPA could benefit from data enhancements, user support
and flow-specific modeling capability to help estimate nutrient loading in watersheds. Further resources
would also allow the federal government to deploy early warning systems based on satellite imagery
and/or citizen scientist reporting to forecast blooms around the country (these technologies are
currently under development in discrete pilot sites/regions).
20
Future work could include working collaboratively with the EPA’s regional offices to promote awareness
amongst the public drinking water systems on the monitoring, screening techniques and source water
protection practices that can identify and reduce cyanotoxins that may impact public drinking water
supplies.
Nutrient monitoring is critical to SWP planning to address cyanotoxins. Future work could include EPA
and partners increasing the coverage and frequency of monitoring both up and downstream of key
sources of point and nonpoint source phosphorus and nitrogen pollution. Monitoring data could also
contribute to modeling efforts such as USGS SPARROW or evaluation of ORD’s Mississippi River Basin’s
multimedia system, which estimates the discharge, fate and transport of nutrients (USGS, 2011; U.S.
EPA, 2015m).
Contingent upon available resources, the EPA may continue to provide logistical and technical support
to the formation and maintenance of state, local and hydrologically based collaboratives of PWSs,
scientists, elected officials and citizens such as the Salmon Falls Source Water Collaborative. As noted
above, the EPA encourages place-based and issue-specific stakeholder workshops to address source
water contaminants of concern to local communities, and hopes to continue this effort. Workshops may
leverage planning tools such as Source Water Assessments and Watershed-Based Plans, as well as
frameworks like Water Safety Plans from the World Health Organization, to identify nutrient sources
and apply cost-effective discharge controls.
Future work could include the EPA conducting an analysis of the economic value of SWP. Analyzing and
articulating the economic value of SWP is necessary for PWSs to justify their investment in these
measures. While early case studies indicate that SWP is less expensive compared to plant-level
treatment methods, more comprehensive research is required (WRI, 2013; Winiecki, 2012).
Objectives for holistic watershed planning, involving a variety of stakeholders at the federal, state, and
local level, include:
 Decision-support and GIS mapping tools allow states and PWSs to assess source water
vulnerability to HABs.
 Ubiquitous source water monitoring in vulnerable watersheds provides states and PWSs with
data necessary to identify the most significant risk factors for HABs and design SWP treatment
options accordingly.
 Point sources help monitor for and reduce nutrient loading in source waters, where appropriate.
 Nonpoint sources of nutrients are mitigated through conservation and other SWP practices.
 CWA programs help protect sources of drinking water.

Activities related to source water protection are ongoing. EPA hosted one regional HAB-related source
water protection workshop in fall of 2015, and plans to host at least two more in 2016. The preliminary
deployment for the citizen science tracking mobile app is estimated to take one year to complete and
two to three years to complete the nationwide deployment. A preliminary version of DWMAPS is
currently available (DWMAPS, 2015); more advanced versions are expected to be available for user
testing by a focus group of states and utilities within three to six months. The satellite detection of algal
blooms is estimated to take approximately one to three years to complete. The nutrient sensor
development and pilots are estimated to take approximately two years and the HABs community
workshops are estimated to take one year to complete.
21
h. Cooperative Agreements and Technical Assistance
This section of the strategy is responsive to §1459(a)(1)(F) of the SDWA directing the EPA to develop a
strategic plan to “enter into cooperative agreements with, and provide technical assistance to, affected
States and public water systems, as identified by the Administrator, for the purpose of managing risks
associated with algal toxins included on the list published [by the EPA].”
This section of the strategy identifies past efforts undertaken by the EPA on cooperative agreements
and technical assistance, as well as ongoing, planned and potential future activities related to
cooperative agreements and technical assistance. This section also describes the goals of this strategy
with regard to meeting these provisions.
A key tool that the EPA utilizes to provide states the opportunity for technical assistance is the Drinking
Water State Revolving Fund (DWSRF), created under the 1996 Amendments to the SDWA. The program
provides financing to water systems for infrastructure improvements needed to achieve the health
protection objectives of the SDWA. Through annual appropriations to the EPA, states receive
capitalization grants for their state’s DWSRF program, which then revolve at the state level. States have
the flexibility to take up to 31% of their capitalization grants in the form of set-asides to provide non-
infrastructure assistance. There are broad eligibilities under the four set-asides including capacity
development, source water protection and technical assistance and training. The four set-asides include
small system technical assistance, administrative and technical assistance, state program management,
and local assistance and other state programs. Each year, states develop work plans outlining how much
in set-asides they plan to take from their capitalization grants and what activities they plan to conduct
with those funds. States could also elect to use some of their funds for source water protection and
technologies related to the control of HABs.
Other cooperative agreements and technical assistance include utilizing the tools and authorities of both
the SDWA and the CWA. For instance, the Clean Water State Revolving Fund program allows a state to
provide, in addition to critical wastewater infrastructure financing, funding options for source water
protection projects. Performance partnership agreements also occur between the EPA and states. These
are two-year agreements that document mutual strategic goals, joint priorities, objectives and
commitments. These partnership agreements can provide flexibility in determining how federal grant
money can be used at the state level to fund source water protection measures and source water
monitoring efforts to help prevent and detect HABs.
The EPA also has tools for cooperative agreements and technical assistance for states that are more
informal in nature. These include the EPA’s working relationships with state agencies and their
associations, drinking water research organizations and the EPA regional efforts in assisting states in
efforts to protect the quality of drinking water.
The EPA has had formal and informal cooperative agreements with states and various organizations in
the drinking water industry, as well as provided technical assistance states and PWSs. For instance, with
regard to formal agreements, states have used DWSRF set-asides to fund the following activities:
• Obtain test kits/laboratory equipment for systems to test for newly recognized contaminants of
concern and training to use that equipment;
• Review and approve laboratory protocols to ensure these laboratories meet new/existing drinking
water analytical method requirements;
22
• Provide technical assistance to laboratories related to data management and timely delivery of
drinking water quality results;
• Obtain laboratory equipment for conducting drinking water sample tests;
• Plan and implement surface water source assessment and protection activities, including source
water management plans, buffer establishment and upkeep, road and storm water management
and reconstruction activities, developing public outreach and educational programs and materials;
• Provide a source water protection ordinance template for city and county governments; and
• Support source water protection education and workshops.
Another example is the relationships the EPA has with the state drinking water regulatory agencies and
their associations. For instance, the EPA has a long-standing cooperative relationship with the
Association of State Drinking Water Administrators (ASDWA), a national professional association of state
drinking water programs. Examples of successful cooperation with ASDWA include the sharing of
information between the EPA and ASDWA, the participation of ASDWA on the EPA Federal Advisory
Committees and input on potential implementation concerns that may arise as a result of regulations
developed by the EPA. The EPA benefited greatly from the input of state representatives and ASDWA
during a May 11, 2015, public meeting on cyanotoxins, and on the EPA document “Recommendations
for Public Water Systems to Manage Cyanotoxins in Drinking Water.” The technical assistance provided
within the document has assisted states and utilities in better preparing for and responding to
cyanotoxins in drinking water.
In addition, the EPA has utilized informal relationships with states and provided emergency technical
assistance to states in times of crisis. For example, the EPA provided analytical support and technical
assistance to the State of Ohio during the Toledo cyanotoxin bloom of 2014.
The EPA has also been an active participant in water industry research planning activities carried out by
the Water Research Foundation (WRF) and others. For instance, the EPA has participated on WRF
research advisory committees, which lead to an enhanced state of knowledge on a range of drinking
water issues, including cyanotoxins. WRF has funded several projects on cyanotoxins, such as
“Optimizing Conventional Treatment for Removal of Cyanobacteria and Toxins” (2010).
The EPA co-chairs the Interagency Workgroup on the Harmful Algal Blooms and Hypoxia Research and
Control Act, which was tasked by Congress with developing a Report to Congress on Harmful Algal
Blooms and Hypoxia Comprehensive Research Plan and Action Strategy. In addition, the IWG continues
to coordinate activities within the federal agencies on harmful bloom activities. The EPA is also involved
in the collaborative efforts of the National HABs Committee whose mission is to facilitate coordination
and communication of activities on a national level between the U.S. HAB community including
researchers and government agencies.
Ongoing Activities
The EPA has several activities in which the Agency is participating in cooperative agreements and
providing technical assistance in areas that may enhance drinking water protection from cyanotoxin
risks. These activities include assistance related to water monitoring, sample analysis, treatment and
capacity development. For instance, a state could use its DWSRF funds to help tackle cyanotoxin
challenges. DWSRF set-asides may be used as part of a state’s strategy to build technical, financial and
managerial capacity of public water systems. For example, a state may use set-asides for demonstration
purposes to build the capacity of the system for activities such as monitoring and training for analysis of
toxins associated with HABs. One example of the use of DWSRF set-asides is from the Ohio
23
Environmental Protection Agency (Ohio EPA), which established a fund in 2015 of $1 million to award
grants to surface water treatment plants to reimburse the purchase of cyanotoxin investigative
monitoring equipment (up to $10,000). Having the capacity to analyze samples at the water supply
instead of sending samples to an outside laboratory allows flexibility in monitoring and timely response
to any potential finished water detections.
Ohio EPA also plans in 2016 to spend another $1 million from its 15% Local Assistance and Other State
Programs set-aside to provide technical assistance to PWSs using surface water to help prevent impacts
from cyanobacteria. In addition, Ohio EPA is encouraging PWSs to acquire training from the provider on
the specific test kit purchased. Ohio EPA staff will also be available to provide guidance and technical
assistance on sample collection and analysis.
There are many areas in which the EPA has provided technical assistance and engaged in cooperative
agreements that fall outside of the scope of the DWSRF. For instance, the EPA provides technical
assistance to states and PWSs on a variety of challenges to drinking water quality, including preventing
algal toxin formation and in addressing algal toxins when they occur to mitigate adverse human health
risks from PWSs. The EPA has played a key role in the development of analytic and decision support
tools for drinking water quality protection. The EPA anticipates continuing to include the development
of analytic and decision support tools in future efforts to assist in collecting and analyzing algal toxin
data.
As described earlier, the EPA has a sustained and cooperative relationship with the states and state
representative associations (e.g., ASDWA). The EPA will continue to participate in the established data
and information sharing activities with ASDWA and other state partners as appropriate. These
relationships and activities are particularly important with regard to cyanotoxin concerns, as they can
facilitate the understanding of the potential risks posed if cyanotoxin blooms occur, as well as provide a
quicker and more accurate response to cyanotoxin detections. The EPA also provides logistical and
technical support for the formation and maintenance of state, local and hydrologically-based
collaboratives of PWSs, scientists, elected officials and citizens such as the Source Water Collaborative as
discussed in Source Water Protection Practices (Section III, g).
As an example of the EPA partnering with states to provide technical assistance, the EPA hosted a
workshop in South Dakota for Region 8 state SDWA and Clean Water Act program managers and staff to
address the formation of algal toxins. This workshop facilitated collaboration between states and federal
agencies, including the EPA, by exploring topics including how to prevent HAB occurrence through
source water protection and pollution reduction measures, and how to manage HAB occurrence
through enhanced ambient water quality monitoring and drinking water treatment. The EPA Region 2
and Environment Canada formed a Lake Ontario nutrients task team under the Great Lakes Water
Quality Agreement Annex 4. This task team is preparing a white paper that will, among other things,
characterize algal conditions in Lake Ontario and recommend data and information needs. The EPA
Region 5 Great Lakes Restoration Initiative (GLRI) provides funding to federal and state agencies to
identify collaboration project opportunities to minimize HABs in the Western Basin of Lake Erie. The EPA
Region 9 is working to assist tribes in HAB response, including targeted technical assistance, analytical
support and resources for infrastructure improvements to tribes. The EPA Region 9 worked with the
Hoopa Tribe in response to detection of anatoxin and microcystin in the Trinity River (source water for
Hoopa drinking water) to coordinate analyses, and later provided source water protection grant and
drinking water Tribal set-aside funds to support ozone treatment for the Tribe’s drinking water system.
24
An area of collaboration with other federal partners includes the EPA’s ongoing work with the Natural
Resources Conservation Service (NRCS), the Agricultural Research Service, the United States Forest
Service, and USGS, among others, to help states leverage federal technical and financial resources in
applying the most cost-effective techniques to reduce the pollution of drinking water sources by HAB
precursors such as through natural treatment of cropland runoff. Additional ongoing activities are
described in Appendix 2.

The EPA plans to continue the ongoing efforts detailed above as well having an active role in filling the
information gaps and research needs. In particular, the EPA has specific capabilities for assisting in
identifying HAB causes, development of analytical methods, enhancing monitoring and modeling
programs and sharing information with the public.
The EPA anticipates that the DWSRF will continue to be a source of funds available for mitigating and
preventing cyanotoxins in drinking water. While operation and maintenance are ineligible costs for both
the project loan fund and the set-asides, a state may finance one-time monitoring associated with
newly-installed equipment to ensure that the equipment is operating properly and meets equipment
specifications as part of the equipment delivery and installation process.
The EPA may also be able to provide technical assistance in the following areas:
 Development of watershed models to better predict nutrient loadings.

 Continued collaboration with ASDWA on providing technical assistance to states.
 Technical assistance to systems experiencing HABs.
 Continued input into the development of research recommendations to the WRF and other research
organizations.
 Workshops on opportunities to engage systems on the EPA recommendations to prepare for and
respond to cyanotoxins in drinking water.
 Pilot studies to provide technical assistance to a limited number of individual systems in preparing
for and responding to cyanotoxins in drinking water.
 Revisions to a document the EPA published in June 2015 on recommendations to prepare for and
respond to cyanotoxins in drinking water.
 Partnerships between the EPA regional laboratories with the goal of developing HAB analytical
capacity and analytical technical points of contact for state or PWS laboratory assistance.
 Coordination of state level information among states and stakeholders.
Additional assistance in these areas is anticipated to greatly enhance the ability of systems and states to
prepare for and respond to cyanotoxins in drinking water, as well as strengthen the activities of the
EPA’s federal partners.
The goal of the EPA’s activities on cooperative agreements and technical assistance is to provide
mechanisms for assistance to states and utilities to prepare for and, if necessary, respond to cyanotoxins
in drinking water. Establishing these agreements and relationships facilitates the responses needed if
and when a system is at risk to cyanotoxins in their water. In addition, the research-related activities
better positions the EPA to identify the most appropriate means to provide technical assistance.
Furthermore, financial assistance mechanisms described in this section enables systems to secure
resources to respond to cyanotoxins in cases where systems may lack the necessary expertise or other
resources. Utilizing cooperative agreements and providing technical assistance helps reduce the
25
potential health, environmental and economic impacts of cyanotoxins in finished drinking water and
drinking water sources. Additional intended future activities are described in Appendix 3, EPA’s Intended
Future Activities Directly Related to Freshwater HABs.

The EPA will continue outreach efforts with states to communicate about possible DWSRF opportunities,
including communicating with partners over the next several months about these opportunities prior to
the next HAB season. The EPA plans to continue to exploring other partnership options with federal
government agencies, states, tribes, PWSs and utility member organizations such as the American Water
Works Association, the Association of Metropolitan Water Agencies and the National Rural Water
Association.
26
IV. Information Coordination
a. Information Gaps
This section of the strategy is responsive to §1459(b)(1) of the SDWA directing the EPA, as part of their
strategic plan, to “identify gaps in the Agency’s understanding of algal toxins, including—(A) the human
health effects of algal toxins included on the list published [by the EPA]; and (B) methods and means of
testing and monitoring for the presence of harmful algal toxins in source water of, or drinking water
provided by, public water systems.”
The EPA has previously worked to identify research gaps in the development of current and future
research plans such as the ORD’s Safe and Sustainable Water Resources Strategic Research Plans. The
EPA finalized its 2016‒2019 project plans in October 2015, which proposed several key research
questions the Agency intends to address in the coming years. EPA has also previously collaborated in
identifying research needs as part of the proceedings of the Interagency International Symposium on
Cyanobacterial Harmful Algal Blooms (ISOC-HAB, 2008). Research needs were identified in Harmful Algal
Research and Response: A National Environmental Science Strategy for 2005‒2015 developed by the
Ecological Society of America, supported by NOAA (HARRNESS, 2005).
The EPA has also worked to develop research needs and challenges as part of the IWG, on the HABHRCA
Amendments of 2014. Public Law 113-124, §603A(e)(6) directs the IWG to identify additional needs and
priorities relating to HABs. The IWG developed a report on a comprehensive research plan and action
strategy that includes information gaps. The IWG, co-chaired by the EPA and NOAA, developed a
Comprehensive Research Plan and Action Strategy to address marine and freshwater HABs and hypoxia.
This plan will be submitted to Congress in 2015 and includes research gaps as described below. For more
information on information gaps discussed in the IWG, please see
http://coastalscience.noaa.gov/research/habs/habhrca.
Information gaps exist regarding the impact of drinking water contaminated with algal toxins on human
health. Additional research is needed on human health effects of existing and emerging cyanotoxins for
which no health data currently exist. Further research is also needed on the human health impacts for
which limited health effects data are available, and to better understand the various exposure pathways
of cyanotoxins, including ingestion, inhalation and dermal exposures, that occur through household use
of tap water provided by PWSs. For example, in June 2015, a Health Advisory document and Health
Effects Support Document (U.S. EPA, 2015f, 2015d) were released for microcystins. Microcystin-LR was
used as a surrogate for all the other microcystin congeners. More than 100 microcystin congeners exist,
which vary based on amino acid composition. Microcystin-LR may be one of the most potent congeners
and the majority of toxicological data on the effects of microcystins are available for this congener;
however the potential health risks from exposure to mixtures of microcystin congeners is unknown.
Thus, additional research is needed to understand the human health impacts of the other congeners,
both existing and emerging, as new congeners continue to be isolated and identified.
At present, limited health effects information is available to derive guideline values for the broader
range of cyanotoxins that may be present in drinking water. Other research gaps include information
from both short- and longer-term studies and carcinogenicity bioassays in experimental animals. One of
the challenges in conducting toxicological studies on cyanotoxins is the difficulty and cost of obtaining
the individual purified toxins that are needed to conduct the toxicological studies. Human health effects
information from cyanotoxin exposures in sensitive populations is needed, for example individuals with
preexisting liver conditions, individuals on dialysis, the elderly, pregnant women, and nursing mothers.
27
There is an information gap regarding toxin transfer through the placental wall as well as through breast
milk. There is also a need to establish a rapid sample collection and response protocol for detecting HAB
toxins in humans and animals, specialized so that preparation procedures are compatible with analytical
methods for detecting HAB toxins in humans and animals.
Where and when HABs will occur remains an information gap that prevents us from fully understanding
the human exposure risks from cyanotoxins in drinking water provided by PWSs. There is a knowledge
gap regarding the occurrence and formation of blooms in surface waters, including rivers. Occurrence
information in all surface waters could be collected using planned and event response monitoring for
HABs, cyanotoxins and HAB predictors, such as nutrients. Understanding the factors leading to HAB and
cyanotoxin formation can help provide insight into occurrences of HABs and cyanotoxins, provide
information for recommendations for monitoring frequency, and better inform HAB prevention
strategies. For example, although research has shown nutrients, specifically phosphorous and nitrogen,
play key roles leading to HAB formation (WHO, 1999; Jacoby et al., 2000) additional information is
needed to fill information gaps on understanding the relationships among nutrient levels, bloom
formation, toxin release and other factors such as temperature and precipitation. This information could
be used to determine threshold values for various indicators.
Information gaps regarding analytical methods include the need for comparing the results obtained
using various cyanotoxin methods and developing cost-effective screening and monitoring methods. As
identified by stakeholders, better understanding of current methods and development of new methods
is a near-term need. At present, the standardized analytical methods that can be used in a national
monitoring program are limited to analyses of a few specific cyanotoxins or cannot speciate groups of
related cyanotoxins, such as the microcystins. Analytic standard production is limited, which, in turn,
limits capacity for monitoring and research, even when there are measurement methods available. The
methods also differ fundamentally in their detection capabilities, and additional research is warranted
to better understand the quantitative ability of immunological assays (that measure the interactions of
cyanotoxins with antibodies) versus that of LC/MS/MS techniques (that measure the mass-to-charge
abundances of ionized cyanotoxin fragments). Each technique has a unique set of advantages and
limitations. Additional methods will be needed in the future to measure new and emerging toxins for
which there are currently no methods. Methods that are more cost-effective and less lab-intensive
would allow for more widespread use in event response and screening. Developing methods for all
analyses, screening and monitoring needed to holistically confront the HAB challenge exceeds the scope
of any one agency, and may be best served by continued interagency partnerships and establishment of
a network of several reference laboratories for standardized and validated methods.
Additional research is needed to better understand congener-specific cyanotoxin removal capabilities of

currently available water treatment processes. It is also necessary to evaluate the potential
consequences that cyanobacteria and cyanotoxin treatment techniques have on the ability of a
treatment facility to comply with existing drinking water quality regulations. For example, the
application of high oxidant concentrations to high concentrations of bloom material may negatively
impact the ability of a facility to comply with the disinfection byproduct rules. Other information gaps
exist regarding cyanobacteria and cyanotoxin treatment such as absorption capacity of powdered
activated carbon, contact time (CT) tables for cyanotoxins removal and the effects of permanganate.
Application of source water treatments, such as algicides, is also an area where information gaps exist
with respect to the impacts of these treatments on treatment efficacy, source water quality,
environmental impacts and the efficiency of downstream treatment infrastructure. Prevention and
treatment activities can involve a multi-barrier approach as well as adaptive management to fully
28
address the HABs issue. Information is needed to provide support to states and PWSs on developing and
incorporating these activities at the PWS level to ensure the best course of action is tailored specifically
to the PWSs specific circumstances.
Source water protection information gaps involve better understanding of the causes of blooms as well
as better understanding of how source water protection activities can prevent or reduce them.
Understanding the impacts of current source water protection practices (both short-term and long-term
practices) can help with the development of future protection practices and best management practices
within a source water’s watershed.
The relationship among factors that promote algal bloom and subsequent toxin production are not well
understood. Those factors include both environmental conditions such as water clarity, meteorological
conditions, alteration of water flow, vertical mixing, temperature and water quality conditions such as
pH changes, nutrient loading (principally in various forms of nitrogen and phosphorus) and trace metals.
Developing approaches for open communication and engagement between specific stakeholders is also
needed for cooperation and support for SWP practices.
More information is also needed to better understand how climate change will affect the geospatial and
temporal distribution of HABs. For example, studies have shown that increases in temperature, altered
rainfall patterns, and anthropogenic nutrient loading may lead to an increase in bloom frequency,
intensity, duration and geographic distribution (O’Neil et al., 2012; Paerl and Huisman, 2009; Paerl et al.,
2011). Another information gap is understanding how the interactions of multiple future climatic
changes will impact HAB and cyanotoxins in fresh water systems. Given the potential increase in
cyanobacterial blooms due to both the direct and indirect effects of climate change, understanding the
effects at a regional scale can help water systems prepare for potential blooms that could occur due to
changes in regional climate.
A better understanding of risk communication in the context of risk management is also needed for
cyanotoxins and HABs. The HAs for microcystin and cylindrospermopsin established two advisory levels,
one for bottle-fed infants and young children of pre-school age and one for all other ages. This can
create confusion for the public, and additional tools would support water systems in communicating this
risk. The advisory levels are based on 10 days of exposure, which may also create difficulty in risk
communication. Additional support would help PWSs handle various scenarios such as short duration
exposures or low levels of exposures. The EPA has released recommendations regarding communication
language that can be found in the recommendations document (http://www2.epa.gov/nutrient-policy-
data/guidelines-and-recommendations) based on varying levels cyanotoxins found in the finished water.
The EPA will update this language and develop other tools as appropriate. Currently the EPA is also
working with the CDC and other stakeholders on updating the Drinking Water Advisory Communication
Toolkit to include cyanotoxins specific information
(http://www.cdc.gov/healthywater/emergency/toolkit/drinking-water-outbreak-toolkit.html).
Developing training tools to assist in answering the key questions specific to PWSs are warranted.
Although systems have been dealing with algal blooms for some time, additional training is needed
regarding the cyanotoxin-producing blooms, on preventing the toxins from reaching finished water as
well as training on how to handle communication situations as described above once cyanotoxins occur
in finished water. PWS training can also help systems understand the impacts of the management cost
consequences to the PWS for preparation and response measures to cyanotoxin occurrence.
29
Additional development is also needed on how HABs data and information are managed and shared.
Many systems are collecting HABs and cyanotoxin information and it would be beneficial to have
commonalities among the data being generated such as all relevant metadata would need to be
included. Using available tools such as the EPA’s Water Quality Exchange or the Water Quality Portal, a
cooperative service that is jointly sponsored by USGS and the EPA, can be used to assist the data
management of cyanotoxin information. Stakeholders identified needs to develop and use other
resource friendly information sources such as creating monitoring networks for sharing data.
b. Information from Other Federal Agencies
This section of the strategy is responsive to §1459(b)(3) of the SDWA directing the EPA, as part of its
strategic plan, to “assemble and publish information from each Federal agency that has—(A) examined
or analyzed cyanobacteria or algal toxins; or (B) addressed public health concerns related to harmful
algal blooms.”
The HABHRCA IWG coordinates and convenes with relevant federal agencies to discuss HABs and
hypoxia events in the United States, and to develop a number of reports and assessments of these
situations. For more information on HABHRCA and the Interagency Workgroup please visit
http://coastalscience.noaa.gov/research/habs/habhrca.
Since 2013, the EPA is an ex-officio member of the National HABs Committee (NHC). The NHC is an
elected body with members representing the HAB research and state and local management community
with non-voting ex-officio members from the EPA, NOAA, USGS and CDC.
In addition to the ongoing EPA efforts described in Appendix 2 and on the EPA’s website that details EPA
activities on cyanotoxins (http://www2.epa.gov/nutrient-policy-data/cyanohabs), several federal
agencies are conducting activities and projects to advance the research on toxin-producing
cyanobacteria and algal toxins in drinking water. Federal agencies, such as USDA, are collaborating to
address nonpoint sources of nutrients that can contribute to the rise of HABs. Other agencies support
research to better understand HABs, including ways to prevent, control and mitigate them. Health and
food safety agencies at the federal and state levels are studying and monitoring the health effects on
people and pets. In some cases, government agencies at all levels are engaging the public to conduct
citizen science to monitor water quality and the occurrence of HABs in local waters. These activities are
listed in the HABHRCA Comprehensive Research Plan and Action Strategy and in Appendix 4 of this
strategic plan. Appendix 4 was compiled from interagency efforts based on input and feedback from
other federal agencies. This information will be further explored with the release of the HABHRCA
Report to Congress anticipated to be released by the end of 2015.
Timeline
EPA intends to publish information on federal agency efforts on HABs in late 2015 through its
collaboration on the HABHRCA Congressional Report.
c. Stakeholder Involvement
This section of the strategy is responsive to §1459(b)(2) of the SDWA directing the EPA, as part of its
strategic plan, to “consult, as appropriate, (A) other Federal agencies that—(i) examine or analyze
cyanobacteria or algal toxins; or (ii) address public health concerns related to harmful algal blooms; (B)
States; (C) operators of public water systems; (D) multinational agencies; (E) foreign governments; (F)
30
research and academic institutions; and (G) companies that provide relevant drinking water treatment
options.”
The EPA held a public listening session on September 16, 2015, to provide an opportunity for
stakeholders to present their views on the key issues that may inform the strategic plan on assessing
and managing risks from cyanotoxins to drinking water. Over 300 people participated and 13 individuals
provided written or oral input. Registrants of that session included members and stakeholders of the
drinking water community, such as PWS operators, state and local governments, academic institutions,
federal agencies, industry representative groups, environmental groups, technology manufacturers and
developers (see Appendix 5 for Summary of Stakeholder Input). Comments submitted during the
listening session were considered in the development of this strategic plan. On September 17, 2015, the
EPA heard clarification of input provided by states and water utilities in a meeting with ASDWA and
AWWA and several of their members. Participants provided additional input regarding key information
gaps related to our understanding of managing algal toxins in drinking water. The consultation focused
on discussions of activities to include in the strategic plan.
In April 2015, the EPA sought input on the most recent draft Contaminant Candidate List (CCL) 4. The list
contained cyanotoxins including anatoxin-a, cylindrospermopsin and microcystins. In May 2015, the EPA
held a public meeting to provide an opportunity for public input on potential actions states and PWSs
could take to prepare for and respond to cyanotoxin health risks in drinking water. The EPA engaged
with stakeholders on what information the Agency could provide to best support states and PWSs in
addressing their risks to cyanotoxins.
The IWG also conducted a series of webinars in all major regions of the United States and a public
meeting in Ohio to initiate a conversation with stakeholders on topics related to HABs and hypoxia.
Input received was used by the IWG to inform the development of the comprehensive research plan and
action strategy for dealing with and responding to HABs and hypoxia that will be published in fall 2015.

As part of future efforts to evaluate risks to drinking water from cyanotoxins, the EPA will continue to
engage stakeholders, including states, ASDWA, AWWA, PWSs, the environmental community and others
as appropriate to ensure timely, useful and valid products. The EPA also intends to participate in
additional public meeting(s) after the current algal bloom season ends to obtain feedback on the EPA’s
recommendation document for PWSs.
31
V. References
Alliance for Coastal Technologies (ACT). 2015. Nutrient Sensor Challenge: A Water Sensor Market
Stimulation Challenge. Available online at: http://www.act-us.info/nutrients-challenge/.
American Water Works Association (AWWA) and Water Research Foundation (WRF). 2015. A Water
Utility Manager’s Guide to Cyanotoxins. American Water Works Association. Denver, CO.
Association of State Drinking Water Administrators (ASDWA), Association of Clean Water Administrators
(ACWA), Groundwater Protection Council (GWPC) and U.S. EPA. 2014. Opportunities to Protect Drinking
Water Sources and Advance Watershed Goals Through the Clean Water Act: A Toolkit for State,
Interstate, Tribal and Federal Water Program Managers. Available online at:
http://www.asdwa.org/document/docWindow.cfm?fuseaction=document.viewDocument&documentid
=3007&documentFormatId=3779.
AWWA Research Foundation. 2001. Assessment of Blue-Green Algal Toxins in Raw and Finished Drinking
Water. Final report #256. Prepared by Carmichael, W. W. AWWA Research Foundation and American
Water Works Association. Denver, CO.
Conley, D.J., Paerl, H.W., Howarth, R.W., Boesch, D.F., Seitzinger, S.P., Havens, K.E., Lancelot, C., Likens,
G.E. 2009. Controlling eutrophication: nitrogen and phosphorus. Science 323(5917): 1014-1015.
Doblin, M. A., Coyne, K. J., Rinta-Kanto, J. M., Wilhelm, S. W., Dobbs, F. C. 2007. Dynamics and short-
term survival of toxic cyanobacteria species in ballast water from NOBOB vessels transiting the Great
Lakes—implications for HAB invasions. Harmful Algae 6(4): 519-530.
Drinking Water MAPS (DWMAPS). 2015. Available online at: http://dwmapspublic.rti.org/.
Duy, T. N., Lam, P. K. S., Shaw, G. R., and Connell, D. W. 2000. Toxicology and risk assessment of
freshwater cyanobacterial (blue-green algal) toxins in water. Reviews of Environmental Contamination
and Toxicology, 163: 113-186.
Fawell, J. K., Mitchell, R. E., Everett, D. J., and Hill, R. E. 1999. The toxicity of cyanobacterial toxins in the
mouse. 1. Microcystin-LR. Human and Experimental Toxicology, 18(3): 162-167.
Fitzgeorge, N. L. M., Clark, S. A. and Kelvin, C. W. 1994. Routes of intoxication. In: G. A. Codd, T. M.
Jeffreies, C. W. Kelvin and E. Potter, (Eds.). Detection Methods for Cyanobacterial (Blue-Green Algal)
Toxins and First International Symposium on Detection Methods for Cyanobacterial (Blue-Green Algal)
Toxins. Royal Society of Chemistry, Cambridge, U.K. p. 69-74. (As cited in Kuiper-Goodman et al., 1999
and WHO 1999)
Funari, E. and Testai, E. 2008. Human health risk assessment related to cyanotoxins exposure. Critical
Reviews in Toxicology, 38: 97-125
Glibert, P. M., Maranger, R., Sobota, D. J., & Bouwman, L. 2014. The Haber Bosch–harmful algal bloom,
(HB–HAB) link. Environmental Research Letters 9(10): 105001.
Ground Water Protection Council (GWPC). CWA-SDWA Coordination Toolkit. 2012. Available online at:
http://www.gwpc.org/cwa-sdwa-coordination-toolkit.
Haddix, P. L., Hughley, C. J., and LeChevallier, M. W. 2007. Occurrence of microcystins in 33 US water
supplies. Journal American Water Works Association, 99(9): 118-125.
HARRNESS, 2005. Harmful Algal Research and Response: A National Environmental Science Strategy
2005–2015. Ramsdell, J.S., D.M. Anderson and P.M. Glibert (Eds.), Ecological Society of America,
32
Washington DC, 96 pp. Available online at:
http://www.esa.org/HARRNESS/harrnessReport10032005.pdf.
Heinze, R. 1999. Toxicity of the cyanobacterial toxin microcystin-LR to rats after 28 days intake with the
drinking water. Environmental Toxicology, 14(1): 57-60.
Humpage, A. R. and Falconer, I. R. (2002). Oral Toxicity of Cylindrospermopsin: No Observed Adverse
Effect Level Determination in Male Swiss Albino Mice. The Cooperative Research Centre for Water
Quality and Treatment, Salisbury, South Australia. Research Report No. 13. (93 pages).
Humpage, A. R. and Falconer, I. R. (2003). Oral toxicity of the cyanobacterial toxin cylindrospermopsin in
male Swiss albino mice: Determination of no observed adverse effect level for deriving a drinking water
guideline value. Environmental Toxicology, 18(2): 94-103.
Interagency, International Symposium on Cyanobacterial Harmful Algal Blooms (ISOC-HAB) Proceedings.
2008. Cyanobacterial harmful algal blooms: state of the science and research needs. Available online at:
https://www.cdph.ca.gov/HealthInfo/environhealth/water/Documents/BGA/ISOCHABdocument.pdf.
Ito, E., Kondo, F., Terao, K., and Harada, K-I. 1997. Neoplastic nodular formation in mouse liver induced
by repeated i.p. injections of microcystin-LR. Toxicon, 35(9): 1453-1457.
Izydorczyk, K. Tarczynska, M., Jurczak, T., Mrowczynski, J., Zalewski, M. 2005. Measurement of
Phycocyanin Fluorescence as an Online Early Warning System for Cyanobacteria in Reservoir Intake
Water. Environmental Toxicology 20(4): 425-430.
Jacoby, J. M., Collier, D. C., Welch, E. B., Hardy, F. J., and Crayton, M. 2000. Environmental factors
associated with a toxic bloom of Microcystis aeruginosa. Canadian Journal of Fisheries and Aquatic
Sciences 57: 231-240.
Jensen, G. S., Ginsberg, D. I., and Drapeau, C. 2001. Blue-green algae as an immuno-enhancer and
biomodulator. Journal of the American Medical Association, 3: 24–30.
Lunetta, R. Schaeffer, B.A., Keith, D., Jacobs, S., Stumpf, R., & Murphy, M. 2015. Evaluation of
cyanobacteria cell count detection derived from MERIS imagery across the eastern USA. Remote Sensing
of Environment. 157: 24-34.
National Oceanic and Atmospheric Association (NOAA), 2015. National Centers for Coastal Ocean
Science, “HAB Forecasting”. Available online at:
http://coastalscience.noaa.gov/research/habs/forecasting. Accessed September 22, 2015.
O’Neil, J.M., Davis, T.W., Burford, M.A., Gobler, C.J. 2012. The rise of harmful cyanobacteria blooms: The
potential roles of eutrophication and climate change. Harmful Algae. 14: 313-334.
Ohio EPA. 2010. Harmful Algal Blooms in Ohio Waters. Available online at:
http://epa.ohio.gov/portals/35/inland_lakes/HABBrochure.pdf.
Ohio EPA. 2015. Ohio Nonpoint Source Pollution Control Program. Available online at:
http://epa.ohio.gov/dsw/nps/index.aspx.
Paerl, H.W. and Huisman, J. 2009. Climate change: a catalyst for global expansion of harmful
cyanobacterial blooms. Environmental Microbiology Reports 1(1): 27-37.
Paerl, H.W., Hall, N.S, Calandrino, E.S. 2011. Controlling harmful cyanobacteria blooms in a world
experiencing anthropogenic and climatic-induced change. Science of Total Environment 4: 1739-1745.
Paerl, H.W. and Otten, T.G. 2013. Harmful Cyanobacterial Blooms: Causes, Consequences, and Controls.
Microbial Ecology 65(4): 995-1010.
33
Reisner, M., Carmeli, S., Werman, M., and Sukenik, A. (2004). The cyanobacterial toxin
cylindrospermopsin inhibits pyrimidine nucleotide synthesis and alters cholesterol distribution in mice.
Toxicological Sciences, 82(2): 620-627.
Schaeffer, B.A., Loftin, K., Stumpf, R., and Werdell, J. Accepted. EPA, NASA, NOAA, and USGS collaborate
to develop a Cyanobacteria Assessment Network (CyAN). Eos, Transactions, American Geophysical
Union.
Source Water Collaborative (SWC). 2015a. How to Collaborate Toolkit. Available online at:
http://www.sourcewatercollaborative.org/how-to-collaborate-toolkit/.
SWC. 2015b. Source Water Protection and Conservation Partners Toolkit. Available online at:
http://www.sourcewatercollaborative.org/swp-conservation-partners-toolkit/.
Stevens, D. K. and Krieger, R. I. 1991a. Stability studies on the cyanobacterial nicotinic alkaloid anatoxin-
a. Toxicon, 29: 167–179.
Sukenik, A., Reisner, M., Carmeli, S., and Werman, M. (2006). Oral Toxicity of the Cyanobacterial Toxin
Cylindrospermopsin in Mice: Long-Term Exposure to Low Doses. Environmental Toxicology, 21(6): 575-
582.
Szlag, D., Sinclair, J., Southwell, B. and Westrick, J. (2015) Cyanobacteria and Cyanotoxins Occurrence
and Removal from Five High-risk Conventional Treatment Drinking Water Plants. Toxins 7(6), 2198-2220
United States Environmental Protection Agency (U.S. EPA). 1986. Quality Criteria for Water, EPA 440/5-
86-001.
U.S. EPA. 1997. State Source Water Assessment and Protection Programs, EPA 816-R-97-009.
U.S. EPA. 2010. National Lakes Assessment: A Collaborative Survey of the Nation’s Lakes. Office of Water
and Office of Research and Development, Washington, D.C.
U.S. EPA. 2012a. Unregulated Contaminant Monitoring Program. Available Online at:
http://water.epa.gov/lawsregs/rulesregs/sdwa/ucmr/.
U.S. EPA. 2012b. Cyanobacteria and Cyanotoxins: Information for Drinking Water Systems. Fact Sheet
Office of Water. EPA-810F11001.
U.S. EPA. 2012c. Recovery Potential Screening: Tools for Comparing Impaired Waters’ Restorability.
Available online at: http://water.epa.gov/lawsregs/lawsguidance/cwa/tmdl/recovery/index.cfm.
U.S. EPA. 2013. National Lakes Assessment. Available online at:
http://water.epa.gov/type/lakes/lakessurvey_index.cfm.
U.S. EPA. 2014. Cyanobacteria and Cyanotoxins: Information for Drinking Water Systems. Fact Sheet.
Office of Water. EPA-810F11001.
U.S. EPA. 2015a. Contaminant Candidate List (CCL) and Regulatory Determination- Draft CCL 4 Chemical
Contaminants. Available online at: http://www2.epa.gov/ccl/chemical-contaminants-ccl-4.
U.S. EPA. 2015b. Health Effects Support Document for the Cyanobacterial Toxin Anatoxin-a. EPA
820R15104, Washington, DC; June, 2015. Available from:
http://water.epa.gov/drink/standards/hascience.cfm
U.S. EPA. 2015c. Health Effects Support Document for the Cyanobacterial Toxin Cylindrospermopsin.
EPA 820R15103, Washington, DC; June, 2015. Available from:
34
U.S. EPA. 2015d. Health Effects Support Document for the Cyanobacterial Toxin Microcystins. EPA
U.S. EPA. 2015e. Drinking Water Health Advisory for the Cyanobacterial Toxin Cylindrospermopsin. EPA
U.S. EPA. 2015f. Drinking Water Health Advisory for the Cyanobacterial Toxin Microcystin. EPA
U.S. EPA. 2015g. Recommendations for Public Water Systems to Manage Cyanotoxins in Drinking Water.
Office of Water, Washington, D.C.
U.S. EPA. 2015h. Frequently Asked Questions and Resources for Harmful Algal Blooms and
Cyanobacterial Toxins. July 2015, Version 1. U.S. EPA, Region 9.
U.S. EPA. 2015i. Method 544. Determination of Microcystins and Nodularin in Drinking Water by Solid
Phase Extraction and Liquid Chromatography/Tandem Mass Spectrometry (LC/MS/MS). Version 1.0,
EPA/600/R-14/474, Cincinnati, OH. Available online at:
http://www.epa.gov/nerlcwww/documents/Method544_Final.pdf.
U.S. EPA. 2015j. Method 545. Determination of Cylindrospermopsin and Anatoxin-a in Drinking Water by
Liquid Chromatography Electrospray ionization Tandem Mass Spectrometry (LC/ESI-MS/MS). EPA-815-R-
15-009, Cincinnati, OH. Available online at:
http://water.epa.gov/scitech/drinkingwater/labcert/upload/epa815r15009.pdf.
U.S. EPA. 2015k. Mississippi River/Gulf of Mexico Watershed Nutrient Task Force 2015 Report to
Congress. Available online at: http://www2.epa.gov/sites/production/files/2015-
10/documents/htf_report_to_congress_final_-_10.1.15.pdf
U.S. EPA. 2015l. Ecoregional Nutrient Criteria Documents for Lakes and Reservoirs. Available online at:
http://www2.epa.gov/nutrient-policy-data/ecoregional-nutrient-criteria-documents-lakes-reservoirs.
Accessed September 7, 2015.
U.S. EPA. 2015m. The Mississippi-Atchafalaya River Basin (MARB). Available online at:
http://water.epa.gov/type/watersheds/named/msbasin/marb.cfm.
United States Geological Survey (USGS). 2011. SPARROW Surface Water-Quality Modeling. Available
online at: http://water.usgs.gov/nawqa/sparrow/.
Winiecki, E. 2012. Economics and source water protection.
WHO (World Health Organization). 1999. Toxic Cyanobacteria in Water: A Guide to their Public Health
Consequences, Monitoring, and Management, I. Chorus and J. Bartram, (Eds.), E&FN Spon, London, UK.
World Resources Institute (WRI). 2013. Natural Infrastructure: Investing in Forested Landscapes for
Source Water Protection in the United States. Available online at:
http://www.wri.org/sites/default/files/wri13_report_4c_naturalinfrastructure_v2.pdf.
Yoshida, T., Makita, Y, Nagata, S., et al. (1997). Acute oral toxicity of microcystin-LR, a cyanobacterial
hepatotoxin in mice. Natural Toxins, 5(3): 91-95.
Yuan, L.L., Pollard, A.I., Pather, S., Oliver, J.L., and D'Anglada, L. 2014. Managing microcystin: identifying
national-scale thresholds for total nitrogen and chlorophyll a. Freshwater Biology 59(9): 1970-1981.
35
Yuan, L. L. and Pollard, A. I. 2015. Deriving nutrient targets to prevent excessive cyanobacterial densities
in U.S. lakes and reservoirs. Freshwater Biology. (60)(9): 1901–1916.
36
VI. Appendix 1. Text of Public Law No: 114-45
[114th Congress Public Law 45]

[From the U.S. Government Publishing Office]
Public Law 114-45

114th Congress
An Act
To amend the Safe Drinking Water Act to provide for the assessment and
management of the risk of algal toxins in drinking water, and for other
purposes. <<NOTE: Aug. 7, 2015 - [H.R. 212]>>
Be it enacted by the Senate and House of Representatives of the

United States of America in Congress assembled, <<NOTE: Drinking Water
Protection Act. 42 USC 201 note.>>
SECTION 1. SHORT TITLE.
This Act may be cited as the “Drinking Water Protection Act”.

SEC. 2. AMENDMENT TO THE SAFE DRINKING WATER ACT.
(a) Amendment.--Part E of the Safe Drinking Water Act (42 U.S.C.

300j et seq.) is amended by adding at the end the following new section:
“SEC. 1459. <<NOTE: 42 USC 300j-19.>> ALGAL TOXIN RISK
ASSESSMENT AND MANAGEMENT.
“(a) Strategic Plan.--

“(1) <<NOTE: Deadline. Health and health care.>>
Development.--Not later than 90 days after the date of enactment
of this section, the Administrator shall develop and submit to
Congress a strategic plan for assessing and managing risks
associated with algal toxins in drinking water provided by
public water systems. The strategic plan shall include steps and
timelines to--
“(A) evaluate the risk to human health from
drinking water provided by public water systems
contaminated with algal toxins;
“(B) establish, publish, and update a comprehensive
list of algal toxins which the Administrator determines
may have an adverse effect on human health when present
in drinking water provided by public water systems,
taking into account likely exposure levels;
“(C) summarize--
“(i) the known adverse human health effects
of algal toxins included on the list published
under subparagraph (B) when present in drinking
water provided by public water systems; and
“(ii) factors that cause toxin-producing
cyanobacteria and algae to proliferate and express
37
toxins;
“(D) with respect to algal toxins included on the
list published under subparagraph (B), determine whether
to--
“(i) publish health advisories pursuant to
section 1412(b)(1)(F) for such algal toxins in
drinking water provided by public water systems;
“(ii) establish guidance regarding feasible
analytical methods to quantify the presence of
algal toxins; and
“(iii) establish guidance regarding the
frequency of monitoring necessary to determine if
such algal toxins are present in drinking water
provided by public water systems;
“(E) recommend feasible treatment options,
including procedures, equipment, and source water
protection practices, to mitigate any adverse public
health effects of algal toxins included on the list
published under subparagraph (B); and
“(F) enter into cooperative agreements with, and
provide technical assistance to, affected States and
public water systems, as identified by the
Administrator, for the purpose of managing risks
associated with algal toxins included on the list
published under subparagraph (B).
“(2) Updates.--The Administrator shall, as appropriate,
update and submit to Congress the strategic plan developed under
paragraph (1).
“(b) <<NOTE: Health and health care.>> Information Coordination.--

In carrying out this section the Administrator shall--
“(1) identify gaps in the Agency's understanding of algal
toxins, including--
“(A) the human health effects of algal toxins
included on the list published under subsection
(a)(1)(B); and
“(B) methods and means of testing and monitoring
for the presence of harmful algal toxins in source water
of, or drinking water provided by, public water systems;
“(2) as appropriate, consult with--
“(A) other Federal agencies that--
“(i) examine or analyze cyanobacteria or
algal toxins; or
“(ii) address public health concerns related
to harmful algal blooms;
“(B) States;
“(C) operators of public water systems;
“(D) multinational agencies;
“(E) foreign governments;
“(F) research and academic institutions; and
“(G) companies that provide relevant drinking water
treatment options; and
“(3) assemble and publish information from each Federal
agency that has--
38
“(A) examined or analyzed cyanobacteria or algal
toxins; or
“(B) addressed public health concerns related to
harmful algal blooms.
“(c) Use of Science.--The Administrator shall carry out this

section in accordance with the requirements described in section
1412(b)(3)(A), as applicable.
“(d) Feasible.--For purposes of this section, the term ‘feasible'
has the meaning given such term in section 1412(b)(4)(D).''.
(b) Report to Congress.--Not later than 90 days after the date of
enactment of this Act, the Comptroller General of the United States
shall prepare and submit to Congress a report that includes--
(1) an inventory of funds--
(A) expended by the United States, for each of
fiscal years 2010 through 2014, to examine or analyze
toxin-producing cyanobacteria and algae or address
public health concerns related to harmful algal blooms;
and
(B) that includes the specific purpose for which the
funds were made available, the law under which the funds
were authorized, and the Federal agency that received or
spent the funds; and
(2) recommended steps to reduce any duplication, and improve
interagency coordination, of such expenditures.
39
VII. Appendix 2. EPA’s Current Activities Directly Related to Freshwater HABs
The efforts listed below include efforts by the EPA to manage and research harmful algal blooms in
freshwater systems. While extensive, this list is not exhaustive and additional efforts are ongoing at the
Agency. Better understanding of the science behind HABs is necessary to protect the public from
cyanotoxins in drinking water and their adverse health effects. Resources permitting, the EPA plans to
close informational gaps and provide helpful tools through research to better identify, monitor, and
manage HABs and toxins.
EPA/ORD Research Activities

o Ohio is the first state in the United States to implement a state-wide program of cyanobacteria
toxin monitoring in raw and finished drinking waters. The EPA collaborated with the Ohio EPA
and collected water samples at intermediate locations within drinking water treatment
facilities. The researchers employed enzyme linked immunosorbent (ELISA) assays for measuring
cyanobacteria toxin. The goals of the project were to: (1) provide a baseline estimate of the
efficacy of currently installed drinking water treatment infrastructure, (2) provide data to inform
cost-effective process upgrades, and (3) provide samples to support the development of a
chromatographic/mass-spectrometric method, which is robust enough to handle the matrix
variations commonly encountered in a water treatment facility. Preliminary results from the in-
plant sampling study indicated the release of intracellular cyanobacterial toxins into aqueous
solution during the addition of a powerful oxidizer (potassium permanganate). Potassium
permanganate is added early in the treatment process for zebra mussel and taste and odor
control. The release of intracellular toxins into a water treatment plant is potentially
problematic because the bulk of the existing treatment infrastructure is not designed to remove
dissolved chemical contaminants. This study also investigates the impacts of pH, suspended
particulate concentration, oxidant dose, oxidant contact time, powdered activated carbon (PAC)
type, PAC dose, temperature, and subsequent control of intracellular toxins.
o Four federal agencies (U.S. EPA, USGS, NOAA, and NASA) are participating in the Cyanobacterial
Assessment Network (CyAN) project to (1) develop a uniform and systematic approach for
identifying cyanobacteria blooms using ocean color satellites across the contiguous United
States; (2) create a strategy for evaluation and refinement of algorithms across satellite
platforms; (3) identify landscape linkage postulated causes of chlorophyll-a and cyanobacteria
blooms in freshwater systems; (4) characterize exposure and human health effects using ocean
color satellites in drinking water sources and recreational waters; (5) characterize behavioral
responses and economic value of the early warning system using ocean color satellites and
mobile dissemination platform; and (6) disseminate satellite data through an Android mobile
application and EnviroAtlas. The EPA anticipates that the use of uniform satellite data products
will improve the decision-making ability of managers. In addition, satellite data products may
augment federal, state, tribal, and municipal monitoring and research efforts. At the conclusion
of this project, there should be an increase in the applied use of remotely sensed water quality
data for water quality management. The use of this technology has tremendous potential owing
to the temporal and spatial coverage of the imagery and the current lack of data available for
many systems. Using satellite data to monitor and report blooms throughout a region or state
would provide a novel robust tool and assist in holistic management of events that may involve
significant risk to the public. Ultimately this project will reduce resource needs and potential
exposures of the public.
40
o The EPA Office of Research and Development’s National Risk Management Research Laboratory,
in partnership with Ohio EPA, USGS and local municipalities, sampled monthly during the 2013
and 2014 summers water throughout the treatment trains of 7 water treatment facilities that
use Lake Erie as a drinking water source. Sampling and testing was done for cyanotoxins,
chlorophyll-a, and other chemical and microbiological markers commonly associated with HABs.
The purpose of this project was to evaluate the effectiveness of toxin removal during water
treatment, detect cyanotoxins, and try to identify water quality indicators that predict the onset
of future HABs.
o The EPA currently conducts monitoring and modeling research in the East Fork of the Little
Miami River Watershed overlaid by five southwestern Ohio counties, including Clermont,
Brown, Highland, Clinton, and Warren. This collaborative research supports the Ohio EPA
surface water modeling division who is responsible for writing the TMDL for the system. Harsha
Lake is a 2000 acre flood control reservoir that bisects the watershed and receives significant
loads of nitrogen and phosphorus pollution from the predominant agricultural land use, failing
septic systems, and 10 small waste water treatment plants in the upper watershed. But the
loading is not static over seasonal or inter annual time periods. While the U.S. Army Corps of
Engineers (USACE) has historically funded monthly lake monitoring at six lentic sites within the
system, typically from May to August, the EPA’s research objectives needed more temporal and
spatial coverage to completely understand the controls over the nutrient budgets. The entire
lake is now being sampled by EPA-ORD every three weeks throughout the entire year and a
continuous water quality sensing buoy is deployed from March through November. The buoy
sensing platform is paired with online monitors located within the intake structure to a 12
million gallons per day (mgd) drinking water treatment plant that include a fluoroprobe
configured to characterize divisional-level dynamics of the algal community. As a result of the
sampling intensity, the U.S. Geological Survey’s Ohio Water Science Center has included Harsha
in its intensive molecular-based study to characterize HABs at beaches of inland lakes and Lake
Erie. The EPA visits the main beach site at Harsha weekly to establish a temporally dense time
series for this collaborative effort. Data resources now existing for Harsha Lake serve to help
verify remote sensing algorithms that the USACE is promoting for early HAB detection and
management in the Ohio River Basin. USACE funded an aerial imaging flyover and supported
permanent monitoring of lake inflows and outflows, as well as algal taxonomic analyses for the
project. The monitoring buoy is located near the drinking water treatment plant (DWTP) intake
but was specifically positioned to pair water quality data with satellite imaging. Synoptic
sampling methods are being used at 22 other USACE Louisville District reservoirs and at DWTP
intake locations on Lake Erie.
o The EPA’s National Center for Environmental Research currently supports research that uses
molecular tools and satellite remote sensing to quantify water quality and human health risks of
harmful algal blooms and disinfection byproducts associated with extreme weather in Lake Erie
drinking water. This research is investigating the impacts of extreme precipitation on urban
runoff and urban water quality by integrating a set of models that down-scale climate
simulations to spatial scales relevant to urban hydrology and land cover products. Products from
this work include molecular tools for quantifying cyanotoxins; remote sensing indicators for
modeling water quality and human health; and visualization products that demonstrate future
changes in drinking water quality (in both long-term forecasting predictions, and short-term
forecasts immediately following an extreme event).
41
o There are many barriers preventing the success of water quality trading (WQT) in the United
States. The EPA research focuses on two major barriers that hinder WQT: uncertainties in
modeling the watershed and thin markets (too few participants). The research will determine
whether any non-traditional participants would have an incentive to purchase nutrient
abatement credits from agricultural producers (traditional participants). Researchers have
examined a drinking water treatment plant’s incentive, and will now assess the impact of HABs
on treatment costs. In addition, recreationalists and local property owners affected by problems
of HABs may also have incentive to purchase nutrient abatement credits from upstream
agricultural producers. A considerable lake modeling effort will be undertaken to better link HAB
dynamics to watershed management scenarios and socioeconomic factors along with the WQT
research.
o The EPA currently collaborates with USGS through an interagency agreement to characterize
cylindrospermopsin and saxitoxin occurrence in U.S. lakes included in the 2007 National Lakes
Assessment. Analyses will include assessing risk to human health via multiple exposure scenarios
to recreational and drinking waters.
o The EPA performs research on the detection of unique cyanobacteria organisms using
fluorescence-based technologies including micro spectrophotometer and flow cytometry.
Different types of algae and cyanobacteria occur in surface water. Occasionally these organisms
produce toxins with are harmful to organisms that live in the water or other organisms that are
exposed to the water. This research aims to correlate the specific spectra of the organism with
its unique morphology. It is anticipated that the specific spectra and changes in the spectra may
be an early predictor for toxin production. Initial preliminary research has identified unique
cyanobacteria that have distinct spectra in the 650 nm range that are different than the algae
that fluoresce in the 690 nm range. It is hoped that a specific signature of different
cyanobacteria can be developed to identify the cyanobacteria that may be producing toxins.
o The EPA-ORD works with the OW’s National Coastal Condition Assessment (NCCA) program and
EPA Region 5. The researchers recently mapped cyanobacteria concentrations across the coastal
zone of the Great Lakes. States and the EPA collected whole water samples and analyzed the
samples for nutrients, chlorophyll-a and phytoplankton species composition, including
Microcystis. A set of about 400 sites across the coastal zone of the Great Lakes were sampled in
2010. Plans are to repeat the effort in 2015. The research included phytoplankton indicators and
mapping of cyanobacteria levels according to WHO thresholds. Phytoplankton will again be
included 2015 in the NCCA survey. Results will contribute to the development of empirical
models linking water quality and plankton levels in coastal waters to watershed disturbance
levels across the Great Lakes Basin, including 762 coastal watersheds.
o The EPA tested cylindrospermopsin (CYN) for mutagenicity in the Salmonella (Ames)
mutagenicity assay using the standard plate-incorporation method in strains TA98 and TA100
with rat-liver metabolic activation (S9). Because studies in the literature showed the CYN
induced chromosomal mutations in vitro only in the presence of S9, and because of the small
amount of sample available, the EPA evaluated the mutagenicity of CYN in Salmonella with S9
and did not do any experiments without S9. The researchers performed two experiments. The
first was exploratory, with a dose range of 1 – 20 µg of CYN per Petri plate; the second
experiment had a slightly higher range of 25 – 100 µg/plate. The results were all negative for
mutagenicity in both of the strains tested. There was not enough sample to repeat the assays at
42
even higher doses, and EPA’s source for the CYN (GreenWater Laboratories, Palatka, FL) could
not provide additional sample at the time. The EPA plans to test CYN one time more at higher
doses (perhaps 1000 µg/plate) if additional sample can be obtained. Otherwise, at this point,
the results show that CYN is not mutagenic in Salmonella, which means that it does not induce
gene mutations. However, the literature does show that in the presence of S9 or in vivo, CYN
can induce chromosomal mutations (i.e., micronuclei) and DNA damage (the comet assay). Thus,
it may have carcinogenic potential through these mechanisms of chromosomal DNA damage.
Thus, CYN clearly causes chromosomal mutation, but until it is tested at somewhat higher doses,
it is unclear if CYN can also cause gene mutation.
o The EPA continues to develop analytical methods for cyanobacterial toxins. The only liquid
chromatographic/mass spectrometric (LC/MS) cyanobacterial toxin analytical method currently
published by the EPA is intended as a finished water method to support the Unregulated
Contaminant Monitoring Rule (UCMR) as appropriate. The EPA is developing
chromatographic/mass spectrometric methods that can be applied with equal confidence
throughout the treatment process, from raw to finished water. If development proceeds with
sufficient speed, ELISA results from the treatment plant sampling study will be compared with
LC/MS results, with the ultimate goal of determining the optimum monitoring trade-off
between ELISA and chromatographic/mass spectrometric analysis.
o The EPA is exploring the impact of algal blooms, including HABs, on disinfection by-products
(DBPs) formation potential in drinking water treatment plants (DWTPs). The EPA began
collecting water quality information at a DWTP intake with the intent to examine what water
quality parameters are most applicable to predicting the water treatment impacts of HABs.
Included in this work are online toxicity monitor testing for HAB toxins and development of
treated water testing protocols for toxin detection.
o The overall health effects caused by cyanobacteria remain poorly elucidated. Our current
understanding of the individual toxicological, dermatological and allergenic effects of
cyanobacterial toxins (cyanotoxins) and their components (including metabolites and by-
products) as well as their possible synergistic interactions is lacking. Numerous species of
cyanobacteria are capable of producing a wide variety of structurally and biochemically diverse
metabolites (some of which have proven to be toxic to other organisms). Animal and cellular
studies have shown the presence of toxicity despite the lack of measurable known cyanotoxins.
The EPA will identify and characterize cyanobacteria peptide(s) responsible for allergic
sensitization in susceptible individuals and to investigate the functional interactions between
cyanobacterial toxins and their co-expressed immunogenic peptides. This effort is a
collaboration between the EPA, Northern Kentucky University and the University of Cincinnati
(UC, Department of Internal Medicine, Division of Immunology and Department of
Environmental Health Gene Environmental Interactions Training Program). Data collected from
EPA and UC will lead to a better assessment of the toxicological and allergic response potential
from cyanobacteria. The outcomes of this study will provide researchers with expertise in (1) the
identification of cyanobacteria and their toxins, (2) the isolation and culturing of cyanobacteria
from the environment, (3) the purification and characterization of lipopolysaccharides (LPS) and
(4) the performance of the in vitro beta-hexosaminidase release assay for allergens using sera
from atopic patients skin-prick positive for M. aeruginosa extract. The data provided by the
effort will be used by the researchers to determine if there is a potential allergenic component
to the health outcomes using animal models and possibly develop a generic screening method
43
to determine exposure to cyanobacteria. This collaborative study provides the opportunity to
characterize cyanotoxins, cyanobacteria-derived allergenic components and their possible roles
in the presence or absence of synergistic interactions.
o Immunoassays are widely used biochemical techniques to detect microcystins in environmental

samples. The use of immunoassays for the detection of microcystins is vulnerable to matrix
components and other interfering substances. The EPA research evaluates the effects of
interfering substances commonly found in drinking and ambient water samples using
commercially available immunoassay kits for microcystin toxins. The microplate and strip test
immunoassay formats were tested in the study. Results of this study may assist in the further
refinement of existing assays and the development of practical antibody-based methods to
detect cyanotoxins in water.
EPA Regional Activities

o In Region 1, the EPA has convened a region-wide cyanobacteria monitoring and “bloom watch”
workgroup consisting of state agencies, tribes, public water suppliers, NGOs, citizen monitoring
groups, and academics. During the 2014 pilot, over 100 water bodies were sampled and in 2015
the program was expanded, including 10 public drinking water system sources and additional
recreational water bodies. Workgroup members participate in a variety of ways – all designed to
ensure sampling and data collection are performed in a uniform and consistent manner for
analyzing regional cyanobacteria occurrence. Participants use monitoring kits complete with
portable microscopes and smartphone adaptors so samplers can identify cyanobacteria in the
field and directly send images to taxonomy experts to confirm their initial identifications. The
smartphone app also allows sampling crews to electronically submit monitoring data to a central
database. A second app is currently in development for tracking the occurrence of blooms
across the region and the mid-west. Portable fluorometers are available on loan as a rapid
assessment tool to detect changes in cyanobacteria. 2016 project enhancements will include
refining data collection and analysis efforts, formatting designs for relaying information to the
general public, enhancing the citizen science program components and recruitment of drinking
water systems. Funding is, in part, through a recently awarded the EPA-ORD Ideation grant.
o EPA Region 1 staff developed a GIS-based method to identify potential risks from nutrient
related impairments, including cyanobacteria blooms in New Hampshire’s drinking water
sources. Information was gathered on drinking water intakes that were in close proximity to
surface waters that have been listed as impaired on the state’s 303(d) list for nutrient related
parameters such as total phosphorus, total nitrogen, chlorophyll-a, dissolved oxygen, excess
algal growth, algal toxins and turbidity. Nine drinking water intakes were identified using
geospatial analysis where at least one nutrient related impairment existed in a waterbody that
was within 200 feet of the intake. A map was produced that shows the identified water systems
along with nutrient impaired water bodies. The same analysis and mapping will be conducted
for the other five New England states. This effort is helping the region and states to gain a better
understanding of the connection between drinking water source waters, CWA 303(d) impaired
waters and algal blooms and is a fundamental step to aligning Clean Water Act and Safe Drinking
Water Act priorities.
o EPA Region 1’s Regional Laboratory established in 2010 monitoring buoys in the Charles and
Mystic Watersheds to track cyanobacterial blooms and water quality conditions. The buoys
44
measure for chlorophyll and use fluorescence sensors to measure for phycocyanin. Field
samples are collected for chlorophyll-a and cyanobacterial cells to correct and evaluate data.
o Renegotiation of the Great Lakes Water Quality Agreement requires that the Parties (EPA and
Environment Canada) re-examine and establish phosphorus loading targets and associated in-
lake endpoints and metrics associated with hypoxia, hazardous algal blooms, and Cladophora for
each of the Great Lakes. In addition, phosphorus load allocations must be determined by
country, state and province, and for priority watersheds. Because of the severe symptoms being
experienced in Lake Erie, it has been designated as the first and lead lake for evaluation; it is
anticipated that Lakes Ontario and Michigan will follow. The EPA is managing phosphorus and
nutrient loading data to ensure consistent use and interpretation for the purposes of setting
loading and other associated targets in Lake Erie. Work is being conducted linking watersheds
with coastal receiving waters. The loading datasets are by source category including municipal
point sources, industrial point sources, atmospheric and nonpoint sources. Preliminary results
indicate that the phosphorus loads of the Maumee and Detroit Rivers are among the largest for
all of the Great Lakes and are high priority watersheds requiring attention to abate the various
symptoms being observed in western Lake Erie. Total phosphorus and dissolved phosphorus
both are greatest from these major sources. The phosphorus loading dataset will be used for
satisfying other requirements of the Agreement through various empirical and statistical
assessments and modeling applications. An ensemble modeling approach is being used by the
Parties consisting of federal and academic partners and is beginning a Science Advisory Peer
Review on December 2014. For the Interagency Task Force, responses from Region 5 and the
Great Lakes National Program Office are pending.
o EPA Region 5 co-leads a binational workgroup to develop and implement the Nutrients Annex
(“Annex 4”) of the 2012 Great Lakes Water Quality Agreement. Under Annex 4, the United
States and Canada are charged with establishing binational phosphorus targets for the
nearshore and offshore waters of the Great Lakes needed to meet several ecosystem objectives,
including minimizing the extent of hypoxic zones associated with excessive phosphorus loading
and maintaining cyanobacteria biomass at levels that do not produce concentrations of toxins
that pose a threat to human or ecosystem health. This effort is focused on Lake Erie in the near
term, with specific milestones in the next 3-5 years (see below). In addition, EPA Region 2 has
begun working with Annex 4 to develop strategies to address phosphorus targets for Lake
Ontario, which is the next Great Lake that will receive focused attention by Annex 4. Region 2 is
conducting a nutrient monitoring protocol that will provide baseline monitoring and modeling
data to help establish phosphorus loading targets for Lake Ontario.
o On September 3, 2014, the EPA Administrator Gina McCarthy announced that the Great Lakes
Restoration Initiative (GLRI) will provide nearly $12M to federal and state agencies for projects
identified as a result of an August 2014 meeting held by Region 5 to identify collaboration
opportunities to minimize HABs in the Western Basin of Lake Erie. These projects include:
 Farmer incentives
 Soil testing and fertilizer recommendations
 Planting of winter crops
 Upgrades to controlled drainage systems
 Funding of best management practices (BMP) at livestock facilities
 Expanding Environmental Quality Incentives Program (EQIP) funding
 Improve HAB monitoring and forecasting by NOAA
45
 Tributary monitoring for phosphorus
o Stakeholder consultation is an explicit requirement in the 2012 Great Lakes Water Quality
Agreement. The EPA will solicit input from stakeholders on the new phosphorus loading targets
for Lake Erie prior to ratification in 2016.
o EPA Region 5 has been working on the Grand Lake St. Marys Project to identify present
conditions and model the Grand Lakes St. Marys watershed in order to identify problem areas
and assist watershed managers with useful information to assist in decision making. The project
started in July 2011 and is funded by the Regional Applied Research Effort (RARE), ORD’s
Regional Science Program, which responds to high-priority, near-term research needs of the
EPA’s regional offices. The EPA is assessing lake conditions and using USDA/Agricultural
Research Services (ARS) models to identify problem areas in the watershed. USDA provided
recommendations in land use management and BMP selection. Another RARE project between
the EPA Region 5 and the EPA involves methods for assessing the water quality degradation
through water treatment plants during algal blooms, which will evaluate cyanotoxin analytical
methods, identify relationships between water quality parameters and algal toxin
production/release, and evaluate treatment effectiveness of different processes on algal toxins.
The project started in September 2014 and will run through 2015. The EPA has been collecting
monthly samples from Lake Erie drinking water treatment plants including raw water, finished
water, and effluents of all intermediate unit processes. Samples are analyzed for cyanobacterial
toxins, mycotoxins, chlorophyll-a, phosphate, ammonia, nitrate, nitrite, dissolved organic
carbon, total nitrogen, and trace metals. Bench-scale studies will evaluate the impact of oxidant
dose, powdered activated carbon dose and pH on algal toxin control.
o EPA Region 6 is working closely with states and encouraging them to develop numeric nutrient
criteria for causal (nitrogen and phosphorus) and response (chlorophyll-a; water clarity)
variables for multiple water body categories (streams/rivers, lakes/reservoirs, estuaries/coastal
waters). Increasing frequency of HABs and cyanotoxins in drinking water supplies further
underscores the importance of regions and states stepping up their efforts to develop these
criteria or translators of narrative nutrient criteria in a timely fashion and at levels protective of
all uses, including the drinking water use. This also points to the need for eventual criteria
development (once national criteria are available) and routine ambient monitoring for
cyanotoxins in waters with drinking water uses.
o EPA Region 7 coordinates with the state drinking water programs who are working with their
respective state recreational monitoring programs. In 2015, the EPA Region 7 laboratory
increased its capabilities to analyze for cyanotoxins and will be collecting algal toxin samples at
the end of September in the source water and the finished water of targeted treatment plants in
tribal lands.
o EPA Region 8 purchased several Abraxis test strip kits to distribute to drinking water operators
for raw water (intake) sampling because these strips were key to a successful HAB response in
the Boysen Reservoir (in Wyoming). Abraxis test strips can be used for drinking water systems as
a screening tool to determine if the ELISA method should be used. These types of field methods
are useful in this part of the country, as laboratory capacity for analyzing algal toxins is limited.
The Region is working with groups in each state to accelerate the development of lab capacity
for cyanotoxin analysis.
46
o EPA Region 8 hosted a harmful algal blooms workshop September 30 – October 1, 2015, in
Rapid City, South Dakota. The workshop was designed to bring together state environmental
agencies, health departments, drinking water utility managers, and public water supply
operators to discuss HABs issues in Region 8. Agenda items included topics such as: impacts
associated with HABs; what causes HABs; HABs monitoring for drinking water and recreational
impacts; new technologies for tracking HABs; and opportunities for outreach and education. The
workshop also provided a forum for sharing updates on state and regional HABs-activities and
building partnerships with other agencies.
o EPA Region 9 is working to assist tribes in HAB response including targeted technical assistance,
analytical support, and resources for infrastructure improvements to tribes. EPA staff worked
with the Hoopa Tribe in response to anatoxin and microcystin in the Trinity River (source water
for Hoopa drinking water) to coordinate analyses, and later provided a source water protection
grant and drinking water Tribal set-aside funds to support ozone treatment for the Tribe’s
drinking water system.
o EPA Region 9 and ORD-National Exposure Research Laboratory were awarded an internal,
competitive 2015 ORD Safe and Healthy Communities Regional Sustainability and Environmental
Sciences Research Program (RESES) program project for their proposal entitled, “Floating
Vegetation Islands: Using Traditional Ecological Knowledge (TEK) for Development of Leading
Indicators of Ecosystem Function for BMP Effectiveness, Water Quality Standards, Biological
Criteria, and Harmful Algal Blooms (HABs).” This pilot research project will develop leading
indicators of ecosystem function to determine the need for and effectiveness of best
management practices. Leading indicators help decision makers be proactive in developing
adaptive management plans. Leading indicators will be correlated to alterations in ecosystem
functions and water quality with changes in land-use practices and climate variability. TEK from
the Chemehuevi Indian Tribe and the Colorado Indian Tribes will be used to help determine
ecosystem potential condition for current restoration projects.
o EPA Region 9 supports its states to address HAB concerns including: updating state guidance and
thresholds for recreational exposures; providing training to agencies and waterbody managers
for recognizing and responding to HABs; developing statewide field monitoring protocols;
establishing lab networks (state and federal labs, identifying capabilities and sharing analytical
methods/ protocols); developing a database for tracking HAB occurrence, toxin data, etc.; and
coordinating with veterinary labs for tissue analysis of affected animals (e.g., dogs, cattle).
o The EPA Region 9 laboratory provides microcystin analysis by ELISA to support program
requests, including: (1) analysis for numerous state and local agencies for initial assessment of
HAB-impacted waters, (2) ongoing monitoring since 2006 in the Klamath River Watershed and
(3) analytical support for monitoring of the 2015 bloom season at Clear Lake for 2015, (the latter
two are two of the region’s priority watersheds). The Region 9 Lab has analyzed 300 - 700
samples annually since approximately 2008. EPA Region 9 also has a Risk Management Program
Grant through an interagency agreement with USGS and ORD/NERL to analyze and optimize
cyanotoxin sample preparation methods for ELISA and LC/MS analysis.
o EPA Region 10 recently funded the Southeast Alaska Tribal Toxins (SEATT) project. The SEATT is
a partnership represented by eight Alaska tribes that was funded to conduct monitoring and
develop better predictive tools for HABs. With over $225K in Indian Environmental General
Assistance Program (IGAP) funding from EPA, together with training support from NOAA and
47
financial support from the Administration for Native Americans’ Environmental Regulatory
Enhancement Program, the partner tribes will monitor HAB events that pose a human health
risk to shellfish harvesters, such as paralytic shellfish poisoning (PSP). This monitoring effort will
provide weekly data on the timing and distribution of HABs, along with measurements of
environmental conditions, indicators, and potential mechanisms that trigger HAB events. The
data collected will be used to create a more rigorous framework for mitigating the impacts of
HAB events on fisheries, rather than traditional rules of thumb which are no longer effective due
to changes in the type, magnitude, frequency and duration of HABs in the region.
o In 2012-2013, EPA funding through the Puget Sound National Estuary Program supported the
Washington Department of Health and Washington Department of Ecology in conducting a
comprehensive sampling effort for diarrhetic HABs throughout Puget Sound and along the
Washington Coast (28 sites in 2012 and 72 sites in 2013). The main HAB sampling target was
Dinophysis spp. in shellfish tissue. Ancillary measurements collected during the project included
temperature, salinity, and nutrients. The goal of the project is to work toward developing a
HABs early warning system. One of the EPA’s approaches has been to use Bayesian regression
models to estimate the effect of nutrient concentrations on chlorophyll-a concentrations above
/below a threshold given nutrient inputs. The estimated marginal densities include use of the
National Lakes Assessment (NLA) sample weights, and represent the estimated Ecoregion 8
marginal densities for log10(Total Phosphorus) and log10(Total Nitrogen). The idea is this type of
approach could serve as some of the basis for empirical modeling of the likelihoods of
cyanobacteria blooms, whether toxic or not, including in freshwater systems that serve as
drinking water sources. Additionally, the NLA (2007) data will be used for the contiguous U.S.
Some modeling aspects of this work can be applied in the Northeastern U.S. where ORD is
interacting with EPA Region 1 and two New England states (MA and RI).
o EPA regions and states are working together to protect the public from exposure to HABs in
coastal and freshwater systems. EPA Region 1 has been working with the State of Vermont on
the Lake Champlain cyanobacteria monitoring, a qualitative and quantitative monitoring
program on Lake Champlain for cyanobacteria during the 2014 summer season. This project is
funded through a grant to the State of Vermont Department of Environmental Conservation
with the purpose of identifying areas of high concentrations of cyanobacteria, particularly toxin
levels, and provide warnings to the public.
48
VIII. Appendix 3. EPA’s Intended Future Activities Directly Related to Freshwater
HABs
Information in this Appendix is based on the EPA’s proposed research project (i.e., research area),
Reducing Impacts of Harmful Algal Blooms, for its 2016-2019 research cycle. Work completed under this
project will provide stakeholders and decision makers with improved scientific information and tools to
assess, predict and manage the risk of HABs, associated toxicity events and the ensuing ecological,
economic and health impacts. The project directly addresses legislative mandates, Agency research
needs, Agency Program Office initiatives, National Water Program (NWP) needs and community and
other stakeholder needs as follows:
 Improve the science of HAB and toxin detection by developing HAB-specific analytical methods
and sampling strategies.
 Assist the NWP in developing new HAB indicators, sampling designs and protocols for use in
national-scale assessments.
 Develop improved approaches to understanding the interactive effects of increasing water
temperatures and nutrient loads on HAB development and toxin production.
 Develop improved models to project risk of HABs under warming climate scenarios.
 Improve understanding of the human health and ecosystem effects resulting from toxin
exposure.
 Provide drinking water treatment system operators with improved methods for detecting and
treating toxins in order to limit or prevent human exposures.
This project will be focused on four intertwined research areas:
Area 1: Management strategies.
Research needs exist to develop new, market-ready treatment technologies, and to optimize existing
technologies for the removal of toxins present in drinking water systems. Ideally, these methods would
minimize capital, maintenance, and operational expenses, and be scalable to such a degree that they
could be implemented in communities ranging from large and wealthy to small and economically
marginalized. Active collaborations with water managers and other private and public sector
stakeholders will help ensure these goals are met and streamline transfer and adoption of viable
management strategies and technologies. Work in this area would be predicated on the assumption that
there are no significant policy or institutional barriers to adoption.
In the area of drinking water treatment, removal effectiveness for various unit operations have been
documented for a subset of the small group of toxins for which commercial standards are available.
However, knowledge gaps exist for (1) the large set of toxins for which standards are currently
unavailable, and (2) how to implement process and operational changes for maximum protection and
cost-effectiveness under a variety of site-specific constraints.
In the area of reservoir management, existing research indicates that modifications of reservoir
hydrology may help to reduce the frequency, intensity, duration and toxicity of bloom events. However,
the efficacy of these efforts is site-specific, and gaps remain in the knowledge of the optimal method(s)
to apply for any given set of reservoir conditions. ORD scientists and engineers will develop a scientific
basis for the development and application of reservoir management strategies. In the domain of
recreational area management, the primary research needs are the development of body contact
49
exposure standards for the entire suite of known toxins as well as the development of scientifically
based guidance for optimal sampling strategies.
Area 2: Health, ecosystem and economic effects.
One of the strongest drivers for changes that may be required to prevent future HABs, and/or mitigate
those that occur, is the threat of serious adverse health effects in exposed populations. Research gaps to
evaluate sources and routes of human exposures and their potential toxicity will need to be addressed.
When HABs and toxins occur in drinking water and recreational water sources, exposed human and
animal populations will need to be evaluated for health effects. The identification of exposure
biomarkers that are simple to obtain are necessary for timely evaluation of exposure levels. The types of
toxicity (critical organ system, chronic, developmental, and reproductive) are not known for most
identified toxins and these potential endpoints will be the focus of research efforts. Mammalian effects
from exposure to widespread fish toxins are also an area that needs focused research efforts, since
these widespread compounds have not been evaluated in mammals. The identification of ichthyotoxins
and their mechanisms of action are needed since these have had a serious effect on fish stocks, both
wild and in aquaculture. The potential of freshwater algal toxins to cause adverse health effects after
transport from lakes and streams into the coastal environment, and subsequent bioaccumulation in
marine organisms, is known to have occurred and requires further research.
HABs have the potential to affect aquatic ecosystems. Gaps in the following research areas need to be
addressed: food web disturbances resulting from toxin production and hypoxic areas, toxicity thresholds
for sentinel species, and the potential for toxin bioaccumulation in fish populations, both wild and aqua-
cultured.
Questions include: 1) “What are the ecological impacts of algal toxins on aquatic life through direct
exposure and through food chain bioaccumulation? 2) How sensitive are real-time biomonitoring
systems that use larval fish, daphnia and algae in comparison to traditional toxicity test organisms used
in whole effluent toxicity testing? And 3) What are the nutrient and other environmental conditions that
are conducive to establishment of toxin producing species?”
Assessment approaches will include determination of whether algal toxins inhibit zooplankton grazing
behavior and population dynamics, as well as the impact on benthic filters; whether simultaneous and
sequential exposure to multiple toxins, particularly the combination of multiple cyanotoxins, pose
cumulative or synergistic risks to aquatic life; the potential for bioaccumulation, bioconcentration, and
biomagnification of different cyanotoxins and other cyanobacterial bioactive compounds in food webs;
development of algal reference toxicant tests using the top 4 toxins found during algal blooms;
comparison of results of reference toxicant tests using standard species to the results obtained from
real-time monitoring systems; and the culturing of toxin-producing species under laboratory conditions
using various combinations of environmental conditions in order to observe the effect on toxin
production.
An accurate assessment of economic effects is a critical piece of the puzzle as the Agency works to craft
a response that is cost-effective and protective of public, economic, and societal health. To the best of
the authors’ knowledge, such an assessment does not currently exist. The assessment would be broken
down into two parts:
1. A nationally representative random sample survey to estimate the direct costs generated by
HABs: these may include, but are not limited to extra monitoring expenses, water treatment
50
plant upgrades and chemical costs and lost revenue from beach closures and drinking water
advisories. The planning and implementation of such a survey, using traditional tools of
economic research, represents an opportunity for cross-agency collaboration.
2. A nationally representative random sample survey to estimate the degree to which public
confidence in the safety of drinking water, natural and recreational assets is affected by
scientific data, general-audience news from traditional media outlets, and information across
the quality spectrum circulated on social media outlets. The motivation for such a survey is the
fact that information circulated through these channels has the potential to quickly shape public
perceptions, and these perceptions, in turn, drive behavior at the individual and family level
with potentially significant negative economic consequences. It is envisioned that such a census
would employ data from a variety of information and social media platforms to track the spread
of information within a strictly delineated subject area.
Area 3: Temperature impacts and bloom modeling
The scientific community generally agrees that HABs have been increasing in frequency, duration and
geographical range. The factors responsible for these postulated increases are thought to include ease
of global transport of species, rapid evolutionary response of algal/bacterial species to changing
environments, increased nutrient loads in aquatic environments, perturbations in rainfall, and increases
in the overall average temperatures of aquatic bodies. These factors all enhance the ability of algal and
cyanobacterial species to move, spread and form blooms with increased temporal, locational and spatial
dimensions, including different water depths. A contributing factor in bloom formation, or duration, is
thought to be increased average water temperatures, which provide a suitable environment for algal
growth. Both laboratory and environmental studies on harmful bloom dynamics are necessary to
understand the extent of effects of increased water temperatures on bloom formation, and tendency of
such blooms to generate toxins that may have adverse environmental and health effects.
Improved modeling capabilities are needed for an assessment of the risk associated with HABs under
the dynamic of different climate scenarios. An understanding of the species, temporal and spatial
dynamics of HABs will improve the capability to anticipate the course of HABs and their potential
adverse effects. The vast majority of HABs are not comprised of one species throughout the course of
the bloom, multiple species are the usual case, either at the same time or sequentially. Detailed
knowledge of the roles different environmental factors play on species identity, toxic vs non-toxic bloom
formation, persistence of blooms, and spatial/temporal extent of blooms is needed in order to increase
the accuracy of bloom forecasts. This is also true of the types of toxins that will be formed in specific
blooms. Together, an increased ability to predict the character of blooms will enable regulatory agencies
at the national, state, tribal and local level to better predict the course of blooms and, therefore,
respond appropriately.
Area 4: Analysis and monitoring in fresh and coastal/estuarine environments.
Effective response to HABs must be based on accurate and timely assessments of the species that
comprise the bloom, the toxins, if any, that are being produced, and the ecosystem impacts resulting
from the presence of HAB biomass.
Morphological, culture-based, molecular biology, and optical sensing (flow cytometry, satellite imaging)
approaches have been used to identify and quantify the primary algal, and related bacterial, species in
blooms. All of these strategies have strengths and weaknesses. Consequently, it is important to improve
51
existing monitoring and analytical methods, and to develop and validate new cost- and time-effective
methods that can be used by Program Offices, Regions, states and other stakeholders.
For ecosystem impacts, existing methods need to be improved, and new methods need to be
developed, all with the goal of delivering the greatest possible amount of analytical power into the
hands of small, local, laboratories, operating on modest budgets.
For toxins, the accuracy and precision of existing methods needs to be improved for different aqueous
matrices (e.g., fresh, treated drinking water, brackish, marine, etc.) and animal tissues. Toxin analytical
methods need to be standardized and, where possible, simplified in order to promote their adoption
across the widest possible range of laboratories. New methods, capable of being employed from a
variety of physical platforms, such as lab benches, field kits, buoys and flow-through monitors, need to
be developed.
Finally, guidance needs to be developed that allows water managers to set up site-specific monitoring
programs that take advantage of the existing suite of analytical methods, and potential in situ
monitoring networks, in order to maximize protection while minimizing sampling and analytical effort.
In recent years, HAB-driven adverse environmental and health effects have been observed in the
estuarine and marine environments of all coastal areas. Adverse health effects have been recorded in
humans through direct exposures and consumption of toxin-containing seafood. Serious adverse health
effects have also been recorded in marine mammals, fish and birds, some of which are endangered.
These effects are largely caused by algal species with toxins that are different from those found in fresh
waters. The factors that act to favor the formation of HABs are largely unknown in marine and estuarine
(saline) environments. The development of estuarine- and marine-specific analytical methods and
indicators is essential for the protection of the environment as well as human populations.
Analytical and monitoring efforts in fresh, estuarine, and marine environments have the potential to
generate data sets across a range of temporal and spatial scales. These data sets would encompass
direct readings on riparian, lake and coastal bodies of water as well as remote sensing from satellites.
The monitoring of HABs is ongoing by the Agency, a number of other federal entities including the U.S.
Geological Survey (USGS) and the National Oceanic and Atmospheric Administration (NOAA), and by
State, Local and academic entities. Data from these monitoring efforts exist in both published and
unpublished form. The utility of these large data sets depends upon their consistency and availability.
Developing a data portal that integrates existing and future data into a programmatic data base would
result in a more cohesive HAB program. This portal would allow data sharing, promote collaborative
research and speed the development of a comprehensive view of HAB extent throughout the United
States. It is recognized that the technical challenges of developing and maintaining a data portal are
significant. However, the potential benefits are so significant that laying the groundwork for such a
portal is an aspirational goal of this project area.
52
IX. Appendix 4. Federal Agencies’ Current and Proposed Activities Directly Related
to HABs
The information in this appendix is compiled from interagency efforts based on input and feedback from
other federal agencies. It is intended to be representative rather than a comprehensive listing of
HABHRCA-related work. This information will be further explored with the release of the HABHRCA
Report to Congress, anticipated to be released by the end of 2015.
Federal Agencies’ Current and Proposed Activities Directly Related to HABs

HABs/
Office/ Program Title
Agency Hypoxia/ Program Activities
Dept. (brief description)
Both
CDC initiated waterborne and foodborne disease
outbreak surveillance systems in the 1970s. U.S. states
and territories voluntarily report to these systems via
the electronic National Outbreak Reporting System
(NORS), which receives aggregate data on human cases
and their exposures, including exposures to harmful
HAB-related
algal blooms (HABs) or HAB toxins. The One Health
DHHS CDC HABs Outbreak and Illness
Harmful Algal Bloom System (OHHABS) is being
Surveillance
developed for single case-level reporting of human and
animal illness, and relevant environmental data.
OHHABS is being programmed to inform restoration
activities in the Great Lakes but will accessible to all
states via NORS. The pilot version of the system is being
tested in preparation for a 2016 launch.
CDC has partnered with the Council of State and

Territorial Epidemiologists (CSTE) since 2013 to place
and provide technical support for epidemiology fellows
in Great Lakes states, including Indiana, Illinois,
Michigan, Minnesota, New York, Ohio, and Wisconsin.
Great Lakes State The activity is supported by the Great Lakes Restoration
DHHS CDC HABs Health Surveillance Initiative. Fellows focus on waterborne disease
Capacity detection, investigation, response and reporting. The
fellowship has expanded state waterborne disease
reporting and analytic capacity; improved state health
surveillance for harmful algal blooms; and ensured
dedicated staff time for waterborne disease
surveillance and coordination activities.
CDC’s health communications activities related to HABs

include the preparation of a HAB website with
information for public health practitioners, clinicians,
and the general public, and the expansion of the
Drinking Water Advisory Communications Toolbox
Health (DWACT) to include information about HAB-related
DHHS CDC HABs
Communications drinking water advisories. The DWACT was created
through a collaborative effort among CDC, EPA, the
American Water Works Association, the Association of
State and Territorial Health Officials, the Association of
State Drinking Water Administrators (ASDWA), and the
National Environmental Health Association (NEHA).
53
HABs/
Both
Additional research is needed to fully characterize and
understand the health risks from drinking water
provided by public water systems when that water is
contaminated with cyanobacterial toxins. There is a
need to establish standardized biological sample
collection and analysis protocols to support assessment
of toxin-associated health effects. Multiple federal
agencies are working together to assess sampling and
Interagency Analytic
Multiple CDC, EPA, NOAA HABs analytical capabilities related to analysis of biological
Workgroup
specimens collected from human and animals exposed
to cyanobacteria toxins via contaminated water,
including drinking water. The goal is to combine
expertise to develop robust analytic methods to detect
biological evidence of exposure to cyanobacterial
toxins, to optimize laboratory and emergency response
capacity in the collection, analysis, and response to
harmful algal bloom-related illnesses.
Method development, refinement, and validation for

detecting human exposures to HAB toxins through the
detection of toxins and specific biomarkers in clinical
DHHS CDC HABs samples. Current methods approved for use include the
detection of saxitoxin, neosaxitoxin, tetrodotoxin, and
gonyautoxins (1-4), which have been applied to
individual cases to confirm suspected HAB exposures.
Method development, refinement, and validation for

detecting HAB toxins; improving understanding of HAB
DHHS FDA HABs
toxin sources and vectors that impact seafood and
dietary supplement safety.
Developed, evaluated, and validated rapid screening

for HAB toxins in seafood, thereby improving
regulatory monitoring, surveillance programs, and
outbreak response. For example, FDA developed an
DHHS FDA HABs
onboard screening dockside testing program for PSP
toxins in shellfish, which led to reopening of a large
portion of Georges Bank in 2013 to safe commercial
harvest of clams.
The Northwest Fisheries Science Center (NWFSC) has

established a new monitoring partnership called
SoundToxins for the early warning of marine harmful
algal blooms in Puget Sound. The NOAA National
Centers for Coastal Ocean Science ECOHAB program
DOC NOAA HABs SoundToxins provided 3 years of funding to develop the Puget Sound
Harmful Algal Bloom (PS-AHAB) project to understand
environmental controls on the benthic (cyst) and
planktonic life stages of the toxic marine dinoflagellate
Alexandrium, and evaluate the effects of climate
change on the timing and location of blooms.
54
HABs/
Both
The Phytoplankton Monitoring Network (PMN) was
established to monitor phytoplankton and harmful
algal blooms and promote environmental stewardship
though the use of citizen volunteers. PMN volunteers
are trained by NOAA staff on sampling techniques and
identification methods for over 50 genera, including 10
National potentially toxin-producing genera, of dinoflagellates
DOC NOAA HABs Phytoplankton and diatoms on the volunteers watch list. Currently,
Monitoring Network 250 marine and Great Lakes sites in 22 states and U.S.
territories including 52 schools, 15 universities, 298
civic groups and 40 state and federal agencies collect
phytoplankton and environmental data. Since the
inception of the program in 2001, more than 275 algal
blooms and 15 toxic events have been reported by
PMN volunteers.
NOAA’s Analytical Response Team (ART) provides rapid

and accurate identification and quantification of
marine algal toxins in suspected harmful algal blooms
(HABs), and related marine animal mortality events and
human poisonings. From 2009 to 2014, ART received
National Analytical over 4000 samples from state and federal government
DOC NOAA HABs
Response Team agencies, NGOs, and academic partners for
determination of toxins associated with harmful algae.
In addition to water samples, marine toxins were
analyzed in samples from marine and freshwater algae,
shellfish, fish, cetaceans, pinnipeds, birds and sea
turtles.
The Technology Transfer Team completed rigorous,

international inter-laboratory trials in partnership with
interagency organizations, federal agencies and private
businesses to bring the receptor binding assay for
paralytic shellfish poisoning (PSP) toxins to U.S. and
international regulatory approval; and guided its
commercialization to assure U.S. marine shellfish are
Technology Transfer
DOC NOAA HABs safe for U.S. citizens and export throughout the world.
Team
The team also provided training on use of the method
to more than 30 countries through formal agreement
with the International Atomic Energy Agency to
promote safe food supply and increased economic
growth through the export of fisheries products and to
the Southeast Alaska Tribal partnership to enable
monitoring of subsistence resources.
The National Centers for Coastal Ocean Science

(NCCOS) develops and transitions HAB forecasts for
coastal and Great Lakes waters. NOAA is also working
Ecological
DOC NOAA HABs with EPA on systematic approach to either warning
Forecasting
state health/water quality on cyanobacteria blooms
and allowing them to evaluate patterns and trends in
lakes and estuaries that are at risk.
55
HABs/
Both
Continue to make satellite coverage of ocean and
coastal zones more comprehensive and combine it with
existing data to enable quantifiable estimates of HABs
(much of this has been funded by NASA). Plan to
DOC NOAA HABs transfer promising new monitoring and prediction
technology and approaches from research to
operational HAB forecasts for Gulf of Mexico, Lake Erie,
Chesapeake Bay, Puget Sounds, Pacific Northwest, and
California.
Ecology and
Developing a better understanding of marine HAB
Oceanography of
DOC NOAA HABs causes and impacts that form the basis for better
Harmful Algal Blooms
management to reduce HABs and their impacts.
(ECOHAB)
Monitoring and National, competitive extramural research program

Event Response for that builds capacity for enhanced coastal and Great
DOC NOAA HABs
Harmful Algal Blooms Lakes HAB monitoring and response in state, local, and
(MERHAB) tribal governments.
National, competitive extramural research program

Prevention, Control,
that develops new methods of coastal and Great Lakes
and Mitigation of
DOC NOAA HABs HAB prevention, control, and mitigation. It also
addresses the socioeconomic impact of HABs and
(PCMHAB)
efforts to reduce HAB impacts.
Provides modest support to supplement monitoring of

DOC NOAA HABs Event Response coastal and Great Lakes HAB events, and advance the
understanding of HABs when they occur.
National, competitive extramural research program

that develops understanding of hypoxia causes and
impacts that form the basis for better management to
Coastal Hypoxia
DOC NOAA Hypoxia reduce hypoxia and its ecological and socioeconomic
Research (CHRP)
impacts. Includes all coastal systems except the large
hypoxic zone along the northern Gulf of Mexico
continental shelf.
Northern Gulf of National, competitive extramural research program

Mexico Ecosystems that develops understanding of the causes and impacts
DOC NOAA Hypoxia and Hypoxia of the northern Gulf of Mexico hypoxic zone, that form
Assessment Program the basis for better management to reduce the hypoxic
(NGOMEX) zone and its ecological and socioeconomic impacts.
Continue to convene workshops to obtain stakeholder

needs that drive research prioritization, and
disseminate advanced knowledge and tools for hypoxia
DOC NOAA Hypoxia
mitigation to regional managers and interagency
management networks such as the Gulf Hypoxia Task
Force or the Landscape Conservation Cooperative.
56
HABs/
Both
Studies molecular ecology of HABs in the Great Lakes
and makes improvements to monitoring HABs and
toxicity in the Great lakes. Monitors six routine stations
in the western basin of Lake Erie weekly during blooms
season and supplies data that supports the predictive
DOC NOAA HABs
models in Lake Erie. Developing a three dimensional
lagrangian particle transport model to effectively
predict HAB advection as part of the Lake Erie
Operational Forecasting System, which is set to go
operational through fiscal year 2015.
Responding to HABs in response to public

reports/complaints in close coordination with state
water quality/public health agencies. Response
programs developed by individual USACE
Divisions/Districts. The Engineer Research and
DOD USACE HABs Development Center (ERDC) is available to support
Divisions/Districts in assessing HAB impacts to USACE
Civil Works Projects (e.g., WQ modeling, remote
sensing, and technical assistance). General WQ
monitoring and HAB response to meet authorized
project purposes and recreation mission requirements.
Studies the variability of in situ and remotely sensed

spectral optical properties to identify dinoflagellates
through field sampling and improvement of remote
Oceanographer
DOD HABs sensing techniques. Dinoflagellate information has
of the Navy
been incorporated into Naval Research Laboratory’s
ecological-circulation models for better
understanding/prediction.
Study plan pending approval: To address noted data

Gulf Oxygen gaps that addresses deepwater oxygen dynamics such
DOI BOEM Hypoxia Deepwater as in the Oxygen Minimum Zone in the Gulf, whereas
Experiment the Louisiana-Texas (LATEX) shelf studies were on the
shallower hypoxic zone.
57
HABs/
Both
Of the 407 NPS units, there are 86 units that are
considered ocean, coastal, or Great Lake parks, in
addition to other park units that have extensive surface
water bodies. HABs have the potential to influence all
of these park units at various levels, and it is therefore
important to prepare for these events in order to
preserve our resources. The National Park Service is
creating a website containing a public health and
HABs and Outreach and ecological HAB events reporting system. It also
DOI NPS
hypoxia Education provides a point of contact for park managers to
partner with local and state health and environmental
agencies who will provide park personnel with
technical assistance for the management of HAB
events. Outreach materials (brochures, interpretive
displays and materials) on HABs, their causes, the
effects on the ecosystem, and the many ways to reduce
or stop nonpoint source pollution, many of which are
simple to implement.
USGS conducts long-term monitoring of nutrients and

other water-quality characteristics in surface and
groundwater networks. The sources and quantities of
nutrients delivered by streams and groundwater to
coastal areas and the Great Lakes are monitored at 106
sites. Annual updates from the monitoring sites are
made available to the public, including nutrient
concentrations, loads, and yields. These data, along
with data aggregated from numerous other agencies,
are used to evaluate trends in critical water quality
parameters including nutrients and sediment. Real-
time measurements for dissolved oxygen and
HABs and National Water
DOI USGS temperature are collected at over 500 and 2000
hypoxia Quality Program
locations, respectively. USGS is pioneering new field
sensor methods and systems for monitoring and
delivering real-time nutrient data, with over 100 nitrate
sensors deployed. The USGS SPARROW model
quantifies nutrient sources and sediment loads to
coastal areas, the Great Lakes, and inland lakes in the
Eastern U.S. SPARROW has also been linked to an
online Decision Support System, which allows direct
exploration of the potential benefits of nutrient
management for systems including the Chesapeake
and the Mississippi, other coastal rivers, and the Great
Lakes.
National Water USGS collects fish-, aquatic macroinvertebrate-, and

Quality algae-community samples, and conducts stream
HABs and
DOI USGS Program/National physical habitat surveys to assess the effects of multiple
hypoxia
Water Quality stressors—including algal toxins—on aquatic
Assessment organisms in streams in several ecoregions.
58
HABs/
Both
HAB research is conducted in at least 20 USGS Water
Science Centers. Studies include both short- and long-
term projects focused on quantifying blooms and
associated toxins and taste-and-odor compounds, and
National Water
understanding causal factors. Many studies employ
Quality Program/
DOI USGS HABs new and developing sensor technology to detect algal
Cooperative Water
pigments. For example, a study of the primary drinking
Program
water supply for Wichita, Kansas combined long-term
discrete and continuous water-quality data to develop
models that estimate the probability of microcystin
occurrence in near real time.
Assesses the impacts of agricultural management
practices, climate change, and land use change on the
timing and magnitude of delivery of nutrients and
sediments to the Great Lakes at 30 sites. Works with
NOAA, EPA, states, universities, and NGOs to
understand how nutrient and sediment loading from
USGS, USDA- HABs and the Great Lakes watershed affect hypoxia, HABs and
DOI, USDA GLRI
NRCS hypoxia biological communities in the near-shore environment.
Edge-of-field studies in GLRI priority watershed
quantify phosphorus, nitrogen, and sediment to
evaluate nutrient reduction projects on agricultural
land. Rapid sharing of edge-of-field monitoring results
with local stakeholders allows for adaptive
implementation.
Pioneer new field monitoring methods (sensors),

Energy, Mineral, and assessment techniques, and laboratory methods
Environmental needed to address harmful algal bloom issues in
DOI USGS HABs Health/Toxic freshwaters. New methods include a multi-toxin
Substances method that can quantify cyanotoxin mixtures, and
Hydrology Program DNA- and RNA-based molecular methods for detecting
microcystin and microcystin producers.
USGS has ongoing research characterizing ecological

and food web impacts of cyanotoxins. For example, a
DOI USGS HABs Ecosystems USGS study in Upper Klamath Lake demonstrated a link
between microcystin and reduced young-of the year
recruitment of federally endangered suckers.
The NIEHS supports multiple studies focused on the

effects of HAB toxins on human and mammalian
physiology, development of biomarkers for chronic
toxin exposure, and the design and testing of novel
Ocean and Human technologies for in situ detection of algal toxins in fresh
Health (OHH) and salt water environments. For example, a number of
HHS NSF/NIEHS HABs Initiative and the ongoing studies are supported that analyze the effects
NSF's Division of of domoic acid on neurotoxicity as well as cognitive
Ocean Sciences impacts in human cohorts, non-human primates and
rodent models. Also, NIEHS is accepting unsolicited
applications for support and use of time-sensitive
mechanisms to allow research support for
unanticipated bloom events.
59
HABs/
Both
Ocean Observing
HABs and Initiative and the Provides environmental data for studies of HABS (both
NSF NSF
hypoxia National Ecological marine and freshwater) and hypoxia.
Observatory Network
Division of Ocean
Sciences (OCE), NSF Observational capabilities for research in marine
NSF NSF HABs
Ocean Observing systems.
Initiative (OOI)
Directorate of
Geosciences,
Prediction and
NSF NSF HABs Focused interdisciplinary research projects.
Resilience Against
Extreme Events
(PREEVENTS)
Division of Biological
Infrastructure,
NSF NSF HABs National Ecological Observational capabilities for ecological research.
Observatory Network
(NEON)
Division of Ocean Research Support, unsolicited proposal in marine

NSF NSF HABs
Sciences ecology.
Program supporting interdisciplinary research to

Collaboration understand and predict the interactions between the
between NSF GEO, water system and climate change, land use, the built
HABs and
NSF NSF SBE, and ENG environment, and ecosystem function and services
hypoxia
directorates, as well though research and models. Several research projects
as USDA NIFA. are focused on nutrient movement and hypoxia
mitigation strategies.
Ocean and Human

Health Initiative, a Studies of the effects of HAB toxins on human and
collaboration mammalian physiology, development of biomarkers for
between NSF's chronic toxin exposure, and the design and testing of
Division of Ocean novel technologies for in situ detection of algal toxins
NSF NSF HABs
Sciences (OCE), and in fresh- and salt-water environments. Also accepting
the National Institute unsolicited applications for support and use of time
for Environmental sensitive mechanism to allow research support for
Health Sciences unanticipated bloom events.
(NIEHS)
Support of extramural and intramural research on the

USDA NIFA and ARS HABs effects of HABs and HAB toxins on food safety,
aquaculture, and livestock.
Research on nutrient management, nutrient

contribution to hypoxia, and aquaculture. Long-Term
USDA ARS Hypoxia
Agro-Ecosystem Research (LTAR) and Watershed
Research Centers.
60
HABs/
Both
Research support for studies of the effects of nutrient
USDA NIFA Hypoxia cycling, climate change, and nutrient management for
agriculture.
NRCS provides conservation planning assistance to

agricultural producers on cropland, grazing land, and
for confined livestock operations. NRCS also has
Conservation financial assistance programs to help producers
Technical Assistance implement and install practices. These programs are all
(planning); voluntary and are incentive-based. For confined
livestock systems, this includes, but is not limited to,
Environmental practices such as waste storage structures, and
Quality Incentives associated practices like roofs and covers, roof runoff
Program; management, diversions, and a nutrient management
Conservation plan for the utilization of manure.
HABs and
USDA NRCS Stewardship On cropland, this may include agronomic practices such
hypoxia
Program; as residue management, cover crops, conservation
Agricultural cropping systems, and nutrient management; buffer
Conservation practices like filter strips and riparian forest buffers;
Easement Program water management practices such as grassed
waterways, grade stabilization structures, drainage
Regional water management, blind inlets (to replace surface
Conservation inlets), wetland restoration and creation; and
Partnership Program prescribed grazing systems and associated practices for
grazing land. The Natural Resources Conservation
Service (NRCS) also assists farmers financially with
edge-of-field water quality monitoring.
Under various water quality initiatives, NRCS and its

partners help producers in selected watersheds to
Great Lakes
voluntarily implement conservation practices that
Restoration Initiative,
avoid, control, and trap nutrient runoff; improve
Mississippi River
wildlife habitat; and maintain agricultural productivity.
USDA NRCS Hypoxia Basin Healthy
These initiatives utilize NRCS programs such as the
Watershed Initiative,
Environmental Quality Incentives Program (EQIP) and
National Water
the Conservation Stewardship Program (CSP) within
Quality Initiative, etc.
targeted watersheds to provide technical and financial
assistance.
Supports research on best management practices for

nutrient management, aquaculture, and plant
HABs and breeding, among others. Specific concerns addressed
USDA NIFA and ARS
hypoxia by this research include manure management from
animal feeding operations and water use and
conservation on irrigated cropland.
61
HABs/
Both
Diversified approach to better understand
cyanobacterial HABs ecology and the development of
watershed and source water management techniques,
including the development of models for nutrients
loadings, the optimization of watershed placement of
phosphorus and sediment BMPs, and the use of water
quality trading (WQT) to cost-effectively reduce
HABs and Water Quality nutrient loadings. It also includes an assessment of the
EPA EPA
hypoxia Management impact of land use and infrastructure on watershed
changes, and the evaluation of ecological contributors
to cyanobacterial HAB development and toxin
production. This research program also includes the
use of molecular methods to characterize the risk in a
reservoir for toxin and algal blooms, and the analysis of
the impact of HABs on creating disinfection byproducts
(DBPs) precursors.
Research support to address data gaps associated with

health, ecosystem, and economic effects of HABs.
Research activities include the characterization of
cyanobacteria and their toxins and allergic
components, the evaluation of the toxicity of multiple
congeners of microcystins, and identification of
biomarkers of exposure for human health risk
assessments. Epidemiology studies to characterize
Human and toxin occurrence in U.S. inland lakes, and studies to
EPA EPA HABs
Ecological Health determine that bioaccumulation, bioconcentration,
and biomagnification of cyanotoxins in mammalian
tissues and food web are also in place. EPA is also
assessing occurrence and health information for the
inclusion of cyanotoxins in the Contaminant Candidate
List (CCL) and the Unregulated Contaminant
Monitoring Rule (UCMR) program. In addition, EPA is
developing Human Health Water Quality Criteria
(HHWQC) for cyanotoxins in recreational waters.
62
HABs/
Both
A collaborative effort of EPA, NASA, NOAA, and USGS to
provide an approach for mainstreaming satellite ocean
color capabilities into U.S. fresh and brackish water
quality management decisions. The Cyanobacteria
Assessment Network (CyAN) for freshwater systems
will develop approaches to relate nutrient loads and
land use to the frequency, location, and severity of
cyanobacterial blooms in lakes of the United States. It
will include assessing risk to human health from
satellite multispectral data to assess biological
conditions and risk to human health in lakes and
reservoirs in the United States.
EPA also provides nationally consistent and
scientifically defensible assessments of aquatic
Monitoring and resources through the National Aquatic Resource
EPA EPA HABs Analytical Methods Surveys (NARS), including indicators associated with
Development cyanotoxin exposure. EPA and its regions are also
working on monitoring efforts such as the Lake
Champlain Cyanobacteria Monitoring, Great Lakes
Restoration Initiative projects and Phosphorus
Reduction Strategy, Southeast Alaska Tribal Toxins
(SEATT) project, and the Puget Sound Toxins Project.
EPA is also working on monitoring projects to improve
identification and removal of HAB toxins in drinking
water, and evaluating the impact of temperature on
bloom development.
EPA is developing analytical tools including the use of
real-time sensors, qPCR and fluorescence based
technologies of micro spectrophotometer and flow
cytometry to detect cyanobacteria organisms in source
water.
EPA is working collaboratively with regional offices to

Drinking Water
EPA EPA HABs characterize the effectiveness of drinking water
Treatment
treatment techniques in reducing toxins.
EPA conducts webinars and provides online resources

EPA EPA HABs Outreach
to promote public awareness and information sharing.
The Ocean Biology

Basic HABs research resulting in publications and new
NASA NASA HABs and Biogeochemistry
retrieval algorithms.
Program
Improve the forecast resolution and frequency of risk

Health and Air
of Karenia brevis toxins on every beach, every day,
NASA NASA HABs Quality Applications
rather than every county, twice a week. The methods
Program
would be applicable across the Gulf of Mexico.
63
HABs/
Both
Monitoring and surveillance of cyanobacterial harmful
algal blooms (CyanoHABs) in drinking and recreational
water supplies. Satellite derived products that were
developed for western Lake Erie are being analyzed for
Health and Air
their use in other regions (e.g., Chesapeake Bay and
NASA NASA HABs Quality Applications
inland lakes in Ohio and Florida). This project has
Program
established methods to identify environmental
thresholds that indicate the potential for
cyanobacterial blooms to form or persist, and these
data sets are also being made available to CDC.
Co-sponsors of the Water Quality Portal, a cooperative

Multiple
data service that makes data publically available. The
agencies and
data are derived from the USGS National Water Quality
partners,
Information System (NWIS), the EPA Storage and
including but
HABs and Retrieval data warehouse (STORET), and the USDA ARS
not limited to Water Quality Portal
hypoxia Sustaining the Earth's Watersheds - Agricultural
EPA, FWS,
Research Database System (STEWARDS). With data
NOAA, NPS,
from over 400 federal, state, tribal, and local agencies,
USACE, USDA,
this efforts will improve understanding of progress in
USGS
nutrient reduction efforts.
Exposure assessment is instrumental in helping to

forecast, prevent, and mitigate exposure that leads to
Multiple ES21 Federal adverse human health or ecological outcomes. This
agencies: CDC, Working Group on vision expands exposures from source to dose, over
NASA, NOAA, HABs Exposure Science time and space, to multiple stressors, and from the
NSF, USDA, and molecular to ecosystem level. HAB exposure
USGS assessment is addressed by ES21 Working Groups on
Biomonitoring, Citizen Engagement/Citizen Science and
Sensors/Dosimeters.
Volunteer monitoring program that collects baseline

data on harmful algal species and builds capacity by
providing data to NOAA Phytoplankton Monitoring
Network and EPA. Volunteers are trained to identify
Volunteer
algae, collect water samples, conduct basic water
Multiple Freshwater
quality analyses, and preserve samples for further
Agencies, EPA HABs Phytoplankton
analysis by the NOAA Analytical Response Team.
and NOAA Monitoring Program
Network became operational in 2015 with stations in
the Western Basin of Lake Erie in seven lakes in EPA
Region 8, with plans to expand to Lakes Michigan,
Superior, Huron and Grand Lake St. Mary in 2016.
64
HABs/
Both
The Conservation Effects Assessment Project (CEAP) is
a collaborative, multi-agency effort to quantify the
environmental effects of conservation practices and
programs and develop the science base for managing
the agricultural landscape for environmental quality.
Project findings are used to guide USDA conservation
policy and program development, and help
conservationists, farmers, and ranchers make more
informed conservation decisions. USGS will incorporate
conservation data collected by CEAP into its surface
water quality monitoring.
CEAP-Croplands developed a National Resources
Inventory (NRI) statistical approach that combines
information voluntarily collected through NASS
producer surveys, and conservation practice data as
inputs into two process based models, the Agricultural
USDA/Multiple Policy Environmental eXtender (APEX) field-scale
agencies, led by model and the Soil and Water Assessment Tool (SWAT)
USDA NRCS, watershed scale model. In addition to determining
ARS, NIFA, FSA, conservation practice adoption trends, the CEAP
and NASS. Also modeling team is able to estimate the environmental
includes USGS, benefits of conservation practices and conservation
NOAA, FWS, Both CEAP treatment needs within major drainage basins of the
EPA, BLM, United States. In the first CEAP-Croplands National
NASA, USDA Assessment, current conservation conditions and
Economic outstanding needs were assessed in twelve major
Research basins, including the Mississippi River Basin,
Service and US Chesapeake Bay and Great Lakes. Since the 2003-06
Forest Service national survey CEAP-Croplands has revisited
watersheds through special studies, including
Chesapeake Bay (2011), Western Lake Erie Basin
(WLEB) (2012), California Bay Delta (2013), and the St.
Francis and Lower Mississippi River Basin (2014). A
second CEAP-Croplands National Assessment was
initiated in 2015. In addition, the Watershed
Assessment Component of CEAP continues to conduct
small watershed-scale studies across the United States
to quantify water and soil resource outcomes of
conservation practices and systems and enhance
understanding of processes. Interactions among
practices are investigated as well as modeling
enhancements, watershed targeting approaches, and
socioeconomic factors. Practice standards are
developed or updated to improve effectiveness and
address gaps.
In 2012, NASS worked with NRCS to administer a CEAP

Multiple Cropland-survey focused on the Western Lake Erie
agencies, led by Basin (WLEB). Data from the survey and other sources
USDA USDA NRCS, CEAP is being used to assess conservation effects in the WLEB
ARS, NASS and and compare trends and progress in conservation as
FSA well as evaluate additional treatment needs in that
region. The assessment report is forthcoming.
65
X. Appendix 5. Summary of Stakeholder Input
The information transcribed below is based upon oral and written comments received during the
Listening Session Webinar on September 16, 2015. This summary is the EPA’s best effort to accurately
record input received by stakeholders. The EPA utilized the input received and incorporated elements
into this strategic plan.
Public Statements
Scott Biernat, Association of Metropolitan Water Agencies (AMWA)
AMWA represents large publically owned systems. AMWA appreciates the technical and
implementation challenges EPA faces in addressing the risks posed by HABs and toxin exposure via
drinking water sources. Plans for addressing must be carefully crafted and implemented to achieve
optimal public risk reduction and public health benefits.
HR 212 presents an important opportunity to set the path for a thoughtful collaborative approach for
addressing challenges posed by algal toxins in drinking water. The process that is established for further
evaluation and reevaluation of algal toxins risks is important to assure optimal risk reduction. This
process must include continuous in-depth consultation with stakeholders to ensure all necessary
expertise and practical experience is brought to the table. Based on the information gathered that lead
to the cyanotoxin Health Advisories, EPA has a good understanding of the nature of the additional
information needed to assess next steps to further reduce algal toxin risk. AMWA notes several points:
 First, the best and most cost effective long range strategy to protect the public from algal toxins
is to prevent bloom-causing nutrients from entering waterway in the first place. In that regard, a
meaningful reduction in algal blooms must begin with the agricultural sector. The development
of a bolder, more innovative strategy for managing nonpoint source water pollution, particularly
from the agricultural sector, must be a part of the strategic plan.
 Second, the strategic plan must place an emphasis on developing robust analytical methods in
time for including cyanotoxins on the Unregulated Contaminant Monitoring Rule (UCMR 4).
Collection of occurrence data under UCMR 4 can fill key information gaps related to algal toxin
occurrence and provide a vital foundation for additional risk assessment as mandated by SDWA.
These analyses will inform future stakeholder discussions and policy decisions from EPA and
other local, state and federal agencies intended to ensure algal toxins do not pose human health
risks if they reach drinking water supplies.
 Third, the assessment of existing guidance and support documents, including all existing health
criteria documents should be identified as an iterative process within the strategic plan, drawing
on lessons learned and new data as they become available to make appropriate and timely
updates. With the summer bloom season under the new cyanotoxin health advisories now
behind us, it is a particularly good time to reengage stakeholders to evaluate how to best
address algal toxin challenges.
 Additional consideration of communication challenges regarding algal toxin health risks should
be a focus of the plan. The setting of the health advisories based on a 10-day exposure and at
two different age-based levels poses a unique public communication challenge. Further
66
guidance must offer robust interpretation of how to evaluate the health advisory levels within
this 10-day timeframe that corresponds to actual risk involved. Going forward it will be the
strength of the collaborative process put in place to augment existing programs and processes
that will determine how efficiently and effectively any data and information gaps are filled, and
will help in identifying the action that will most effectively manage algal toxin risks. Combined
with the already robust process for evaluating contaminant risk and evaluating the need for
further regulation required under SDWA, such evaluation will ensure sound policies are
developed.
In closing, AMWA thanks EPA for initial efforts in getting ahead of algal toxin issues and looks forward to
collaboration on its next steps.
Steve Via, American Water Works Association (AWWA)
AWWA appreciates the opportunity for stakeholder involvement and input. AWWA understands the
agency is working on a tight schedule and there is a lot of work to do to meet the deadline. Congress
provided good direction, and the health advisory documents and their recommendations are important
guides to taking on the task to assess and manage algal toxins. Some detailed suggestions:
 Health effects: look beyond microcystins LR, right now working on assumption that all
congeners have equal health effects, and that it is true of natural congeners and degradates.
That assumption might not prove valid in the long haul. Useful for agency to communicate next
steps for anatoxin-a, the health assessment only covered microcystins and cylindrospermopsin
while there were a number of identified data gaps with anatoxin-a.
 Important to realize goal for occurrence is both assessment and management of cyanotoxins.
One of the key gaps from utilities is the need for ongoing and reliable monitoring systems, for
toxins and the parameters that inform their management, both in water supplies and influent
waters.
 Look to other strategies under other vehicles like the Clean Water Act, Unites States Geological
Survey (USGS) and others to develop data systems to inform risk management at water supply
level. Technical tools and support for action at utility level – clear awareness what information is
actually useful for decision making. For example, lysing source water and using that observation
as assessment of risk. To make a treatment change, that lyse data may or may not be most
important especially for removal by coagulation and settling.
 Help make strong technical decisions and provide good support for a regulator rather than
utility itself. There has been a lot of discussion about what methods should be used, need robust
testing in relevant concentration range which is between 0.1-3 ug/L for microcystins rather than
data driven by higher level concentrations.
 EPA source water collaborative with the United States Department of Agriculture (USDA)
conservation division has identified a number of practices for achieving land use management,
and implementation of best management practices.
 We really need to think about risk communication in context of risk management. One doesn’t
occur well without the other. A 10-day HA is challenging construct. Understanding impacts of
67
management action, cost consequences for a community as they are thrown in preparation and
response measures.
Beth Messer, Ohio EPA
Ohio EPA Responses to U.S. EPA HAB Webinar Questions – verbal comments
In an effort to keep the comments brief, Ohio EPA will provide verbal comments based on our
experiences and what we consider the highest priority.
1) What do you consider to be the key information gaps in understanding the human health
effects of specific algal toxins?
Additional research is needed for all microcystin variants, as well as saxitoxin for acute, short-
term, and long-term exposures.
More information is needed on potential sensitive population exposures such as the elderly,
immuno-compromised or individuals with pre-existing liver disease or on dialysis, and data is
needed on whether cyanotoxins cross the placental wall and potential exposure via breastmilk
2) What do you consider to be the key information gaps in understanding the occurrence of
blooms and toxin formation?
More information is needed on:
 the triggers for cyanotoxin production, release and degradation;

 triggers for cyanobacteria to release cyanotoxins to the extracellular form;
 movement of blooms within the water column, which may assist public water systems
with avoidance strategies
 the role of resting cells/akinetes in the probability of future bloom formation; and
 factors that contribute to a shift in phytoplankton community dominance to HABs.
3) Please identify effective technical support or tools, including outreach and education efforts,
that could benefit states’ and PWSs’ ability to predict, prevent and mitigate the occurrence of
algal blooms and inform management decisions.
Guidance is needed on effective reservoir management strategies, in particular the impacts of

algaecide application. Many public water systems rely on algaecide as a source control strategy.
The effectiveness of this strategy could be improved through guidance on proper application
rates and timing for different types of blooms, the effect of algaecides on akinetes, and
information on the effect on community dynamics and possible long-term implications of use
(specifically copper resistance).
Finally, Ohio has found remote sensing data to be useful tool, and recommends continued
support for remote sensing projects including NOAA’s Lake Erie HAB Bulletins and forecasts,
CyAN collaborative efforts, and NASA and other research efforts on use of multi-spectral sensors
(aircraft or drones).
68
4) In your opinion, what are the most important steps that can be taken to improve strategies
for use of HAB-related analytical methods, monitoring, and treatment of harmful algal toxins
for drinking water?
More information is needed to optimize treatment strategies, including:
 CT tables or cyanotoxin reaction kinetics for microcystin (including variants other than
MC-LR) and saxitoxin for commonly used oxidants such as chlorine and permanganate
under variable pH, temperature, and concentration ranges;
 the effect of permanganate on cell lysis for genera other than microcystis;
 the effectiveness and operational guidance for granular activated carbon (GAC),
including different types of carbon on saxitoxin, MC-LR and other common microcystin
variants;
 saxitoxin adsorption capacity for different types of powdered activated carbon; and
 short-term plant optimization guidance for removal/destruction of cyanotoxins via
conventional surface water treatment processes.
Thank you for providing Ohio EPA the opportunity to comment.
Rob Blair, Kentucky Division of Water
The comments Kentucky put together have been addressed by other presenters.
Van McClenden and Ken Hudnell, North American Lakes Management Society
(Comments were read, and also submitted electronically. Comments appear here as they were
submitted electronically)
1. I commend EPA for heightened concern about protecting the public from the health risks posed by
cyanotoxins produced by cyanobacteria. The Agency has taken a number of steps in this direction,
including:
- April 2015 - Teaming with NASA, NOAA, EPA and USGS on developing a satellite surveillance
system
- June 2015 - Producing Drinking Water Health Advisories for Two Cyanobacterial Toxins
- June 2015 - Recommending that Public Water Systems Manage Cyanotoxins in Drinking Water
- By having two cyanotoxins on the UCMR4 candidate list and apparently moving toward
implementing testing at utilities to better understand the scope of cyanotoxins in source water
and finished drinking water
- And finally, hosting this webinar to get public input on developing a strategic plan for
addressing cyanotoxins in drinking water
2. I also commend EPA for realizing that the incidence of cyano HABs is increasing even as we’ve spent
decades and many millions of dollars on the watershed management point- and nonpoint-source
programs to reduce new nutrient inputs into freshwaters. I hope and think that the Agency is moving
toward re-implementing the CWA’s Clean Lakes program that calls for treating impaired waterbodies.
69
Even Ben Grumbles who made the decision to de-emphasize waterbody treatments in the early 1990s
has told me he regrets that decision and no longer believes that watershed management alone can
reverse the trend of increasing freshwater impairment. Movement toward fully implementing the CWA
by complementing watershed management with waterbody management is indicated by:
- December 2013 – EPA produced A Long-Term Vision for Assessment, Restoration, and
Protection under the Clean Water Act Section 303(d) Program (Webpage to PDF) – that allows
for Alternative Approaches – “By 2018, States use alternative approaches …that incorporate
adaptive management and are tailored to specific circumstances where such approaches are
better suited to implement priority watershed or water actions that achieve the water quality
goals….”
- May 2014 – Webinar, A Systems Approach to Freshwater Management: Waterbody Treatments
- September 2014 – Putting up a webpage on Nutrient Policy, Controls & Waterbody Treatments
3. As the Agency develops a strategic plan for addressing Algal Toxins in Drinking Water, I urge the
Agency to seriously consider using an Adaptive Systems Approach to Freshwater Management as
described by the North American Lake Management Society or NALMS. I’ll be glad to send you a
description of an ASA, but the core of an ASA is using rigorous science and cost-benefit analyses in
putting together a feasible plan that uses the best of watershed and waterbody management tools. An
ASA uses:
- rigorous science in consideration of the physical, chemical, and biological characteristics of

freshwaters to identify the direct and contributing causes of impairments, in this case
eutrophication and cyano HABs
- cost-benefit analyses of waterbodies’ designated uses, in this case drinking source water, and
of all watershed and waterbody management tools
- an implementation strategy for drinking water protection should reduce the risks of adverse
health effects from cyanotoxins in the near term at an affordable cost. If a strategy is not being
effective enough, the ASA process is reiterated, and the strategy is adapted to produce a more
effective strategy
4. A drinking water protection strategy should focus on preventing cyanobacteria from proliferating in
source waters by using sustainable waterbody treatments to suppress cyanobacteria and remove or
deactivate nutrients in the waterbody where they are highly concentrated and easy to get to. If
cyanotoxins are not in the source water, they cannot be in the finished drinking water. It is much
cheaper to combine technologies to treat source waters than it is to annually spend millions of dollars
on removing cyanotoxins from drinking water using activated carbon, ultraviolet irradiation, or
microfiltration. And none of these ensures that all cyanotoxins will be removed from drinking water.
5. There is currently a movement towards combining waterbody treatments to suppress cyanobacteria

and remove or deactivate nutrients. Projects are being planned to assess the efficacy of combining
water circulation to suppress cyanobacteria and synergize nutrient removal or deactivation by other
technologies. For example, circulation and ultrasound guns can likely suppress cyanobacteria in even the
most difficult waterbodies where shallow, weeded areas continually seed cyanobacteria into open
70
waters. And circulation increases nutrient uptake by plants on floating artificial wetlands, and growth of
periphyton on curtains that transfer nutrients from the water column to higher trophic levels.
Circulation can keep specialized micro-pellets suspended that stimulate blooms of diatoms that also
transfer nutrients from the water column to higher trophic levels. A treatment system could also meter
out flocculants and keep them suspended to capture and deactivate nutrients as they enter a waterbody
from a stream.
6. EPA should contribute funds to the 3 NOAA HAB research grant programs as directed by Congress in
last year’s reauthorized and expanded HABHRCA law. These funds could support studies such as that
just mentioned, and enable utilities to make better informed decisions about approaches to source
water protection.
7. To assist states and utilities, EPA should revive the Section 314 grant program that provided money
for waterbody treatments. EPA can encourage states to include waterbody management treatments in
their watershed management plans that EPA approves. Currently, the Agency only encourages states to
use 5% of Section 319 funds for waterbody treatments.
8. Finally, EPA should request line-item funding for funds to support the 3 NOAA HAB research-grant
programs and the Section 319 waterbody treatment program.
I would also like to comment and commend the EPA for moving forward on addressing such an
important public health risk and to emphasize the need to move expeditiously on creating regulatory
requirements for testing and treating impaired water bodies. We feel the Adaptive Systems Approach
will bring a science based assessment to the problem and provide an effective cost-beneficial solution to
an issue that has been under addressed in the public arena.
Thank you for allowing me to provide input on this very important project. Both I and NALMS will gladly
supply any assistance that we can.
Rebecca Gorney, NY State Department of Environmental Conservation, Department of Lake

Management
NY requests assistance on question #3: What are the definitions of appropriate thresholds in variety of
contexts like algal biomass in terms of determining bloom concentrations and bloom existence as well as
nutrient concentration and how these can be used to prevent and mitigate blooms once occurring? Also
need help understanding secondary health factors related to toxin removal in drinking water. Useful to
better understand secondary health factors, such as disinfection by-products and what happens after
toxins are removed.
Clayton Creager, CA North Coast Regional Water Quality Control Board
Comments are not from the drinking water division, but CA is dealing with nuisance levels of blue green
algae and have a total maximum daily load (TMDL) for microcystins for rivers. We have compiled
endpoint numbers for public health warning across states and found quite a bit of variance. Is this an
artifact of risk calculation equations, or a difference in toxicity data available to the states doing the
calculations? Has anyone looked at consistency both for drinking water and health advisories and why
they vary? Drinking water systems draw from rivers that are seeing more frequent occurrence of
nuisance levels in flowing water. Lower flows and higher temps and more abundant growth of
71
filamentous algae as a substrate are three things to cause issues. There is little information to evaluate
the implications of this apparently increasing trend. CA has found a high correlation of 10 ug/L
chlorophyll-a as inflection point to where species composition dramatically shifted to cyanobacteria.
That became our threshold as a biomass indicator.
Lynn Thorp, Clean Water Action
Clean Water Action recognizes the ongoing work by EPA and what Congress has asked them to do, but
still thinks there is value in addressing info gaps in the contribution of different kinds of pollution and
source water contamination to occurrence of cyanotoxins. Also information gaps on how reductions in
pollutants will help in short and long terms with the jobs that utilities and EPA now have to take on. We
think this is a really important and interesting example of the integration of CWA and SDWA which has
been a priority both inside and outside the agency for some time. Consider these connections as part of
way to respond to Congress. Agree about risk communication as a part of the response, and need to
include non-drinking water impacts of HABs, both other human health and ecological.
Jessica Glowczewski, City of Akron, Ohio
(Comments appear here as they were submitted electronically)
If source water protection is going to be a focus on preventing algal blooms, someone needs to have
authority to enforce and follow up on potential pollution situations which bridge gaps between
townships, villages, cities, municipalities, counties and states. It doesn't have to be a utility managing
the watershed they use, but there needs to be more cooperation and more enforcement and support
from other jurisdictions and stakeholders, as well as enforceable penalties for pollution events.
Kim Ward, California State Water Resources Control Board
(Comments appear here as they were submitted electronically)
Helpful overview of the changes in agricultural production methods which seem to play an important
role in changes observed in toxigenic blooms observed in the Great Lakes in recent decades:
https://dl.sciencesocieties.org/files/publications/crops-and-soils/keeping-farm-based-p-out-of-lake-
erie.pdf
"Source water protection" should include consideration of sediment/soil conditions in surface

waterbodies, e.g., the possible occurrence of microcystins/other cyanotoxins in biological soil crusts
(often found in arid environments) and as cyanotoxin-producing symbionts/endosymbionts in aquatic
plants (e.g. cyanolichens). http://apsjournals.apsnet.org/doi/pdfplus/10.1094/MPMI-22-6-0695;
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3942747/;
https://www.ncbi.nlm.nih.gov/pubmed/25752635;
http://www.sciencedirect.com/science/article/pii/S0048969712001349
A key information gap concerns potential aerosol exposures in raw & finished water, e.g. research by
CDC on microcystin aerosols, etc.:
Lyda Hakes, Alameda County Water District
(Submitted electronically)
72
1) More information is needed on the growth/death rates of harmful algal blooms and the half-life
of the cyanotoxins (both extracellular and intracellular). Also, more information about what
drives extracellular versus intracellular concentrations would be useful.
2) Consider regulating cyanotoxins (specifically microcystin) by grouping them as Total

Trihalomethanes (TTHMs) and haloacetic acids (HAA5) are. This would better allow for the use
of a quantitative instrument like Liquid Chromatography/Tandem Mass Spectrometry (LC/MS-
MS) so that PWSs and regulators know exactly what is in the water and at what quantities.
3) Investigate further the nexus between climate change and the increased presence and
persistence of HABs. Also, research on the relationship between drought conditions and HABs
and cyanotoxins.
Amy Little, California State Water Resources Control Board
1) In the event of a public notification event due to a health advisory exceedance, what, if any,
information should be communicated to hemodialysis centers.
2) There is literature to describe how toxins are distributed in cyanobacteria cells (intra- vs. extra-
cellular) but this is likely changing throughout a bloom. If a public water system was faced with
exceeding a health advisory, knowing how toxins are distributed during the exceedance would
likely provide valuable information to the utility in order to target treatment optimization.
3) We have utilized fluorometers (when keeping cells intact is a treatment strategy) to evaluate cell
lysis at different stages of treatment, UV254 instruments to evaluate jar tests, provided
systematic technical assistance during bloom onsets on a case-by-case basis (e.g. shift intake to
a lower position where the pH is lower, add acid to lower the pH, add a filter aid to improve
filter performance), and we would like to explore the value of bench top charge analyzer to
enhance coagulation/flocculation treatment performance.
4) We are very grateful for the steps the USEPA has taken thus far to provide comprehensive
guidance and recommendations; as compiling this information was likely a tremendous task. In
our opinion, expanding on satellite information to provide up-to-date information and/or
predict when the blooms are toxic to govern efficient monitoring strategies would be of
tremendous value.
5) Peter Moyle, a fisheries biologist in California, developed a ranking system to prioritize which
dams should modify operations to significantly improve fishery habitat. This was the
introduction of indexing ecological parameters as a means to rank where to target effort in
order to get the most effective results. Many have expanded on this approach in nutrient loaded
systems. In our opinion, developing a way to measure (and prioritize) the most effective efforts
would be valuable.
Don Jensen, City of Highland Park, Illinois
73
As manager of a water utility blessed with Lake Michigan as a source, algal toxins are relatively low on
my list of worries.
I am, however, concerned when the news accounts of toxic algal blooms in the Great Lakes alarm our
residents.
Watershed risk assessment tools including indicators of risk (water temperature, sunlight hours, nutrient
loading, dissolved organic carbon, total organic carbon etc.) and associated mapping products would be
quite helpful in allaying those fears.
Of course, they would be most useful to water system managers in watersheds with higher risk of such
blooms.
74
Darci L. Meese, WaterOne (Submitted electronically)
75
76
Office of Water
820F15003
June 2015
2015 Drinking Water Health Advisories for

Two Cyanobacterial Toxins
Summary (cyanobacterial toxins or “cyanotoxins”) that are
EPA has issued 10-Day Drinking Water Health harmful to the environment, animals and human
Advisories (HAs) for the cyanobacterial toxins health. Winds and water currents can transport
microcystins and cylindrospermopsin. cyanobacterial blooms within proximity to drinking
water intakes at treatment plants that, if not
EPA recommends HA levels at or below 0.3 removed during treatment, can cause odor, taste
micrograms per liter for microcystins and 0.7 and color problems in treated drinking water and
micrograms per liter for cylindrospermopsin in can be harmful to human health.
drinking water for children pre-school age and
younger (less than six years old). For school-age What is a health advisory?
children through adults, the recommended HA levels The Safe Drinking Water Act provides the authority
for drinking water are at or below 1.6 micrograms for EPA to publish health advisories for contaminants
per liter for microcystins and 3.0 micrograms per not subject to any national primary drinking water
liter for cylindrospermopsin. Young children are regulation. Health advisories describe non-
more susceptible than older children and adults as regulatory concentrations of drinking water
they consume more water relative to their body contaminants at or below which adverse health
weight. effects are not anticipated to occur over specific
exposure durations (e.g., one-day, 10-days, several
HAs are non-regulatory values that serve as informal
years, and a lifetime). They serve as informal
technical guidance to assist federal, state and local
technical guidance to assist federal, state and local
officials, and managers of public or community
officials, and managers of public or community
water systems to protect public health from
water systems by providing information on the
contaminants. EPA has also published health effects
health effects of and methods to sample and treat
support documents for the cyanobacterial toxins
cyanobacterial toxins in drinking water. HAs are not
microcystins and cylindrospermopsin. These
legally enforceable federal standards and are subject
documents contain the health effects basis for the
to change as new information becomes available.
development of HAs for the protection of human
health. In addition, EPA has published a health Why has EPA taken this action?
effects support document for anatoxin-a but There are no U.S. federal guidelines, water quality
concluded that there was not adequate information criteria, standards or regulations for cyanobacteria
to support a health advisory for this toxin. or cyanotoxins in drinking water under the Safe
Drinking Water Act or in surface waters under the
Background
Clean Water Act. However, EPA has listed
What are cyanobacterial toxins? cyanotoxins including microcystin-LR,
Cyanobacteria, common to freshwater and marine cylindrospermopsin, and anatoxin–a on the previous
ecosystems, can under certain conditions (high and current Contaminant Candidate Lists (CCL),
nutrient concentrations and high light intensity) which identify contaminants that may need
form scums or “blooms” at the surface of a water regulation under the Safe Drinking Water Act.
body. These blooms can produce toxic compounds
EPA found there are adequate health effects data to liver weight, liver serum enzymes, and lesions in the
develop HAs for microcystins and liver. The lowest observed adverse effect level
cylindrospermopsin, but found the data inadequate (LOAEL) based on these effects was 50 micrograms
to develop an HA for the cyanobacterial toxin per kilogram per day, a no observed adverse effect
anatoxin-a. level (NOAEL) was not identified. This dose was
selected as the basis for deriving a reference dose
How Can I Be Exposed to Cyanobacterial Toxins?
(RfD) for microcystins. A total uncertainty factor of
For the cyanotoxin HAs, drinking water is the 1000 (10 to account for differences between humans
primary source of exposure. Exposure may also and animals, 10 to account for variability in humans,
occur by ingestion of toxin contaminated food, 3 for extrapolation from a LOAEL, and 3 to address
including consumption of fish; by inhalation and database deficiencies) was applied to determine the
dermal contact during bathing or showering; and RfD for microcystins. These values were used along
during recreational activities. Effects due to these with body weight and drinking water intake for
other routes of exposure cannot be quantified at this infants and adults to derive the 10-Day HA values.
time, however, they are assumed to be less than The 10-day HA of 0.3 µg/L is considered protective of
from drinking water ingestion. non-carcinogenic adverse health effects for bottle-
What information was used to develop the fed infants and young children of pre-school age
health advisories for cyanobacterial toxins? over a ten-day exposure to microcystins in drinking
water. The 10-day HA of 1.6 µg/L is considered
EPA worked with Health Canada and conducted a
protective of non-carcinogenic adverse health
comprehensive search of the literature from January
effects for children of school age through adults over
2013 to May 2014. The HA includes information on
a 10-day exposure to microcystins in drinking water.
occurrence; environmental fate; mechanisms of
toxicity; acute, short term, subchronic and chronic The critical study supporting the cylindrospermopsin
toxicity and cancer in humans and animals; 10-day advisory was conducted by Humpage and
toxicokinetics; health effects and exposure. The HA Falconer (2002, 2003). This study is an 11-week
also includes information on methods for analysis study with cylindrospermopsin administered to male
and treatment techniques for removal in drinking mice by gavage. Effects observed included increases
water treatment plants. in kidney weight. The NOAEL from this study was 30
micrograms per kilogram per day and the LOAEL
Health Effects Information
based on kidney weight changes was 60 micrograms
Effects including gastroenteritis, and liver and kidney per kilogram per day. The NOAEL of 30 micrograms
damage have been reported in humans following per kilogram per day was selected as the basis for
short-term exposure to cyanotoxins in drinking the RfD. A total uncertainty factor of 300 (10 to
water. Recreational exposure to cyanobacterial account for differences between humans and
blooms has been reported to lead to allergic animals, 10 to account for variability in humans, and
reactions, including hay fever-like symptoms; skin 3 to address database deficiencies) was applied to
rashes; and gastrointestinal distress. Animal studies determine the RfD for cylindrospermopsin. These
have shown that long-term adverse effects from values were used along with body weight and
cyanotoxins include liver and kidney damage. drinking water intake for infants and adults to derive
However, more research is needed to quantify these the 10-Day HA values. The 10-day HA of 0.7 µg/L is
effects. considered protective of non-carcinogenic adverse
Critical Studies Used health effects for bottle-fed infants and young
The critical study supporting the microcystins 10-day children of pre-school age over a 10-day exposure to
HA was conducted by Heinze (1999). This study is a cylindrospermopsin in drinking water. The 10-day HA
28-day study in rats, whose drinking water contained of 3 µg/L is considered protective of non-
microcystin-LR. Effects observed included changes in carcinogenic adverse health effects for children of
school age through adults over a 10-day exposure to U.S. EPA (United States Environmental Protection
cylindrospermopsin in drinking water. Agency). 2015. Drinking Water Health Advisory for the
Cyanobacterial Toxin Microcystin. EPA 820R15100,
As the science on the health impacts of algal toxins Washington, DC; June, 2015. Available from:
continues to improve, EPA will track developments http://water.epa.gov/drink/standards/hascience.cfm
and update recommendations as appropriate.
U.S. EPA (United States Environmental Protection
Additional EPA support document to assist Agency). 2015. Drinking Water Health Advisory for the
states and utilities in managing Cyanobacterial Toxin Cylindrospermopsin. EPA
cyanobacterial toxins http://water.epa.gov/drink/standards/hascience.cfm
EPA has also published a cyanotoxin management U.S. EPA (United States Environmental Protection
document as a companion to the HAs. The Agency). 2015. Health Effects Support Document for the
document is designed to provide information and a Cyanobacterial Toxin Microcystins. EPA 820R15102,
framework that Public Water Systems and others Washington, DC; June, 2015. Available from:
can consider to inform their decisions on managing http://water.epa.gov/drink/standards/hascience.cfm
the risks from cyanotoxins to drinking water. It U.S. EPA (United States Environmental Protection
includes a potential stepwise approach these Agency). 2015. Health Effects Support Document for the
systems could use to inform their decisions on Cyanobacterial Toxin Cylindrospermopsin. EPA
whether and how to monitor and treat water, and 820R15103, Washington, DC; June, 2015. Available from:
communicate with stakeholders. http://water.epa.gov/drink/standards/hascience.cfm
Where can I find more information? U.S. EPA (United States Environmental Protection
Agency). 2015. Health Effects Support Document for the
To learn more about the HAs for microcystins and Cyanobacterial Toxin Anatoxin-a. EPA 820R15104,
cylindrospermopsin and to view the health effects Washington, DC; June, 2015. Available from:
support documents for these and anatoxin-a in http://water.epa.gov/drink/standards/hascience.cfm
drinking water, visit EPA’s Health Advisory webpage:
http://water.epa.gov/drink/standards/hascience.cf
m. To learn about additional strategies Public Water
Systems and others could consider in managing
cyanotoxins, visit EPA’s CyanoHABs website:
http://www2.epa.gov/nutrient-policy-
data/guidelines-and-recommendations
References
Heinze, R. 1999. Toxicity of the cyanobacterial toxin
microcystin-LR to rats after 28 days intake with the
drinking water. Environ. Toxicol. 14(1):57-60.
Humpage, A.R. and I.R. Falconer. 2002. Oral Toxicity of
Cylindrospermopsin: No Observed Adverse Effect Level
Determination in Male Swiss Albino Mice. The
Cooperative Research Centre for Water Quality and
Treatment, Salisbury, South Australia. Research Report
No. 13. (93 pages).
Humpage, A.R. and I.R. Falconer. 2003. Oral toxicity of the
cyanobacterial toxin cylindrospermopsin in male Swiss
albino mice: Determination of no observed adverse effect
level for deriving a drinking water guideline value.
Environ. Toxicol. 18(2):94-103.
United States Office of Water EPA- 820R15100
Environmental Mail Code 4304T June 2015
Protection Agency
Drinking Water Health Advisory

for the Cyanobacterial
Microcystin Toxins
Drinking Water Health Advisory
for the Cyanobacterial Microcystin Toxins
Prepared by:
U.S. Environmental Protection Agency

Office of Water (4304T)
Health and Ecological Criteria Division
Washington, DC 20460
EPA Document Number: 820R15100

Date: June 15, 2015
ACKNOWLEDGMENTS
This document was prepared by U.S. EPA Scientists Lesley V. D’Anglada, Dr.P.H. (lead) and
Jamie Strong, Ph.D. Health and Ecological Criteria Division, Office of Science and Technology,
Office of Water. EPA gratefully acknowledges the valuable contributions from Health Canada’s
Water and Air Quality Bureau, in developing the Analytical Methods and Treatment Technologies
information included in this document.
This Health Advisory was provided for review and comments were received from staff in the
following U.S. EPA Program Offices:
U.S. EPA Office of Ground Water and Drinking Water
U.S. EPA Office of Science and Technology
U.S. EPA Office of Research and Development
U.S. EPA Office of Children’s Health Protection
U.S. EPA Office of General Counsel
This Health Advisory was provided for review and comments were received from the following
other federal and health agencies:
Health Canada
U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention
Drinking Water Health Advisory for Microcystins-June 2015 i

TABLE OF CONTENTS
ACKNOWLEDGMENTS .................................................................................................................I
TABLE OF CONTENTS ................................................................................................................. II
LIST OF TABLES ..........................................................................................................................IV
LIST OF FIGURES ........................................................................................................................IV
ABBREVIATIONS AND ACRONYMS ........................................................................................ V
EXECUTIVE SUMMARY .............................................................................................................. 1
1.0 INTRODUCTION AND BACKGROUND ......................................................................... 3
1.1 Current Criteria, Guidance and Standards ........................................................................ 3
2.0 PROBLEM FORMULATION.............................................................................................. 6
2.1 Cyanobacteria and Production of Microcystins ................................................................ 6
2.2 Physical and Chemical Properties ..................................................................................... 6
2.3 Sources and Occurrence .................................................................................................... 9
2.3.1 Occurrence in Surface Water ...................................................................................... 10
2.3.2 Occurrence in Drinking Water .................................................................................... 13
2.4 Environmental Fate ......................................................................................................... 14
2.4.1 Persistence................................................................................................................... 14
2.4.2 Mobility....................................................................................................................... 15
2.5 Nature of the Stressor-Characteristics of the Microcystin Toxins .................................. 15
2.5.1 Toxicokinetics ............................................................................................................. 15
2.5.2 Noncancer Health Effects Data ................................................................................... 16
2.5.2.1 Human Studies .................................................................................................... 16
2.5.2.2 Animal Studies .................................................................................................... 17
2.5.3 Mode of Action for Noncancer Health Effects ........................................................... 18
2.5.4 Carcinogenicity Data .................................................................................................. 18
2.6 Conceptual Model for Microcystins ............................................................................... 19
2.6.1 Conceptual Model Diagram ........................................................................................ 20
2.6.2 Factors Considered in the Conceptual Model for Microcystins ................................. 22
2.7 Analysis Plan .................................................................................................................. 23
3.0 HEALTH EFFECTS ASSESSMENT ................................................................................ 26
3.1 Dose-Response ................................................................................................................ 26
3.1.1 Study Selection ........................................................................................................... 26
3.1.2 Endpoint Selection ...................................................................................................... 28
3.2 Ten-day Health Advisory................................................................................................ 28
3.2.1 Bottle-fed Infants and Young Children of Pre-school Age ........................................ 28
3.2.2 School-age Children through Adults .......................................................................... 29
3.2.3 Uncertainty Factor Application................................................................................... 29
4.0 RISK CHARACTERIZATION .......................................................................................... 31
4.1 Use of microcystin-LR as a surrogate for total microcystins ......................................... 31
Drinking Water Health Advisory for Microcystins-June 2015 ii

4.2 Consideration of Study Duration .................................................................................... 31
4.3 Consideration of Reproductive Effects as Endpoint ....................................................... 32
4.4 Allometric Scaling Approach ......................................................................................... 33
4.5 Benchmark Dose (BMD) Modeling Analysis................................................................. 33
4.6 Carcinogenicity Evaluation............................................................................................. 33
4.7 Uncertainty and Variability............................................................................................. 34
4.8 Susceptibility................................................................................................................... 35
4.9 Distribution of Body Weight and Drinking Water Intake by Age .................................. 35
4.10 Distribution of Potential Health Advisory Values by Age ............................................. 36
5.0 ANALYTICAL METHODS .............................................................................................. 38
6.0 TREATMENT TECHNOLOGIES ..................................................................................... 41
6.1 Management and Mitigation of Cyanobacterial Blooms in Source Water ..................... 41
6.2 Drinking Water Treatment .............................................................................................. 42
6.2.1 Conventional Treatment for Microcystins .................................................................. 43
6.2.2 Adsorption................................................................................................................... 44
6.2.3 Chemical Oxidation .................................................................................................... 45
6.2.4 Other Filtration Technologies ..................................................................................... 46
6.2.5 Combined Treatment Technologies ............................................................................ 47
6.3 Point-of-Use (POU) Drinking Water Treatment Units ................................................... 48
7.0 REFERENCES ................................................................................................................... 49
Drinking Water Health Advisory for Microcystins-June 2015 iii

LIST OF TABLES
Table 1-1. International Guideline Values for Microcystins ............................................................ 4

Table 1-2. State Guideline Values for Microcystins ......................................................................... 5
Table 2-1. Abbreviations for Microcystins (Yuan et al., 1999) ........................................................ 8
Table 2-2. Chemical and Physical Properties of Microcystin-LR .................................................... 9
Table 2-3. States surveyed as part of the 2007 National Lakes Assessment with water body
Microcystin concentrations above the WHO advisory guideline level for recreational water
of 10 µg/L (U.S. EPA, 2009) .................................................................................................. 11
Table 3-1. Liver Weights and Serum Enzyme Levels in Rats Ingesting Microcystin-LR in
Drinking Water (Heinze, 1999) .............................................................................................. 27
Table 3-2. Histological Evaluation of the Rat Livers after Ingesting Microcystin-LR in Drinking
Water (Heinze, 1999) .............................................................................................................. 27
LIST OF FIGURES
Figure 2-1. Structure of Microcystin (Kondo et al., 1992) ............................................................... 7

Figure 2-2. Structure of the amino acids Adda and Mdha (Harada et al., 1991) .............................. 8
Figure 2-3. Conceptual Model of Exposure Pathways to Microcystins in Drinking Water ........... 21
Figure 4-1. 90th Percentile Drinking Water Ingestion Rates by Age Group ................................... 35
Figure 4-2. Ten-day Health Advisories for Microcystins by Age Group ....................................... 37
Drinking Water Health Advisory for Microcystins-June 2015 iv

ABBREVIATIONS AND ACRONYMS
Γgt Γ-Glutamyltransferase
A Alanine
Adda 3-Amino-9-Methoxy-2,-6,-8-,Trimethyl-10-Phenyldeca-4,-6-Dienoic Acid
ALT Alanine Aminotransferase
ALP Alkaline Phosphatase
AST Aspartate Aminotransferase
AWWARF American Water Works Association Research Foundation
BMD Benchmark Dose
BMDL Benchmark Dose Level
BW Body Weight
Clo2 Chlorine Dioxide
CAS Chemical Abstracts Service
CASA Computer-Assisted Sperm Analysis
CCL Contaminant Candidate List
CWA Clean Water Act
DBP Disinfection By-Products
DL Detection Limit
EBCT Empty Bed Contact Time
ELISA Enzyme-Linked Immunosorbent Assay
EPA U.S. Environmental Protection Agency
FSH Follicle Stimulating Hormone
g Gram
GC/MS Gas Chromatograph/Mass Spectrometry
GAC Granular Activated Carbon
GLERL Great Lakes Environmental Research Laboratory
H2O2 Hydrogen Peroxide
HA Health Advisory
HAB Harmful Algal Bloom
HESD Health Effects Support Document
HPLC High-Performance Liquid Chromatography
IARC International Agency for Research on Cancer
i.p. Intraperitoneal
Kg Kilogram
L Leucine
LC/MS Liquid Chromatography/Mass Spectrometry
LDH Lactate Dehydrogenase
LD50 Lethal Dose to 50% of Organisms
LH Luteinizing Hormone
LOAEL Lowest-Observed-Adverse-Effect Level
MC-LA Microcystin-LA
MC-LR Microcystin-LR
MC-RR Microcystin-RR
MC-YR Microcystin-YR
MC-YM Microcystin-YM
Drinking Water Health Advisory for Microcystins-June 2015 v

Mdha Methyldehydroalanine
MERHAB-LGL Monitoring and Event Response to Harmful Algal Blooms in the Lower
Great Lakes
µg Microgram
µm Micromole
LOQ Level of Quantification
Mdls Method Detection Limit
mg Milligram
ml Milliliter
MMPB 2-methyl-3-methoxy-4-phenylbutyric acid
MOA Mode of Action
MF Microfiltration
MWCO Molecular Weight Cut-Off
NDEA N-Nitrosodiethylamine
NF Nanofiltration
NLA National Lakes Assessment
NOAEL No-Observed-Adverse-Effect Level
NOD Nodularin
NOM Natural Organic Material
OATp Organic Acid Transporter Polypeptides
PAS Periodic Acid-Schiff
PBS Phosphate-Buffered Saline
PDA Photodiode Array Detector
P-GST glutathione S-transferase placental form-positive
POD Point of Departure
POU Point-of-Use
PP2 Protein Phosphatase 2A
PP1 Protein Phosphatase 1
PPIA Protein Phosphatase Inhibition Assays
RfD Reference Dose
RO Reverse Osmosis
ROS Reactive Oxygen Species
SPE Solid-Phase Extraction
TEF Toxicity Equivalency Factors
TOC Total Organic Carbon
TOXLINE Toxicology Literature Online
TUNEL Terminal Deoxynucleotidyl Transferase-Mediated dUTP-Biotin Nick End-
Labeling Assay
UF Uncertainty Factor
UF Ultrafiltration
UV Ultraviolet
Drinking Water Health Advisory for Microcystins-June 2015 vi

EXECUTIVE SUMMARY
Microcystins are toxins produced by a number of cyanobacteria species, including

members of Microcystis, Anabaena, Nodularia, Nostoc, Oscillatoria, Fischerella, Planktothrix,
and Gloeotrichia. Approximately 100 microcystin congeners exist, which vary based on amino
acid composition. Microcystin-LR is one of the most potent congeners and the majority of
toxicological data on the effects of microcystins are available for this congener.
Many environmental factors such as the ratio of nitrogen to phosphorus, temperature,

organic matter availability, light attenuation and pH play an important role in the development of
microcystin blooms, both in fresh and marine water systems and could encourage toxin
production. Microcystins are water soluble and tend to remain contained within the cyanobacterial
cell (intracellular), until the cell breaks and they are released into the water (extracellular).
This Health Advisory (HA) for microcystins is focused on drinking water as the primary
source of exposure. Exposure to cyanobacteria and their toxins may also occur by ingestion of
toxin-contaminated food, including consumption of fish, and by inhalation and dermal contact
during bathing or showering and during recreational activities in waterbodies with the toxins.
While these types of exposures cannot be quantified at this time, they are assumed to contribute
less to the total cyanotoxin exposures than ingestion of drinking water. Due to the seasonality of
cyanobacterial blooms, exposures are not expected to be chronic.
Limited data in humans and animals demonstrate the absorption of microcystins from the
intestinal tract and distribution to the liver, brain, and other tissues. Elimination from the body
requires facilitated transport using receptors belonging to the Organic Acid Transporter
polypeptide (OATp) family. Data for humans and other mammals suggest that the liver is a
primary site for binding these proteins (i.e., increased liver weight in laboratory animals and
increased serum enzymes in laboratory animals and humans). Once inside the cell, these toxins
covalently bind to cytosolic proteins (PP1 and PP2) resulting in their retention in the liver.
Limited data are available on the metabolism of microcystins, but most of the studies indicate that
microcystins can be conjugated with glutathione and cysteine to increase their solubility and
facilitate excretion.
The main source of human health effects data for microcystins is from acute recreational
exposure to cyanobacterial blooms. Symptoms include headache, sore throat, vomiting and
nausea, stomach pain, dry cough, diarrhea, blistering around the mouth, and pneumonia.
However, human data on the oral toxicity of microcystins are limited and confounded by:
potential co-exposure to other contaminants; a lack of quantitative information; and other
confounding factors. Reports of human intravenous exposure to dialysate prepared with
microcystin-contaminated water indicated acute liver failure and death in a large number of the
exposed patients.
Studies in laboratory animals demonstrate liver, kidney, and reproductive effects

following short-term and subchronic oral exposures to microcystin-LR. Studies evaluating the
chronic toxicity of microcystins have not shown clinical signs of toxicity and are limited by study
design and by the lack of quantitative data.
Drinking Water Health Advisory for Microcystins-June 2015 1

The U.S. Environmental Protection Agency (EPA) identified a study by Heinze (1999)
conducted on rats as the critical study used in the derivation of the reference dose (RfD) for
microcystins. The critical effects identified in the study are increased liver weight, slight to
moderate liver lesions with hemorrhages, and increased enzyme levels as a result of exposure to
microcystin-LR. The lowest-observed-adverse-effect level (LOAEL) was determined to be 50
μg/kg/day, based on these effects. The drinking water route of exposure matches potential
drinking water exposure scenarios in humans. The total uncertainty factor (UF) applied to the
LOAEL was 1000. This was based on a UF of 10 for intraspecies variability, a UF of 10 for
interspecies variability, a UF of 3 (10½) for extrapolation from a LOAEL to no-observed-adverse-
effect level (NOAEL), and a UF of 3 (10½) to account for deficiencies in the database. EPA is
using microcystin-LR as a surrogate for other microcystin congeners. Therefore, the HA based on
this critical study applies to total microcystins.
EPA is issuing a Ten-day HA for microcystins based on the Heinze (1999) short-term, 28-
day study. Studies of a duration of 7 to 30 days are typically used to derive Ten-day HAs. The HA
is consistent with this duration and appropriately matches human exposure scenarios for
microcystins in drinking water. Cyanobacterial blooms are usually seasonal, typically occurring
from May through October. Microcystins typically have a half-life of 4 days to 14 days in surface
waters, (depending on the degree of sunlight, natural organic matter, and the presence of bacteria)
and can be diluted via transport. In addition, concentrations in finished drinking water can be
reduced by drinking water treatment and management measures.
The Ten-day HA value for bottle-fed infants and young children of pre-school age is 0.3
µg/L and for school-age children through adults is 1.6 µg/L for microcystins. The two advisory
values use the same toxicity data (RfD) and represent differences in drinking water intake and
body weight for different life stages. The first advisory value is based on the summation of the
time-weighted drinking water intake/body weight ratios for birth to <12 months of age. The
second advisory value is based on the mean body weight and 90th percentile drinking water
consumption rates for adults age 21 and over (U.S. EPA’s Exposure Factors Handbook (2011a)),
which is similar to that of school-aged children. Populations such as pregnant women and nursing
mothers, the elderly, and immune-compromised individuals or those receiving dialysis treatment
may be more susceptible than the general population to the health effects of microcystins. As a
precautionary measure, individuals that fall into these susceptible groups may want to consider
following the recommendations for children pre-school age and younger. This HA is not a
regulation, it is not legally enforceable, and it does not confer legal rights or impose legal
obligations on any party.
Applying the U.S. EPA (2005) Guidelines for Carcinogen Risk Assessment, there is
inadequate information to assess carcinogenic potential of microcystins. The few available
epidemiological studies are limited by their study design, poor measures of exposure, potential
co-exposure to other contaminants, and the lack of control for confounding factors. No long term
animal studies were available to evaluate dose-response for the tumorigenicity of microcystins
following lifetime exposures. Other studies evaluating the tumor promotion potential of
microcystin found an increase in the number and/or size of GST-P positive foci observed. In two
promotion studies, microcystin-LR alone showed no initiating activity.

1.0 INTRODUCTION AND BACKGROUND
EPA developed the non-regulatory Health Advisory (HA) Program in 1978 to provide
information for public health officials or other interested groups on pollutants associated with
short-term contamination incidents or spills for contaminants that can affect drinking water
quality, but are not regulated under the Safe Drinking Water Act (SDWA). At present, EPA lists
HAs for 213 contaminants (http://water.epa.gov/drink/standards/hascience.cfm).
HAs identify the concentration of a contaminant in drinking water at which adverse health
effects are not anticipated to occur over specific exposure durations (e.g., one-day, ten-days, and a
lifetime). HAs serve as informal technical guidance to assist Federal, State and local officials, and
managers of public or community water systems in protecting public health when emergency
spills or contamination situations occur. An HA provides information on the environmental
properties, health effects, analytical methodology, and treatment technologies for removal of
drinking water contaminants.
The Health Effects Support Document for Microcystins (U.S.EPA, 2015a) is the peer-
reviewed, effects assessment that supports this HA. This document is available at
http://www2.epa.gov/nutrient-policy-data/health-and-ecological-effects. The HAs are not legally
enforceable Federal standards and are subject to change as new information becomes available.
The structure of this Health Advisory is consistent with EPA’s Framework for Human Health
Risk Assessment to Inform Decision Making (U.S.EPA, 2014).
EPA is releasing the Recommendations for Public Water Systems to Manage Cyanotoxins
in Drinking Water (U.S. EPA, 2015b) as a companion to the HAs for microcystins and
cylindrospermopsin. The document is intended to assist public drinking water systems (PWSs)
that choose to develop system-specific plans for evaluating their source waters for vulnerability to
contamination by microcystins and cylindrospermopsin. It is designed to provide information and
a framework that PWSs and others as appropriate may consider to inform their decisions on
managing the risks from cyanotoxins in drinking water.
1.1 Current Criteria, Guidance and Standards
Currently there are no U.S. federal water quality criteria, or regulations for cyanobacteria
or cyanotoxins in drinking water under the SDWA or in ambient waters under the Clean Water
Act (CWA). The Safe Drinking Water Act (SDWA), as amended in 1996, requires the EPA to
publish a list of unregulated contaminants every five years that are not subject to any proposed or
promulgated national primary drinking water regulations, which are known or anticipated to occur
in public water systems, and which may require regulation. This list is known as the Contaminant
Candidate List (CCL). The EPA’s Office of Water included cyanobacteria and cyanotoxins on the
first and second CCL (CCL 1, 1998; CCL 2, 2005). EPA included cyanotoxins, including
anatoxin-a, cylindrospermopsin, and microcystin-LR, on CCL 3 (2009) and the draft CCL 4
(April 2015 for consideration).

SDWA requires the Agency to make regulatory determinations on at least five CCL
contaminants every five years. When making a positive regulatory determination, EPA
determines whether a contaminant meets three criteria:
• The contaminant may have an adverse effect on the health of persons,

• The contaminant is known to occur or there is substantial likelihood the contaminant
will occur in public water systems with a frequency and at levels of concern, and
• In the sole judgment of the Administrator, regulating the contaminant presents a
meaningful opportunity for health risk reductions.
To make these determinations, the Agency uses data to analyze occurrence (prevalence
and magnitude) and health effects. EPA continues gathering this information to inform future
regulatory determinations for cyanotoxins under the SDWA. The SDWA also provides the
authority for EPA to publish non-regulatory HAs or take other appropriate actions for
contaminants not subject to any national primary drinking water regulation. EPA is providing this
HA and the HA for cylindrospermopsin to assist State and local officials in evaluating risks from
these contaminants in drinking water.
Internationally, eighteen countries and three U.S. states have developed drinking water
guidelines for microcystins, as shown in Table 1.1 and Table 1.2, respectively, based on lifetime
exposures.
Table 1-1. International Guideline Values for Microcystins
Country Guideline Value Source

Brazil, China, Czech Republic, Based on the World Health
1.0 μg/L
Denmark, Finland, France, Germany, Organization (WHO) Provisional
microcystin-LR
Italy, Japan, Korea, Netherlands, Guideline Value of 1ug/L for
Norway, New Zealand, Poland, drinking water
South Africa, and Spain (WHO, 1999; 2003)
1.3 μg/L microcystin- Australian Drinking Water
Australia LR (toxicity Guidelines 6
equivalents) (NHMRC, NRMMC, 2011)
Guidelines for Canadian
Drinking Water Quality:
1.5 μg/L Supporting Documentation
Canada
microcystin-LR Cyanobacterial Toxins-
Microcystin-LR
(Health Canada, 2002)

Table 1-2. State Guideline Values for Microcystins
State Guideline Value Source

Minnesota 0.04 µg/L Microcystin-LR Minnesota Department of Health (MDH, 2012)
Public Water System Harmful Algal Bloom
Ohio 1 µg/L Microcystin
Response Strategy (Ohio EPA, 2014)
Public Health Advisory Guidelines, Harmful Algae
Oregon 1 µg/L Microcystin-LR
Blooms in Freshwater Bodies. (OHA, 2015)
For drinking water, the provisional WHO Guideline value for microcystin-LR of 1 μg/L
(or the underlying Tolerable Daily Intake (TDI) of 0.04 μg/kg) has been widely used as the basis
for national standards or guideline values (WHO, 1999, 2003). Following the release of the WHO
provisional guideline, drinking-water standards or national guideline values were adopted in 16
countries. Australia and Canada have used the TDI, but have adapted other factors in the
calculation to reflect their national circumstances (e.g. body weight or amounts of water
consumed), thus reaching somewhat higher guidance values or standards (Chorus, 2012). A few
countries have issued guideline values specifically for microcystin-LR while others use
microcystin-LR as a surrogate for all microcystin congeners (i.e. toxicity equivalents). Values are
similar across all countries, ranging between 1.0 and 1.5 μg/L based on lifetime exposures.

2.0 PROBLEM FORMULATION
The development of the HA begins with problem formulation, which provides a strategic
framework by focusing on the most relevant cyanotoxin properties and endpoints identified in the
Health Effects Support Document for Microcystins (U.S. EPA, 2015a).
2.1 Cyanobacteria and Production of Microcystins
Cyanobacteria, formerly known as blue-green algae (Cyanophyceae), are a group of

bacteria with chlorophyll-a capable of photosynthesis (light and dark phases) (Castenholz and
Waterbury, 1989). Most cyanobacteria are aerobic photoautotrophs, requiring only water, carbon
dioxide, inorganic nutrients and light for survival, while others have heterotrophic properties and
can survive long periods in complete darkness (Fay, 1965). Some species are capable of nitrogen
fixation (diazotrophs) (Duy et al., 2000), producing inorganic nitrogen compounds for the
synthesis of nucleic acids and proteins. Cyanobacteria can form symbiotic associations with
animals and plants, such as fungi, bryophytes, pteriodophytes, gymnosperms and angiosperms
(Rai, 1990), supporting their growth and reproduction (Sarma, 2013; Hudnell, 2008; Hudnell,
2010).
Under the right conditions of pH, nutrient availability, light, and temperature,
cyanobacteria can reproduce quickly forming a bloom. Although studies of the impact of
environmental factors on cyanotoxin production are ongoing, nutrient (N, P and trace metals)
supply rates, light, temperature, oxidative stressors, interactions with other biota (viruses, bacteria
and animal grazers), and most likely, the combined effects of these factors are all involved (Paerl
and Otten 2013a; 2013b). Fulvic and humic acids reportedly encourage cyanobacteria growth
(Kosakowska et al., 2007).
Microcystins are produced by several cyanobacterial species, including Anabaena,

Fischerella, Gloeotrichia, Nodularia, Nostoc, Oscillatoria, members of Microcystis, and
Planktothrix (Duy et al., 2000; Codd et al., 2005; Stewart et al., 2006a; Carey et al., 2012).
2.2 Physical and Chemical Properties
The cyclic peptides include around 100 congeners of microcystins. Table 2-1 lists only the
most common microcystins congeners. Figure 2-1 provides the structure of microcystin where X
and Y represent variable amino acids. Although substitutions mostly occur in positions X and Y,
other modifications have been reported for all of the amino acids (Puddick et al., 2015). The
amino acids are joined end-to-end and then head to tail to form cyclic compounds that are
comparatively large (molecular weights ranging from ~800 to 1,100 g/mole).

Figure 2-1. Structure of Microcystin (Kondo et al., 1992)
Microcystin congeners vary based on their amino acid composition and through
methylation or demethylation at selected sites within the cyclic peptide (Duy et al., 2000). The
variations in composition and methylation account for the large number of toxin congeners. The
microcystins are named based on their variable amino acids, although they have had many other
names (Carmichael et al., 1988). For example, microcystin-LR, the most common congener,
contains leucine (L) and arginine (R) (Carmichael, 1992). The letters used to identify the variable
amino acids are the standard single letter abbreviations for the amino acids found in proteins. The
variable amino acids are usually the L-amino acids as found in proteins. In this HA, the term
microcystin may be followed by the abbreviations for the variable amino acids. For example,
microcystin-LR is for the microcystin with leucine in the X position of Figure 2-1 and arginine in
the Y position. Most research has concentrated on microcystin-LR, with lesser amounts of data
available for the other amino acid combinations. For the purpose of this HA, microcystin-LR is
used as the surrogate for total microcystins.
Structurally, the microcystins are monocyclic heptapeptides that contain seven amino
acids: two variable L-amino acids, three common D-amino acids or their derivatives, and two
novel D-amino acids. These two D-amino acids are: 3S-amino-9S-methoxy-2,6,8S,-trimethyl-10-
phenyldeca-4,6-dienoic acid (Adda) and methyldehydroalanine (Mdha). Adda is characteristic of
all toxic microcystin structural congeners and is essential for their biological activity (Rao et al.,
2002; Funari and Testai, 2008). Mdha plays an important role in the ability of the microcystins to
inhibit protein phosphatases. Figure 2-2 illustrates the structures of these two unique amino acid
microcystin components.
Microcystins are water soluble. In aquatic environments, the cyclic peptides tend to
remain contained within the cyanobacterial cell and are released in substantial amounts only upon
cell lysis. The microcystins are most frequently found in cyanobacterial blooms in fresh and
brackish waters (WHO, 1999). Table 2-2 provides chemical and physical properties of
microcystin-LR.

Table 2-1. Abbreviations for Microcystins (Yuan et al., 1999)
Microcystin Congeners Amino Acid in X Amino Acid in Y

Microcystin-LR Leucine Arginine
Microcystin-RR Arginine Arginine
Microcystin-YR Tyrosine Arginine
Microcystin-LA Leucine Alanine
Microcystin-LY Leucine Tyrosine
Microcystin-LF Leucine Phenylalanine
Microcystin-LW Leucine Tryptophan
Figure 2-2. Structure of the amino acids Adda and Mdha (Harada et al., 1991).
Adda Mdha

Table 2-2. Chemical and Physical Properties of Microcystin-LR
Property Microcystin-LR
Chemical Abstracts Registry (CAS) # 101043-37-2

Chemical Formula C49H74N10O12
Molecular Weight 995.17 g/mole
Color/Physical State Solid
Boiling Point N/A
Melting Point N/A
Density 1.29 g/cm3
Vapor Pressure at 25°C N/A
Henry’s Law Constant N/A
Kow N/A
Koc N/A
Solubility in Water Highly
Other Solvents Ethanol and methanol
Sources: Chemical Book, 2012; TOXLINE, 2012
2.3 Sources and Occurrence
Cyanotoxin production is strongly influenced by the environmental conditions that

promote growth of particular cyanobacterial species and strain. Nutrient concentrations, light
intensity, water turbidity, temperature, competing bacteria and phytoplankton, pH, turbulence,
and salinity are all factors that affect cyanobacterial growth and change in cyanobacteria
population dynamics. Although environmental conditions affect the formation of blooms, the
numbers of cyanobacteria and toxin concentrations produced are not always closely related.
Cyanotoxin concentrations depend on the dominance and diversity of strains within the bloom
along with environmental and ecosystem influences on bloom dynamics (Hitzfeld et al., 2000;
Chorus et al., 2000; WHO, 1999). Extracellular microcystins (either dissolved in water or bound
to other materials) typically make up less than 30% of the total microcystin concentration in
source water (Graham et al., 2010). Most of the toxin is intracellular, and released into the water
when the cells rupture or die. Both intracellular and extracellular microcystins may also be
present in treated water, depending on the type of treatment processes in place.

2.3.1 Occurrence in Surface Water
Microcystins are the most common cyanotoxin found worldwide and have been reported
in surface waters in most of the U.S. and Europe (Funari and Testai, 2008). Dry-weight
concentrations of microcystins in surface freshwater cyanobacterial blooms or surface freshwater
samples reported worldwide between 1985 and 1996 ranged from 1 to 7,300 µg/g. Water
concentrations of extracellular plus intracellular microcystins ranged from 0.04 to 25,000 µg/L.
The concentration of extracellular microcystins ranged from 0.02 to a high of 1,800 µg/L reported
following treatment of a large cyanobacteria bloom with algaecide (WHO, 1999) and the U.S.
Geological Survey (USGS) reported a concentration of 150,000 µg/L total microcystins, in a lake
in Kansas (Graham et al., 2012).
According to a survey conducted in Florida in 1999 between the months of June and
November, the most frequently observed cyanobacteria were Microcystis (43.1%),
Cylindrospermopsis (39.5%), and Anabaena spp (28.7%) (Burns, 2008). Of 167 surface water
samples taken from 75 waterbodies, 88 samples were positive for cyanotoxins. Microcystin was
the most commonly found cyanotoxin in water samples collected, occurring in 87 water samples.
In 2002, the Monitoring and Event Response to Harmful Algal Blooms in the Lower Great
Lakes (MERHAB-LGL) project evaluated the occurrence and distribution of cyanobacterial
toxins in the lower Great Lakes region (Boyer, 2007). Analysis for total microcystins was
performed using Protein Phosphatase Inhibition Assay (PPIA). Microcystins were detected in at
least 65% of the samples, mostly in Lake Erie, Lake Ontario, and Lake Champlain. The National
Oceanic and Atmospheric Administration (NOAA) Center of Excellence for Great Lakes and
Human Health (CEGLHH) continues to monitor the Great Lakes and regularly samples algal
blooms for microcystin in response to bloom events.
A 2004 study of the Great Lakes found high levels of cyanobacteria during the month of
August (Makarewicz et al., 2006). Microcystin-LR was analyzed by PPIA (limit of detection of
0.003 µg/L) and was detected at levels of 0.084 μg/L in the nearshore and 0.076 μg/L in the bays
and rivers. This study reported higher levels of microcystin-LR (1.6 to 10.7μg/L) in smaller lakes
in the Lake Ontario watershed.
In 2006, the USGS conducted a study of 23 lakes in the Midwestern U.S. in which
cyanobacterial blooms were sampled to determine the co-occurrence of toxins in cyanobacterial
blooms (Graham et al., 2010). This study reported that microcystins were detected in 91% of the
lakes sampled. Mixtures of all the microcystin congeners measured (LA, LF, LR, LW, LY, RR,
and YR) were common and all the congeners were present in association with the blooms.
Microcystin--LR and –RR were the dominant congeners detected with mean concentrations of
104 and 910 µg/L, respectively.
EPA’s National Aquatic Resource Surveys (NARS) generate national estimates of

pollutant occurrence every 5 years. In 2007, the National Lakes Assessment (NLA) conducted the
first-ever national probability-based survey of the nation's lakes, ponds and reservoirs (U.S.EPA,
2009). This baseline study of the condition of the nation’s lakes provided estimates of the
condition of natural and man-made freshwater lakes, ponds, and reservoirs greater than 10 acres
and at least one meter deep. A total of 1,028 lakes were sampled in the NLA during the summer

of 2007. The NLA measured microcystins using Enzyme Linked Immunosorbent Assays (ELISA)
with a detection limit of 0.1 µg/L as well as cyanobacterial cell counts and chlorophyll-a
concentrations, which were indicators of the presence of cyanobacterial toxins. Samples were
collected in open water at mid-lake. Due to the design of the survey, no samples were taken
nearshore or in other areas where scums were present.
A total of 48 states were sampled in the NLA, and states with lakes reporting microcystins
levels above the WHO’s moderate risk 1 threshold in recreational water (>10 µg/L) are shown in
Table 2-3. Microcystins were present in 30% of the lakes sampled nationally, with sample
concentrations that ranged from the limit of detection (0.1 µg/L) to 225 µg/L. Two states (North
Dakota and Nebraska), had 9% of samples above 10 µg/L. Other states including Iowa, Texas,
South Dakota, and Utah also had samples that exceeded 10 µg/L. Several samples in North
Dakota, Nebraska, and Ohio exceeded the WHO high risk threshold value for recreational waters
of 20 µg/L (192 and 225 µg/L, respectively). EPA completed a second survey of lakes in 2012,
but data have not yet been published.
Microcystins have been detected in most of the states of the U.S., and over the years many
studies have been done to determine their occurrence in surface water. USGS, for example, did a
study in the Upper Klamath Lake in Oregon in 2007 and detected total microcystin concentrations
between 1 µg/L and 17 µg/L (VanderKooi et al., 2010). USGS also monitored Lake Houston in
Texas from 2006 to 2008, and found microcystins in 16% of samples with concentrations less
than or equal to 0.2 µg/L (Beussink and Graham, 2011). In 2011, USGS conducted a study on the
upstream reservoirs of the Kansas River, a primary source of drinking water for residents in
northeastern Kansas, to characterize the transport of cyanobacteria and associated compounds
(Graham et al., 2012). Concentrations of total microcystin were low in the majority of the
tributaries with the exception of Milford Lake, which had higher total microcystin concentrations,
some exceeding the Kansas recreational guidance level of 20 µg/L. Upstream from Milford Lake,
a cyanobacterial bloom was observed with a total microcystin concentration of 150,000 µg/L.
When sampled a week later, total microcystin concentrations were less than 1 µg/L. The study
authors indicated that this may be due to dispersion of microcystins through the water column or
to other areas, or by degradation of microcystins via abiotic and biological processes. Samples
taken during the same time from outflow waters contained total microcystin concentrations of 6.2
µg/L.
In 2005, Washington State Department of Ecology developed the Ecology Freshwater

Algae Program to focus on the monitoring and management of cyanobacteria in Washington
lakes, ponds, and streams (WSDE, 2012). The data collected have been summarized in a series of
reports for the Washington State Legislature (Hamel, 2009, 2012). Microcystin levels ranged
from the detection limit (0.05 µg/L) to 4,620 µg/L in 2008, 18,700 µg/L in 2009, 853 µg/L in
2010, and 26,400 µg/L in 2011.
1
The WHO established guideline values for recreational exposure to cyanobacteria using a three-tier approach: low
risk (<20,000 cyanobacterial cells/ml corresponding to <10 µg/L of MC-LR); moderate risk (20,000-100,000
cyanobacterial cells/ml corresponding to 10-20 µg/L of MC-LR); and high risk (>100,000 cyanobacterial cells/ml
corresponding to >20 µg/L for MC-LR).

Table 2-3. States Surveyed as Part of the 2007 National Lakes Assessment with Water Body
Microcystins Concentrations above the WHO Advisory Guideline Level for Recreational
Water of 10 µg/L (U.S. EPA, 2009)
Number Percentage of Samples Maximum Detection

State of Sites with Detection of of
Sampled Microcystins >10 µg/L Microcystins
North Dakota 38 9.1% 192 µg/L
Nebraska 42 9.1% 225 µg/L
South Dakota 40 4.9% 33 µg/L
Ohio 21 4.5% 78 µg/L*
Iowa 20 4.5% 38 µg/L*
Utah 26 3.6% 15 µg/L*
Texas 51 1.8% 28 µg/L *
* Single Sample
Other surveys and studies have been conducted to determine the occurrence of
microcystin in lakes in the United States. A survey conducted during the spring and summer of
1999 and 2000 in more than 50 lakes in New Hampshire found measureable microcystin
concentrations in all samples (Haney and Ikawa, 2000). Microcystins were analyzed by ELISA
and were found in all of the lakes sampled with a mean concentration of 0.1 µg/L. In 2005 and
2006, a study conducted in New York, including Lake Ontario, found variability in microcystin-
LR concentrations within the Lake Ontario ecosystem (Makarewicz et al., 2009). Of the samples
taken in Lake Ontario coastal waters, only 0.3% of the samples exceeded the WHO provisional
guideline value for drinking water of 1 µg/L. However, 20.4% of the samples taken at upland
lakes and ponds within the Lake Ontario watershed, some of them sources of drinking water,
exceeded 1 µg/L. During 2008 and 2009, a study was conducted in Kabetogama Lake, Minnesota
which detected microcystin concentrations in association with algal blooms (Christensen et al.,
2011). Microcystin concentrations were detected in 78% of bloom samples. Of these, 50% were
above 1 μg/L, and two samples were above the high risk WHO recreational level of 20 μg/L.
A study from 2002 evaluated water quality including chlorophyll-a concentration,

cyanobacterial assemblages, and microcystin concentrations in 11 potable water supply reservoirs
within the North Carolina Piedmont during dry summer growing seasons (Touchette et al., 2007).
Microcystins concentrations were assessed using ELISA. The study found that cyanobacteria
were the dominant phytoplankton community, averaging 65-95% of the total phytoplankton cells.
Although microcystin concentrations were detected in nearly all source water samples,
concentrations were <0.8 µg/L.
Since 2007, Ohio EPA (OHEPA, 2012) has been monitoring inland lakes for cyanotoxins.
Of the 19 lakes in Ohio sampled during the NLA, 36% had detectable levels of microcystins. In

2010, OHEPA sampled Grand Lake St. Marys for anatoxin-a, cylindrospermopsin, microcystins,
and saxitoxin. Toxin levels ranged from below the detection limit (<0.15 µg/L) to more than
2,000 µg/L for microcystins. Follow-up samples taken in 2011 for microcystins indicated
concentrations exceeding 50 µg/L in August. During the same month, sampling in Lake Erie
found microcystins levels exceeding 100 µg/L.
In 2008, NOAA began monitoring for cyanobacterial blooms in Lake Erie using high
temporal resolution satellite imagery. Between 2008 and 2010, Microcystis cyanobacterial blooms
were detected associated with water temperatures above 18°C (Wynne et al., 2013). Using the
Great Lakes Coastal Forecast System (GLCFS), forecasts of bloom transport are created to
estimate the trajectory of the bloom, and these are distributed as bulletins to local managers,
health departments, researchers and other stakeholders. To evaluate bloom toxicity, the Great
Lakes Environmental Research Laboratory (GLERL) collected samples at six stations each week
for 24 weeks, measuring toxin concentrations as well as chlorophyll biomass and an additional 18
parameters (e.g., nutrients) to improve future forecasts of these blooms. In 2014, particulate toxin
concentrations, collected from 1 meter depth, ranged from below detection to 36.7
μg/L. Particulate toxin concentrations peaked in August, 2014 at all sites, with the Maumee Bay
site yielding the highest toxin concentration of the entire sampling period. Dissolved toxin
concentrations were collected at each site from September until November when the field season
ended. During the final months of sampling (October-November), dissolved toxin concentrations
were detected with peak concentrations of 0.8 μg/L (mean: 0.28 +/- 0.2 μg/L) whereas particulate
toxin concentrations were below detection limits on many dates, indicating that a majority of the
toxins (mean: 72% +/- 37%) were in the dissolved pool as the bloom declined in intensity.
Concentrations of microcystins were detected during sampling in 2005 and 2006 in lakes
and ponds used as a source of drinking water within the Lake Ontario watershed (Makarewicz et
al., 2009). A microcystin-LR concentration of 5.07 µg/L was found in Conesus Lake, a source of
public water supply that provides drinking water to approximately 15,000 people. Microcystin-LR
was also detected at 10.716 µg/L in Silver Lake, a public drinking water supply for four
municipalities.
2.3.2 Occurrence in Drinking Water
The occurrence of cyanotoxins in drinking water depends on their levels in the raw source
water and the effectiveness of treatment methods for removing cyanobacteria and cyanotoxins
during the production of drinking water. Currently, there is no program in place to monitor for the
occurrence of cyanotoxins at surface-water treatment plants for drinking water in the U.S.
Therefore, data on the presence or absence of cyanotoxins in finished drinking water are limited.
The American Water Works Association Research Foundation (AWWARF) conducted a

study on the occurrence of cyanobacterial toxins in source and treated drinking waters from 24
public water systems in the United States and Canada in 1996-1998 (AWWARF, 2001). Of 677
samples tested, microcystin was found in 80% (539) of the waters sampled, including source and
treated waters. Only two samples of finished drinking water were above 1 μg/L. A survey
conducted in 2000 in Florida (Burns, 2008) reported that microcystins were the most commonly

found toxin in pre- and post-treated drinking water. Finished water concentrations ranged from
below detection levels to 12.5 µg/L.
During the summer of 2003, a survey was conducted to test for microcystins in 33 U.S.
drinking water treatment plants in the northeastern and Midwestern U.S. (Haddix et al., 2007).
Microcystins were detected at low levels ranging from undetectable (<0.15 µg/L) to 0.36 µg/L in
all 77 finished water samples.
In August 2014, the city of Toledo, Ohio issued a “do not drink or boil advisory” to nearly
500,000 customers in response to the presence of total microcystins in the city’s finished drinking
water at levels up to 2.50 µg/L. The presence of the toxins was due to a cyanobacterial bloom
near Toledo’s drinking water intake located on Lake Erie. The advisory was lifted two days later,
after treatment adjustments led to the reduction of the cyanotoxin concentrations to concentrations
below the WHO guideline value of 1 µg/L in all samples from the treatment plant and distribution
system.
2.4 Environmental Fate
Different physical and chemical processes are involved in the persistence, breakdown, and
movement of microcystins in aquatic systems as described below.
2.4.1 Persistence
Microcystins are relatively stable and resistant to chemical hydrolysis or oxidation at or

near neutral pH. Elevated or low pH or temperatures above 30°C may cause slow
hydrolysis. Microcystin is not destroyed by boiling (Rao et al., 2002). In natural waters kept in the
dark, microcystins have been observed to persist for 21 days to 2-3 months in solution and up to 6
months in dry scum (Rapala et al., 2006; Funari and Testai, 2008).
In the presence of full sunlight, microcystins undergo photochemical breakdown, but this
varies by microcystin congener (WHO, 1999; Chorus et al., 2000). The presence of water-soluble
cell pigments, in particular phycobiliproteins, enhances this breakdown. Breakdown can occur in
as few as two weeks to longer than six weeks, depending on the concentration of pigment and the
intensity of the light (Tsuji et al., 1993; 1995). According to Tsuji et al, microcystin-LR was
photodegraded with a half-life (time it takes half of the toxin to degrade) of about 5 days in the
presence of 5 mg/L of extractable cyanobacterial pigment. Humic substances can also act as
photosensitizers and can increase the rate of microcystin breakdown in sunlight. In deeper or
turbid water, the breakdown rate is slower.
Microcystins are susceptible to degradation by aquatic bacteria found naturally in rivers

and reservoirs (Jones et al., 1994). Bacteria isolates of Arthrobacter, Brevibacterium,
Rhodococcus, Paucibacter, and various strains of the genus Sphingomonas (Pseudomonas) have
been reported to be capable of degrading microcystin-LR (de la Cruz et al., 2011; Han et al.,
2012). These degradative bacteria have also been found in sewage effluent (Lam et al., 1995),
lake water (Jones et al., 1994; Cousins et al., 1996; Lahti et al., 1997a), and lake sediment (Rapala

et al., 1994; Lahti et al., 1997b). Lam et al., in 1995 reported that the biotransformation of
microcystin-LR followed a first-order decay with a half-life of 0.2 to 3.6 days (Lam et al., 1995).
In a study done by Jones et al. (1994) with microcystin-LR in different natural surface waters,
microcystin-LR persisted for 3 days to 3 weeks; however, more than 95% loss occurred within 3
to 4 days. A study by Christoffersen et al., 2002, measured half-lives in the laboratory and in the
field of approximately 1 day, driven largely by bacterial aerobic metabolism. These researchers
found that approximately 90% of the initial amount of microcystin disappeared from the water
phase within 5 days, irrespective of the starting concentration. Other researchers (Edwards et al.,
2008) have reported half-lives of 4 to 14 days, with longer half-lives associated with a flowing
stream and shorter half-lives associated with lakes.
2.4.2 Mobility
Microcystins may adsorb onto naturally suspended solids and dried crusts of
cyanobacteria. They can precipitate out of the water column and reside in sediments for months
(Han et al., 2012: Falconer, 1998). Ground water is generally not expected to be at risk of
cyanotoxin contamination, however, ground water under the direct influence of surface water can
be vulnerable. A study conducted by the USGS and the University of Central Florida determined
that microcystin and cylindrospermopsin did not sorb in sandy aquifers and were transported
along with ground water (O’Reilly et al., 2011). The authors suggested that the removal of
microcystin was due to biodegradation.
2.5 Nature of the Stressor-Characteristics of the Microcystin Toxins
2.5.1 Toxicokinetics
No data were available that quantified the intestinal, respiratory or dermal absorption of
microcystins. Available data indicate that the Organic Acid Transporter polypeptide (OATp)
receptors facilitate the absorption of toxins from the intestinal tract into liver, brain, and other
tissues. The OATp family transporters facilitate the cellular, sodium-independent uptake and
export of amphipathic compounds such as bile salts, steroids, drugs, peptides and toxins (Cheng et
al., 2005; Fischer et al., 2005; Svoboda et al., 2011). This facilitated transport is necessary for
both uptake of microcystins into organs and tissues as well as for their export. Microcystins
compete with bile acids for uptake by the liver and is limited in the presence of bile acids and
other physiologically-relevant substrates for the transporter (Thompson and Pace, 1992). Other
studies following in vitro or in vivo exposures have shown that inhibition of microcystin uptake
by its OATp transporter reduces or eliminates the liver toxicity observed (Runnegar et al., 1981,
1995; Runnegar and Falconer, 1982; Hermansky et al., 1990a, b).
Limited information is available on the metabolism of microcystins. Some studies have

found that metabolism of microcystin-LR in mice occurs in the liver (Robinson et al., 1991; Pace
et al., 1991). Most of the available studies show minimal if any catabolism (process of breaking
down molecules into smaller units to release energy). Microcystins can be conjugated with
cysteine and glutathione to increase their solubility and facilitate excretion (Kondo et al., 1996).

However, it is not clear whether hepatic cytochromes, such as cytochrome P450-facilitated
oxidation, precedes conjugation (Cote et al., 1986; Brooks and Codd, 1987). Both in vivo and in
vitro studies have shown biliary excretion (Falconer et al., 1986; Pace et al., 1991; Robinson et
al., 1991).
2.5.2 Noncancer Health Effects Data
2.5.2.1 Human Studies
The human data on the oral toxicity of microcystin-LR are limited by the potential co-
exposure to other pathogens and toxins, by the lack of quantitative information, and by the failure
to control for confounding factors.
Only a few epidemiological and case studies are available on the toxicity of microcystins
in humans. An outbreak among army recruits who had consumed reservoir water with a
cyanobacteria bloom with M. aeruginosa reported symptoms of headache, sore throat, vomiting
and nausea, stomach pain, dry cough, diarrhea, blistering around the mouth, and pneumonia
(Turner et al., 1990). Microcystins, including microcystin-LR, were present in bloom samples.
However, high levels of Escherichia coli were also found in reservoir water after two weeks. The
authors suggested that exposure to microcystins may have had a role in some of the clinical
symptoms.
An epidemiology study done in Australia compared the hepatic enzyme levels from
patients served by a public water supply contaminated with a M. aeruginosa bloom with enzyme
levels from patients living in areas served by water supplies uncontaminated by cyanobacteria
(Falconer et al., 1983). Although the authors observed significant variability in enzyme levels
between the two groups, the findings were attributed by the authors to the imprecise method of
study participant selection and confounding factors such as alcoholism and chronic kidney disease
among some of the participants.
A cross-sectional study done in China assessed the relationship between the consumption
of water and food (carp and duck) contaminated with microcystins and liver damage in children
(Li et al., 2011a). The authors found that mean serum levels of microcystins ranged from below
detection to 1.3 μg microcystin-LR equivalents/L. According to the authors, hepatitis B infection
was a greater risk for liver damage among these children than the microcystin exposure.
Acute intoxication with microcystin-producing cyanobacteria blooms in recreational water

was reported in Argentina in 2007 (Giannuzzi et al., 2011). A single person was immersed in a
Microcystis bloom with concentrations of 48.6 µg/L. After four hours of exposure, the patient
exhibited fever, nausea, and abdominal pain, and three days later, presented dyspnea and
respiratory distress and was diagnosed with an atypical pneumonia. A week after the exposure,
the patient developed a hepatotoxicosis with a significant increase of alanine aminotransferase
(ALT), aspartate aminotransferase (AST) and γ-glutamyltransferase (γGT). The patient
completely recovered within 20 days.

An outbreak of acute liver failure occurred in a dialysis clinic in 1996 in Caruaru, Brazil
where dialysis water was contaminated with microcystins, and possibly cylindrospermopsin. Of
the 130 patients who received their routine hemodialysis treatment (intravenously) at that time,
116 reported symptoms of headache, eye pain, blurred vision, nausea and vomiting. Subsequently,
100 of the affected patients developed acute liver failure and, of these, 76 died (Carmichael et al.,
2001; Jochimsen et al., 1998). Analyses of blood, sera, and liver samples from the patients
revealed microcystins. Although the patients in the study had pre-existing diseases, the direct
intravenous exposure to dialysate prepared from surface drinking water supplies put them at risk
for cyanotoxin exposure and resultant adverse effects (Hilborn et al., 2013).
In another contamination event at a dialysis center in Rio de Janeiro, Brazil in 2001, 44

dialysis patients were potentially exposed to microcystin concentrations of 0.32 µg/L, detected in
the activated carbon filter used in an intermediate step for treating drinking water to prepare
dialysate (Soares et al., 2005). Concentrations of 0.4 µg/L microcystin-LR were detected in the
drinking water. Serum samples were collected from 13 dialysis patients 31 to 38 days after the
detections in water samples, and patients were monitored for eight weeks. Concentrations of
microcystin-LR in the serum ranged from 0.46 to 0.96 ng/mL. Although the biochemical
outcomes varied among the patients, markers of hepatic cellular injury chlolestasis (elevations of
AST, ALT bilirubin, ALP and GGT) in serum during weeks one to eight after treatment
frequently exceeded normal values. Since microcystin-LR was not detected during weekly
monitoring after the first detection, the authors suggested that the patients were not continuously
exposed to the toxin and that the toxin detected in the serum after eight weeks may have been
present in the form of bound toxin in the liver (Soares et al., 2005). Results were consistent with a
mild to moderate mixed liver injury.
2.5.2.2 Animal Studies
Most of the information on the noncancer effects of microcystins in animals is from oral
and intraperitoneal (i.p.) administration studies in mice and rats exposed to purified microcystin-
LR. Liver effects are observed following acute oral exposure to microcystin-LR (Yoshida et al.,
1997; Ito et al., 1997b; Fawell et al., 1999). Effects on the liver, kidney, and male reproductive
system (testicular function and sperm quality), including changes in organ weights and
histopathological lesions, are observed following short-term and subchronic oral exposure to
microcystin-LR (Heinze, 1999; Fawell et al., 1999; Huang et al., 2011; Chen et al., 2011). Oral
and i.p. developmental toxicity studies in mice provide some evidence for fetal body weight
changes and maternal mortality (Fawell et al., 1999; Chernoff et al., 2002).
According to the authors, no clinical signs of toxicity were observed in a chronic study
done in mice for 18 months by Ueno et al. (1999). Although histopathology from a 280 day study
in mice revealed infiltrating lymphocytes and fatty degeneration in the livers, no quantitative data
were provided in the study (Zhang et al., 2012).

2.5.3 Mode of Action for Noncancer Health Effects
Mechanistic studies have shown the importance of membrane transporters for systemic
uptake and tissue distribution of microcystin by all exposure routes (Fischer et al. 2005; Feurstein
et al., 2010). The importance of the membrane transporters to tissue access is demonstrated when
a reduction in, or lack of, liver damage happens following OATp inhibition (Hermansky et al.,
1990 a,b; Thompson and Pace, 1992).
The uptake of microcystins causes protein phosphatase inhibition and a loss of

coordination between kinase phosphorylation and phosphatase dephosphorylation, which results
in the destabilization of the cytoskeleton. This event initiates altered cell function followed by
cellular apoptosis and necrosis (Barford et al., 1998). Both cellular kinases and phosphatases keep
the balance between phosphorylation and dephosphorylation of key cellular proteins controlling
metabolic processes, gene regulation, cell cycle control, transport and secretory processes,
organization of the cytoskeleton and cell adhesion. Each of the microcystin congeners evaluated
(LR, LA, and LL) interacts with catalytic subunits of protein phosphatases PP1 and PP2A,
inhibiting their functions (Craig et al., 1996).
As a consequence of the microcystin-induced changes in cytoskeleton, increases in

apoptosis and reactive oxygen species (ROS) occur. In both in vitro and in vivo studies, cellular
pro-apoptotic Bax and Bid proteins increased while anti-apoptotic Bcl-2 decreased (Fu et al.,
2005; Weng et al., 2007; Xing et al., 2008; Takumi et al., 2010; Huang et al., 2011; Li et al.,
2011b). Mitochondrial membrane potential and permeability transition pore changes (Ding and
Ong, 2003; Zhou et al., 2012) lead to membrane loss of cytochrome c, a biomarker for apoptotic
events. Wei et al., (2008) found a time-dependent increase in ROS production and lipid
peroxidation in mice after exposure to microcystin-LR. After receiving a 55 µg/kg of body weight
i.p. injection of microcystin-LR, the levels of hepatic ROS increased rapidly within 0.5 hours and
continued to accumulate for up to 12 hours in a time-dependent manner.
2.5.4 Carcinogenicity Data
Several human epidemiological studies from China have reported an association between
liver or colon cancer and consumption of drinking water from surface waters containing
cyanobacteria and microcystins (Ueno et al., 1996; Zhou et al., 2002). In these studies, a
concentrations measured in a surface drinking water supply were used as a surrogate for exposure
to microcystins. Individual exposure to microcystins was not estimated, and there was no
examination of numerous possible confounding factors, such as co-occurring chemical
contaminants or hepatitis infections in the population.
A study done by Flemming et al. (2002, 2004) in Florida failed to find a significant
association for primary liver cancer between populations living in areas receiving their drinking
water from a surface water treatment plant (with the potential for microcystin exposures), and the
Florida general population, or those receiving their water from ground-water sources. The one
significant association observed was between those people in the surface water service areas,
versus those in their surrounding areas described as buffer zones. However, the nature of the
water supply for the buffer zones were not identified.

The only longer-term oral animal study of purified microcystin-LR was conducted by Ito
et al. (1997b). Ito et al. (1997b) administered 80 µg microcystin-LR/kg/day by gavage to mice for
80 or 100 days over 28 weeks (7 months). This single dose failed to induce neoplastic nodules of
the liver. The lack of hyperplastic nodules at 7 months suggests that microcystins are not a
mutagenic initiator of tumors, however, the fairly short duration may have been a limiting factor.
Several studies suggest that microcystin-LR is a tumor promoter. In these studies, animals
were first exposed to substances known to be tumor initiators (e.g. N-methyl-N-nitroso urea or
NDEA) alone, or in combination with microcystin-LR at i.p. doses known to have no significant
impact on liver weight. The combination of the initiator and the microcystin-LR significantly
increased the number and area of glutathione S-transferase placental form-positive (GST-P) foci
when compared to treatment with the initiator alone. The same was true for situations where the
initiator treatment was combined with a partial hepatectomy (to stimulate tissue repair) and then
exposed to microcystins i.p. (Nishiwaki-Matsushima et al., 1992; Ohta et al., 1994). GST-P foci
are regarded as indicators for potential tumors formation. The results from these studies support
the classification of microcystin as a tumor promoter.
2.6 Conceptual Model for Microcystins
The conceptual model is intended to explore potential links of exposure to a contaminant

or stressor with the adverse effects and toxicological endpoints important for management goals,
including the development of HA values. The conceptual model demonstrates the relationship
between exposure to microcystins in drinking water and adverse health effects in the populations
at risk.
HAs describe non-regulatory concentrations of drinking water contaminants at which

adverse health effects are not anticipated to occur over specific exposure durations (e.g., one-day,
ten-days, and a lifetime). HAs also contain a margin of safety to protect sensitive members of the
population. They serve as informal technical guidance to assist federal, state and local officials, as
well as managers of public or community water systems, in protecting public health. They are not
to be construed as legally enforceable federal standards.
Assessment endpoints for HAs can be developed for both short-term (one-day and ten-
day) and lifetime exposure periods using information on the non-carcinogenic and carcinogenic
toxicological endpoints of concern. Where data are available, endpoints will reflect susceptible
and/or more highly exposed populations.
• A One-day HA is typically calculated for an infant (0-12 months or 10kg child), assuming
a single acute exposure to the chemical and is generally derived from a study of less than
seven days’ duration.
• A Ten-day HA is typically calculated for an infant (0-12 months or 10kg child), assuming
a limited period of exposure of one to two weeks, and is generally derived from a study of
7 to 30-days duration.

• A Lifetime HA is derived for an adult (>21 years or 80kg adult), and assumes an exposure
period over a lifetime (approximately 70 years). It is usually derived from a chronic study
of two years duration, but subchronic studies may be used by adjusting the uncertainty
factor employed in the calculation. For carcinogens, the HA documents typically provide
the concentrations in drinking water associated with risks for one excess cancer case per
ten thousand persons exposed up to one excess cancer case per million exposed for Group
A and B carcinogens and those classified as known or likely carcinogens (U.S. EPA,
1986, 2005). Cancer risks are not provided for Group C carcinogens or those classified as
“suggestive”, unless the cancer risk has been quantified.
For each assessment endpoint EPA uses one or more measures of effect (also referred to
as a point of departure), which describe the change in the attribute of the assessment endpoint in
response to chemical exposure, to develop acute, short-term, longer term (subchronic) or chronic
reference values when the data are available. The measures of effect selected represent impacts on
survival, growth, system function, reproduction and development.
This conceptual model provides useful information to characterize and communicate the
potential health risks related to exposure to cyanotoxins in drinking water. The sources of
cyanotoxins in drinking water, the route of exposure for biological receptors of concern (e.g., via
various human activities such as drinking, food preparation and consumption) and the potential
assessment endpoints (i.e., effects such as kidney and liver toxicity, and reproductive and
developmental effects) due to exposure to microcystins are depicted in the conceptual diagram
below (Figure 2-3).
2.6.1 Conceptual Model Diagram
Cyanobacteria are a common part of freshwater and marine ecosystems. An increase in

water column stability, high water temperatures, elevated concentrations of nutrients, and low
light intensity have been associated with an increase and or dominance of microcystin-producing
cyanobacteria in surface waters (or aquatic ecosystems). The presence of detectable
concentrations of cyanotoxins in the environment is closely associated with these blooms. Winds
and water currents can potentially transport cyanobacterial blooms to areas within the proximity
of water intakes for drinking water treatment plants. If not managed in source waters, or removed
during drinking water treatment, cyanobacteria and cyanotoxins may result in exposure that could
potentially affect human health.

Figure 2-3. Conceptual Model of Exposure Pathways to Microcystins in Drinking Water

2.6.2 Factors Considered in the Conceptual Model for Microcystins
Stressors: For this HA, the stressor is microcystins concentrations in finished drinking water.
Sources: Sources of microcystins include potential sources of drinking water such as rivers,
reservoirs, and lakes in the U.S. where blooms producing microcystins occur. Shallow private
wells under the direct influence of surface water (in hydraulic connection to a surface water body)
can also be impacted by microcystins-producing blooms, if the toxins are drawn into the well
along with the water from the surface water. There is substantially less information on exposure
from this source.
Routes of exposure: Exposure to cyanotoxins from contaminated drinking water sources may
occur via oral exposure (drinking water, cooking with water, and incidental ingesting from
showering); dermal exposure (contact of exposed parts of the body during bathing or showering,
washing dishes, or outside activities); inhalation exposure (during bathing, showering or washing
dishes,); or intravenous exposure (e.g. via dialysis). Toxicity data are available for the oral route
of exposure from drinking water, but are not available to quantify dose response for other
exposure routes (inhalation, dermal, dietary, and intravenous exposures).
Receptors: The general population (adults and children) could be exposed to cyanotoxins through
dermal contact, inhalation and/or ingestion. Infants and pre-school age children can be at greater
risk to microcystins because they consume more water per body weight than do adults. Other
individuals of potential sensitivity are persons with kidney and/or liver disease due to the
compromised detoxification mechanisms in the liver and impaired excretory mechanisms in the
kidney. There are no human data to quantify risk to pregnant woman or to evaluate the transfer of
cyanotoxins across the placenta. Data are also not available on the transfer of cyanotoxins through
the milk from nursing mothers or regarding the risk to the elderly. Given this lack of information,
pregnant women, nursing mothers, and the elderly may also be potentially sensitive populations.
Data from the episode in a dialysis clinic in Caruaru, Brazil where microcystins were not removed
by treatment of dialysis water, identify dialysis patients as a population of potential concern in
cases where the drinking water source for the clinic is contaminated with cyanotoxins. Data are
not available to derive a One-day HA for children because studies with single oral dosing do not
provide dose-response information. A lifetime HA for microcystins is not recommended as the
types of exposures being considered are short-term and episodic in nature. Although the majority
of the cyanobacterial blooms in the U.S. occur seasonally, usually during late summer, some
toxin-producing strains can occur early in the season and can last for days or weeks.
Endpoints: Human data on oral toxicity of microcystins are limited, but suggest the liver as the
primary target organ. Acute, short-term, and subchronic studies in animals also demonstrate that
the liver and kidney are target organs. In addition, some studies suggest that microcystins may
lead to reproductive and developmental effects. Studies have suggested that microcystins have
tumor promotion potential if there has been co-exposure to a carcinogen or cellular organ damage.
However, these data are limited, and there has been no long term bioassay in animals to evaluate
cancer. Available toxicity data are described in the Health Effects Support Document (HESD) for
Microcystins (U.S. EPA, 2015a), and indicate that the primary target organ for microcystins is the
liver. Kidney and reproductive effects in male mice were also observed, but were either not as

sensitive as the liver or lack confirmation from more than one laboratory. Data are inadequate to
assess the carcinogenic potential of microcystins at this time.
2.7 Analysis Plan
The Health Effects Support Document for Microcystins (HESD, U.S. EPA, 2015a),
provides the health effects basis for development of the HA, including the science-based
decisions providing the basis for estimating the point of departure. To develop the HESD for
microcystins, a comprehensive literature search was conducted from January 2013 to May 2014
using Toxicology Literature Online (TOXLINE), PubMed component, and Google Scholar to
ensure the most recent published information on microcystins was included. Some of the search
terms included in the literature search were microcystin, microcystin congeners, human toxicity,
animal toxicity, in vitro toxicity, in vivo toxicity, occurrence, environmental fate, mobility, and
persistence. EPA assembled available information on occurrence, environmental fate,
mechanisms of toxicity, acute, short-term, subchronic and chronic toxicity and cancer in humans
and animals, toxicokinetics, and exposure. Additionally, EPA considered information from the
following risk assessments during the development of the microcystins health risk assessment:
• Health Canada (2012) Toxicity Profile for Cyanobacterial Toxins

• Enzo Funari and Emanuela Testai (2008) Human Health Risk Assessment Related to
Cyanotoxins Exposure
• Tai Nguyen Duy, Paul Lam, Glen Shaw and Des Connell (2000) Toxicology and Risk
Assessment of Freshwater Cyanobacterial (Blue-Green Algal) Toxins in Water
The toxicity data available for an individual pollutant vary significantly. An evaluation of
available data was performed by EPA to determine data acceptability. The following study quality
considerations from U.S. EPA’s (2002) A Review of the Reference Dose and Reference
Concentration Processes were used in selection of the studies for inclusion in the HESD and
development of the HA.
• Clearly defined and stated hypothesis.

• Adequate description of the study protocol, methods, and statistical analyses.
• Evaluation of appropriate endpoints. Toxicity depends on the amount, duration, timing,
and pattern of exposure and may range from frank effects (e.g., mortality) to more subtle
biochemical, physiological, pathological, or functional changes in multiple organs and
tissues.
• Application of the appropriate statistical procedures to determine an effect.
• Establishment of dose-response relationship (i.e., no observed adverse effect level
(NOAEL) and/or lowest observed adverse effect level (LOAEL) or data amenable to
modeling of the dose-response in order to identify a point of departure for a change in the
effect considered to be adverse (out of the range of normal biological viability). The
NOAEL is the highest exposure level at which there are no biologically significant

increases in the frequency or severity of adverse effect between the exposed population
and its appropriate control. The LOAEL is the lowest exposure level at which there are
biologically significant increases in frequency or severity of adverse effects between the
exposed population and its appropriate control group.
After the available studies were evaluated for inclusion in the HESD and HA, the critical
study was selected based on consideration of factors including exposure duration (comparable to
the duration of the HA being derived), route of exposure (oral exposure via drinking water,
gavage, or diet is preferred), species sensitivity, comparison of the point of departure with other
available studies demonstrating an effect, and confidence in the study (U.S. EPA, 1999). Once, a
point of departure is chosen for quantification, uncertainty factors appropriate for the study
selected are then applied to the point of departure to account for variability and uncertainty in the
available data.
For microcystins, toxicity and exposure data are available to develop a Ten-day HA. EPA
used measures of effect and estimates of exposure to derive the Ten-day HAs using the following
equation:
NOAEL or LOAEL or BMDL

HA =
UF × DWI/BW
Where:
NOAEL or = No- or Lowest-Observed-Adverse-Effect Level (mg/kg bw/day) from a study

LOAEL of an appropriate duration (7 to 30 days).
BMDL = When the data available are adequate, benchmark dose (BMD) modeling can
be performed to determine the point of departure for the calculation of HAs.
The benchmark dose approach involves dose-response modeling to obtain
dose levels corresponding to a specific response level near the low end of the
observable range of the data (U.S.EPA, 2012). The lower 95% confidence
limit is termed the benchmark dose level (BMDL).
UF = Uncertainty factors (UF) account for: (1) intraspecies variability (variation in

susceptibility across individuals); (2) interspecies variability (uncertainty in
extrapolating animal data to humans; (3) uncertainty in extrapolating from a
LOAEL to a NOAEL; and (4) uncertainty associated with extrapolation when
the database is incomplete. These are described in U.S. EPA, 1999 and U.S.
EPA, 2002.
DWI/BW = For children, a normalized ratio of drinking water ingestion to body weight
(DWI/BW) was calculated using data for infants (birth to <12 months). The
estimated drinking water intake body weight ratio (L/kg/day) used for birth to

<12 months of age are the 90th percentile values of the consumers only
estimates of direct and indirect water ingestion based on 1994-1996, 1998
CSFII (Continuing Survey of Food Intakes by Individuals) (community water,
mL/kg/day) in Table 3-19 in the U.S. EPA (2011a) Exposure Factors
Handbook. The time weighted average of DWI/BW ratios values was derived
from multiplication of age-specific DWI/BW ratios (birth to <1 month, 1 to
<3 months, 3 to <6 months, and 6 to <12 months) by the age-specific fraction
of infant exposures for these time periods.
For adults (>21 years of age), EPA updated the default BW assumption to 80
kg based on National Health and Nutrition Examination Survey (NHANES)
data from 1999 to 2006 as reported in Table 8.1 of EPA’s Exposure Factors
Handbook (U.S. EPA, 2011a). The updated BW represents the mean weight
for adults ages 21 and older.
EPA updated the default DWI to 2.5 L/d, rounded from 2.546 L/d, based on
NHANES data from 2003 to 2006 as reported in EPA’s Exposure Factors
Handbook (U.S. EPA 2011a, Table 3-33). This rate represents the consumer’s
only estimate of combined direct and indirect community water ingestion at
the 90th percentile for adults ages 21 and older.

3.0 HEALTH EFFECTS ASSESSMENT
The health effects assessment provides the characterization of adverse effects and includes
the hazard identification and dose-response assessment. The hazard identification includes
consideration of available information on toxicokinetics; identification, synthesis and evaluation
of studies describing the health effects of microcystins; and the potential modes of action
(MOAs), or toxicity pathways related to the health effects identified.
3.1 Dose-Response
3.1.1 Study Selection
The critical study chosen for determining the guideline value is a short-term study by
Heinze (1999) in which 11-week-old male hybrid rats (F1 generation of female WELS/Fohm x
male BDIX) were administered microcystin-LR via drinking water for 28 days at concentrations
of 0 (n=10), 50 (n=10) or 150 (n=10) μg/kg body weight (Heinze, 1999). Water consumption was
measured daily, and rats were weighed at weekly intervals. The dose estimates provided by the
authors were not adjusted to account for incomplete drinking water consumption (3-7% of
supplied water was not consumed over the 28-day period). Rats were sacrificed by exsanguination
under ether anesthesia after 28 days of exposure, and evaluation of hematology, serum
biochemistry plus histopathology of liver and kidneys, and measurement of organ weights (liver,
kidneys, adrenals, spleen and thymus) was performed.
Hematological evaluation showed an increase of 38% in the number of leukocytes at the

highest dose group (150 μg/kg body weight). Serum biochemistry showed a significant increase in
both treatment groups in mean levels of alkaline phosphatase (ALP) and lactate dehydrogenase
(LDH); 84 and 100% increase in LDH, and 34 and 33% increase in ALP, in the low and high dose
groups respectively. No changes were observed in mean levels of AST (aspartate
aminotransferase), and ALT (alanine aminotransferase). An increase in relative liver weights was
observed in a dose-dependent manner; 17% at 50 μg/kg body weight, and 26% at 150 μg/kg body
weight. Mean enzyme levels and relative liver weights are shown in Table 3-1.
A dose-dependent increase in absolute liver weight was also reported, and data on the liver
weights were provided by the author in a personal communication. A dose-dependent increase in
the average absolute liver weights was also observed in all groups: 8.8 grams at the control group,
9.70 grams at the lower dose and 10.51 grams at the high dose (Table 3-1). No statistically
significant changes in other organ weights or body weights were reported, and no effects on the
kidneys were observed. Table 3-2 summarizes the histological observations of liver lesions. Liver
lesions were considered toxic and spread diffusely throughout the parenchyma indicating cell
damage expressed by an increase in cell volume, an increase in mitochondria, cell necrosis, the
activation of Kupffer cells, and an increase in the amounts of periodic acid-Schiff (PAS)-positive
substances. Liver lesions were observed in both treatment groups. No kidney effects were
observed in either dose groups. The LOAEL was determined to be 50 μg/kg/day. The selection of
Heinze (1999) as the critical study was based on the appropriateness of the study duration, the use

Table 3-1. Liver Weights and Serum Enzyme Levels in Rats Ingesting Microcystin-LR in
Drinking Water (Heinze, 1999)
Control 50 μg/kg 150 μg/kg
(Mean ± SD) (Mean ± SD) (Mean ± SD)
Serum Enzymes
Alkaline phosphatase (ALP)
9.67 ± 2.20 13.00 ± 3.81* 12.86 ± 1.85*
(microkatals/L)
Lactate dehydrogenase (LDH)
16.64 ± 4.48 30.64 ± 5.05* 33.58 ± 1.16*
(microkatals/L)
Liver Weight
Relative (g/100 g body weight) 2.75 ± 0.29 3.22 ± 0.34* 3.47 ± 0.49*
Absolute (g)** 8.28 ± 1.37 9.70 ± 1.32 10.51 ± 1.02
* p<0.05 when compared with control; katal=conversion rate of 1 mole of substrate per second.
**Information provided by the author through a personal communication.
Table 3-2. Histological Evaluation of the Rat Livers after Ingesting Microcystin-LR in
Drinking Water (Heinze, 1999)
Degenerative Degenerative
and Necrotic and Necrotic
Activation of PAS-positive
Hepatocytes Hepatocytes
Kupffer Cells Material
with without
Hemorrhage Hemorrhage
Control
Slight 0 0 0 1
Moderate 0 0 0 0
Intensive damage 0 0 0 0
50 μg/kg
Slight 0 4 0 5
Moderate 10 6 0 5
150 μg/kg
Slight 0 0 0 0
Moderate 10 6 1 8

of multiple doses, dose-related toxicological responses, and histopathological evaluations of
toxicity.
3.1.2 Endpoint Selection
The point of departure selected from the Heinze (1999) study is the LOAEL (50
μg/kg/day) for liver effects (increased liver weight, slight to moderate liver necrosis lesions, with
or without hemorrhages at the low dose and increased severity at the high dose, and changes in
serum enzymes indicative of liver damage). For the lesions, incidence increases from one animal
impacted in the control group to ten animals impacted in the dosed groups. This dose-response is
more dramatic than the difference in liver weight between the control and low dose (1.17 fold)
and the differences in the ALP and LDH levels between the control and low dose group (1.34 and
1.84-fold, respectively). Therefore, the liver lesions are identified as the endpoint of greatest
concern. These differences also advise against application of benchmark dose modeling for these
effects. The male and female mice in the Fawell et al (1999) study displayed liver lesions, but the
difference between controls and the low dose group (40 μg/kg/day) was less than two-fold. In an
i.p. infusion study by Guzman and Solter (1999) with a more direct delivery of dose to the liver,
necrosis was observed at doses of 32 and 48 μg/kg/day, but not at a dose of 16 μg/kg/day, thus
providing support for the critical effect and dose.
3.2 Ten-day Health Advisory
This Ten-day HA is applied to total microcystins using microcystin-LR as a surrogate.

The Ten-day HA is considered protective of non-carcinogenic adverse health effects over a ten-
day exposure to microcystins in drinking water.
3.2.1 Bottle-fed Infants and Young Children of Pre-school Age
The Ten-day HA for bottle-fed infants and young children of pre-school age is calculated as
follows:
50 μg/kg/day
Ten -day HA = = 0.3 μg/L
1000 × 0.15 L/kg/day
Where:
50 µg/kg/day = The LOAEL for liver effects in 11-week-old male hybrid rats
exposed to microcystin-LR in drinking water for 28 days
(Heinze, 1999).
1000 = The composite UF including a 10 for intraspecies variability
(UFH), a 10 for interspecies differences (UFA), a 3 for LOAEL to
NOAEL extrapolation (UFL), and a 3 for uncertainties in the
database (UFD).
0.15 L/kg/day = Normalized drinking water intake per unit body weight over the
first year of life based on the 90th percentile of drinking water
consumption and the mean body weight (U.S. EPA, 2011a).

The Ten-day HA of 0.3 µg/L is considered protective of non-carcinogenic adverse health effects
for bottle-fed infants and young children of pre-school age over a ten-day exposure to
microcystins in drinking water.
3.2.2 School-age Children through Adults
The Ten-day HA for school-age children through adults is calculated as follows:
50 μg/kg/day
Ten -day HA = = 1.6 μg/L
1000 × 0.03 L/kg/day
Where:
50 µg/kg/day = The LOAEL for liver effects in 11-week-old male hybrid
rats exposed to microcystin-LR in drinking water for 28
days (Heinze, 1999).
1000 = The composite UF including a 10 for intraspecies
variability (UFH), a 10 for interspecies differences (UFA), a
3 for LOAEL to NOAEL extrapolation (UFL), and a 3 for
uncertainties in the database (UFD).
0.03 L/kg/day = Drinking water intake per unit body weight based on adult
default values of 2.5 L/day and 80 kg (U.S. EPA, 2011a).
The Ten-day HA of 1.6 µg/L is considered protective of non-carcinogenic adverse health effects
for children of school age through adults over a ten-day exposure to microcystins in drinking
water.
3.2.3 Uncertainty Factor Application
• UFH - A Ten-fold value is applied to account for variability in the human population. No
information was available to characterize interindividual and age-related variability in the
toxicokinetics or toxicodynamics among humans. Individuals with pre-existing liver
problems could be more sensitive to microcystins exposures than the general population.
Pregnant woman, nursing mothers, and the elderly could also be sensitive to microcystins
exposures.
• UFA - A Ten-fold value is applied to account for uncertainty in extrapolating from

laboratory animals to humans (i.e., interspecies variability). Information to quantitatively
assess toxicokinetic or toxicodynamic differences between animals and humans is
unavailable for microcystins. Allometric scaling is not applied in the development of the
Ten-day HA values for microcystins. The allometric scaling approach is derived from the
relationship between body surface area and basal metabolic rate in adults (U.S. EPA,
2011b). This approach is not appropriate for infants and children due to the comparatively
slower clearance during these ages and the limited toxicokinetic data available to assess
the appropriateness of body weight scaling in early life.

• UFL - An uncertainty factor of 3 (100.5 = 3.16) is selected to account for the extrapolation
from a LOAEL to a NOAEL. The threefold factor is justified based on the evidence
suggesting that the uptake of microcystins by tissues requires membrane transporters.
Uptake from the intestines involves both apical and basolateral transporters, uptake by the
microvilli capillaries and portal transport to the liver. Transporters are again necessary for
hepatic uptake. When there is slow infusion into the peritoneum and into the portal
intraperitoneal capillaries, uptake is described as rapid because of the rich blood supply
and large surface area of the peritoneal cavity (Klassen, 1996). Delivery of the
microcystins to the intraperitoneum increases the amount of the dose that reaches the liver
for three additional reasons: 1) the apical and basolateral intestinal barriers to uptake are
eliminated with the i.p. infusion; 2) there is no dilution of dose by the gastric plus
intestinal fluids as when food residues are in the gastrointestinal track; and 3) there is no
delay in reaching the site of absorption because of gastric emptying time (Klassen, 1996).
In addition, facilitated transporter kinetics are similar to Michaelis Menton enzyme
kinetics in that there are Km and Vmax components that are defined by the affinity of the
transported substance for the transporter.
In the Guzman and Solter (1999) intraperitoneal infusion study in rats, the NOAEL is 16
µg/kg/day and the LOAEL is 32 µg/kg/day, a two-fold difference. There is no reason to
believe that the less direct delivery from the intestines to the liver following oral
exposures through drinking water (as was used in Heinze, 1999) would have a more than
3-fold separation between a NOAEL and LOAEL had there been one in the Heinze (1999)
study.
• UFD - An uncertainty factor of 3 (100.5 = 3.16) is selected to account for deficiencies in the
database for microcystins. The database includes limited human data, including studies
evaluating the association between microcystin exposure and cancers in liver and colon,
and systemic effects including liver endpoints such as elevated liver enzymes. Oral and
i.p. acute and short-term studies on mice and rats, and subchronic studies done in mice are
available. Chronic data are also available for microcystin, however, are limited by the lack
of quantitative data provided in the study. Additionally, there are limited neurotoxicity
studies (including a recent publication on developmental neurotoxicity) and several i.p.
reproductive and developmental toxicity studies. The database lacks a multi-generation
reproductive toxicity study.
The default factors typically used cover a single order of magnitude (i.e., 101). By convention,
in the Agency, a value of 3 is used in place of one-half power (i.e., 10½) when appropriate (U.S.
EPA, 2002).

4.0 RISK CHARACTERIZATION
The following topics describe important conclusions used in the derivation of the health
advisory. This section characterizes each topic and its impact on the health advisory.
4.1 Use of microcystin-LR as a surrogate for total microcystins
Among the approximately 100 different congeners of microcystins known to exist,

microcystin-LR is the most common. The difference in toxicity of microcystin congeners depends
on the amino acid composition (Falconer, 2005). Stoner et al. (1989) administered by
intraperitoneal (i.p.) purified microcystin congeners (-LR, -LA, -LY and -RR) into ten or more
adult male and female Swiss albino mice. Necropsies were performed to confirm the presence of
the pathognomonic hemorrhagic livers. The authors reported 50% lethal doses (LD50) of 36 ng/g-
bw for -LR, 39 ng/g-bw for -LA, 91 ng/g- bw for -LY and 111 ng/g-bw for –RR. Similarly, Gupta
et al., (2003) determined LD50 for the microcystin congeners LR, RR and YR in female mice
using DNA fragmentation assay and histopathology examinations of the liver and lung. The acute
LD50 determination showed that the most toxic congener was microcystin-LR (43.0 µg/kg),
followed by microcystin-YR (110.6 µg/kg) and microcystin-RR (235.4 µg/kg). The most toxic
microcystins are those with the more hydrophobic L-amino acids (-LA, -LR, -and -YM), and the
least toxic are those with hydrophilic amino acids, such as microcystin-RR.
Wolf and Frank (2002) proposed toxicity equivalency factors (TEFs) for the four major
microcystin congeners based on LD50 values obtained after i.p. administration. The proposed
TEFs, using microcystin-LR as the index compound (TEF=1.0) were 1.0 for microcystin-LA and
microcystin-YR and 0.1 for microcystin-RR. The application of TEFs based on i.p. LD50 values to
assessment of risk from oral or dermal exposure is questionable given that differences in
lipophilicity and polarity of the congeners may lead to variable absorption by non-injection routes
of exposure.
The potential health risks from exposure to mixtures of microcystin congeners is

unknown, and since microcystin-LR is one of the most potent congeners and has the majority of
toxicological data on adverse health effects, microcystin-LR is used as a surrogate for all
microcystins in the health advisory.
4.2 Consideration of Study Duration
EPA used a 28-day study conducted by Heinze (1999) to derive the Ten-day HA for
microcystins. It is standard to use studies that are 7 to 30 days in duration to derive a 10-day
advisory value. In the study conducted by Heinze (1999), rats were dosed daily via drinking water
with microcystin and sacrificed at the conclusion of the study. No interim sacrifices were
performed to evaluate effects at 10 days or any other time less than the full 28 days. At the
conclusion of the 28-day study, adverse effects observed in the liver included increases in liver
weight, slight to moderate liver necrosis lesions accompanying hemorrhages at the low dose with
increased severity at the high dose, and changes in serum enzymes indicative of liver damage.
Given the lack of interim effects data, it is not known when during the 28-day study these effects
were manifested.

4.3 Consideration of Reproductive Effects as Endpoint
Upon consideration of all available studies, liver effects were considered the most
appropriate basis for quantitation as it was a common finding among oral toxicology studies
(Falconer et al., 1994; Fawell et al., 1999; Ito et al., 1997b). However, while the liver is the
primary target of microcystin toxicity, there have been reports of effects of microcystin-LR on the
male reproductive system and sperm development following oral exposures (Chen et al., 2011).
In a study conducted by Chen et al. (2011), oral exposures to low concentrations of

microcystin-LR for 3 to 6 months showed reproductive toxicity including decreased sperm counts
and sperm motility, as well as an increase in sperm abnormalities, decreased serum testosterone
and increased serum luteinizing hormone (LH) levels. Because these effects were observed at
doses lower (0.79 μg/kg/day) than those observed for liver effects in Heinze (1999), EPA
evaluated Chen et al. (2011) and the lesions in the testes and effects on sperm motility as the
potential critical study and points of departure for the derivation of the RfD for microcystins.
The Chen et al., 2011 study has several limitations in the experimental design and
reporting. There was a lack of data reported on testis weights and sperm motility. The authors
reported “no significant differences in testis weights,” but no information was provided on the
weights of the testis or whether there was a trend toward decreasing weights that failed to be
statistically significant. Also, no information was given on the methodology used for sperm
motility evaluation. No information was provided on how samples were handled and what
measurements were made to determine the percentage of sperm motility. Although body weight
and amount of water consumed were measured, these data were not presented, and doses to the
animals were not calculated by the study authors. In addition, the purity of microcystin-LR and
the species and age of the mouse used were not reported. Male sperm characteristics such as
volume, motility, and structure of sperm differ developmentally by age. Therefore, not knowing
the age of the mice in the study introduces uncertainty in the quantification of the reproductive
effects.
The fixation and staining of the testes used for microscopic examination
(paraformaldehyde in phosphate-buffered saline (PBS) and paraffin), could result in the
generation of artifacts, such as disruption of the testicular tubes. Cytoplasmic shrinkage and
chromatin aggregations were observed in both control and experimental groups. In order to
preserve the microstructure of the testis, dual fixation such as Davidson’s or Bouin’s fixation
followed by PAS staining should have been done. In addition, the histopathology analysis of the
testes reported by the authors did not provide sufficient detail to adequately assess the degree of
damage.
The quality of the medium used for the sperm analysis, and the lack of additional data
from the sperm analysis measurements carried out through the computer-assisted sperm analysis
(CASA) are additional limitations in experimental design for this study. Very few details of the
serum hormone assay protocol and the quantitative parameters of sperm motility from the CASA
analysis were provided. Therefore, the calculation for the motility of the sperm was unclear and
could not be verified.

Based on the limitations in study design, report and methods used by Chen et al. (2011),
EPA concluded that the quantitative data on decreased sperm counts and sperm motility were not
appropriate for determining the point of departure for the derivation of the RfD for microcystin.
4.4 Allometric Scaling Approach
Allometric scaling was not applied in the development of the short term RfD for
microcystins. In the development of short-term advisory values (One-day and Ten-day),
parameters are used that reflect exposures and effects for infants up to one year of age, rather than
for adults. The body weight scaling approach is derived from the relationship between body
surface area and basal metabolic rate in adults. Infants/children surface area and basal metabolic
rates are very different than adults with a slower metabolic rate. In addition, limited toxicokinetic
data are available to assess the appropriateness of body weight scaling in early life. The body
weight scaling procedure has typically been applied in the derivation of chronic oral RfDs and
cancer assessments, both of which are concerned with lifetime repeated exposure scenarios (U.S.
EPA, 2012). Thus, given the short term duration of the critical study and the development of a
short term RfD for determination of a Ten-day HA value, and the application of the Ten-day HA
to infants and pre-school age children, the application of the body weight scaling procedure is not
appropriate for this scenario.
In addition, for short-term advisories (one-day and ten-day duration), EPA assumes all
exposure is derived from drinking water and, therefore, no Relative Source Contribution (RSC)
term is applied. For lifetime health advisory values, EPA does include an RSC that reduces the
advisory value to account for other potential sources.
4.5 Benchmark Dose (BMD) Modeling Analysis
The data set reported by Heinze (1999) was evaluated for BMD modeling. Heinze (1999)
demonstrated dose-related liver changes and statistically significant effects at the lowest dose (50
µg/kg/day). Histological changes were also observed in all the animals (ten) in each dose group
(Table 3-2). Although differences in the degree of necrosis were observed with or without
hemorrhage related to dose, all the histological effects including Kupffer cell activation and PAS
staining showed no dose-response since all ten animals at the low and high doses displayed liver
damage associated with each effect. Therefore, the dose-response for the sum of the incidence
categories (slight, moderate, and intensive damage), are not amenable to BMD modeling. As a
result, the LOAEL of 50 µg/kg/day described by Heinze (1999) was used as the POD for
development of the HA.
4.6 Carcinogenicity Evaluation
While there is evidence of an association between liver and colorectal cancers in humans
and microcystins exposure and some evidence that microcystin-LR is a tumor promoter in
mechanistic studies, there is inadequate information to assess carcinogenic potential of
microcystins in humans (U.S. EPA, 2005). The human studies are limited by lack of exposure

information and the uncertainty regarding whether or not these studies adequately controlled for
confounding factors such as Hepatitis B infection. No chronic cancer bioassays for microcystins
in animals are available.
The only oral study that examined the tumorgenicity of microcystin-LR failed to find
preneoplastic nodules in the livers of groups of 22 mice receiving up to 100 doses of 0 or 80
μg/kg/day over 7 months. Some studies suggest that microcystin-LR is a tumor promoter. Given
the potential impact on the cell cytoskeleton, necrotic effects on liver cell generation of reactive
oxygen species (ROS), and other biochemical changes, this finding is not surprising. The work by
Nishiwaki-Matsushima et al., 1992 that compares glutathione S-transferase placental form-
positive (P-GST) foci from 10 µg/L microcystin-LR to that from the phenobarbital (0.05% in the
diet) as a positive control suggests that it is at best a weak promoter. The results from the second
part of the same study that compare P-GST foci following initiation with DEN followed by
microcystin-LR (10 µg/kg), both before and after a partial hepatectomy, support this conclusion.
The International Agency for Research on Cancer (IARC) classified microcystin-LR as a

Group 2B (possibly carcinogenic to humans) based on the conclusion that there was strong
evidence supporting a plausible tumor promoter mechanism for these liver toxins. U.S. EPA’s
Cancer Guidelines (2005) state that the descriptor of “inadequate information to assess
carcinogenic potential” is appropriate when available data are judged inadequate for applying one
of the other descriptors or for situations where there is little or no pertinent information or
conflicting information. The guidelines also state that (p. 2-52) “Descriptors can be selected for
an agent that has not been tested in a cancer bioassay if sufficient other information, e.g.,
toxicokinetic and mode of action information, is available to make a strong, convincing, and
logical case through scientific inference”. In the case of microcystins, the data suggest that
microcystin-LR may be a tumor promoter but not an initiator. Without stronger epidemiology
data and a chronic bioassay of purified microcystin-LR, the data do not support classifying
microcystin-LR as a carcinogen.
4.7 Uncertainty and Variability
Several uncertainty factors were applied in several areas to account for incomplete
information. Human data on the toxic effects of microcystins are limited. Quantification of the
absorption, distribution, and elimination of microcystins in humans following oral, inhalation or
dermal exposure is not well understood. The clinical significance in humans for biological
changes observed in experimental animals such as decreased sperm count and motility, and
microscopic lesions in the testes needs further analysis. In animal studies with oral exposures to
microcystins, some adverse effects in males such as reduced testosterone levels, as well as
toxicity to the female reproductive tissues and those of offspring have not been fully
characterized. No data are available to quantify the differences between humans and animals for
the critical health endpoints. There is uncertainty regarding susceptibility and variability in the
human population following exposure to microcystins and the relative toxicity of other
microcystins congeners when compared to microcystin-LR. Additional information is needed on
the potential health risks from mixtures of microcystins with other cyanotoxins, as well as
biological and chemical stressors present in source water and drinking water supplies.

In addition, for short-term advisories (One-day and Ten-day duration), EPA assumes all
exposure is derived from drinking water and, therefore, no Relative Source Contribution (RSC)
term is applied. For lifetime health advisory values, EPA does include an RSC that reduces the
advisory value to account for exposure to other potential sources.
4.8 Susceptibility
Available animal data are not sufficient to determine if there is a definitive difference in
the response of males versus females following oral exposure to microcystins. Fawell et al. (1999)
observed a slight difference between male and female mice in body weight and serum proteins
(ALT and AST), but no sex-related differences in liver pathology.
Studies in laboratory rodents suggest that the acute effects of microcystin-LR may be
more pronounced in adult or aged animals than in juvenile animals (Adams et al., 1985; Ito et al.,
1997a; Rao et al., 2005). In these studies, young animals showed little or no effect at microcystin-
LR doses found to be lethal to adult animals. Age-dependent differences in toxicity were observed
after both oral and i.p. exposure, suggesting that differences in gastrointestinal uptake were not
entirely responsible for the effect of age. The relevance of these age-related differences to acute
toxicity in humans is unknown. However, for infants to one-year olds fed exclusively with
powdered formula prepared with tap water, drinking water is the dominant route of exposure to
cyanotoxins. There are significant differences in exposure between these life-stages that impact
risk.
Based on the available studies in animals, individuals with liver and/or kidney disease
may be more susceptible than the general population since the detoxification mechanisms in the
liver and impaired excretory mechanisms in the kidney may be compromised. Data from an
episode in a dialysis clinic in Caruaru, Brazil where microcystins were not removed by treatment
of dialysis water, identify dialysis patients as a population of potential concern in cases where the
drinking water source for the clinic used to prepare the dialysate is contaminated with
cyanotoxins. Other potentially sensitive individuals include pregnant woman, nursing mothers,
and the elderly.
4.9 Distribution of Body Weight and Drinking Water Intake by Age
Both body weight and drinking water intake are distributions that vary with age. EPA has
developed two health advisory values, a Ten-day HA of 0.3 µg/L based on exposure to infants
over the first year of life, and a Ten-day HA of 1.6 µg/L based on exposure to adults, over 21
years of age. Section 4.10 discusses how EPA recommends application of these values to other
age groups.
The U.S. EPA (2011a) Exposure Factors Handbook provides values for drinking water
ingestion rate and corresponding body weight. The estimated 90th percentile of community water
ingestion for the general population (males and females of all ages) has been used as the default
value for water ingestion. EPA plotted the 90th percentile of drinking water intake using Table 3-
19 for ages ≤3 years, and Table 3-38 for ages >3 years due to sample size in the respective studies.
Age groups <3 months in Table 3-19 were combined due to insufficient sample sizes. Figure 4.1

Figure 4-1. 90th Percentile Drinking Water Ingestion Rates by Age Group
0.25 0.23
90%tile Drinking Water Ingestion Rate (L/kg/day)
0.2
0.148
0.15
0.112
0.1
0.056 0.052
0.043
0.05 0.035
0.026 0.023 0.026 0.034
Adapted from U.S. EPA 2011 Exposure Factors Handbook (U.S.EPA, 2011a).
represents the 90th percentile drinking water ingestion rates (L/kg/day) for each age group (located
on top of the columns). Bottle-fed ages are shown in red (first three columns on the left).
Based on the drinking water intake rates for children <12 months (0.15 L/kg-day), the
exposure of children is over 4 times higher than that of adults >21 years old on a body weight
basis (0.034 L/kg-day). Infants from birth to 3 months may be exclusively bottle-fed and
therefore, have a higher ingestion rate. After 3 months of age, typically around 4 to 6 months of
age, other food and liquids are introduced into the infant diet, lowering the ingestion rate of
drinking water. Drinking water contributes the highest risk of the total cyanotoxin intake for
infants to one-year-olds fed exclusively with powdered formula prepared with tap water
containing cyanotoxins. At the age of 6, children’s intake of drinking water relative to their body
weight is approximately the same as those of an adult (>21 years). Data evaluating the transfer of
microcystins through breast milk are not available for humans.
4.10 Distribution of Potential Health Advisory Values by Age
Using the ingestion rates for each age-group (from Figure 4-1), EPA estimated Ten-day
HA values for microcystins for each age group (plotted on Figure 4-2) to demonstrate the
variability due to body weight and drinking water intake by age.

Figure 4-2. Ten-day Health Advisories for Microcystins by Age Group
EPA decided to apply the Ten-day HA value calculated for infants over the first year of
life (0.3 µg/L) to all bottle-fed infants and young children of pre-school age because these age
groups have higher intake per body weight relative to adults. As Figure 4.2 demonstrates, when
the Ten-day HA is estimated by age group, the calculated HA value for infants from birth to 3
months old is 0.2 µg/L, slightly below the infant health advisory value of 0.3 µg/L. EPA believes
that infants from birth to 3 months old are not at a disproportionate risk at a 0.3 µg/L advisory
value because a 30-fold safety factor is built into this calculation to account for human variability
and deficiencies in the database. The estimated Ten-day HA values for infants from 3 months old
through pre-school age groups (less than 6 years old) are at or above the advisory value of 0.3
µg/L. Therefore, children within these age groups are adequately protected by the advisory value
for bottle-fed infants and young children of pre-school age. EPA decided to apply the adult Ten-
day HA value of 1.6 µg/L to school age children (children older than or equal to 6 years) through
adulthood because children’s intake of drinking water relative to body weight in this age group is
almost the same as those of an adult (>21 years).

5.0 ANALYTICAL METHODS
This Health Advisory (HA) for the Cyanobacterial Microcystin Toxins is applied to total
microcystins which should include all of the measureable microcystin congeners within the
cyanobacterial cells (intracellular) and outside the cell (extracellular).
Extracellular microcystins (either dissolved in water or bound to other materials) typically

make up less than 30% of the total microcystin concentration in source water (Graham et al.,
2010). Most of the toxin is intracellular, and released into the water when the cells rupture or die.
Both intracellular and extracellular microcystins may also be present in treated water, depending
on the type of treatment processes in place. Therefore, it is important to note that analysis for
microcystins should account for both intracellular and extracellular toxins in samples when intact
cells may be present. Release of intracellular microcystins is achieved by rupturing or lysing the
cell walls in order to expose the intracellular microcystins. Cell lysis can be achieved by a variety
of methods including sequential freeze-thawing, freeze drying, and mechanical or sonic
homogenization. Following cell lysis, microcystins may need to be extracted for some analytical
methods. At low concentrations, the direct determination of microcystins may not be feasible, and
a preconcentration step may be required. Typically samples are filtered and/or centrifuged after
cell lysis to remove cell fragments and particulates. This may be followed by freeze-drying or
solid-phase extraction (SPE). Typical elution solvents are dilute acid, methanol, acidified
methanol/water mixtures, and butanol/methanol/water mixtures.
Preconcentration is generally needed when techniques such as liquid chromatography are

used in order to achieve limits of detection in the low-µg/L and ng/L range. Extraction efficiency
has been shown to vary depending on the type of solvent, the hydrophobicity of the congener, the
water content of the cells (freeze-dried versus frozen) and differences between field samples and
laboratory cultures. Variations in extraction efficiency may impact the accurate quantitation of
microcystins so the use of a surrogate compound to monitor the extraction efficiency is strongly
recommended. Responsible authorities should ensure that the appropriate methods and
preparation techniques (extraction, concentration and separation) are being used in the laboratory
depending on the type of sample and the analytical method selected.
Analytical methods available for the detection of microcystins in drinking water include
reversed phase high performance liquid chromatography (HPLC) coupled with mass
spectrometric (MS, MS/MS) or ultraviolet/photodiode array detectors (UV/PDA), Enzyme Linked
Immunosorbent Assays (ELISA), and Protein Phosphatase Inhibition Assays (PPIA).
EPA has developed a liquid chromatography/tandem mass spectrometry (LC/MS/MS)

method for microcystins and nodularin (combined intracellular and extracellular) in drinking
water (Method 544; U.S. EPA, 2015). Accuracy and precision data have been generated in
reagent water, and finished ground and surface waters for the following compounds: microcystin-
LA (microcystin-LA), -LF (microcystin-LF), -LR (microcystin-LR), -LY (microcystin-LY), -RR
(microcystin-RR), -YR (microcystin-YR), and nodularin-R (NOD). This method is intended for
use by analysts skilled in solid phase extractions, operation of LC/MS/MS instruments, and the
interpretation of associated data. The single laboratory lowest concentration minimum reporting
levels (LCMRLs) for this method range from 2.9 to 22 ng/L (0.0029-0.022 µg/L). The Detection
Limit (DL) for analytes in this method range from 1.2 to 4.6 ng/L. In this method, a 500 mL water

sample (fortified with an extraction surrogate) is filtered, and both the filtrate and the filter are
collected. The filter is placed in a solution of methanol containing 20% reagent water and held for
at least one hour at -20 ºC to release the intracellular toxins from cyanobacteria cells captured on
the filter. The liquid is drawn off the filter and added back to the 500-mL aqueous filtrate. The
500-mL sample (plus the intracellular toxin solution) is passed through a SPE cartridge to extract
the method analytes and surrogate. Analytes are eluted from the solid phase with a small amount
of methanol containing 10% reagent water. The extract is concentrated to dryness by evaporation
with nitrogen in a heated water bath, and then adjusted to a 1-mL volume with methanol
containing 10% reagent water. A 10-μL injection is made into an LC equipped with a C8 column
that is interfaced to an MS/MS. Analytes are separated and identified by comparing the acquired
mass spectra and retention times to reference spectra and retention times for calibration standards
acquired under identical LC/MS/MS conditions. The concentration of each analyte is determined
by external standard calibration. To download Method 544 Determination of Microcystins and
Nodularin in Drinking Water by Solid Phase Extraction and Liquid Chromatography/tandem
Mass Spectrometry (LC/MS/MS), please go to: http://www.epa.gov/nerlcwww/ordmeth.htm
High performance liquid chromatography (HPLC) is widely used to separate microcystin

congeners. A variety of stationary phases have been used including reversed-phase C18 columns,
amide C18 columns, internal surface reversed-phase columns or ion exchange columns.
Optimization of chromatographic parameters is needed to ensure a good resolution of analytes. In
addition to mass spectrometry, ultraviolet/visible absorbance is a commonly used detection
techniques with HPLC. Most microcystin congeners have similar absorption profiles between 200
and 300 nm. The wavelength of the UV/visible detector can be set at these values to record the
responses of microcystins in sample extracts separated by the HPLC. The retention time, UV
spectra and peak area of commercially available or laboratory standards is the basis of
identification and quantification of microcystins using HPLC-UV/visible detection. However, due
to the limited number of commercially available standards, the toxins are often quantified by
comparison to an microcystin-LR standard and reported in terms of microcystin-LR equivalence.
HPLC-UV/visible is susceptible to interferences from natural organic materials (NOMs).
Detection limits will depend partially on the sample volume extracted, the concentration of the
toxins, and the presence of interfering contaminants.
A variety of antibodies have been isolated against microcystin-LR and microcystin-RR, as

well as recombinant antibody fragments and antibodies against the amino acid ADDA.
Commercial ELISA kits that contain all of the reagents needed for analysis have also been
developed and typically provide a cross reactivity chart for some of the congeners (i.e.,
microcystin-LR, -RR, YR, nodularin) that are commonly found in water. These range from 50-
85% for microcystin-RR, 35-181% for microcystin-YR and 10-124% for microcystin-LA.
Detection of the total microcystins will be expressed as the sum of the congeners provided from
ADDA ELISA. The methods detection limit (MDLs) of several commercial laboratory ELISA
kits have been reported to range from 0.04 to 0.2 µg/L for microcystin-LR. Commercial ELISA
kits generally have quantitation ranges from 0.2 (LOQ) to an upper limit of 5 µg/L. Two high
sensitivity ELISA plate kits have become commercially available with MDLs ranging from 0.04
to 0.05 µg/L.

PPIAs are used with a variety of detection methods and substrates including radioactive
detection assays using 32P-radiolabelled substrates and colorimetric assays using p-nitrophenol
phosphate as the substrate. The method has also been adopted for fluorescence measurements
using the substrates methylumbelliferyl phosphate. The detection limit of total microcystins,
reported as microcystin-LR equivalents (microcystin-LRequiv) using radiometric protein
phosphatase assays is approximately 0.1 µg/L or less, and using colorimetric PP1 inhibition
assays range between 10 to 20 ng/mL (0.01 to 0.02 µg/L).
Rapid tests for the identification of the presence of microcystins in water have been
developed for use in the field. Field test kits can be used as a presence/absence tool for
determining if a bloom is toxic or if treatment plant operations need to be adjusted during a bloom
event but do not currently have sufficient sensitivity at microcystin concentrations below 1 µg/L
to be used for treated water analyses. Commercially-available test kits use a variety of methods
including immunochromatography (test strips), ELISA, and phosphatase inhibition to estimate the
level of microcystins in a water sample. In general, the results of field test kits should be
considered qualitative and should only be used to conduct a preliminary assessment of
microcystin levels. The applicability of test kits is between 1 and 5 µg/L of microcystins with a
detection limit of approximately 0.5 µg/L. Several field test kits do not include a lysing agent and,
therefore, only determine the presence of extracellular microcystins. When using these field test
kits, users should consult the manufacturer regarding an appropriate lysing technique if the
detection of both intracellular and extracellular microcystins is required.
A new approach using laser diode thermal desorption-atmospheric pressure chemical

ionization interface coupled to tandem mass spectrometry (LDTD-APCI-MS/MS) has been
developed for the analysis of total microcystins in complex environmental matrices. The method
is based on oxidation of the MCs in a sample using potassium permanganate under alkaline
conditions to produce 2-methyl-3-methoxy-4-phenylbutyric acid (MMPB). MMPB is then
extracted and directly injected (no chromatographic separation) into the LDTD-APCI-MS/MS
system. This approach results in ultra-fast sample analysis with simple sample preparation,
reducing time and material costs associated with chromatographic separation. This method does
not require individual MC standards, but similar to ELISA and PPIA, the results do not provide
information on the identity of the individual MC congeners. The MDL and LOQ are 0.2 and 0.9
µg/L, respectively (Roy-Lachapelle et al., 2014).

6.0 TREATMENT TECHNOLOGIES
The information below is adapted from the Health Canada Guidelines for Cyanobacteria
Toxins in Drinking Water, available later in 2015.
Detailed information on the operational considerations of a variety of treatment methods

can be found in the EPA Drinking Water Treatability Database for Microcystins (U.S. EPA,
2007); the International Guidance Manual for the Management of Toxic Cyanobacteria (GWRC,
2009) available at http://www.waterra.com.au/cyanobacteria-
manual/PDF/GWRCGuidanceManualLevel1.pdf, and Management Strategies for Cyanobacteria
(Blue-Green Algae): A Guide for Water Utilities (Newcombe et al., 2010) available at
http://www.researchgate.net/profile/Lionel_Ho/publication/242740698_Management_Strategies_for_Cyan
obacteria_(Blue-Green_Algae)_A_Guide_for_Water_Utilities/links/02e7e52d62273e8f70000000.pdf
For additional information on treatment strategies commonly used or being considered by

water systems vulnerable to cyanotoxins, please see Recommendations for Public Water Systems
to Manage Cyanotoxins in Drinking Water (U.S. EPA, 2015b).
6.1 Management and Mitigation of Cyanobacterial Blooms in Source Water
Algaecides can be applied to lakes and reservoirs to mitigate algal blooms, including
cyanobacteria. In most cases, depending on the cyanobacteria species present, the application of
algaecides has the potential to compromise cell integrity releasing cyanotoxins into the source
waters. Chemical treatment to control blooms in drinking water sources in the early stages of the
bloom when cyanobacterial concentrations are still relatively low (usually under 5,000 to 15,000
cells/mL) (WHO, 1999), are less likely to release significant cyanotoxin concentrations upon cell
lysis and is able to mitigate or prevent a cyanobacterial bloom from proliferating as the season
progresses. If harmful cyanobacterial blooms occur, utilities may take action to investigate
alternative source water sources, change intake locations or levels to withdraw source water with
minimal cyanotoxin concentrations, or investigate methods of destratification in the water source.
Purchasing water from a neighboring interconnected water system that is unaffected by the bloom
may also be an option for some systems.
Clays and commercial products such as aluminum sulfate (alum) have been used for the
management of blooms in source waters. Alum treatment efficiency depends on the alum dose
and the type of flocculant. Aeration and destratification have also been used to treat
cyanobacterial blooms, usually in smaller water bodies (from one acre to several tens of
acres). Active mixing devices, diffuse air bubblers, and other means of reducing stratification
have proven to be effective in controlling outbreaks and persistence of blooms in relatively small
shallow impoundments (around <20 feet deep). These strategies can be applied to the entire
source water body or to just a portion of the lake depending on the need, and size and depth of the
water body relative to the source water intake(s).
Hydrogen peroxide (H2O2) has been used as an algaecide in source water because of a
rapid reaction time (90% of bloom collapsed in 3 days and 99% in 10 days), and environmentally
safe reaction products (oxygen and water) (Wang et al., 2012; Matthjis et al; 2012). The

drawbacks (aside from cell lysis) are that oxidant breakdown is so rapid that repeated applications
are needed. Further understanding of this technique is needed (Matthjis et al., 2012).
The use of ultrasonic sound waves to disrupt cyanobacterial cells has also been
investigated as a potential source water treatment option (Rajasekhar et al., 2012). It is
environmentally friendly compared to chemical treatment strategies. The technique has also been
reported to be capable of degrading microcystin-LR (Song et al., 2005). Drawbacks include that
application frequencies are difficult to calculate and are system-specific; and that applications on
large scale require more powerful and, therefore, more expensive equipment. Sonication shows
potential for use in cyanobacterial bloom management, but further study to determine effective
operating procedures is needed before it can be considered as a feasible approach (Rajasekhar et
al., 2012).
Excess nutrients are thought to be a primary driver of cyanobacterial blooms. Long-term

prevention of cyanobacterial blooms likely requires reductions in nutrient pollution. Excess
nitrogen and phosphorus in aquatic systems can stimulate blooms and create conditions under
which harmful cyanobacteria thrive. Thus, managing nutrient pollution sources within a
watershed in addition to waterbody-specific physical controls (in systems that are amenable to
those controls) tends to be the most effective strategy. Nutrient pollution can be from urban,
agricultural, and atmospheric sources and, therefore, reductions can be achieved through a variety
of source control technologies and best management practices.
6.2 Drinking Water Treatment
Effective treatment of cyanotoxins in drinking water includes the evaluation and selection
of appropriate treatment methods. Any variation in treatment methods aimed at reducing toxins
concentrations need to be tailored to the type(s) of cyanobacteria present and the site-specific
water quality (e.g. pH, temperature, turbidity, presence of natural organic material (NOM), etc.),
the treatment processes already in place, and the utility’s multiple treatment goals (e.g., turbidity
and total organic carbon (TOC) removal, disinfection requirements, control of disinfection by-
products (DBP) formation, etc.). Utilities need to have an understanding of the type and
concentration of cyanotoxins present in the source water and should conduct site-specific
evaluations such as jar testings and piloting in order to determine the most effective treatment
strategy. Potential target parameters include: chlorophyll-a, turbidity, cyanobacterial cells and
extracellular and intracelluar toxins.Care should be taken to avoid cell lysisTo remove both
intracellular and extracellular toxins from drinking water, a multi-barrier approach is required,
which may consist of conventional filtration for intracellular cylindrospermopsin removal and
additional processes such as activated carbon, biodegradation, advanced oxidation, and small-pore
membrane processes (e.g. nanofiltration and reverse osmosis), for the removal or oxidation of
extracellular cylindrospermopsin. The most effective way to deal with cyanobacteria cells and
their toxins, is to remove the cells intact, without damaging them, to prevent the release
of additional extracellular toxins into the water.

6.2.1 Conventional Treatment for Microcystins
In the absence of cell damage, conventional treatment employing coagulation,

flocculation, clarification (sedimentation or dissolved air flotation) and rapid granular filtration
can be effective at removing intact cells and the majority of intracellular toxins (cell bound)
(Chow et al., 1998; Newcombe et al., 2015). However, if toxins are released into solution, a
combination of conventional treatment processes with oxidation, adsorption, and/or advanced
treatment needs to be considered to treat both intracellular and extracellular cyanotoxins.
The efficiency of the conventional treatment processes to remove cyanobacterial cells and
intracellular microcystins has been shown to vary from 60 to 99.9%. Factors that impact removal
include the cyanobacterial species and cell density, coagulant type and dose, pH, NOM, and
operational parameters such as flocculation time, frequency of filter backwashing and clarifier
sludge removal (Vlaski et al., 1996; Hoeger et al., 2004; Jurczak et al., 2005; Zamyadi et al.,
2012a, 2013c; Newcombe et al., 2015). Typically, 60 to 95% of cells and intracellular
microcystins can be removed during sedimentation with as much as 99.9% removal achieved
through filtration (Lepisto et al., 1994; Drikas et al., 2001; Hoeger et al., 2004; Newcombe et al.,
2015). The efficiency of coagulation and clarification for cell removal is dependent on pH,
coagulant type and dose and the morphological characteristics of the cyanobacteria. Rapid sand
filtration without pre-treatment (i.e., direct filtration, without coagulation/clarification) is not
effective for cyanobacterial cell removal.
If operated properly, conventional treatment (coagulation, flocculation, clarification and

filtration), does not cause cell lysis or increases in the extracellular microcystin concentrations of
treated water (Chow et al., 1998, 1999; Drikas, 2001; Sun et al., 2012). Drinking water treatment
plants utilizing conventional treatment followed by oxidation or activated carbon may remove
both intracellular and extracellular microcystins up to 99.99% of total microcystins to achieve
concentrations below 0.1 µg/L in treated water (Karner et al., 2001; Lahti et al., 2001; Hoeger et
al., 2005; Jurczak et al., 2005; Rapala et al., 2006; Zamyadi et al., 2013a). Conventional treatment
is generally considered to have limited effectiveness for the removal of the extracellular
microcystins. Therefore, additional processes such as adsorption, chemical oxidation,
biodegradation or reverse osmosis, and nanofiltration are required to remove extracellular
microcystins.
Although microfiltration and ultrafiltration membranes can remove both cyanobacterial

cells and intracellular microcystins, removal of extracellular microcystins using ultrafiltration is
variable (35 to 70%) and microfiltration is not effective (Gijsbertsen-Abrahamse et al., 2006;
Dixon et al., 2011a, b). Nanofiltration and reverse osmosis membranes can achieve high removals
of intracellular and extracellular microcystins, from 82% to complete removal (Westrick et al.,
2010; Dixon et al., 2010). Pore size, among others, is an important factor in removal efficiency
for these processes.
Successful removal of cyanobacterial cells and intracellular microcystins will depend on

proper operations of the conventional treatment processes (Hoeger et al., 2004; Dugan and
Williams, 2006; Ho et al., 2013; Zamyadi et al., 2012a, 2013c). Operational considerations for
removing cyanobacterial cells using coagulation, flocculation and clarification are similar to

considerations for achieving effective particle removal. The appropriate coagulant and
coagulation pH should be determined through jar-testing to maximize cell removal. In jar-testing,
the NOM, chlorophyll-a, or cyanobacterial cell count can be used to optimize the coagulation
conditions for cyanobacterial cell removal (Sklenar et al., 2014; Newcombe et al., 2015).
Sufficient mixing should be provided at the point of chemical addition to ensure rapid and
uniform contact, and an appropriate mixing speed should be determined to optimize the
flocculation process (GWRC, 2009). It is important to minimize the potential for the
accumulation of cyanobacterial cells as scums at the surface of sedimentation basins and filters
(Zamyadi et al., 2012a, 2013c).
Effective sludge removal from sedimentation/clarification processes is important to

minimize the release of intracellular and extracellular microcystins into the surrounding waters, as
significant cell numbers can accumulate within the sludge, and cells contained within the sludge
can lyse rapidly (Drikas et al., 2001; Ho et al., 2013; Zamyadi et al. 2012a). It has been reported
that accumulation of cyanobacterial cells and microcystins in clarifiers can lead to their
breakthrough into filter effluent. In addition, cell lysis can occur in the clarifier sludge, increasing
the extracellular concentration of microcystins in the treatment plant. Therefore sludge
management (decreased sludge age) in clarifiers and increased frequency of backwashing of
filters is important because settled/filtered cells can remain viable and possibly multiply over a
period of at least 2 to 3 weeks. Within 1 day, some cells in the sludge can lyse and release NOM
and taste and odor compounds, in addition to cyanotoxins (Newcombe et al., 2015). Additionally,
backwash water from the filters may contain cyanobacterial cells and/or extracellular
microcystins; hence, care needs to be taken if spent backwash water is recirculated to the
beginning of the treatment process to prevent the reintroduction of cells and toxins into the
treatment train. Although longer filter run-times are typically desirable between backwashing,
during periods of high algal concentrations, cells can accumulate in the filter, which can
potentially lead to a significant amount of extracellular microcystins released into the filtered
water. The optimum balance between maximizing water production and minimizing the risk of
toxin breakthrough will be plant-specific.
6.2.2 Adsorption
Adsorption processes, such as granular activated carbon (GAC) or powdered activated

carbon (PAC), are effective at removing extracellular microcystins but are not capable of
removing intact cells and intracellular toxins (Lambert et al., 1996; Newcombe, 2002; Newcombe
et al., 2003). Removal through adsorption depends on many factors including the type of activated
carbon used, the microcystin congener and water quality conditions. In general, mesoporous
carbons (such as chemically-activated wood-based carbons) are the most effective for the removal
of microcystins (Newcombe et al., 2010). Other factors such as the type of microcystin congener
present, the raw water quality (i.e., NOM and pH) and contact time affect microcystins removal
efficiency when using activated carbon processes. In addition, shortened filter run times or filter
overload may happen during cyanobacteria blooms. Therefore, water treatment plants should
conduct jar-testing to determine the most effective activated carbon dose, type, and feed point
prior to the application without affecting other water quality parameters and treatment processes
(Sklenar et al., 2014).

The performance of GAC filtration for extracellular microcystin removal depends upon
the empty bed contact time (EBCT), carbon age, carbon pore size, and raw water quality
characteristics such as NOM and pH, as well as the microcystin variant (Newcombe, 2002;
Newcombe et al., 2003; Ho and Newcombe, 2007; Wang et al., 2007). Solution chemistry can
also affect microcystin-LR adsorption onto GAC. Enhanced removal of microcystin-LR has been
observed at lower pH (2.5 versus 6.5) due to either precipitation or reduced solvency effect
(Pendleton et al., 2001).
Removal of extracellular microcystins by PAC can be highly effective (up to 95%)

depending on the microcystin congener and concentration, the PAC type and dose, the contact
time and the water quality characteristics such as TOC (Newcombe et al., 2003; Cook and
Newcombe; 2008; Ho et al., 2011). According to Newcombe et al. (2010), a PAC dose of 20
mg/L and a contact time of at least 45 minutes should be considered for removal of most
extracellular microcystins (with the exception of microcystin-LA).
6.2.3 Chemical Oxidation
Chemical oxidation using chlorine, potassium permanganate, or ozonation can be effective

at oxidizing extracellular microcystins, but can also impaired cell integrity, resulting in an
increase in concentrations of extracellular microcystins in drinking water. By applying
conventional filtration (or other filtration process) first to remove the majority of intact cells, the
extracellular microcystin concentration is less likely to increase due to cell lysis when water is
treated with oxidants. In cases where pre-oxidation (oxidant applied anywhere along the treatment
process prior the filter influent) is practiced, it may need to be discontinued during an algal bloom
or adjustments to the oxidant type and doses may be needed to minimize cell rupture prior to
filtration (Newcombe et al., 2015).
The effectiveness of chemical oxidation of microcystins depends on the type of oxidant,

dose, contact time, microcystin congener and water quality characteristics such as pH and
dissolved organic carbon (DOC) (GWRC, 2009; Sharma et al., 2012). Laboratory-scale
experiments have demonstrated that the general trend for the effectiveness of cyanobacterial cell
and extracellular microcystin oxidation to be: ozone>permanganate>chlorine>chlorine-based
oxidants (Acero et al., 2005; Rodriguez et al., 2007a, b; Ding et al., 2010; Sharma et al., 2012;).
However, selection of the most appropriate oxidant for microcystins should be based on the
characteristics of each water source, the disinfection requirements, and potential formation of
disinfection by-products (DBPs) (Sharma et al., 2012).
It is also important to recognize that the use of oxidants may result in the formation of
DBPs and should be considered when selecting a strategy for oxidizing microcystins (Merel et al.,
2010; Zamyadi et al., 2012b; Wert et al., 2013). For example, ozone and chlorine dioxide can
result in the formation of inorganic DBPs, such as bromate and chlorite/chlorate, respectively.
Additionally, modifying pre-oxidation practices may compromise other treatment objectives (e.g.,
turbidity removal), and should be considered.

The oxidation of microcystins by chlorine has been found to be highly effective (>90%
removal) under experimental conditions (Ho et al., 2006a; Acero et al., 2008; Merel et al., 2009;
Sorlini and Collivignarelli, 2011). However, the effectiveness of chlorination on the oxidation of
microcystins depends upon the chlorine dose, contact time, pH, temperature, and other water
quality characteristics (Sharma et al., 2012). Several studies have found that microcystins are
efficiently oxidized if pH is maintained below 8, the chlorine dose is greater than 3 mg/L and 0.5
to 1.5 mg/L of free chlorine residual is present after 30 minutes of contact time (Nicholson et al.,
1994; Acero et al., 2005; Ho et al., 2006a; Xagoraraki et al., 2006; Newcombe et al., 2010).
However, much higher chlorine doses (2 to 10 mg/L) are required to lyse the cyanobacterial cell
and then oxidize the previously cell-bound microcystins (Zamyadi et al., 2013b).
The oxidation of microcystins in water by permanganate is one of the more effective

processes for oxidizing extracellular microcystins in water (Sharma et al., 2012). Rodriguez et al.
(2007a) exhibited a 90% oxidation of microcystin-LR at a dose of 1.0 mg/L, a contact time of 60
minutes, a pH of 8, and a temperature of 20°C. Complete oxidation occurred at a dose of 1.5
mg/L (Rodriguez et al., 2007a). Treatment plants considering potassium permanganate for
oxidation of microcystins should be aware that permanganate can discolor water when it is
present in excess of 0.05 mg/L. Therefore, dosage control and process monitoring (e.g., visual
inspection of the basin effluent color, measuring permanganate residual) is important in avoiding
consumer complaints (MWH, 2012).
The oxidation of microcystins in water by ozone has been shown to be highly effective
(greater than 90% oxidation) in laboratory-scale studies (Rositano et al., 2001; Shawwa and
Smith, 2001; Brooke et al., 2006). The efficacy depends on temperature, pH, ozone dose, contact
time, and other water quality characteristics such as DOC and alkalinity (Sharma et al., 2012).
Utilities should also be aware that the use of ozone may result in the formation of bromate and
other DBPs.
Monochloramine is a weaker oxidant than chlorine and is not an effective treatment

barrier for microcystins (Westrick et al., 2010).
Most laboratory studies have found that chlorine dioxide (ClO2) is not effective for
oxidizing extracellular microcystins (Kull et al., 2004, 2006; Ding et al., 2010; Sorlini and
Collivignarelli, 2011) or cyanobacterial cells and intracellular microcystins (Ding et al., 2010;
Wert et al., 2014) at dosages (1-2 mg/L) and contact times typically applied to drinking water.
6.2.4 Other Filtration Technologies
Biological filtration, using either biologically-active sand or activated carbon, has been
shown to be effective for the removal of extracellular microcystins in bench- and pilot-scale
studies (Keijola et al., 1998; Bourne et al., 2006; Ho et al., 2006b, 2008, 2012) and in limited full-
scale studies (Grutzmacher et al., 2002, Rapala et al., 2006). The removal of intact cyanobacterial
cells and their associated intracellular toxins through physical straining in slow sand filters has
also been documented (Grutzmacher et al., 2002; Pereira et al., 2012). Biological filters also have
the capability to remove particulate including intact cyanobacterial cells. Bank filtration may also
be effective for the removal of microcystins (Lahti et al., 1998; Schijven et al., 2002). A detailed

review of biological treatment options for cyanotoxin removal conducted by Ho et al., (2012b)
identified the type and concentration of microcystin-degrading bacteria, concentration of
microcystins, and temperature as key factors that influence the efficiency of biological filtration
for the removal of microcystins. In addition, the concentration of other organic matter within the
source water may inhibit biodegradation, as microcystins may be a secondary substrate in the
presence of NOM. Particle size, chemical composition and roughness or topography of the
surface of the media used for filtration have also been identified as important factors for biofilm
growth and ultimately the biodegradation of microcystins (Wang et al., 2007, Ho et al., 2012).
Membrane filtration including microfiltration (MF) and ultrafiltration (UF) can achieve
greater than 98% removal of cyanobacterial cells and intracellular microcystins (Chow et al.,
1997; Gijsbertsen-Abrahamse et al., 2006; Campinas and Rosa, 2010: Sorlini et al., 2013).
Nanofiltration (NF), reverse osmosis (RO) and, to a lesser extent UF, can be used for both
intracellular and extracellular microcystin removal (Neumann and Weckesser, 1998; Lee and
Walker, 2008; Dixon et al., 2011a,b). The performance of membrane filtration for microcystin
removal depends on characteristics of the membrane such as molecular weight cut-off (MWCO)
and hydrophobicity, initial concentration, size and molecular weight of the microcystins, and
operating parameters such as flux, recoveries and degree of fouling. It is recommended that
cyanobacterial cells are removed prior to reverse osmosis to prevent membrane clogging and
fouling.
Laboratory and pilot-scale studies have demonstrated that MF and UF can remove greater
than 98% of cyanobacterial cells (Chow et al., 1997, Gijsbertsen-Abrahamse et al., 2006;
Campinas and Rosa, 2010; Sorlini et al., 2013), and ultrafiltration can be moderately effective
(35-70%) for removal of extracellular microcystins (Lee and Walker, 2008). Several studies have
also demonstrated that the release of intracellular microcystins from the shear stress on
cyanobacterial cells during MF and UF is possible, although it generally results in permeate
microcystin concentration increases of less than 12 percent (Gijsbertsen-Abrahamse et al., 2006;
Campinas and Rosa, 2010).
The removal of extracellular microcystins by NF and RO is very effective (greater than

90%) and depends on the MWCO, as the filtration of microcystins occurs via size exclusion
(Gijsbertsen-Abrahamse et al., 2006).
6.2.5 Combined Treatment Technologies
In practice, full-scale treatment plants use a combination of treatment technologies (i.e.,

conventional filtration and chemical oxidation) in order to remove both intracellular and
extracellular microcystins. Data indicate that utilities can effectively remove both intracellular and
extracellular microcystins to achieve concentrations below 0.1 µg/L (Lahti et al. 2001; Boyd and
Clevenger, 2002; Zurawell, 2002; Hoeger et al., 2005, Jurczak et al., 2005; Rapala et al., 2006;
Haddix et al., 2007; Nasri et al., 2007; Zamyadi et al., 2013c). However, some studies have
shown that the presence of high concentrations of cells (i.e., 105 cells/mL) and/or microcystins in
raw water (100 µg/L) may be challenging for treatment plants to reduce concentrations to below
0.1 µg/L (Tarczyriska et al., 2001; Zamyadi et al., 2012a).

In most cases, utilities will be able to effectively remove intracellular microcystins with
processes that are already in place (e.g., conventional treatment) when they are operated with a
focus on cyanobacteria cell or NOM removal. Extracellular microcystins may also be removed in
many treatment plants by using existing treatment such as chlorination after filtration or by the
addition of PAC following conventional treatment (Carriere et al., 2010). Although it is possible
to remove both intracellular and extracellular microcystins effectively using a combination of
treatment processes, the removal efficiency can vary considerably. Utilities need to ensure that
they are utilizing their existing treatment processes to their fullest capacity for removal of both
cyanobacterial cells and extracellular microcystins and that appropriate monitoring is being
conducted to ensure that adequate removal is occurring at each step in the treatment process.
6.3 Point-of-Use (POU) Drinking Water Treatment Units
Limited information is available on residential treatment units for the removal of

cyanobacteria cells and microcystins. A study using common water filtration and purification
systems found that the efficacy of POU filtration devices to remove microcystin (LR) varies
considerably with the type of device being used (Pawlowicz et al., 2006). Microcystin-LR was
successfully removed using carbon filters allowing only 0.05 to 0.3% of the toxin load to pass
through the filter. However, more than 90% of microcystin-LR passed through string-wound
filters and pleated paper. According to the authors, the use of carbon home filter devices tested in
this study may provide additional protection beyond that provided by the drinking water treatment
plant against human exposure to microcystin-LR. Additional studies are recommended to assess
the efficacy of POU drinking water treatment units for other cyanotoxins and under other
conditions. Third-party organizations are currently developing certification standards to test POU
devices to evaluate how well they remove cyanotoxins from drinking water treatment units. Those
standards are expected in the near future.
More information about treatment units and the contaminants they can remove can be found at
http://www.nsf.org/Certified/DWTU/.

7.0 REFERENCES
Acero, J. L., Rodriguez, E., Meriluoto, J. (2005). Kinetics of reactions between chlorine and the
cyanobacterial toxins microcystins. Water Research, 39(8): 1628-1638.
Acero, J. L., Rodriguez, E., Majado, et al. (2008). Oxidation of microcystin with chlorine and
permanganate during drinking water treatment. Journal of Water Supply: Research and
Technology, 57(6): 371-380.
Adams, W. H., Stoner, R. D., Adams, D. G., et al. (1985). Pathophysiologic effects of a toxic
peptide from Microcystis aeruginosa. Toxicon. 23(3): 441-447.
AWWA Research Foundation. (2001). Assessment of Blue-Green Algal Toxins in Raw and
Finished Drinking Water. Final report #256. Prepared by Carmichael, W. W. AWWA
Research Foundation and American Water Works Association. Denver, CO.
Barford, D., Das, A. K., and Egloff, M. (1998). The structure and mechanism of protein
phosphatases: Insights into catalysis and regulation. Annual Review of Biophysics and
Biomolecular Structure, 27: 133-164.
Beussink, A. M., and Graham, J. L. (2011). Relations between hydrology, water quality, and
taste-and-odor causing organisms and compounds in Lake Houston, Texas, April 2006–
September 2008. U.S. Geological Survey Scientific Investigations Report 2011-5121,
pp.27. Reston, VA.
Bourne, D. G., Blakeley, R. L., Riddles, P., and Jones, G. J. (2006). Biodegradation of the
cyanobacterial toxin microcystin-LR in natural water and biologically active slow sand
filters. Water Research, 40(6): 1294-1302.
Boyd, E. and Clevenger, T. (2002). Removal of cyanobacterial toxins in Midwestern States. In:
Proceedings of the Water Quality and Technology Conference. American Water Works
Association, Denver, CO.
Boyer, G. L. (2007). Cyanobacterial toxins in New York and the Lower Great Lakes ecosystems.
In: H. K. Hudnell, (Ed.), Proceedings of the Interagency, International Symposium on
Cyanobacterial Harmful Algal Blooms (ISOC-HAB): State of the Science and Research
Needs, Advances in Experimental Medicine and Biology. Springer Press, New York, NY.
pp 151-163.
Brooke, S., Newcombe, G., Nicholson, B., and Klass, G. (2006). Decrease in toxicity of
microcystins LA and LR in drinking water by ozonation. Toxicon, 48(8): 1054-1059.
Brooks, W. P. and Codd, G. A. (1987). Distribution of Microcystis aeruginosa peptide toxin and
interactions with hepatic microsomes in mice. Pharmacology and Toxicology, 60(3): 187-
191. (Cited in WHO 1999).

Burns, J., (2008). Toxic cyanobacteria in Florida waters. In: In: H. K. Hudnell, (Ed.), Proceedings
of the Interagency, International Symposium on Cyanobacterial Harmful Algal Blooms
(ISOC-HAB): State of the Science and Research Needs, Advances in Experimental
Medicine and Biology. Chapter 5. Springer Press, New York, NY. pp.139-152.
Campinas, M. and Rosa, M. J. (2010). Evaluation of cyanobacterial cells removal and lysis by
ultrafiltration. Separation and Purification Technology, 70(3): 345-352.
Carey, C. C., Ibelings, B. W., Hoffmann, E. P., et al. (2012). Eco-physiological adaptations that
favour freshwater cyanobacteria in a changing climate. Water Research, 46(5): 1394-
1407.
Carmichael, W. W. (1992). A Status Report on Planktonic Cyanobacteria (Blue Green Algae) and
their Toxins. EPA/600/R-92/079, Environmental Monitoring Systems Laboratory, Office
of Research and Development, U.S. Environmental Protection Agency, Cincinnati, OH.
(Cited in WHO 1999).
Carmichael, W. W., Beasley, V., Bunner, D. L., et al. (1988). Naming of cyclic hepatapeptide
toxins of cyanobacteria (blue-green algae). Toxicon 26(11): 971-973.
Carmichael W. W., Azevedo S. M., An J. S., et al. (2001). Human fatalities from cyanobacteria:
chemical and biological evidence for cyanotoxins. Environmental Health Perspectives
109(7): 663-668
Carriere, A., Prevost, M., Zamyadi, A., et al. (2010). Vulnerability of Quebec drinking-water
treatment plants to cyanotoxins in a climate change context. Journal of Water and Health,
8(3): 455-465.
Castenholz, R. W. and Waterbury, J. B. (1989). In: J. T. Staley, M. P. Bryant, N. Pfennig and J.

G. Holt (Eds.), Bergey's Manual of Systematic Bacteriology. Vol. 3, Williams & Wilkins,
Baltimore, MD. pp. 1710-1727. (Cited in WHO 1999).
Chen, Y., Xu, J., Li, Y., and Han, X. (2011). Decline of sperm quality and testicular function in
male mice during chronic low-dose exposure to microcystin-LR. Reproductive
Toxicology, 31(4): 551-557.
Cheng, X., Maher, J., Dieter, M. Z., and Klaassen, C. D. (2005). Regulation of mouse organic
anion-transporting polypeptides (OATPs) in liver by prototypical microsomal enzyme
inducers that activate transcription factor pathways. Drug Metabolism Disposition, 33(9):
1276-1282.
Chernoff, N., Hunter, E. S., III, Hall, L. L, et al. (2002). Lack of teratogenicity of microcystin-LR
in the mouse and toad. Journal of Applied Toxicology, 22(1): 13-17.
Christensen, V. G., Maki, R. P., and Kiesling, R. L. (2011). Relation of Nutrient Concentrations,
Nutrient Loading, and Algal Production to Changes in Water Levels in Kabetogama Lake,

Voyageurs National Park, Northern Minnesota, 2008–09. U.S. Geological Survey Scientific
Investigations Report 2011–5096, 50 p.
Christoffersen, K., Lyck, S., and Winding, A. (2002). Microbial activity and bacterial community
structure during degradation of microcystins, Aquatic Microbial Ecology, 27: 125-136.
Chow, C. W. K., Panglisch, S., House, J., et al. (1997). A study of membrane filtration for the
removal of cyanobacterial cells. Journal of Water Supply: Research and Technology,
46(6): 324-334.
Chow, C., Drikas, M. and Ho, J. (1999). The impact of conventional water treatment processes on
cells of the cyanobacterium Microcystis aeruginosa. Water Research, 33(15): 3253-3262.
Chow, C. W. K., House, J., Velzeboer, R. M. A., et al. (1998). The effect of ferric chloride
flocculation on cyanobacterial cells. Water Research, 32(3): 808-814.
Chorus, I (Ed.). (2012). Current Approaches to Cyanotoxin Risk Assessment, Risk Management
and Regulations in Different Countries. Federal Environment Agency. Dessau-Roβlau,
Germany. Online publication: www.uba.de/uba-info-medien-e/4390.html
Chorus, I., Falconer, I., Salas, H. S., and Bartram, J. (2000). Health risks caused by freshwater
cyanobacteria in recreational waters. Journal of Toxicology and Environmental Health-
Part B, Critical Reviews, 3(4): 323-347.
Codd, G. A., Morrison, L. F., and Metcalf, J. S. (2005). Cyanobacterial toxins: risk management
for health protection. Toxicology and Applied Pharmacology, 203(3): 264-272.
Cook, D. and Newcombe, G. (2008). Comparison and modeling of the adsorption of two
microcystin analogues onto powdered activated carbon. Environmental Technology, 29(5):
525-534.
Cote, L-M., Lovell, R. A., Jeffrey, E. H., et al. (1986). Failure of blue-green algae (Microcystis
aeruginosa) hepatotoxin to alter in vitro mouse liver enzymatic activity. Journal of
Toxicology Toxin Reviews, 52(2): 256
Cousins, I. T., Bealing, D. J., James, H. A., and Sutton, A. (1996). Biodegradation of microcystin-
LR by indigenous mixed bacterial populations. Water Research, 30(2): 481-485. (Cited in
WHO 1999).
Craig, M., Luu, H. A., McCready, T. L., et al. (1996). Molecular mechanisms underlying the
interaction of motuporin and microcystins with type-1 and type-2A protein phosphatases.
Biochem. Cell Biology, 74(4): 569-578.
de la Cruz, A., Antoniou, M., Hiskia, A., et al. (2011). Can We Effectively Degrade
Microcystins? - Implications on Human Health. Anti-Cancer Agents in Medicinal
Chemistry, 11:19-37.

Ding, J. Shi, H., Timmons, T., et al. (2010). Release and removal of microcystins from
Microcystis during oxidative, physical and UV-based disinfection. Journal of
Environmental Engineering, 136(1): 2-11.
Ding, W. X. and Ong, C. N.(2003). Role of oxidative stress and mitochondrial changes in
cyanobacteria-induced apoptosis and hepatotoxicity. FEMS Microbiology Letters, 220(1):
1-7.
Dixon, M. B., Falconet, C., Ho, L., et al. (2010). Nanofiltration for the removal of algal
metabolites and the effects of fouling. Water Science Technology 61(5): 1189-99.
Dixon, M. B., Falconet C, Ho L., et al.(2011a). Removal of cyanobacterial metabolites by

nanofiltration from two treated waters. Journal of Hazardous Materials, 188(1-3): 288-95.
Dixon, M. B., Richard, Y., Ho, L., et al. (2011b). A coagulation-powdered activated carbon
ultrafiltration, multiple barrier approach for removing toxins from two Australian
cyanobacterial blooms. Journal of Hazardous Materials, 186(2-3): 553-1559.
Drikas, M., Chow, C., Christopher, W. K., et al. (2001). Using coagulation, and settling to remove
toxic cyanobacteria. Journal American Water Works Association, 93(2): 100-111.
Dugan, N. R. and Williams, D. J. (2006). Cyanobacteria passage through drinking water filters
during perturbation episodes as a function of cell morphology, coagulant and initial filter
loading rate. Harmful algae, 5: 26-35.
Duy, T. N., Lam, P. K. S., Shaw, G. R., and Connell, D. W. (2000). Toxicology and risk
assessment of freshwater cyanobacterial (blue-green algal) toxins in water. Reviews of
Envrionemental Contamination and Toxicology, 163: 113-186.
Edwards C., Graham, D., Fowler, N., and Lawton, L. A. (2008). Biodegradation of microcystins
and nodularin in freshwaters. Chemosphere, 73(8):1315–1321.
Falconer, I. R. (1998). Algal toxins and human health. In: J. Hubec (Ed.), Handbook of
Environmental Chemistry, Vol. 5, Part C, Quality and Treatment of Drinking Water. pp.
53-82.
Falconer, I. R., Beresford, A. M., and Runnegar, M. T. C. (1983). Evidence of liver damage by
toxin from a bloom of the blue-green alga, Microcystis aeruginosa. Medical Journal of
Australia, 1(11): 511-514.
Falconer, I. R., Buckley, T., and Runnegar, M. T. C. (1986). Biological half-life organ
distribution and excretion of 125I-labeled toxic peptide from the blue-green alga
Microcystis aeruginosa. Australian journal of biological sciences, 39(1): 17-21.

Falconer, I. R., Burch, M. D., Steffense, M., et al. (1994). Toxicity of the blue-green alga
(Cyanobacterium) Microcystis aeruginosa in drinking water to growing pigs, as an animal
model for human injury and risk assessment. Environmental Toxicology and Water
Quality, 9(2): 131-139.
Falconer, I.R. (2005). Cyanobacterial Toxins of Drinking Water Supplies: Cylindrospermopsins

and Microcystins. CRC Press Boca Raton, Florida. 263p
Fawell, J. K., Mitchell, R. E., Everett, D. J., and Hill, R. E. (1999). The toxicity of cyanobacterial
toxins in the mouse. 1. Microcystin-LR. Human and Experimental Toxicology, 18(3): 162-
167.
Fay, P. (1965). Heterotrophy and nitrogen fixation in Chlorogloea fritschii. Journal of General
Microbiology, 39: 11-20. (Cited in WHO 1999).
Feurstein, D., Kleinteich, J., Heussner, A. H., et al. (2010). Investigation of Microcystin
Congener–Dependent Uptake into Primary Murine Neurons. Environmental Health
Perspectives, 118(10): 1370-1375.
Fischer, W. J., Altheimer, S., Cattori, V., et al.(2005). Organic anion transporting polypeptides
expressed in liver and brain mediate uptake of microcystin. Toxicology and Applied
Pharmacology, 203(3): 257-263.
Fleming, L. E., Rivero, C., Burns, J., et al. (2002). Blue green algal (cyanobacterial) toxins,
surface drinking water and liver cancer in Florida. Harmful Algae, 1(2): 157-168.
Fleming, L. E., Rivero, C., Burns, J., et al. (2004). Cyanobacteria exposure, drinking water and
colorectal cancer. In: K. A. Steidinger, J. H. Landsberg, C. R. Tomas and G. A.Vargo,
(Eds.), Harmful Algae 2002. Proceedings of the Xth International Conference on Harmful
Algae. Florida Fish and Wildlife Conservation Commission and Intergovernmental
Oceanographic Commission of UNESCO, Tallahassee, FL. p. 470-472.
Fu, W. Y., Chen, J. P., Wang, X. M. and Xu, L. H. (2005). Altered expression of p53, Bcl-2 and
Bax induced by microcystin-LR in vivo and in vitro. Toxicon, 46(2): 171-177.
Funari, E. and Testai, E. (2008). Human Health Risk Assessment Related to Cyanotoxins
Exposure. Critical Reviews in Toxicology, 38: 97-125.
Giannuzzi, L., Sedan, D., Echenique R., and Andrinolo, D. (2011). An Acute Case of Intoxication
with Cyanobacteria and Cyanotoxins in Recreational Water in Salto Grande Dam,
Argentina. Marine Drugs, 2011(9): 2164-2175.
Gijsbertsen-Abrahamse, A. J., Schmidt, W., Chorus, I., and Heijman, S. G. J. (2006). Removal of
cyanotoxins by ultrafiltration and nanofiltration. Journal of Membrane Science, 276: 252-
259.

Graham, J. L., Ziegler, A. C., Loving, B. L., and Loftin, K. A.(2012). Fate and Transport of
Cyanobacteria and Associated Toxins and Taste-and-Odor Compounds from Upstream
Reservoir Releases in the Kansas River, Kansas, September and October 2011. U.S.
Geological Survey Scientific Investigations Report 2012-5129, pp. 65.
Graham, J., Loftin, K., Meyer, M., and Ziegler, A. (2010). Cyanotoxin mixtures and taste-and-
odor-compounds in cyanobacterial blooms from the midwestern United States.
Environmental Science and Technology, 44: 7361-7368.
Grutzmacher, G. Bottcher, G., Chorus, I., and Bartel, H. (2002). Removal of microcystins by slow
sand filtration. Environmental Toxicology, 17(4): 386-394.
Global Water Research Coalition (GWRC). (2009). International Guidance Manual for the
Management of Toxic Cyanobacteria. G. Newcombe (Ed.). Global Water Research
Coalition and Water Quality Research Australia. London, United Kingdom.
Gupta, Nidhi, et al. (2003). Comparative toxicity evaluation of cyanobacterial cyclic peptide toxin
microcystin congeners (LR, RR, YR) in mice. Toxicology 188; 2: 285-296.
Guzman, R. E., Solter, P. F. (1999). Hepatic oxidative stress following prolonged sublethal
Microcystin LR exposure. Toxicologic Pathology, 27: 582–588.
Haddix, P. L., Hughley, C. J., and Lechevallier, M. W. (2007). Occurrence of microcystins in 33

US water supplies. Journal American Water Works Association, 99(9): 118-125.
Han, J., B. Jeon, and H. Park. 2012. Cyanobacteria cell damage and cyanotoxin release in
response to alum treatment Water Science & Technology: Water Supply, 12.5, pp. 549-
555.
Haney, J. and Ikawa, M. (2000). A Survey of 50 NH Lakes for Microcystin (MCs). Final Report
Prepared for N.H. Department of Environmental Services by University of New
Hampshire. Retrieved September 25, 2012 from the World Wide Web:
http://water.usgs.gov/wrri/AnnualReports/2000/NHfy2000_annual_report.pdf
Harada K., Ogawa, K., Kimura, Y., et al. (1991). Microcystins from Anabaena flos-aquae NRC
525-17. Chemical Research in Toxicology, 4(5): 535-540.
Health Canada. (2002). Guidelines for Canadian Drinking Water Quality: Supporting
Documentation - Cyanobacterial Toxins–Microcystin-LR. Water Quality and Health
Bureau, Healthy Environments and Consumer Safety Branch, Health Canada, Ottawa,
Ontario. Available at http://www.hc-sc.gc.ca/ewh-semt/pubs/water-eau/doc_sup-
appui/index_e.html
Health Canada. (2012). Toxicity Profile for Cyanobacterial Toxins. Prepared for Water Quality
and Science Division of Health Canada by MTE GlobalTox. MTE File No.: 36348-100.
January 27, 2012. pp. 48.

Health Canada. (2015). Guidelines for Canadian Drinking Water Quality: Guideline technical
document for public comment - Cyanobacterial toxins. Water and Air Quality Bureau,
Healthy Environments and Consumer Safety Branch, Ottawa, Ontario. Available in 2015.
Heinze, R. (1999). Toxicity of the cyanobacterial toxin microcystin-LR to rats after 28 days
intake with the drinking water. Environmental Toxicology, 14(1): 57-60.
Hermansky, S. J., Casey, P. J., and Stohs, S. J. (1990a). Cyclosporin A - a chemoprotectant

against microcystin-LR toxicity. Toxicology Letters, 54(2-3): 279-285.
Hermansky S. J., Wolff, S. N., and Stohs, S. J. (1990b). Use of rifampin as an effective
chemoprotectant and antidote against microcystin-LR toxicity. Pharmacology 41(4): 231-
236.
Hilborn E. D., Soares R. M., Servaites J. C., et al. (2013). Sublethal Microcystin Exposure and
Biochemical Outcomes among Hemodialysis Patients. PLoS ONE, 8(7): e69518
Hitzfeld, B. S., Höeger, J., and Dietrich, D. R. (2000). Cyanobacterial Toxins: Removal during
Drinking Water Treatment, and Human Risk Assessment. Environmental Health
Perspectives, 108: 113-122.
Ho, L., Sawade, E., and Newcombe, G. (2012b). Biological treatment options for cyanobacteria
metabolite removal-a review. Water Research, 46: 1536-1548.
Ho, L. and Newcombe, G.(2007). Evaluating the adsorption of microcystin toxins using granular
activated carbon (GAC). Journal of Water Supply: Research and Technology-Aqua, 56(5):
281-291.
Ho, L., Dreyfus, J., Boyer, J., et al. (2012a). Fate of cyanobacteria and their metabolites during
water treatment sludge management processes. Science of the Total Environment,
424:232-238.
Ho, L., Barbero, A., Dreyfus, J., et al. (2013). Behaviour of cyanobacterial bloom material
following coagulation and/or sedimentation. Journal of Water Supply: Research and
Technology-Aqua, 62(2): 350-358.
Ho, L., Onstand, G., von Gunten, U., et al. (2006a). Differences in the chlorine reactivity of four
microcystin analogues. Water Research, 40(6): 1200-1209.
Ho, L., Meyn, T., Keegan, A., et al. (2006b). Bacterial degradation of microcystin toxins within a
biologically active sand filter. Water Research, 40(4): 768-774.
Ho, L., Wijesundara, S., Shaw, G., et al. (2008). Biological Filtration Processes for the Removal
of Algal Metabolites. The Cooperative Research Center for Water Quality and Treatment.
Research Report 64. Salisbury, Australia.

Ho, L., Lambling, P., Bustamante, H., et al. (2011). Application of powdered activated carbon for
the adsorption of cylindrospermopsin and microcystin toxins from drinking water
supplies. Water Research, 45(9): 2954-2964.
Hoeger, S. J., Shaw, G., Hitzfeld, B. C., Dietrich, D. R. (2004). Occurrence and elimination of
cyanobacterial toxins in two Australian drinking water treatment plants. Toxicon, 43: 639-
649.
Hoeger, S. J., Hitzfeld, B. C., Dietrich, D. R. (2005). Occurrence and elimination of

cyanobacterial toxins in drinking water treatment plants. Toxicology and Applied
Pharmacology, 203: 231-242.
Huang, W. J., Cheng, B. L. and Cheng, Y. L. (2007). Adsorption of microcystin-LR by three

types of activated carbon. Journal of Hazardous Materials, 141: 115-122.
Huang, P., Zheng, Q., and Xu, L-H. (2011). The apoptotic effect of oral administration of
microcystin-RR on mice liver. Environmental Toxicology, 26(5): 443-452.
Hudnell, H. K. (Ed.) (2008). Cyanobacterial Harmful Algae Blooms, State of the Science and
Research Needs. Proceedings of the Interagency, International Symposium on
Cyanobacterial Harmful Algal Blooms. RTP North Carolina, Sept. 2005. Advances in
Experimental Medicine & Biology. Springer Science. Vol. 619, 948 pp.
Hudnell, H. K. (2010). The state of U.S. freshwater harmful algal blooms assessments policy and
legislation. Toxicon, 55: 1024-1034.
Ito, E., Kondo, F., and Harada, K-I. (1997a). Hepatic necrosis in aged mice by oral administration
of microcystin-LR. Toxicon, 35(2): 231-239.
Ito, E., Kondo, F., Terao, K., and Harada, K-I. (1997b). Neoplastic nodular formation in mouse
liver induced by repeated i.p. injections of microcystin-LR. Toxicon, 35(9): 1453-1457.
Jochimsen, E. M., Carmichael W. W., An J. S., et al. (1998). Liver failure and death after
exposure to microcystins at a hemodialysis center in Brazil. New England Journal of
Medicine, 338(13): 873-8.
Jones, G. J., Blackburn, S. I., and Parker, N. S. (1994). A toxic bloom of Nodularia spumigena
Mertens in Orielton Lagoon, Tasmania. Australian Journal of Marine and Freshwater
Research, 45: 787-800. (Cited in WHO 1999).
Jurczak, T., Tarczynska, M., Izydorczyk, K., et al. (2005). Elimination of microcystins by water
treatment processes-examples from Sulejow Resevoir, Poland. Water Research, 39(3):
2394-2406.

Karner, D. A., Standridge, J. H., Harrington, G. W., and Barnum, R. P. (2001). Microcystin algal
toxins in source and finished drinking water. Journal American Water Works Association,
93(8): 72-81.
Keijola, A., Himber, K, Esala, A. L., et al. (1998). Removal of cyanobacterial toxins in water
treatment processes: laboratory and pilot plant experiments. Toxicity Assessment, 3(5):
643-656.
Klassen, C. D. and Aleksunes, L. M. (1996). Xenobiotic bile acid and cholesterol transporters:
function and regulation. Pharmacological Reviews, 62: 1-96. (As cited in Zhou et al.,
2012).
Kondo, F., Ikai, Y, Oka, H., et al. (1992). Formation, characterization, and toxicity of the
glutathione and cysteine conjugates of toxic heptapeptide microcystins. Chemical
Research in Toxicology, 5(5): 591-596.
Kondo, F., Matsumoto, H, Yamada, S., et al. (1996). Detection and identification of metabolites
of microcystins in mouse and rat liver. Chemical Research in Toxicology, 9: 1355–1359.
Kosakowska, A., Nedzi, M. and Pempkowiak, J. (2007). Responses of the toxic cyanobacterium
Microcystis aeruginosa to iron and humic substances. Plant Physiology and Biochemistry
45: 365-370.
Kull, T. P., Sjovall, O. T., Tammenkoski, M. K., et al. (2006). Oxidation of the cyanobacterial
hepatotoxin microcystin-LR by chlorine dioxie: influence of natural organic matter.
Environmental Science and Technology, 40(5): 1504-1510.
Kull, T. P., Backlund, P. H., Karlsson, K. M., and Meriluoto, J. A. (2004). Oxidation of the
cyanobacterial hepatotoxin microcystin-LR by chlorine dioxide: reaction kinetics,
characterization, and toxicity of reaction products. Environmental Science and
Technology, 38(22): 6025–6031.
Lahti, K., Niemi, M. R., Rapala, J., and Sivonen, K. (1997a). Biodegradation of cyanobacterial
hepatotoxins - characterization of toxin degrading bacteria. Proceedings of the VII
International Conference on Harmful Algae. (Cited in WHO 1999).
Lahti, K., Rapala, J., Fardig, M., Niemela, M., and Sivonen, K. (1997b). Persistence of
cyanobacterial hepatotoxin, microcystin-LR, in particulate material and dissolved in lake
water. Water Research, 31(5): 1005-1012. (Cited in WHO 1999).
Lahti, K., Rapala, J., Kivimaki, A. L., et al. (2001). Occurrence of microcystins in raw water
sources and treated drinking water of Finnish waterworks. Water Science and
Technology, 43(1): 225-228.

Lahti, K.J. et al. (1998). Fate of cyanobacterial hepatotoxins in artificial recharge of groundwater
and in bank filtration. In: J. H. Peters, et al. (Eds), Artifical recharge of groundwater. A.S.
Balkema, Rotterdam, The Netherlands.
Lam, A. K., Fedorak, P. M, and Prepas, E. E. (1995). Biotransformation of the cyanobacterial

hepatotoxin microcystin-LR, as determined by HPLC and protein phosphatase bioassay.
Environmental Science and Technology, 29: 242-246.
Lambert, T. W., Holmes, C. F. B., and Hrudey, S. E. (1996). Adsorption of microcystin-LR by

activated carbon and removal in full scale water treatment. Water Research, 30(6): 1411-
1422.
Lee, J. and Walker, H. W. (2008). Mechanisms and factors influencing the removal of
microcystin-LR by ultrafiltration membranes. Journal of Membrane Science, 320: 240-
247.
Lepisto, L., Lahti, K., Niemi, J., and Fardig, M. (1994). Removal of cyanobacteria and other
phytoplankton in four Finish waterworks. Algological Studies, 75: 167-181.
Li, Y., Chen, J. A., Zhao, Q., et al. (2011a). A Cross-Sectional Investigation of Chronic Exposure
to Microcystin in Relationship to Childhood Liver Damage in the Three Gorges Reservoir
Region, China. Environmental Health Perspectives. 119(10): 1483-1488.
Li, T., Huang, P., Liang, J., et al. (2011b). Microcystin-LR (MCLR) induces a compensation of
PP2A activity mediated by alpha4 protein in HEK293 cells. International Journal of
Biological Sciences, 7(6): 740-52.
MWH. (2012). Water treatment principles and design. 3rd edition. John Wiley & Sons. New
York, NY.
Makarewicz, J., Boyer, G., Guenther, W., et al. (2006). The Occurrence of Cyanotoxins in the
Nearshore and Coastal Embayments of Lake Ontario. Great Lakes Research Review, 7:
25-29.
Makarewicz, J., Boyer, G. Lewis, T., et al. (2009). Spatial and temporal distribution of the
cyanotoxin microcystin-LR in the Lake Ontario ecosystem: Coastal embayments, rivers,
nearshore and offshore, and upland lakes. Journal of Great Lakes Research, 35: 83-89.
Matthijs, H. C., Visser, P. M., Reeze, B., et al. (2012). Selective suppression of harmful
cyanobacteria in an entire lake with hydrogen peroxide. Water Research, 46(5): 1460-
1472.
Merel, S., LeBot, B., Clement, M., et al. (2009). MS identification of microcystin-LR chlorination
by-products. Chemosphere, 74(6): 832-839.

Merel, S., Clement, M., and Thomas, O. (2010). State of the art on cyanotoxins in water and their
behaviour towards chlorine. Toxicon, 55: 677-691.
Minnesota Department of Health (MDH) (2012). Health Based Value for Groundwater. Health
Risk Assessment Unit, Environmental Health Division, No.651-201-4899. Retrieved April
25, 2014 from the World Wide Web:
http://www.health.state.mn.us/divs/eh/risk/guidance/gw/microcystin.pdf
Nasri, H., Bouaicha, N, and Harche, M. K. (2007). A new morphospecies of Microcystis sp.
forming bloom in the Cheffia Dam (Algeria): seasonal variation of microcystin
concentrations in raw water and their removal in a full-scale treatment plant.
Environmental Toxicology, 22(4): 347-356.
Neumann, U., and Weckesser, J. (1998). Elimination of microcystin peptide toxins from water by
reverse osmosis. Environmental Toxicology, 13: 143-148.
Newcombe, G. (2002). Removal of Algal Toxins from Drinking Water Using Ozone and GAC.
American Water Works Association Reseach Foundation and American Water Works
Association. Denver, CO.
Newcombe, G., Cook, D., Brooke, S., et al. (2003). Treatment options for microcystin toxins:
similarities and differences between variants. Environmental Toxicology, 24: 299-308.
Newcombe, G., House, J., Ho, L., et al. (2010). Management Strategies for Cyanobacteria (Blue-
Green Algae): A Guide for Water Utilities. Water Quality Research Australia, Research
Report 74. Adelaide, Australia.
Newcombe, G., Dreyfus, J., Monrolin, Y., et al. (2015). Optimizing Conventional Treatment for
the Removal of Cyanobacteria and Toxins. Water Research Foundation, Denver, CO.
NHMRC, NRMMC. (2011). Australian Drinking Water Guidelines Paper 6 National Water
Quality Management Strategy. National Health and Medical Research Council, National
Resource Management Ministerial Council, Commonwealth of Australia, Canberra.
http://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/eh52_australian_drinkin
g_water_guidelines_150413.pdf
Nicholson, B. C., Rositano, J., and Burch, M. D. (1994). Destruction of cyanobacterial peptide
hepatotoxins by chlorine and chloramine.Water Research, 28(6): 1297–1303.
Nishiwaki-Matsushima, R., Ohta, T., Nishiwaki, S. et al. (1992). Liver tumor promotion by the
cyanobacterial cyclic peptide toxin microcystin-LR. Journal of Cancer Research and
Clinical Oncology, 118(6): 420-424.
O’Reilly, A., Wanielista, M., Loftin, K., and Chang, N. (2011). Laboratory simulated transport of
microcystin-LR and cylindrospermopsin in groundwater under the influence of stormwater
ponds: implications for harvesting of infiltrated stormwater. GQ10: Groundwater Quality
Management in a Rapidly Changing World (Proc. 7th International Groundwater Quality

Conference held in Zurich, Switzerland, 13–18 June 2010). IAHS Publ 342, 2011, 107-
111.
Ohio EPA (OHEPA) (2012). 2011 Grand Lake St. Marys Algal Toxin Sampling Data. Retrieved
September 25, 2012 from the World Wide Web:
http://www.epa.state.oh.us/dsw/HAB.aspx
Ohio EPA (OHEPA). (2014). Public Water System Harmful Algal Bloom Response Strategy.
Retrieved April 25, 2014 from the World Wide Web:
http://epa.ohio.gov/Portals/28/documents/HABs/PWS_HAB_Response_Strategy_2014.pd
f
Ohta, T., Sueoka, E., Iida, N., et al. (1994). Nodularin, a potent inhibitor of protein phosphatases
1 and 2A, is a new environmental carcinogen in male F344 rat liver. Cancer Research,
54(24): 6402-6406.
Oregon Health Authority (OHA) (2015). Public Health Advisory Guidelines, Harmful Algae
Blooms in Freshwater Bodies. Harmful Algae Bloom Surveillance (HABS) Program,
Public Health Division, Center for Health Protection, Retrieved April 25, 2014 from the
World Wide Web:
http://public.health.oregon.gov/HealthyEnvironments/Recreation/HarmfulAlgaeBlooms/D
ocuments/HABPublicHealthAdvisoryGuidelines%2020150424x.pdf
Pace, J. G., Robinson, N. A., Miura, G. A., et al. (1991). Toxicity and kinetics of 3H microcystin-
LR in isolated perfused rat livers. Toxicolgy and Applied Pharmacology, 107(3): 391-401.
Paerl, H. W. and Otten, T. G. (2013a). Blooms Bite the Hand That Feeds Them. Science, 342(25):
433-434.
Paerl, H. W. and Otten, T. G. (2013b). Harmful Cyanobacterial Blooms: Causes, Consequences,

and Controls. Microbial Ecology, 65: 995-1010.
Pawlowicz, M. B., Evans, J. E., Johnson, D. R., and Brooks, R. G. (2006). A study of the efficacy
of various home filtration substrates in the removal of microcysin-LR from drinking
water. Journal of Water and Health, 4(1): 99-107.
Pendleton, P., Schumann, R. and Wong, S. H. (2001). Microcystin-LR adsorption by activated

carbon. Journal of Colloid and Interface Science, 240: 1-8.
Pereira, S. P., Martins, F. C., Gomes, L. N. L., et al. (2012). Removal of cyanobacteria by slow
sand filtration for drinking water. Journal of Water, Sanitation, and Hygiene for
Development, 2(3): 133-145.
Puddick, J., Prinsep, M., Wood, S., et al. (2015). Further Characterization of Glycine-Containing
Microcystins from the McMurdo Dry Valleys of Antarctica. Toxins, 7: 493-515.

Rai, A. N. (1990). CRC Handbook of Symbiotic Cyanobacteria. CRC Press, Boca Raton, FL. pp.
253. (Cited in WHO 1999).
Rajasekhar, P., Fan, L., Nguyen, T., and Roddick, F. A. (2012). A review of the use of sonication
to control cyanobacterial blooms. Water Research, 46: 4319-4329.
Rao, P. V. L., Gupta, N., Bhaskar, A. S. B., and Jayaraj, R. (2002). Toxins and bioactive
compounds from cyanobacteria and their implications on human health. Journal of
Environmental Biology, 3: 215-224.
Rao, P. V. L., Gupta, N., Jayaraj, R., et al. (2005). Age-dependent effects on biochemical
variables and toxicity induced by cyclic peptide toxin microcystin-LR in mice.
Comparative Biochemistry and Physiology Part C: Toxicology and Pharmacology,
140(1): 11-19.
Rapala, J., Lahti, K., Sivonen, K., and Niemeld, S. (1994). Biodegradability and adsorption on
lake sediments of cyanobacterial hepatotoxins and anatoxin-a. Letters in Applied
Microbiology, 19: 423-428. (Cited in WHO 1999).
Rapala, J., Niemela, M., Berg, K., et al.(2006). Removal of cyanobacteria, cyanotoxins,
heterotrophic bacteria and endotoxins at an operating surface water treatment plant. Water
Science and Technology, 54(3): 23-28.
Robinson, N. A., Pace, J. G., Matson, C. F., et al. (1991). Tissue distribution, excretion and
hepatic biotransformation of microcystin-LR in mice. Journal of Pharmacology and
Experimental Therapeutics, 256(1): 176-182.
Rodriguez, E., Onstad, G. D., Kull, T. P. J., et al. (2007a). Oxidative elimination of cyanotoxins:
comparison of ozone, chlorine, chlorine dioxide and permanganate. Water Research,
41(15): 3381-3393.
Rodriguez, E., Majado, M. E., Meriluoto, J., et al. (2007b). Oxidation of microcystins by
permanganate: reaction kinetics and implications for water treatment. Water Research,
41(1): 102-110.
Rositano, J., Nicholson, B. C., and Pieronne, P. (1998). Destruction of cyanobacterial toxins by
ozone . Ozone: Science and Engineering: The Journal of the International Ozone
Association, 20(3): 223-238.
Rositano, J., Newcombe, G., Nicholson, B., and Sztanjnbok. (2001). Ozonation of NOM and algal
toxins in four treated waters. Water Research, 35(1): 23-32.
Roy-Lachapelle, A., Fayad, P. B., Sinotte, M., et al. (2014). Total microcystin analysis in water
using laser diode thermal desorption-atmospheric pressure chemical ionization-tandem
mass spectrometry. Analytical Chimica Acta, 820: 76-83.

Runnegar, M. T. C. and Falconer, I. R. (1982). The in vivo and in vitro biological effects of the
peptide hepatotoxin from the blue-green alga Microcystis aeruginosa. South African
Journal of Science, 78: 363-366.
Runnegar, M., Berndt, N., Kong, S. M., et al. (1995). In vivo and in vitro binding of microcystin
to protein phosphatases 1 and 2A. Biochemical and Biophysical Research
Communications, 216(1): 162-169.
Runnegar, M. T. C., Falconer, I. R., and Silver, J. (1981). Deformation of isolated rat hepatocytes
by a peptide hepatoxin from the blue-green alga Microcystis aeruginosa. Naunyn-
Schmiedeberg's Archives of Pharmacology, 317(3): 268-272.
Sarma, T.A. (2013). Cyanobacterial Toxins in Handbook of Cyanobacteria. CRC Press. Taylor
and Francis Group, pp. 487-606.
Schijven, J., Berger, P, and Miettinen, I. (2002). Removal of pathogens, surrogates, indicators and
toxins using riverbank filtration. In: C. Ray (Ed), Riverbank Filtration: Improving Source-
Water Quality. Kluwer Academic Publishers, Netherlands. pp. 73-116.
Sharma, V. K., Triantis, T. M., Antoniou, M. G., et al. (2012). Destruction of microcystins by
conventional and advanced oxidation processes: a review. Separation and Purification
Technology, 91: 3-17.
Shawwa, A. R. and Smith, D. W. (2001). Kinetics of microcystin-LR oxidation by ozone. Ozone:

Science and Engineering: The Journal of the International Ozone Association, 23: 161-
170.
Sklenar, K., Westrick, J., and Szlag, D.(2014). Managing and Mitigating Cyanotoxins in Water
Supplies. Water Research Foundation Webcast. August 28, 2014.
Soares, R.M. Yuan, M., Servaites, C. Delgado, A., V.F. Magalhães, V., Hilborn, E., Carmichael,
W., and Azevedo, S. (2005). Sublethal exposure from microcystins to renal insufficiency
patients in Rio de Janeiro, Brazil. Environmental Toxicology, 21(2): 95-103
Song, W., Teshiba, T., Rein, K., and O’Shea, K. E. (2005). Ultrasonically induced degradation
and detoxification of microcystin-LR (cyanobacterial toxin). Environmental Science and
Technology, 39(16): 6300-6305.
Sorlini, S. and Collivignarelli, C.(2011). Microcystin-LR removal from drinking water supplies
by chemical oxidation and activated carbon adsorption. Journal of Water Supply Research
and Technology, 60(7): 403–411.
Sorlini, S., Gialdini, F., and Collivignarelli, C. (2013). Removal of cyanobacterial cells and
microcystin-LR from drinking water using a hollow fiber microfiltration pilot plant.
Desalination, 309: 106-112.

Stewart, I., Schluter, P., and Shaw, G. (2006a). Cyanobacterial lipopolysaccharides and human
health - a review. Environmental Health, 5(1): 1-23.
Stoner, R. D.; Adams, W. H.; Slatkin, D. N.; and Siegelman, H. W. (1989). The effects of single
L-amino acid substitutions on the lethal potencies of microcystins. Toxicon, 27(7): 825-
828.
Sun, F., Pei, H.-Y., Hu, W.-R., and Ma, C.-X. (2012). The lysis of Microcystis aeruginosa in
AlCl3 coagulation and sedimentation processes. Chemical Engineering Journal, 193-194:
196-202.
Svoboda, M., Riha, J., Wlcek, K., et al. (2011). Organic Anion Transporting Polypeptides
(OATPs): regulation of expression and function. Current Drug Metabolism, 12(2): 139-
153.
Takumi, S., Komatsu, M., Furukawa, T., et al.(2010). p53 plays an important role in cell fate
determination after exposure to microcystin-LR. Environmental Health Perspectives,
118(9): 1292-1298.
Tarczyriska, M., Romanowska –Duda, Z., Jurczak, T., and Zalewski, M. (2001). Toxic
cyanobacterial blooms in drinking water reservoir, causes, consequences, and
management strategy. Water Science and Technology: Water Supply, 1(2): 237-246.
Thompson, W. L. and Pace, J. G. (1992). Substances that protect cultured hepatocytes from the
toxic effects of microcystin-LR. Toxicology In vitro, 6(6): 579-587.
Touchette, B. W., Burkholder, J. M., Allen, E. H., et al. (2007). Eutrophication and cyanobacteria
blooms in run-of-river impoundments in North Carolina, U.S.A. Lake and Reservoir
Management, 23(2): 179-192.
Toxicology Literature Online (TOXLINE) (2012). Toxicology Data Network, National Institute of
Health. Retrieved on September 25, 2012 from the World Wide Web:
http://toxnet.nlm.nih.gov/cgi-bin/sis/htmlgen?TOXLINE
Tsuji, K., Naito, S., Kondo, F., et al. (1993). Stability of microcystins from cyanobacteria: Effect
of light on decomposition and isomerization. Environmental Science and Technology, 28:
173-177. (Cited in WHO 1999).
Tsuji, K., Naito, S., Kondo, F., et al. (1995). Stability of microcystins from cyanobacteria--II.
Effect of UV light on decomposition and isomerization. Toxicon, 33(12): 1619-1631.
Turner, P. C., Gammie, A. J., Hollinrake K., and Codd, G. A. (1990). Pneumonia associated with
contact with cyanobacteria. British Medical Journal, 300(6737): 1440-1441.

Ueno, Y., Makita, Y., Nagata, S., et al. (1999). No chronic oral toxicity of a low-dose of
microcystin-LR, a cyanobacterial hepatoxin, in female Balb/C mice. Environmental
Toxicology, 14(1): 45-55.
Ueno, Y., Nagata, S., Tsutsumi, T., et al. (1996). Detection of microcystins, a blue-green algal
liver cancer in China, by highly sensitive immunoassay. Carcinogenesis 17(6): 1317-
1321.
U.S. EPA (United States Environmental Protection Agency). (1986). Guidelines for Mutagenicity
Risk Assessment. Fed. Reg. 51(185):34006-34012. Available from:
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=23160
U.S. EPA (United States Environmental Protection Agency) (1999). Drinking Water Health
Advisories: Pesticides. Lewis Publishers. ISBN: 978-0-87371-235-4.
U.S. EPA (United States Environmental Protection Agency) (2002). A Review of the Reference
Dose and Reference Concentration Processes. Risk Assessment Forum, Washington, DC;
EPA/630/P-02/0002F. Available from: http://www.epa.gov/iris/backgrd.html
U.S. EPA (United States Environmental Protection Agency). (2005). Guidelines for Carcinogen
Risk Assessment. Risk Assessment Forum, Washington, DC; EPA/630/P-03/001B.
Available from: http://www.epa.gov/iris/backgrd.html
U.S. EPA (United States Environmental Protection Agency). (2007). Drinking Water Treatability
Database: Microcystins. Available from:
http://iaspub.epa.gov/tdb/pages/contaminant/contaminantOverview.do?contaminantId=-
1336577584
U.S. EPA (United States Environmental Protection Agency) (2009). National Lakes Assessment:
A Collaborative Survey of the Nation’s Lakes. EPA 841-R-09-001. Available from:
http://www.epa.gov/owow/LAKES/lakessurvey/pdf/nla_report_low_res.pdf
U.S. EPA (United States Environmental Protection Agency). (2011a). Exposure Factors
Handbook 2011 Edition (Final). Washington, DC, EPA/600/R-09/052F. Available from:
http://www.epa.gov/ncea/efh/pdfs/efh-complete.pdf
U.S. EPA (United States Environmental Protection Agency). (2011b). Recommended Use of
Body Weight 3/4 as the Default Method in Derivation of the Oral Reference Dose.
EPA/100/R11/0001. http://www2.epa.gov/sites/production/files/2013-
09/documents/recommended-use-of-bw34.pdf.
U.S. EPA (United States Environmental Protection Agency). (2012). Benchmark dose technical
guidance document [external review draft]. EPA/630/R-00/001. Available from:
http://www.epa.gov/iris/backgrd.html

U.S. EPA (United States Environmental Protection Agency). (2014). Framework for Human
Health Risk Assessment to Inform Decision Making. Office of the Science Advisor, Risk
Assessment Forum, Washington, DC; EPA/100/R-14/001. Available from:
http://www2.epa.gov/programs-office-science-advisor-osa/framework-human-health-risk-
assessment-inform-decision-making
U.S. EPA (United States Environmental Protection Agency). (2015a). Health Effects Support
Document for the Cyanobacterial Toxin Microcystin. EPA-820R1502, Washington,
DC; 2015. Available from: http://www2.epa.gov/nutrient-policy-data/health-and-
ecological-effects
U.S. EPA (United States Environmental Protection Agency). (2015b). Recommendations for
Public Water Systems to Manage Cyanotoxins in Drinking Water. EPA-815R15010,
Washington, DC; 2015. Available from: http://www2.epa.gov/nutrient-policy-data/
guidelines-and-recommendations
U.S. EPA (United States Environmental Protection Agency). (2015c) Method 544. Determination
of Microcystins and Nodularin in Drinking Water by Solid Phase Extraction and Liquid
Chromatography/Tandem Mass Spectrometry (LC/MS/MS). Version 1.0, EPA/600/R-
14/474, Cincinnati, OH. Available from:
http://www.epa.gov/nerlcwww/documents/Method544_Final.pdf
VanderKooi, S., Burdinck, Echols, K., et al. (2010). Algal Toxin in Upper Klamath Lake, Oregon:
Linking Water Quality to Juvenile Sucker Health. U.S. Geological Survey Fact Sheet
2009-3111, pp. 2.
Vlaski, A. van Breemen, A. N., Alaerts, G. J., et al. (1996). Optimisation of coagulation
conditions for the removal of cyanobacteria by dissolved air flotation or sedimentation.
Journal of Water Supply: Research and Technology, 45(5): 253-261.
Wang, H., Ho., L., Lewis, D. M., et al. (2007). Discriminating and assessing adsorption and
biodegradation removal mechanisms during granular activated carbon filtration of
microcystin toxins. Water Research, 41(18): 4262-4270.
Wang, Z. Li., D, Qin, H., and Li, Y. (2012). An integrated method for removal of harmful
cyanobacterial blooms in eutrophic lakes. Environmental Pollution, 160: 34-41.
WSDE (Washington State Department of Ecology). (2012). Freshwater Algae Control Program.
Accessed December 12,
2012; http://www.ecy.wa.gov/programs/wq/plants/algae/index.html
Wei, Y., Weng, D., Li, F., et al. (2008). Involvement of JNK regulation in oxidative stress-
mediated murine liver injury by microcystin-LR. Apoptosis, 13(8): 1031-1042.
Weng, D., Y. Lu, Wei, Y., et al. (2007). The role of ROS in microcystin-LR-induced hepatocyte
apoptosis and liver injury in mice. Toxicology, 232(1-2): 15-23.

Wert, E. C., and Rosario-Ortiz, F. L. (2013). Intracellular organic matter from cyanobacteria as a
precursor for carbonaceous and nitrogenous disinfection byproducts. Environmental
Science and Technology, 47: 6332-6340.
Wert, E.C., Dong, M. M., Rosario-Ortiz, F. L., and Korak, J. (2014). Release of Intracellular
Metabolites from Cyanobacteria During Oxidation Processes. Water Research
Foundation, Denver, CO.
Westrick, J. A., Szlag, D. C., Southwell, B. J., and Sinclair, J. (2010). A review of cyanobacteria
and cyanotoxins removal/inactivation in drinking water treatment. Analytical and
Bioanalytical Chemistry, 397(5): 1705-1714.
WHO (World Health Organization). (1999). Toxic Cyanobacteria in Water: A Guide to their
Public Health Consequences, Monitoring, and Management, I. Chorus and J. Bartram,
(Eds.). E&FN Spon, London, UK.
WHO (World Health Organization). (2003). Cyanobacterial Toxins: Microcystin-LR in Drinking-

Water. Background document for development of WHO Guidelines for Drinking-water
Quality, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland.
Wolf, H.-U. and Frank, C. (2002). Toxicity assessment of cyanobacterial toxin mixtures.
Environmental Toxicology, 17(4): 395-399.
Wynne, T., Stumpf, R., Tomlinson, M. et al. (2013). Evolution of a cyanobacterial bloom forecast
system in western Lake Erie: Development and initial evaluation. Journal of Great Lakes
Research, Supplement 39(9): 90-99.
Xagoraraki, I., Harrington, G., Zulliger, K., et al. (2006). Inactivation kinetics of the
cyanobacterial toxin microcystin-LR by free chlorine. Journal of Environmental
Engineering, 132(7): 818-823.
Xing, M. L., Wang, X. F., and Xu, L. H. (2008). Alteration of proteins expression in apoptotic FL
cells induced by MCLR. Environmental Toxicology, 23(4): 451-458.
Yuan, M., Namikoshi, M., Otsuki, A., et al. (1999). Electrospray Ionization Mass Spectrometric
Analysis of Microcystins, Cyclic Heptapeptide Hepatotoxins: Modulation of Charge
States and [M 1 H]1 to [M 1 Na]1 Ratio. Journal of the Amereican Society for Mass
Spectrometry, 10: 1138–1151.
Yoshida, T., Makita, Y, Nagata, S., et al. (1997). Acute oral toxicity of microcystin-LR, a
cyanobacterial hepatotoxin in mice. Natural Toxins, 5(3): 91-95.
Zamyadi, A., MacLeod, S. L., Fan, Y., et al. (2012a). Toxic cyanobacterial breakthrough and
accumulation in a drinking water plant: A monitoring and treatment challenge. Water
Research, 46(5): 1511-1523.

Zamyadi, A., Ho, L., Newcombe, G., et al. (2012b). Fate of toxic cyanobacterial cells and
disinfection by-products formation after chlorination. Water Research, 46(5): 1524-1535.
Zamyadi, A., Dorner, S., Sauve, S., et al. (2013a). Species dependence of cyanobacteria removal
efficiency by different drinking water treatment processes. Water Research, 47(8): 2689-
2700.
Zamyadi, A., Dorner, S., Ndong, M., et al. (2013b). Low-risk cyanobacterial bloom sources: cell
accumulation within full-scale treatment plants. Journal American Water Works
Association, 105(11): 651-663
Zamyadi, A., Fan, Y, Daly, R. I., Prevost, M. (2013c). Chlorination of Microcystis aeruginosa:
toxin release and oxidation, cellular chlorine demand and disinfection by-products
formation. Water Research, 47(3): 1080-1090.
Zhang, Z., Zhang, X. X., Qin, W., et al. (2012). Effects of microcystin-LR exposure on matrix
metalloproteinase-2/-9 expression and cancer cell migration. Ecotoxicology and
Environmental Safety, 77: 88-93.
Zhou, L., Yu, H., and Chen, K. (2002). Relationship between microcystin in drinking water and
colorectal cancer. Biomedical and Environmental Sciences, 15(2): 166-171.
Zhou, Y., Yuan, J., Wu, J., and Han, X. (2012). The toxic effects of microcystin-LR of rat
spermatogonia in vitro. Toxicology Letters, 212(1): 48-56.
Zurawell, R. (2002). An Initial Assessment of Microcystin in Raw and Treated Municipal

Drinking Water Derived from Eutrophic Surface Waters in Alberta. Alberta Environment,
Edmonton, AB. ISBN No. 0-7785-2417-5.

Recommendations for
Public Water Systems to
Manage Cyanotoxins in
Drinking Water
June 2015
Office of Water (4606M)
EPA 815-R-15-010
June 2015
Disclaimer
This document was prepared by the U.S. Environmental Protection Agency (EPA) as an
informational resource for public water systems and primacy agencies to prepare for and respond
to the risk of cyanotoxins in finished water.
This document is not a regulation; it is not legally enforceable; and it does not confer legal rights
or impose legal obligations on any party, including EPA, States, or the regulated community.
While EPA has made every effort to ensure the accuracy of the discussion in this document, the
obligations of the regulated community are determined by statutes, regulations, or other legally
binding requirements. The recommendations discussed are not a substitute for applicable legal
requirements. In the event of a conflict between the discussion in this document and any statute
or regulation, this document would not be controlling.
Although this document describes suggestions for managing cyanotoxin issues in raw and
finished water, the recommendations may not be appropriate for all situations and alternative
approaches may be applicable.
Mention of trade names or commercial products does not constitute an EPA endorsement or
recommendation for use.
This Page Intentionally Left Blank
Table of Contents
Table of Contents ................................................................................................................... i
Table of Exhibits .................................................................................................................... ii
Abbreviations and Acronyms ................................................................................................ iii
Acknowledgements .............................................................................................................. iv
Preparers and Reviewers ....................................................................................................... v
Executive Summary .............................................................................................................. vi
A. Introduction .......................................................................................................................1
B. What is a Health Advisory?..................................................................................................2
C. Cyanotoxin Management Plan Development .......................................................................3
D. Potential Cyanotoxin Management Steps ............................................................................5
1 Step 1: Conduct System-Specific Surface Water Evaluation ..................................................6
2 Step 2: Preparation and Observation ...................................................................................7
2.1 Preparation ........................................................................................................................8
2.2 Observation ........................................................................................................................9
2.2.1 Visual Inspection and Phytoplankton Identification ............................................................... 9
2.2.2 System Effects ....................................................................................................................... 11
2.2.3 Other Bloom Indicators ......................................................................................................... 12
2.3 Communication ................................................................................................................13
2.4 Source Water Mitigation ...................................................................................................13
3 Step 3: Monitor for Cyanotoxins in Raw Water and Treatment Adjustments....................... 15
3.1 Sampling and Analysis for Cyanotoxins ..............................................................................15
3.1.1 Raw Water Sampling ............................................................................................................. 15
3.1.2 Sampling Logistics ................................................................................................................. 16
3.1.3 Analytical Methods ............................................................................................................... 17
3.2 Communications ...............................................................................................................17
3.3 Treatment ........................................................................................................................17
4 Step 4: Monitor for Cyanotoxins in Raw and Finished Water and Treatment Adjustments .. 22
4.1 Monitoring .......................................................................................................................22
4.2 Communications ...............................................................................................................23
4.3 Treatment ........................................................................................................................23
5 Step 5: Monitor for Cyanotoxins in Finished Water, Treatment Adjustments or Additions,
and Public Communications ..............................................................................................23
5.1 Monitoring .......................................................................................................................24
5.2 Communications ...............................................................................................................25
5.3 Treatment ........................................................................................................................26
6 References ........................................................................................................................... 28
Appendix A System-Specific Evaluation Information, Tools, and Resources .............................38
Appendix B Examples of Water Utility and Community Responses to Harmful Algal Blooms.....48
Appendix C Key Questions and Answers .................................................................................50
Appendix D Potential Language for Use in a Cyanotoxin Public Notification and Social Media....54
Appendix E Long Term Mitigation Strategies and Treatment Options .......................................56
i
Table of Exhibits
Figure 1. Potential management steps public surface water systems may use to determine
whether cyanotoxins are present in raw water or finished drinking water. .................................... 5
Figure 2. Traffic light approach to support communication and other actions in response to
elevated concentrations of cyanotoxins in finished drinking water. ............................................. 27
ii
Abbreviations and Acronyms
ASDWA Association of State Drinking Water Administrators

AWWA American Water Works Association
CDC Centers for Disease Control and Prevention
CMP Cyanotoxin Management Plan
CT Concentration (oxidant) x Contact Time
CWA Clean Water Act
DAF Dissolved Air Flotation
DBP Disinfection Byproduct
DWACT Drinking Water Advisory Communication Toolbox
DWMAPS Drinking Water Mapping Application for Protecting Source Waters
ECHO Enforcement and Compliance History Online
ELISA Enzyme Linked Immunosorbent Assay
EPA Environmental Protection Agency
GAC Granulated Activated Carbon
HAAs Haloacetic Acids (HAA5)
HAs Health Advisories
HABs Harmful Algal Blooms
LC/MS/MS Liquid Chromatography/Mass Spectrometry/Mass Spectrometry
MCL Maximum Contaminant Level
MF Microfiltration
NF Nanofiltration
MIB Methylisoborneol
NPDAT Nitrogen and Phosphorus Data Access Tool
NPDES National Pollution Discharge Elimination System
NTU Nephelometric Turbidity Unit
PWS Public Water System
RO Reverse Osmosis
STORET STOrage and RETrieval
SWC Source Water Collaborative
TMDL Total Maximum Daily Load
TTHMs Total Trihalomethanes
UF Ultrafiltration
USACE United States Army Corps of Engineers
USDA United States Department of Agriculture
U.S. EPA United States Environmental Protection Agency
UV Ultraviolet
WRF Water Research Foundation
iii
Acknowledgements
EPA’s Office of Ground Water and Drinking Water would like to thank the Association of State
Drinking Water Administrators, the States of California, Massachusetts, New York, Ohio,
Oregon, Rhode Island and the U.S. Centers for Disease Control for their contributions to this
document. Additionally, over 500 participants joined EPA at a May 11, 2015 public meeting
discussing steps the Agency could take to support states and utilities in managing cyanotoxin
risks in drinking water. EPA thanks these participants for their meaningful input which was
critical for the development of this document.
iv
Preparers and Reviewers
The following individuals helped to develop and review this document:
Hannah Holsinger (U.S. EPA)1 James Hyde (New York State Department
Kenneth Rotert (U.S. EPA)1 of Health)
Lili Wang (U.S. EPA)1 Holly Kaloz (Ohio EPA)
Rachel Carlson (U.S. EPA)1 Kirk Leifheit (Ohio EPA)
Ryan Albert (U.S. EPA) Darren Lytle (U.S. EPA)
Becky Allenbach (U.S. EPA) Casey Lyon (Oregon Health Authority)
Steve Allgeier (U.S. EPA) Beth Messer (Ohio EPA)
Blake Atkins (U.S. EPA) Mike Messner (U.S. EPA)
Michael Baker (Ohio EPA) Jini Mohanty (U.S. EPA)
Michael Beach (CDC) Thomas Poy (U.S. EPA)
Phil Berger (U.S. EPA) Aditi Prabhu (U.S. EPA)
Kimberly Burnham (Ohio EPA) Heather Raymond (Ohio EPA)
Andrea Candara (New York State Stig Regli (U.S. EPA)
Department of Health) Crystal Rodgers-Jenkins (U.S. EPA)
Gregory Carroll (U.S. EPA) Karen Sklenar
Michael Celona (Massachusetts Glynda Smith (U.S. EPA)
Department of Public Health) Thomas Speth (U.S. EPA)
Jimmy Chen (U.S. EPA) June Swallow (Rhode Island Department of
Robert Clement (U.S. EPA) Health)
Lesley D’Anglada (U.S. EPA) Jim Taft (Association of State Drinking
Kristin Divris (Massachusetts Department Water Administrators)
of Environmental Protection) Nicole Tucker (U.S. EPA)
Katharine Dowell (U.S. EPA) Tom Waters (U.S. EPA)
Allison Dugan (U.S. EPA) Steve Wendelken (U.S. EPA)
Nicholas Dugan (U.S. EPA) Lloyd Wilson (New York State Department
Michael Elovitz (U.S. EPA) of Health)
Mike Finn (U.S. EPA) Jonathan Yoder (CDC)
Stephanie Flaharty (U.S. EPA) Lester Yuan (U.S. EPA)
Dan Hautman (U.S. EPA) George Zoto (Massachusetts Department of
Austin Heinrich (ORISE participant)2 Environmental Protection)
Susan Holdsworth (U.S. EPA)
1
Lead Preparer
2
Oak Ridge Institute for Science and
Education (ORISE) participant
v
Executive Summary
This cyanotoxin management document is a companion to the United States Environmental
Protection Agency’s (EPA) Health Advisories (HAs) for microcystins and cylindrospermopsin.
Human exposure to cyanotoxins can result in a host of adverse health effects, including
gastroenteritis, liver damage and kidney damage. The HA values represent concentrations in
drinking water below which adverse noncarcinogenic effects are not expected to result from the
ingestion of drinking water. Derivation of the HAs is described in detail in the final EPA HA
documents for these cyanotoxins.
Cyanotoxins can enter drinking water supplies as a result of the growth of harmful algal blooms
(HABs) in surface water sources or ground water sources under the direct influence of surface
water. The formation of algal blooms is dependent upon a number of environmental conditions,
including the presence of nutrients (such as nitrogen, phosphorus), climate, and stratification of
the water source.
This document is intended to assist public drinking water systems (PWSs) that choose to develop
system-specific plans for evaluating their source waters for vulnerability to contamination by
microcystins and cylindrospermopsin. It could also serve as a model for addressing potential
concerns from other cyanotoxins in the future. The document provides a stepwise approach
PWSs could use to inform their decisions on whether and how to monitor and (or) treat for
microcystins and cylindrospermopsin and when and how to communicate with stakeholders. The
stepwise approach includes the following five steps:
• Step one involves conducting a system-specific evaluation for vulnerability to blooms;

• Step two suggests activities for preparing and observing for potential blooms;
• Step three describes monitoring activities to determine whether cyanotoxins are present
in the raw water, and recommended communication and treatment activities if
cyanotoxins are found in the raw water;
• Step four describes monitoring activities to determine whether cyanotoxins are present in
finished water and recommended communication and treatment activities if cyanotoxins
are found; and
• Step five describes continued finished water monitoring (confirming the initial finished
water sample in Step 4), treatment and communication activities if cyanotoxins are found
in the finished water above acceptable levels.
This document provides information and a framework that PWSs may consider if they choose to
develop a system-specific cyanotoxin management plan (CMP). This document includes a three-
tiered, traffic light strategy as an example of an approach that could be used to provide
information to the public about algal toxin levels in local drinking water. Clear and effective
communication is critical to support informed choices about how to best protect the public from
adverse health impacts from HABs. It can also support the states in assisting PWSs and other
stakeholders in their cyanotoxin risk management efforts. The recommendations in this
document may be updated over time as EPA receives new relevant information related to
vi
effective strategies for monitoring, treatment and communication with stakeholders regarding
management of cyanotoxins in drinking water.
vii
A. Introduction
Cyanobacteria, also known as blue-green algae, naturally occur within marine and fresh water
ecosystems. Some cyanobacteria are capable of producing toxins, called cyanotoxins, which
can pose a risk to human health (U.S. EPA, 2014a). Under certain conditions cyanobacteria can
grow rapidly, producing cyanobacterial blooms, often referred to as HABs. A bloom is a rapid
and excessive growth of cyanobacteria (AWWA and WRF, 2015). It is not possible to
determine solely upon visual observation if a bloom is producing toxins, thus any bloom is
potentially dangerous. When blooms occur, the risk of cyanotoxin contamination of the surface
water increases, placing potential risk to drinking water sources (U.S. EPA, 2014a).
The purpose of this document is to provide information to public water systems (PWSs), state
and local authorities, and other stakeholders to assist with the management of cyanotoxin
occurrence in drinking water. This document is a companion to EPA’s HAs for microcystins and
cylindrospermopsin.
This document is not a regulation; it is not legally enforceable; and it does not confer legal rights
or impose legal obligations on any party, including EPA, states, or the regulated community.
This document describes approaches PWSs can consider in developing a system-specific plan.
This document includes recommendations for coordination, preparation, monitoring, treatment
responses and communication. Some aspects of this approach may also be useful when
responding to the occurrence of other cyanotoxins in addition to microcystins and
cylindrospermopsin. The recommendations in this document may be updated over time as EPA
receives new relevant information related to effective strategies for monitoring, treatment and
communication with stakeholders regarding management of cyanotoxins in drinking water.
In some states, primacy agencies may already have existing programs for addressing
cyanotoxins; the information provided in this document can be used to supplement these
programs, as appropriate. EPA also recognizes that states and PWSs may have collected relevant
information through source water monitoring or assessments and may have other information
(for example, watershed-level information) that can help the PWS manage risks from
cyanotoxins in drinking water.
Increases in cyanobacterial blooms are driven by a number of factors, including excess nutrient
loading from anthropogenic sources (Paerl and Otten, 2013; Conley et al., 2009, Glibert et al.
2014) and climate change that produces conditions that favor bloom formation (Paerl and
Huisman, 2009). In addition, Doblin et al. (2007) demonstrated that cyanobacteria that produce
HABs can be transported in ballast water from shipping from a port where active blooms occur
to other locations when ballast water is discharged.
A critical step for reducing human health risks from algal toxins is coordinated action by
multiple groups and organizations to reduce the anthropogenic inputs that can lead to algal
blooms. Reducing nitrogen and phosphorus pollution can reduce cyanobacterial blooms, thereby
reducing treatment costs to utilities and human health risks caused by cyanotoxins in sources of
drinking water (Paerl, 2014). This document discusses tools that can be useful for protecting
source waters from excess nutrient loading.
1
B. What is a Health Advisory?
EPA Health Advisories (HAs) provide technical guidance on health effects, analytical
methodologies and treatment technologies associated with contaminants that are known or
expected to occur in drinking water. Under the Safe Drinking Water Act (SDWA), EPA may
publish Health Advisories for contaminants that are not subject to any national primary drinking
water regulation. 42 § 300g-1(b)(1)(F) 1. HAs are not legally enforceable under SDWA, but serve
as technical guidance to assist federal, state, and local officials and drinking water system owners
and operators in managing drinking water resources and achieving public health goals (U.S.
EPA, 2012a). While EPA recognizes there are multiple water quality concerns related to
cyanotoxins, EPA has only developed drinking water HAs at this time. EPA intends to develop
criteria under Section 304(a) of the Clean Water Act (CWA) to address recreational exposure to
cyanotoxins in the future.
Typically, HA values are developed for One-Day, Ten-Day, and (or) Lifetime exposure
durations. HA values are an estimate of the concentration of a chemical in drinking water that is
not expected to cause any adverse noncarcinogenic effects for the period of exposure. HAs are
intended to serve as informal recommendations for federal, state, and local officials and water
system managers during emergency spills or contamination situations for a specific chemical that
is otherwise not often found in drinking water supplies (U.S. EPA, 2008). A HA value is
determined using the best available information on health effects, exposure and other relevant
data. For more information on HAs, please visit the EPA Health Advisory website (U.S. EPA,
2014b) or view the 2012 Edition of Drinking Water Standards and Health Advisories (U.S. EPA,
2012a).
EPA released Ten-day HA values for microcystins and cylindrospermopsin concurrent with the
release of this document. EPA recommends that systems take actions to protect the public from
exposure to microcystins and cylindrospermopsin as soon as practicable, recognizing that the
response to the detection of cyanotoxins may take a few days. Development of a system-specific
management plan can help water systems prevent cyanotoxin levels from reaching levels of
public health concern in drinking water. The Ten-day HA recommended concentrations for total
microcystins are 0.3 µg/L for bottle-fed infants and young children of pre-school age (less than
six years of age) and 1.6 µg/L for all other ages. The HA for microcystins was developed based
on studies of microcystin-LR; for the purposes of the HA, microcystin-LR is considered a
surrogate for all microcystins. The Ten-day HA recommended concentrations for
cylindrospermopsin are 0.7 µg/L for bottle-fed infants and young children of pre-school age and
3.0 µg/L for all other ages. For more information on Health Advisories for the microcystins and
cylindrospermopsin, please see (U.S. EPA, 2015a,b). For more in-depth discussion of the
science used to develop the Health Advisories, please see the Cyanotoxin Health Effects
Support Documents (U.S. EPA, 2015c,d,e).
1
“The Administrator may publish health advisories (which are not regulations) or take other appropriate actions for
contaminants not subject to any national primary drinking water regulation.” 42 § 300g-1(b)(1)(F)
2
C. Cyanotoxin Management Plan Development
EPA encourages PWSs to carefully consider their potential vulnerability to HABs and to
consider developing a system-specific cyanotoxin management plan (CMP) prior to any
projected algal bloom occurrence. In much of the U.S., blooms typically occur seasonally, but in
some climates blooms can occur throughout the year. PWSs may want to periodically evaluate
and modify their CMPs as their understanding of the specific challenges related to HABs facing
their system evolves.
The first step for some systems to consider in developing a CMP is coordination across the
various parties that would participate in the response to a HAB event. Systems developing CMPs
are likely to benefit from coordination with individuals with experience in all aspects of the
drinking water treatment process including source water intake, sample collection, treatment and
distribution, as well as on-site laboratory personnel or contacts from outside laboratories capable
of analyzing cyanotoxins. Systems are also likely to benefit from coordination with
communications specialists, representatives from state and local public health agencies, state and
local environmental agencies, local government personnel and other local entities that are likely
to be involved in a response to a HAB event. The system may also want to designate a team
leader or other official to identify the roles and responsibilities of each member of the team,
evaluate and update these roles and responsibilities as needed and take responsibility for
overseeing development of the CMP.
This document presents EPA recommendations for one possible approach to developing a CMP,
(see Figures 1 and 2) including components for determining if and when a PWS is vulnerable to
cyanotoxins, monitoring for cyanotoxins, suitable treatment actions, and communication
strategies. There is a model of this approach in section D of this document. The potential
management steps are intended to provide a stepwise process that allows a PWS, as it deems
appropriate, to take action to reduce the likelihood of cyanotoxin occurrence in finished water.
The following paragraphs discuss each of the elements of this CMP approach.
Monitoring
A PWS may benefit from incorporating different types of monitoring in its CMP. Source water
and system observations can inform system decisions about when to start cyanotoxin monitoring
in raw and finished water, when and how to adjust treatment plant operations, and when to
communicate with external stakeholders and the public. EPA does not currently regulate
cyanotoxins and PWSs are not required to monitor for cyanotoxins in their drinking water
(unless required by their primacy agencies). If a PWS decides to monitor, it should consider
maintaining records of any monitoring occurring as part of a CMP, as historical data can be
valuable to a PWS (and nearby systems) for determining their vulnerability to cyanotoxins.
Sampling frequencies are useful to include in a CMP. Monitoring information and related
resources are discussed in each of the potential cyanotoxin management steps.
Treatment
EPA recommends that a PWS identify treatment and management strategies as part of a CMP to
address cyanotoxins in drinking water in the context of its other drinking water treatment goals
(for example, turbidity control, disinfection byproduct (DBP) control, disinfection, taste and odor
3
control, corrosion control, etc.). The PWS may also want to consider potential strategies for
source water protection as well as control measures at the water treatment facilities to remove
cyanotoxins from the drinking water. A PWS can evaluate the existing treatment capabilities and
make short- and long-term improvements as needed before the bloom season so that it can
respond to detections of cyanotoxins in raw and (or) finished water as soon as possible. This
document provides an overview of different treatment and management options, and general
information on treatment adjustments and improvements based on information from published
literature, research reports, guidance manuals and other resources. Additional treatment
information is provided with each step in the potential management steps. More comprehensive
literature review of treatment technologies and a water utilities’ guide can be found elsewhere
(WHO, 1999; Newcombe, 2009; Newcombe et al., 2010; Westrick et al., 2010; AWWA and
WRF, 2015).
Communication
EPA recommends that drinking water systems consider communications to be an integral part of
every step of a CMP. Important communications to consider as a part of a CMP include sharing
of information with the primacy agency, contract laboratories, neighboring drinking water
systems, local officials and the public. For public communications, it may also be beneficial to
ensure that communication strategies take into consideration the media and non-English
speakers, as well as segments of the public that are likely to take the greatest interest in
messaging on cyanotoxins (such as parents of bottle-fed infants and young children of pre-school
age [less than six years old], including bottle-fed infants). Partnerships to aid in communication
with sensitive populations would be helpful to include as part of a CMP, such as with day care
centers and pediatricians.
A useful tool to utilize when developing a communication strategy is the Drinking Water
Advisory Communication Toolbox (DWACT) (CDC, 2013) and EPA’s Developing Risk
Communication Plans for Drinking Water Contamination Incidents (U.S. EPA, 2013a). These
tools can help to prepare for communicating about cyanobacterial blooms and cyanotoxin
occurrence. The toolbox includes tips on what to do before, during, and after issuing public
notices. It also describes with whom to consider collaborating before issuing public notices
and how to work with the media. The risk communication guidance document has information
and templates for communicating with the public during a drinking water contamination
incident. PWSs can also consult with their primacy agency for additional available
communication tools and resources as appropriate. For example, Ohio and Oregon have public
messaging templates available (Ohio EPA, 2014; Oregon Health Authority, 2013). Appendix
D contains potential language for use in cyanotoxin public notification and social media alerts.
Local source water assessment or protection organizations may also be leveraged to
communicate key messages to the drinking water community; a few of these watershed groups
can be found through the Source Water Collaborative “How to Collaborate Toolkit” (SWC,
2015a). Additional communication strategies are discussed throughout this document as the
different steps in the potential management steps are discussed.
4
D. Potential Cyanotoxin Management Steps
Figure 1. Potential management steps public surface water systems may use to determine
whether cyanotoxins are present in raw water or finished drinking water.
Figure 1 depicts potential management steps public surface water systems may use to determine
whether cyanotoxins are present in raw water or finished drinking water. The potential
management steps contain suggested actions for monitoring, treatment and communication. Step
1 involves conducting a system-specific evaluation to determine if and when a PWS is
5
vulnerable to cyanotoxin occurrence. Step 2 involves preparing for possible cyanotoxin
occurrence and observations to detect potential for cyanotoxin occurrence in source waters,
including visual bloom observation, treatment plant effects and source water indicators. If a PWS
observes indications of a HAB that may impact their drinking water system, raw water
monitoring is recommended (Step 3). If raw water monitoring indicates the presence of
cyanotoxins, the PWS may elect to initiate finished water monitoring (Step 4). If cyanotoxins are
detected in finished water, PWSs are encouraged to confirm the presence of cyanotoxins in
finished drinking water (Step 5). Treatment and potential communication actions are provided
based on the concentration of algal toxins detected in finished water. Each of the steps will be
described in further detail in subsequent sections. The primary goal of a CMP is to prevent
cyanotoxins from entering the finished drinking water. The potential management steps follow a
multi-barrier approach starting with source water protection to ensure that public health is not
impacted by the presence of cyanotoxins in finished drinking water. Drinking water systems that
develop a CMP are encouraged to tailor their approach based on local knowledge of bloom
history and growth conditions to ensure effective use of resources and be flexible to
accommodate changing conditions the PWS may experience.
1 Step 1: Conduct System-Specific Surface Water Evaluation
An essential component of the long-term solution to impacts of algal toxins on drinking water
supplies will be effective source water protection strategies to limit excess nutrients in surface
water. Step 1 of the potential management steps involves a system-specific evaluation of source
water to determine if there are vulnerabilities to cyanotoxin occurrence. A variety of information
can be considered in this evaluation including the type of source water, historical cyanotoxin
occurrence, weather data, seasonal patterns of cyanobacterial blooms, land use patterns, nutrient
levels, chlorophyll-a and phycocyanin levels, point and nonpoint sources of contamination
upstream, water quality impairments and information gathered as part of source water
assessments. Determining source water vulnerability to cyanotoxins is important so that the PWS
can be prepared to respond to cyanotoxin occurrence if needed. The outcome of the system-
specific evaluation of source water will help a PWS determine whether to proceed to Step 2 of
the potential management steps (preparation and observations for possible bloom events).
Cyanobacterial blooms can increase the amount of cyanotoxins in source waters to levels that
can be harmful to human health. Blooms occur when conditions in source waters are conducive
to growth based on a variety of factors (WHO, 1999). A PWS may wish to take a weight of
evidence approach to identifying source water vulnerability, as there are no single predictors of
the likelihood of bloom occurrence. Examples of the types of information to review in the
system-specific evaluation include:
• Source Water Characteristics

• Water Quality Parameters
• Source Water Assessment Information
• Climate and Weather Information
• Land Use
• Nutrient Levels
6
Some source waters will have greater vulnerabilities than others based on source water
characteristics. For example, ground water systems (not directly influenced by surface water) are
not anticipated to have vulnerabilities to cyanotoxins. Likewise, fast flowing, nutrient-poor rivers
are less vulnerable than nutrient-rich lakes and reservoirs. Water quality parameters can help to
determine if the source water has had a history of blooms or bloom indicators such as
cyanobacterial cell counts or chlorophyll-a levels. Elevated nitrogen and phosphorus levels will
be important to consider in a system-specific evaluation. Source water assessments, including a
consideration of the predominant land use in the watershed and potential nutrient sources that
may lead to cyanobacterial growth, will provide useful information for a system-specific
evaluation. Similarly, climate and weather information such as water temperature and intensity
of precipitation events can help a system determine if a source water has conditions conducive to
cyanobacterial growth. Each of these factors are discussed in greater detail in Appendix A of this
document. Additionally, links to data sources and tools that may be helpful in conducting a
system-specific vulnerability evaluation are provided in Appendix A.
A PWS may wish to conduct a system-specific evaluation for all drinking water sources utilized
by the PWS and collaborate with other PWSs using the same source water. A PWS may also
consider collaborating with other source water stakeholders who may have additional
information that could be included in a system-specific evaluation. The Source Water
Collaborative’s “How to Collaborate Toolkit”, as discussed in the introduction, is also a useful
resource to help watershed stakeholders form partnerships (SWC, 2015a). EPA recognizes that
some of the information discussed in this section and Appendix A may not be available for every
drinking water source.
EPA recommends systems evaluate available data on their source water to make a weight of
evidence determination about their vulnerability to cyanotoxins. If the PWS determines their
source water is vulnerable, EPA recommends that the system proceed to Step 2 of the process.
If a PWS determines that their source water is not vulnerable to cyanotoxins, the PWS may want
to consider periodically reassessing its source water as watershed characteristics could change
over time.
2 Step 2: Preparation and Observation

If a PWS has determined they have vulnerabilities to cyanotoxins, EPA recommends that the
PWS consider preparing and observing for possible cyanobacterial blooms and cyanotoxin
occurrence. A PWS can prepare by determining when blooms are most likely to occur,
evaluating the current treatment process to determine susceptibility and vulnerabilities, and
determining if long-term mitigation strategies are available to prevent the blooms from occurring
and cyanotoxins from reaching drinking water sources. A PWS can observe source waters for
possible bloom occurrence by visual inspection, evaluating system effects, and recognizing
bloom indicators.
7
2.1 Preparation
Seasonal Variation
If the PWS determines that their watershed is vulnerable to HABs, it will be useful for the PWS
to determine when their source water is likely to be at greatest risk for the presence of
cyanotoxins. Cyanobacteria and cyanotoxin occurrence can have seasonal variation depending
on the location and condition of the watershed. In many parts of the U.S., this peak season occurs
from late May to early October, but may be longer or shorter, depending on local conditions.
Some cyanobacteria can persist during colder temperatures; for example, in some cases, the State
of Ohio has documented peak toxin concentrations occurring in November and December.
Climate conditions and short-term weather events can greatly impact the timing and duration of
bloom occurrences. If the season of greatest vulnerability is unknown in a watershed that has had
previous bloom events, monitoring can help the PWS determine when this is likely to occur.
EPA encourages the PWS to use all available information and consult with the primacy agency
to perform the system-specific evaluation described in this section. The combined information
can help the PWS determine if and when blooms may occur. Once the PWS has identified the
timing of greatest vulnerability to cyanotoxins, the PWS can take steps to observe for possible
blooms during this period (included in this Step).
Existing Treatment Evaluation
If a PWS finds that it is vulnerable to cyanotoxins, it may want to consider evaluating whether it
has effective measures in place to respond to cyanotoxins in drinking water that are compatible
with meeting other treatment goals. For example, the PWS may want to examine its raw water
supply and treatment process to determine the likelihood of toxin release from intact cells either
in the reservoir or at the raw water intake and the level of protection provided by the existing
treatment (WHO, 1999). Examples of treatment evaluation questions that could be asked include:
• Is the existing treatment a conventional coagulation, clarification and filtration process

that is likely to be effective for intact cell removal?
• Can the existing system handle more frequent backwashing and more sludge in the event
of a cyanobacterial bloom?
• Are there any conditions such as pre-oxidation that could lead to cell lysis?
• Is chlorination being operated adequately to oxidize cyanotoxins?
• Can powdered activated carbon (PAC) be added at adequate doses?
• Are there any advanced water treatment facilities with ozonation and (or) granular
activated carbon (GAC) that can be used to effectively remove dissolved toxins?
A PWS that is particularly concerned with the potential presence of algal toxins in their source
water may want to consider performing bench and pilot studies to simulate the full-scale system
operation under a cyanobacterial bloom condition.
In general, systems that are impacted by cyanotoxins on a seasonal basis may want to consider
the use of temporary supplementary treatment, such as PAC, which can be added intermittently
to the existing treatment process as part of an immediate response to cyanotoxins in drinking
8
water. Systems that are impacted by cyanotoxins on a recurring basis or throughout the year may
want to consider installing permanent treatment as a long-term, cost-effective alternative, such as
GAC, ozonation, and membrane filtration, if they have not already done so. It is important to
remember that all treatment options have specific trade-offs that must be considered before
implementation (U.S. EPA, 2014d). More information on treatment options are described later in
this document, as well as in EPA’s HAs for microcystins and cylindrospermopsin (U.S. EPA,
2015a,b).
2.2 Observation
If a PWS has determined that its source water is vulnerable to cyanotoxins, the PWS may want to
begin regular observation for possible blooms (Step 2). A bloom can have extremely high cell
densities of phytoplankton (extremely high densities are typically defined as greater than 20,000
to 100,000 cells per mL). Proliferation of phytoplankton is typically dominated by a single or a
few species (Loftin et al., 2008). Cyanobacterial blooms may float on the water surface or be
mixed throughout the photic zone, epilimnion or water column (Loftin et al., 2008).
There are multiple indicators of the potential presence of a HAB, including: 1) visual indicators
and phytoplankton identification, 2) system effects and 3) other bloom indicators. Visual
indicators include both visual confirmation of a bloom at or near the raw water intake, and
confirmed reports of blooms by the public and phytoplankton identification. Routine microscopic
phytoplankton identification can provide information when blooms are not visually apparent and
to help determine the type of bloom. System effects of bloom occurrence involves a weight of
evidence approach looking at multiple elements within the drinking water treatment system such
as increased pH, shortened filter run times or taste and odor events. Potential source water
indicators of bloom events include increased turbidity, increased nutrient levels, increased
cyanobacterial cell counts, increased chlorophyll-a or phycocyanin levels and increased
temperature in source water. Each of these parameters is discussed in greater detail below.
Tracking multiple indicators can help the PWS prepare for cyanotoxin occurrence in source
water.
If no indication of a bloom has occurred, EPA encourages the PWS to continue observing for
possible blooms throughout the vulnerable season determined previously as part of this Step. If
any of the three types of observations indicates a bloom is occurring near the source water
intake, the PWS may want to begin monitoring its raw water for toxins (Step 3).
2.2.1 Visual Inspection and Phytoplankton Identification
What is visual inspection?
Visual inspection involves looking for visual signs of a possible bloom, such as water clarity,
discoloration and scum formation near a water intake (Newcombe, 2009). Colors can range from
grey or tan, to blue-green or bright blue or reddish. The appearance of blooms may also be
described as fine grass clippings or small clumps. In general, a healthy cyanobacterial scum will
appear like bright green or olive green paint on the surface of the water. Scums look blue or red
in color when some or all of the cells are dying and release their pigments into the surrounding
water (Newcombe et al., 2010). Some cyanobacteria produce a distinctive earthy and musty
odor, related to the production of geosmin or Methylisoborneol (MIB) that can often be smelled
9
at some distance before the bloom can be seen. Not all blooms will give off a recognizable odor,
as many cyanobacteria are not capable of producing taste and odor compounds. Visual inspection
provides valuable information on cyanobacterial growth and is an important part of any
monitoring program (Newcombe et al., 2010).
Who should conduct a visual inspection?
A bloom may be observed or reported by the PWS operators, state or local stakeholders, or the
general public. Properly trained professional staff might conduct field inspection of drinking
water sources regularly throughout the year, focusing on specific seasons the source water was
previously determined to be vulnerable.
The general public is also encouraged to notify their states or the PWSs as soon as they see a
bloom. For example, Ohio EPA encourages individuals reporting potential blooms to fill out a
Bloom Report Form on their website and email the form, with attached digital photographs (if
available), to a designated mailbox (Ohio EPA, 2014). Individuals are encouraged to report the
bloom location, color, size, and appearance, nearby public beach or drinking water plant
intake(s) (if any), as well as any other available water quality information. The State of New
York has a Suspicious Algae Bloom Report Form and a program for citizens to send in photos of
the suspected blooms (New York, 2015). The PWS may want to consider creating outreach tools
for their community to educate the public on blooms and what to do if they see a bloom. PWSs
should also consider partnering with outside organizations, such as the local health department,
who may be conducting recreational water monitoring for potential blooms and can inform the
PWS when blooms may be headed towards intake structures.
When, where, and how often should visual inspection be conducted?
A PWS may want to conduct a visual inspection regularly throughout the vulnerable season as
determined by the PWS in the system-specific evaluation (Step 1). This peak season usually
occurs in the Midwest from late May to early October, but may be longer in other areas,
depending on local conditions. The distribution of cyanobacteria depends on the morphological,
hydrogeological, meteorological and geographic characteristics of a given water body. For
example, accumulations are normally observed at the downwind end of a reservoir, lake or river
reach. Cyanobacteria may also migrate throughout the water column during the day; inspection
could focus on those times the bloom is most likely to be at the surface. A visual tool that could
be used includes the Secchi disk, a tool used to measure water transparency (Tilzer, 1988). A
change in Secchi disk depth over time (where the water becomes more turbid or has decreased
clarity) could indicate bloom occurrence (WHO, 2000).
The frequency of visual inspection may vary depending on seasonal and weather conditions.
PWSs may want to conduct a visual site inspection at daily, weekly, or biweekly intervals during
bloom season and increase to more frequent intervals if cyanobacteria begin to proliferate.
Limitations of visual inspection
The visual inspection method may not detect some cyanobacteria, such as Planktothrix or
Cylindrospermopsis, which do not form a scum or those cyanobacteria that bloom along the
thermocline (Newcombe, 2009). The only indication of a Planktothrix or Cylindrospermopsis
10
bloom may be a slight green or brown discoloration of the water. For non-bloom forming
cyanobacteria, it is important to collect samples for analysis to determine the abundance of
cyanobacteria in the water body if the PWS believes it is vulnerable to cyanotoxin occurrence.
Cyanobacterial blooms may be confused with scums or mats of filamentous green algae. There
are blooms of other phytoplankton that look very similar to cyanobacterial blooms, but these
cannot be readily distinguished without microscopic evaluation. It is important to note that not
all blooms will produce cyanotoxins, and cyanotoxins can occur when blooms are not visible. It
may be useful for systems to conduct some monitoring if conditions are favorable for the
production of algal blooms but visual inspection is insufficient to determine the potential
presence of algal toxins in source water.
Routine Phytoplankton Monitoring
In addition to visual inspection, water systems may want to consider routine phytoplankton
identification. Routine phytoplankton monitoring can:
• Help visually distinguish green algae and diatom blooms from potentially harmful
cyanobacterial blooms.
• Provide information on cyanobacteria that may be present at intake depths, but not
visually apparent at the water surface.
• Catch Planktothrix and other non-scum forming blooms that would be less likely to be
visually apparent.
• Detect lower concentrations of cyanobacteria that may not be visually apparent. This
could aid in reservoir management strategies and help trigger early treatment
adjustments.
• Aid in identifying targets for monitoring of cyanotoxins in raw and finished water.
• Help meet other treatment objectives. Increases in diatoms can cause fishy odors, clog
filters, increase chlorine demand and lead to problems with disinfection byproducts.
A PWS should ensure that staff are properly trained prior to attempting phytoplankton
identification.
2.2.2 System Effects
A bloom may impact water quality and treatment plant operations. The changes commonly
associated with a cyanobacterial bloom are listed below. If a PWS determines that a bloom may
be occurring through a weight of evidence evaluation of system effects, the PWS may want to
proceed with raw water monitoring (Step 3), even if the plant intake does not show visual signs
of a bloom being present.
• Increased taste and odor. Some cyanobacteria can produce MIB and geosmin, which
can cause taste and odor issues within the treatment plant and distribution system during
cyanobacterial blooms. Other taste and odor impacts may not be associated with
cyanobacteria (for example a fishy odor is typically associated with a diatom bloom,
which would not produce cyanotoxins).
• Increased pH. As cyanobacteria draw carbon dioxide out of the water during
photosynthesis, pH tends to rise. Therefore, pH increases are more often observed when
11
cyanobacterial growth is expanding in water. Diurnal pH swings may also occur. As
cyanobacteria draw carbon dioxide out of the water during photosynthesis, pH may
increase throughout the day. Increases in pH could also be caused by green algae blooms,
but routine phytoplankton monitoring would help distinguish between the two.
• Increased turbidity. Treatment plants may experience higher turbidity, in some cases, in
filter influent and effluent due to cyanobacterial growth and cell production.
• Decreased filter run times. Treatment plants may experience shortened filter run times
during cyanobacterial blooms to varying degrees depending on the species. For example,
although not entirely the result of cyanobacteria, one water treatment plant reported that
the average filter run time was every 24 to 48 hours in the summer – notably less than the
72 hours or more in the winter (Kommineni et al., 2009).
• Need for increased coagulant dose. A higher coagulant dose is often needed, potentially
due to increased turbidity and total organic matter in water during a cyanobacterial bloom
or the tendency for some cyanobacteria to float and inhibit settling. One treatment plant
reported an approximately 50 percent increase in its average alum dose in the summer
(Kommineni et al., 2009).
• Increased chlorine demand or decreased chlorine residual. The increased organic
matter loading during a cyanobacterial bloom, if not adequately removed, could result in
a higher chlorine demand. Treatment plants reported that the chlorine residual was
decreased during algal growth events (Kommineni et al., 2009).
2.2.3 Other Bloom Indicators
Environmental monitoring of physical, chemical and biological variables indicating bloom-

formation potential is important but can be resource intensive if data are not already available.
Those key indicators can include cyanobacterial cell counts, biovolumes (the volume of cells in a
unit amount of water, mm3/L), chlorophyll-a and phycocyanin concentrations, presence of
cyanotoxin production genes in source water, nutrient concentrations (nitrogen and phosphorus),
changes in hydrophysical conditions and new weather patterns (such as increased temperature
and (or) rain) (Izydorczyk et al., 2005; Ohio EPA, 2014). National Oceanic and Atmospheric
Association (NOAA) satellite imagery data are also being used to predict blooms in western
Lake Erie and the Gulf of Mexico (Florida and Texas). NOAA, U.S. EPA, United States
Geological Survey (USGS), and NASA are partnering on the Cyanobacteria Assessment
Network (CyAN) project, which includes making satellite data processed for cyanobacteria
abundance available for large inland lakes nationwide once new satellite sensors come online and
the data has been evaluated. EPA encourages PWSs to use all available data, as discussed in Step
1, as part of a weight of evidence approach to determine if recent changes have occurred,
possibly indicating bloom occurrence. Bloom indicators can be used to inform a decision
whether a PWS should proceed with toxin analysis. The World Health Organization has a
resource available that contains some details on specific parameters such as cyanobacterial cell
counts or chlorophyll-a levels, which may indicate a severe bloom compared to a moderate or
minor bloom (WHO, 1999).
The partnerships discussed in section 1 could be helpful for gathering data on the occurrence of
certain HAB indicators. For example, having partnerships in place with watershed stakeholders
could allow for quick dissemination of monitoring information and potential changes in source
water quality within the watershed. As an example of one possible volunteer monitoring
program, the Sierra Club of Ohio has a Water Sentinel Program, which enlists volunteers to
12
monitor waterways for a variety of pollutants (Ohio Sierra Club, 2015). Another example of a
volunteer monitoring program is the Ohio Lake Management Society’s Citizen Lake Assessment
and Monitoring (CLAM) program. Citizen-collected cyanotoxin data from this program
triggered cyanotoxin sampling at a public water supply that was not experiencing any operational
or water quality issues at the time. Even though a visible bloom was not apparent at the intake,
the water system sampling revealed microcystins were present at the intake depth, and continued
to be detected at elevated concentrations for several months (OLMS, 2014). The U.S. Army
Corps of Engineers (USACE) also has an extensive lake monitoring program that could provide
valuable information to water systems with USACE source waters (USACE, 2015).
2.3 Communication
If any of the key indicators suggest the potential for bloom formation, the PWS may want to
begin sampling the raw water for cyanotoxins (Step 3). Systems may also want to consider
communicating with the primacy agency and state and local officials to make them aware of the
potential bloom so that those agencies can be ready to assist if cyanotoxins are detected at the
PWS. Communications might include notifying the laboratory of pending raw water samples (if
using an outside laboratory), and informing the system’s public affairs personnel so they can be
prepared in the event cyanotoxins are detected at levels of concern in the finished water. The
PWS may also want to consider informing other users of the source water to let them know that a
bloom has been detected or has likely occurred, such as for recreational purposes or animal uses.
If a PWS determines they may be impacted by a bloom, the PWS could consider working with
public health officials to raise awareness about HABs. Information on source water protection
activities that could reduce the likelihood of HABs could also be provided.
2.4 Source Water Mitigation

Systems that have observed a possible bloom in the source water may also want to consider
taking initial actions to eliminate or mitigate the bloom before it impacts the drinking water
intake. Various treatment and management strategies are available to control cyanobacterial
blooms. EPA recommends PWSs consult with their state and local governments and primacy
agencies as some of these treatment and management strategies discussed below could have
various requirements, such as permit requirements, as well as unintended impacts on the source
water.
Intake Relocation and Alternative Source

Some systems have the ability to adjust the depths of their intakes (WHO, 1999) and draw from
multiple intake depths at one intake tower, which can be used to minimize the intake of
cyanobacterial cells that have accumulated on the surface or at certain depths (Newcombe et al.,
2010). For example, a reservoir may have multiple intake depths to choose from and can use
water quality monitoring data to determine which intake to utilize. However, this may not be an
option for avoiding cell uptake in some shallow waters (Vermont Environmental Conservation,
2014). Also, some systems may be able to locate their raw water intakes away from areas where
blooms have accumulated, such as sheltered bays, or provide temporary extensions to existing
intakes (WHO, 1999). Some systems may have multiple reservoirs, and can discontinue use of
13
one source during a bloom event and rely on other sources, including blending with ground
water.
Bypassing pre-sedimentation ponds
Some systems on rivers have pre-sedimentation ponds or reservoirs. Many of these pre-
sedimentation ponds were not designed to be drained and cleaned and are now their own sources
of nutrients. Even when the original source waters do not contain blooms, it is possible that
conditions within the ponds can support the formation of blooms. Systems that are experiencing
bloom formation in pre-sedimentation ponds can develop a process to bypass these pre-
sedimentation ponds for the duration of the bloom. Long-term solutions include eliminating pre-
sedimentation ponds from the treatment train, dividing the pond or reservoir in half, or installing
drains and linings. This will allow the system to take one side out of service while power
washing and draining the sediments out of the other side.
Ultrasonic Treatment
Ultrasonic treatment has the potential to help prevent blooms from forming and has been used as
an inexpensive measure in some instances to disrupt gas vesicles within the cells, as well as to
interfere with photosynthesis and cell division (Rajasekhar et al., 2012). While this treatment
demonstrated a 93.5 percent destruction of M. aeruginosa when coupled with coagulation
(Tokodi et al., 2012), in some cases this treatment can increase the release of intracellular toxins,
(U.S. EPA, 2014d) and practicing coagulation in the raw water source may be difficult.
Therefore, PWSs are encouraged to consult with their state and local governments before
installing this treatment in the source water.
Algaecides
One strategy that has seen widespread use is the addition of algaecides to the source water,
which may kill off the cyanobacteria and prevent operational problems in the water treatment
plant. Examples of algaecides include copper sulfate, copper citrate and hydrogen peroxide
formulations (U.S. EPA, 2014d). However, EPA does not encourage the use of algaecides in
drinking water sources. Until recently algaecides were added without a complete understanding
of the potential environmental concerns (for example, toxicity to aquatic life). Copper-
containing compounds may create water quality concerns for both the aquatic environment and
the drinking water source. Furthermore, microcystin-producing cyanobacteria have been
demonstrated to have the potential to develop resistance to copper if treatments are repeatedly
applied (Garcia-Villadra et al., 2004).
Another concern with using algaecides in general is that cell death can lead to the release of the
intracellular toxins; hence, algaecides are only recommended to be used as an emergency
measure in the early stages of a bloom, when the resultant toxin concentrations that may be
released are likely to be low (U.S. EPA, 2014d). Australian guidance recommends against using
copper sulfate due to the potential ecological effects (Newcombe, 2010). The World Health
Organization suggests using copper sulfate only in dedicated water supply reservoirs (WHO,
1999). To protect against environmental concerns, some state governments require coverage
under a general permit prior to applying algaecides to a source of drinking water, and some states
may have specific algaecide prohibitions. PWSs should consult with their state and local
governments (and primacy agency) before the application of algaecides.
14
Coagulants
The addition of a coagulant (such as alum) to the source water has been shown to lead to the
precipitation of phosphorus to the source water’s sediment layer and can coagulate cells out of
the water. There are mixed opinions on whether cell lysis results from coagulation applied in the
source water, and there have been reported depth limitations. Once the phosphorus is settled, the
sediment can be capped to prevent re-release (U.S. EPA, 2014d).
Skimming
Skimming the surface of a source water containing a bloom can remove the cells but the
effectiveness is dependent on the species of cyanobacteria present. This strategy is often used
as an emergency measure to respond to later stages of a bloom (U.S. EPA, 2014d), but may
also present a possible additional strategy for the initial response to early bloom detection.
Aeration
Aeration can be an effective cyanobacteria management tool in source water. Aeration pumps air
through a diffuser near the bottom of a source water body, releasing a plume that rises to the
surface. This plume causes mixing of the water column that disrupts the migration behavior of
the cyanobacterial cells and limits the accessibility of nutrients. Aeration has been shown to be
successful in small water bodies, but is highly dependent on airflow rate and the degree of
stratification of the water body (U.S. EPA, 2014d).
Mechanical Mixing
Typical mechanical mixers are surface mounted, and either move water from the surface
downwards, or draw water from the bottom to the surface. This mixing of the water column
disrupts the cyanobacteria migration and limits the availability of nutrients. Mechanical mixing
has been found to have some success in water bodies in the U.S. The devices can have a
limited range, so areas further away from the device may remain stratified (U.S. EPA, 2014d).
3 Step 3: Monitor for Cyanotoxins in Raw Water and Treatment

Adjustments
3.1 Sampling and Analysis for Cyanotoxins
3.1.1 Raw Water Sampling
If a cyanobacterial bloom is observed and (or) inferred by means of visual inspection, system
effects, or other bloom indicators (Step 2), EPA encourages the PWS to sample the raw water for
cyanotoxins (Step 3). EPA suggests the PWS collect raw water samples at the plant intake prior
to any treatment. Samples that may have been exposed to chlorine or other oxidants should be
quenched immediately upon sampling. Temporary surface blooms may be observed early in the
morning, but some blooms may disperse as winds increase and (or) mix back into the water
column during the day. PWSs may want to consider sampling at the worst case area of the bloom
in the source water around the intake to determine the maximum levels of toxins produced. If
water systems are contemplating algaecide application, the surface scum (if visible) should also
15
be sampled to provide an indication of the potential for release of toxins as a result of the
algaecide application.
Where a bloom is observed or inferred, PWSs may want to consider sampling the raw water at
least two to three times per week. If there is limited evidence of a bloom, or the source water
does not have a history of periodic cyanobacterial blooms, systems could consider monitoring
less frequently and increase monitoring if cyanotoxins are found in raw water. A PWS is
encouraged to choose a raw water sampling frequency considering the following factors: past
frequency of occurrence of blooms and cyanotoxins in the water source or nearby water bodies;
current toxin concentrations in source water (elevated concentrations could trigger increased
sampling frequency), bloom dynamics (history of highly variable toxin concentrations, impacts
of wind-induced mixing, currents, etc.), characteristics of the water body (for example, size,
depth, thermal stratification); source water quality (for example, nutrient levels); growth rate of
the cyanobacteria; weather and seasonal influences (for example, temperature rainfall); and
adequacy of treatment and capacity of the treatment plant to treat cyanotoxins.
If monitoring results indicate the presence of cyanotoxins in the raw water, EPA recommends
that the PWS continue to Step 4, monitoring for cyanotoxins in raw and finished water. EPA
encourages the PWS to conduct sampling under Step 4 within 24 hours after the detection of
cyanotoxins in the raw water collected under Step 3. Furthermore, the PWS may want to
communicate with their stakeholders as described in section 3.2 and adjust treatment as
described in section 3.3. If no cyanotoxins are found in the raw water, the PWS may want to
continue to observe for possible blooms (Step 2), unless a bloom is observed visually. In cases
where raw water monitoring (Step 3) is triggered by visual confirmation of blooms near the
intake (that is microscopically confirmed to be caused by cyanobacteria), EPA encourages the
PWS to continue raw water sampling until the bloom is no longer visually identifiable.
3.1.2 Sampling Logistics
Samples should be handled properly to ensure reliable results, whether analyzing the samples
using a field kit or shipping to a laboratory. EPA recommends that a PWS follow sample
collection and handling procedures established by the method or laboratory performing the
analysis (U.S. EPA, 2014a). For laboratory analysis, EPA encourages the PWS to use laboratory-
provided sample containers to collect water samples. Laboratories may not accept containers not
provided by the laboratory, or they may invalidate results. Amber glass containers are typically
used to avoid potential cyanotoxin adsorption associated with some plastic containers and to
minimize exposure to sunlight (U.S. EPA, 2014a). Raw water samples that have been exposed to
any oxidants should be quenched immediately upon sampling. Samples should be cooled
immediately after collection, during shipping, and pending analysis at the laboratory. Ideally,
samples should be shipped on the same day they are collected. Samples generally should be
analyzed within five days from the time of collection. EPA encourages systems to contact the
appropriate laboratory prior to shipping samples for additional sample handling instructions.
More information is available in USGS Guidelines for Design and Sampling for Cyanobacterial
Toxin and Taste-and-Odor Studies in Lakes and Reservoirs (2008).
16
3.1.3 Analytical Methods
Analysis of cyanotoxins may require sample preparation, depending on the form of the
cyanotoxins (intra- or extracellular), and the specific analytical methods (U.S. EPA, 2015a,b).
Analytical methods measure dissolved (extracellular) cyanotoxins. Therefore, to determine total
(intracellular and extracellular) cyanotoxin concentrations, sample preparation should include
cell lysis so that the intracellular toxins can be quantified (U.S. EPA, 2015a,b). It can be helpful
to analyze for both extracellular toxin (may involve an additional filtration step) as well as total
toxin concentrations for a raw sample, to inform treatment adjustments.
Enzyme-linked immunosorbent assays (ELISA) are commonly used to detect cyanotoxins.

ELISA results quantify total microcystins and can directly be compared to the HA level for total
microcystins. ELISAs are non-specific in that they cannot, however, identify and quantify
various individual microcystin variants (sometimes referred to as “congeners”). For raw water
monitoring (Step 3), ELISA field kits could be used if they can meet the quantitation limits the
PWS or state deems necessary as outlined by a CMP or alternative cyanotoxin management
approach. These kits can provide rapid results of the potential presence and semi-quantitative
amounts of microcystins or cylindrospermopsin in raw water samples.
The PWS could also consider using a quantitative laboratory ELISA test for total microcystins,
such as the ADDA specific ELISA. Standard Operating Procedures for this method developed by
Ohio EPA are a useful resource in providing additional helpful advice for quality-control and
sample-handling measures (Ohio EPA, 2015). For additional information on analytical methods
for microcystins and cylindrospermopsin, please also see the analytical methods discussion
under Section 5 of the HAs (U.S. EPA, 2015a,b). To assess the performance of unit processes
within treatment plants, the same analytical method should be used for the paired raw and
finished water samples.
3.2 Communications
If a PWS detects cyanotoxins in the raw water, the PWS may want to consider communicating
their findings with state and local officials. The PWS may also want to consider working with
public health officials to develop messaging for the public on the proactive measures the PWS is
taking. This would keep the public health officials aware of the nature and degree of the concern
prior to detections in finished water. The PWS could also prepare a communication message for
the public in the event the PWS receives public inquiries on visible blooms in source waters. If
the source water has public access, such as for recreational purposes or animal uses, the PWS
may want to consider informing other source water managers to let them know cyanotoxins have
been detected.
3.3 Treatment
If cyanotoxins are detected in the raw water in Step 3, it is important to determine whether they
are intracellular or extracellular before a PWS begins implementing any treatment strategies.
Raw water samples collected in this Step should be analyzed for both extracellular and total
(intracellular and extracellular) toxins to inform proper treatment strategies. Information on how
to collect and analyze samples for both total and extracellular toxins is provided in section 3.1.
17
The following treatment strategies are commonly used by PWSs to respond to cyanotoxins in
raw or finished drinking water:
1. Removing intact cells first

2. Minimizing pre-oxidation of raw water
3. Adding or increasing powdered activated carbon
4. Increasing post-chlorination
These treatment strategies can be implemented easily and quickly to provide immediate response
to any cyanotoxins detected in raw or finished water and prevent cyanotoxins from breaking
through into treated water. A system that has detected cyanotoxins in their source water will
likely not be able to design, construct and start up a new permanent treatment system in time to
address the HAB.
When selecting these treatment strategies, it is important for a PWS to evaluate any potential
impacts on their ability to meet all other treatment goals (for example, turbidity removal, DBPs
precursor control, disinfection, taste and odor control, corrosion control, etc.) and associated
operational issues (such as filter backwash and sludge handling). Each of the treatment strategies
is discussed individually below in more detail.
Treatment Strategy 1: Removing intact cells first
Conventional treatment, defined as coagulation, flocculation, sedimentation and filtration

processes, is widely used by surface water treatment plants to reduce particulate material.
Conventional treatment can be very effective to remove whole cells that contain intracellular
cyanotoxins (WHO, 1999; Newcombe, 2009; Newcombe et al., 2010; Newcombe et al., 2015). If
cyanotoxins are released from the cells (the extracellular form), treatment becomes more
complicated and costly because conventional treatment has limited ability to remove
extracellular cyanotoxins (WHO, 1999; Westrick et al., 2010; Hitzfeld, et al., 2000). Therefore,
ensuring that cells are not lysed before they are removed should be considered as the first
treatment response taken by a PWS during a cyanobacterial bloom (Newcombe et al., 2015).
Operational considerations for removing whole cells through the conventional treatment process
are similar to those considered for achieving effective particulate removal, such as coagulants,
pH, and mixing (Newcombe et al., 2015). The source water quality and cyanobacterial
morphology (such as individual cells, filamentous, etc.) also strongly affect the treatment
efficiency (Kommineni et al., 2009; Newcombe, 2009). Detailed operational guidance on how to
improve the removal of cyanobacterial cells (containing intracellular toxin) by conventional
treatment can be found in Newcombe (2009) and Newcombe et al. (2010; 2015).
During a cyanobacterial bloom, water plant operators may want to consider altering treatment
processes and operational parameters (such as chemical doses, coagulation, pH, filter backwash
frequency and loading rates) to account for the increased loading of whole cells. Jar tests are
commonly used to simulate conventional treatment and can be used to help determine
appropriate coagulation chemicals, pH level, and operational parameters at a coagulation and
filtration plant to improve toxin reduction. Jar tests are relatively simple, low-cost, and can be
completed in a short timeframe (Lytle, 1995). When conducting jar tests, the operator can
18
measure the decrease in settled water turbidity (for high turbidity water > 10 Nephelometric
Turbidity Units (NTU)), chlorophyll-a (or phycocyanin), or cell count as a surrogate for cell
removal (Sklenar et al., 2014; Newcombe et al., 2015). However, the use of turbidity as an
indicator of treatment efficiency is not recommended for low turbidity water (< 10 NTU) due to
poor correlations between the decrease in turbidity and cell removal (Newcombe et al., 2015). A
PWS is encouraged to conduct its own tests to determine appropriate treatment parameters
before full-scale application. Once applied to the full-scale operation, the PWS is encouraged to
measure toxins by ELISA to help understand what is happening in the treatment train or identify
treatment breakdown.
Residuals produced by the conventional treatment process, including sludges and backwash
water, should be properly handled and removed from the system to minimize the release of
intracellular and extracellular toxins into the surrounding water (Kommineni et al., 2009;
Zamyadi et al., 2012; Newcombe et al., 2015). During a cyanobacterial bloom, treatment sludge
should be frequently removed from the sedimentation basin (daily if possible) and any sludge or
backwash supernatant recycling should be discontinued until the cyanotoxins have degraded or
been diluted (Newcombe et al., 2015). Cells can accumulate in the filters, which can potentially
lead to a significant amount of extracellular microcystins released into the filtered water.
Frequent backwashing has been recommended to minimize the risk of cells breaking through
into filtered water (Newcombe, 2009). Additionally, backwash water from the filters may
contain cyanobacterial cells and (or) extracellular microcystins. Therefore, operators may wish to
consider a filter-to-waste cycle following backwashing that is long enough to flush out any
residual toxins remaining in the filter.
Treatment Strategy 2: Minimizing pre-oxidation of raw water
Oxidants, such as chlorine, ozone or potassium permanganate, applied to raw water containing
intact cells can lyse cells or stimulate the release of intracellular toxins in un-lysed cells,
resulting in the release of cyanotoxins. However, the amount of oxidant dosed may not be
sufficient to oxidize the released toxins. Therefore, caution should be taken when using pre-
oxidation. For example, Australian guidelines recommend not practicing pre-chlorination or pre-
ozonation without additional processes to remove the released toxins (Newcombe et al., 2010),
or adding a sufficient dose to oxidize the cyanotoxins as well as to lyse the cells (Newcombe et
al., 2015). Another study indicated that pre-oxidation would only be considered acceptable when
the total intracellular and extracellular toxin concentrations are so low as to be irrelevant (House
et al., 2004).
Some water systems practice pre-oxidation with potassium permanganate followed by PAC to
improve coagulation, control zebra mussels and (or) reduce taste and odor compounds. However,
EPA’s research demonstrated that potassium permanganate at low levels (for example, at a dose
of 1 mg/L) could have the potential to stimulate the release of intracellular toxins from
cyanobacteria, thus increasing potential downstream risks. Therefore, the PWS can consider
discontinuing the pre-oxidation with potassium permanganate during a cyanobacterial bloom.
Treatment Strategy 3: Adding or increasing powdered activated carbon
PAC is regarded as an effective treatment for microcystins and cylindrospermopsin as well as

taste and odor compounds (Westrick et al., 2010; Cyanocenter UBA, 2015). PAC is generally
19
considered an episodic treatment for cyanotoxins and can be added intermittently to the
conventional treatment process to respond to periodic or seasonal spikes of cyanotoxins in a
fairly cost-effective approach (Westrick et al., 2010). PAC can be added either at the intake (after
pre-oxidation with potassium permanganate or other oxidants) and removed during clarification,
or it can be added to the settling tanks and removed through filtration. When placed after
potassium permanganate, PAC adsorbs not only the released toxins but also the permanganate
residual. Thus, a higher PAC dose may be needed. EPA is conducting research to provide water
utilities with a better understanding of how pre-oxidation with potassium permanganate followed
by PAC affect cyanobacterial cells and their toxins.
The efficiency of PAC adsorption depends on the carbon type (for example, pore size) and the
presence of natural organic matter (NOM) in water (U.S. EPA, 2014a). Mesoporous carbon
(such as wood-based PAC) has been demonstrated to be the most effective at removing
microcystins and cylindrospermopsin (WHO, 1999; U.S. EPA, 2014a; Sklenar et al., 2014;
Westrick et al., 2010; Drikas et al., 2002). A PAC dose in excess of 20 mg/L would be required
for cyanotoxin removal, significantly higher than what is typically used in drinking water
treatment (Jurczak et al., 2005; Tokodi et al., 2012; U.S. EPA, 2014a). Since the removal varies
by carbon type, source water quality and treatment objectives, systems may consider conducting
jar tests to select the appropriate PAC type, dose and feed point prior to full-scale application
(Sklenar et al., 2014).
Treatment Strategy 4: Increasing post-chlorination
In general, microcystins are readily oxidized by chlorine (Newcombe et al., 2010). Effectiveness
of chlorine in microcystin oxidation is highly dependent on pH, temperature and the initial
microcystin concentration (Acero et al., 2005; Ho et al., 2006a). Oxidation of extracellular
cyanotoxins is most effective when the pH is below 8 (Acero et al., 2005; U.S. EPA, 2014a).
This is especially the case for microcystins (Westrick et al., 2010; Acero et al., 2005).
Cylindrospermopsin can be effectively oxidized by chlorine when the pH ranges from 6 to 9
(Westrick et al., 2010). Chloramines and chlorine dioxide are not effective for microcystins or
cylindrospermopsin oxidation at typical contact times used in drinking water application (U.S.
EPA, 2014a; Nicholson et al., 1994; Westrick et al., 2010).
It is important to understand the effectiveness of the existing chlorination process for oxidation
of cyanotoxins. Disinfectant effectiveness (to inactivate pathogenic organisms, such as Giardia),
is commonly expressed as CT (in mg/L-min), which is calculated by multiplying the disinfectant
concentration (C, in mg/L) by the contact time (T, in minutes). Researchers developed chlorine
CT values for oxidizing microcystin-LR (Acero et al., 2005) and compared these CT values with
those needed for inactivating Giardia cysts (Acero et al., 2005; Ho et al., 2006a; Westrick,
2008). (See Section 4.10 of the “Drinking Water Health Advisory for the Cyanobacterial Toxin
Microcystin” (U.S. EPA, 2015a) for the use of microcystin-LR as a surrogate for total
microcystins). Table 3-1 shows the chlorine CT values required to reduce microcystin-LR
concentrations from 50 or 10 μg/L to 1 μg/L in a batch or plug-flow reactor (Acero et al., 2005).
Table 3-2 shows a subset of the chlorine CT values required for 99.9% (3-log) of inactivation of
Giardia cysts at a residual chlorine concentration of 1 mg/L (typical of the primary disinfection)
(U.S. EPA, 2003).
20
Table 3-1. Chlorine CT values for reducing microcystin-LR concentration to 1 μg/L in a batch or
plug-flow reactor
pH Initial Microcystin- CT Value (mg/L-min)

LR Concentration
(μg/L)
10 °C 15 °C 20 °C 25 °C
6 50 46.6 40.2 34.8 30.3
10 27.4 23.6 20.5 17.8
7 50 67.7 58.4 50.6 44.0
10 39.8 34.4 29.8 25.9
8 50 187 161 140 122
10 110 94.9 82.3 71.7
9 50 617 526 459 399
10 363 310 270 235
Source: (Acero et al., 2005)
CT (mg/L-min) = residual disinfectant concentration (mg/L) x contact time (min)
Table 3-2. Chlorine CT values for 99.9% (3-log) inactivation of Giardia cysts at a residual
chlorination concentration of 1 mg/L
pH CT Value (mg/L-min)
10 °C 15 °C 20 °C 25 °C
6 79 53 39 26
7 112 75 56 37
8 162 108 81 54
9 234 156 117 78
Researchers suggested that, at a pH value below 8, the chlorine CT values required to reduce
microcystin-LR to below 1 μg/L were comparable to those required for achieving 99.9 percent (3
log) inactivation of Giardia cyst (Acero et al., 2005). However, well-operated surface water
filtration plants can achieve at least a 2 to 2.5-log removal of Giardia cysts through filtration and
only need 1 or 0.5-log additional removal via disinfection to meet the overall treatment
requirement of 3-log Giardia removal and inactivation (U.S. EPA, 1991). Therefore, the
chlorine CT values applied at many water treatment plants, such as to achieve 1 or 0.5-log
inactivation of Giardia cysts, will likely be a fraction of the CT values required for 3-log
inactivation shown in Table 3-2.
Since the chlorine CT values in Table 3-1 are based on reducing microcystin-LR concentrations
from 50 or 10 μg/L to 1 μg/L, higher CT values would be required to further reduce the
microcystin-LR concentration to below the HA value for bottle-fed infants and young children of
pre-school age (0.3 μg/L). Assuming that the same reaction kinetics and the same equation for
calculating CT values still apply when reducing microcystin-LR to 0.3 μg/L, the CT values in
Table 3-1 should be modified by a multiplier of 1.3 and 1.5 for the initial microcystin-LR
concentration of 50 and 10 μg/L, respectively. These multipliers are derived from the original
equation (Acero et al., 2005):
21
CT = -ln ([MC] / [MC]0) / kapp, where [MC] and [MC]0 are the final and initial
concentrations of microcystin-LR, respectively. kapp is the apparent second-order rate constant
for the chlorination of microcystins at a given temperature.
Using this equation, CT values for reducing microcystin-LR from 50 to 0.3 μg/L can be
calculated by:
CT0.3-µg/L = ln (50/0.3) / ln (50/1) x CT1-µg/L = 1.3 CT1-µg/L
Similarly, CT values for reducing microcystin-LR from 10 to 0.3 μg/L can be calculated by:
CT0.3-µg/L = ln (10/0.3) / ln (10/1) x CT1-µg/L = 1.5 CT1-µg/L
Therefore, the CT values required for oxidizing microcystin-LR (and by assumption, total
microcystins) may be higher than those required for inactivation of Giardia cysts, depending on
pH, temperature and initial concentration of microcystins. The PWSs may consider increasing
their chlorine dose or contact time to reduce total microcystins to below the HA value for bottle-
fed infants and young children of pre-school age. Literature reported CT values for oxidizing
microcystins may be used by water utilities as a guide to benchmark their existing treatment
practices (Westrick, 2008); however, any CT recommendations should be treated with caution
and not be used to replace direct sampling and analysis of treated water for cyanotoxins.
The presence of NOM will decrease the efficiency of chlorine oxidation of cyanotoxins
(Rodriguez et al., 2007a) and increase the formation potential of chlorinated DBPs. The systems
should assess the impact of an increased chlorine dose on the DBP formation potential and avoid
the running annual average concentrations of total trihalomethanes (TTHMs) and haloacetic
acids (HAA5) exceeding the respective Maximum Contaminant Levels (MCLs) of 0.080 mg/L
and 0.060 mg/L.
4 Step 4: Monitor for Cyanotoxins in Raw and Finished Water and

Treatment Adjustments
4.1 Monitoring
If cyanotoxins are detected in raw water, EPA suggests that the PWS monitor both raw water and
finished water for cyanotoxins. EPA recommends using a quantitative laboratory ELISA test for
total microcystins for finished water samples, using sampling and analysis procedures described
in section 3.1. As noted for raw water samples, finished water samples that have been exposed to
any oxidants should be quenched immediately upon sampling. EPA suggests PWSs continue
collecting raw water samples at the same locations as described in Step 3 (section 3) to allow for
comparison. PWSs may want to consider collecting finished water samples at the entry point to
the distribution system. If paired raw and finished water samples are collected to assess the
treatment performance, then samples should be staggered to account for the time it takes for the
increment of sampled raw water to travel through the treatment plant.
Cyanotoxin analysis in raw and finished water can result in one of three outcomes. If no
cyanotoxins are detected in either raw or finished water, EPA recommends that the PWS
continue raw water sampling two to three times per week until the cyanobacterial bloom is no
longer visually detectable, Step 3. If cyanotoxins are detected in the raw water but not the
22
finished water, EPA recommends that the PWS continue raw water and finished water sampling
two to three times per week, until cyanotoxins are no longer found in the raw water (Step 4).
Factors to consider for sampling frequency include: toxin concentration, treatment capabilities
(historically able to remove high levels of toxins or not) and bloom dynamics (history of highly
variable toxin concentrations, impacts of wind and currents, etc.). If a system detects cyanotoxins
in the finished water, EPA recommends that the system monitor to confirm the presence of
cyanotoxins in finished water within 24 hours (Step 5). The PWSs may also consider providing
communications to their stakeholders (see section 4.2) and treatment (see section 4.3).
4.2 Communications
If PWSs confirm the presence of cyanotoxins in finished water, EPA encourages PWSs to
continue to Step 5 and communicate their findings with state and local officials to help prepare
for possible actions they may choose to take after the confirmation sampling of finished water
has been completed. EPA recommends PWSs confirm the presence of cyanotoxins in finished
water with at least one additional finished water sample before notifying the public. Section 5.2
discusses an approach for communication depending on the concentrations of cyanotoxins found
in finished water. If additional raw water sampling continues to detect cyanotoxins, the PWS
may wish to consider informing other source water managers as well.
4.3 Treatment
If a system detects and positively confirms cyanotoxins in the finished water, it indicates that
cyanotoxins have broken through the treatment barriers. It is important to monitor the
performance of individual unit processes across the treatment train to help understand what is
happening in the treatment train or identify possible treatment breakdown. PWSs should
continue implementing the treatment strategies described in section 3.3 and consider further
testing and adjustments to improve treatment performance. These treatment changes can be
implemented easily and quickly and will decrease the likelihood of toxin occurrence at the
household tap.
5 Step 5: Monitor for Cyanotoxins in Finished Water, Treatment

Adjustments or Additions, and Public Communications
If cyanotoxins have been detected in finished water (Step 4), EPA recommends that PWSs take
at least one additional sample to confirm the results as soon as possible within the first 24 hours
under Step 5. Although the HAs for microcystins and cylindrospermopsin are Ten-Day HAs, the
PWSs may want to consider proactively communicating the findings to state, local and public
health officials as soon as cyanotoxins are confirmed in finished drinking water. EPA anticipates
it may take multiple days to go from observing a bloom to detecting cyanotoxins in raw water to
confirming their presence above HA values in a finished water sample. Therefore, the PWS
would have the opportunity to adjust and add treatment as necessary prior to communicating
with stakeholders and issuing notices to the public as appropriate. A CMP can help a PWS be
prepared to take action to help reduce the likelihood of cyanotoxins reaching the finished water.
Step 5 contains suggested communication actions, treatment actions and additional monitoring
based on the concentrations of cyanotoxins found in the finished water. Figure 2 depicts a
recommended three-tier traffic light system, including actions EPA recommends once
23
cyanotoxins have been confirmed in finished drinking water. (This figure focuses on
microcystins, but a similar approach would also be appropriate for cylindrospermopsin.) A PWS
may want to develop a similar response plan, to allow the PWS to act quickly if cyanotoxins are
found in finished water.
5.1 Monitoring
EPA suggests PWSs continue to analyze finished water samples with a quantitative laboratory
ELISA test for total microcystins. If a PWS detects microcystins in its finished water, it could
also consider sampling for cylindrospermopsin to ensure no co-occurrence in finished water. If a
system wants to detect and quantify individual microcystin variants, a more selective method,
such as liquid chromatography/tandem mass spectrometry (LC/MS/MS) can be used
(recognizing this method does not identify the majority of microcystin congeners and may
underestimate the total concentration of microcystins in the sample). More information on
analytical methods is available in section 5 of the EPA HAs for microcystins and
cylindrospermopsin (U.S. EPA, 2015a,b).
If cyanotoxins have been initially detected in finished water, the PWS is encouraged to confirm
the presence of cyanotoxins in finished water with additional finished water samples as soon as
possible within 24 hours of detection. EPA suggests the PWS consider continuing monitoring the
finished water until cyanotoxins are no longer detected in finished water. This includes
monitoring during and after implementing treatment and (or) management strategies to remove
cyanotoxins from the finished water. PWSs are encouraged to base their sampling frequency on
the concentration of cyanotoxins detected in the finished water. The PWS may want to collect
finished water samples at least two to three times per week for systems with cyanotoxin
concentrations that have been reduced below detection levels in the finished water sampling
under Step 5, until levels are below quantification in at least 2-3 consecutive samples in raw
water. For systems that continue to detect concentrations above the HA value for bottle-fed
infants and young children of pre-school age but below the HA value for school-age children
through adults in the finished water sampling under Step 5, the PWS may want to consider daily
finished water sampling. PWSs that detect concentrations above the HA value for school-age
children through adults should consider sampling at least daily. Using the same sampling and
analytical procedures for all finished water sampling conducted under Step 4 and 5 of the
potential management steps will help provide comparable results.
If an increase in cyanotoxin concentrations is seen in finished water, the PWS may want to
increase sampling based on the new concentrations detected. For example, PWSs sampling
finished water at least two to three times per week might consider increasing to daily monitoring
if concentrations increase to levels above the HA value for bottle-fed infants and young children
of pre-school age but below the HA value for school-age children through adults. The PWS may
want to take into consideration local conditions (such as high proportion of susceptible
populations and weather events) to determine the need for samples in excess of the
recommended frequencies. Also, distribution system monitoring could be considered, for
example Ohio EPA encourages PWSs in Ohio to address distribution modeling and sampling in
their contingency plan (Ohio EPA, 2014). Depending on the concentration of cyanotoxins
initially detected in finished water, toxins could persist in the distribution system at levels of
concern even after it is non-detect at the entry point. Distribution system monitoring may be a
key component of any effort to characterize potential threats to the public.
24
It is important to consider specifying (within a CMP) when finished water monitoring is no
longer recommended. EPA suggests that the PWSs continue finished water sampling until at
least two consecutive samples, 24 hours apart, show concentrations of cyanotoxins below 0.3
μg/L for microcystins and 0.7 μg/L for cylindrospermopsin. A PWS may also want to consider
characterizing cyanotoxin levels throughout the distribution system while considering
appropriate communications with the primacy agency and the public. After a PWS has two
consecutive samples below 0.3 μg/L for microcystins, PWSs may want to consider returning to
Step 3 to determine if cyanotoxins are still present in the raw water.
5.2 Communications
PWSs are not required to notify their customers of any bloom or cyanotoxin occurrence and are
not required to include detections as part of a system’s Consumer Confidence Report under any
National Primary Drinking Water Regulations . PWSs should consult with their primacy agency
to determine if they are subject to any state or tribal notification requirements. Although not
required, PWSs may want to consider communicating with their consumers if cyanotoxins in
finished water are confirmed in additional samples. This communication may be received more
positively if PWSs have engaged in prior communication with the public about HABs. The PWS
is encouraged to tailor their communications based on the levels detected as detailed below. EPA
anticipates that multiple days may pass between the initial observation of a bloom and a
confirmation of the presence of cyanotoxins above HA values in finished water. This framework
allows for time to adjust or supplement treatment prior to using the communication strategies
discussed below.
Below HA Value for Bottle-Fed Infants and Young Children of Pre-School Age
If a PWS confirms concentrations at or below the HA value for bottle-fed infants and young
children of pre-school age (less than six years old), (0.3 μg/L for microcystins; 0.7 μg/L for
cylindrospermopsin), in finished water in either an initial or secondary sample, EPA
recommends that the PWS communicate with their primacy agency and local public health
agencies on the monitoring results and that the PWS does not issue any advisories to the public,
unless otherwise directed by the primacy agency.
Above HA Value for Bottle-Fed Infants and Young Children of Pre-School Age but Below HA
Value for School-Age Children through Adults
If a PWS confirms concentrations above the HA value for bottle-fed infants and young children
of pre-school age but below the HA value for school-age children through adults (above 0.3 μg/L
but below 1.6 μg/L for microcystins; above 0.7 μg/L but below 3.0 μg/L for cylindrospermopsin)
in additional finished water samples under Step 5, EPA recommends that the PWS consult with
their primacy agency and the local public health agency to determine when and how to notify
drinking water consumers who may be more susceptible to adverse outcomes (such as bottle-fed
infants and young children of pre-school age) within 24 hours, to advise them to use alternate
sources of drinking water. After at least two consecutive finished water samples are below the
HA level for bottle-fed infants and young children of pre-school age, EPA recommends notifying
consumers that drinking water has returned to acceptable levels. PWSs can consider developing
targeted outreach for sensitive populations though partnerships with others, such as
communicating with pediatricians and day care centers. Tools are available to help PWSs in
25
develop their communication messages such as Center for Disease Control’s (CDC)
DWACT(CDC, 2013) and EPA’s Developing Risk Communication Plans for Drinking Water
Contamination Incidents (U.S. EPA, 2013a). The CDC is currently updating their Drinking
Water Advisory Toolbox and intends to include cyanotoxin specific information in that update.
Above HA Value for School-Age Children through Adults

If a PWS confirms concentrations above the HA value for school-age children through adults
(1.6 μg/L for microcystins; 3.0 μg/L for cylindrospermopsin) in the additional finished water
samples collected under Step 5, EPA recommends that the PWS consult with their primacy
agency as well as the local public health agency to determine when and how to issue a ‘Do Not
Drink/Do Not Boil Water’ advisory to the general public served by that water supply within 24
hours. After at least two consecutive samples are below the HA level for school-age children
through adults, EPA recommends that the PWS remove the ‘Do Not Drink/Do Not Boil Water’
advisory. EPA suggests the PWS notify water customers who may be more susceptible to
adverse outcomes (such as bottle-fed infants and young children of pre-school age) within 24
hours, to advise them to use alternate sources of drinking water until two consecutive samples
are below the HA value for bottle-fed infants and young children of pre-school age. The PWS
may want to also consider evaluating the distribution system levels before removing the
notification to ensure these levels have been reduced as well.
PWS may want to describe in their communication efforts they are undertaking to address the
problem and the expected duration of the elevated levels of cyanotoxins. EPA also encourages
PWSs to identify alternatives customers have available if they receive a ‘Do Not Drink/Do Not
Boil Water’ advisory. For PWSs where source waters have public access for recreation, the
system’s notice may include statements about recreational use of waters with cyanobacterial
blooms to prevent exposure of humans and animals to cyanotoxins. (Note: EPA is developing
water quality criteria for recreational water that will provide additional information about levels
of cyanotoxins in source waters.)
5.3 Treatment
If a system detects cyanotoxins in the additional finished water samples collected under Step 5,
the PWS is encouraged to continue adjusting the treatment as discussed in section 3.3, to return
cyanotoxin concentrations to below HA levels as soon as possible. The PWS might also monitor
individual treatment processes to better understand or to determine which treatments are
effective at cyanotoxin removal.
Providing an alternate water source may be useful as a temporary management strategy

following the detection of cyanotoxins in the finished water. This can include other sources
under the purview of the system or providing for an interconnection to another system. When
using this strategy, EPA recommends that systems be aware of any primacy agency regulatory
implications that may result.
If the PWS is frequently challenged by cyanotoxins in source waters and modifications to its
existing treatment system do not sufficiently reduce cyanotoxin levels to below HA values, it
may consider installing permanent treatment as appropriate to address cyanotoxin occurrence in
future years (see Appendix E).
26
Figure 2. Traffic light approach to support communication and other actions in response to
elevated concentrations of cyanotoxins in finished drinking water.
27
6 References
Acero, J.L., Rodriguez, E., Meriluoto, J. 2005. Kinetics of Reactions between Chlorine and the
Cyanobacterial Toxins Microcystins. Water Research 39(8): 1628-38.
Alvarez, M.B., Rose, J.B., and Bellamy, B. 2010. Treating algal toxins using oxidation,
adsorption, and membrane technologies. Water Research Foundation.
American Water Works Association (AWWA) and Water Research Foundation (WRF). 2015. A
Water Utility Manager’s Guide to Cyanotoxins. American Water Works Association. Denver,
CO.
Association of State Drinking Water Administrators (ASDWA) and U.S. EPA. 2014.
Opportunities to Protect Drinking Water Sources and Advance Watershed Goals Through the
Clean Water Act: A Toolkit for State, Interstate, Tribal and Federal Water Program Managers.
Available online at:
http://www.asdwa.org/document/docWindow.cfm?fuseaction=document.viewDocument&docum
entid=3007&documentFormatId=3779. Accessed February 25, 2015.
Beaver, J. R., Manis, E. E., Loftin, K. A., Graham, J. L., Pollard, A. I., and Mitchell, R. M. 2014.
Land use patterns, ecoregion, and microcystin relationships in U.S. lakes and reservoirs: A
preliminary evaluation. Harmful Algae 36:57-62.
Bourke, A.T. C., Hawes, R.B., Neilson, A., and Stallman, N. D. 1983. An outbreak of hepato-
enteritis (the Palm Island mystery disease) possibly caused by algal intoxication. Toxicon 21(3):
45-48.
Bowling, L.C., and Baker, P.D. 1996. Major cyanobacterial bloom in the Barwon-Darling River,
Australia, in 1991, and underlying limnological conditions. Marine and Freshwater Research 47:
643-57.
Carmichael, W.W. 1986. Algal toxins. Advances in Botanical Research 12: 14-101.
CDC. 2013. Drinking Water Advisory Communication Toolbox. Centers for Disease Control,
Department of Health and Human Services, United States Environmental Protection Agency, and
American Water Works Association. Available online at:
http://www.cdc.gov/healthywater/emergency/dwa-comm-toolbox/. Accessed February 25, 2015.
Cheung, M. Y., Liang, S., and Lee, J. 2013. Toxin-producing cyanobacteria in freshwater: a
review of the problems, impact on drinking water safety, and efforts for protecting public health.
Journal of Microbiology 51(1): 1-10.
Chow, C. W. K., Panglisch, S., House, J., Drikas, M., Burch, M. D., and Gimbel, R. 1997. A
study of membrane filtration for the removal of cyanobacterial cells. Aqua 46(6): 324-334.
Codd, G.A. 1995. Cyanobacterial toxins: occurrence, properties and biological significance.
Water Science and Technology 32(4): 149-156.
28
Conley, D.J., Paerl, H.W., Howarth, R.W., Boesch, D.F., Seitzinger, S.P., Havens, K.E.,
Lancelot, C., Likens, G.E. 2009. Science 323(5917): 1014-1015.
Connecticut Department of Public Health. 2015. Available online at:

http://www.ct.gov/dph/site/default.asp. Accessed April 7, 2015.
Cyanocenter UBA. 2015. Decision support tool: Background information on drinking water
Treatment. Available online at: http://toxische-cyanobakterien.de/en/background-
information/drinking-water-treatment/. Accessed February 19, 2015.
Dixon, M. B., Falconet, C., Ho, L., Chow, C. W., O’Neill, B. K., Newcombe, G. 2010.
Nanofiltration for the removal of algal metabolites and the effects of fouling. Water Science
Technology 61(5): 1189-99.
Dixon, M. B., Richard, Y., Ho, L., Chow, C. W., O’Neill, B. K., and Newcombe, G. 2011. A
coagulation–powdered activated carbon–ultrafiltration–Multiple barrier approach for removing
toxins from two Australian cyanobacterial blooms. Journal of hazardous materials 186(2): 1553-
1559.
Doblin, M. A., Coyne, K. J., Rinta-Kanto, J. M., Wilhelm, S. W., Dobbs, F. C. (2007). Dynamics
and short-term survival of toxic cyanobacteria species in ballast water from NOBOB vessels
transiting the Great Lakes—implications for HAB invasions. Harmful Algae 6(4): 519-530.
Dolman, A.M., Rucker, J., Pick, F.R., Fastner, J., Rohrlack, T., Mischke, U., Wiedner C. 2012.
Cyanobacteria and cyanotoxins: The influence of nitrogen versus phosphorus. PLoS One. 7(6):1-
14.
Drikas, M., Newcombe, G. Nicholson, B. 2002. Water Treatment Options for Cyanobacteria and
their Toxins. In: Blue-Green Algae, Their significance and management within water. The
Cooperative Research Centre for Water Quality and Treatment.
Ganf, G.G. and Oliver, R.L. 1982. Vertical separation of light and available nutrients as a factor
causing replacement of green algae by blue-green algae in the plankton of a stratified lake.
Journal of Ecology 70: 829-844.
Garcı́a-Villada, L., Rico, M., Altamirano, M., Sánchez-Martı́n, L., López-Rodas, V., & Costas,
E. (2004). Occurrence of copper resistant mutants in the toxic cyanobacteria Microcystis
aeruginosa: characterisation and future implications in the use of copper sulphate as algaecide.
Water Research 38(8): 2207-2213.
Gijsbertsen-Abrahamse, A.J., Schmidt, W., Chorus, I., Heijman, S.G.J. 2006. Removal of
cyanotoxins by ultrafiltration and nanofiltration. Journal of Membrane Science 276(1-2): 252-
259.
Glibert, P. M., Maranger, R., Sobota, D. J., & Bouwman, L. (2014). The Haber Bosch–harmful
algal bloom (HB–HAB) link. Environmental Research Letters 9(10): 105001.
Griffiths, D.J., and Martin, S.L. 2003. The Palm Island mystery disease 20 years on: a review of
research on the cyanotoxin cylindrospermopsin. Environmental Toxicology 18(2): 78-93.
29
Grützmacher, G., Böttcher, G., Chorus, I., and Bartel, H. 2002. Removal of microcystins by slow
sand filtration. Environmental Toxicology 17(4): 386-394.
Herath, G. 1995. The Algal Bloom Problem in Australian Waterways: an Economic Appraisal.
Review of Marketing and Agricultural Economics. 63(1): 77-86.
Hitzfeld, B.C., Höger, S. J., and Dietrich, D.R. 2000. Cyanobacterial toxins: removal during
drinking water treatment, and human risk assessment. Environmental Health Perspectives
108(1): 113.
Ho, L., Onstad, G, Gunten, U, Rinck-Pfeiffer, S, Craig, K, and Newcombe, G. 2006a.

Differences in the chlorine reactivity of four microcystin analogues. Water Res, 40:1200-1209.
Ho, L., Meyn, T., Keegan, A., Hoefel, D., Brookes, J., Saint, C. P., and Newcombe, G. 2006b.
Bacterial degradation of microcystin toxins within a biologically active sand filter. Water
Research 40(4): 768-774.
Ho, L., Sawade, E., Newcombe, G. 2012. Biological treatment options for cyanobacteria
metabolite removal-a review. Water Research 46:1536-1548.
House, J., Ho, L., Newcombe, G., and Burch, M. 2004. Management strategies for toxic blue-
green algae: literature review. Australian Water Quality Centre, Cooperative Research Centre
for Water Quality and Treatment, Salisbury, South Australia, 5108.
Ibelings, B.W., Mur, L.R., Walsby, A. 1991. Diurnal changes in buoyancy and vertical
distribution in populations of Microcycstis in two shallow lakes. Journal of Plankton Research
13(2): 419-436.
Izydorczyk, K. Tarczynska, M., Jurczak, T., Mrowczynski, J., Zalewski, M. 2005. Measurement
of Phycocyanin Fluorescence as an Online Early Warning System for Cyanobacteria in Reservoir
Intake Water. Environmental Toxicology 20(4): 425-430.
Jacoby, J. M., Collier, D. C., Welch, E. B., Hardy, F. J., and Crayton, M. 2000. Environmental
factors associated with a toxic bloom of Microcystis aeruginosa. Canadian Journal of Fisheries
and Aquatic Sciences 57: 231-240.
Jurczak, T., Tarczynska, M., Izydorczyk, K., Mankiewicz, J., Zalewski, M., and Meriluoto, J.
2005. Elimination of microcystins by water treatment processes – examples from Sulejow
Reservoir, Poland. Water Research 39: 2394-2406.
Knoll, L.B., Sarnelle, O., Hamilton, S.K., Kissman, C.E.H., Wilson, A.E., Rose, J.B., Morgan,
M.R. 2008. Invasive zebra mussels (Dreissena polymorpha) increase cyanobacterial toxin
concentrations in low-nutrient lakes. Canadian Journal of Fisheries and Aquatic Sciences. 65(3):
448-455.
Kommineni, S., and Water Research Foundation. 2009. Strategies for Controlling and Mitigating
Algal Growth within Water Treatment Plants. Water Research Foundation Report.
30
Lahti, K., Vaitomaa, J., Livimaki, A.L., Sivonen, K. 1998. Fate of cyanobacterial hepatotoxins in
artificial recharge of groundwater and in bank filtration. In: Artifical recharge of groundwater,
J.H. Peters et al., eds. A.S. Balkema, Rotterdam, The Netherlands.
Lee, T., Tsuzuki, M., Takeuchi, T., Yokoyama, K., Karube, I. 1995. Quantitative determination
of cyanobacteria in mixed phytoplankton assemblages by an in vivo fluorimetric method.
Analytica Chimica Acta 302: 81-87.
Loftin, K. A., Ziegler, A. C., and Meyer, M. T. 2008. Guidelines for design and sampling for
cyanobacterial toxin and taste-and-odor studies in lakes and reservoirs. U.S. Department of the
Interior, U.S. Geological Survey.
Lytle, D.A. 1995. “How Do I Run a Proper Jar Test?” Proceedings of the 1995 AWWA Water
Quality and Technology Conference. New Orleans, LA. Nov 12-16.
Markensten, H., Moore, K., and Persson, I. 2010. Simulated lake phytoplankton composition
shifts towards cyanobacteria dominance in a future warmer climate. Ecological Applications
20(3): 752-767.
Markham, L., Porter, M., and Schofield, T. 1997. Algae and zooplankton removal by dissolved
air flotation at severn trent Ltd. Surface Water Treatment Works. In Proceedings: CIWEM
Dissolved Air Flotation International Conference (pp. 112-119).
Mehnert, G., Leunert, F., Cirés, S., Jöhnk, K. D., Rücker, J., Nixdorf, B., and Wiedner, C. 2010.
Competitiveness of invasive and native cyanobacteria from temperate freshwaters under various
light and temperature conditions. Journal of Plankton Research 32(7): 1009-1021.
Mitsui, A., Kumazawa, S., Takahashi, A., Ikemoto, H., Cau, S., and Arai, T. 1986. Strategy by
which nitrogen-fixing unicellular cyanobacteria grow photoautotrophically. Nature 323: 720-
722.
Mouchet, P., and Bonnélye, V. 1998. Solving algae problems: French expertise and world-wide
applications. Journal of Water Supply: Research and Technology. Aqua 47: 125–141.
Multi-Resolution Land Characteristics Consortium (MRLC). 2015. National Land Cover

Database. Available online at http://www.mrlc.gov/index.php. Accessed May 14, 2015.
National Hydrography Dataset Plus (NHDPlus). 2015. Available online at: http://www.horizon-
systems.com/NHDPlus/NHDPlusV1_tools.php. Accessed on May 14, 2015.
National Oceanic and Atmospheric Association (NOAA), Great Lakes Environmental Research
Laboratory. 2014. Harmful Algal Blooms. Available online at:
http://www.glerl.noaa.gov/res/waterQuality/. Accessed February 22, 2015.
Newcombe, G. (Ed). 2009. International Guidance Manual for the Management of Toxic
Cyanobacteria: A Guide for Water Utilities. Global Water Research Coalition.
31
Newcombe, G., House, J., Ho, L., Baker, P., and Burch, M. 2010. Managing Strategies for
Cyanobacteria (Blue-green Algae): A Guide for Water Utilities. Water Quality Research
Australia: Research Report 74.
Newcombe, J., Dreyfus, J., Monrolin, Y., Pestana, C., Reeve, P., Sawade, E., Ho, L., Chow, C.,
Krasner, S.W., and Yates, R.S. 2015. Optimizing Conventional Treatment for the Removal of
Cyanobacteria and Toxins. Water Research Foundation: Order Number 4315.
New York State Department of Environmental Conservation. 2015. Suspicious Algae Bloom
Report. Available online at: http://www.dec.ny.gov/docs/water_pdf/algaereportform.pdf
Nicholson, B.C., Rositano, J., and Burch, M.D. 1994. Destruction of cyanobacterial peptide
hepatotoxins by chlorine and chloramine. Water Research 28(6): 1297-1303.
Ohio EPA. 2014. Public Water System Harmful Algal Bloom Response Strategy. Available
online at:
http://epa.ohio.gov/Portals/28/documents/HABs/PWS_HAB_Response_Strategy_2014.pdf.
Accessed February 4, 2015.
Ohio EPA. 2015. Ohio EPA Total (Extracellular and Intracellular) Microcystins – ADDA by
ELISA Analytical Methodology. Version 2.0. Available online at:
http://epa.ohio.gov/Portals/28/documents/habs/HAB_Analytical_Methodology.pdf. Accessed
April 8, 2015.
Ohio Lake Management Society (OLMS). 2014. CLAM Program. Available online at:
http://olms.org/citizen-lake-awareness-and-monitoring/. Accessed May 15, 2015.
Ohio Sierra Club (2015). Ohio Water Sentinel Program. Available online at:
http://ohiosierraclub.org/all-committees/conservation-committee/water-committee/ohio-water-
sentinel-program/. Accessed May 15, 2015.
Oregon Health Authority. 2013. Algae Resources for Drinking Water: Best Management
Practices for Harmful Algae Blooms, Exhibit 5. Available online at:
https://public.health.oregon.gov/HealthyEnvironments/DrinkingWater/Operations/Treatment/Pag
es/algae.aspx. Accessed February 25, 2015.
Paerl, H.W. 2008. Nutrient and other environmental controls of harmful cyanobacterial blooms
along the freshwater-marine continuum. In: Hudnell, editor. Cyanobacterial Harmful Algal
Blooms: State of the Science and Research Needs (pp. 217-237). Springer New York.
Paerl, H.W. and Huisman, J. 2009. Climate change: a catalyst for global expansion of harmful
cyanobacterial blooms. Environmental Microbiology Reports 1(1): 27-37.
Paerl, H.W. and Otten, T.G. 2013. Harmful Cyanobacterial Blooms: Causes, Consequences, and
Controls. Microbial Ecology 65(4): 995-1010.
Paerl, H.W. 2014. Mitigating Harmful Cyanobacterial Blooms in a Human- and Climatically-
Impacted World. Life 4: 988-1012.
32
Parmesan, C. 2006. Ecological and Evolutionary Response to Recent Climate Change. Annual
Review of Ecology, Evolution, and Systematics 37: 637-369.
Rajasekhar, P., Fan, L., Nguyen, T., and Roddick, F.A. 2012. A review of the use of sonication
to control cyanobacterial blooms. Water research 46(14): 4319-4329.
Rapala, J., Miemela, M., Berg, K.A., Lepisto, L., Lahti, K. 2006. Removal of cyanobacteria,
cyanotoxins, heterotrophic bacteria and endotoxins at an operating surface water treatment plant.
Water Science & Technology 54(3): 23-28.
Reynolds, C.S., Oliver, R.L., and Walsby, A.E. 1987. Cyanobacterial dominance: the role of
buoyancy regulation in dynamic lake environments. New Zealand Journal of Marine and
Freshwater Research 21(3): 379-390.
Robarts, R.D., and Zohary, T. 1987. Temperature effects on photosynthetic capacity, respiration,
and growth rates of bloom-forming cyanobacteria. New Zealand Journal of Marine and
Freshwater Research 21(3): 391-399.
Rodríguez, E.M., Acero, J.L., Spoof, L., and Meriluoto, J. 2007a. Oxidation of MC-LR and RR
with chlorine and potassium permanganate: Toxicity of the reaction products. Water Research
42(6): 1744-1752.
Rodríguez, E., Onstad, G.D., Kull, T.P., Metcalf, J.S., Acero, J.L., and Von Gunten, U. 2007b.
Oxidative elimination of cyanotoxins: comparison of ozone, chlorine, chlorine dioxide and
permanganate. Water Research 41(15): 3381-3393.
Schijven, J., Berger, P., Miettinen, I. 2002. Removal of pathogens, surrogates, indicators and
toxins using riverbank filtration. In: Riverbank Filtration. Improving source water quality. Ray,
C., Melin, G., Linksy, R.B. (Eds.), Kluwer Academic, Netherlands.
Source Water Collaborative (SWC). 2015a. How to Collaborate Toolkit. Available online at:
http://www.sourcewatercollaborative.org/how-to-collaborate-toolkit/. Accessed April 8, 2015.
SWC. 2015b. Source Water Protection and Conservation Partners Toolkit. Available online at:
http://www.sourcewatercollaborative.org/swp-conservation-partners-toolkit/. Accessed May 5,
2015.
SWC. 2015c. A Planner’s Guide. Available online at: http://sourcewatercollaborative.org/guide-

for-land-use-planners/. Accessed May 5, 2015.
Sklenar, K., Westrick, J., and Szlag, D. 2014. Managing and Mitigating Cyanotoxins in Water
Supplies. Water Research Foundation Webcast. August 28, 2014.
Teixeira, M.R., and Rosa, M.J. 2006. Comparing dissolved air flotation and conventional
sedimentation to remove cyanobacterial cells of Microcystis aeruginosa: part I: the key operating
conditions. Separation and Purification Technology 52(1): 84-94.
33
Thomas, R.H., and Walsby, A.E. 1985. Buoyance Regulation in a Strain of Microcystis. Journal
of General Microbiology 131(4): 799-809.
Tilzer, M.M. 1988. Secchi disk – chlorophyll relationships in a lake with highly variable
phytoplankton biomass. Hydrobiologia 162(2): 163-171.
Tokodi, N., Drobac, D., Svirčev, Z., and Lazić, D. 2012. Cyanotoxins in Serbia and water
treatment procedures for their elimination. Geographica Pannonica 16(4): 155-163.
Ueno, Y., Nagata, S., Tsutsumi, T., Hasegawa, A., Watanabe, M. F., Park, H. D., Chen, G.C.,
Chen, G., Yu, S. Z. 1996. Detection of microcystins, a blue-green algal hepatotoxin, in drinking
water sampled in Haimen and Fusui, endemic areas of primary liver cancer in China, by highly
sensitive immunoassay. Carcinogenesis 17(6): 1317-1321.
United States Army Corps of Engineers (USACE). 2015. Environment: Lakes. Available online
at: http://www.corpsresults.us/environment/envlakes.cfm.
United States Environmental Protection Agency (U.S. EPA), 1991. “Guidance Manual for
Compliance with the Filtration and Disinfection Requirements for Public Water Systems Using
Surface Water Sources”. Office of Drinking Water, Washington, DC.
U.S. EPA. 1997. State Source Water Assessment and Protection Programs, EPA 816-R-97-009.
U.S. EPA. 2003. “LT1ESWTR Disinfection Profiling and Benchmarking Technical Guidance
Manual”. EPA 816-R-03-004. Office of Water, Washington, DC.
U.S. EPA. 2008. Drinking Water Health Advisory for Boron, EPA 822-R-08-013.
U.S. EPA. 2011. My Waters Mapper. Available online at:

http://www.epa.gov/waters/mywatersmapper/. Accessed May 14, 2014.
U.S. EPA. 2012a. 2012 Edition of the Drinking Water Standards and Health Advisories. EPA
822-S-12-001.
U.S. EPA. 2012b. STORET. Available online at: http://www.epa.gov/storet/. Accessed

February 27, 2015.
U.S. EPA. 2013a. Developing Risk Communication Plans for Drinking Water Contamination
Incidents. Available online at:
http://water.epa.gov/infrastructure/watersecurity/lawsregs/upload/epa817f13003.pdf. Accessed
April 9, 2015.
U.S. EPA. 2013b. National Lakes Assessment. Available online at:

http://water.epa.gov/type/lakes/lakessurvey_index.cfm. Accessed May 15, 2015.
U.S. EPA. 2013c. Climate Resilience Evaluation & Awareness Tool (CREAT). Available online
at: http://water.epa.gov/infrastructure/watersecurity/climate/creat.cfm. Accessed May 15, 2015.
34
U.S. EPA. 2013d. Impacts of Climate Change on the Occurrence of Harmful Algal
Blooms. Fact Sheet. Office of Water. EPA 820-S-12-001. Available online at:
http://www2.epa.gov/sites/production/files/documents/climatehabs.pdf.
U.S. EPA. 2013e. BASINS (Better Assessment Science Integrating point & Nonpoint Sources).
Available online at: http://water.epa.gov/scitech/datait/models/basins/index.cfm. Accessed May
17, 2015.
U.S. EPA. 2014a. Cyanobacteria and Cyanotoxins: Information for Drinking Water
Systems. Fact Sheet. Office of Water. EPA-810F11001.
U.S. EPA. 2014b. Drinking Water Science and Regulatory Support. Available online at:
http://water.epa.gov/drink/standards/hascience.cfm. Accessed February 25, 2015.
U.S. EPA. 2014c. Nitrogen and Phosphorus Pollution Data Access Tool. Available online at:
http://www2.epa.gov/nutrient-policy-data/nitrogen-and-phosphorus-pollution-data-access-
tool. Accessed February 28, 2015.
U.S. EPA. 2014d. Control and Treatment. Available online at: http://
www2.epa.gov/nutrient-policy-data/control-and-treatment. Accessed February 22,
2015.
U.S. EPA. 2015a. Drinking Water Health Advisory for the Cyanobacterial Toxin Microcystin.
EPA 820R15100. Available online at:
http://www2.epa.gov/sites/production/files/2015-06/documents/microcystins-report-2015.pdf.
U.S. EPA. 2015b. Drinking Water Health Advisory for the Cyanobacterial Toxin
Cylindrospermopsin. EPA 820R15101. Available online at:
http://www2.epa.gov/sites/production/files/2015-06/documents/cylindrospermopsin-report-2015.pdf.
U.S. EPA. 2015c. Health Effects Support Document for the Cyanobacterial Toxin Microcystins. EPA
820R15102. Available online at:
http://www2.epa.gov/sites/production/files/2015-06/documents/microcystins-support-report-2015.pdf.
U.S. EPA. 2015d. Health Effects Support Document for the Cyanobacterial Toxin Cylindrospermopsin.
EPA 820R15103. Available online at: http://www2.epa.gov/sites/production/files/2015-06/documents/
cylindrospermopsin-support-report-2015.pdf.
U.S. EPA. 2015e. Health Effects Support Document for the Cyanobacterial Toxin Anatoxin-a.
EPA 820R15104. Available online at:
http://www2.epa.gov/sites/production/files/2015-06/documents/anatoxin-a-report-2015.pdf
U.S. EPA. 2015f. Climate Ready Water Utilities (CRWU). Available online at:
http://water.epa.gov/infrastructure/watersecurity/climate/index.cfm. Accessed May 18, 2015.
U.S. EPA 2015g. How’s My Waterway. Available online at:

http://watersgeo.epa.gov/mywaterway/. Accessed May 14, 2015.
35
U.S. EPA. 2015h. Enforcement and Compliance History Online. Available
online at: http://echo.epa.gov/. Accessed May 8, 2015.
U.S. EPA 2015i. Discharge Monitoring Report (DMR) Pollutant Loading Tool.
Available online at: http://cfpub.epa.gov/dmr/. Accessed May 8, 2015.
United States Geological Survey (USGS). 2011. SPARROW Surface Water-Quality Modeling.
Available online at: http://water.usgs.gov/nawqa/sparrow/.
USGS. 2015. National Water-Quality Assessment (NAWQA) Program. Available online at:
http://water.usgs.gov/nawqa/.
van Rijn, J. and Shilo, M. 1985. Carbohydrate fluctuations, gas vacuolation, and vertical
migration of scum-forming cyanobacteria in fishponds. Limnology and Oceanography 30(6):
1219-1228.
Vermont Department of Environmental Conservation. 2014. Available online at:

http://www.anr.state.vt.us/dec/dec.htm. Accessed April 7, 2015.
Walsby, A.E., Hayes, P.K., Boje, R., and Stal,L.J. 1997. The selective advantage of buoyance
provided by gas vesicles for planktonic cyanobacteria in the Baltic Sea. New Phytologist 136(3):
407-417.
Wang, H., Ho, L., Lewis, D.M., Brookes, J.D., and Newcombe, G. 2007. Discriminating and
assessing adsorption and biodegradation removal mechanisms during granular activated carbon
filtration of microcystin toxins. Water Research 41:4262-4270.
Westrick, J.A. 2008. “Chapter 13: Cyanobacterial toxin removal in drinking water processes and
recreational waters.” In Cyanobacterial Harmful Algal Blooms: State of the Science and
Research Needs. Springer New York, 2008. 275-290.
Westrick, J.A., Szlag, D.C., Southwell, B.J., and Sinclair, J. 2010. A review of cyanobacteria and
cyanotoxins removal/inactivation in drinking water treatment. Analytical and Bioanalytical
Chemistry 397(5): 1705-1714.
World Health Organization (WHO). 1999. Toxic Cyanobacteria in Water: A guide to their public
health consequences, monitoring and management. Bartram, J. and Chorus, I., (eds). CRC Press,
London.
WHO. 2000. Monitoring Bathing Waters - A Practical Guide to the Design and Implementation
of Assessments and Monitoring Programs. Bartram, J. and Rees, G. (eds). E & FN Spon,
London.
World Resources Institute (WRI). 2013. Natural Infrastructure: Investing in Forested Landscape
for Source Water Protection in the United States. Gartner, T., Mulligan, J., Schmidt, R., Gunn, J.
(eds.) WRI, Washington, D.C.
36
Yoshida, M., Yoshida, T., Takashima, Y., Hosoda, N., Hiroishi, S. 2007. Dynamics of
microcystin-producing and non-microcystin-producing Microcystis populations is correlated
with nitrate concentrations in a Japanese lake. FEMS Microbiological Letters. 266:49-53.
Yuan, L.L., Pollard, A.I., Pather, S., Oliver, J.L., and D'Anglada, L. 2014. Managing
microcystin: identifying national-scale thresholds for total nitrogen and chlorophyll a.
Freshwater Biology 59(9): 1970-1981.
Zamyadi, A., et al., 2012. Toxic cyanobacterial breakthrough and accumulation in a drinking
water plant: A monitoring and treatment challenge. Water Research 46:1511-1523.
Zhang, X.J., Chen, C., Ding, J.Q., Hou, A., Li, Y., Niu, Z.B., Su, X.Y., Xu, Y.J. 2010. The 2007
water crisis in Wuxi: China: Analysis of the origin. Journal of Hazardous Materials 182(1): 130-
135.
37
Appendix A System-Specific Evaluation Information, Tools, and
Resources
Appendix A contains a more detailed discussion of the types of information a PWS could
consider when conducting a system-specific evaluation for cyanotoxin vulnerability in source
waters. Resources and tools that could be useful in conducting the system-specific evaluation and
developing a CMP are also discussed; a more extensive table is available at the end of this
appendix. Tools listed vary in their complexity and purpose. Some users may prefer user-friendly
tools that provide summary information about key areas, while others may prefer more complex
and (or) comprehensive databases or models. This list of tools discussed in this appendix may
not be fully comprehensive and other useful tools and information sources are available. For
example, some U.S. states also have resources and tools that could also provide PWSs with
assistance in developing a CMP.
Source Water Characteristics

Source water type is an important factor in a system-specific evaluation. Certain types of source
waters will have greater vulnerabilities than others. For example, lakes and reservoirs are
expected to have greater vulnerabilities than moving waters such as free-flowing rivers. Lakes
and reservoirs typically have conditions that are more favorable to cyanobacterial growth (as
discussed below) than rivers, although some rivers can have vulnerabilities under certain
conditions such as drought causing slow moving water. Most ground water is generally not
expected to be at risk; however, ground water under the direct influence of surface water may be
at risk. If toxins are released after a surface water bloom in a lake, a proportion of the dissolved
toxins may accompany any surface water that is induced by pumping well action to passage from
lake water to nearby well water. A study by Ueno et al., (1996) found microcystins in four
percent of shallow wells, with no occurrence in deeper wells.
As part of a system-specific evaluation, PWSs might want to consider other source water body
characteristics, such as size and depth of the source water. In some cases where the source is a
large body of water, such as the Great Lakes, different locations within the body of water can
have different characteristics and therefore different vulnerabilities to cyanotoxins. Information
on source water type and hydrology can often be found in a state’s Source Water Assessment, a
source water analysis that states completed for all PWSs after the 1996 Amendments to the
SDWA, as explained in “Source Water Assessment Information” below. A useful tool on surface
water and watershed characteristics is the NHDPlus (National Hydrography Dataset). This tool
can provide for general mapping and analysis of surface water systems (see discussion below)
(NHDPlus, 2015).
Vertical stratification can be a contributing factor to bloom formation in source waters. Vertical
stratification can be determined by measuring vertical profiles of temperature within the source
water (WHO, 1999). Some common cyanobacteria can move throughout the water column due
to their buoyancy regulation ability. It is believed that cyanobacteria use this buoyancy to move
throughout the water column to overcome the separation of nutrients and light (Ganf and Oliver,
1982; Reynolds et al., 1987). During low mixing events in the source water, such as those due to
low wind, the buoyant cyanobacteria can float towards the surface (Walsby et al., 1997; Paerl
and Huisman, 2009). Some cyanobacterial strains, such as strains of Microcystis, regulate their
buoyancy based on light intensity, leading to increased cyanobacterial growth (Thomas and
Walsby, 1985). Studies have shown that cyanobacteria can migrate on a diurnal pattern with cells
38
rising to the surface overnight and sinking throughout the day (Ganf and Oliver, 1982; van Rijn
and Shilo, 1985; Ibelings et al., 1991). Wind patterns, slow moving waters, and reduced source
water mixing, along with vertical stratification, are important factors to consider as part of the
weight of evidence evaluation (Jacoby et al., 2000).
Water Quality Parameters

Data on the PWS’s source water quality is helpful for a system-specific evaluation. Historical
data showing cyanobacterial cell occurrence and (or) cyanotoxin occurrence in source water or
finished water is the most direct indication of source water vulnerabilities to cyanotoxins. Other
useful information in determining source water vulnerability include: phycocyanin, chlorophyll-
a, Secchi depths, geosmin and MIB taste and odor problems (sometimes caused by
cyanobacteria) and nutrient concentrations (see nutrient section for more specific information on
evaluating nutrient impacts).
For example, cyanobacteria contain chlorophyll-a as well as other pigments for photosynthesis
that can serve as an indicator for cyanotoxins in source water (Cheung et al., 2013). The
pigments allow cyanobacteria to produce energy when light intensity is low as well as in varying
light spectra not typically used by other phytoplankton species. This gives cyanobacteria an
advantage over other phytoplankton in turbid waters (WHO, 1999). Source waters with sustained
high levels of chlorophyll-a may have vulnerabilities to cyanotoxin occurrence. Phycocyanin is
another pigment produced by cyanobacteria that is not produced by other algae (Lee et al., 1995).
A study by Izydorczyk et al., (2005) found a positive correlation between phycocyanin
fluorescence and cyanobacterial biomass in a drinking water reservoir during a Microcystis
aeruginosa bloom. Phycocyanin could also be an indicator of cyanotoxin occurrence. Satellite or
hyperspectral imagery from aircraft that have been processed based on chlorophyll-a or
phycocyanin concentrations are also a useful indicator of cyanobacteria occurrence.
Systems without their own historical data can consider consulting with their primacy agency to
determine if other sources of historical data are available. Multiple federal agencies (EPA,
USACE, United States Geological Survey (USGS)) and state agencies, have assessed waters for
cyanotoxins as part of assessment and research projects and may have additional data not known
to the PWS. The EPA and State National Aquatic Resource Surveys include algal toxin and
indicators; for example cyanobacteria and microcystins are part of the National Lakes
Assessment (U.S. EPA, 2013b). Many state water quality agencies monitor for cyanobacteria or
microcystins and make that data available through the Water Quality Data Portal. Another source
of water quality data is satellite imaging, such as the Lake Erie HABs tracker by National
Oceanic and Atmospheric Association (NOAA, 2014). Historical information on past wild and
domestic animal poisonings surrounding the source water could also provide information on past
blooms (Carmichael, 1986; Codd, 1995). Other water quality parameter information that can be
helpful include the dominant cyanobacterial species within a source water (to help determine
which cyanotoxins may be produced) as well as presence of zebra mussels, which could be a
possible contributing factor of blooms (Knoll et al., 2008). In addition, it would be helpful to
consult with nearby PWSs to determine if they have had past issues with blooms or cyanotoxins
in their raw or finished water.
Source Water Assessment Information

In the absence of historical data on cyanobacterial blooms, or cyanotoxin occurrence, a useful
starting place is to review the PWS’s Source Water Assessment information. The 1996
39
amendments to the Safe Drinking Water Act (SDWA) Section 1453 required primacy agencies
to develop and implement Source Water Assessments. The Assessment delineates the Source
Water Protection Area of every PWS, creates an inventory of the significant potential sources of
contamination within the Protection Area, and evaluates the susceptibility of each system to
contamination (U.S. EPA, 1997). Some states update their Source Water Assessments on a
regular basis, while others have not conducted updates in recent years. While updates may vary
by state and system, Source Water Assessments remain a useful resource to help PWSs devise an
approach to conduct system-specific evaluations. The Assessment may contain relevant
information to cyanotoxin vulnerability, such as the locations of nutrient dischargers and flow
patterns. The Source Water Assessment may also contain information relevant to the other types
of information discussed later in this section that PWSs might consider for inclusion in the
system-specific evaluation. States may also want to update Source Water Assessments with new
information on cyanotoxin vulnerabilities, as local stakeholders often use the Assessment as a
baseline for source water protection plans and activities. For example, the Colorado Department
of Public Health and Environment (CDPHE) has helped Colorado water providers and their
communities develop nearly 150 protection plans with an additional 50 plans in progress. These
plans lay out a roadmap of targeted protection activities that can help curb cyanotoxin incidence.
EPA is developing a new tool to help states and utilities update source water assessment and
protection plans called Drinking Water Mapping Application for Protecting Source Waters
(DWMAPS). DWMAPS is a Web-based mapping tool that will allow users to identify and
analyze potential risks to local source waters. For example, the tool will display possible bloom
risk factors within Source Water Protection Areas, such as point sources of nutrients upstream,
land use factors (integrated from the National Land Cover Dataset), and county-level nitrogen
and phosphorus loading from agricultural lands. DWMAPS can also show which source waters
are listed as impaired for nutrients (and other causes) under the Clean Water Act (CWA) and
allow PWSs to easily and securely retrieve drinking water data specific to their own system on
intakes, wells, treatment plants and source water protection areas. Features of the map also
inform protective actions: for example, one feature of the tool will map projects funded through
CWA Section 319 to reduce nonpoint sources of pollution. DWMAPS will also offer “Web
services” that allow users to import data and GIS shapefiles from DWMAPS into their own GIS
platforms. The tool is expected to be available to states, utilities and the public in 2015.
Another useful piece of information is the impairment status of a source water. Under section
303(d) of the CWA, states, territories and authorized tribes develop lists of impaired waters.
These are waters that are too polluted or otherwise degraded to meet the water quality standards
set by states, territories or authorized tribes. Waters are prioritized and a Total Maximum Daily
Load (TMDL), must be developed for those waters. EPA’s MyWaters Mapper (U.S. EPA, 2011)
and How’s My Waterway each provide information about whether a water is listed as impaired
under section 303(d) of the CWA (U.S. EPA, 2015g). See the below table for links to these tools.
Climate and Weather Information

Evaluating climate and weather related events are also a key factor in assessing a PWS’s
vulnerability to cyanobacterial blooms and cyanotoxin occurrence. Warm ambient temperatures,
increased source water temperatures, rainfall events, drought and heavy winds can impact the
timing and duration of cyanotoxin occurrence. Some cyanobacteria have optimal temperatures at
which they will grow, increasing the likelihood of cyanotoxin occurrence. Optimal temperatures
will depend on the cyanobacteria genus and species (Robarts and Zohary, 1987; Mehnert et al.,
40
2010). Climates can impact vertical stratification within water bodies with temperate climates
having seasonal stratification, whereas tropical climates more typically exhibit a diurnal
stratification pattern (WHO, 1999).
Weather patterns, such as high intensity rainfall events, can increase runoff into source waters,
thus potentially increasing nutrient loading levels that may lead to increased production of
cyanobacteria and cyanotoxins (Paerl, 2008). Drought can also lead to conditions that favor
cyanobacterial production. For example, drought was a major factor that caused low flow
conditions in the Barwon-Darling River, contributing to the growth of a large bloom in 1991
(Bowling and Baker, 1996). Climate change can also increase bloom occurrence and expand the
geographic range as water temperatures increase and new weather patterns emerge, such as
changes in the frequency and intensity of rainfall events (Parmesan, 2006; Paerl and Huisman,
2009; Markensten et al., 2010).
PWSs are encouraged to take into account changes in climactic conditions impacting their source
water when conducting the system-specific evaluation. EPA has developed a Climate Ready
Water Utilities (CRWU) initiative to assist the water sector, including drinking water, in
addressing climate change impacts (U.S. EPA, 2015f). There are multiple tools available within
the initiative including the Climate Resilience Evaluation and Awareness Tool (CREAT)
(U.S. EPA, 2013c). CREAT is a software tool designed to assist users (including PWSs)
understand the potential impact of climate change on their utility. The tool can assist users by
identifying climate change projections for a utility, such as annual total precipitation and annual
average temperature, and evaluate possible adaptation options. Additional information can be
found in EPA’s factsheet on Impacts of Climate Change on the Occurrence of Harmful Algal
Blooms (U.S. EPA, 2013d).
Land Use
EPA suggests PWSs consider land use information in the source watershed as part of the system-
specific evaluation. Certain land uses are more closely linked to cyanotoxin occurrence,
particularly those that result in excess nutrient discharges. A study by Beaver et al. (2014) found
strong associations between agricultural land cover and microcystins occurrence in three
ecoregions. Certain other land uses can increase nutrients in source waters, leading to increased
bloom activity. For example, if a PWS’s source water is vulnerable to nutrient rich runoff from
agriculture or urban areas, the PWS may be vulnerable to cyanotoxins as well. Modification of
watersheds by agriculture, urban and industrial development can lead to cyanobacterial blooms
(Paerl, 2008). Hydrological changes, such as construction of impoundments, damming rivers,
and water extraction for irrigation purposes, can also impact the sources waters (Bowling and
Baker, 1996; Paerl, 2008; WHO, 1999). A helpful tool to determine land use activities around
source waters is the National Land Cover Database (NLCD). This database was developed by a
consortium of federal agencies that generate land cover information at the national scale for a
wide variety of applications such as land management and modeling (MRLC, 2015). NLCD
indicates land cover by wetlands, open water, forested lands, pasture lands, cropland, urban areas
and many more use categories. Another useful tool that incorporates some of the features of the
NLCD is the NHDPlus. NHDPlus offers a watershed delineation tool that could be helpful in
determining the entire watershed area surrounding a source water. A user can also look at the
local catchments that buffer the water body and calculate various attributes available in
NHDPlus including percent land use, drainage area features, and flow volume and velocity
estimates through source water protection areas (NHDPlus, 2015).
41
Several tools exist that help communities plan land use in order to mitigate impacts on source
waters. Advice Worth Drinking: A Planner’s Guide outlines a variety of simple steps that land
use planners can take to integrate source water protection into regular planning activities and
“Smart Growth” strategies like Visioning and zoning. Many local groups like the Salmon Falls
Watershed Collaborative have successfully implemented land conservation practices to curb the
impacts of urbanization on water quality. The Source Water Collaborative website outlines
funding mechanisms that can help PWSs work with local communities to finance land
conservation in source water protection areas. In addition, the Collaborative’s Conservation
Partners Toolkit outlines steps for working with U.S. Department of Agriculture (USDA)
programs and Conservation Districts to implement agricultural “best management practices” in
lands impacting drinking water.
Nutrient Levels
As part of the weight of evidence evaluation for source water cyanotoxin vulnerabilities, PWSs
may want to include water quality parameters such as nutrient levels. These factors can be
anthropogenic or naturally occurring. For the purpose of a system-specific evaluation, PWSs
might evaluate water quality parameters based on available data and on the conditions typically
seen in the source water. A change in water quality parameters that may occur leading to a bloom
is discussed earlier in Appendix A.
Source water nutrient levels could be important to examine as part of a system-specific

evaluation. Nutrients, specifically nitrogen and phosphorus, play a large role in bloom
occurrence (WHO, 1999; Jacoby et al., 2000, Dolman et al., 2012, Yoshida et al., 2007).
Increases in nitrogen and phosphorus are mainly due to nonpoint source pollution (such as from
agricultural or urban runoff), point source pollution (such as wastewater treatment plants and
agricultural and municipal discharges) and subsurface drainage from groundwater and septic
systems (Paerl, 2008). A recent analysis of a national dataset indicated that total nitrogen and
chlorophyll-a concentrations were strong predictors of microcystin occurrence (Yuan et al.,
2014). Cyanobacteria are also capable of fixing nitrogen in low nitrogen conditions and can store
phosphorus (Mitsui et al., 1986, Paerl and Otten, 2013).
There are many available tools a PWS can use to evaluate water quality parameters to inform
vulnerability findings such as EPA’s MyWaters Mapper, which provides snapshots of Office of
Water program data including water impairments and water monitoring data (U.S. EPA, 2011)
and How’s My Waterway, which provides conditions of local waters quickly and in plain
language (U.S. EPA, 2015g). The Water Quality Data Portal contains source (raw) water
monitoring data for a range of physical, chemical and biological parameters. The data in the
Portal come from federal, state and tribal water quality agencies, volunteer groups and academia
(U.S. EPA, 2012b). The DWMAPS mapping tool to be released in 2015, as described above, also
compiles information on point and nonpoint sources of nutrients in source waters, which can be
used to help a system evaluate source water. For those systems wishing to conduct modeling
analyses, the USGS has a modeling tool called SPARROW that is useful in modeling and
mapping nutrient loading in watersheds across the U.S. (USGS, 2011). Similarly, EPA’s BASINS
modeling framework provides a platform for analyses for users wishing to conduct simple
nutrient loading analyses, to more complex water quality modeling (U.S. EPA, 2013e). Yet
another available tool is EPA’s Nitrogen and Phosphorus Pollution Data Access Tool (NPDAT).
This tool draws from multiple sources to provide focused information on the extent and
42
magnitude of nitrogen and phosphorus pollution in U.S. water and potential sources of the
pollutants (U.S. EPA, 2014c). USGS has a National Water Quality Assessment Program
(NAWQA) that can help provide information on nutrients, among other water quality
parameters, including how conditions are improving or getting worse over time (USGS, 2015).
Additional tools can be used to identify point sources of pollution in source waters. EPA’s
Enforcement and Compliance History Online (ECHO) tool contains information and maps of
facility inspections, enforcement and violation history. For example ECHO can provide
information on National Pollutant Discharge Elimination System (NPDES) dischargers in
noncompliance and significant noncompliance (U.S. EPA, 2015h). EPA’s Discharge
Monitoring Report (DMR) Loading Tool can provide information on nutrient discharge
volumes at NPDES-permitted facilities, as well as facilities under the Toxics Release Inventory.
It is designed to help determine who is discharging and what, where and how much they are
discharging (U.S. EPA, 2015i). This may be a useful source of information about upstream
contributions of nutrients to source water protection areas.
43
Table A-1. List of Resources and Tools
Resource and Tool Organization Link
CyanoHABs: Cyanobacterial United States Environmental http://www2.epa.gov/nutrient-

Harmful Algal Blooms Website Protection Agency policy-data/cyanohabs
Drinking Water Advisories and United States Environmental http://www2.epa.gov/nutrient-

Health Effects Support Protection Agency policy-data/health-and-
Documents (2015) ecological-effects
Opportunities to Protect Association of Clean Water http://www.asdwa.org/docume

Drinking Water Sources and Administrators, Association nt/docWindow.cfm?fuseaction
Advance Watershed Goals of State Drinking Water =document.viewDocument&d
Through the Clean Water Act: Administrators, Ground ocumentid=3007&documentF
A Toolkit for State, Interstate, Water Protection Council ormatId=3779
Tribal and Federal and United States
Water Program Managers Environmental Protection
(2014) Agency
Drinking Water Advisory United States Centers for http://www.cdc.gov/healthywa

Communication Toolbox Disease Control and ter/pdf/emergency/drinking-
(Updated 2013) (includes Prevention water-advisory-
communication outreach tools) communication-toolbox.pdf
How to Collaborate Toolkit Source Water Collaborative http://www.sourcewatercollab

orative.org/how-to-
collaborate-toolkit/
Developing Risk United States Environmental http://water.epa.gov/infrastruct

Communication Plans for Protection Agency ure/watersecurity/lawsregs/upl
Drinking Water Contamination oad/epa817f13003.pdf
Incidents (2013)
Water Security Initiative: United States Environmental http://www.epa.gov/watersecu

Interim Guidance on Protection Agency rity/pubs/guide_interim_cmp_
Developing Consequence wsi.pdf
Management Plans for Drinking
Water Utilities (2008)
44
A Water Utility Manager’s American Water Works http://www.waterrf.org/Public

Guide to Cyanotoxins Association and Water ReportLibrary/4548a.pdf
Research Foundation
Optimizing Conventional Water Research Foundation WRF Report 4315

Treatment for the Removal of
Cyanobacteria and Toxins
Toxic Cyanobacteria in Water: World Health Organization http://www.who.int/water_san

A guide to their public health itation_health/resourcesquality
consequences, monitoring and /toxcyanbegin.pdf
management (1999)
Cyanobacterial Toxins Health Canada http://www.hc-sc.gc.ca/ewh-

Microcystin – LR Guideline semt/alt_formats/hecs-
(Updated 2002, currently sesc/pdf/pubs/water-
undergoing revision) eau/cyanobacterial_toxins/cya
nobacterial_toxins-eng.pdf
International Guidance Manual Global Water Research http://www.waterra.com.au/cy

for the Management of Toxic Coalition, Water Quality anobacteria-
Cyanobacteria (2009) Research Australia manual/PDF/GWRCGuidance
ManualLevel1.pdf
Management Strategies for Water Quality Research http://www.waterra.com.au/pu

Cyanobacteria (blue-green Australia blications/document-
algae): A Guide for Water search/?download=106
Utilities (2010) (includes
drinking water cyanotoxin
values for various countries)
Bloom Characterization Guide Ohio Environmental http://epa.ohio.gov/portals/28/

Protection Agency Documents/HAB/BloomChara
cterizationGuide-DRAFT.pdf
Drinking Water Mapping United States Environmental (Anticipated release in 2015)

Application for Protecting Protection Agency
Source Waters (DWMAPS)
National Hydrography Dataset Horizon Systems http://www.horizon-

(NHDPlus) Corporation (in coordination systems.com/NHDPlus/NHDP
with EPA and USGS) lusV1_tools.php
Climate Ready Water Utilities United States Environmental http://water.epa.gov/infrastruct

(CRWU) Protection Agency ure/watersecurity/climate/inde
x.cfm
45
Climate Resilience Evaluation United States Environmental http://water.epa.gov/infrastruct

and Awareness Tool (CREAT) Protection Agency ure/watersecurity/climate/creat
.cfm
National Land Cover Database Multi-Resolution Land http://www.mrlc.gov/

(NLCD) Characteristics Consortium
MyWaters Mapper United States Environmental http://www.epa.gov/waters/my

Protection Agency watersmapper/
How’s My Waterway United States Environmental http://watersgeo.epa.gov/myw

Protection Agency aterway/
STORET (STOrage and United States Environmental http://www.epa.gov/storet/

RETrieval) Protection Agency
BASINS (Better Assessment United States Environmental http://water.epa.gov/scitech/da

Science Integrating point & Protection Agency tait/models/basins/index.cfm
Nonpoint Sources)
Nitrogen and Phosphorus United States Environmental http://www2.epa.gov/nutrient-

Pollution Data Access Tool Protection Agency policy-data/nitrogen-and-
(NPDAT) phosphorus-pollution-data-
access-tool
Preventing Eutrophication: United States Environmental http://www2.epa.gov/sites/pro

Scientific Support for Duel Protection Agency duction/files/documents/nandp
Nutrient Criteria (2015) factsheet.pdf
SPARROW (Surface Water United States Geological http://water.usgs.gov/nawqa/s

Quality Monitoring) Survey parrow/
National Water Quality United States Geological http://water.usgs.gov/nawqa/

Assessment Program Survey
Enforcement and Compliance United States Environmental http://echo.epa.gov/

History Online (ECHO) Protection Agency
Discharge Monitoring Report United States Environmental http://cfpub.epa.gov/dmr/

(DMR) Loading Tools Protection Agency
Advice Worth Drinking: A Source Water Collaborative http://www.sourcewatercollab

Planner’s Guide orative.org/guide-for-land-
use-planners/
46
Source Water Protection Source Water Collaborative http://www.sourcewatercollab

funding tools orative.org/how-to-
collaborate-
toolkit/maintaining/sustainable
-funding/
Source Water Protection and Source Water Collaborative http://www.sourcewatercollab

Conservation Partners Toolkit orative.org/swp-conservation-
partners-toolkit/
47
Appendix B Examples of Water Utility and Community Responses to
The following short examples highlight instances where a bloom impacted drinking water
supplies and may have resulted in adverse public health effects. These case studies highlight the
potential technological and policy options that were implemented. These studies represent
examples that have been cited in peer-reviewed publications; there are likely many more
anecdotal examples.
Carroll Township, Ohio

In September 2013, microcystins concentrations at Carroll Township’s intake on Lake Erie
increased to >5 μg/L, the highest concentration observed at the intake in four years of
monitoring. A finished water sample collected at the same time had a microcystins concentration
of 1.4 μg/L, which exceeded Ohio EPA’s microcystins threshold of 1 μg/l. Repeat samples had a
source water concentration of 13 μg/L and a finished water concentration of 3.6 μg/L. Ohio EPA
recommended that the water system issue a Do Not Drink Advisory and transition to an
emergency connection with a neighboring utility. The advisory impacted over 2,200 people and
lasted approximately 48 hours. The water system remained on their emergency connection for
several weeks, until microcystins source water concentrations improved and treatment was
demonstrated to be effective. The water system utilizes conventional surface water treatment
with pre-ozonation. After the event, the system spent approximately $250,000 on plant upgrades,
including new ozone generators and concentrators that enabled them to increase their ozone
dose. In 2014, the upgraded and optimized plant was able to effectively treat source water
microcystins concentrations of up to 18 μg/L, with no finished water detections.
Toledo, Ohio
In early August 2014, microcystins concentrations at Toledo’s intake on Lake Erie rapidly
increased to 14 μg/L. The microcystins concentration in the finished water was 2.5 μg/L and
repeat samples run by two different analysts verified the above-threshold levels. Ohio EPA
recommended that the city issue a Do Not Drink Advisory. Microcystins were also detected
above Ohio EPA thresholds in eight distribution samples.
After the finished water detections, Toledo optimized the plant for microcystins removal by
increasing powder activated carbon (PAC) from 6.3 to 15 mg/L, and increasing alum and
chlorine doses. Microcystins concentrations in the finished water decreased to near the detection
limit and the city lifted the advisory. The advisory affected approximately half a million people
and lasted 55 hours. Eleven days after the advisory was lifted, microcystins concentrations at
Toledo’s intake increased to >50 μg/L, the highest concentration observed at the intake in five
years of monitoring. At that point the water system was already optimizing for microcystins
removal, with increased carbon dose and chemical feeds. While microcystins was again detected
in finished water samples, concentrations did not exceed the 1.0 μg/L threshold. The city is
undergoing extensive plant rehabilitation and upgrades, including increasing the PAC feed
capacity to 40 mg/L and installing additional PAC feed locations. Costs for initial upgrades were
estimated at $4.4 million. The city is also planning to pilot the use of ozone as part of more
comprehensive long-term plant upgrades.
48
Lake Taihu, Wuxi, China (Zhang et al., 2010)
A bloom of Microcystis aeruginosa developed on Lake Taihu in May 2007, resulting in color,
taste, and odor issues for approximately two million people who depend on the lake for drinking
water. The bloom was likely the result of eutrophication and industrial and domestic wastewater
discharges. The drinking water issues led to a shortage of bottled water, public panic, and
economic impacts in the area. In response to the bloom, an emergency drinking water treatment
process was developed to prepare for future incidents and the local government developed
stricter standards for waste discharges and increased efforts for wetland restoration surrounding
the lake. The treatment process resolved the odor issues and consisted of adding potassium
permanganate at the intake and powdered activated carbon in the treatment plant. Microcystin-
LR levels were reduced in the source waters from 7.59 μg/L on May 30 to 0.73 μg/L on June 2.
Sulejow Reservoir, Lodz, Poland (Jurczak et al., 2005)

Sulejow Reservoir is frequently subject to blooms due to eutrophication, which impacts the water
quality of the nearby cities of Lodz and Tomaszow. Testing was conducted in the summers of
2002 and 2003 to determine the microcystins removal effectiveness of water treatment processes.
During this time, the peak concentrations of microcystins in raw water samples were 4.7 μg/L
dissolved and 3.3 μg/L cell-bound. The treatment plant for the Sulejow-Lodz system employed
pre-oxidation, coagulation, filtration, ozonation and chlorination. This treatment train sufficiently
removed microcystin levels to below detection limits in drinking water samples, especially
during the filtration process. To further reduce the levels of microcystins in the raw water, the
water utility utilized more ground water sources.
Solomon Dam, Queensland, Australia (Griffiths and Saker, 2003)

Over 100 children on Palm Island exhibited gastroenteritis in November 1979. Following the
hospital admissions, local officials hypothesized that the outbreak could be connected to the local
water supply source at Solomon Dam and cyanotoxins such as cylindrospermopsin. An
epidemiological study (Bourke et al., 1983) later examined the link between the outbreak and the
water supply and suggested that the use of copper sulfate in the water supply may have increased
the exposure to cyanotoxins in the treated water. The reservoir had recently been treated for an
algal bloom using copper sulfate, which can lyse cyanobacterial cells and release extracellular
toxins, which are not effectively removed by filtration.
Darling-Barwon River, New South Wales, Australia (Herath, 1995)

A cyanobacterial bloom affected more than 1,000 kilometers of the Darling-Barwon River in
Australia in 1991. Estimates for the economic losses were more than $1.3 million in direct
treatment and water supply costs, one million people-days of drinking water, and 2,000 site-days
of recreation that were valued to be at least $10 million. The local government appointed a Blue-
Green Algae Tasks Force in response, which recommended various policies to reduce the
impacts and occurrence of blooms in the future. One of the primary targets was reducing
phosphorus loadings in the river basin that originated from detergents and other cleaning agents
that end up in sewage treatment plants. A phosphorus permit trading program was also suggested
as a potential policy solution to control nutrient inputs.
49
Appendix C Key Questions and Answers
Where in the country are harmful algal blooms a problem?

Harmful algal blooms (HABs) are a national concern. HABs have impacted waters across many
regions of the U.S. EPA recommends that drinking water systems in all areas of the country that
use surface water sources, such as lakes and reservoirs, assess their water source’s vulnerability
to HABs. EPA estimates that lakes and reservoirs that serve as sources of drinking water for
between 30 and 48 million people may be periodically contaminated by algal toxins.
How do cyanotoxins produced by some harmful algal blooms affect drinking water
quality?
HABs can occur in source waters used for drinking water. Winds and water currents can
potentially transport HABs within proximity to drinking water intakes at treatment plants. If not
removed during drinking water treatment, exposure to cyanotoxins in tap water could potentially
affect human health. Algal blooms may also cause aesthetic problems (earthy and musty smell)
and affect the taste of treated drinking water.
What are the health effects from exposure to cyanotoxins in drinking water?
Effects including gastroenteritis, and liver and kidney damage have been reported in humans
following acute or short-term exposure to cyanotoxins in drinking water. Recreational exposure
to cyanobacterial blooms has been reported to lead to allergic reactions, including hay fever-like
symptoms; skin rashes; and gastrointestinal distress. However, more research is needed to
quantify these effects.
What about using water with elevated algal toxins for showering and other uses?
The Health Advisory values for two key algal toxins (microcystins and cylindrospermopsin) are
specifically for consumption of drinking water. Exposure to cyanobacteria and their toxins may
also occur by ingestion of toxin-contaminated food, including consumption of fish, and by
inhalation and dermal contact during bathing or showering. While these types of exposures
cannot be quantified at this time, they are assumed to contribute less to the total cyanotoxin
exposures than ingestion of drinking water. While information is not currently available to
determine safe concentrations for showering, bathing, or other uses, EPA expects that it is
unlikely that showering or bathing in water with cyanotoxin levels near or below the Health
Advisory will present a health risk. As our understanding of algal toxin health effects continue to
develop, EPA will continue to evaluate their health effects for other uses.
Are immunocompromised individuals or infants fed by nursing mothers at risk?

Populations such as nursing mothers and pregnant women, the elderly, and immune-
compromised individuals or those receiving dialysis treatment may be more susceptible than the
general population to the health effects of microcystins. As a precautionary measure,
immunocompromised individuals and nursing mothers may want to consider following the
recommendations for bottle-fed infants and young children of pre-school age.
For additional information please see Section 4 on Risk Characterization in the Health
Advisories for microcystins and cylindrospermopsin for additional information available at
http://www2.epa.gov/sites/production/files/2015-06/documents/microcystins-report-2015.pdf
and http://www2.epa.gov/sites/production/files/2015-06/documents/cylindrospermopsin-
report-2015.pdf
50
Are the Health Advisory values safe for all children (regardless of age)?
The microcystins and cylindrospermopsin HA values are set at levels at which adverse health
effects are not expected to occur. EPA would not expect adverse health effects for children who
are school-aged when microcystins concentrations in drinking water are at or below 1.6 µg/L and
cylindrospermopsin concentrations are at or below 3 µg/L. EPA would not expect adverse health
effects for bottle-fed infants and young children of pre-school age (less than six years old) when
microcystins concentrations in drinking water are at or below 0.3 µg/L and cylindrospermopsin
concentrations are at or below 0.7 µg/L.
Why is EPA’s value different from the World Health Organization’s (WHO) value?
In 1998, the World Health Organization (WHO) released a provisional guideline of 1 μg/L for
microcystin-LR in drinking water. The derivation of this value differs from EPA’s Health
Advisory guideline derivation in two ways: the duration of exposure and exposure parameters
(body weight and drinking water consumption) used for the calculation of the values.
WHO’s guideline is based on risk from a lifetime of consumption and was calculated using data
from a study of longer duration (13 weeks). EPA based the HA on a more recent study of shorter
duration (28 days). EPA believes that this shorter exposure duration is more representative of
how people may be exposed to cyanotoxins in their drinking water, from sporadic blooms rather
than exposure over a lifetime of consumption.
The exposure parameters used for the WHO guideline for body weight (60 kg) and average water
intake (2 L/day) are different from the parameters EPA used (adult average body weight of 80 kg
consuming 2.4 L/day). EPA’s exposure values are based on statistics of the U.S. population, and
are routinely used for risk assessment purposes.
Why has EPA developed Health Advisories for two cyanotoxins: microcystins and
cylindrospermopsin?
Cyanobacteria and their toxins have been found in drinking water systems and recreational
waters in the U.S. Many states have expressed concerns regarding the presence of cyanotoxins in
tap water and in surface waters, causing use impairments due to near-shore algae buildup.
Currently, there are no U.S. federal water quality criteria or standards, or regulations concerning
the management of HABs in drinking water under the Safe Drinking Water Act (SDWA) or in
ambient waters under the Clean Water Act (CWA).
Based on the toxicology and epidemiology data, EPA found there are adequate data to develop
HAs for microcystins and cylindrospermopsin. EPA issued these Health Advisories to assist state
and local authorities in their efforts to address cyanotoxin risks.
What happens if you exceed the Health Advisory level? If it is a Ten-day value, what
happens if you exceed for a smaller number of days?
The Health Advisory levels for microcystins and cylindrospermopsin are non-regulatory
concentrations of drinking water contaminants at which adverse health effects are not anticipated
to occur over a Ten-day exposure period. Because it is difficult to determine in advance the
duration of elevated algal toxin levels, EPA recommends that water systems begin to take actions
once the elevated levels have been confirmed by additional samples. Additionally, because of
time needed to process sequential analytical tests, it can take several days following the detection
of a bloom and/or cyanotoxins before concentrations above the HA values are confirmed in
51
finished water. Therefore, EPA recommends initiating the response activities as soon as
practicable.
What about pets exposed to cyanotoxins through drinking water?

Pets are at greatest risk from exposure to cyanotoxins from consuming scum and mats, licking
their fur after swimming in contaminated water, and drinking water from a water body
contaminated by cyanobacteria. However, pets could also get sick if they drink tap water
contaminated with high concentrations of cyanotoxins.
What is EPA doing to address problem of cyanotoxins in drinking water sources and other
waters?
EPA has developed Health Advisories that will provide states, drinking water utilities and the
public with information on health effects of cyanotoxins, and methods to sample and treat
cyanotoxins in drinking water. EPA will continue to work with states, local communities and
stakeholders to provide technical support on key steps that can be taken to protect the public
from exposure to algal toxins.
EPA is planning to develop ambient water quality criteria for the protection of human health for
recreational uses for microcystins, cylindrospermopsin, and anatoxin-a, if adequate data are
available.
EPA is working diligently with its partners to address nitrogen and phosphorus pollution (known
to create environmental conditions favorable to HABs) including:
• Providing states with technical guidance and resources to help them develop water
quality criteria for nitrogen and phosphorus as part of their water quality standards for
surface waters.
• Working with states to identify waters with nitrogen and phosphorus pollution and to
develop Total Maximum Daily Loads (TMDLs) to restore the waters by limiting
allowable nutrient inputs.
• Awarding grants to states for operating nonpoint source management programs.
• Administering a permit program that restricts the amount of nitrogen and phosphorus
released to the environment from point sources, such as wastewater treatment plants.
• Providing funding for the construction and upgrading of municipal wastewater facilities
and the implementation of nonpoint source pollution control and estuary protection
projects.
• Working with state and federal partners on the Mississippi River and Gulf of Mexico
Watershed Nutrient Taskforce to reduce hypoxia.
• Conducting and supporting research on nitrogen and phosphorus pollution-related topics.
EPA has worked closely with the U.S. Department of Agriculture (USDA) to focus investment in
priority watersheds across the country through the National Water Quality Initiative. In addition,
USDA funds are available through a variety of programs to control agricultural sources of runoff
through suites of conservation practices, and many states are partnering with USDA to do so.
EPA is working to help educate the public and stakeholders about nutrients and HABs, leading a
HABs public awareness campaign, including expert webinars and publications.
52
Are there point-of-use treatment devices consumers can use at their home or workplace to
protect themselves from cyanotoxins in drinking water?
EPA is unaware of point-of-use devices that have been demonstrated to be effective for removal
of cyanotoxins from drinking water. Third-party organizations are currently developing
certification standards to test point-of-use devices and evaluate how reliably they can remove
cyanotoxins from drinking water.
Can algal blooms and their toxins affect ground water wells?
Typically no, unless they are “Ground Water Under the Direct Influence of Surface Water”
wells. Cyanobacteria require sunlight to survive. Most groundwater wells are not expected to be
impacted by cyanotoxins.
What research is EPA doing on harmful algal blooms?

EPA researchers are conducting HAB research on water quality, human and ecological health
effects, monitoring and analytical methods for rapid detection, and drinking water treatment
research related to HABs.
Specifically for drinking water treatment, EPA researchers are conducting a Lake Erie field study
that monitors cyanobacteria at numerous treatment plants to define the start and end of bloom
events, and the water quality changes that take place through the treatment plant. In addition, the
researchers are working on science to improve the ability of existing treatment processes to
remove cyanobacterial toxins and to improve the performance of existing operations by
modifying the locations where treatment chemicals are applied, and the types and concentrations
of chemicals applied.
EPA is partnering with NOAA, USGS, and NASA on the Cyanobacteria Assessment Network
(CyAN) project that includes making satellite data processed for cyanobacteria abundance
available for large inland lakes nationwide once new satellite sensors come online and the data
has been evaluated.
What funding sources are available to states for HAB monitoring?

Eligible federal funding sources for source water monitoring include CWA Section 106 base
funding and the 106 monitoring initiative enhancement funds received annually, CWA Section
604(b) and potentially CWA Section 319. There are also funds for specific geographic programs,
like the Great Lakes, Lake Champlain and Lake Pontchartrain, which received targeted funds
that may be available for HAB monitoring. In addition, the Drinking Water State Revolving
Fund (DWSRF) set-asides may be used as part of a state's strategy to build technical, financial,
and managerial capacity of public water systems. For example, these funds may be used for
demonstration purposes to build the capacity of the system for activities such as monitoring.
53
Appendix D Potential Language for Use in a Cyanotoxin Public
Notification and Social Media
Above Health Advisory Value for Bottle-Fed Infants and Young Children of Pre-School
Age but Below Health Advisory Value for School-Age Children through Adults
We are providing this notice to make you aware of the presence of low levels of [insert specific
cyanotoxin] in your treated drinking water. [insert specific cyanotoxin here] is a cyanotoxin,
created by certain types of harmful algal blooms in the source water of your drinking water
supply. The cyanobacteria in harmful algal blooms that produce [insert specific cyanotoxin here]
can grow rapidly when certain environmental conditions are favorable for their growth. Since not
all ‘blooms’ produce these cyanotoxins, we sampled the treated drinking water to determine if
cyanotoxins were present when a suspected bloom was occurring. The sampling conducted on
insert date and confirmed on insert date indicated that the cyanotoxins are present in treated
drinking water. We are adjusting our treatment operations to reduce concentrations of [insert
specific cyanotoxin here] as quickly as possible.
Insert boundaries of service area affected
The cyanotoxins were detected at levels that may present a risk to bottle-fed infants and young
children of pre-school age if ingested. Some consumers who have a suppressed immune system
may also be at risk. We recommend that bottle-fed infants and young children of pre-school age
and consumers with suppressed immune systems use an alternative source of water for ingestion
and the preparation of infant formula use an alternative source of water until further notice.
Boiling water will not remove the cyanotoxins. The water is considered safe for humans for
bathing, showering, washing hands, watering yards and gardens, washing dishes, flushing toilets,
cleaning, laundry, and shaving.
We are working diligently to correct this problem, and do not expect this problem to last more
than insert number of days. We will provide further notice when the water is again safe to use for
all purposes. Additional information can be found at insert link to additional information. If you
have any questions or concerns please contact us at insert phone number.
Above Health Advisory Value All Consumers
We are providing this notice to make you aware of the presence of [insert specific cyanotoxin] in
your treated drinking water. [insert specific cyanotoxin here] is a cyanotoxin, created by certain
types of harmful algal blooms in the source water of your drinking water supply. The
cyanobacteria in harmful algal blooms that produce [insert specific cyanotoxin here] can grow
rapidly when certain environmental conditions are favorable for their growth. Since not all
‘blooms’ produce these cyanotoxins, we sampled the treated drinking water to determine if
cyanotoxins were present when a suspected bloom was occurring. The sampling conducted on
insert date and confirmed on insert date indicated that the cyanotoxins are present in treated
drinking water. We are adjusting our treatment operations to reduce concentrations of [insert
specific cyanotoxin here] as quickly as possible.
54
Insert boundaries of service area affected
The cyanotoxins were detected at levels which may present a risk to all consumers. We
recommend that you use an alternative source of water for ingestion and in the preparation of
infant formula until further notice.
Boiling water will not remove the cyanotoxins. The water is considered safe for humans for
bathing, showering, washing hands, watering yards and gardens, washing dishes, flushing toilets,
cleaning, laundry, and shaving.
We are working diligently to correct this problem, and do not expect this problem to last more
than insert number of days. We will provide further notice when the water is again safe to use for
all purposes. Additional information can be found at insert link to additional information. If you
have any questions or concerns, please contact us at insert phone number.
Potential Twitter Notice

Above Health Advisory Value for Bottle-Fed Infants and Young Children of Pre-School Age but
Below Health Advisory Value for School-Age Children through Adults
Insert name of area served by the contaminated water’s tap water may contain low levels of
[insert specific cyanotoxin here], a toxin. Please see insert website link for more information.
(Link would be to the full notices described at the beginning.)
Above Health Advisory Value All Consumers
Insert name of area served by the contaminated water’s tap water may contain [insert specific
cyanotoxin here], a toxin. Do not drink until further notice. See insert website link for more
information. (Link would be to the full notices described at the beginning.)
55
Appendix E Long Term Mitigation Strategies and Treatment Options
Source Water Protection
PWSs may want to consider including long-term source water protection actions as part of a
CMP to help prevent blooms and reduce source water vulnerability. This is especially important
when source waters are determined to have ongoing cyanotoxin vulnerabilities and recurrence of
blooms is likely (WHO, 1999). These actions occur at the watershed level to provide sustained
protection of the source water and mitigate conditions that are conducive to cyanobacterial
growth. Perhaps the most important source water protection step to address blooms is reduction
of nutrient inputs, including both nitrogen and phosphorus.
Implementing source water protection strategies can reduce costs to the PWS, as watershed
protection programs are often much less expensive than having to employ additional drinking
water treatment (WRI, 2013). The Source Water Collaborative, a group of 26 national
organizations dedicated to protecting sources of drinking water, provides detailed information on
specific actions that can help prevent nutrient pollution. The Collaborative’s “How to
Collaborate Toolkit”, previously described in the introduction and section 1 of this document,
helps watershed stakeholders form partnerships to implement source water activities and
includes information on funding source water protection work and planning for investment
(SWC, 2015a). The Source Water Collaborative’s “How to Collaborate Toolkit” also includes a
section on implementing pilot projects. It provides profiles on examples of source water
management practices like manure storage systems, land conservation and GIS decision-support
(SWC, 2015a). Another resource is the Source Water Collaborative’s Conservation Partners
toolkit, designed for source water protection officials. This toolkit offers a step-by-step guide for
understanding conservation programs and how to collaborate with key partners like Soil and
Water Conservation Districts and U.S. Department of Agriculture (USDA) State
Conservationists (SWC, 2015b). The Source Water Collaborative also provides a Planner’s
Guide that outlines land use decisions that can affect current and future drinking water supplies,
either intentionally or inadvertently. This guide describes how urban and land use planners can
integrate source water protection into regular planning activities (SWC, 2015c).
Controlling nutrient inputs requires the cooperation of many programs and stakeholders.
Opportunities to Protect Drinking Water Sources and Advance Watershed Goals through the
Clean Water Act is a toolkit that describes ways PWSs can coordinate with SDWA and CWA
programmatic activities, such as point source permitting, water quality standards, listings, Total
Maximum Daily Loads (TMDLs) and Section 319 program funding to protect drinking water
sources (ASDWA, 2014).
Please see Appendix A for a list of tools and their links referenced in this document, including
the tools mentioned above.
Alternative Drinking Water Sources
A long-term option PWSs could consider is to switch to an alternative drinking water source
either permanently or during known bloom seasons, if available. PWSs could also consider
utilizing or installing multiple drinking water intakes (at various depths or locations) in a source
water, especially when the source water is a large body of water. For example, a bloom occurring
56
on a lake may not be impacting the entire lake and multiple intakes may allow water to be drawn
from an unaffected section of the lake or unaffected depth within the lake. Intakes in shallow
water may be at more risk than deeper intakes, especially in areas prone to high cyanobacterial
densities (Loftin et al., 2008).
Permanent Treatment Options
If PWSs are frequently challenged by cyanotoxins in source waters and modifications to their
existing treatment systems do not sufficiently reduce cyanotoxin levels to below HA values, they
may consider installing permanent treatment systems as long-term, cost effective alternatives to
address cyanotoxin occurrence in future years. These permanent treatment systems will likely
require significant capital investment and long-term resource planning. PWSs may want to
consult with the states and primacy agencies on plan review and approval requirements. PWSs
may need to conduct an evaluation of different treatment technologies to select the most cost-
effective option and associated design and operational parameters to achieve multiple treatment
goals including cyanotoxins removal.
Permanent treatment options that are effective for removing intracellular cyanotoxins include
dissolved air flotation (DAF), microfiltration (MF), and ultrafiltration (UF). Permanent treatment
options that can be effective for removing extracellular cyanotoxins include ozone, GAC,
biological filtration, nanofiltration (NF), reverse osmosis (RO), ultraviolet (UV) with hydrogen
peroxide, and other emerging technologies. Each of these technologies is briefly described
below. More details can be found in the literature (Westrick et al., 2010; Newcombe et al., 2010;
WHO, 1999).
Treatment Options for Intracellular Toxin Removal
Dissolved Air Flotation. DAF is effective particularly for light cells and species containing gas
vesicles, which typically form surface scums. Waters of high color and low turbidity are best
suited for flotation processes (Sklenar et al., 2014; U.S. EPA, 2014a; Tokodi et al., 2012,
Markham et al., 1997; Mouchet and Bonnélye, 1998).
MF and UF. Membrane filtration technologies have demonstrated the ability to remove
cyanobacterial cells and their extracellular toxins from drinking water to varying degrees. For
systems having these technologies at their disposal, using them during a bloom could prove
beneficial. MF and UF are highly effective and can be used alone or as a replacement for
conventional filtration in the removal of intact cyanobacterial cells (Westrick et al., 2010). These
membranes have been reported to remove more than 98 percent of whole cells (Chow et al.,
1997; Teixiera and Rosa, 2006).
Treatment Options for Extracellular Toxin Removal
Ozonation. Use of ozone, after removing the cells through a physical process, is the most
efficient process for the destruction of extracellular microcystins and cylindrospermopsin
(Newcombe et al., 2010; WHO, 1999; U.S. EPA, 2014a; Rodríguez et al., 2007b). Ozone is
effective under most water quality conditions, contact times and doses (0.5-1.1 mg/L)
encountered in drinking water treatment (Alvarez et al., 2010). The effectiveness of ozone is
negatively impacted by the dissolved organic carbon concentrations (Alvarez et al., 2010).
57
Australian guidance recommends a CT value of 1.0 mg/L-min at a pH of 7 or higher to treat for
microcystins and cylindrospermopsin (Newcombe et al., 2010). Use of ozone can result in
biodegradable organics and bromate. Bromate is an inorganic DBP regulated by EPA (EPA,
1999). Water systems should conduct system-specific tests to determine the optimal ozone dose
that maximizes toxin degradation and also ensures that the running annual average bromate MCL
of 0.01 mg/L would not be exceeded.
Granular Activated Carbon (GAC). GAC removes microcystins through adsorption to the media
and (or) biodegradation by microorganisms residing on the filter media (Wang et al., 2007). Like
PAC, GAC effectiveness is dictated by the carbon type, source water chemistry, contact time and
pore capacity. An empty bed contact time of no less than 10 minutes is recommended for
adequate microcystin-LR removal in most cases (Alvarez et al., 2010). While GAC can remove
cylindrospermopsin, its effectiveness is less than that for microcystins (Alvarez et al., 2010; U.S.
EPA, 2014a). Australian guidelines recommend using an adsorption step (for example, GAC) or
oxidation as a follow-up to conventional treatment for removing cylindrospermopsin
(Newcombe et al., 2010). The GAC media can become spent within weeks (Sklenar et al., 2014)
to at least six months, depending on water quality and other factors (Alvarez et al., 2010). Like
many other treatments, PWSs may want to conduct tests prior to full-scale use to account for
variations in carbon type and water quality conditions (Sklenar et al., 2014). GAC can also be
used as a substrate for a biological process by allowing bacterial growth on GAC media in rapid
gravity filters to degrade cyanotoxins.
Biological Filtration. Biologically active slow sand filtration, river bank filtration and GAC
have been reported to effectively remove extracellular microcystins and taste and odor
compounds due to very long contact times and high biological activities in these processes (Ho et
al., 2006b, 2012), but full-scale studies are limited (Grützmacher et al., 2002; Rapala et al.,
2006). These processes would also be effective for cylindrospermopsin removal (Ho et al.,
2012). Cyanobacteria and toxins are biodegraded in these processes to varying degrees
depending on conditions, such as type and concentration of microcystin degrading bacteria,
concentration of microcystins, natural organic matter in source water and temperature (Ho et al.,
2012).
Slow sand filtration can remove cyanobacterial cells, but also has the ability to biodegrade some
toxins in the schmutzdecke on the surface of the filtration media. However, waters containing
cyanobacteria can lead to rapid blocking of the filtration media (WHO, 1999), which can result
in the retention of a large proportion of cells (Cyanocenter UBA, 2015) and blocking of the filter.
The ability of slow sand filters to degrade some cyanotoxins may vary with the season; achieving
a 95 percent removal of extracellular microcystins and cylindrospermopsin in the summer but
less than 65 percent in the autumn (Westrick et al., 2010). River bank filtration may also be
effective for the removal of microcystins (Lahti et al., 1998; Schijven et al., 2002; Grützmacher
et al., 2002).
Membrane Filtration. NF and RO are tighter membranes than MF and UF and can remove a
high percentage of extracellular cyanotoxins. Tokodi et al. (2012) found that NF can completely
remove cyanobacterial cells and their associated extracellular toxins. Removal of extracellular
cyanotoxins by NF and RO is important because cell lysis is highly likely during the process
(Connecticut Department of Public Health, 2015). Westrick et al. (2010) reported a range of
microcystins removal from 82 percent to complete removal by NF and RO. Dixon et al. (2010,
58
2011), found that cylindrospermopsin was removed at a 90-100 percent efficiency using NF and
RO. The exact removal efficiencies by NF depend on the membrane material for NF (Westrick et
al., 2010), and on the membrane pore size and water quality for NF and RO (Gijsbertsen-
Abrahamse et al., 2006). Systems should test cyanotoxin removal through individual membrane
pilot tests (Sklenar et al., 2014).
Ultraviolet with Hydrogen Peroxide. Ultraviolet (UV) treatment alone requires impractically
high UV doses for effective cyanotoxin oxidation (U.S. EPA, 2014c; Westrick et al., 2010;
WHO, 1999). For example, only 58 percent microcystin oxidation was achieved at 300 mJ/cm2
using UV (Alvarez et al., 2010). While UV alone is ineffective in cyanotoxin removal, requiring
dosages that are orders of magnitude higher (as high as 20,000 mJ/cm2) than needed for
disinfection (Westrick et al., 2010), UV treatment at 100 mJ/cm2 with hydrogen peroxide
addition at 2 mg/L as part of an advanced oxidation process showed 50 percent microcystin-LR
removal in one study (Alvarez et al., 2010). Alvarez et al. (2010) studied a range of UV doses
from 33–1,000 mJ/cm2 and hydrogen peroxide doses of 1-4 mg/L, and reported that the
effectiveness of UV with hydrogen peroxide was dictated by the hydrogen peroxide
concentrations and availability. UV and hydrogen peroxide can achieve disinfection, oxidation
and photolysis. The needed dose should be determined by bench-scale studies (Alvarez et al.,
2010).
59
APPENDIX I:
CDC DOCUMENTS
APPENDIX J:
ORCA STUDY SUMMARY
Tracking Cyanotoxins in the Aquatic Food Web in Martin County
In 2015, researchers identified a correlation between cyanobacterial blooms and high levels of non-
alcoholic liver disease in a cluster of Florida counties including those along the Treasure Coast and
Okeechobee County. The most common cyanotoxin that results from the type of cyanobacteria
bloom seen in these counties is microcystin. Human exposure to microcystin (via ingestion of
contaminated water, inhalation, recreational exposure, and consumption of contaminated foods)
can result in both short and long-term human health impacts. Short term symptoms include
abdominal cramps, nausea, vomiting, diarrhea, fever and sore throat, while chronic exposure to
microcystins could result in liver disease.
In 2018, ORCA began tracking cyanotoxins from the waters in Martin County, Florida into the local
aquatic food web. This exploratory research began with a convenience survey (i.e. testing all fish
caught at any site) of 54 fish caught throughout Martin County. Analysis of these fish found that
27.8% of fillets and 69.8% of livers contained microcystin concentrations above ORCA’s established
detection limits. The average microcystin level in fillets was 7.4 ng/g with a range of 0.8-39 ng/g, and
the average microcystin level of liver samples was 17.2 ng/g with a range of 0.6-149 ng/g. Following
the positive identification of microcystin in fish in Martin County, ORCA began a study designed to
estimate the degree of exposure of subsistence fishers to microcystin. Focusing on fishing locations
surrounding Indiantown, ORCA researchers have interviewed 27 subsistence fishers and tested 22
fish from the Port Mayaca Locks to date. Initial findings from interviews revealed that subsistence
fishers eat 3 to 4 times more fish than the average US citizen and depend on fishing for up to seven
meals each week. Most fish caught at the Port Mayaca locks had detectable microcystin in the fillets
(63.6%) and livers (54.5%) with average levels of 3.1 ng/g and 13.6 ng/g, respectively. These fishers also
face additional cyanotoxin exposure while they fish through skin contact with contaminated water
and inhalation of aerosolized microcystin.
Fishing in Martin County is a crucial food source as well as an important social and cultural activity.
These studies show that fishers are being exposed to cyanotoxins through fish consumption. Due to
the dependence on fishing for food and recreation, the solution cannot be to stop fishing. Instead,
ORCA’s data should be used to support the need to improve water quality in Martin County.
*Zhang, F., Lee, J., Liang, S., & Shum, C. (2015). Cyanobacteria blooms and non-alcoholic liver
disease: evidence from a county level ecological study in the United States.
Environmental Health, 1-11.
NOTE: ORCA’s subsistence fishing study will be completed, with a paper submitted for publication, in
the summer of 2019.
FUNDING: ORCA received funding for the survey of fish in Martin County from the Guardians of
Martin County, and for the subsistence fishing study from the Frances Langford Fund through the
Community Foundation Martin – St. Lucie.
APPENDIX K:
ENVIRONMENTAL HEALTH STUDY
Zhang et al. Environmental Health (2015) 14:41
DOI 10.1186/s12940-015-0026-7
RESEARCH Open Access
Cyanobacteria blooms and non-alcoholic liver

disease: evidence from a county level ecological
study in the United States
Feng Zhang1, Jiyoung Lee1,2,3*, Song Liang4,5 and CK Shum6,7
Abstract
Background: Harmful cyanobacterial blooms present a global threat to human health. There is evidence suggesting
that cyanobacterial toxins can cause liver damage and cancer. However, because there is little epidemiologic research
on the effects of these toxins in humans, the excess risk of liver disease remains uncertain. The purpose of this study is
to estimate the spatial distribution of cyanobacterial blooms in the United States and to conduct a Bayesian statistical
analysis to test the hypothesis that contamination from cyanobacterial blooms is a potential risk factor for non-alcoholic
liver disease.
Methods: An ecological study design was employed, in which county-specific gender and age standardized mortality
rates (SMR) of non-alcoholic liver disease in the United States were computed between 1999 and 2010. Bloom coverage
maps were produced based on estimated phycocyanin levels from MERIS (Medium Resolution Imaging Spectrometer)
water color imageries from 08/01/2005 to 09/30/2005. A scan statistical tool was used to identify significant clusters of
death from non-alcoholic liver disease. A map of local indicator of spatial association (LISA) clusters and a Bayesian
spatial regression model were used to analyze the relationship between cyanobacterial bloom coverage and death
from non-alcoholic liver disease.
Results: Cyanobacterial blooms were found to be widely spread in the United States, including coastal areas;
62% of the counties (1949 out of 3109) showed signs of cyanobacterial blooms measured with MERIS.
Significant clusters of deaths attributable to non-alcoholic liver disease were identified in the coastal areas
impacted by cyanobacterial blooms. Bayesian regression analysis showed that bloom coverage was significantly
related to the risk of non-alcoholic liver disease death. The risk from non-alcoholic liver disease increased by
0.3% (95% CI, 0.1% to 0.5%) with each 1% increase in bloom coverage in the affected county after adjusting for
age, gender, educational level, and race.
Conclusions: At the population level, there is a statistically significant association between cyanobacterial
blooms and non-alcoholic liver disease in the contiguous United States. Remote sensing-based water monitoring
provides a useful tool for assessing health hazards, but additional studies are needed to establish a specific association
between cyanobacterial blooms and liver disease.
* Correspondence: lee.3598@osu.edu
1
Environmental Science Graduate Program, The Ohio State University,
Columbus, OH, USA
2
College of Public Health, Division of Environmental Health Sciences, The
Ohio State University, Columbus, OH, USA
Full list of author information is available at the end of the article
© 2015 Zhang et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain
Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,
unless otherwise stated.
Zhang et al. Environmental Health (2015) 14:41 Page 2 of 11
Background Aerosol samples, taken during recreational activities on

Liver disease constitutes a rapidly increasing global bur- bloom impacted lakes, have been found with detectable
den to society and is an important cause of morbidity levels of microcystins [17,18]. Although the levels of
and mortality in the United States, accounting for up to aerosolized toxin were generally low, laboratory investiga-
2% of all deaths in the US [1]. Economically, approxi- tions have found that treatment of mice by the intranasal
mately 1% of the total national health care expenditure route to microcystin-LR, the most toxic known variant of
is spent on the care of patients with liver disease, which microcystin, was an effective method for toxin exposure
appears to be on the rise in the US [1]. In the United [19]. In addition, there is evidence that liver disease has
Kingdom, liver disease is the fifth most common cause been associated with the consumption of seafood, such as
of death with increasing mortality rates [2]. Major risk fish, and water contaminated with microcystin [20]. The
factors include hepatitis C and B viruses, heavy alcohol relationship between cyanobacterial toxin and liver cancer
consumption, and non-alcoholic liver disease [3]. Non- has been presumed from several epidemiological studies
alcoholic liver disease refers to a collection of liver diseases in developing countries [21,22]. Deaths in Brazil have been
in people who drink little or no alcohol. Non-alcoholic attributed to exposure to cyanobacterial hepatotoxins
liver disease is developing into a worldwide major health (microcystins) via hemodialysis water [11] and chronic ex-
problem [2] with age, gender, and obesity as potential risk posure to microcystins has been identified as a risk factor
factors [4]. Accumulating evidence suggests that non- for childhood liver damage in China [23]. However, these
alcoholic liver disease is rapidly becoming another im- and other epidemiological studies do not conclusively
portant cause of hepatocellular carcinoma [5]. prove the etiological effect of cyantoxins, therefore, more
Cyanobacterial blooms have been reported to be a severe studies are warranted to fully understand the health
problem in many water bodies and coastal areas around impact of these toxins.
the world. Recent research suggests that eutrophication, In recent decades, the incidence and intensity of toxic
coupled with climate change, promotes the worldwide cyanobacterial blooms, as well as the associated eco-
proliferation and expansion of cyanobacterial harmful nomic impact, have increased in the United States and
algal blooms [6]. These blooms can affect water quality, worldwide [6,24]. Even though cyanobacterial blooms
producing a variety of toxins, such as microcystins, have become a serious problem for water resources in
nodularin, and anatoxin [7]. It has also been shown that the United States, no federal regulatory guidelines for
the neurotoxic amino acid, beta-methylamino alanine cyanobacteria or their toxins in drinking or recreational wa-
(BMAA), is widely produced by cyanobacteria [8]. Hu- ters exist at this time. Some states such as Iowa, Minnesota,
man exposure to these toxins occurs through ingestion, Nebraska, Wisconsin, California, Oregon and Ohio have
skin contact, and inhalation [9]. Despite the potential established monitoring programs and routinely issue alerts
health risks of cyanobacterial toxins, shown by animal for harmful cyanobacterial blooms [25]. Many states and
studies, a limited number of epidemiological studies other jurisdictions rely on WHO guidelines to manage
have been reported in humans. Microcystins usually cyanobacterial blooms and toxins, whereas other states
accumulate in vertebrate liver cells and are suggested have developted their own guidelines to support public
to cause liver damage [10], are the most common and health decision-making, such as posting advisories or
more thoroughly studied of the cyanobacterial toxins, closing access to contaminated water bodies [26]. Add-
and have been identified as being hepatotoxins [11]. itional legislation is recommended to promote research and
Microcystins are resistant to digestion in the gastro- establish guidelines regarding cyanobacterial blooms [24].
intestinal tract and are concentrated in the liver by an Optical/Infrared remote sensing has been used to
active transport system [12]. Acute poisoning results in monitor algal blooms, mostly by quantifying the concen-
destruction of the liver architecture, leading to blood tration of pigments in water bodies, such as chlorophyll-a
loss in the liver and hemorrhagic shock [13]. Chronic (chl-a) or phycocyanin [27]. However, as chl-a is common
exposure to these toxins causes an ongoing active liver to almost all phytoplankton, its retrieval from remotely
injury in mice [14] and there is experimental evidence sensed data cannot be used to specifically determine the
suggesting that microcystins can cause tumor promotion abundance of cyanobacteria, especially where other groups
[15]. Human exposure to microcystins occurs through: 1) of eukaryotic algae co-occur. In contrast to chl-a, phyco-
ingestion of microcystins from tap water due to cyanon- cyanin is a pigment only found at high concentrations in
bacterial blooms in the source water; 2) recreational cyanobacterial blooms; therefore phycocyanin has been
exposure through accidental ingestion or inhalation shown to be a better indicator of cyanobacterial blooms
and dermal contact; and 3) consumption of seafood [28] and has been proposed as a tool for inferring elevated
with accumulated microcystin. Algal cells and water- microcystin levels [29]. Phycocyanin, other than chl-a and
borne toxins can be aerosolized by a bubble-bursting carotenoid, is the most measureable pigment-protein
process via wind-driven, white-capped waves [16]. complex in Microcystis spp [30]. The Medium Resolution
Imaging Spectrometer (MERIS) onboard the European used to identify the linkage between cyaonobacterial
Space Agency ENVIronmental SATellite (ENVISAT) is blooms and non-alcoholic liver disease. A Bayesian
suitable for retrieving data on phycocyanin concentrations regressional analysis was then used to quantitatively
because one of its 15 VIS_NIR (Visible-Near Infrared) measure the linkage between cyanobacterial bloom and
programmable spectral bands between 390 nm and non-alcoholic liver disease mortality.
1040 nm, can be used to detect the phycocyanin absorption
peak that is near 620 nm. ENVISAT MERIS on a push- MERIS-observed bloom coverage data
broom detector generates an observation swath width of All MERIS L1B full resolution images covering the con-
1,150 km, and with an exact repeat-orbit of 35-days; the ef- tiguous United States from 08/01/2005 to 9/30/2005
fective temporal sampling or Earth revisit time is 2–3 days were retrieved from the National Aeronautics and
with a spatial resolution of ~300 m. Different types of Space Administration (NASA)’s Goddard Space Flight
models, such as semi-empirical and single reflectance ratio, Center (GSFC) Ocean Color Science Team that are
have been developed to quantify phycocyanin levels [31,32]. available from http://oceancolor.gsfc.nasa.gov/. August
A nested semi-empirical band ratio model [33], based on and September were chosen to match the common sea-
MERIS, has been proven to statistically outperform other sonal peak (late summer or early fall) of cyanobacterial
MERIS-based algorithms [30]. A previous study, based on blooms (http://www.cdc.gov/nceh/hsb/hab/default.htm).
Lake Erie beaches, also suggested that the semi-empirical After downloading the data, we used the Basic ERS &
band ratio model performed well even with relatively low Envisat (A) ATSR and MERIS (BEAM) VISAT toolbox
phycocyanin levels [34]. Although MERIS ceased its oper- provided by ESA and Brockmann Consult and its supple-
ation on 9 May 2012, due to a sudden failure in communi- mentary Regional Case-2 Water Processor [36] to further
cation in the ENVISAT satellite, it is still useful in terms of refine the data from the US. In particular, we applied the
retrieving historical bloom conditions over Lake Erie back Case-2 Regional Processor (C2R) v1.5.2 to convert the top
to 2002 when it was launched. The advantages of using sat- of atmosphere (TOA) radiance (archived in the original
ellite images for water quality parameters include: a) near L1B data) to water leaving radiance (R_LW) above the sur-
continuous spatial coverage of satellite imagery allowing for face. Phycocyanin levels were estimated using the nested
estimates over large areas, and b) a record of archived semi-empirical band ratio model [33], which has proven
imagery giving an estimation of historical bloom conditions. to be quite reliable [37]. The nested semi-empirical band
In other studies, the linkage between some satellite mea- ratio model used MERIS bands 6, 7, 9, and 12. The ab-
sured environmental factors and health risks showed a sorption of phycocyanin, was calculated as below:
potential for satellite imagery use [35]. However, satellite
imagery data have a limitation for near lake coastal regions Bð705Þ
apc ð620Þ ¼ ðaw ð709Þ þ bb Þ −bb −aw ð620Þ
because of land contamination, and the spatial (pixel) reso- Bð620Þ
lution of about 300 m. δ −1 −ðε achl ð665ÞÞ
The current study adopted an ecological method using
In this model, the absobtion of chlorophyll a is cal-
aggregate disease mortality data at the county level for
culated as:
the contiguous US and MEIRS-derived data for cyano-

bacterial bloom coverage. Exploratory spatial analysis Bð705Þ
methods and regression models were used to test the hy- achl ð665Þ ¼ ðaw ð709Þ þ bb Þ
Bð665Þ
pothesis that non-alcoholic liver disease mortality rates
are related to satellite-observed algal bloom coverage. −bb −aw ð665Þ γ −1
The identification of cyanobacterial blooms as potential
risk factors for non-alcoholic liver disease will help to Where B(X) = water-leaving reflectance centered at X
address the prevention of this disease worldwide, includ- nm [unit: dimensionless];
ing the US, and assist in drawing attention to mitigating aw(665) = pure water absorption at 665 nm;
cyanobacterial blooms throughout the world. γ = 0.68, estimated chlorophyll-a absorption;
aw(709) = pure water absorption at 709 nm [unit: m−1];
Methods bb = backscattering coefficient estimated by a single
First, coverage maps of cyanobacterial blooms in the US band (Band 12 of MERIS in this study);
were produced using estimated phycocyanin levels from aw(620) = pure water absorption at 620 nm [unit: m−1];
MERIS images. Second, standardized county level non- δ = 0.82, correction factor; and
alcoholic liver disease mortality rates were computed ɛ= 0.24
using the mortality data from Multiple Cause of Death Maximum value composite technique was used to com-
data. Subsequently, exploratory methods including spatial bine all images into one large image with each pixel being
clustering and local indicator of spatial association were the highest value for that pixel location. This generates a
cloud-free phycocyanin image for the spatial monitoring of Death data contained in the Centers for Disease
of cyanobacterial blooms over water bodies in the con- Control and Prevention (CDC) Wide-ranging Online
tiguous US. Data for Epidemiologic Research (WONDER) online
Cyanobacteria blooms can present serious risks to human database (http://wonder.cdc.gov/mcd.html). Non-alcoholic
and animal health due to their ability to produce toxins. liver disease mortality data and population-at-risk were
The World Health Organization (WHO) has provided retrieved by county, gender, and age (10 year intervals).
guideline levels of 1 and 20 μg/L of microcystin in drinking Aggregated, non-alcoholic liver disease mortality counts
and recreational water, respectively, in order to protect and population-at-risk were also retrieved by gender and
public health [38]. The WHO guideline for recreational age groups for the US to be used as the standard popula-
exposure to cyanobacteria uses a three-tier approach tion in calculating non-alcoholic liver disease mortality
based on cyanobacterial density and chl-a level [39]. rates, adjusting for effects of gender and age. Gender
For protection of health, due to the irritative or aller- and age adjusted rates were calculated using indirect
genic effects of cyanobacterial toxins, a guideline level standardization for each county. Rate adjustment removes
of 20,000 cyanobacterial cells/ml (corresponding to the effects of gender and age from crude rates in order to
10 mg chl-a/liter under conditions of cyanobacterial allow meaningful comparisons across populations with
dominance) has been derived [39]. We chose the level different underlying race and age structures. Population
of 20,000 cyanobacterial cells/ml to be the threshold of data from the US, during the study period, were used as
significant blooms. To transform this threshold to a standard populations to obtain a standardized rate for
phycocyanin level, we used a linear relationship between each county. Some other potential confounders, such as
the log transformed parameters suggested in Ahn et al. [40] educational level and race, were also adjusted by putting
and derived a level of 4.11 μg/L as equivalent to 20,000 the two factors as covariates in the regression model. The
cells/ml of cyanobacteria. The relationship between cyano- percentage of people over 25 with a college degree or
bacterial cell abundance and phycocyanin is given as: above was used as an indicator of educational level and
the percentage of black people was used to adjust for race.
LogðcyanobacteriaÞ ¼ 0:360 logðphycocyannÞ þ 4:08 The percentage of people over 25 with a college degree
was from the US Census Bureau, 2006–2010 American
As a precaution, 4 μg/L phycocyanin was used as our Community Survey and the percentage of black people
actual threshold for identifying water bodies affected by was retrieved from the Multiple Cause of Death data in
cyanobacterial blooms that could have potential ad- the CDC WONDER online database.
verse health effects on their neighborhoods. By over- With the indirect standardization, the expected number
laying the county boundary polygon GIS layer from US of non-alcoholic liver disease deaths was first calculated
Census (http://www.census.gov/geo/maps-data/data/tiger- for each county, which was determined by the number
cart-boundary.html), with the MERIS bloom coverage im- of cases that would be expected if people in the study
agery with a spatial resolution of 260 m × 290 m, (http:// population had the same mortality rate as people in the
badc.nerc.ac.uk/data/meris/index-old.html), we calculated standard population with the same age and gender.
the bloom coverage as the percentage of county area Standardized mortality rates (SMRs) were calculated by
covered by cyanobacterial blooms. The maximum value dividing the observed count by the expected value.
composite and the zonal statistics were performed using Death counts were “suppressed” when the data met the
ArcGIS 10.0. Although the bloom coverage data are only criteria for confidentiality constraints. Rates were sup-
from 2005, they were intended to represent the bloom dis- pressed for sub-national data representing zero to nine
tributions in that decade since the development of eu- (0–9) deaths. Thereafter, all counties with suppressed
trophication and algal blooms is gradual over years [41,42] data were not included for standardized rate calculation,
and studies show that eutrophication conditions in US es- spatial analysis, and modeling. The number of counties in
tuaries remain nearly the same over a decade [43]. How- the study area was 3109, with 195 counties being omitted
ever, in some areas the cyanobacterial bloom situation may that had suppressed disease data. Consequently, the
change rapidly due to eutrophication, mitigation or climate number of data points (counties) for the statistical
factors. Using only 2005 data may underestimate the bloom modeling was 2914. The suppressed data may lead to a
areas for the time period of 1999 to 2010 if the bloom situ- bias of the model and the results needs to be inter-
ation increased exponentially during this time period. preted with caution.
Non-alcoholic liver disease data Testing the hypothesis that non-alcoholic liver disease is
Non-alcoholic liver disease data (ICD-10 codes: R74.0, related to bloom coverage
K71.0 – K77.8) [44] at the county level were extracted A flexible-shaped spatial scan statistic (FlexScan) was
for the period from 1999–2010 from the Multiple Cause performed to identify spatial clusters of non-alcoholic
liver disease deaths. Tango and Takahashi [45] showed number of cases via the formula (μi) = pit/r where r is the
that FlexScan detects irregular shaped clusters by using over-dispersion parameter. We modeled the average num-
a limited exhaustive search that would detect arbitrarily- ber of deaths (μi) as a function of potential risk factors as
shaped clusters by aggregating their nearest circular in the following:
neighboring areas.
Exploratory spatial data analysis and Bayesian regression Log ðμiÞ ¼ log Ei þ a þ Xi β þ ei þ ϕi
models were used to assess the association between cyano-
bacterial blooms and the SMR of non-alcoholic liver dis- where ui denotes expected number of deaths in county
ease using GeoDa software [46,47] and WinBUGS [48]. I; α is the incidence rate when all covariates have zero
For exploratory spatial data analyses, a bivariate global value; Xi is a vector of covariates in county I; β is a vec-
Moran’s I statistic and local indicator of spatial association tor of the regression coefficients; ei is the unobserved
(LISA) were used. Bivariate global Moran’s I value (i.e., uncorrelated) heterogeneity; and øi is the struc-
determines the overall strength and direction of the re- tured spatial random effect. County-specific random ef-
lationship between the two variables, SMR and bloom fects were modeled via a conditional autoregressive
coverage in each county. LISA provides information (CAR) process, which implies that each øi, conditional
relating to the location of spatial clusters and outliers. on its neighbors, follows a normal distribution with a
Local statistics are important because the magnitude of mean equal to the average of neighboring spatial effects,
spatial autocorrelation is not necessarily uniform over and variance is inversely proportional to the number of
the study area. The LISA analysis by GeoDa presented neighbors.
a cluster map and identified clusters of High-High non-
alcoholic liver disease clusters (units of significantly Results
high disease mortality rates surrounded by significantly Spatial distribution of cyanobacterial blooms in the
high bloom coverage), Low-Low clusters (units of signifi- contiguous US
cantly low disease mortality rates surrounded by signifi- Based on the estimated phycocyanin concentrations
cantly low bloom coverage), High-Low or Low-High from MERIS, it was observed that cyanobacterial blooms
outliers and insignificant areas (units where the rela- were widely spread in US water bodies, including lakes
tionship between disease mortality rates and bloom and rivers (Figure 1). From the maximum value compos-
coverage were not significant). Significance was tested ite image, it is evident that a large part of Lake Erie was
by comparison to a reference distribution obtained by covered by cyanobacterial bloom, mostly in the western
random permutations; 999 permutations were used to basin. Other parts of the Great Lakes, such as Saginaw
determine a significance level for the differences be- Bay, also showed significant bloom coverage. The largest
tween spatial units. Spatial contiguity was assessed as lakes in the contiguous US (i.e. Great Salt Lake, Lake of
Queen’s contiguity that defines spatial neighbors as the Woods, Lake Oahe, Lake Okeechobee, and Lake
those areas with shared borders and vertices. Pontchartrain) were all afflicted with cyanobacterial
A negative binomial regression analysis was performed blooms. Coastal areas in Texas, Louisiana, North Carolina,
using STATA 13.0 (Stata Corp., College Station, TX, Virginia, Maryland and Delaware also had significant
USA) to assess the relationship between non-alcoholic cyanobacterial blooms. Based on our satellite estimations,
liver disease deaths and bloom coverage, adjusting for the occurrence of cyanobacterial blooms in US waters was
educational level and race. Negative binomial regression shown to be a common and serious problem. When the
was used instead of Poisson regression because of the data were aggregated at the county level (Figure 2), it was
over-dispersed data. Thereafter, Bayesian negative bino- observed that counties in coastal areas, as well as counties
mial models were fitted in WinBUGS [48] to examine the in the mid-north areas, have substantial bloom coverage;
association between non-alcoholic liver disease deaths and overall, 1,949 counties showed some bloom coverage,
bloom coverage using a conditional autoregressive (CAR) which is 62% of all the counties assessed. Due to the limi-
process. Basically, spatial random effects were used at a tation of the spatial resolution of MERIS images, large
county level to account for spatial correlation present in lakes and estuaries were better represented than relatively
the data. Markov Chain Monte Carlo simulation (MCMC) small ponds. It is possible that the bloom situation was
was applied to estimate model parameters [49]. After the underestimated in areas where most water bodies were
initial burn-in of 5,000 iterations, another 10,000 iterations relatively small ponds.
were used for the summaries of the posterior distribution
of the parameters. It was assumed that the observed Spatial clusters of non-alcoholic liver disease
counts of non-alcoholic liver disease deaths (Yi) in county In total, 773,828 non-alcoholic liver disease deaths in the
i follow a negative binomial distribution with parameters US from 1999 to 2010 were reported; a spatial variation
pi and r; i.e., Yi ~ NB (pi, r), where pi relates to the average in non-alcoholic liver disease mortality was observed
(A) Southern part of the US (B) Midwestern part of the US
(C) Western part of the US (D) Northeastern part of the US
Figure 1 The spatial distribution of cyanobacteria blooms in different parts of the Contiguous US in 2005 as estimated by MERIS. (A) Southern
part of the US (B) Midwestern part of the US (C) Western part of the US (D) Northeastern part of the US.
(Figure 3). FlexScan identified 65 significant spatial clus- autocorrelation. The High-High and Low-Low counties
ters of non-alcoholic liver disease (p < 0.01), which represent spatial clusters, while the High-Low and Low-
included 432 counties. There were 26 significant clusters High counties represent spatial outliers. The legend also
along the coastal areas versus 39 significant clusters in shows the number of counties in each category. The clus-
the inland areas. The most likely clusters were located ters were observed in those places with significant positive
along the coastal area of Texas and included 14 counties spatial relationships between the two variables, while the
(p = 0.001). Counties in the clusters also showed higher outliers showed significant negative spatial relationships.
bloom coverage than counties from the non-clusters High-High areas shown in Figure 4 tend to have high
according to the Wilcoxon signed-rank test (p < 0.001). bloom coverage that is shown in Figure 2 as well as high
SMR of nonalcoholic liver disease SMR shown in Figure 3.
Exploratory spatial analysis on the relationship between In contrast, High-Low areas shown in Figure 4 tend to
non-alcoholic liver disease and bloom coverage have high bloom coverage shown in Figure 2, but low
The global Moran’s I value is 0.001 (p = 0.001), which nonalcoholic liver disease SMR as shown in Figure 3.
indicates an overall positive spatial correlation of non- There were significant clusters around coastal areas near
alcoholic liver disease SMR and bloom coverage. The Texas.
bivariate LISA cluster map is shown in Figure 4 (per-
mutations = 999, p <0.05), which shows local patterns Bayesian regression of non-alcoholic liver disease on
of spatial correlation at the county level between SMR and bloom coverage
average bloom coverage for its neighbors. Significant clus- Bayesian regression revealed a significant relationship
ters, as well as outliers, are color coded by type of spatial between non-alcoholic liver diseases and bloom coverage
Figure 2 Bloom coverage area (percentage by county) in the US in 2005 as estimated by MERIS.
using a negative binomial model (Table 1), accounting alcoholic liver disease death rate of 468 per 1,000,000
for the spatial correlated parttern of the data. According people per year. The results show that bloom coverage
to this model, risk for non-alcoholic liver disease death was a significant factor influencing the rate of non-
increased 0.3% (95% Bayesian confidence interval, 0.1% alcoholic liver diseases.
to 0.5%) for each 1% increase in bloom coverage of the
county; adjusting for age, gender, educational level, and Discussion
race. In the US, if the bloom coverage per county in- We estimated the overall spatial distribution of cyanobac-
creases by 1%, the estimated number of deaths per year terial blooms in the contiguous US using MERIS-based
will increase by about 440, given the current non- phycocyanin levels. Harmful cyanobacterial blooms may
Figure 3 SMR of nonalcoholic liver disease of each county. The FlexScan identified significant clusters of death counts due to nonalcoholic liver
disease from 1999 to 2010.
Figure 4 Bivariate LISA cluster map of nonalcoholic liver disease and cyanobacterial bloom coverage. The numbers of counties by the categories
are shown in the parentheses.
be more common than previously estimated [50], as most rates for non-alcoholic liver disease; counties in the spatial
large lakes in the US and coastal areas showed cyanobac- clusters also showed higher cyanobacterial bloom coverage
terial blooms indicating a serious environmental problem than counties in the non-clusters, indicating that environ-
in wide areas. In the contiguous US, 1,949 counties mental risk is associated with cyanobacterial bloom and
showed at least some blooms in their water bodies (due to could be contributing to the spatial clusters of non-
the limitation of spatial resolution, small lakes could not alcoholic liver disease. By Bayesian spatial regression, we
be assessed). For monitoring of algal blooms and their found a significant positive association between the rela-
toxins, it appears that remote sensing is a useful, quick, tive risk of non-alcoholic liver disease mortality and
and cheap method for evaluation of large areas and can cyanobacterial bloom coverage after adjusting for gender,
serve as a supplement to in situ monitoring of water bodies age, race, and educational level; there was an excess risk of
with more extensive coverage. Currently, a number of states non-alcoholic liver disease mortality in those areas with
regularly monitor cyanobacteria and cyanotoxins in water high bloom coverage. The results show that spatial distri-
bodies (e.g. New York State implemented the “Citizens bution of cyanobacterial blooms, estimated by remote
Statewide Lake Assessment Program” to monitor lake con- sensing, was associated with non-alcoholic liver disease
ditions, including harmful algal blooms and Ohio regularly mortality, strongly suggesting that cyanobacterial blooms
monitors microcystin levels at recreational beaches) (http:// are an important risk factor.
epa.ohio.gov/ddagw/HAB.aspx). In the US, the distribution Monitoring of phycocyanin data, obtained from remote
of mortality from non-alcoholic liver disease seems to vary sensing, can be used as an indicator of cyanobacterial
geographically, which could be the result of possible blooms over a large area and can aid in assessing health
environmental risk factors. In the contiguous US, we risks due to theseblooms; cyanobacterial blooms may pro-
have identified 65 spatial clusters with high mortality duce toxins, such as microcystins that have been shown to
be liver toxins [23]. Living in bloom areas increases the
Table 1 Model estimates of the ajusted association probability that persons will be exposed to excess micro-
between cyanobacterial bloom coverage and nonalcholic cystin through inhalation, recreational exposure, or inges-
liver disease mortality in the Coutigious US by Baysian tion of contaminated food or water. Several cyanotoxins,
negative binomial regression such as microcystins, Nodularins and Cylindrospermopsin
Variable Risk Ratio 95% credible have shown to cause liver damage [51] and microcystins
Estimate interval
can cause liver hemorrhage and chronic effects, such as
County bloom coverage 1.003 (1.001,1.005) tumor promotion [51]. Microcystins have over 80 variants
Percentage of adults with higher 0.385 (0.350, 0.422) and could be produced by Microcystis, Anabaena, Nostoc,
degree
Oscillatoria, and Hapalosphon [9]. Nodularins are more
Percentage of black population 1.60 (1.46,1.78) commonly isolated from the filamentous, planktonic
percentage
cyanobacterium, Nodularia spumigena are structurally
similar to microcystins and can induce similar toxic effects losses due to cyanobacterial blooms on health could be
[9,51]. Cylindrospermopsin causes liver hemorrhage [52] reliably estimated.
and can be produced by Aphanizomenon, Cylindrosper- There are some limitations related to this study: 1)
mopsin, Umezakia [9]. Cyanobacterial blooms have also while effects were adjusted for gender, age, race, and
been implicated as a potential risk factor for amyotrophic educational level, other potential confounding factors
lateral sclerosis [53]. were not included (e.g. obesity, smoking, Hepatitis B
This current study shows a possible association be- infection, and diet); 2) the study used aggregated data
tween non-alcoholic liver disease and spatial distribution and therefore inferences based on the analysis cannot be
of cyanobacterial blooms. Although such a study is gen- directly transferred to an individual level, ecological
erally suitable to show an association, it is not suitable studies, as presented here, i do not have the ability to
to prove or disprove an etiological cause for disease. distinctively incorporate individual information, as satel-
However, it can be used for hypothesis generation and lite measurements do not represent individual exposure
testing. Other studies are needed to investigate the level due to differences in diet, recreational activities, etc.; 3)
of exposure through different routes that are sufficient ENVISAT MERIS imageries have a spatial resolution of
to cause disease. approximately 300 m, which limits the ability to assess
The significant association shown herein, between small lakes and ponds, as well as data outages near
cyanobacterial blooms and non-alcoholic liver disease coastal regions, which could lead to some biases in the
provides some evidence for a potential health risk, but estimation of exposure levels; 4) the spatial resolution
more epidemiological research is warranted in order to limitation may lead to an underestimation of the bloom
more accurately assess this risk. Exposures and possible coverages in inland areas with more small ponds; 5) the
health effects (both acute and chronic) of cyanobacteria population-based ecological study does not consider
and their toxins need to be evaluated more extensively population dynamics during the study period (e.g. people
than under current conditions [54], especially now that may have migrated during the study period) and the
global warming will be more favorable for cyanobacterial residence at time of death may not be the location where
bloom-forming events. In developed countries, where the disease was initiated. According to the US 2000 cen-
people are collecting water from surface sources to sus data, between March 1999 and March 2000, 43.4
drink, more actions should be taken to control bloom million Americans moved and 39% of all moves were
formation as microcystins are highly stable in water and cross county [57]. We might underestimate the effect of
resistant to boiling. bloom on nonalcoholic liver disease if impacted people
This study highlights several important points for moved away from the bloom areas; and 6) we used one
consideration. First, large scale ecological studies, suchas year’s cyanobacterial bloom situation to represent a
that presented herein, are particularly useful under 12 year bloom situation. It is possible that the bloom
conditions where disease data at individual levels are situation could change quickly in a few years in some lo-
not available and individual levels of exposure are diffi- cations; however, it is rare. We could have potentially
cult to obtain [55]. Second, the statistically significant underestimated the bloom situation in areas where eu-
positive association between non-alcoholic liver disease trophication increased very rapidly and miss some areas
mortality rates and cyanobacterial coverages can be with both high bloom coverages and high non-alcoholic
taken as a probable indication of a potential health effect. liver disease mortality. We may have underestimated the
This association justifies the need for further studies to effect of bloom if the missing high bloom areas (remotely
investigate the biological mechanism(s) responsible for sensed) tend to have high nonalcoholic liver disease mor-
the adverse effects of cyanobacterial toxins on human tality rates. Alternatively, we may have overestimated the
health, especially liver damage and liver disease. Third, effect of bloom, if the missing high bloom areas (remotely
the data show that satellites offer tremendous spatial sensed) tend to have low nonalcoholic liver disease mor-
coverage and provide a great resource for regional en- tality rates. Finally, the mismatch of the temporal window
vironmental monitoring, pollution event warnings, and of remote sensing images and disease data could lead to
environmental health studies. Satellite-estimated envir- potential bias in the results. In addition, coastal areas may
onmental factors could be used for studying potential suffer from bloom contamination resulting from adjacent
health risks as demonstrated in this study (association waters, whose effects were not considered in this study
between satellite bloom data and liver disease mortal- and which may lead to an underestimation of the effects
ity). In the United States alone, toxic cyanobacterial of bloom on non-alcoholic liver disease in coastal areas.
blooms result in substantial losses of $2.4-4.6 billion Although most coastal areas close to blooms, also showed
annually in recreational, drinking, and agricultural high bloom percentage, it is possible we may underesti-
water resources [56]. The economic costs of toxic cya- mate the bloom level for the coastal areas by using bloom
nobacteiral blooms would be understandably more, if coverage. Due to these study limitations, the association
between cyanobacterial bloom and non-alcoholic liver dis- 2. Williams R. Global challenges in liver disease. Hepatology. 2006;44(3):521–6.
ease should be interpreted with caution. It is obvious that 3. Ruhl CE, Everhart JE. Determinants of the association of overweight with
elevated serum alanine aminotransferase activity in the United States.
more research is needed to confirm the effects of cyano- Gastroenterology. 2003;124(1):71–9.
bacterial blooms on liver disease as described herein. 4. Preiss D, Sattar N. Non-alcoholic fatty liver disease: an overview of prevalence,
diagnosis, pathogenesis and treatment considerations. Clin Sci.
2008;115(5–6):141–50.
Conclusions 5. Younossi ZM, Stepanova M, Afendy M, Fang Y, Younossi Y, Mir H, et al.
This ecological study in the contiguous US, using satellite Changes in the prevalence of the most common causes of chronic liver
data and data of multiple causes of death, found a signifi- diseases in the United States from 1988 to 2008. Clin Gastroenterol Hepatol.
2011;9(6):524–U109.
cant positive association between risk of non-alcoholic liver 6. O’Neil JM, Davis TW, Burford MA, Gobler CJ. The rise of harmful
disease mortality and cyanobacterial bloom coverage. We cyanobacteria blooms: The potential roles of eutrophication and climate
identified clusters of non-alcoholic liver disease mortality change. Harmful Algae. 2012;14:313–34.
7. Dittmann E, Wiegand C. Cyanobacterial toxins - occurrence, biosynthesis
in clusters in those counties that also had higher bloom and impact on human affairs. Mol Nutr Food Res. 2006;50(1):7–17.
coverage. 8. Holtcamp W. The emerging science of BMAA: do cyanobacteria contribute
The evidence for excess non-alcoholic liver disease in to neurodegenerative disease? Environ Health Perspect. 2012;120(3):A110–6.
9. Cheung MY, Liang S, Lee J. Toxin-producing cyanobacteria in freshwater: A
areas with high cyanobacterial bloom coverage suggests review of the problems, impact on drinking water safety, and efforts for
that more attention should be centered around the public protecting public health. J Microbiol. 2013;51(1):1–10.
health impact of harmful cyanobacterial blooms. Add- 10. Wiegand C, Pflugmacher S. Ecotoxicological effects of selected cyanobacterial
secondary metabolites a short review. Toxicol Appl Pharmacol. 2005;203(3):201–18.
itionally, remote sensing could be used to efficiently
11. Carmichael WW, Azevedo SMFO, An JS, Molica RJR, Jochimsen EM, Lau S,
monitor the distribution of algal blooms over a national et al. Human fatalities from cyanobacteria: Chemical and biological
or global level and serve as a possible early warning evidence for cyanotoxins. Environ Health Perspect. 2001;109(7):663–8.
tool for public health alerts. 12. Runnegar M, Berndt N, Kaplowitz N. Microcystin uptake and inhibition of protein
phosphatases - effects of chemoprotectants and self-inhibition in relation to
Abbreviations known hepatic transporters. Toxicol Appl Pharm. 1995;134(2):264–72.
SMR: Standardized mortality rate; LISA: Local indicator of spatial association 13. Falconer IR, Jackson ARB, Langley J, Runnegar MT. Liver pathology in mice
(LISA); BMAA: Beta-methylamino alanine (BMAA); NASA: National Aeronautics in poisoning by the blue-green-alga microcystis-aeruginosa. Aust J Biol Sci.
and Space Administration; GSFC: Goddard Space Flight Center; BEAM: Basic 1981;34(2):179–87.
ERS & Envisat (A) ATSR and MERIS (BEAM); TOA: Top of atmosphere; 14. Falconer IR, Smith JV, Jackson ARB, Jones A, Runnegar MTC. Oral toxicity of
R_LW: Water leaving radiance; WHO: World Health Organization; a bloom of the cyanobacterium microcystis-aeruginosa administered to
CDC: Centers for Disease Control and Prevention; WONDER: Wide-ranging mice over periods up to 1 year. J Toxicol Env Health. 1988;24(3):291–305.
Online Data for Epidemiologic Research; FlexScan: flexible-shaped spatial 15. Humpage AR, Falconer IR. Microcystin-LR and liver tumor promotion: Effects
scan statistic; CAR: Conditional autoregressive; MCMC: Markov Chain Monte on cytokinesis, ploidy, and apoptosis in cultured hepatocytes. Environ Toxicol.
Carlo simulation; MERIS: The Medium Resolution Imaging Spectrometer; 1999;14(1):61–75.
ENVISAT: Environmental Satellite; ESA: The European Space Agency. 16. Blanchar D, Syzdek LD. Concentration of bacteria in jet drops from bursting
bubbles. J Geophys Res. 1972;77(27):5087–99.
Competing interests 17. Backer LC, Carmichael W, Kirkpatrick B, Williams C, Irvin M, Zhou Y, et al.
The authors declare that they have no competing interests. Recreational exposure to low concentrations of microcystins during an algal
bloom in a small lake. Mar Drugs. 2008;6(2):389–406.
Authors’ contributions 18. Backer LC, McNeel SV, Barber T, Kirkpatrick B, Williams C, Irvin M, et al.
JL and FZ conceived and designed the study; FZ performed the data collection Recreational exposure to microcystins during algal blooms in two California
and analyzed the data; FZ and JL interpreted the results; FZ wrote the paper; lakes. Toxicon. 2010;55(5):909–21.
and JL, SL, and CS edited the paper. All authors read and approved the final 19. Fitzgeorge RB, Clark SA, Keevil CW. Routes of intoxication. In: Codd GA,
manuscript. Jefferies TM, Keevil CW, Potter E, editors. Detection methods for cyanobacterial
toxins. Cambridge, UK: Royal Society of Chemistry; 1994. p. 69–74.
Acknowledgements 20. Pawlik-Skowronska B, Toporowska M, Rechulicz J. Simultaneous accumulation of
The editorial assistance from Dr. Bruce Casto and technical help from anatoxin-a and microcystins in three fish species indigenous to lakes affected by
Dr. Kuo-Hsin Tseng is greatly appreciated. cyanobacterial blooms. Oceanol Hydrobiol Stud. 2012;41(4):53–65.
21. Ueno Y, Nagata S, Tsutsumi T, Hasegawa A, Watanabe MF, Park HD, et al.
Author details Detection of microcystins, a blue-green algal hepatotoxin, in drinking water
1 sampled in Haimen and Fusui, endemic areas of primary liver cancer in
Environmental Science Graduate Program, The Ohio State University,
Columbus, OH, USA. 2College of Public Health, Division of Environmental China, by highly sensitive immunoassay. Carcinogenesis. 1996;17(6):1317–21.
Health Sciences, The Ohio State University, Columbus, OH, USA. 3Department 22. Svircev Z, Krstic S, Miladinov-Mikov M, Baltic V, Vidovic M. Freshwater
of Food Science and Technology, The Ohio State University, Columbus, OH, cyanobacterial blooms and primary liver cancer epidemiological studies
USA. 4Department of Environmental and Global Health, College of Public in Serbia. J Environ Sci Health Part C-Environ Carcino Ecotoxicol Rev.
Health and Health Professions, University of Florida, Gainesville, FL 32610, 2009;27(1):36–55.
USA. 5Emerging Pathogens Institute, University of Florida, Gainesville, FL 23. Li Y, Chen JA, Zhao Q, Pu CW, Qiu ZQ, Zhang RP, et al. A cross-sectional
32610, USA. 6Division of Geodetic Science, School of Earth Sciences, The investigation of chronic exposure to microcystin in relationship to childhood liver
Ohio State University, Columbus, OH, USA. 7Institute of Geodesy & damage in the Three Gorges Reservoir Region. China Environ Health Perspect.
Geophysics, Chinese Academy of Sciences, Wuhan, China. 2011;119(10):1483–8.
24. Hudnell HK. The state of U.S. freshwater harmful algal blooms assessments,
Received: 29 December 2014 Accepted: 22 April 2015 policy and legislation. Toxicon. 2010;55(5):1024–34.
25. Pelaez M, Antoniou MG, He X, Dionysiou DD, de la Cruz AA, Tsimeli K, et al.
Sources and occurrence of cyanotoxins worldwide. In: Fatta-Kassinos D,
References editor. Xenobiotics in the Urban Water Cycle: Mass Flows, Environmental
1. Kim WR, Brown RS, Terrault NA, El-Serag H. Burden of liver disease in the Processes, Mitigation and Treatment Strategies. Dordrecht Heidelberg
United States: Summary of a workshop. Hepatology. 2002;36(1):227–42. London New York: Springer; 2010.
26. Chorus I. Current approaches to cyanotoxin risk assessment, risk 51. Otten TG, Paerl HW. Health effects of toxic cyanobacteria in U.S. drinking
management and regulations in different countries. Federal Environment and recreational waters: our current understanding and proposed direction.
Agency: Deutschland Umweltbundesamt; 2005. Curr Environ Health Rep. 2015;2(1):1–10.
27. Hunter PD, Tyler AN, Gilvear DJ, Willby NJ. Using remote sensing to aid the 52. Runnegar MT, Kong SM, Zhong YZ, Ge JL, Lu SC. The role of glutathione in
assessment of human health risks from blooms of potentially toxic the toxicity of a novel cyanobacterial alkaloid cylindrospermopsin in
cyanobacteria. Environ Sci Technol. 2009;43:2627–33. cultured rat hepatocytes. Biochem Biophys Res Commun. 1994;201(1):235–41.
28. McQuaid N, Zamyadi A, Prévost M, Bird DF, Dorner S. Use of in vivo phycocyanin 53. Torbick N, Hession S, Stommel E, Caller T. Mapping amyotrophic lateral
fluorescence to monitor potential microcystin-producing cyanobacterial sclerosis lake risk factors across northern New England. Int J Health
biovolume in a drinking water source. J Environ Monitor. 2011;13:455–63. Geograph. 2014;13:1.
29. Marion JW, Lee J, Wilkins JR, Lemeshow S, Lee C, Waletzko EJ, et al. In vivo 54. Paerl HW, Paul VJ. Climate change: links to global expansion of harmful
phycocyanin flourometry as a potential rapid screening tool for predicting cyanobacteria. Water Res. 2012;46(5):1349–63.
elevated microcystin concentrations at eutrophic lakes. Environ Sci Technol. 55. Best N. Bayesian ecological modeling. In: Lawson A, Biggeri A, Böhning D,
2012;46(8):4523–31. Lesaffre E, Viel JF, Bertollini R, editors. Disease mapping and risk assessment for
30. Ruiz-Verdu A, Simis SGH, De Hoyos C, Gons HJ, Pena-Martinez R. An evaluation public health. West Sussex, England: John Wiley & Sons Ltd; 1999. p. 193–201.
of algorithms for the remote sensing of cyanobacterial biomass. Remote Sens 56. Dodds WK, Bouska WW, Eitzmann JL, Pilger TJ, Pitts KL, Riley AJ, et al.
Environ. 2008;112(11):3996–4008. Eutrophication of us freshwaters: analysis of potential economic damages.
31. Dekker AG. Detection of optical water quality parameters for entrophic waters Environ Sci Technol. 2009;43(1):12–9.
by high resolution remote sensing. Amsterdam: Free University; 1993. 57. Schachter J. Geographical mobility march 1999 to march 2000. U.S.
32. Schalles JF, Yacobi YZ. Remote detection and seasonal patterns of phycocyanin, Department of Commerce, Economics and Statistics Administration: U.S.
carotenoid, and chlorophyll pigments in eutrophic waters. Arch fur Hydrobiol. Census Bureau; 2001.
2000;55:153–68. Special Issues Advancements in Limnology.
33. Simis SGH, Peters SWM, Gons HJ. Remote sensing of the cyanobacterial pigment
phycocyanin in turbid inland water. Limnol Oceanogr. 2005;50(1):237–45.
34. Lee J, Tseng K-H, Zhang F, Lee C, Marion J, Liang S, et al. From Satellite to
Genes: An Interdisciplinary Approach for Quantifying Microcystin at Lake
Erie Beaches, American Society for Microbiology 114th General Meeting,
Boston, MA. 2014.
35. Hu ZY, Rao KR. Particulate air pollution and chronic ischemic heart disease
in the eastern United States: a county level ecological study using satellite
aerosol data. Environ Health. 2009;8:26.
36. Chawira M. Monitoring blue-green algae in the IJsselmeer using remote
sensing and in-situ measurements. Enschede, The Netherlands: University of
Twente; 2012.
37. Randolph K, Wilson J, Tedesco L, Li L, Pascual DL, Soyeux E. Hyperspectral
remote sensing of cyanobacteria in turbid productive water using optically
active pigments, chlorophyll a and phycocyanin. Remote Sens Environ.
2008;112(11):4009–19.
38. Falconer I, Bartram J, Chorus I, Kuiper-Goodman T, Utkilen H, Burch M, et al.
Safe levels and safety practices. In: Chorus I, Bartram J, editors. Toxic cyanobacteria
in water: A guide to their public health consequences, monitoring and
management. London: E & FN Spon; 1999.
39. Organization WH. Guidelines for safe recreational water environments.
Geneva: World Health Organization; 2003.
40. Ahn CY, Joung SH, Yoon SK, Oh HM. Alternative alert system for cyanobacterial
bloom, using phycocyanin as a level determinant. J Microbiol. 2007;45(2):98–104.
41. Paerl HW, Xu H, McCarthy MJ, Zhu G, Qin B, Li Y, et al. Controlling harmful
cyanobacterial blooms in a hyper-eutrophic lake (Lake Taihu, China): the
need for a dual nutrient (N & P) management strategy. Water Res.
2011;45(5):1973–83.
42. Millie DF, Fahnenstiel GL, Bressie JD, Pigg RJ, Rediske RR, Klarer DM, et al.
Late-summer phytoplankton in western Lake Erie (Laurentian Great Lakes):
bloom distributions, toxicity, and environmental influences. Aquat Ecol.
2009;43(4):915–34.
43. Bricker SB, Longstaf B, Dennison W, Jones A, Boicourt K, Wicks C, et al.
Effects of nutrient enrichment in the nation’s estuaries: A decade of change.
Harmful Algae. 2008;8(1):21–32.
44. Sogaard KK, Horvath-Puho E, Gronbaek H, Jepsen P, Vilstrup H, Sorensen HT.
Risk of venous thromboembolism in patients with liver disease: a nationwide Submit your next manuscript to BioMed Central
population-based case–control study. Am J Gastroenterol. 2009;104(1):96–101. and take full advantage of:
45. Tango T, Takahashi K. A flexibly shaped spatial scan statistic for detecting
clusters. Int J Health Geogr. 2005;4:11.
• Convenient online submission
46. Anselin L. Exploring spatial data with GeoDaTM: a workbook. Urbana-Champaign,
IL, USA: Center for Spatially Integrated Social Science; University of Illinois; 2005. • Thorough peer review
47. Anselin L. Local Indicators of Spatial Association - Lisa. Geogr Anal. 1995;27(2):93–115. • No space constraints or color figure charges
48. Lunn DJ, Thomas A, Best N, Spiegelhalter D. WinBUGS-a Bayesian modelling
framework: concepts, structure, and extensibility. Stat Comput. 2000;10:325–37. • Immediate publication on acceptance
49. Gelfand AE, Smith AFM. Sampling-based approaches to calculating marginal • Inclusion in PubMed, CAS, Scopus and Google Scholar
densities. J Am Stat Assoc. 1990;85(410):398–409. • Research which is freely available for redistribution
50. Fristachi A, Sinclair JL. Occurrence of cyanobacterial harmful algal blooms
workgroup report. In: Hudnell KH, editor. Adv Exp Med Biol. 619th ed. New
York: Springer; 2008. p. 45–103. Submit your manuscript at
www.biomedcentral.com/submit
APPENDIX L:
THE ROYAL SOCIETY PUBLISHING STUDY
Dietary exposure to an environmental
rspb.royalsocietypublishing.org toxin triggers neurofibrillary tangles and
amyloid deposits in the brain
Paul Alan Cox1, David A. Davis2, Deborah C. Mash2, James S. Metcalf1
Research and Sandra Anne Banack1
Cite this article: Cox PA, Davis DA, Mash DC, 1
Institute for Ethnomedicine, Jackson Hole, WY, USA
2
Metcalf JS, Banack SA. 2016 Dietary exposure Department of Neurology, University of Miami, Miller School of Medicine, Miami, FL, USA
to an environmental toxin triggers neurofi-
brillary tangles and amyloid deposits in the Neurofibrillary tangles (NFT) and b-amyloid plaques are the neurological hall-
marks of both Alzheimer’s disease and an unusual paralytic illness
brain. Proc. R. Soc. B 283: 20152397.
suffered by Chamorro villagers on the Pacific island of Guam. Many Chamor-
http://dx.doi.org/10.1098/rspb.2015.2397 ros with the disease suffer dementia, and in some villages one-quarter of the
adults perished from the disease. Like Alzheimer’s, the causal factors of
Guamanian amyotrophic lateral sclerosis/parkinsonism dementia complex
(ALS/PDC) are poorly understood. In replicated experiments, we found that
Received: 6 October 2015 chronic dietary exposure to a cyanobacterial toxin present in the traditional
Accepted: 14 December 2015 Chamorro diet, b-N-methylamino-L-alanine (BMAA), triggers the formation
of both NFT and b-amyloid deposits similar in structure and density to
those found in brain tissues of Chamorros who died with ALS/PDC. Vervets
(Chlorocebus sabaeus) fed for 140 days with BMAA-dosed fruit developed
NFT and sparse b-amyloid deposits in the brain. Co-administration of the diet-
Subject Areas:
ary amino acid L-serine with L-BMAA significantly reduced the density of NFT.
environmental science, health and disease These findings indicate that while chronic exposure to the environmental toxin
and epidemiology, neuroscience BMAA can trigger neurodegeneration in vulnerable individuals, increasing the
amount of L-serine in the diet can reduce the risk.
Keywords:
Alzheimer’s, amyotrophic lateral sclerosis,
L-serine, cyanobacteria, BMAA, tau
1. Introduction
Author for correspondence:
(a) Toxins and neurodegenerative illness
The relationship between environmental toxins and neurological disease has been
Paul Alan Cox of interest since residents of Minamata Bay, Japan, were sickened by chronic
e-mail: paul@ethnomedicine.org dietary exposure to methyl-mercury-laden fish. Parkinson’s disease (PD) has
been linked to rotenone or paraquat exposures in agricultural workers [1].
PD also was diagnosed in ‘frozen addicts’, users of a recreational drug
contaminated with 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine [2]. Exposures
to pesticides, metals, solvents and certain types of volatile anaesthetics have
additionally been linked to PD, while exposures to lead, mercury and pesticides
have been suggested as risk factors for amyotrophic lateral sclerosis (ALS) [1].
However, the role of naturally occurring environmental toxins in progressive
neurodegenerative disease has not been extensively studied.
(b) A paralytic disease among Pacific Islanders

In the 1950s, US Army physicians described a puzzling ALS-like disease among
the indigenous Chamorro villagers of Guam [3]. In the 1960s, amyotrophic
We dedicate this study to the memory of Oliver lateral sclerosis/Parkinsonism dementia complex (ALS/PDC) was described
Sacks, friend and mentor. based on histopathology and clinical symptoms which resemble aspects of
Alzheimer’s disease (AD), ALS and PD. Neurofibrillary tangles (NFT) in the
Electronic supplementary material is available brains of individuals with ALS/PDC have similar immunohistology and struc-
ture as those found in the brains of AD patients but are biochemically and
at http://dx.doi.org/10.1098/rspb.2015.2397 or
via http://rspb.royalsocietypublishing.org.
& 2016 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution
License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original
author and source are credited.
regionally more heterogeneous [4,5]. Many afflicted villagers through inhalation of desert dust has been suggested as trig- 2
suffered from dementia. Histopathology of Chamorros who gering the increased incidence of ALS a decade subsequent to
rspb.royalsocietypublishing.org
died prior to manifesting clinical symptoms of ALS/PDC the deployment of military personnel in Operation Desert
suggests that depositions in the brain pre-dated clinical Storm [26]. Similarly, inhalation of aerosolized BMAA from
onset [4]. No clear pattern of inheritance for the disease has wave break has been proposed to explain the increased risk
been ascertained [6]. Since even outsiders who adopted a of ALS in individuals who live near lakes with persistent
Chamorro lifestyle experienced an increased risk of illness cyanobacterial blooms [18,19]. Exposure through ingestion
[7], a common environmental exposure seemed likely. Deter- of drinking water has not been ruled out [27]. Maternal
mining the nature of the toxin, however, was difficult due to exposures to BMAA may also increase the risk of ALS in
a significant delay between exposure and clinical symptoms, neonates later in their life [20,21].
extending years or even decades [8].
Proc. R. Soc. B 283: 20152397

(f ) Mechanisms of BMAA-induced neurodegeneration
(c) BMAA: a neurotoxic amino acid in cycad seeds Through activation of metabotropic glutamate receptors such
In the 1960s, consumption of flour made from the gameto- as mGluR5 [28] or ionotropic glutamate receptors including
phyte of cycad seeds (Cycas micronesica Hill) was proposed the N-methyl-D-aspartate receptor, kainate or the a-amino-
as a cause of the disease. Interest increased when a novel neu- 3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor, BMAA
rotoxic amino acid, L-BMAA, was isolated from cycad seeds selectively kills subpopulations of NADPH-diaphorase-positive
by Bell [9]. In the 1980s, BMAA fed to macaques was found motor neurons [29]. It is toxic to glial cells [30] and causes motor
to cause acute neurological symptoms [10], a finding that neuron damage and astrogliosis in the ventral horn [31]. BMAA
was discounted when it was argued that an equivalent potentiates different neurotoxic insults including methyl mer-
human dose would require the consumption of unreasonable cury [32], which co-occurs in some fish. BMAA rapidly
amounts of cycad seed flour [11]. BMAA was subsequently crosses the blood –brain barrier, where it is captured by the
identified as a cyanobacterial product [12]. The toxin is bio- central nervous system (CNS) in a time period consistent
magnified in flying foxes, which are eaten by Chamorros with protein misincorporation [33]. BMAA can be mistaken
[13]. Equally important was the discovery that a majority of by cellular machinery for L-serine and be misincorporated
BMAA in cycad seeds binds to proteins and cannot be into proteins, leading to protein misfolding, aggregation
released by washing with water, but only on hydrolysis, and subsequent apoptosis [34]. Misincorporation of BMAA
suggesting that BMAA doses ingested by the Chamorros into proteins has been proposed as a mechanism for bioaccu-
had been previously underestimated [14,15]. Evidence mulation as well as a mechanism for slow release of BMAA
continued to build for the link between BMAA and neuro- within the CNS over years depending on rates of protein
degenerative disease with respect to cyanobacterial exposure turnover [15]. Misincorporation of even the 20 canonical
and epidemiology [15–20]. amino acids at error rates as low as 1/10 000 can lead to neu-
A key missing puzzle piece has been experimental rodegeneration in laboratory animals [35]. BMAA exposure
evidence that chronic dietary exposure to BMAA triggers results in hyperphosphorylated tau, possibly by decreasing
neuropathological changes consistent with ALS/PDC, which activity of protein phosphatase 2A (PP2A) through activation
presumably should occur prior to the onset of clinical symp- of the mGluR5 receptor and subsequent dissociation of the
toms. It is now known that in vivo BMAA exposure generates catalytic subunit PP2Ac [36]. In Chamorro ALS/PDC
fibril formation and cognitive deficits in rodents [21], brains, PP2A activity is significantly decreased, resulting in
although some earlier animal studies that focused on acute a significant increase in hyperphosphorylated tau [36].
rather than chronic exposure were inconclusive [22]. This
finding suggests that chronic BMAA exposure more closely
models early disease. (g) Producing an animal model of BMAA-induced
neuropathology
(d) BMAA in the Chamorro diet The occurrence of BMAA in post-mortem brain tissues of
BMAA is produced by symbiotic cyanobacteria of the genus Chamorro ALS/PDC patients but generally not in non-Cha-
Nostoc harboured in specialized cycad roots emergent in the morro control patients suggests that chronic dietary exposure
leaf litter above the soil. BMAA accumulates in the gameto- to BMAA is an environmental risk factor for ALS/PDC
phytes of cycad seeds, which, after washing, are used by [12,15,37]. To satisfy Koch’s postulates of disease causation
villagers to prepare tortilla flour, dumplings and to thicken [38], it is necessary to show that chronic exposure to BMAA
soups and stews. Animals, including flying foxes, feral deer causes healthy individuals to develop neurodegenerative dis-
and pigs that feed on cycad seeds, which in turn are ease and that BMAA can be re-isolated from those
consumed by villagers, also accumulate BMAA in their tis- individuals in which neurodegeneration has been induced.
sues [13]. BMAA is biomagnified up to 10 000-fold from its NFT and b-amyloid deposits have not both been produced
production by cyanobacteria to its concentration in volant in human neuronal cell culture, so in vivo experiments are
mammals [12,13,15]. necessary. However, no animal species other than humans is
known to develop ALS/PDC or AD. Furthermore, NFT and
b-amyloid deposits have not both previously been produced
(e) BMAA exposures beyond Guam in any single animal model, with the exception of a triple trans-
Diverse taxa of cyanobacteria produce BMAA [23,24], which genic mouse model [39] in which the structure and density
is biomagnified in some marine ecosystems, accumulating in of the NFT significantly differ from the human condition
sharks, bottom-dwelling fish and shellfish. BMAA also (K. Iqbal 2015, personal communication). Some non-human
occurs in cyanobacterial soil crusts [25]. BMAA exposure primates including squirrel monkeys, chimpanzees, gorillas
and orangutans of great age as well as lemurs develop senile spinal fluid (CSF) samples were taken under ketamine anaesthesia 3
plaques that are immunopositive for b-amyloid, and a single to confirm BMAA exposures in the vervets and absence of BMAA
exposure in the controls.
41-year-old chimpanzee was found to produce paired helical
tau filaments [40]. Vervets are known to accumulate vascular
b-amyloid deposits with age, but not NFT and other tau
(b) Neuropathology
inclusions [41]. We therefore decided to chronically expose ver- In both experiments, one hemisphere of each vervet brain was
vets to BMAA for an extended period and to examine their frozen. The other hemisphere was immersion fixed in buffered for-
brain tissues for tau inclusions and amyloid deposition consist- malin for histopathology. This hemisphere was freeze-sectioned at
ent with ALS/PDC pathology. Since matching the duration of 40 mm and an adjacent series of coronal sections were processed
chronic exposure to the years—even decades—required for with antibody stains using the MultiBrain Services of Neuro-
Chamorros to develop ALS/PDC [8] is unfeasible, we shor- Science Associates (Tennessee, USA). In both experiments,
tened chronic exposure to BMAA to 140 days. Since L-serine adjacent sections were stained with AT8 immunohistochemistry
Proc. R. Soc. B 283: 20152397

has been found to prevent misincorporation of BMAA and (IHC) stain with a thionine Nissl counterstain for hyperphos-
apoptosis in human neuronal cell culture [34], we added a phorylated tau and b-amyloid (1– 42) IHC stain for b-amyloid
deposits. In the first experiment, NFT (100 magnification) and
cohort of vervets which daily received equal amounts of
b-amyloid deposits (10 magnification) were identified from
BMAA and L-serine. Finally, to increase statistical rigour, we
blinded review of the stained sections and were quantified using
replicated the first experiment. We used an oral dose manual counts in three sections in series from non-overlapping
(210 mg kg21 d21) that previous investigators found using brain regions. In the second experiment, stained sections
gavage could be tolerated by macaques [10] and in the second were examined using automated images prepared with a Tissue-
experiment added a cohort of vervets with a 10-fold dose Scope LE (Huron Digital Pathology, Ontario, Canada). Stained
reduction (21 mg kg21 d21) to produce a cumulative BMAA serial sections were digitally scanned at 20 using a 350 mm2
exposure closer to total lifetime Chamorro exposure. grid for NFT and b-amyloid deposits. Thioflavine-S with a thio-
nine Nissl counterstain was used to confirm the presence of NFT
and plaques. The regions of interest (ROI) for each case were
initially drawn on the Nissl section and the ROI was mapped to
2. Material and methods the immunostained slides. The ROI was marked with an array
tool to identify regional boundaries of the amygdala, hippo-
(a) In vivo studies campus, entorhinal, frontal, temporal, motor, occipital and
The vervets studied in this report were housed in groups in large cingulate cortices. Digital images were measured using NIH
outdoor enclosures at the Behavioural Science Foundation (BSF) IMAGE J64 software (1.44) converted from RGB colour to 8-bit fol-
in St Kitts, West Indies. The BSF is a fully accredited biomedical lowed by applying a threshold to eliminate non-specific
research facility with approvals from the Canadian Council on background staining. After threshold correction, the images
Animal Care. The animal use protocol was approved by the Insti- were converted to binary allowing for quantification of pathologi-
tutional Animal Care and Use Committee (IACUC) of BSF and cal features detected above background. The high-contrast images
McGill University (Quebec, Canada). The vervet low-protein were highly suited for digital quantification of pixel counts. Repre-
diet was supplemented with fruit dosed with L-BMAA or other sentative sections were examined in parallel to validate the digital
test substances. Doses were prepared at the Institute for Ethno- measurements by comparison to manually derived b-amyloid
medicine, Jackson Hole, using a Mettler Toledo balance with a deposits and NFT counts [42].
Quantos automated powder dispensing module at a tolerance
of +0.1% of target dose. A small cavity was made in each
piece of fruit, and the test substance was placed inside. In the (c) Analytical chemistry
first experiment, 16 juvenile vervets were presented daily with Blinded samples of brain tissue, blood serum and CSF were ana-
a dosed piece of fruit, approximating a 210 mg kg21 d21 dose lysed for BMAA content using triple quadrupole tandem mass
based on average weight (3.1 kg) of the vervets. One cohort of spectrometry (LC-MS/MS) with a precolumn 6-aminoquinolyl-
four was fed daily for 140 days a piece of fruit containing 651 mg N-hydroxysuccinimidyl carbamate (AQC) derivatization using
of L-BMAA, a second cohort was fed fruit with 651 mg of L-serine, the validated method determined by the Association of Analyti-
a third cohort was fed fruit dosed with 651 mg of L-BMAA plus cal Communities AOAC International [43]. Negative controls
651 mg of L-serine, and a fourth control cohort received a piece of included matrix blanks from control vervets with no detectable
fruit dosed with 651 mg of rice flour as a placebo. BMAA, AQC-derivatized blanks, internal standards and solvent
For both the original experiment on the younger vervets and blanks (HCl, TCA). Product ion analysis of BMAA used m/z 459
the replication experiment on the adult vervets, 14 regions of as the precursor ion for collision-induced dissociation (CID) and
each vervet’s brain were investigated for neuropathology with two-step mass filtering was performed during selective reaction
immunostaining for tau AT8 and b-amyloid 1– 42. Three serial sec- monitoring of BMAA after CID in the second quadrupole, moni-
tions were studied on a 3 4 grid with 100 magnification. In the toring the following transitions: m/z 459 to 119 CE 25 eV; 459 to
replication experiment, cohorts of eight adult vervets were fed 289 CE 23 eV; 459 to 171 CE 45 eV. The resultant product ions
dosed fruit for 140 days. These 7-year-old vervets, which were were detected after passing the third quadrupole and their
colony-born, were somewhat larger than the younger vervets in relative abundances were quantified. BMAA was analytically
the first experiment, so to approximate a 210 mg kg21 d21 dose distinguished from its isomers using m/z 459 to 188 CE 38 eV
for these larger animals, the L-BMAA dose was increased from (2,4-diaminobutyric acid); 459 to 214 CE 35 eV (N-(2-ami-
the first experiment in order to adjust for weight. A cohort of ver- noethyl)glycine); 459 to 258 CE 36 eV (BMAA) [44,45]. Double
vets was added in which the L-BMAA dose was reduced 10-fold to ionized AQC-derivatized BMAA was also monitored with a pre-
approximate 21 mg kg21 d21 to be closer to a lifetime Chamorro cursor ion of m/z 230 and a product ion of 171 CE 27 eV.
exposure. Thus in the replication experiment one cohort of eight Additionally, the following amino acids were monitored:
vervets received daily 987 mg of L-BMAA, a second cohort single derivatized lysine m/z 317, double derivatized lysine m/z
received 98.7 mg of L-BMAA, a third cohort received 987 mg of 487, leucine m/z 302, serine m/z 276 and an internal standard
L-BMAA and 987 mg of L-serine, and a fourth control cohort (b-N-methyl-d3-amino-DL-alanine-15N2) with a precursor ion
received 987 mg of rice flour. Periodic blood serum and cerebral of m/z 464, and product ions m/z 124 CE 25 eV, 171 CE 45 eV,
259 CE 36 eV and 294 CE 23 eV. BMAA tissue concentrations 4
(a) (b)
were determined relative to concentration curves run daily in
spiked matrix samples from a control animal using b-N-
methyl-d3-amino-DL-alanine-15N2 as an internal standard.
Sample preparation techniques and complete analytical methods
appear in the electronic supplementary material.
(d) Statistical analysis

Because of the small sample size inherent in the experimental
design, and to avoid assumptions of normal distribution of resul-
tant data, we used non-parametric methods to compare medians.
To determine if chronic dietary exposure to BMAA results in
(c) (d)
a greater density of NFT, the hypotheses H0 ¼ there is no differ-
Proc. R. Soc. B 283: 20152397

ence in median NFT density between treatment groups and H1 ¼
there is a difference in median NFT density between treatment
groups were evaluated with a Kruskal – Wallis H-test, a non-
parametric analogue of an analysis of variance. A Jonckheere–
Terpstra trend test was used to test the hypotheses H0 ¼
median NFT density is independent of BMAA dose versus
H1 ¼ median NFT density increases with BMAA dose. Different
hypotheses comparing median amounts of BMAA per cohort
treatment group were evaluated with a Kruskal – Wallis H-test: (e) (f)
H0 ¼ there is no difference in median BMAA concentrations
between treatment groups within plasma, brain or CSF, and
H1 ¼ there is a difference between median BMAA concentrations
between treatment groups in plasma, brain or CSF. (In this case,
medians of protein-bound BMAA were used for plasma and
brain samples, while medians of total BMAA content were
used for the CSF; see methods in electronic supplementary
material.) Other hypotheses evaluated with a Kruskal –Wallis
H-test were H0 ¼ there is no difference in the median ratios of
protein to total BMAA concentrations between treatment
groups in plasma or brain, and H1 ¼ there is a difference in the Figure 1. Neuropathology of vervet brain tissue with chronic dietary BMAA
median ratios of protein to total BMAA concentrations between exposures; a comparison of thioflavine-S and b-amyloid (1 – 42) immuno-
treatment groups in plasma or brain. To determine if chronic reactivity. (a) Thioflavine-S stained cells and neuropil threads in the motor
dietary exposure to BMAA is related to the presence of b-amy- cortex; scale bar, 150 mm. (b) Intraneuronal b-amyloid accumulation in
loid deposits, two alternative hypotheses were evaluated with neurons in motor cortex. (c) Vervet extracellular thioflavine-S deposits in
a x 2 test: H0 ¼ there is no difference between treatment types the frontal cortex. (d ) Localized b-amyloid immunostained neocortical
on the number of vervets that develop b-amyloid deposits, and deposits in vervet brains. (e) Thioflavine-S positive senile plaques and tangles
H1 ¼ there is a difference between treatment types on the in human AD temporal cortex. (f ) b-amyloid senile plaques in human tem-
number of vervets that develop b-amyloid deposits. Spearman’s poral cortex of AD patient (86-year-old male; 400 magnification). Human
rank correlation coefficients were calculated to evaluate H0 ¼
brain sections from AD patients were run as reference controls.
there is no relationship between protein-bound or protein-
bound/total ratio of BMAA concentrations in the brain and
NFT counts, and H1 ¼ there is a relationship between protein-
In the replication experiment, chronic L-BMAA exposures
bound or protein-bound/total ratio of BMAA concentrations in
for 140 days again led to hyperphosphorylated tau deposits
the brain and NFT counts.
and NFT formation in all BMAA-fed vervets (figure 2).
Median NFT density differed significantly between treatment
groups (Kruskal –Wallis H statistic ¼ 16.4, p , 0.001). Fur-
3. Results thermore, there was a clear dose relationship between
In the first experiment, AT8-positive tangles and neuronal chronic dietary exposure to L-BMAA and density of NFT
processes, as well as sparse b-amyloid plaque-like deposits, (Jonckheere–Terpstra trend test, Z ¼ 4.4, p , 0.00001). NFT
were observed in brain tissues of the L-BMAA-dosed vervets. were abundant in vervets with chronic dietary exposure
AT8-positive NFT were observed in the perirhinal and to BMAA in the superior frontal, temporopolar (dorsal
entorhinal cortices, amygdala ( paralaminar nucleus), motor and ventral), perirhinal, occipital and entorhinal (anterior
cortex, frontal cortex, temporopolar cortex and occipital and posterior) cortices, and in the amygdala (figure 3). In
cortex of the BMAA-fed animals (figures 1 –3). In contrast, these brain areas, there was a highly significant dose relation-
no AT8 immunopositive inclusions were visualized in the ship between increasing dietary exposure to L-BMAA and
hippocampus (CA1 or dentate gyrus). Sparse immunoposi- NFT density (Jonckheere–Terpstra trend test, Z-scores for
tive b-amyloid deposits were observed primarily in the the 14 brain regions range between 3.13 and 4.87, p ,
frontal, temporal and motor cortices. In the first experiment, 0.001–0.00001; figure 3). The regional differences in NFT
the L-serine treated cohort and the control cohort of four ver- and b-amyloid deposit counts in the brain areas examined
vets were generally negative for tau AT8 and b-amyloid 1–42 were profound. For example, in the occipital cortex, other
neuropathology, while there was an 80–90% reduction of NFT than controls, vervets in the low-dose treatment had the
and plaques in the cohort fed equal amounts of L-BMAA and lowest median count (65) NFT density, while the median
L-serine; these results will be published elsewhere. density (136) of the high-dose BMAA cohort was more than
(a) (b) 5
(c)
Bmc
L Bpc
cgs
M1
PL
cd
Pu
Proc. R. Soc. B 283: 20152397

ac (d)
LF
Amy EC
STS
PrC
(e) (f) (g) (h) (i)
( j) (k) (l) (m)
(n) (o) (p)
Figure 2. Microscopic pathology of chronic dietary L-BMAA exposures in vervets. Representative low-power images (5 magnification) of hyperphosphorylated tau
(AT8) immunostained coronal hemisections from control (a,c) and L-BMAA-fed vervets (b,d ). AT8 immunostaining is seen in the amygdala (Amy), entorhinal (EC),
perirhinal (PrC), primary motor (M1) and temporal cortices of L-BMAA-fed vervets. Higher-power images show predominant tau AT8 staining in superficial cortical
layers II and III with more robust staining over the entorhinal and perirhinal cortices (25 magnification) (d ). Microscopic images (original magnification 120)
show NFT in vervets fed L-BMAA. Tangle-like tau aggregates are seen in the temporal gyrus (e,f ). Dense intracellular tau immunolabelling (g – i) and extracellular
deposits ( j,k) were seen in the parahippocampal gyrus. Abundant neuropil threads, tangles and dystrophic neuronal processes are observed in layers II and III of the
perirhinal cortex (I, high-power images shown in l,m) and the paralaminar nucleus of the amygdala (n). Tau plaques were seen in L-BMAA-fed vervets ranging from
large and diffuse (o) to small dense aggregates ( p). ac, anterior commissure; Amy, amygdala; Bmc, basal nucleus of the amygdala, magnocellular region; Bpc, basal
nucleus of the amygdala, parvicellular subdivision; Cd, caudate; cgs, cingulate gyrus sulcus; EC, entorhinal cortex; L, lateral nucleus of the amygdala; LF, lateral
fissure; M1, primary motor cortex; PrC, perirhinal cortex; PL, paralaminar nucleus; Pu, putamen; STS, superior temporal sulcus.
twice the low-dose NFT density. Co-administration of Supplementing the diet with L-serine resulted in more than
L-serine with high-dose BMAA significantly reduced a 50% NFT reduction in median NFT densities within five
median NFT density (124). This reduction in NFT induced brain regions: temporal (dorsal and ventral), primary motor,
by L-serine occurred in all measured areas of the brain. entorhinal (posterior) and insula cortices. In the perirhinal
(a) 6
effects of BMAA
250 dose on density
of tangles
per cohort of eight vervets

200
median NFT density

150
100
50 amygdala (paralaminar nucl.)

entorhinal cortex (anterior)
Proc. R. Soc. B 283: 20152397

occipital cortex
perirhinal cortex
0 temporopolar cortex (ventral)
high temporopolar cortex (dorsal)
high low superior frontal gyrus
+SER control
(b)
effects of BMAA
250
dose on density
of tangles
per cohort of eight vervets
200
median NFT density
150
100
50 entorhinal cortex (posterior)

insula
primary motor cortex
anterior cingulate gyrus
0 caudate nucleus
high substantia nigra
high low dentate gyrus
+SER control
Figure 3. Median counts for density of AT8 IHC positive staining inclusions plus NFT per brain area by treatment type. Each horizontal surface represents the median of an eight-
vervet cohort statistically significant for dose using the Jonckheere–Terpstra trend test. (a) Brain regions in which AT8 IHC positive density counts from the 210 mg kg21 d21 BMAA
treatment are greatest compared with other treatment types. (b) Brain regions in which AT8 IHC positive density counts from the 210 mg kg21 d21 BMAA plus 210 mg kg21d21
L-serine (highþSER) treatment is less than low-dose (low) BMAA (entorhinal cortex posterior) or in which low-dose NFT density is similar to controls (all other brain areas).
cortex, amygdala and anterior cingulate gyrus, L-serine occurred in brain tissues of individual L-BMAA-fed vervets
reduced NFT densities more than 35%. at concentrations between 0.24 and 2.2 mg mg21 (see elec-
A highly significant ( p , 0.00001) dose relationship tronic supplementary material), similar to Chamorro ALS/
between chronic dietary exposure to L-BMAA and NFT PDC brain tissues (median ¼ 0.6 mg mg21, range ¼ 0.2–
density was also found in other brain regions, but no pro- 1.2 mg mg21) [46], and was detected in blood plasma and
found differences in NFT densities were found between the CSF. Even within the low-dose cohorts, protein-bound
low-dose and control cohorts in these regions which included BMAA within vervet brain tissues (0.24 –0.78 mg mg21)
the dentate gyrus, substantia nigra, caudate nucleus, anterior reached concentrations consistent with the Guam disease.
cingulate gyrus, primary motor cortex and the insula cortex There was no significant difference in protein-bound BMAA
(figure 3b). Finally, dose relationships by treatment group concentrations in blood plasma between treatment groups, but
were also significant in the entorhinal cortex, but this brain there were significant differences for BMAA concentrations for
area differed from the others in that co-administration of brain and CSF samples (Kruskal–Wallis H statistics (corrected
L-serine led to an NFT density not only lower than high- for ties) of 8.69 and 9.09 ( p , 0.05). There was no significant
dose BMAA, but also lower than low-dose BMAA (figure 3b). difference in the ratio of protein to total BMAA concentrations
Chronic dietary exposure to BMAA significantly increased in brain, but there was in blood plasma (H statistic ¼ 13.24,
the likelihood of a vervet developing b-amyloid deposits p , 0.01). Finally, no significant relationship was found between
(x 2 ¼ 15, p , 0.01). One of the eight low-dose BMAA vervets, protein-bound BMAA and NFT density as well as in the protein-
three of the eight high-dose BMAA vervets and two of the bound/total BMAA ratio and NFT density in vervet brains as
eight high-dose BMAA plus L-serine vervets had b-amyloid determined by Spearman’s rank correlation coefficients.
deposits. These b-amyloid deposits were diffuse and sparse
in distribution (figure 1). b-amyloid deposits were not found
in any of the control vervets. 4. Discussion
The relationship between NFT counts and measured con-
centrations of BMAA in the occipital cortex was also of (a) L-BMAA triggers neuropathology
interest. BMAA could not be detected in control vervets or Chronic dietary exposure to L-BMAA results in the formation
baseline samples using LC-MS/MS. Protein-bound BMAA of NFT and b-amyloid deposits in a clear dose relationship.
O 7
(BMAA)
H3C
N OH
H NH2
vervet
tau proteins hold microtubule together

1
extracellular fluid
HOOC
42
cytosol amyloid precursor
Proc. R. Soc. B 283: 20152397

protein (APP)
hyperphosphorylated tau separate 1

from the microtubule
Ab-42 a helix
fragment
hyperphosphorylated tau filaments HOOC
42
b-pleated sheet
paired helical filaments (PHF)
neurofibrillary tangles (NFT ) amyloid plaque
Figure 4. Theoretical pathways of development of ALS/PDC and AD neuropathology from chronic dietary BMAA exposure. (a) Tau proteins which bind microtubules
become hyperphosphorylated, leading to dissociation of hyperphosphorylated tau fragments. These form paired helical filaments, leading to the formation of
neurofibrillary tangles. (b) The APP is cleaved, producing b-amyloid (Ab-42) fragments which are in an a-helix conformation. These change to a b-pleated
sheet conformation, oligomerize, forming amyloid plaques.
Other protein inclusions similar to those found in brain were observed in any of these vervets in the two experiments.
tissues from Chamorros who died with ALS/PDC were Specific immunological methods (AT8) permit evaluation of
also found. Chronic dietary exposure to L-BMAA triggered neuronal changes before the actual formation of NFT and neu-
tauopathies in all BMAA-dosed vervets including those at ropil threads (figure 2). In vervets with chronic dietary
the low-dose treatment but NFT densities varied between exposure to BMAA, we observed changes in the transentorh-
brain regions. Concentrations of BMAA in vervet brains fell inal region of the temporal lobe, but none in Ammon’s horn
within the range measured in post-mortem brain tissues of of the hippocampus. Extensively distributed NFT formations
Chamorros who died with ALS/PDC confirming BMAA with gliosis characterize ALS/PDC. Dementia in these cases
exposures in the vervets that are clinically relevant. Further- is attributable to tangles and neuronal dropout in the neo-
more, the regional densities of NFT are similar in both the cortex, resembling the pattern reported for AD, but far more
Chamorros and L-BMAA-fed vervets [47,48]. widely distributed [4]. NFT in ALS/PDC brain tissues stain
There was a significant dose relationship between BMAA positively with antibodies to hyperphosphorylated tau protein
and NFT density in all affected regions of the brains [5]. Thus, the distribution of AT8-positive tangles following
(figure 3). The distribution of NFT and their relationship to chronic dietary BMAA exposure in vervets is similar to the
dose exposure in the temporal lobe is similar to Braak 1 early histopathology reported previously in ALS/PDC.
stage AD pathology [49] (figure 3). Consistent with the neuro- The paucity of clinical symptoms in the BMAA-fed vervets
pathology of preclinical AD, no profound clinical symptoms corresponds to the finding of NFT in 5/29 asymptomatic
Chamorro patients who died without ALS or PD being recog- exposures of aged vervets to BMAA results in more profound 8
nized clinically [4]. Furthermore, b-amyloid plaques have been histopathology.
detected in human ALS/PDC patients who remain cognitively We have sponsored FDA-approved human clinical trials
intact [50]. The fact that BMAA-dosed vervets produced NFT (ClinicalTrials.gov Identifier NCT01835782) to determine if
and rare b-amyloid deposits in both experiments supports L-serine is a safe and efficacious treatment to reduce disease
the theory that BMAA in the traditional diet is a cause of the progression in ALS patients. We hope to initiate human clini-
Chamorro disease. cal trials of L-serine for mild cognitive impairment and early
onset AD in the near future.
(b) Neurofibrillary tangles and b-amyloid deposit In conclusion, Koch’s postulates [38] have been satisfied
with respect to establishing chronic dietary exposure to
formation BMAA as a cause of a neurodegenerative illness: (i) BMAA
Normal microtubules which serve as the pathways for antero- has been identified in post-mortem brain tissue from ALS/
Proc. R. Soc. B 283: 20152397

grade and retrograde transport within the neuron unravel as the PDC patients from Guam who consume a BMAA-rich diet
soluble monomeric tau proteins become hyperphosphorylated but not in control patients who have not been exposed to the
and detach from the microtubule. These hyperphosphorylated tau traditional Chamorro diet, (ii) vervets fed BMAA over 140
proteins form paired helical filaments, and thence aggregates days developed NFT and b-amyloid deposits, and (iii)
leading to NFT (figure 4a). In the b-amyloid plaque pathway BMAA was isolated and identified in BMAA-fed vervets that
(figure 4b), amyloid precursor protein (APP) is cleaved by b-secre- had NFT and b-amyloid deposits in their brains. This study
tase and g-secretase into b-amyloid fragments. Although the indicates that chronic exposure to BMAA can trigger neurode-
initial confirmation of Ab-42 is an a-helix, transformation to a generative illness and that adding L-serine to the diet can
b-pleated sheet conformation is a necessary step in plaque reduce the risk of disease.
formation. When the b-pleated sheets oligomerize, they can even-
tually join into polymers which form plaques (figure 4b). Our data Ethics. This work was carried out with permission from the Behaviour-
suggest that chronic dietary exposure to BMAA triggers both the al Sciences Foundation (BSF) in St. Kitts and all procedures were
approved by the Institutional Animal Care and Use Committee
NFT and b-amyloid pathways.
(IACUC) of BSF and McGill University (Quebec, Canada). Dosing
and animal health was monitored by an on-site DVM veterinarian.
(c) Neuroprotective mechanisms of L-serine Authors’ contributions. P.A.C., D.C.M. and S.A.B. participated in the con-
Larger BMAA doses resulted in increased protein-bound ception and design of the experiment, D.C.M. and D.A.D. performed
histopathological analyses, S.A.B. conducted the chemical analyses,
BMAA concentrations in the brain. However, although
and all authors participated in conducting the experiments, data
L-serine reduced NFT density, it did not alter the ratio of
analysis and writing the manuscript.
protein-bound to total BMAA, which may be physiologically Competing interests. The Institute for Ethnomedicine has applied for
invariant. Possible neuroprotective mechanisms of L-serine patents for the use of L-serine to treat neurodegenerative illness
include prevention of BMAA misincorporation in specific (US 13/683,821) and for screening potential drug candidates using
proteins involved in NFT formation. Misincorporation at BMAA-induced neurodegeneration (US 14/229,624).
rates as low as 1/10 000 can result in neurodegeneration Funding. This study was supported by the Josephine P. and John
[35], but such levels may be below our ability to differentiate. J. Louis Foundation, the William Stamps Farish Fund, Douglas and
Elizabeth Kinney, and Patrick and Heather Henry.
There may also be additional neuroprotective mechanisms
Acknowledgements. We thank Prof. Roberta Palmour, the late Prof. Frank
other than prevention of misincorporation. R. Ervin, Dr Amy Beierschmitt, Mr James Powell and the staff of the
Behavioural Science Foundation in St. Kitts for overseeing the care
and dosing of the vervets, Dr Peter Wyatt of Queen Mary, University
(d) Implications of chronic BMAA exposure for of London for technical advice, Ms Jane Cox and Mr James Powell for
neurodegenerative disease dose preparation, Mr W. Broc Glover for sample preparation, Dr
Robert Switzer and his team at NeuroScience Associates in Tennessee
BMAA-producing cyanobacteria occur globally, perhaps for helpful discussions at the initiation of the study and choice of
causing similar neuropathologies. Our finding that all of antibodies, Huron Digital Pathology for technical advice regarding
the low-dose vervets developed tauopathies with NFT has automated digital scanning of the brain sections, Dr Walter Bradley
implications for human health. BMAA may serve as an at the Department of Neurology, Miller School of Medicine, Miami
for clinical observations of the vervets, the staff of the Miami Brain
environmental trigger for some forms of other neurodegen-
Endowment Bank at the Miller School of Medicine for manage-
erative illnesses including sporadic ALS and AD. In human ment of the biospecimen repository, Ms Marilyn Asay for helping
beings, increasing age is a risk factor for ALS, AD and PD. to prepare the manuscript and graphs, and Mr Michael Rothman
We have initiated experiments to determine if chronic dietary for the drawing.
References
1. Cannon JR, Greenamyre JT. 2011 The role of 3. Arnold A, Edgren DC, Palladino VS. 1953 neurology symposium vol II: motor neuron diseases:
environmental exposures in neurodegeneration Amyotrophic lateral sclerosis: fifty cases research on amyotrophic lateral sclerosis and related
and neurodegenerative diseases. Toxicol. Sci. 124, observed on Guam. J. Nerv. Ment. Dis. disorders (eds FH Norris, LT Kurland), pp. 51 –60.
225–250. (doi:10.1093/toxsci/kfr239) 117, 135 –139. (doi:10.1097/00005053195 New York, NY: Grune and Stratton.
2. Langston JW, Ballard P, Tetrud JW, Irwin I. 1983 311720-00005) 5. Buée-Scherrer V, Buee L, Hof PR, Leveugle B, Gilles
Chronic Parkinsonism in humans due to a product 4. Hirano A, Malamud N, Kurland LT, Zimmerman HM. C, Loerzel AJ, Perl DP, Delacourte A. 1995
of meperidine-analog synthesis: Science 219, 1968 A review of the pathologic findings in Neurofibrillary degeneration in amyotrophic lateral
979–980. (doi:10.1126/science.6823561) amyotrophic lateral sclerosis. In Contemporary sclerosis/parkinsonism-dementia complex of Guam.
Immunochemical characterization of tau proteins. to cyanobacteria an environmental risk factor for 31. Yin HZ, Yu S, Hsu C-I, Liu J, Acab A, Wu R, Tao A, 9
Am. J. Pathol. 146, 924. amyotrophic lateral sclerosis and other Chiang BJ, Weiss JH. 2014 Intrathecal infusion of
6. Kurland LT, Radhakrishnan K, Williams DB, Waring neurodegenerative diseases? Amyotroph. Lateral BMAA induces selective motor neuron damage and
SC. 1994 Amyotrophic lateral sclerosis-Parkinsonism- Scler. Frontotemp. Degener. 14, 325–333. (doi:10. astrogliosis in the ventral horn of the spinal cord.
dementia complex on Guam: epidemiologic and 3109/21678421.2012.750364) Exp. Neurol. 261, 1–9. (doi:10.1016/j.expneurol.
etiological perspectives. In Motor neuron disease 19. Torbick N, Hession S, Stommel E, Caller T. 2014 2014.06.003)
(ed. AC Williams), pp. 109– 130. London, UK: Mapping amyotrophic lateral sclerosis lake risk 32. Rush T, Liu X, Lobner D. 2012 Synergistic toxicity of
Chapman and Hall Medical. factors across northern New England. Int. J. Health the environmental neurotoxins methylmercury and
7. Garruto RM, Gajdusek DC, Chen KM. 1981 Geogr. 13, 1 –14. (doi:10.1186/1476-072x-13-1) b-N-methylamino-L-alanine. Neuroreport 23,
Amyotrophic lateral sclerosis and Parkinsonism- 20. Sabel CE, Boyle PJ, Löytönen M, Gatrell AC, Jokelainen 216–219. (doi:10.1097/wnr.0b013e32834fe6d6)
dementia among Filipino migrants to Guam. Ann. M, Flowerdew R, Maasilta P. 2003 Spatial clustering of 33. Xie X, Basile M, Mash DC. 2013 Cerebral uptake and
Neurol. 10, 341 –350. (doi:10.1002/ana.410100405) amyotrophic lateral sclerosis in Finland at place of birth protein incorporation of cyanobacterial toxin b-N-
Proc. R. Soc. B 283: 20152397

8. Garruto RM, Gajdusek DC, Chen KM. 1980 and place of death. Am. J. Epidemiol. 157, 898–905. methylamino-L-alanine. Neuroreport 24, 779– 784.
Amyotrophic lateral sclerosis among Chamorro (doi:10.1093/aje/kwg090) (doi:10.1097/wnr.0b013e328363fd89)
migrants from Guam. Ann. Neurol. 8, 612–619. 21. Karlsson O, Berg AL, Hanrieder J, Arnerup G, 34. Dunlop RA, Cox PA, Banack SA, Rodgers KJ. 2013
(doi:10.1002/ana.410080612) Lindström AK, Brittebo EB. 2014 Intracellular fibril The non-protein amino acid BMAA is
9. Bell EA. 2009 The discovery of BMAA, and examples formation, calcification, and enrichment of misincorporated into human proteins in place of
of biomagnification and protein incorporation chaperones, cytoskeletal, and intermediate filament L-serine causing protein misfolding and
involving other non-protein amino acids. proteins in the adult hippocampus CA1 following aggregation. PLoS ONE 8, e75376. (doi:10.1371/
Amyotroph. Lateral Scler. 10, 21 –25. (doi:10.3109/ neonatal exposure to the nonprotein amino acid journal.pone.0075376)
17482960903268700) BMAA. Arch. Toxicol. 89, 423–436. (doi:10.1007/ 35. Lee JW et al. 2006 Editing-defective tRNA
10. Spencer PS, Nunn PB, Hugon J, Ludolph AC, Ross s00204-014-1262-2) synthetase causes protein misfolding and
SM, Roy DN, Robertson RC. 1987 Guam amyotrophic 22. Karamyan VT, Speth RC. 2008 Animal models of neurodegeneration. Nature 443, 50 –55. (doi:10.
lateral sclerosis-Parkinsonism-dementia linked to a BMAA neurotoxicity: a critical review. Life Sci. 82, 1038/nature05096)
plant excitant neurotoxin. Science 237, 517–522. 233 –246. (doi:10.1016/j.lfs.2007.11.020) 36. Arif M, Kazim SF, Grundke-Iqbal I, Garruto RM, Iqbal
(doi:10.1126/science.3603037) 23. Cox PA et al. 2005 Diverse taxa of cyanobacteria K. 2014 Tau pathology involves protein phosphatase
11. Duncan MW. 1992 b-methylamino-L-alanine produce b-N-methylamino-L-alanine, a 2A in Parkinsonism-dementia of Guam. Proc. Natl
(BMAA) and amyotrophic lateral sclerosis- neurotoxic amino acid. Proc. Natl Acad. Sci. Acad. Sci. USA 111, 1144–1149. (doi:10.1073/pnas.
parkinsonism dementia of the Western Pacific. Ann. USA 102, 5074–5078. (doi:10.1073/pnas. 1322614111)
NY Acad. Sci. 648, 161 –168. (doi:10.1111/j.1749- 0501526102) 37. Pablo J, Banack SA, Cox PA, Johnson TE,
6632.1992.tb24534.x) 24. Downing S, Banack SA, Metcalf JS, Cox PA, Downing Papapetropoulos S, Bradley WG, Buck A,
12. Cox PA, Banack SA, Murch SJ. 2003 Biomagnification TG. 2011 Nitrogen starvation of cyanobacteria Mash DC. 2009 Cyanobacterial neurotoxin BMAA in
of cyanobacterial neurotoxins and neurodegenerative results in the production of b-N-methylamino-L- ALS and Alzheimer’s disease. Acta Neurol. Scand.
disease among the Chamorro people of Guam. Proc. alanine. Toxicon 58, 187 –194. (doi:10.1016/j. 120, 216–225. (doi:10.1111/j.1600-0404.2008.
Natl Acad. Sci. USA 100, 13 380–13 383. (doi:10. toxicon.2011.05.017) 01150.x)
1073/pnas.2235808100) 25. Richer R, Banack SA, Metcalf JS, Cox PA. 2015 The 38. Koch R. 1880 Investigations into the etiology of
13. Banack SA, Murch SJ, Cox PA. 2006 Neurotoxic persistence of cyanobacterial toxins in desert soils. traumatic infective diseases (trans. WW Cheyne).
flying foxes as dietary items for the Chamorro J. Arid Environ. 112, 134 –139. (doi:10.1016/j. London, UK: The New Sydenham Society.
people, Marianas Islands. J. Ethnopharmacol. 106, jaridenv.2014.01.023) 39. LaFerla FM, Oddo S. 2005 Alzheimer’s disease: Ab,
97 –104. (doi:10.1016/j.jep.2005.12.032) 26. Cox PA, Richer R, Metcalf JS, Banack SA, Codd GA, tau and synaptic dysfunction. Trends Mol. Med. 11,
14. Cheng R, Banack SA. 2009 Previous studies Bradley WG. 2009 Cyanobacteria and BMAA exposure 170–176. (doi:10.1016/j.molmed.2005.02.009)
underestimate BMAA concentrations in cycad flour. from desert dust: a possible link to sporadic ALS 40. Heuer E, Rosen RF, Cintron A, Walker LC. 2012
Amyotroph. Lateral Scler. 10, 41 –43. (doi:10.3109/ among Gulf War veterans. Amyotroph. Lateral Scler. Nonhuman primate models of Alzheimer-like
17482960903273528) 10, 109–117. (doi:10.3109/17482960903286066) cerebral proteopathy. Curr. Pharm. Des. 18,
15. Murch SJ, Cox PA, Banack SA. 2004 A mechanism 27. Metcalf JS, Banack SA, Lindsay J, Morrison LF, Cox 1159– 1169. (doi:10.2174/138161212799315885)
for slow release of biomagnified cyanobacterial PA, Codd GA. 2008 Co-occurrence of b-N- 41. Lemere CA et al. 2004 Alzheimer’s disease Ab
neurotoxins and neurodegenerative disease in methylamino-L-alanine, a neurotoxic amino acid vaccine reduces central nervous system Ab levels in
Guam. Proc. Natl Acad. Sci. USA 101, 12 228– with other cyanobacterial toxins in British a non-human primate, the Caribbean vervet.
12 231. (doi:10.1073/pnas.0404926101) waterbodies, 1990–2004. Environ. Microbiol. 10, Am. J. Pathol. 165, 283–297. (doi:10.1016/s0002-
16. Jonasson S, Eriksson J, Berntzon L, Spáčil Z, Ilag L, 702 –708. (doi:10.1111/j.1462-2920.2007.01492.x) 9440(10)63296-8)
Ronnevi L, Rasmussen U, Bergman, B. 2010 Transfer 28. Liu X, Rush T, Zapata J, Lobner D. 2009 b-N- 42. Dunn WD, Gearing M, Park Y, Zhang L, Hanfelt J,
of a cyanobacterial neurotoxin within a temperate methylamino-L-alanine induces oxidative stress and Glass JD, Gutman DA. In press. Applicability of
aquatic ecosystem suggests pathways for human glutamate release through action on system Xc2. digital analysis and imaging technology in
exposure. Proc. Natl Acad. Sci. USA 107, 9252 – Exp. Neurol. 217, 429 –433. (doi:10.1016/j. neuropathology assessment. Neuropathology.
9257. (doi:10.1073/pnas.0914417107) expneurol.2009.04.002) (doi:10.1111/neup.12273)
17. Brand LE, Pablo J, Compton A, Hammerschlag N, 29. Rao SD, Banack SA, Cox PA, Weiss JH. 2006 BMAA 43. Glover WB, Baker TC, Murch SJ, Brown PN. 2015
Mash DC. 2010 Cyanobacterial blooms and the selectively injures motor neurons via AMPA/kainate Determination of b-N-methylamino-L-alanine,
occurrence of the neurotoxin, beta-N-methylamino- receptor activation. Exp. Neurol. 201, 244– 252. N-(2-aminoethyl)glycine, and 2,4-diaminobutyric
L-alanine (BMAA), in South Florida aquatic food (doi:10.1016/j.expneurol.2006.04.017) acid in food products containing cyanobacteria by
webs. Harmful Algae 9, 620 –635. (doi:10.1016/j. 30. Chiu AS, Gehringer MM, Braidy N, Guillemin GJ, ultra-performance liquid chromatography and
hal.2010.05.002) Welch JH, Neilan BA. 2013 Gliotoxicity of the tandem mass spectrometry: single-laboratory
18. Bradley WG, Borenstein AR, Nelson LM, Codd GA, cyanotoxin, b-methyl-amino-L-alanine (BMAA). validation. J. AOAC Int. 98, 1559–1567. (doi:10.
Rosen BH, Stommel EW, Cox PA. 2013 Is exposure Sci. Rep. 3, 1482. (doi:10.1038/srep01482) 5740/jaoacint.15-084)
44. Banack SA, Downing TG, Spácil Z, Purdie E, Metcalf 46. Murch SJ, Cox PA, Banack SA, Steele JC, Sacks OW. the neuropathological findings. Acta Neuropathol. 10
JS, Downing S, Esterhuizen M, Codd GA, Cox PA. 2004 Occurrence of b-methylamino-L-alanine 88, 405 –412. (doi:10.1007/bf00389491)
2010 Distinguishing the cyanobacterial neurotoxin (BMAA) in ALS/PDC patients from Guam. Acta 49. Braak H, Braak E. 1995 Staging of Alzheimer’s
b-N-methylamino-L-alanine (BMAA) from its Neurol. Scand. 110, 267–269. (doi:10.1111/j.1600- disease-related neurofibrillary changes. Neurobiol.
structural isomer 2,4-diaminobutyric acid (2,4-DAB). 0404.2004.00320.x) Aging 16, 271–278. (doi:10.1016/0197-
Toxicon 56, 868 –879. (doi:10.1016/j.toxicon.2010. 47. Hof PR, Perl DP, Loerzel AJ, Steele JC, Morrison JH. 4580(95)00021-6)
06.006) 1994 Amyotrophic lateral sclerosis and 50. Schmidt ML, Lee VY, Saido T, Perl D, Schuck T,
45. Banack SA, Metcalf JS, Jiang L, Craighead D, parkinsonism-dementia from Guam: differences in Iwatsubo T, Trojanowski JQ. 1998 Amyloid
Ilag LL, Cox PA. 2012 Cyanobacteria produce neurofibrillary tangle distribution and density in the plaques in Guam amyotrophic lateral sclerosis/
N-(2-aminoethyl)glycine, a backbone for hippocampal formation and neocortex. Brain Res. parkinsonism-dementia complex contain species of
peptide nucleic acids which may have been 650, 107–116. (doi:10.1016/0006-8993(94)90212-7) Ab similar to those found in the amyloid plaques
the first genetic molecules for life on earth. 48. Oyanagi K, Makifuchi T, Ohtoh T, Chen KM, van der of Alzheimer’s disease and pathological aging.
Proc. R. Soc. B 283: 20152397

PLoS ONE 7, e49043. (doi:10.1371/journal.pone. Schaaf T, Gajdusek DC, Chase TN, Ikuta F. 1994 Acta Neuropathol. 95, 117– 122. (doi:10.1007/
0049043) Amyotrophic lateral sclerosis of Guam: the nature of s004010050774)
APPENDIX M:
WETLANDS CONCEPTUAL ECOLOGICAL MODEL
APPENDIX N:
DR. GARY GOFORTH LETTER TO U.S. ARMY
CORPS OF ENGINEERS
Suggestions for Consideration by the U.S. Army Corps of Engineers to Reduce the Destructive
Discharges to the St. Lucie River and Estuary – February 2014
Gary Goforth, P.E., Ph.D.

February 7, 2014
Executive Summary
The first reported discharge from Lake Okeechobee to the St. Lucie River occurred on June 15,
1923. For the ensuing 90 years, the coastal communities have sacrificed their health, economy,
and environment in order to protect the health, safety and welfare of residents around Lake
Okeechobee. 2013 was no exception, as over 1.5 million acre feet of Lake untreated water was
released to the estuaries, carrying millions of pounds of harmful nutrients, sediments and blue
green algae. The damage to the coastal systems was monumental, including mortality of marine
organisms and toxic algae blooms necessitating the health department to issue advisories for
residents and tourists to avoid contact with the water.
I commend the Corps for holding “After Action Review” public workshops in January 2014 to
receive public input. Moving forward in collaboration, and without criticizing any party,
suggestions are offered below to reduce the destructive regulatory releases to the estuaries.
The suggestions are organized into four broad categories:
A. Complete the Documentation of the 2013 Lake Releases

B. Actions That Could Be Initiated Within The Next Year
C. When the LORS guidance is to send “Up to the maximum practicable releases to the
WCAs”
D. When Regulatory Releases To The Estuaries Are Unavoidable
A. COMPLETE THE DOCUMENTATION OF THE 2013 LAKE RELEASES
1. Consider preparing an “After Action” report covering the regulatory releases to the
estuaries. The following components could be included in this report.
a. An evaluation of the short-term and long-term environmental damages, and the
risk to public health and safety, resulting from the discharge of millions of pounds
of nitrogen and phosphorus from the Lake into the rivers, estuaries, lagoons and
near-shore reef. Impacts from the following should be identified and methods to
mitigate the damages should be suggested.
i. Freshwater
ii. Nitrogen (it is estimated that over 1.4 million pounds were discharged
from the Lake to the St. Lucie River between May 8 and October 21, 2013)
iii. Phosphorus (it is estimated that over 130,000 pounds were discharged
from the Lake to the St. Lucie River between May 8 and October 21, 2013)
iv. Blue green algae – present at many samples of water discharged from the
Lake since May 8
2
b. An evaluation of the short-term and long-term economic damage resulting from

the Lake discharges to the rivers, estuaries, lagoons and near-shore reef.
c. An analysis of the sediment transported from the Lake into the C-43 and C-44
canals, and subsequently into the St. Lucie and Caloosahatchee Estuaries.
According to data obtained from the SFWMD database, over 15 million pounds of
TSS (suspended sediment) was discharged to the Lake into the St. Lucie River and
Estuary during the 2013 releases. The analysis could document the transport and
current location of this sediment gone. The analysis could evaluate options,
including expediting the muck removal project recommended in the Indian River
Lagoon – South project, with subsequent routine dredging to remove
accumulated sediment (perhaps disposal on surplus lands obtained for the C-44
Reservoir/STA) and establishing sediment sump areas such as was recently
completed in the C-51 canal upstream of S-155.
d. A Lessons Learned document summarizing operational decisions, flow summaries
(see Figure 1 and Tables 1 and 2), nutrient load summaries, etc. from the current
event, along with identification of ways to reduce discharges to the estuaries in
future events.
2. If a permit was issued to the Corps to discharge water from Lake Okeechobee into the St.
Lucie River and Indian River Lagoon, portions of which are classified as Outstanding
Florida Waters, suggest documenting adherence to the permit conditions. If a permit
was not issued by the State that authorizes these discharges, consider requesting follow-
up action from the Florida Department of Environmental Protection. Some of the
Outstanding Florida Waters affected by the Lake discharges to the St. Lucie River and
Indian River Lagoon include the following:
1. Hobe Sound National Wildlife Refuge
2. St. Lucie Inlet State Preserve
3. Indian River State Aquatic Preserve
4. Jenson Beach State Aquatic Preserve
From Chapter 62.302-700, Florida Administrative Code:
(1) It shall be the Department policy to afford the highest protection to
Outstanding Florida Waters and Outstanding National Resource Waters. No
degradation of water quality, other than that allowed in subsections 62-4.242(2)
and (3), F.A.C., is to be permitted in Outstanding Florida Waters and Outstanding
National Resource Waters, respectively, notwithstanding any other Department
rules that allow water quality lowering.
3
Suggestions for Consideration by the U.S. Army Corps of Engineers to Reduce the Destructive Discharges to the St. Lucie River and
Estuary – February 2014
Figure 1. Summary of flows between May 8 and October 21, 2013.
From Lake Okeechobee

Flows between May 8, 2013 and October 21, 2013
All flows in 1000 acre feet Local basin runoff
All estimates are draft and subject to revision.
418.2 417.7
St. Lucie River, Estuary,
Lake Okeechobee 3.1 250.8 IRL and coastal waters
Total LO releases to estuaries

1,455.2
Total Lake releases sent to the south
189.7
977 Lake water used in EAA and L-8 basins
24.8 52.6
0.0 L-8 Canal Total Lake water sent through EAA and L-8 to STAs and WCAs
966 77.1 65.8
17.3 1.4 Basin EAA and L-8 basin runoff sent to the Lake
79.5 City of WPB 52.3
15.8 26.1 2.4 Net Lake flow to the south (Lake pass
1,211 0.0 through minus southern runoff into Lake)
Caloosahatchee 71.2 13.5
Estuary and coastal 127.4 71.0
Lake Worth Lagoon
waters 304.1
21.5
To Rotenberger Everglades Agricultural Area (EAA) 185.9 46.1
0 38.2
Ratio of Lake releases to estuaries
to net Lake flow to the south
108
To Holey Land
That is, 108 times more Lake water has been
0 225.1
sent to the estuaries than to the south, when
0.1
adjusted for runoff sent to the Lake.
144.1
5.2 217.0 0.8 DRAFT. Other sources of water to and from the Lake exist and can be added later
Stormwater Treatment Areas and Water Conservation Areas NOT TO SCALE
4
Table 1. Summary of 2013 Lake Releases to Estuaries and WCAs (excludes Lake releases for
water supply to adjacent basins).
Source Jan 1- May 7 May 8-Oct 21 Oct 22-Dec 31 Total 2013
Lake Release to STA/WCAs 41,334 65,800 39,375 146,509

Basin Runoff to STA/WCAs 148,370 1,002,127 12,374 1,162,871
Total to STAs/WCAs 189,704 1,067,927 51,749 1,309,380
Lake as % of Total 22% 6% 76% 11%
Lake Release to St. Lucie R/E 1,450 417,733 768 419,951

Basin Runoff to St. Lucie R/E 41,283 719,281 30,438 791,002
Total to St. Lucie R/E 42,733 1,137,015 31,206 1,210,953
Lake as % of Total 3% 37% 2% 35%
Lake Release to Caloos. R/E 65,957 966,463 23,808 1,056,228

Basin Runoff to Caloos. R/E 66,301 1,762,350 45,587 1,874,237
Total to Caloos. R/E 132,257 2,728,813 69,395 2,930,465
Lake as % of Total 50% 35% 34% 36%
Lake Release to LW Lagoon 5,058 71,030 24,877 100,966

Basin Runoff to LW Lagoon 19,502 304,100 7,442 331,045
Total to LW Lagoon 24,560 375,130 32,320 432,010
Lake as % of Total 21% 19% 77% 23%
Total Lake Releases 113,799 1,521,027 88,828 1,723,653

Notes:
1. Estimates are preliminary and subject to revision; all flows in acre feet.
2. Basin Runoff to STAs/WCAs include only those basins that discharge to the STAs.
3. Basin Runoff to St. Lucie River/Estuary includes runoff from C-23, C-24, C-25 and C-44
4. Basin Runoff to Caloosahatchee River/Estuary includes runoff from S-78 and S-79 basins only.
5. Basin Runoff to Lake Worth Lagoon includes only runoff from C-51 Basin.
Table 2. Summary of 2013 flows to STAs/WCAs.
Lake Stormwater Total STA Hydraulic

STA Period Releases Flows to WCAs Area Load from Lake
AF AF AF acres cm per day
Jan 1- May 7 6,344 4,718 11,062 0.30
May 8-Oct 21 21,476 83,315 104,791 0.78
STA-1E 5,000
Oct 22-Dec 31 4,828 102 4,931 0.41
Total 2013 32,649 88,135 120,783 0.55
Jan 1- May 7 1,179 25,791 26,971 0.04
May 8-Oct 21 38,217 154,563 192,780 1.07
STA-1W 6,500
Oct 22-Dec 31 8,466 3,572 12,038 0.56
Total 2013 47,862 183,926 231,789 0.61
Jan 1- May 7 9,352 51,809 61,160 0.14
May 8-Oct 21 395 288,434 288,829 0.00
STA-2 15,500
Oct 22-Dec 31 8,420 13,941 22,361 0.23
Total 2013 18,167 354,184 372,350 0.10
Jan 1- May 7 24,358 60,229 84,587 0.36
May 8-Oct 21 5,712 362,266 367,978 0.06
STA-3/4 16,300
Oct 22-Dec 31 17,657 -5,903 11,755 0.47
Total 2013 47,728 416,592 464,320 0.24
Jan 1- May 7 0 5,805 5,805 0.00
May 8-Oct 21 0 96,144 96,144 0.00
STA-5/6 13,700
Oct 22-Dec 31 0 662 662 0.00
Total 2013 0 102,612 102,612 0.00
Jan 1- May 7 41,233 148,352 189,585 0.17
May 8-Oct 21 65,800 984,723 1,050,523 0.21
Total 57,000
Oct 22-Dec 31 39,372 12,374 51,746 0.30
Total 2013 146,405 1,145,448 1,291,854 0.21
Jan 1- May 7 22% 78%
May 8-Oct 21 6% 94%
Portion of STA Inflows
Oct 22-Dec 31 76% 24%
Total 2013 11% 89%
Jan 1- May 7 100 19 119
Diverted directly to May 8-Oct 21 0 17,404 17,405
WCAs Oct 22-Dec 31 3 0 3
Total 2013 104 17,423 17,527
Jan 1- May 7 41,334 148,370 189,704
Total Lake to STAs and May 8-Oct 21 65,800 1,002,127 1,067,927
WCAs Oct 22-Dec 31 39,375 12,374 51,749
Total 2013 146,509 1,162,871 1,309,380
Notes:
1. "Stormwater Flows" is total STA outflow to WCAs minus Lake inflows to STAs; assumed all
Lake flows pass through STAs
2. Negative value appears for STA-3/4 during Oct. 22-Dec. 31 because not all Lake flows left
the STA during this period
3. Some STA-3/4 and STA-5/6 outflow was sent to Rotenberger WMA instead of the WCAs.
6
B. ACTIONS THAT COULD BE INITIATED WITHIN THE NEXT YEAR
3. Consider sponsoring an Emergency Management Workshop for staff of the U.S Army
Corps of Engineers (Corps), SFWMD and stakeholders.
a. Participants could be asked to develop an Operation Plan for the following
scenario:
“The U.S. Army Corps of Engineers has decided to close the gates at Port
Mayaca (S-308), which will terminate discharges to the St. Lucie River and
Indian River Lagoon. No additional discharges will occur to the C-43 canal
(Caloosahatchee River) above the rates in the 2008 Lake Okeechobee
Regulation Schedule”
b. Workshop participants could be asked to develop an Operation Plan and a list of
projects necessary to safely manage the waters of Lake Okeechobee, which
could include reducing inflows to the Lake and increasing the amount of water
sent to the south.
c. Participants could be asked to identify physical, legal or discretionary operational
constraints to moving more water to the south, and to identify courses of action
necessary to resolve the constraints.
4. Consider preparing an updated risk assessment on the potential risk of failure of the
Herbert Hoover Dike, now that the initial 21.4 mile section of the dike has been
rehabilitated. Issues regarding the structural integrity of the Herbert Hoover Dike led to
lowering the upper limit of the Lake Okeechobee regulation schedule (2008 LORS) from
18.5 ft NGVD to 17.25 ft NGVD, resulting in increased pressure to send Lake releases to
the estuaries. From a presentation by Lt. Col. Greco at the August 8, 2013 WRAC
meeting in Stuart, it is my understanding that the Corps estimated the following annual
risks of failure before the rehabilitation:
7
Documents available on the Corps website indicate the risks are much less after the
present rehabilitation:
a. “Since 2007, the Corps has made a significant investment, over $300 million, in
projects designed to reduce the risk of catastrophic failure of the aging
structure.”
b. “The implementation of the 21.4 mile cutoff wall component in Reach 1 satisfies
the majority of the risk reduction goals. 1”
5. Consider initiating the re-evaluation of the Lake Okeechobee regulation schedule with a
goal to minimize Lake releases to the estuaries.
a. As part of this re-evaluation, consider performing a cost benefit analysis of
making regulatory releases to the estuaries.
i. Costs would include the economic and environmental costs of
discharging nutrient and sediment-laden freshwater to the estuaries,
including
1. Immediate loss of income and jobs for people who make their
living on or in the water,
2. Near-term loss of income as potential tourists and others decide
not to visit the coastal regions, or buy homes in the area
3. Loss of real estate value
4. Ecological costs to the rivers, estuaries, lagoons and near-shore
reef systems.
ii. Benefits of releasing water from the Lake have been described as two-
fold:
1. Benefit to the region around the Lake: reducing the risk of breach
of the Herbert Hoover Dike, and
2. Benefit to the Lake: improving the health of Lake Okeechobee
iii. Request examining the balance of potential economic costs of a dike
breach with the certainty of economic and environmental costs of
these Lake discharges to the St. Lucie River and Indian River Lagoon.
This analysis may likely indicate that the coastal regions are suffering the
majority of the economic and environmental costs, while the Lake and the
region south of the Lake are receiving the majority of the benefits. If so,
consider determining how the Lake Okeechobee regulation schedule could
be revised to more fairly align the majority of the impacts (costs) to the
region receiving the majority of the benefits.
b. Now that the dike rehabilitation efforts have satisfied the majority of the risk
reduction goals (USACE 2012), consider raising the upper limit of the lake
regulation schedule to more equitably balance the impacts to the estuaries, the
1
REVIEW PLAN for HERBERT HOOVER DIKE DESIGN AND CONSTRUCTION PHASES Martin, Palm Beach, Hendry,
Glades and Okeechobee Counties, Florida Jacksonville District Second Revision, April 2012.
8
Lake and the Everglades ecosystems. In addition, consider suspending the

regulatory releases to the estuary when the Lake is below a stage of 16 ft NGVD.
c. Consider eliminating any restriction of moving water south to the STAs and
WCAs.
i. Additional treatment areas are operational that were not operational
when the 2008 LORS was developed.
ii. The 2008 LORS recommended a dramatic decrease in the amount of
regulatory releases to be sent south to the STAs compared to the
design of STA-3/4. In 1994, STA-3/4 was designed to capture and treat
an average of more than 250,000 acre feet per year (80 billion gallons
per year) of Lake regulatory releases.
iii. According to the SFWMD’s report (SFWMD 2013), the 2008 LORS
assumed a limit on Lake Okeechobee releases to the STAs of 60,000 ac-
ft per year. This limit was based on the high TP concentrations in Lake
Okeechobee following the 2004-2005 hurricanes. Lake outflow
concentrations have declined considerably since that time. Using the
updated TP concentrations will allow higher flows to the STAs and not
exceed the TP loads from the Lake as assumed in the 2012 Restoration
Strategies.
iv. Suggest establishing a minimum release target of 250,000 acre feet to
the STAs/WCAs before sending any Lake regulatory releases to the
estuaries.
6. Consider assisting the SFWMD obtain relief from STA operational restrictions imposed
by the federal Migratory Bird Treaty Act of 1918. STA operations protecting certain
ground nesting migratory birds limited the amount of Lake releases that could be sent
to the STAs (i.e., black necked stilts, which are neither threatened nor endangered)
(SFWMD 2013).
7. Consider beginning the conceptual engineering design of new discharge structures along
the south rim of Lake Okeechobee. A physical constraint to moving more water south is
insufficient discharge capacity to the south, and new discharge structures along the
south rim of Lake Okeechobee ideally would be large enough to increase the capacity to
send water to the south up to the combined capacity of the structures that send Lake
water to the east and west estuaries. This conceptual design could identify any
necessary expansion of the EAA Canals and treatment areas necessary to convey and
treat Lake regulatory releases that are currently sent to the St. Lucie and
Caloosahatchee Rivers and estuaries. It is my understanding that the Corps of Engineers
began investigating this “third outlet” in the early 1950s, and in March 1955 published a
Draft report: Partial Definitive Project Report, Central and South Florida Flood Control
and other projects. Serial 19, March 28, 1955. District Office, U.S. Army Corps of
Engineers, Jacksonville, Fla.
9
C. When The LORS Guidance Is To Send “Up To Maximum Practicable Releases To The WCAs”
8. Consider evaluating ways to reduce inflows to Lake Okeechobee before and during
periods when releases are being made to the estuaries. Alternatives could include
a. temporarily raising the normal operating levels in canals adjacent to the Lake.
b. temporarily reducing the use of pump stations S-2 and S-3 in the EAA, and
instead sending this runoff to the south. During June 2013 as harmful Lake
releases were being made to the St. Lucie River and Caloosahatchee River, over 8
billion gallons of water from the EAA were pumped into Lake Okeechobee
through pump station S-2. The pumping capacity within the EAA to drain excess
rainfall to the south is greater now than before the STAs were constructed due
to the additional STA inflow pump stations. In addition, approximately 50,000
acres of land has been removed from the regional drainage area, having been
converted to STAs and the soon-to-be-constructed EAA Flow Equalization Basin.
c. temporarily reducing the use of Culvert 10A on the L-8 Canal, and instead
sending this runoff to the south. During the summer of 2013 as harmful Lake
releases were being made to the estuaries, over 8 billion gallons of stormwater
runoff from the L-8 Canal Basin were sent to the Lake through LC10A.
9. Consider requesting the SFWMD to move more water to the WCAs during periods of
Lake regulatory releases, without sacrificing public health and safety.
a. Stormwater Treatment Areas. Although available STA treatment acreage has
increased dramatically – 15-fold since 1995 (3,815 acres in 1995 to 57,000 acres
today) – the volume of Lake releases sent through the EAA to the STAs and WCAs
has decreased significantly (Figure 2). 2013 flows to the STAs are summarized
above in Table 2. During 2013 several STA flow-ways were not used for
treatment of Lake releases, for example, the Eastern Flow-way of STA-1E
(approximately 1,100 acres) has been off-line for over a year due to structure
repairs by the U.S. Army Corps of Engineers. In addition, the 13,700-acre STA-
5/6 was not used to treat Lake releases, despite available infrastructure to send
the Lake releases to the STA, available treatment capacity and a need for water
supply in the NW corner of WCA-3A directly downstream of STA-5/6. A fear
expressed by SFWMD staff is that sending Lake water to the STAs will overload
them and it will take years for the treatment performance to recover. They
point to an event between July 29, 2002 and February 14, 2003, when STA-1W
received approximately 270,000 acre feet of Lake releases, and approximately 18
months were required for the STA to return to normal performance. That
volume equates to an average hydraulic load of 6.0 ft per month, where the
hydraulic load is equal to the flow divided by the area, e.g., 270,000 acre feet /
201 days / 6,700 acres = 6 ft per month (6.1 cm/day). By comparison, between
January and October 21, 2013, approximately 57,000 acres of STAs received
approximately 107,000 acre feet of Lake releases. This equates to an average
hydraulic load of 0.21 feet per month, or 1/30th of the 2002-2003 loading.
10
Figure 2. Summary of annual Lake releases to the WCAs.
Specific suggestions for each STA are provided in Table 3 below. Adverse impacts to these
areas can be minimized by sending Lake releases to all these areas as opposed to just one or
two of them. Under ideal conditions south of the Lake, if STA-1E, STA-1W, STA-2 and STA-3/4
each receive an average hydraulic load of 1.5 feet of Lake releases per month (e.g., 10,050 acre
feet to the 6,700-acre STA-1W), and STA-5/6 receives an average of 0.5 feet of per month, over
70,000 acre feet per month of Lake releases could be sent south. The SFWMD might also
consider declaring a state of emergency to enable greater flexibility to send more water sent to
south, e.g., to the WCAs and points further south, and to designate Lake regulatory releases as
high of priority as flood control discharges. In light of the public health and safety issues
surrounding Lake regulatory releases, normal STA operations might be suspended.
Table 3. Suggested Lake Releases Targets to the STAs

STA Target Flow (AF/mo) Maximum Flow (AF/mo)
STA Area 1.5 3.0
acres cm/day cm/day
STA-1E 5,000 7,500 15,000
STA-1W 6,500 9,750 19,500
STA-2 15,500 23,250 46,500
STA-3/4 16,300 24,450 48,900
STA-5/6 13,700 6,850 6,850
Total 57,000 71,800 136,750
Notes. 1. 1.5 cm/day = 1.5 ft/month; 3.0 cm/day = 3.0 ft/month

Target and maximum flow to STA-5/6 set at 0.5 cm/day (0.5 ft/mo).
STA-3/4 – This 16,500-acre treatment area is located between the EAA and WCA-3A. As
mentioned above, the 1994 Conceptual Design for the STAs contemplated sending an annual
average of more than 250,000 AF of Lake regulatory releases to STA-3/4. This equates to an
average hydraulic load of 1.3 ft per month over the course of 12 months. Since 2008 actual
volumes of Lake water that have been sent to the STAs have averaged less than 20,000 AF/yr.
12
STA-5/6 – Canals currently exist to convey Lake releases to STA-5/6 and existing water supply
pumps can be supplemented with more pumps to increase the capability to capture and treat
Lake water in STA-5/6. The north portion of WCA-3A routinely needs supplemental water which
can be provided by STA-5/6 treated discharges. With the expansion to over 13,000 acres,
dryout of the treatment area is often cited as a concern for long-term phosphorus removal
performance. When a treatment cell dries out, the soil is exposed to air and a spike in
phosphorus concentration often occurs upon rewetting. Close synchronization with the
operation of the STA-3/4 and Rotenberger structures will be required to ensure the most
effective operation of STA-5/6 to treat Lake releases. The divide structure G-373 could be
opened, the STA-5/6 water supply pumps (G-507, G-349B, G-350B and G-509) could be turned
on and the opening of G-411 could be coordinated with pump operations. The existing water
supply pumps could be supplemented with temporary pumps to increase the volume of Lake
water treated in the STA, if needed. No Lake releases were sent to STA-5/6 in 2013 despite low
water depths in the spring and early summer. STA-5/6 in part discharges into WCA-3A, which
was below its regulation schedule much of the 2013 event. The north portion of WCA-3A
needed supplemental water, which could have been provided by STA-5/6 discharges.
b. Wildlife Management Areas (WMAs)

i. Holey Land Wildlife Management Area is a 35,350-acre area located
between the EAA and WCA-3A. The Holey Land hydroperiod
restoration project was designed for flow-through operation, however,
it has largely been operated as a stagnant system, supplied with water
from rainfall alone. The primary inflow pumps (G-200A) were
destroyed in the 2004 hurricanes and not repaired, hence temporary
pump/culverts would be needed to send Lake releases to the Holey
Land through a connection to the STA-3/4 inflow pump station (G-372).
The Holey Land outlet structures (G-204, G-205 and G-206) could be
operated to effect a flow-through operation that minimizes the
increase in water depth. During 2013, there are no records of Lake
water being sent to the Holey Land.
Suggestion: During periods of Lake regulatory releases, suggest
sending Lake regulatory to the WMA with a target hydraulic
load of 0.5 cm per day (0.5 ft per month). Note that this amount
would not raise water levels by 0.5 ft since the load would be
spread over 30 days. If applied evenly this would amount to an
inflow of 0.5 cm per day, and the actual rise in water depth would
be a function of the outlet capacity, rainfall and local hydrology. If
a target hydraulic load of 0.5 cm per day could be achieved,
approximately 17,500 acre feet of Lake releases could be treated
in this WMA each month.
13
ii. Rotenberger Wildlife Management Area is a 28,760-acre area located

between the EAA and WCA-3A. The Rotenberger WMA currently
receives treated discharges from STA-5/6 and STA-3/4, and has
discharged water containing extremely low phosphorus concentrations
(13 parts per billion), better than any STA. In prior years, Lake water
has been sent to the Rotenberger, but during the 2013 event, no Lake
releases were sent to Rotenberger. The infrastructure (canal and
pumps) exists today to send additional Lake releases to the
Rotenberger WMA through structure G-373 and the STA-5/6 discharge
canal. The Rotenberger inflow pump (G-410) has a capacity of 240
cubic feet per second. The Rotenberger outlet structures (G-302s)
could be operated to effect a flow-through operation that minimizes
the increase in water depth. Close synchronization with the operation
of the STA-3/4 and STA-5/6 structures will be required to ensure the
most effective operation of Rotenberger to treat Lake releases. Pump
station G-410 has historically been used to deliver Lake water to the
Rotenberger Wildlife Management Area, but during 2013, records
indicate no Lake water was sent to the Rotenberger.
Suggestion: During periods of Lake regulatory releases, suggest
sending Lake regulatory to the WMA with a target hydraulic
load of 0.5 cm per day (0.5 ft per month). If a target hydraulic
load of 0.5 cm per day could be achieved, approximately 14,000
acre feet of Lake releases could be treated in this WMA each
month.
c. The Water Conservation Areas. These lands include, but are not limited to the
following.
i. Water Conservation Area 1 (approx. 140,000 acres)
ii. Water Conservation Area 2A and 2B (approx. 130,000 acres)
iii. Water Conservation Area 3A and 3B (approx. 575,000 acres)
The SFWMD’s report on Lake releases included the following:

“Lake Okeechobee regulatory discharges to the WCAs are
constrained by water levels in the conservation areas. Specifically,
federal regulations contained in the Water Control Plan prohibit
lake regulatory releases to the WCAs when WCA water levels
exceed their respective regulation schedules. Therefore, lake
regulatory discharges are not made to the WCAs if WCA stages
exceed their regulation schedules.”
However, EAA stormwater can be sent to the WCAs when WCA

water levels exceed their respective regulation schedules.
14
Suggest considering revising any Water Control Plan that prohibits

sending Lake regulatory releases into the WCAs during emergency
conditions such as during operations that reduce the risk of failure
of the Herbert Hoover Dike. Lake regulatory releases intended to
reduce the risk of failure of the Herbert Hoover Dike (which would
flood the EAA) should be treated with the same priority for public
health and safety as flood protection for the same region.
2013 Operations. Despite the Lake Okeechobee Regulation Schedule

(LORS) direction to send “up to maximum practicable releases to the
WCAs” and despite significant available capacity in the WCAs before the
May 8 commencement of regulatory releases to the estuaries, Lake
releases to the WCAs were not maximized (Table 4).
Suggestions: Suggest establishing a minimum release target of
250,000 acre feet to the WCAs before sending any Lake regulatory
releases to the estuaries. During Lake regulatory releases to the
estuaries, consider non-discretionary Lake releases to the WCAs.
Also, to the extent that Water Control Plans prohibit Lake releases
to the WCAs when WCA water levels exceed their respective
regulation schedules, suggest revising the Plans to treat Lake
releases to the WCAs with the same priority for public health and
safety as flood protection for the same region.
d. Everglades National Park.

Suggestion: Continue working with the SFWMD to remove all flow
restrictions from WCA-3A and WCA-3B into the Park.
15
Table 4. Summary of 2013 LORS guidance for WCA releases.

Lake Lake Lake Total Lake
WCA-1 WCA-2A WCA-3A Total WCA
releases releases releases releases
Tuesday 2008 LORS Part C Release Available Available Available Available
Week sent to sent to sent to sent to
Date Guidance Capacity Capacity Capacity Capacity
WCA-1 WCA-2A WCA-3A WCAs
acre feet acre feet acre feet acre feet acre feet acre feet acre feet acre feet
Up to Maximum
1 1-Jan 60,072 0 109,332 169,404 317 182 71 570
Practicable to WCAs
Up to Maximum
2 8-Jan 56,446 0 120,866 177,312 311 608 975 1,893
Practicable to WCAs
Up to Maximum
3 15-Jan 52,820 0 137,369 190,189 242 1,565 5,543 7,350
Practicable to WCAs
Up to Maximum
4 22-Jan 47,804 0 148,903 196,707 196 2,234 4,615 7,045
Practicable to WCAs
Up to Maximum
5 29-Jan 53,908 0 165,407 219,314 216 5,153 2,734 8,102
Practicable to WCAs
Up to Maximum
6 5-Feb 53,062 0 191,849 244,911 1,828 955 2,664 5,447
Practicable to WCAs
Up to Maximum
7 12-Feb 56,386 0 213,322 269,708 1,078 352 1,273 2,703
Practicable to WCAs
Up to Maximum
8 19-Feb 9,670 0 204,978 214,647 981 0 0 981
Practicable to WCAs
Up to Maximum
9 26-Feb 0 0 216,512 216,512 491 0 0 491
Practicable to WCAs
10 5-Mar No releases to WCAs 0 0 223,076 223,076 213 71 0 284
11 12-Mar No releases to WCAs 12,691 0 249,518 262,210 0 2,234 1,024 3,258
12 19-Mar No releases to WCAs 4,895 0 275,961 280,856 555 0 0 556
13 26-Mar No releases to WCAs 0 0 302,404 302,404 0 91 0 91
14 2-Apr No releases to WCAs 10,153 0 338,786 348,939 544 302 196 1,041
Up to Maximum
15 9-Apr 967 0 285,714 286,681 137 979 86 1,202
Practicable to WCAs
16 16-Apr No releases to WCAs 0 0 247,552 247,552 22 0 0 22
Up to Maximum
19 7-May 0 0 53,550 53,550 0 46 166 212
Practicable to WCAs
Up to Maximum
20 14-May 91,740 0 45,205 136,945 572 491 1,772 2,835
Practicable to WCAs
Up to Maximum
21 21-May 65,330 0 36,860 102,190 0 408 1,472 1,879
Practicable to WCAs
Up to Maximum
22 28-May 31,970 0 33,485 65,455 0 26 94 120
Practicable to WCAs
Up to Maximum
23 4-Jun 0 0 0 0 0 0 0 0
Practicable to WCAs
Up to Maximum
24 11-Jun 0 0 0 0 0 0 0 0
Practicable to WCAs
Up to Maximum
25 18-Jun 0 0 0 0 0 0 0 0
Practicable to WCAs
Up to Maximum
26 25-Jun 68,110 0 0 68,110 0 0 0 0
Practicable to WCAs
Up to Maximum
27 2-Jul 0 0 0 0 0 0 0 0
Practicable to WCAs
Up to Maximum
28 9-Jul 3,083 0 0 3,083 11 0 0 11
Practicable to WCAs
Up to Maximum
29 16-Jul 0 0 0 0 0 0 0 1
Practicable to WCAs
Maximum Practicable to
30 23-Jul 0 0 0 0 0 0 0 0
WCAs
31 30-Jul 0 0 0 0 0 0 0 0
WCAs
32 6-Aug 19,513 0 0 19,513 499 0 0 499
WCAs
33 13-Aug 84,434 0 0 84,434 6,146 334 269 6,749
WCAs
34 20-Aug 129,894 0 0 129,894 3,214 103 373 3,691
WCAs
Up to Maximum
35 27-Aug 115,584 0 0 115,584 4,550 0 0 4,550
Practicable to WCAs
Up to Maximum
36 3-Sep 115,174 0 0 115,174 5,384 0 0 5,384
Practicable to WCAs
Up to Maximum
37 10-Sep 116,154 0 0 116,154 2,296 0 0 2,296
Practicable to WCAs
Up to Maximum
38 17-Sep 128,254 0 0 128,254 2,483 0 0 2,483
Practicable to WCAs
Up to Maximum
39 24-Sep 120,467 0 0 120,467 2,070 0 0 2,070
Practicable to WCAs
Up to Maximum
40 1-Oct 80,620 0 0 80,620 10,711 35 127 10,873
Practicable to WCAs
Up to Maximum
41 8-Oct 96,605 0 0 96,605 11,005 0 0 11,005
Practicable to WCAs
Up to Maximum
42 15-Oct 75,060 0 0 75,060 10,752 191 200 11,143
Practicable to WCAs
Up to Maximum
43 22-Oct 97,995 0 0 97,995 2,204 4,873 421 7,498
Practicable to WCAs
Up to Maximum
44 29-Oct 90,350 0 0 90,350 592 1,308 252 2,152
Practicable to WCAs
Up to Maximum
45 5-Nov 109,810 0 9,939 119,749 0 470 1,688 2,157
Practicable to WCAs
Up to Maximum
46 12-Nov 111,200 0 54,666 165,866 14 139 437 590
Practicable to WCAs
Up to Maximum
47 19-Nov 120,930 0 109,332 230,262 1,090 2,215 1,819 5,123
Practicable to WCAs
Up to Maximum
48 26-Nov 125,100 0 79,514 204,614 2,603 1,109 1,491 5,202
Practicable to WCAs
Up to Maximum
49 3-Dec 109,475 0 94,423 203,899 1,359 646 2,333 4,339
Practicable to WCAs
Up to Maximum
50 10-Dec 105,915 0 144,120 250,034 3,369 499 1,800 5,668
Practicable to WCAs
Up to Maximum
51 17-Dec 100,964 0 188,846 289,810 2,064 604 2,181 4,849
Practicable to WCAs
Up to Maximum
52 24-Dec 98,793 0 228,603 327,396 0 274 989 1,264
Practicable to WCAs
53 31-Dec No releases to WCAs 96,622 0 233,573 330,195 0 116 419 535
16
D. WHEN REGULATORY RELEASES TO THE ESTUARIES ARE UNAVOIDABLE
10. When future regulatory releases to the estuaries are unavoidable, consider supporting
the declaration of a State of Emergency for the coastal regions that will be adversely
affected by the discharge of Lake Okeechobee regulatory releases, and for the region
south of the Lake potentially impacted by a failure of the Herbert Hoover Dike.
a. In 2013, the Department of Health issued numerous public health advisories for
the St. Lucie River and Indian River Lagoon, and hopefully the declaration of an
Emergency would increase support for improving public health and safety.
b. In addition to potentially providing financial support to individuals and
businesses adversely affected by the discharges, declaring a State of Emergency
may give agencies and regions additional flexibility to deal with this crisis.
c. Consider requesting the SFWMD to declare a State of Emergency with regard to
their Works of the District permits (40E-61 and 40E-63), and their operation of
the Stormwater Treatment Areas (STAs), the Water Conservation Areas (WCAs)
and other public lands. This hopefully will give them more flexibility to manage
water levels in the Kissimmee-Okeechobee-Everglades system and reduce
discharges to the coastal estuaries.
11. In light of the public health and safety risk associated with blue green algae, consider
requesting the SFWMD to notify the public health departments when blue green algae
are observed in waters discharged from Lake Okeechobee to the St. Lucie Canal and the
C-43 Canal. According to the SFWMD’s DBHYDRO database, blue green algae were
observed in water samples collected by the SFWMD during many sampling events since
regulatory releases began May 8, 2013.
17
APPENDIX O:
DR. GARY GOFORTH WATER YEAR 2018
ANALYSIS
Water Quality Assessment of the St. Lucie River Watershed – Water Year 2018 - DRAFT
Gary Goforth, P.E., Ph.D. 1
Quis custodiet ipsos custodes? (Who watches the Watchers?) 2
Key Findings:
1. Over the last water year (May 2017 – April 2018), the surface water entering the St. Lucie
River and Estuary (SLRE) in general was of poor water quality. The best water quality
entering the SLRE was from the highly urbanized Tidal Basins. The largest source of
phosphorus, nitrogen and sediment pollution to the SLRE was Lake Okeechobee
discharges. The C-44 Canal Basin contributed poor water quality, and was the only basin
demonstrating a worsening in both nitrogen and phosphorus over the last ten years.
2. It was estimated that stormwater runoff from agricultural land use contributed more flow
and nutrient pollution than any other land use, even contributing more flow and nutrient
pollution than Lake Okeechobee discharges.
3. The annual Basin Management Action Plan (BMAP) progress reports produced by the
Florida Department of Environmental Protection continue to indicate water quality
conditions in the tributaries of the SLRE are better than they actually are. Examples of
flaws in the BMAP assessment process include the omission of Lake Okeechobee pollution
loads, the use of simulated data instead of observed data, the inability to account for
hydrologic variability, and the inability to assess individually each of the major basins
contributing to the SLRE.
4. An alternative to the assessment approach presented in the BMAP progress reports was
developed and used to evaluate water quality conditions of major inflows to the SLRE and
to assess progress towards achieving the Total Maximum Daily Load (TMDL) load reduction
goals. This alternative approach uses observed data, includes Lake discharges, accounts for
hydrologic variability, and is applied to each of the major basins contributing pollution
loads to the SLRE. For WY2018, observed 10-yr average nitrogen loads to the SLRE
exceeded the Phase 1 BMAP target loads (adjusted for hydrologic variability) by 37
percent. Observed 10-yr average phosphorus loads exceeded the Phase 1 BMAP target
loads (adjusted for hydrologic variability) by 19 percent.
5. Recommendations for improving the BMAP progress reports are offered.
1
Gary Goforth, LLC Stuart, Florida 34997
2
Satires of Juvenal, late 1st and early 2nd centuries A.D.
1
EXECUTIVE SUMMARY
The St. Lucie River and Estuary (SLRE), located along Florida’s southeast coast, is one of the most
biologically-diverse estuaries in the nation, and is home to more than three dozen threatened
and endangered species (SFWMD et al. 2009). Unfortunately, the SLRE is also one of the most
ecologically-stressed river and estuarine systems in Florida. For more than 90 years, the regions’
environmental and economic health has been sacrificed by state and federal agencies through
diversion of polluted water from Lake Okeechobee in order to provide flood protection and
irrigation benefits to farms and communities south of Lake Okeechobee. This “tragedy of the
commons” has played out in national and international media during 2016, as more than 200
billion gallons of polluted Lake Okeechobee overflow containing tons of nutrients, sediment, toxic
blue-green algae, poorly oxygenated and low salinity water was diverted from its historical
southerly flow pattern easterly to a major tributary to the Indian River Lagoon 3. Unfortunately,
this tragedy was repeated during the fall of 2017, and again in the summer of 2018.
The spatial extent of the SLRE watershed has more than doubled in the last century as major
agricultural canals were constructed (SFWMD 2002). These canals now contribute significant
loadings of nutrient and unknown quantities of pesticides (FDEP 2008, FDEP 2013). The
dominant land uses in the watershed are agriculture (55%), natural areas (26%), urban areas and
other (19%) (FDEP 2013, FDEP 2015, SFWMD 2016). To protect the designated uses of the SLRE,
the Florida Department of Environmental Protection (FDEP) established TMDLs for total
phosphorus (TP), total nitrogen (TN) and biological oxygen demand (BOD), with concentration
endpoints of 81 parts per billion (ppb) for TP, 720 ppb for TN and 2,000 ppb for BOD (FDEP 2008).
A BMAP was developed in 2013 which established target load reductions for TP and TN.
Unfortunately, the BMAP did not include load reduction targets or projects for Lake Okeechobee
discharges, and the FDEP intentionally ignores pollution loading from the lake in their annual
progress report for the BMAP (FDEP 2016). Further, the BMAP did not use readily available
monitoring data when establishing “Starting Loads,” but instead relied on simulated flows and
loads. The FDEP Starting Loads underestimated loads from basins contributing to the SLRE by up
to 36 percent compared to individual basin monitoring data, creating not only flawed reference
conditions but incorrect load reduction targets and subsequently flawed assessments of annual
progress towards achieving the TMDL. As an example of this flawed assessment, despite an
admitted lack of field verification and monitoring data, all discharges from agricultural land use
are assumed to achieve 100 percent of their load reduction goals once a Notice of Intent has
been signed – clearly an optimistic assumption (FDEP 2016). When presented with this
information in 2015, FDEP’s response was that they plan to start using actual data in 2017 (FDEP
2015). However, the 2017 report did not, and again used prior flawed methods.
3
A similar environmental catastrophe has occurred in the Caloosahatchee Estuary on Florida’s west coast, which has
received even greater volumes of polluted Lake Okeechobee overflow.
2
In lieu of relying on FDEP’s annual progress report, levels of TP, TN and total suspended sediment
(TSS) from Lake Okeechobee and from the SLRE watershed were summarized for the most recent
water year (May 1, 2017 to April 30, 2018, “WY2018”), and are presented in Table ES-1.
As an alternative to the FDEP BMAP targets, hydrologically-adjusted performance measures were

developed to assess progress towards achieving the TMDL for the SLRE. These performance
measures were developed in a manner similar to those utilized in the predominantly agricultural
areas south of the Lake (SFWMD 2016). These performance measures utilized the same
concentration endpoints for TP and TN as the BMAP and were used to assess performance of the
source basins contributing to the SLRE. The results are summarized in Figures ES-1 and ES-2 and
indicate that nitrogen and phosphorus loads increased above the Starting Period levels. By
contrast, FDEP reported that both nitrogen and phosphorus loads decreased from the Starting
Period (FDEP 2018). For the Tidal Basins, reliable flow measurements are not available, and the
flow estimate is based on uncalibrated computer modeling; to avoid propagating this uncertainty
into the assessment approach, the assessment for the Tidal Basins was based on observed
concentrations measured at 29 stations.
Table ES-1. Summary of WY2018 Flows and Loads from SLRE Watershed and Lake Okeechobee.
Entire C-44 Lake Total C-44 Canal to
Water Year 2018 (May 1, 2017 - April 30, 2018) C-23 Canal C-24 Canal Ten Mile Creek Tidal Basins
Canal Basin Okeechobee Watershed SLRE
Basin Discharge, acre feet 219,620 216,954 169,483 152,861 137,763 585,612 1,482,292 79,843
Percent of Total SLRE Watershed 15% 15% 11% 10% 9% 40% 100%
Total Nitrogen load, pounds 1,113,530 979,900 1,177,370 499,901 341,316 2,988,912 7,100,929 560,435
Total Nitrogen concentration, ppb 1,864 1,661 2,555 1,203 911 1,877 1,762 2,581
Total Phosphorus load, pounds 267,633 248,259 218,531 154,954 42,337 358,056 1,289,770 146,320
Total Phosphorus concentration, ppb 448 421 474 373 113 225 320 674
Total Suspended Solids load, pounds 1,482,646 3,095,210 7,455,764 2,174,217 1,595,310 145,531,979 161,335,126 6,818,189
Total Suspended Solids concentration, ppb 2,483 5,246 16,177 5,230 4,258 91,386 40,025 31,403
Note: Tidal Basins flows and loads are estimated – not measured.
Notes:
1. The C-44 Canal Basin includes the smaller S-153 drainage area, which discharges into the C-44 on the west
end near Lake Okeechobee. The FDEP and SFWMD refer to this combined area as the “C-44/S-153 Basin.”
2. “Entire C-44 Canal Basin” includes approximately 89,640 acre feet that was discharged to Lake Okeechobee.
3. Tidal Basins flows and loads are estimated – not measured.
The largest single source of total nitrogen, total phosphorus and sediment load to the SLRE was
Lake Okeechobee discharges. In addition, total phosphorus concentrations in Lake Okeechobee
discharges to the SLRE remained more than 5 times the lake’s TMDL in-lake target concentration
of 40 parts per billion (ppb). Based on data from the South Florida Water Management District
(SFWMD) reported that the 5-yr average annual phosphorus loading to the lake from surrounding
watersheds was more than 5 times the Lake’s TMDL of 105 metric tons, yet staff acknowledged
the agency does not enforce permits that set numeric limits on phosphorus discharges to the
3
lake 4 (SFWMD 2016). Unfortunately, despite the continued and well-publicized pollution of the
lake, the Florida legislature in 2016 enacted a water bill that pushed back deadlines for achieving
the lake’s TMDL by decades (Ch. 2016-1).
Figure ES-1. Progress Towards The St. Lucie River And Estuary TN TMDL For Water Year 2018
(Note: FDEP estimated the loading as 1,897,520 pounds, 63 percent less than estimated herein;
the primary reason for the discrepancy is that FDEP ignores the load from Lake Okeechobee)
Figure ES-2. Progress Towards The St. Lucie River And Estuary TP TMDL For Water Year 2018
(Note: FDEP estimated the loading as 457,816 pounds, 20 percent less than estimated herein;
the primary reason for the discrepancy is that FDEP ignores the load from Lake Okeechobee)
4
Works of the District permits (40E-61, F.A.C.)
4
The best water quality entering the SLRE during WY2018 was observed in the highly urbanized
Tidal Basins, with concentrations of 113 ppb and 911 ppb for TP and TN, respectively. Each of the
remaining source basins, except the C-44 Canal Basin 5, exhibited a slight improvement in nutrient
levels compared to their base periods, however, collectively these WY2018 loads did not achieve
the alternative BMAP Phase 1 load target (Figures ES-1 and ES-2). The C-23 and Tidal Basins met
the alternative BMAP Phase 1 target for TP, while the C-23, C-24 and Tidal Basins met the
alternative BMAP Phase 1 target for TN. The predominantly agricultural C-44 Canal Basin
exhibited poor nutrient conditions, and in fact, continued a trend of deteriorating nutrient
conditions compared to its 1996-2005 base period. As a whole, the water quality entering the
SLRE remains poor, with a slight deterioration compared to the 1996-2005 period (Table ES-2).
The assessment described herein highlights areas of potential improvement in the state’s BMAP
annual assessment and reporting program, and suggests that implementation of regional nutrient
control programs has had variable degrees of success.
Table ES-2. Summary of Water Quality Conditions Entering the St Lucie River and Estuary
Total Nitogen Total Phosphorus

Source Basin
WY2018 Status 10-yr Trend WY2018 Status 10-yr Trend
C-23 Canal Poor Improving Poor Improving
C-44 Canal Poor Worsening Poor Worsening
Ten Mile Creek Fair Improving Poor Improving
Tidal Basins Fair Improving Poor Improving
Lake Okeechobee Poor Improving Poor Worsening
Total Inflow Poor Worsening Poor Worsening
Notes:
“Good” indicates the water year achieved the TMDL.
“Fair” indicates the water year exceeded the TMDL by less than 33%.
“Poor” indicates the water year exceeded the TMDL by more than 33%.
The Tidal Basins and Lake Okeechobee assessment were based on observed concentrations; other source basin
assessments were based on observed loads compared to hydrologically-adjusted base period loads.
5
The C-44 Canal Basin is also known as the “C-44/S-153 Basin.”
5
1. BACKGROUND
Florida is blessed with abundant surface and groundwater resources, however, decades of
excessive water use consumption and weakening environmental policies have resulted in water
quantity and water quality issues in virtually every major river system in the state (FDEP 2016).
The St. Lucie River and Estuary (SLRE), located along Florida’s southeast coastline, is one of the
most biologically-diverse estuaries in the nation, and is home to more than three dozen
threatened and endangered species (SFWMD et al. 2009). Unfortunately, the SLRE is also one of
the most ecologically-stressed river and estuarine systems in Florida.
At the turn of the 20th century, the city of Stuart was not yet incorporated, the area south of the
River was part of Palm Beach County, and the St. Lucie River flowed deep and clear (Lyons 1975).
The estuary was renowned for its inshore tarpon fishing, and was known as the “Tarpon Fishing
Capital of the World.” The watersheds of the North Fork and South Fork of the River extended a
few miles west to a ridge that separated Allapattah Slough, Cane Slough and other areas that
flowed north to the St. John’s River or south to the Loxahatchee River (Figure 1 from SFWMD
2002). In 1913, the State of Florida decided to construct a canal between Lake Okeechobee and
the SLRE to divert overflow water from the lake in order to encourage and enhance agricultural
and community development south of the Lake (Blake 1980). Prior to that time, there was no
natural connection between the Lake and the SLRE. The first discharge of Lake water into the
SLRE is reported to have occurred on June 13, 1923 (Osborn 2012). It wasn’t long before fishing
guides, residents and local governments began to realize the significant environmental
consequences of the destructive Lake releases, and by 1930 Martin County Commissioners
forwarded the first of many requests to terminate the discharges (MBOCC 1930). As tons of
sediment from the Lake muddied the once-clear river and estuary, the inshore tarpon fishing
industry collapsed, and the area re-cast itself as the “Sailfish Capital of the World.” Discharges
from Lake Okeechobee were initially unregulated, however, a series of operating schedules were
eventually put into place by the U.S. Army Corps of Engineers, leading up to today’s Lake
Okeechobee Regulation Schedule 2008 (aka “LORS2008”) (USACE 2016).
For more than 90 years, the regions’ environmental and economic health has been sacrificed by
state and federal agencies through diversion of polluted water from Lake Okeechobee in order to
provide flood protection and irrigation benefits to farms and communities south of Lake
Okeechobee. This “tragedy of the commons” has played out in national and international media
during 2016, as more than 200 billion gallons of polluted Lake Okeechobee overflow containing
tons of nutrients, sediment, toxic blue-green algae and low salinity water was diverted from its
6
historical southerly flow pattern easterly to a major tributary to the Indian River Lagoon 6. In late
2017, more than 190 billion gallons of polluted Lake Okeechobee was discharged to the estuary.
During the 1920s through 1960s, the area of the SLRE watershed more than doubled as large
agricultural drainage canals were constructed by regional drainage districts and the U.S. Army
Corps of Engineers. These canals allowed the waters of Allapattah Slough and Cane Sloughs to be
quickly re-directed to the SLRE, increasing the volume and rapidity of stormwater runoff into the
river and estuary.
Today, nutrient and sediment loads to the SLRE come from Lake Okeechobee, the C-23 Canal
Basin, the C-24 Canal Basin, the C-44 Canal Basin, the Ten Mile Creek Basin and direct runoff from
tidally-influenced portions of the North Fork, South Fork and smaller basins – collectively referred
to as the “Tidal Basins” (Figure 2). In addition, stormwater runoff from the adjacent C-25 Canal
Basin periodically is discharged through the C-24 Canal into the North Fork of the SLRE.
Figure 1. Historical drainage of the St. Lucie River Watershed (SFWMD 2002).
6
A similar environmental catastrophe has occurred in the Caloosahatchee Estuary on Florida’s west coast, which has
received even greater volumes of polluted Lake Okeechobee overflow.
7
Land use. The dominant land use in the SLRE watershed is agriculture, comprising 55 percent of
the contributing area (Table 1). Approximately 26 percent of the watershed is natural (uplands,
wetlands, water, and barren land); while less than 20 percent is urban and built-up area. The
total area for the SLRE excludes the South Coastal basin (approximately 12,000 acres) which does
not drain to the St. Lucie River or Estuary, but to the St. Lucie Inlet area.
Table 1. Characteristics of Basins Contributing to the SLRE (from FDEP 2014, SFWMD 2016)
Natural Urban &
Area Ag Area Ag Area Natural Urban &
Basin Area Other
(acres) (acres) (%) Area (%) Other (%)
(acres) (acres)
C-23 Canal Basin 112,160 84,744 76% 23,706 21% 3,710 3%
C-24 Canal Basin 83,373 67,516 81% 15,701 19% 156 0%
C-44 Canal Basin 132,717 78,351 59% 37,163 28% 17,203 13%
Ten Mile Creek 39,726 32,491 82% 0 0% 7,235 18%
Tidal Basins, composed of the following: 157,840 26,533 17% 59,945 38% 71,362 45%
North Fork, excl. Ten Mile Cr. 92,138 3,968 4% 33,129 36% 55,041 60%
South Fork 50,121 20,120 40% 18,987 38% 11,014 22%
Basin 4-5-6 15,581 2,445 16% 7,830 50% 5,306 34%
Total SLRE Watershed 525,816 289,635 55% 136,516 26% 99,665 19%
2. DATA ANALYSIS
Monitoring data for flow and water quality from the Lake and the agricultural drainage canal
basins are available from the SFWMD public database (DBHYDRO). The periods of record for
flow, water quality and rainfall vary among the contributing basins; for example flow records
extend from April 1931 for discharges from Lake Okeechobee but begin in September 1999 for
the Ten Mile Creek Basin. In general, water quality data for the Lake discharges are available
after 1973 and for the other basins after 1979. For the Tidal Basins, reliable flow data are not
available, however water quality is monitored at 29 stations; see Figure 2 for monitoring
locations.
Daily flow data were downloaded from DBHYDRO for representative stations for each of the
source basins; missing data were filled in using appropriate algorithms. Available water quality
data for TP, TN and TSS were also downloaded from DBHYDRO. Calculations of load followed the
methods used by the SFWMD as described in the 2016 South Florida Environmental Report
(SFWMD 2016). Daily rainfall data were downloaded from DBHYDRO, and precipitation estimates
for each basin were obtained using representative stations and weighting factors as described in
the Draft Technical Support Document: St. Lucie River Watershed Performance Metric
Methodologies (SFWMD et al. 2013). Annual summaries of historical flows and loads for each of
the source basins and Lake Okeechobee are presented in Appendix 1.
8
Figure 2. St. Lucie River Watershed, showing locations of sampling sites (from SFWMD 2016)
Notes: 1. The runoff from the C-25 Basin generally flows east to the Indian River Lagoon, but at times may be directed south into the C-24 Basin. 2. The Tidal
Basins include areas downstream of water control structures and includes the South Fork, Basin 4-5-6, North and South Mid-Estuary, and the North Fork Basin,
excluding the Ten Mile Creek Basin. 3. The South Coastal Basin does not contribute to the St. Lucie River and Estuary, but rather, to the St. Lucie Inlet area.
9
C-44 Canal Basin. Runoff from the C-44 Canal Basin can flow either to the SLRE through the S-
80 structure or to Lake Okeechobee through the S-308 structure, depending on decisions by the
U.S. Army Corps of Engineers (USACE) and the South Florida Water Management District
(SFWMD). Over the long-term (1980-2016), approximately two-thirds of the C-44 Canal Basin
runoff flowed to the SLRE and one-third flowed to the lake. However the annual proportion
varied significantly: from one percent to the SLRE in 2008 (with the balance to the Lake) to 100
percent in 2004. Unfortunately, the FDEP TMDL and BMAP did not account for this annual
variability, and erroneously assumed a constant proportion of “only 76.5% of the runoff in the
C-44/ S-153 sub-basin runoff flows to the St. Lucie Estuary.” (FDEP 2013). This led to the
incorrect decision that “only 76.5% of the C-44/S-153 sub-basin runoff was applied in the St.
Lucie River and Estuary BMAP allocations.” This unnecessary assumption utilized by FDEP
resulted in nutrient TMDL allocations and associated BMAP goals that are too high roughly half
the years and too low the other years. To accurately characterize flow and loads from the C-44
Canal Basin, total basin flows and loads, that is, combining flow to both the SLRE and to the
Lake, were calculated in this present analysis and used to develop basin performance measures.
Lake Okeechobee Pass-through Flows and Loads. The present analysis calculates Lake
Okeechobee discharges through the C-44 Canal, referred to as “pass-through flows,” following
the algorithm used by the SFWMD (SFWMD 2013). However, the SFWMD algorithm for
calculating the remaining C-44 Canal Basin loads generates excessive negative loads and
concentrations, which are physically impossible. Hence a modified algorithm was used in this
analysis that maintained the daily mass balance at the S-80 and S-308 structures and avoided
negative values.
Tidal Basins. The Tidal Basins encompass the area within Basin 4-5-6, South Fork Basin, portions
of North Fork (excluding the Ten Mile Creek Basin), and other small basins; this area
encompasses approximately 158,000 acres. Water quality data were obtained from 29
monitoring stations for the period November 2001 through the current water year from
DBHYDRO. While flow is monitored at some of these locations, the lack of reliable flow data for
all stations prevented calculations of nutrient loads, and hence the water quality assessment
was based on measured concentrations. To ensure data were representative of basin runoff
and not tidal flow, only data collected when positive outflow was observed and when specific
conductance values were below 2,500 µmhos/cm were used in the analysis; for further details
see SFWMD et al. 2013. A single monthly composite concentration for the Tidal Basins was
calculated based on an area weighting of available data during that month. The median
monthly composite concentrations were calculated for each water year as representative of the
Tidal Basins.
10
Water Quality Conditions for Current Year and for the Most Recent 10-yr period
Summaries of the WY2018 nutrient and sediment levels from the SLRE watershed and from
Lake Okeechobee are presented in Table 2 and Figures 4. For the year, approximately 90
percent of the C-44 Canal Basin flows and nutrient loads entered the SLRE while the remainder
entered the lake. Flow and loads were also distributed based on land use 7 (Figure 5). Despite
the large discharges from the Lake during WY2018, stormwater runoff from agricultural lands
represented the single largest source of flow and nutrient loading, accounting for 46 percent
of the flow, 49 percent of the nitrogen loads and 63 percent of the phosphorus loading. By
contrast, runoff from urban and natural areas contributed the smallest amount of pollution
loading, ranging from 5-6 percent.
Table 2. Summary of WY2018 Surface Inflows from SLRE Watershed and Lake Okeechobee.
Entire C-44 Lake Total C-44 Canal to

Water Year 2018 (May 1, 2017 - April 30, 2018) C-23 Canal C-24 Canal Ten Mile Creek Tidal Basins
Canal Basin Okeechobee Watershed SLRE
Basin Discharge, acre feet 219,620 216,954 169,483 152,861 137,763 585,612 1,482,292 79,843
Total Nitrogen load, pounds 1,113,530 979,900 1,177,370 499,901 341,316 2,988,912 7,100,929 560,435
Total Nitrogen concentration, ppb 1,864 1,661 2,555 1,203 911 1,877 1,762 2,581
Total Phosphorus load, pounds 267,633 248,259 218,531 154,954 42,337 358,056 1,289,770 146,320
Total Phosphorus concentration, ppb 448 421 474 373 113 225 320 674
Total Suspended Solids load, pounds 1,482,646 3,095,210 7,455,764 2,174,217 1,595,310 145,531,979 161,335,126 6,818,189
Total Suspended Solids concentration, ppb 2,483 5,246 16,177 5,230 4,258 91,386 40,025 31,403
Average annual flows and loads for each source basin were calculated for the most recent 10-yr
period 8 (Table 3 and Figure 6). Of the multiple basins contributing to the SLRE, Lake
Okeechobee was the largest single source of flow, nitrogen and suspended sediment from the
SLRE Watershed for the most recent 10-yr period. The C-23 Canal Basin was the single largest
source of phosphorus load to the SLRE. The Tidal Basins exhibited the lowest concentration of
nutrients of all sources to the SLRE. Discharges from Lake Okeechobee accounted for more
than three-quarters of the sediment load to the SLRE.
Classified by land use, agricultural runoff was the largest source of flow and nutrient pollution,
contributing 52 percent of the flow, 58 percent of the nitrogen load and 71 percent of the
phosphorus load (Figure 7). By contrast, runoff from urban and natural areas contributed the
7
Land use data from FDEP BMAP spreadsheets. Allocation among different land uses assumed similar load
reductions across land uses within a basin.
8
Monthly surface runoff flows for the Tidal Basin were estimated from SFWMD’s SLE Tidal Basin Lin-Res Model
calibrated to the SLE WaSh Model results. For periods outside the Lin-Res Model simulation period, a regression
2
model was used based on measured monthly rainfall (R =85%). Flows and loads were thus estimated for the Tidal
Basins.
11
smallest amount of pollution loading, ranging from 7-9 percent. Lake Okeechobee discharges
contributed more than 75 percent of the total suspended sediment load.
Table 3. Summary of recent 10-yr average annual flows and loads from the SLRE Watershed
and Lake Okeechobee.
Entire C-44 Lake Total
WY2009-2018 Average C-23 Canal C-24 Canal Ten Mile Creek Tidal Basins
Canal Basin Okeechobee Watershed
Flow, acre feet 107,381 139,810 139,864 104,136 137,317 225,919 854,427
Total Nitrogen load, pounds 500,456 588,706 621,715 319,391 316,801 954,471 3,301,538
Percent of Total SLRE Watershed 15% 18% 19% 10% 10% 29%
Total Nitrogen concentration, ppb 1,714 1,548 1,635 1,128 848 1,554 1,421
Total Phosphorus load, pounds 133,522 128,316 117,225 88,451 36,479 104,246 608,238
Total Phosphorus concentration, ppb 457 337 308 312 98 170 262
Total Suspended Solids load, pounds 1,231,578 1,623,105 1,258,615 1,703,149 1,741,195 27,146,558 34,704,201
Total Suspended Solids concentration, ppb 4,218 4,269 3,309 6,014 4,663 44,187 14,936
See footnote on flow estimation for Tidal Basins.
12
Figure 4. Distribution of WY2018 Flows From the SLRE Watershed by Source Basin.
WY2018 Annual Surface Flow From the SLRE Watershed WY2018 Annual Nitrogen Loads From the SLRE Watershed
C-23 Canal C-23 Canal

15% 16%
Lake Okeechobee Lake Okeechobee
40% C-24 Canal C-24 Canal
15% 42%
14%
Entire C-44 Canal Entire C-44 Canal

Basin Basin
11% 16%
Tidal Basins Ten Mile Creek
9% Ten Mile Creek
10%
7%
Tidal Basins
5%
Flow data for Tidal Basins estimated from SFWMD calculations. Flow data for Tidal Basins estimated from SFWMD calculations.
WY2018 Annual Total Suspended Solids Loads From the SLRE

WY2018 Annual Phosphorus Loads From the SLRE Watershed Watershed
C-24 Canal Entire C-44 Canal

Basin Ten Mile Creek
2%
C-23 Canal 5% 1%
1%
C-23 Canal Tidal Basins
Lake Okeechobee 1%
21%
28%
C-24 Canal
19%
Ten Mile Creek
12% Lake Okeechobee
Entire C-44 Canal
90%
Basin
17%
Tidal Basins
3%
13
Figure 5. Distribution of WY2018 Flows From the SLRE Watershed by Land Use.
WY2018 Annual Surface Flow From the SLRE Watershed WY2018 Annual Nitrogen Loads From the SLRE
Watershed
Lake Lake
Okeechobee Agriculture Agriculture
Okeechobee
39% 46% 42% 49%
Communities Natural
8% 7%
Communities Natural
6% 3%
WY2018 Annual Phosphorus Loads From the SLRE WY2018 Annual Total Suspended Solids Loads From
Watershed the SLRE Watershed
Lake Agriculture
Okeechobee 7%
28%
Natural
Agriculture 2%
63% Lake
Okeechobee Communities
90% 1%
Communities
5%
Natural
4%
14
Figure 6. Distribution of 10-yr Average Flows From the SLRE Watershed by Source Basin.
WY2009-2018 Average Annual Surface Flow From the SLRE Watershed WY2009-2018 Average Annual Nitrogen Loads From the SLRE Watershed
C-23 Canal
Lake Okeechobee 13%
C-23 Canal
27% C-24 Canal 15%
Lake Okeechobee
16% 29%
C-24 Canal
18%
Tidal Basins
16% Entire C-44 Canal Tidal Basins
Basin 9% Entire C-44 Canal
Ten Mile Creek 16% Basin
12% Ten Mile Creek
10% 19%
Flow data for Ten Mile Creek prior to July 1999 estimated from adjacent stations. Flow data for Tidal Flow and water quality data for Ten Mile Creek prior to July 1999 estimated from adjacent stations.
Basins estimated from SFWMD calculations. Flow data for Tidal Basins estimated from SFWMD calculations.
WY2009-2018 Average Annual Total Suspended Solids Loads From the SLRE
WY2009-2018 Average Annual Phosphorus Loads From the SLRE Watershed Watershed
C-23 Canal C-24 Canal Entire C-44 Canal

Tidal Basins 5% Basin
Lake Okeechobee 3%
6% C-23 Canal 4%
17%
22%
Ten Mile Creek
5%
Ten Mile Creek C-24 Canal

15% 21% Tidal Basins
5%
Entire C-44 Canal Lake Okeechobee
Basin 78%
19%
Flow and water quality data for Ten Mile Creek prior to July 1999 estimated from adjacent stations. Flow and water quality data for Ten Mile Creek prior to July 1999 estimated from adjacent stations.
15
Figure 7. Distribution of 10-yr Average From the SLRE Watershed by Land Use.
WY2009-2018 Average Annual Surface Flow From the SLRE WY2009-2018 Average Annual Nitrogen Loads From the
Watershed SLRE Watershed
Lake Lake
Okeechobee Okeechobee
26% 29%
Agriculture Agriculture
52% 58%
Communities Communities
12% 9%
Natural
10%
Natural
4%
WY2009-2018 Average Annual Phosphorus Loads From the WY2009-2018 Average Annual Total Suspended Solids Loads From the SLRE
SLRE Watershed Watershed
Natural
Communities
3%
7%
Communities
Lake Agriculture
Natural 3%
Okeechobee 16%
5%
17%
Lake
Okeechobee
Agriculture
78%
71%
16
Flows and Loads to the St. Lucie River and Estuary
Flows and loads to the St. Lucie River and Estuary (SLRE) are slightly different than the flows
and loads from the SLRE watershed in that a portion of the runoff from the C-44 Canal Basin is
discharged to Lake Okeechobee and not into the SLRE. During WY2018, 53 percent of the C-44
Basin runoff was sent to Lake Okeechobee and the rest sent to the SLRE. The most recent
annual and 10-yr average flows and loads sent to the estuary are presented in Tables 4 and 5
and Appendix 3. Despite the large discharges from the Lake during WY2018, stormwater
runoff from agricultural lands represented the single largest source of flow and nutrient
loading of all the sources to the SLRE, accounting for 51 percent of the flow, 57 percent of the
nitrogen loads and 70 percent of the phosphorus loading. By contrast, runoff from urban and
natural areas contributed the smallest amount of pollution loading, ranging from 7-8 percent.
Lake Okeechobee discharges contributed an astonishing 145 million pounds of suspended
sediment.
Table 4. WY2018 flows and loads to the SLRE.

C-44 Canal to Lake
Water Year 2018 (May 1, 2017 - April 30, 2018) C-23 Canal C-24 Canal Ten Mile Creek Tidal Basins Total to SLRE
SLRE Okeechobee
Flow, acre feet 219,620 216,954 79,843 152,861 137,763 585,612 1,392,652
Total Nitrogen load, pounds 1,113,530 979,900 560,435 499,901 341,316 2,988,912 6,483,994
Total Phosphorus load, pounds 267,633 248,259 146,320 154,954 42,337 358,056 1,217,559
Total Suspended Solids load, pounds 1,482,646 3,095,210 6,818,189 2,174,217 1,595,310 145,531,979 160,697,552
Table 5. Summary of recent 10-yr average annual flows and loads to the SLRE.
C-44 Canal to Lake
WY2009-2018 Average C-23 Canal C-24 Canal Ten Mile Creek Tidal Basins Total to SLRE
SLRE Okeechobee
Flow, acre feet 107,381 139,810 99,939 104,136 137,317 225,919 814,501
Total Nitrogen load, pounds 500,456 588,706 425,464 319,391 316,801 954,471 3,105,287
Total Phosphorus load, pounds 133,522 128,316 93,845 88,451 36,479 104,246 584,858
Total Suspended Solids load, pounds 1,231,578 1,623,105 876,984 1,703,149 1,741,195 27,146,558 34,322,570
17
3. ASSESSMENT OF WATER QUALITY RELATIVE TO TMDL AND BMAP TARGETS
A prior report described the details of development of annual performance measures that
account for hydrologic variability (Goforth 2016). An assessment of the nutrient levels from the
SLRE watershed and Lake Okeechobee was conducted focusing on two aspects:
1. WY2018 water quality conditions; and

2. The most recent 10-yr period compared to the base period (“starting period” in the
BMAP report) and TMDL and BMAP Targets.
WY2018 Water Quality Conditions
Nutrient levels for the most recent water year are compared to base period levels, BMAP Phase
1 goals and TMDL goals in the figures below. The base period loads were adjusted for
hydrologic variability using the regression equations in Goforth (2016).
Total Nitrogen.
Loads. Improvement from the base period loading was observed for all basins except
the C-44 Basin, which discharged 60 percent more nitrogen load than during the base
period, even after adjusting for hydrologic variability (Figure 8). Lake Okeechobee
contributed more than 79 percent more nitrogen than during the base period. The
combined Total Nitrogen load from the C-44 Canal Basin and Lake Okeechobee that
entered the SLRE through the C-44 Canal was 3.5 million pounds – more than 10 times
the TMDL allocation established by FDEP for that entry point into the SLRE.
Concentrations. The Tidal Basins had the lowest TN concentration for WY2017 at 911
ppb. Every other basin and Lake Okeechobee exhibited higher TN concentrations than
were observed during the base period.
18
Figure 8. WY2018 Total Nitrogen Levels Compared to Base Period and Goals
Total Nitrogen Loads Compared to Base Period, TMDL and BMAP Goal , Adjusted for
Hydrologic Variability (Assumes TMDL excludes Lake discharges to SLRE)
Base Period WY2018 BMAP Ph. 1 Goal TMDL Goal
4,000,000
Nitrogen Loads, pounds
3,000,000
2,000,000
1,000,000
0
C-23 Canal C-24 Canal C-44 Canal Ten Mile Creek Lake Okeechobee
Note: C-44 Canal Basin values includes flow to Lake Okeechobee.
Total Nitrogen Concentrations

3,000
2,500
Concentration, ppb
2,000
1,500
1,000
500
0
C-23 Canal C-24 Canal C-44 Canal Ten Mile Creek Tidal Basins Lake
Okeechobee
C-44 Canal Basin values include flow to Lake Okeechobee.
Note: The TMDL and BMAP goals are based on 10-yr averages; they are compared against
annual values in these charts for information purposes only.
Total Phosphorus.
Loads. Improvement from the base period loading was observed for WY2018 for the C-23
Canal, C-24 Canal and Ten Mile Creek basins. The C-44 Canal Basin contributed approximately
139 percent more phosphorus load than during the base period even after adjusting for
hydrologic variability (Figure 9). Lake Okeechobee contributed more than 138 percent more
phosphorus than during the base period. The combined Total Phosphorus load from the C-44
19
Canal Basin and Lake Okeechobee that entered the SLRE through the C-44 Canal was more than
500,000 pounds – more than 17 times the TMDL allocation established by FDEP for that entry
point into the SLRE.
Concentrations. The Tidal Basins had the lowest TP concentration (113 ppb) of the SLRE source
basins during WY2018. No basin achieved the alternative BMAP Phase 1 goal in terms of
concentration reduction.
Figure 9. Total Phosphorus Levels Compared to Base Period and Goals
Total Phosphorus Loads Compared to Base Period, TMDL and BMAP Goal , Adjusted for
Hydrologic Variability (Assumes TMDL excludes Lake discharges to SLRE)
500,000
Phosphorus load, pounds
400,000
300,000
200,000
100,000
0
C-23 Canal C-24 Canal C-44 Canal Ten Mile Creek Lake Okeechobee
Total Phosphorus Concentrations Compared to Base Period, TMDL and BMAP Goal

500
450
400
Concentration, ppb
350
300
250
200
150
100
50
0
Okeechobee
C-44 Canal Basin values includes flow to Lake Okeechobee.
Note: The TMDL and BMAP goals are based on 10-yr averages; they are compared against
annual values in these charts for information purposes only.
20
Assessment of the most recent 10-yr period compared to the Base Period and TMDL and
BMAP Targets.
Assessments of the water quality from the SLRE watershed for the most recent 10-yr period,
adjusted for hydrologic variability using the performance measures described above, are
summarized in Figures 10-11 and discussed below.
Total Nitrogen
Loads. Using the 10-year average annual loads and adjusting for hydrologic variability,
the Phase 1 BMAP TN load reduction goals were met for the C-23 Canal Basin, the C-24
Canal Basin and the Ten Mile Creek Basin, but was not met for the C-44 Canal Basin. 10-
yr trends indicate improving water quality for all basins except the C-44 Canal Basin,
which discharged 7 percent greater loads than during the Base Period.
Concentrations (Tidal Basins and Lake Okeechobee). The 10-yr concentration for the
Tidal Basins achieved the BMAP Phase 1 goal of 7 percent reduction.
Total Phosphorus
Loads. Using the 10-year average annual loads and adjusting for hydrologic variability,
the Phase 1 BMAP phosphorus load reduction goals were met for the C-23 Canal, the C-
24 Canal and the Ten Mile Creek Basins, but was not met for the C-44 Canal Basin. 10-yr
loading trends indicate improving water quality for all basins except the C-44 Canal
Basin, which discharged 31 percent greater load than during the Base Period.
Concentrations (Tidal Basins and Lake Okeechobee). The 10-yr concentration for the
Tidal Basins achieved the BMAP Phase 1 goal of 9 percent reduction. By contrast, the
10-year flow-weighted mean concentration for lake discharges did not met the BMAP
Phase 1 goal, and in fact was 8 percent more than the Base Period. The 10-year trend in
concentrations for the Tidal Basins indicates improving water quality, while the trend for
Lake Okeechobee discharge concentrations indicates a worsening compared to the Base
Period.
21
Figure 10. Comparison of 10-yr Average Annual Nitrogen Loads
10-yr Total Nitrogen Loads Compared to Base Period, TMDL and BMAP Goal , Adjusted
for Hydrologic Variability (Lake discharges not adjusted)
Base Period 10-yr Average BMAP Ph. 1 Goal TMDL Goal
2,000,000
Nitrogen Loads, pounds
1,500,000
1,000,000
500,000
0
C-23 Canal C-24 Canal C-44 Canal Ten Mile Creek Lake Okee (not adj.)
10-yrTotal Nitrogen Concentrations Compared to Base Period, TMDL and BMAP Goal

2,000
1,500
Concentration, ppb
1,000
500
0
Okeechobee
"Sub-Total" includes C-23, C-24, C-44 and Ten Mile Creek. C-44 Canal Basin values include flow to
Lake Okeechobee.
22
Figure 11. Comparison of 10-yr Average Annual Phosphorus Loads
10-yr Total Phosphorus Loads Compared to Base Period, TMDL and BMAP Goal ,
Adjusted for Hydrologic Variability (Lake discharges not adjusted)
Base Period 10-Yr Average BMAP Ph. 1 Goal TMDL Goal
250,000
Phosphorus load, pounds
200,000
150,000
100,000
50,000
0
C-23 Canal C-24 Canal C-44 Canal Ten Mile Creek Lake Okee (not adj.)
10-yr Total Phosphorus Concentrations Compared to Base Period, TMDL and BMAP Goal

500
450
400
Concentration, ppb
350
300
250
200
150
100
50
0
Okeechobee
C-44 Canal Basin values includes flow to Lake Okeechobee.
23
Nitrogen Loading from Septic Tanks. During 2016, several agricultural interests erroneously
stated that the primary source of nutrient loads to the SLRE is the “200,000 septic tanks” that
line Martin County waterways (e.g., saveourfarms.com 2016). In fact, the Florida Department
of Health notes there are less than 1/10th the number of alleged septic tanks in Martin County:
16,172 (FDOH 2015). In addition, a SFWMD Governing Board member erroneously stated that
80 percent of the nutrient loading to the SLRE was due to local septic tanks (Smith 2016).
The fact is that Martin County and the City of Stuart have significantly reduced nutrient loading
to the SLRE from individual and community septage systems and through other local projects.
These include:
• Elimination of seventy (70) wastewater package plants, which annually prevent the discharge
of more than 560,000 pounds of nitrogen and more than 140,000 pounds of phosphorus (Polley
2014).
• Conversion of more than 1,700 septic tanks to centralized sanitary sewers, removing an
estimated 15,400 pounds per year of nitrogen (Fielding 2015).
• Since 2000 Martin County has invested $50+ million in 25 stormwater projects, with
approximately 30,000 pounds per year of nitrogen removal (Fielding 2015).
As a result of these projects, Martin County has exceeded all the nitrogen load reductions
required under the State’s Basin Management Action Plan (BMAP); nonetheless, the County is
actively planning additional septic to sewer conversion projects that will further reduce nutrient
loading to the River (FDEP 2015).
Researchers at Florida Atlantic University (FAU) postulated that malfunctioning septic tanks are
a primary source of nutrient loading to the SLRE (LaPointe and Herren 2016). However,
estimates prepared by consultants to FDEP suggest loadings from septic tanks are about one-
third estimates from FAU, and previous BMAP progress reports have documented that prior
septic-to-sewer conversion projects have been very effective in reducing nutrient loads from
septic systems, particularly in Martin County (Ye and Sun 2013, FDEP 2015). The documented
decline in nutrient concentrations, particularly nitrate-nitrite, in the Tidal Basins shown in
Figure 12 provide further documentation of the effectiveness of these projects. Additional
septic-to-sewer projects in Martin and St. Lucie Counties are underway.
24
Figure 12. Reduction in nitrate-nitrite concentrations in the Tidal Basins.
4. SUMMARY
An overall assessment of the water quality entering the SLRE is summarized in Table 5 and
described below.
Status. The assessment of “status” in each source basin was based on the WY2017 nutrient
levels. For the Tidal Basins and Lake Okeechobee, concentrations were assessed; for all other
source basins loads were assessed.
• “Good” indicates the water year achieved the TMDL;

• “Fair” indicates the water year exceeded the TMDL by less than 33%;
• “Poor” indicates the water year exceeded the TMDL by more than 33%.
25
The overall status of nutrient levels from the SLRE Watershed was “Poor”. The best water
quality was exhibited by the highly urbanized Tidal Basins, which demonstrated a status of
“Fair” for nitrogen.
Table 5. Water Quality Conditions Entering the SLRE.
Total Nitogen Total Phosphorus

Source Basin
WY2018 Status 10-yr Trend WY2018 Status 10-yr Trend
C-44 Canal Poor Worsening Poor Worsening
Ten Mile Creek Fair Improving Poor Improving
Tidal Basins Fair Improving Poor Improving
Lake Okeechobee Poor Improving Poor Worsening
Total Inflow Poor Worsening Poor Worsening
Trend. The assessment of trend in each source basin was based on the most recent 10-yr
average nutrient level compared to its base period. For the Tidal Basins and Lake Okeechobee,
concentrations were assessed; for all other source basins loads were assessed.
• “Improving” indicates the 10-yr average nutrient level was below the base period value,
adjusted for hydrologic variability;
• “Worsening” indicates the 10-yr average nutrient level was above the base period
value, adjusted for hydrologic variability.
All the source basins except the C-44 Canal Basin and Lake Okeechobee exhibited an
“Improving” trend for both nutrients. The C-44 Canal Basin demonstrated a “Worsening” trend
in TP and TN when comparing the recent loads against the base period loads. The overall trend
for both nutrients from the SLRE watershed exhibited a worsening trend for both nitrogen and
phosphorus.
As a result of this assessment, FDEP and other agency staff should identify the successful load
reduction measures being implemented in the better performing basins for potential
application to other basins. In addition, agency staff should prioritize the other basins,
26
particularly the C-44 Canal Basin, for follow-up with landowners to improve nutrient control
measures. During the 2015 stakeholder meeting, SFWMD staff acknowledged that the SFWMD
was not enforcing Works of the District permits (Rule 40E-61) in the C-44 Canal Basin. The
SFWMD is encouraged to start enforcing these permits in the C-44 Canal Basin and other basins
around Lake Okeechobee.
5. COMPARISON TO FDEP BMAP ASSESSMENT APPROACH
The development and application of the performance measures described above offer sharp
contrasts to the methods being utilized by the FDEP in their development of TMDLs and
implementation of the BMAP for the SLRE, including the following.
1. The FDEP TMDL and BMAP method ignored pollution loading from Lake Okeechobee.
2. The FDEP BMAP method used simulated nutrient levels, and ignored abundant available
flow and water quality data for the source basins.
3. The FDEP BMAP method has no annual assessment method.
4. The FDEP BMAP method has no means to account for hydrologic variability.
5. The FDEP BMAP method gives no information on the status and water quality conditions
within individual basins.
6. The FDEP BMAP method ignores the annual variability in the proportion of C-44 Canal
Basin runoff that flows to the SLRE (as opposed to Lake Okeechobee).
In their statewide BMAP Progress Report, FDEP estimated the 10-yr average nitrogen loading
from the St Lucie Basin as 1,897,520 pounds, 63 percent less than estimated herein (FDEP
2018). FDEP also estimated the loading as 457,816 pounds, 20 percent less than estimated
herein (FDEP 2018). The primary reason for these discrepancies is that FDEP ignores the
loading from Lake Okeechobee.
Based on these contrasts, the following recommendations are made.
1. BMAP progress reports should present the observed nutrient and suspended sediment
loads contained in Lake Okeechobee discharges to the St. Lucie River and Estuary.
2. The SLRE BMAP should be revised to identify projects to capture and treat Lake
discharges to the St. Lucie River and Estuary sufficiently to achieve the watershed’s
TMDL and BMAP Goals.
3. BMAP progress reports should utilize available flow and water quality data – and not
simulated values - in assessing and documenting water quality conditions for each basin
and in assessing progress towards achieving the TMDLs and Phase 1 BMAP Goals.
27
4. BMAP Progress Reports should show basin-specific Phase 1 BMAP Goals, measured
loads and load reductions, and document the progress towards achieving the TMDLs
and BMAP Goals.
5. BMAP Progress Reports should present the measured 1996-2005 (i.e., the “Starting
Period”) nutrient loads and concentrations for those basins with measured loads, and
document actual load reductions by comparing measured loads/concentrations with the
1996-2005 values.
6. The TMDL and BMAP should be re-developed for the C-44 Canal Basin as a whole,
recognizing the annual variability in the proportion of C-44 Canal Basin runoff that flows
to the SLRE (as opposed to Lake Okeechobee).
7. The FDEP should recognize the limitations in establishing TMDLs and BMAP goals for the
Tidal Basins and revise the TMDLs and BMAP goals to utilize concentrations: 81 ppb for
TP and 720 ppb for TN.
8. The BMAP method for assessing current water quality conditions and calculating load
reductions should account for hydrologic variability in manner similar to that developed
by the SFWMD for the EAA and C-139 Basins (Rule 40E-63).
9. The BMAP Progress Reports should clearly note that the projected load reductions from
agricultural BMPs have not been field verified, and may overestimate the load
reductions. A similar caveat may be necessary for non-agricultural projects.
10. There should be a balanced set of requirements for municipal and agricultural source
control projects regarding monitoring their effectiveness.
11. BMAP progress reports should document compliance with all nutrient discharge limits of
existing permits issued to landowners in the St. Lucie River and Estuary Basin.
12. BMAP progress reports should use an annual reporting period consistent with the
precedent established by the SFWMD, utilizing a Water Year of May 1 to April 30.
Additional details for each recommendation are described in Goforth 2015.
7. REFRENCES
Blake, N. 1980. Land into Water – Water into Land. University Presses of Florida.
Chapter 2016-1. Florida Water 2016 Bill. Committee Substitute for CS for Senate Bill No. 552.
FDEP 2008. TMDL report: Nutrient and dissolved oxygen TMDL for the St. Lucie Basin.
Tallahassee, FL: Div. of Water Resource Management, Bureau of Watershed Management.
FDEP 2013. BMAP for the St. Lucie River Watershed. Tallahassee, FL: Bureau of Watershed
Restoration, Division of Environmental Assessment and Restoration.
28
FDEP 2015. 2nd Annual Progress Report for the St. Lucie River Watershed. Tallahassee, FL: Bur.
of Watershed Restoration, Div. of Environmental Assessment and Restoration. Dec. 2015.
FDEP 2016. TMDL website. http://www.dep.state.fl.us/water/tmdl/
FDEP 2018. Florida Statewide Annual Report. June 30, 2018.
FDOH 2015. Florida Water Management Inventory. December 4, 2015.
Fielding, E. 2015. Op-ed in Stuart News. Pollution solutions in our own backyard: Septic-to-
sewer conversions. October 13, 2015
Goforth, G. 2015. Comments on the Draft 2015 Progress Report for the St. Lucie River and
Estuary Basin Management Action Plan. Provided to FDEP staff on September 10, 2015.
http://www.garygoforth.net/Comments%20on%20the%20Draft%202015%20Progress%20R
eport%20-%209%2010%202015%20-%20Goforth.pdf
Goforth, G. 2016. Water Quality Assessment of the St. Lucie River Watershed – Water Year
2016. http://www.garygoforth.net/FINAL%20DRAFT%20-
%20Water%20Quality%20Assessment%20of%20the%20SLRW%20-
%20Water%20Year%202016.pdf
LaPointe, B. and Herren, L. 2016. 2015 Martin County Watershed To Reef Septic Study – Final
Report. May 2016.
Lyons, E. 1975. The Last Cracker Barrel.
Martin County Board of County Commissioners 1930. Resolution seeking closure termination
of Lake discharges. December 15, 1930.
Polley, J. 2014. Inter-office Memorandum to Nicki Van Vonno. Subject: Package Wastewater
Plants. August 7, 2014.
Osborn, N. 2012. Oranges And Inlets: An Environmental History Of Florida’s Indian River
Lagoon. Master’s Thesis Florida Atlantic University. August 2012.
Save Our Farms 2016. http://save-our-farms.com/algae/
SFWMD 2002. St. Lucie Minimum Flows and Levels – Draft. May 2002.
SFWMD, FDEP, FDACS, Tetra Tech 2009. St. Lucie River Watershed Protection Plan. January
2009.
29
SFWMD et al. 2013. Draft – Technical Support Document: St. Lucie River Watershed
Performance Metric Methodologies. December 2013.
http://www.garygoforth.net/TSD%20for%20SLRW%20-%2012%2018%202013.pdf
SFWMD 2015. At the BMAP progress meeting in Stuart, staff reported the agency does not
enforce Works (40E-61) of the District permits.
SFWMD 2106. South Florida Environmental Report. March 2016.
Smith, N. 2016. http://www.sunshinestatenews.com/story/groundhog-day-sfwmd
USACE 2016. Draft - Dam Safety Modification Study for Herbert Hoover Dike – Environmental
Impact Statement. Appendix B. June 2016.
Ye, Ming and Huaiwei Sun 2013. Estimation of Nitrogen Load from Removed Septic Systems to
Surface Water Bodies in the City of Port St. Lucie, the City of Stuart, and Martin County.
Prepared for the Florida Department of Environmental Protection. September 2013.
30
APPENDIX 1. HISTORICAL WATER QUALITY DATA
Annual Summary of Lake Okeechobee Discharges to the St Lucie River and Estuary
C-44 Basin Lake FWM FWM FWM
Water Rainfall Discharges TP Load TP Conc TN Load TN Conc TSS Load TSS Conc
Year inches AF pounds µg/L pounds µg/L pounds µg/L
1980 58.25 408,675 118,931 107 2,938,947 2,645 26,470,394 23,818
1981 37.19 63,084 15,027 88 404,360 2,357 825,914 4,814
1982 51.02 4,604 1,233 98 22,878 1,827 5,248,516 419,168
1983 69.84 952,232 483,492 187 4,764,127 1,840 126,107,899 48,700
1984 58.74 177,419 64,988 135 1,007,620 2,088 11,495,846 23,827
1985 47.69 130,776 46,652 131 543,499 1,528 6,219,661 17,489
1986 46.42 6,319 1,620 94 21,993 1,280 2,374,092 138,152
1987 50.94 35,011 8,177 86 194,181 2,040 872,468 9,164
1988 45.94 131,823 48,395 135 625,095 1,744 3,032,427 8,459
1989 41.42 110,357 37,811 126 476,203 1,587 3,544,287 11,810
1990 45.18 40,252 11,750 107 112,164 1,025 5,140,110 46,959
1991 53.84 3,243 1,333 151 12,808 1,452 2,706,317 306,860
1992 41.47 8,894 2,998 124 35,727 1,477 2,271,675 93,924
1993 67.53 125,944 46,909 137 506,692 1,479 3,337,961 9,746
1994 56.85 17,083 7,643 165 68,590 1,477 8,563,721 184,348
1995 66.40 520,631 196,699 139 2,363,613 1,669 23,745,118 16,772
1996 69.99 748,625 221,851 109 2,688,173 1,320 30,256,538 14,862
1997 47.11 74,968 20,169 99 258,973 1,270 3,862,713 18,947
1998 56.93 1,020,158 511,820 184 5,758,028 2,076 98,993,658 35,684
1999 46.22 68,661 24,953 134 267,275 1,431 2,614,497 14,003
2000 60.74 220,120 97,944 164 1,119,743 1,871 16,528,935 27,613
2001 34.60 95,831 34,095 131 398,908 1,531 9,187,068 35,253
2002 57.14 7,622 2,738 132 26,090 1,259 2,843,045 137,162
2003 51.51 210,133 82,484 144 876,800 1,534 6,790,783 11,884
2004 45.70 497,599 175,182 129 1,897,219 1,402 29,945,184 22,130
2005 55.91 568,903 334,378 216 3,435,851 2,221 89,311,749 57,730
2006 65.64 837,413 504,434 222 4,087,935 1,795 131,699,956 57,833
2007 30.61 8,800 3,413 143 29,615 1,238 4,370,521 182,628
2008 53.60 519 215 153 1,964 1,391 11,239,562 7,961,770
2009 42.95 109,296 54,584 184 554,227 1,865 16,508,436 55,544
2010 55.70 48,781 19,916 150 202,622 1,527 7,562,974 57,013
2011 32.28 229,568 80,370 129 822,212 1,317 6,153,579 9,857
2012 40.94 5,094 1,850 134 18,420 1,330 4,717,870 340,593
2013 48.06 73,136 25,325 127 275,894 1,387 7,115,212 35,776
2014 54.19 419,378 156,877 138 1,468,198 1,287 21,409,471 18,773
2015 43.18 80,249 31,315 143 279,058 1,279 2,653,241 12,158
2016 46.19 369,871 186,695 186 1,613,039 1,604 36,742,257 36,530
2017 40.00 338,202 127,468 139 1,322,123 1,438 23,070,557 25,085
2018 47.45 585,612 358,056 225 2,988,912 1,877 145,531,979 91,386
WY1980-2018 Average 50.39 239,869 106,405 163 1,140,764 1,749 24,129,902 36,992
WY1996-2005 Average 52.58 351,262 150,561 158 1,672,706 1,751 29,033,417 30,395
Current 10-yr Average 45.09 225,919 104,246 170 954,471 1,554 27,146,558 44,187
Difference -14% -36% -31% 8% -43% -11% -6% 45%
31
Lake Okeechobee Discharges to SLRE Annual Summary of Flow

Flow (left axis) C-44 Basin Rainfall
1,200,000 100
90
1,000,000
80
Annual Flow (acre feet)

70
800,000
Rainfall (inches)
60
600,000 50
40
400,000
30
20
200,000
10
0 0
Water Year (May - April)
Lake Okeechobee Discharges to SLRE Annual Summary of TP Load and Concentration

Load (left axis) Concentration (right axis)
540,000 300
te legacy TSS to Lake flows results in very high concentrations (e.g., see 6/2007)
480,000
420,000
Concentration (ppb)
Annual Load (pounds)
360,000 200
300,000
240,000
180,000 100
120,000
60,000
0 0
Lake Okeechobee Discharges to SLRE Annual Summary of TN Load and Concentration

6,000,000 3,000
5,000,000 2,500
Concentration (ppb)
4,000,000 2,000
3,000,000 1,500
2,000,000 1,000
1,000,000 500
0 0
Lake Okeechobee Discharges to SLRE Annual Summary of TSS Load and Concentration
150,000,000 450,000
120,000,000 360,000
Concentration (ppb)
90,000,000 270,000
60,000,000 180,000
30,000,000 90,000
0 0
32
Annual Summary of Flows and Loads from the C-23 Canal Basin
Basin Basin FWM FWM FWM
Water Rainfall Flow TP Load TP Conc TN Load TN Conc TSS Load TSS Conc
1996 65.35 280,758 330,424 433 1,227,995 1,608 12,922,456 16,926
1997 46.21 76,826 44,094 211 275,330 1,318 777,968 3,724
1998 53.38 157,211 155,006 363 691,302 1,617 1,100,725 2,575
1999 41.95 75,674 100,720 489 320,679 1,558 571,796 2,779
2000 51.68 175,031 260,173 547 811,320 1,705 2,917,375 6,129
2001 35.93 38,331 37,882 363 191,091 1,833 506,625 4,860
2002 43.16 139,212 202,463 535 616,797 1,629 4,640,951 12,259
2003 51.06 125,215 134,302 394 517,426 1,520 2,449,136 7,193
2004 41.73 139,689 178,613 470 585,625 1,542 2,518,307 6,629
2005 65.32 232,805 414,965 655 907,802 1,434 22,321,213 35,258
2006 67.95 297,209 376,213 465 1,307,351 1,618 2,296,280 2,841
2007 33.97 39,871 39,798 367 210,552 1,942 516,026 4,759
2008 48.53 96,813 131,876 501 436,109 1,657 1,653,257 6,280
2009 40.11 114,820 180,602 578 545,325 1,747 3,238,415 10,372
2010 58.64 112,338 98,382 322 459,942 1,506 1,228,895 4,023
2011 35.94 33,643 27,122 296 136,294 1,490 344,187 3,762
2012 45.93 60,600 60,666 368 291,478 1,769 649,029 3,938
2013 48.95 85,776 114,702 492 383,126 1,643 674,920 2,893
2014 61.24 169,434 205,469 446 795,045 1,726 1,681,869 3,650
2015 48.36 104,991 126,147 442 447,289 1,567 924,204 3,237
2016 56.98 106,699 191,685 661 571,498 1,970 1,215,143 4,188
2017 43.70 65,889 62,811 351 261,029 1,457 876,471 4,892
2018 71.34 219,620 267,633 448 1,113,530 1,864 1,482,646 2,483
WY1996-2018 Average 50.32 128,194 162,685 467 569,736 1,634 2,935,126 8,420
WY1996-2005 Average 49.58 144,075 185,864 474 614,537 1,569 5,072,655 12,947
Difference 3% -25% -28% -4% -19% 9% -76% -67%
33
C-23 Canal Basin Annual Summary of Flow and Rainfall

Flow (left axis) Rainfall (right axis)
500,000 100
450,000 90
400,000 80
Annual Rainfall (inches)

350,000 70
300,000 60
250,000 50
200,000 40
150,000 30
100,000 20
50,000 10
0 0
C-23 Canal Basin Annual Summary of TP Load and Concentration

490,000 700
420,000 600
350,000 500
Concentration (ppb)
280,000 400
210,000 300
140,000 200
70,000 100
0 0
C-23 Canal Basin Annual Summary of TN Load and Concentration

2,500,000 2,500
2,000,000 2,000
Concentration (ppb)
1,500,000 1,500
1,000,000 1,000
500,000 500
0 0
C-23 Canal Basin Annual Summary of TSS Load and Concentration

25,000,000 40,000
20,000,000 32,000
Concentration (ppb)
15,000,000 24,000
10,000,000 16,000
5,000,000 8,000
0 0
34
Annual Summary of Flows and Loads from the C-24 Canal Basin
1980 55.90 166,563 166,037 367 1,097,441 2,423 2,717,690 6,000
1981 29.17 15,174 12,940 314 126,328 3,061 247,585 6,000
1982 49.70 91,552 87,747 352 640,549 2,573 1,983,392 7,967
1983 53.51 283,915 159,070 206 1,299,451 1,683 6,256,287 8,103
1984 46.11 122,741 94,838 284 591,753 1,773 1,920,968 5,755
1985 34.48 92,555 88,643 352 478,219 1,900 530,875 2,109
1986 44.80 121,639 133,241 403 625,405 1,891 868,503 2,626
1987 43.62 113,169 108,770 353 688,723 2,238 8,012,360 26,035
1988 33.71 87,426 100,549 423 403,562 1,697 698,954 2,940
1989 37.44 46,904 43,711 343 187,571 1,471 254,295 1,994
1990 32.14 70,470 61,841 323 306,941 1,602 803,424 4,193
1991 59.54 184,221 217,550 434 754,914 1,507 4,017,878 8,020
1992 50.06 148,360 121,711 302 448,836 1,113 838,651 2,079
1993 65.92 273,309 241,856 325 1,047,481 1,409 2,285,925 3,076
1994 44.95 139,535 130,853 345 1,086,370 2,863 1,414,448 3,728
1995 65.15 264,480 174,365 242 987,526 1,373 1,618,912 2,251
1996 62.96 240,708 165,637 253 930,541 1,422 2,113,692 3,229
1997 42.52 97,160 60,237 228 418,991 1,586 1,934,774 7,323
1998 60.28 209,835 160,051 280 1,046,405 1,834 1,469,743 2,576
1999 43.01 115,946 102,429 325 562,299 1,783 543,555 1,724
2000 52.75 204,809 193,757 348 966,161 1,735 2,415,590 4,337
2001 41.27 50,232 43,510 319 219,827 1,609 553,639 4,053
2002 45.28 206,299 188,233 336 911,650 1,625 1,480,212 2,639
2003 51.86 156,587 145,163 341 662,179 1,555 3,638,067 8,544
2004 42.99 156,125 132,912 313 617,474 1,454 1,320,521 3,110
2005 59.90 239,507 368,059 565 1,119,294 1,719 10,001,612 15,356
2006 60.29 259,530 263,728 374 1,028,892 1,458 2,137,437 3,029
2007 33.99 41,877 56,853 499 212,147 1,863 204,781 1,798
2008 46.22 116,298 128,023 405 512,591 1,621 1,130,353 3,574
2009 44.66 135,732 176,857 479 590,256 1,599 3,336,446 9,039
2010 57.48 159,496 144,789 334 678,537 1,564 1,312,414 3,026
2011 37.23 78,976 57,652 268 365,650 1,703 1,004,423 4,677
2012 49.61 141,246 120,693 314 712,653 1,855 2,005,512 5,221
2013 49.57 141,852 120,396 312 602,513 1,562 1,708,869 4,430
2014 55.36 140,325 125,462 329 550,596 1,443 819,488 2,148
2015 51.67 104,865 86,171 302 394,500 1,383 1,043,552 3,659
2016 62.21 160,646 115,999 266 616,567 1,411 1,263,245 2,892
2017 49.46 118,012 86,880 271 395,884 1,234 641,891 2,000
2018 63.44 216,954 248,012 420 980,181 1,661 3,095,210 5,246
WY1980-2018 Average 48.98 146,539 134,237 337 663,253 1,664 2,042,184 5,125
WY1996-2005 Average 50.28 167,721 155,999 342 745,482 1,634 2,547,141 5,585
Difference 4% -17% -18% -1% -21% -5% -36% -24%
35

500,000 100
450,000 90
400,000 80

350,000 70
300,000 60
250,000 50
200,000 40
150,000 30
100,000 20
50,000 10
0 0

420,000 600
350,000 500
Concentration (ppb)
280,000 400
210,000 300
140,000 200
70,000 100
0 0

1,750,000 3,500
1,500,000 3,000
1,250,000 2,500
Concentration (ppb)
1,000,000 2,000
750,000 1,500
500,000 1,000
250,000 500
0 0

12,000,000 30,000
10,000,000 25,000
Concentration (ppb)
8,000,000 20,000
6,000,000 15,000
4,000,000 10,000
2,000,000 5,000
0 0
36
Annual Summary of Flows and Loads from the C-44 Canal Basin (to Lake and SLRE)
1980 58.25 91,598 111,221 447 324,585 1,303 1,090,623 4,378
1981 37.19 27,676 14,168 188 119,896 1,593 16,755 223
1982 51.02 84,775 50,504 219 538,072 2,334 4,631,969 20,092
1983 69.84 200,608 89,819 165 1,439,689 2,639 54,544,213 99,984
1984 58.74 216,422 71,877 122 906,326 1,540 3,825,464 6,500
1985 47.69 192,012 108,654 208 862,262 1,651 468,496 897
1986 46.42 231,132 120,430 192 1,273,371 2,026 5,747,074 9,144
1987 50.94 157,734 92,635 216 706,659 1,647 1,357,381 3,165
1988 45.94 283,533 114,811 149 1,174,569 1,523 7,869,706 10,207
1989 41.42 280,208 120,205 158 1,123,237 1,474 329,905 433
1990 45.18 245,568 95,382 143 946,214 1,417 575,195 861
1991 53.84 148,997 60,613 150 565,388 1,395 330,676 816
1992 41.47 209,541 85,690 150 497,605 873 1,239,199 2,175
1993 67.53 359,991 168,443 172 1,111,289 1,135 1,293,516 1,321
1994 56.85 177,178 65,464 136 760,889 1,579 1,047,128 2,173
1995 66.40 439,081 195,660 164 1,433,521 1,201 2,771,975 2,322
1996 69.99 307,565 173,363 207 2,283,833 2,731 938,733 1,122
1997 47.11 129,268 44,196 126 675,173 1,921 342,480 974
1998 56.93 157,242 55,000 129 395,149 924 466,281 1,090
1999 46.22 148,620 61,449 152 550,544 1,362 178,338 441
2000 60.74 218,669 152,829 257 797,115 1,340 312,248 525
2001 34.60 104,182 45,807 162 401,498 1,417 388,937 1,373
2002 57.14 135,035 113,746 310 682,648 1,859 341,719 931
2003 51.51 145,180 107,856 273 409,276 1,037 110,931 281
2004 45.70 213,456 176,683 304 746,505 1,286 849,225 1,463
2005 55.91 232,251 259,138 410 1,383,856 2,191 2,739,731 4,338
2006 65.64 381,862 275,750 266 2,587,689 2,492 7,362,561 7,090
2007 30.61 66,532 35,402 196 280,558 1,551 516,233 2,853
2008 53.60 191,652 166,372 319 999,321 1,917 1,401,970 2,690
2009 42.95 121,649 126,455 382 542,174 1,639 1,087,327 3,287
2010 55.70 143,443 84,195 216 542,082 1,390 726,761 1,863
2011 32.28 60,241 45,157 276 206,213 1,259 117,984 720
2012 40.94 53,882 24,256 166 251,376 1,716 576,794 3,936
2013 48.06 160,985 131,395 300 677,515 1,548 636,030 1,453
2014 54.19 284,575 247,135 319 1,280,958 1,655 646,049 835
2015 43.18 124,861 87,709 258 422,392 1,244 178,061 524
2016 46.19 192,202 137,644 263 751,933 1,439 635,951 1,217
2017 40.00 87,323 69,770 294 365,133 1,538 525,430 2,213
2018 47.45 169,483 218,531 474 1,177,370 2,555 7,455,764 16,177
WY1980-2018 Average 50.39 184,005 112,959 226 825,484 1,650 2,966,021 5,928
WY1996-2005 Average 52.58 179,147 119,007 244 832,560 1,709 666,862 1,369
Difference -14% -22% -1% 26% -25% -4% 89% 142%
37

500,000 100
450,000 90
400,000 80

350,000 70
300,000 60
250,000 50
200,000 40
150,000 30
100,000 20
50,000 10
0 0

300,000 500
240,000 400
Concentration (ppb)
180,000 300
120,000 200
60,000 100
0 0

3,000,000 3,000
2,500,000 2,500
Concentration (ppb)
2,000,000 2,000
1,500,000 1,500
1,000,000 1,000
500,000 500
0 0

60,000,000 120,000
50,000,000 100,000
Concentration (ppb)
40,000,000 80,000
30,000,000 60,000
20,000,000 40,000
10,000,000 20,000
0 0
38
Annual Summary of Flows and Loads from the Ten Mile Creek Basin
1996 68.09 76,829 51,198 245 243,274 1,164 611,150 2,925
1997 53.45 46,572 28,814 228 147,466 1,164 625,393 4,938
1998 54.02 86,468 62,136 264 273,795 1,164 612,275 2,604
1999 40.35 42,737 34,319 295 135,323 1,164 253,938 2,185
2000 44.68 115,143 97,509 311 364,591 1,164 1,086,526 3,470
2001 30.22 46,099 49,029 391 139,805 1,115 203,113 1,620
2002 55.31 110,997 128,292 425 419,349 1,389 802,333 2,658
2003 50.48 271,838 273,099 369 819,413 1,108 3,456,852 4,676
2004 48.67 253,018 231,788 337 676,280 983 1,271,027 1,847
2005 53.55 301,142 362,600 443 1,089,565 1,330 6,057,049 7,396
2006 65.01 136,353 146,349 395 475,830 1,283 1,769,843 4,773
2007 38.41 59,524 37,552 232 153,148 946 439,801 2,717
2008 48.01 110,235 120,599 402 352,551 1,176 1,071,333 3,574
2009 50.42 145,966 213,636 538 624,683 1,574 4,379,953 11,034
2010 63.14 91,964 62,820 251 280,036 1,120 831,537 3,325
2011 37.36 50,463 20,266 148 106,644 777 205,842 1,500
2012 51.55 94,488 70,616 275 304,340 1,184 893,846 3,479
2013 48.99 82,746 53,782 239 234,585 1,043 754,968 3,355
2014 57.07 85,081 51,840 224 226,043 977 704,446 3,045
2015 55.45 89,455 46,652 192 206,934 851 388,381 1,597
2016 59.84 132,321 109,224 304 402,076 1,117 562,287 1,563
2017 48.58 116,013 100,718 319 308,666 978 6,136,019 19,450
2018 64.42 152,861 154,954 373 499,901 1,203 2,174,217 5,230
WY1996-2018 Average 51.61 117,318 109,034 342 368,882 1,156 1,534,440 4,810
WY1996-2005 Average 49.88 135,084 131,878 359 430,886 1,173 1,497,966 4,078
Difference 8% -23% -33% -13% -26% -4% 14% 47%
WY2000-2005 49.70 176,370 184,095 384 569,262 1,187 2,092,392 4,363
Note: Flows and loads prior to WY2001 based on adjacent basins.
39
Ten Mile Creek Basin Annual Summary of Flow and Rainfall

500,000 100
450,000 90
400,000 80

350,000 70
300,000 60
250,000 50
200,000 40
150,000 30
100,000 20
50,000 10
0 0
Flow data prior to July 1999 estimated from adjacent stations. Water Year (May - April)
Ten Mile Creek Basin Annual Summary of TP Load and Concentration

420,000 600
350,000 500
Concentration (ppb)
280,000 400
210,000 300
140,000 200
70,000 100
0 0

Flow and water quality data prior to July 1999 estimated from adjacent stations.
Ten Mile Creek Basin Annual Summary of TN Load and Concentration

2,000,000 2,000
1,500,000 1,500
Concentration (ppb)
1,000,000 1,000
500,000 500
0 0

Ten Mile Creek Basin Annual Summary of TSS Load and Concentration
7,000,000 21,000
6,000,000 18,000
5,000,000 15,000
Concentration (ppb)
4,000,000 12,000
3,000,000 9,000
2,000,000 6,000
1,000,000 3,000
0 0

40
Annual Summary of Flows and Loads from the Tidal Basins

1980 155,185 49,912 118 384,499 911 2,205,338 5,226
1981 89,373 28,745 118 221,438 911 1,270,084 5,226
1982 91,439 29,409 118 226,555 911 1,299,434 5,226
1983 256,279 82,426 118 634,977 911 3,641,981 5,226
1984 149,556 48,101 118 370,550 911 2,125,332 5,226
1985 96,851 31,150 118 239,965 911 1,376,345 5,226
1986 121,871 39,197 118 301,957 911 1,731,908 5,226
1987 207,601 66,770 118 514,366 911 2,950,207 5,226
1988 110,097 35,410 118 272,784 911 1,564,584 5,226
1989 97,212 31,266 118 240,861 911 1,381,484 5,226
1990 57,218 18,403 118 141,767 911 813,119 5,226
1991 78.80 215,664 69,363 118 534,344 911 3,064,794 5,226
1992 60.32 165,536 53,241 118 410,145 911 2,352,433 5,226
1993 79.25 216,896 69,759 118 537,398 911 3,082,311 5,226
1994 60.44 165,854 53,343 118 410,932 911 2,356,945 5,226
1995 76.49 209,412 67,352 118 518,854 911 2,975,949 5,226
1996 77.41 204,909 65,904 118 507,698 911 2,911,961 5,226
1997 53.74 147,561 47,460 118 365,609 911 2,096,993 5,226
1998 56.01 192,323 61,856 118 476,513 911 2,733,097 5,226
1999 47.60 139,088 44,734 118 344,616 911 1,976,583 5,226
2000 55.39 167,383 53,835 118 414,722 911 2,378,683 5,226
2001 36.04 90,793 29,201 118 224,956 911 1,290,261 5,226
2002 65.82 171,093 55,028 118 423,913 911 2,431,401 5,226
2003 44.80 93,580 32,646 128 231,861 911 943,883 3,709
2004 45.54 122,723 37,766 113 275,232 825 1,744,016 5,226
2005 55.72 142,713 45,900 118 370,230 954 2,551,184 6,574
2006 66.47 203,662 73,350 132 565,613 1,021 4,322,199 7,804
2007 31.25 76,179 18,394 89 152,766 737 710,542 3,430
2008 63.69 169,686 41,463 90 399,256 865 2,107,771 4,568
2009 44.20 142,576 37,325 96 336,771 869 2,835,350 7,313
2010 58.22 159,555 44,773 103 356,200 821 2,565,076 5,912
2011 32.91 84,663 24,535 107 196,776 855 1,414,223 6,143
2012 47.08 142,871 37,753 97 332,753 856 1,654,459 4,258
2013 51.81 118,440 26,627 83 261,484 812 1,371,547 4,258
2014 55.06 148,505 41,862 104 386,977 958 1,719,707 4,258
2015 51.30 148,971 39,664 98 320,775 792 1,725,105 4,258
2016 60.68 166,501 45,234 100 411,104 908 1,928,095 4,258
2017 44.76 123,321 32,587 97 274,740 819 1,428,070 4,258
2018 50.08 137,763 42,337 113 341,316 911 1,595,310 4,258
WY2003-2018 Average 50.22 136,357 38,889 103 321,460 867 1,811,277 4,885
WY2003-2005 Average 48.68 119,672 38,490 118 296,509 911 1,700,660 5,226
Current 10-yr Average 49.61 137,317 37,270 100 321,890 862 1,823,694 4,884
Difference 2% 15% -3% -16% 9% -5% 7% -7%
Note: For the Tidal Basins, reliable flow data are not available, so flows were estimated based
on SFWMD models (see text for additional details). Concentrations observed in WY2003-2005
were applied prior to that time.
41
Tidal Basins Annual Summary of Flow and Rainfall

250,000 125
200,000 100

150,000 75
100,000 50
50,000 25
0 0

Flows estimated from SFWMD 's SLE Tidal Basin Lin-Res Model .
Tidal Basins Annual Summary of TP Load and Concentration

Load (left axis) Concentration (right axis) TMDL concentration (right axis)
90,000 150
Concentration (ppb)
60,000 100
30,000 50
0 0
Flows estimated from SFWMD 's SLE Tidal Basin Lin-Res Model . Loads prior to WY2003 estimated from mean concentration during WY2003-2005.
Tidal Basins Annual Summary of TN Load and Concentration

1,200,000 1,200
1,000,000 1,000
Concentration (ppb)
800,000 800
600,000 600
400,000 400
200,000 200
0 0

Tidal Basins Annual Summary of TSS Load and Concentration

4,500,000 9,000
4,000,000 8,000
3,500,000 7,000
Concentration (ppb)
3,000,000 6,000
2,500,000 5,000
2,000,000 4,000
1,500,000 3,000
1,000,000 2,000
500,000 1,000
0 0

Note: TSS data collection stopped in WY2012; subsequent load calculations awere based on WY2012 concentrations.
42
Annual Summary of Flows and Loads from the SLRE Watershed

SLRE Total FWM FWM FWM
1996 77.41 1,859,394 1,008,377 199 7,881,514 1,559 49,754,531 9,840
1997 53.74 540,163 234,505 160 1,949,104 1,327 9,388,272 6,391
1998 56.01 1,812,819 1,002,245 203 8,595,988 1,744 105,351,422 21,371
1999 47.60 564,466 361,039 235 2,084,921 1,358 6,053,866 3,944
2000 55.39 1,058,923 819,562 285 4,279,325 1,486 25,454,951 8,840
2001 36.04 330,297 197,642 220 1,206,340 1,343 11,776,297 13,111
2002 65.82 668,024 593,097 326 2,506,281 1,380 12,220,165 6,727
2003 44.80 972,173 731,190 277 3,401,408 1,287 17,314,576 6,549
2004 45.54 1,382,610 932,944 248 4,798,335 1,276 37,648,279 10,013
2005 55.72 1,661,906 1,752,507 388 7,998,479 1,770 131,754,142 29,153
2006 66.47 2,101,536 1,631,022 285 9,976,231 1,746 149,090,225 26,088
2007 31.25 255,094 168,742 243 851,307 1,227 6,288,411 9,065
2008 63.69 495,542 423,390 314 1,709,651 1,269 17,208,815 12,770
2009 44.20 713,631 763,092 393 2,963,550 1,527 30,756,760 15,849
2010 58.22 664,562 431,821 239 2,296,660 1,271 13,717,123 7,590
2011 32.91 529,093 252,658 176 1,802,457 1,253 9,206,182 6,399
2012 47.08 447,950 292,759 240 1,669,621 1,371 9,929,369 8,151
2013 51.81 587,599 418,734 262 2,115,255 1,324 11,713,540 7,331
2014 55.06 1,219,058 815,106 246 4,581,557 1,382 26,673,538 8,046
2015 51.30 636,516 410,372 237 2,010,579 1,162 6,773,845 3,913
2016 60.68 1,113,694 779,389 257 4,299,374 1,420 42,126,069 13,910
2017 44.76 840,260 474,995 208 2,880,711 1,261 32,456,710 14,204
2018 50.08 1,392,652 1,217,559 321 6,483,994 1,712 160,697,552 42,432
WY1996-2018 Average 51.98 949,911 683,163 264 3,840,984 1,487 40,145,854 15,541
WY1996-2005 Average 53.81 1,085,078 763,311 259 4,470,169 1,515 40,671,650 13,784
Current 10-yr Average 49.61 814,501 585,649 264 3,110,376 1,404 34,405,069 15,533
Difference -8% -25% -23% 2% -30% -7% -15% 13%
Notes: 1. "FWM" = Flow-eighted mean
43
SLRE Watershed Annual Summary of Flow and Rainfall

2,500,000 100
90
2,000,000 80

70
1,500,000 60
50
1,000,000 40
30
500,000 20
10
0 0
SLRE Watershed Annual Summary of TP Load and Concentration

2,000,000 500
1,800,000
1,600,000 400
Concentration (ppb)
1,400,000
1,200,000 300
1,000,000
800,000 200
600,000
400,000 100
200,000
0 0
SLRE Watershed Annual Summary of TN Load and Concentration

12,000,000 2,000
1,800
10,000,000
1,600
Concentration (ppb)
1,400
8,000,000
1,200
6,000,000 1,000
800
4,000,000
600
400
2,000,000
200
0 0
SLRE Watershed Annual Summary of TSS Load and Concentration

180,000,000 45,000
40,000
150,000,000
35,000
Concentration (ppb)
120,000,000 30,000
25,000
90,000,000
20,000
60,000,000 15,000
10,000
30,000,000
5,000
0 0
44
APPENDIX 2. Nutrient Trends
Figure 2-1. C-23 Canal Basin Nutrient Load Trends.
Notes: A negative change in loads denotes a reduction in load in comparison to the base period, adjusted for
hydrologic variability. A downward trend in the solid line also denotes a reduction in loads.
45
Figure 2-2. C-24 Basin Nutrient Load Trends.
46
Figure 2-3. C-44 Basin Nutrient Load Trends.
47
Figure 2-4. Ten Mile Creek Basin Nutrient Load Trends.
48
Figure 2-5. Comparison of observed annual concentrations to TMDL and BMAP Goals: Lake
Okeechobee Discharges.
Lake Okeechobee - TN TMDL Goal = 720 ppb
100% 100%
Change in Concentration from Base Period (%)
Base Period
0% 0%
BMAP Ph I Goal
(18%reduction)
TMDL Goal (59%reduction)
-100% -100%
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Water Year (May 1 - April 30)
Annual Difference Series3 10-yr Rolling Average Series4
Lake Okeechobee - TP Lake Okeechobee TMDL Goal = 40 ppb

Change in Concentration from Base Period (%)
100% 100%
Base Period
0% 0%
BMAP Ph I Goal (22%reduction)
TMDL Goal (75%reduction)

-100% -100%
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020

Notes: A negative change denotes a reduction in concentration in comparison to the base period.
A downward trend in the solid line denotes a reduction in concentration.
49
Figure 2-6. Comparison of observed annual concentrations to TMDL and BMAP Goals: Tidal Basins.
Tidal Basins - TN
50% 50%
Base Period
Change from Base Period (%)
0% 0%
BMAP Ph I Goal
(6% reduction)
TMDL Goal (21% reduction)
-50% -50%
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
Tidal Basins - TP
50% 50%
Base Period
Change from Base Period (%)
0% 0%
BMAP Ph I Goal
(9% reduction)
TMDL Goal (32% reduction)
-50% -50%
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019

Notes: A negative change denotes a reduction in concentration in comparison to the base period.
A downward trend in the solid line denotes a reduction in concentration.
50
Appendix 3. Flows and Loads to the St. Lucie River and Estuary
C-44 Canal Basin to SLRE

1980 58.25 88,221 110,390 460 308,027 1,284 1,084,507 4,521
1981 37.19 22,544 12,765 208 93,147 1,519 7,462 122
1982 51.02 6,840 2,300 124 31,367 1,686 68,167 3,665
1983 69.84 200,608 89,819 165 1,439,689 2,639 54,544,213 99,984
1984 58.74 216,422 71,877 122 906,326 1,540 3,825,464 6,500
1985 47.69 133,325 90,542 250 642,138 1,771 204,223 563
1986 46.42 8,160 2,088 94 29,354 1,323 36,958 1,666
1987 50.94 87,327 34,746 146 356,427 1,501 139,607 588
1988 45.94 139,353 52,868 140 540,107 1,425 840,369 2,218
1989 41.42 212,179 97,698 169 831,209 1,441 81,625 141
1990 45.18 40,392 22,126 201 122,023 1,111 79,862 727
1991 53.84 14,384 4,813 123 36,939 944 27,925 714
1992 41.47 156,565 64,954 153 264,779 622 544,834 1,280
1993 67.53 351,243 165,497 173 1,067,243 1,117 1,248,656 1,307
1994 56.85 18,263 5,381 108 47,491 956 26,745 539
1995 66.40 342,851 149,923 161 1,140,226 1,223 1,277,421 1,370
1996 69.99 307,565 173,363 207 2,283,833 2,731 938,733 1,122
1997 47.11 97,077 33,732 128 482,734 1,829 90,431 343
1998 56.93 146,824 51,375 129 349,943 876 441,924 1,107
1999 46.22 122,359 53,883 162 454,730 1,367 93,497 281
2000 60.74 176,437 116,344 242 602,790 1,256 127,842 266
2001 34.60 9,010 3,925 160 31,752 1,296 35,590 1,453
2002 57.14 32,802 16,343 183 108,481 1,216 22,222 249
2003 51.51 114,820 63,495 203 293,728 941 35,854 115
2004 45.70 213,456 176,683 304 746,505 1,286 849,225 1,463
2005 55.91 176,836 226,606 471 1,075,737 2,237 1,511,334 3,143
2006 65.64 367,369 266,947 267 2,510,611 2,513 6,864,511 6,871
2007 30.61 28,842 12,732 162 93,079 1,187 46,741 596
2008 53.60 1,991 1,214 224 7,182 1,327 6,538 1,208
2009 42.95 65,241 100,087 564 312,289 1,760 458,160 2,582
2010 55.70 92,428 61,142 243 319,323 1,270 216,226 860
2011 32.28 51,780 42,713 303 174,881 1,242 83,929 596
2012 40.94 3,652 1,181 119 9,976 1,004 8,653 871
2013 48.06 85,649 77,902 334 357,654 1,536 88,024 378
2014 54.19 256,335 233,596 335 1,154,698 1,657 338,557 486
2015 43.18 107,985 80,421 274 362,023 1,233 39,362 134
2016 46.19 177,656 130,551 270 685,090 1,418 415,041 859
2017 40.00 78,823 64,531 301 318,269 1,485 303,702 1,417
2018 47.45 79,843 146,320 674 560,435 2,581 6,818,189 31,403
WY1980-2018 Average 50.39 123,935 79,817 237 542,365 1,609 2,150,572 6,381
WY1996-2005 Average 52.58 139,719 91,575 241 643,023 1,692 414,665 1,091
Current 10-yr Average 45.09 99,939 93,845 345 425,464 1,566 876,984 3,227
Difference -14% -28% 2% 43% -34% -7% 111% 196%
51
C-44 Canal Basin to SLRE Annual Summary of Flow and Rainfall

500,000 100
450,000 90
400,000 80

350,000 70
300,000 60
250,000 50
200,000 40
150,000 30
100,000 20
50,000 10
0 0
C-44 Canal Basin to SLRE Annual Summary of TP Load and Concentration

360,000 800
700
300,000
600
Concentration (ppb)
240,000
500
180,000 400
300
120,000
200
60,000
100
0 0
C-44 Canal Basin to SLRE Annual Summary of TN Load and Concentration

3,000,000 3,000
2,500,000 2,500
Concentration (ppb)
2,000,000 2,000
1,500,000 1,500
1,000,000 1,000
500,000 500
0 0
C-44 Canal Basin to SLRE Annual Summary of TSS Load and Concentration
60,000,000 120,000
50,000,000 100,000
Concentration (ppb)
40,000,000 80,000
30,000,000 60,000
20,000,000 40,000
10,000,000 20,000
0 0
52
WY2018 Surface Flow to the SLRE WY2018 Nitrogen Load to the SLRE
(Excludes C-44 Basin Flow to Lake) (Excludes C-44 Basin Flow to Lake)
C-23 Canal C-23 Canal

16% 17%
Lake Okeechobee
C-24 Canal Lake Okeechobee C-24 Canal
42%
16% 46% 15%
C-44 Canal to SLRE

9%
Ten Mile Creek Ten Mile Creek
Tidal Basins 11% Tidal Basins 8%
10% 5%
C-44 Canal to SLRE

Flow data for Tidal Basins estimated from SFWMD calculations. 6% Flow data for Tidal Basins estimated from SFWMD calculations.
WY2018 Sediment Load to the SLRE

WY2018 Phosphorus Load to the SLRE (Excludes C-44 Basin Flow to Lake)
(Excludes C-44 Basin Flow to Lake) C-24 Canal
C-44 Canal to SLRE Ten Mile Creek
2%
C-23 Canal 4% 1%
1%
Tidal Basins
C-23 Canal 1%
Lake Okeechobee 22%
29%
C-24 Canal
20%
Lake Okeechobee
Ten Mile Creek 91%
C-44 Canal to SLRE
13%
12%
Tidal Basins
4%
53
WY2018 Surface Flow to the SLRE WY2018 Nitrogen Loading to the SLRE
Lake Agriculture
Agriculture Okeechobee 46%
Lake
45% 46%
Okeechobee
42%
Communities Natural
7% 6%
Communities Natural
5% 3%
WY2018 Phosphorus Loading to the SLRE WY2018 Sediment Loading to the SLRE
Agriculture
7%
Natural
Lake
1%
Okeechobee
29% Communities
Lake 1%
Agriculture Okeechobee
61% 91%
Communities
5%
Natural
5%
54
WY2009-2018 Average Annual Surface Flow to the SLRE WY2009-2018 Average Annual Nitrogen Loads to the SLRE
C-23 Canal
13%
Lake Okeechobee C-23 Canal
28% 16%
C-24 Canal Lake Okeechobee
17% 31%
C-24 Canal
19%
Tidal Basins C-44 Canal to SLRE
17% 12% Tidal Basins C-44 Canal to
10% SLRE
Ten Mile Creek Ten Mile Creek 14%
13% 10%
WY2009-2018 Average Annual Total Suspended Solids Loads

WY2009-2018 Average Annual Phosphorus Loads to the SLRE
to the SLRE
C-24 Canal C-44 Canal to
C-23 Canal
5% SLRE
4%
Tidal Basins Tidal Basins 2%
Lake Okeechobee
6% C-23 Canal 5%
18%
23% Ten Mile Creek
5%
Ten Mile Creek C-24 Canal

15% 22%
C-44 Canal to SLRE
Lake Okeechobee
16%
79%
55
WY2009-2018 Average Annual Surface Flow to the SLRE WY2009-2018 Average Annual Nitrogen Loads to the
SLRE
Lake
Okeechobee Lake
28% Agriculture Okeechobee
51% 31%
Agriculture
Communities 57%
12% Communities
Natural
8%
9%
Natural
4%
WY2009-2018 Average Annual Phosphorus Loads to the WY2009-2018 Average Annual Total Suspended
SLRE Solids Loads to the SLRE
Lake
Communities
Okeechobee
7% Agriculture
18%
15%
Lake
Okeechobee
Agriculture Natural
79%
70% 3%
Natural
5%
Communities
3%
56
APPENDIX P:
FLORIDA DEPARTMENT OF ENVIRONMENTAL
PROTECTION 2018 ALGAL BLOOM SAMPLING
RESULTS
APPENDIX Q:
DR. GARY GOFORTH SEPTIC TANK ANALYSIS
Septic tanks are part of the loading
problem to St Lucie Estuary
Estimated contribution remains

below 10 percent of total nitrogen
loading to estuary
Efforts by counties and

municipalities are resulting in
declining nitrogen levels
• Conversion of more than 8,000
septic tanks and 70 package
treatment plants
• Active septic conversion
program - $155 million
• $85 million spent on other
nutrient reduction projects
• Met and exceeded all nitrogen
reduction goals of BMAP
Revised 10/28/2018
Septic Tank Contributions to Pollutant Loading to the St. Lucie Estuary – 2 4 2017
The major pollutants of concern for the St. Lucie Estuary are toxic blue-green algae, too much
sediment (muck), too much water that lowers the salinity, too much nitrogen, too much
phosphorus, and pesticides. All of these are found in Lake Okeechobee discharges to the estuary,
such as the 220 billion gallons sent between January and November 2016. The 2017 Florida
Legislature is identifying ways to reduce these damaging discharges, consistent with the 2016
Legacy Florida Act that directs state agencies to prioritize projects that reduce the damaging
discharges.
Septic tanks have been purported to be a major source of pollution entering the estuary, although
they contribute no toxic algae, no sediment and no pesticides, and contribute negligible amounts
of water and phosphorus (LaPointe and Herren 2016). The contribution of nitrogen from septic
tanks is uncertain, and is a subject of scientific debate. Dr. LaPointe of Harbor Branch
Oceanographic Institute/FAU maintains that septic tanks are a primary source of nitrogen to the
estuary, while I maintain that septic tanks are not a primary source (Goforth 2016). I examined his
2016 report to Martin County and his 2017 presentation to the Florida Senate and identified the
following potential flaws that led to an overestimation of nitrogen loading to the estuary from
septic tanks in Martin and St. Lucie County.
1. Dr. LaPointe overestimated the number of septic tanks that are in the St. Lucie Estuary
watershed.
a. Some Martin County septic tanks are located in the Lake Okeechobee Watershed.
b. Some Martin County and St. Lucie County septic tanks are in the Indian River Lagoon
watershed or drain to the Atlantic Ocean.
c. In 2013, FDEP consultants estimated the total number of both functioning (28,761)
and removed (6,673) tanks that contribute some level of nitrogen to the estuary, for
a total of 35,439 tanks.
d. Dr. LaPointe assumed all known (43,224) and estimated (20,986) tanks in Martin and
St. Lucie County drain to the estuary, for a total of 64,210 tanks (from FDOH 2015).
e. Bottom line: Dr. LaPointe’s estimated number of septic tanks that contribute
loading to the estuary are 22% - 81% higher than the FDEP’s estimate.
2. Dr. LaPointe overestimated the nitrogen loading from individual septic tanks to the estuary.
a. LaPointe estimated 24.25 pounds/yr of nitrogen to the estuary from each septic
tank. This estimate assumed every septic tank contributed the same nitrogen load,
regardless of how far the tank was located from the estuary, what the soil type is,
what the depth to water table is, the condition of the tank and drainfield, and other
site-specific factors which are critical for accurate estimates.
1
b. FDEP consultants that evaluated septic tank removal for the BMAP used spatial
modeling that accounted for distance to the estuary, soil type, depth to
groundwater and other site-specific factors, and their estimated loading per septic
tank ranged from 3.9 lbs/yr to 16.0 lbs/yr with an average of 6.52 lbs/yr per tank (Ye
and Sun 2013). These loadings are consistent with other Florida studies (IFAS, FDOH,
etc.)
c. Bottom line: Dr. LaPointe’s estimated loading from each septic tank is 272% higher
than the FDEP’s average estimated loading.
3. The cumulative effect of the above potential flaws result in Dr. LaPointe overestimating
nitrogen loading to the estuary from septic tanks by 354% - 574% compared to FDEP’s
estimates.
4. On an annual average basis (1980-2016), FDEP’s estimated loading (0.23 million pounds/yr)
represents about 6% of the total estimated nitrogen loading to the estuary (3.87 million
pounds/yr).
a. By contrast, Dr. LaPointe estimated septic tank loading (1.05 – 1.56 million
pounds/yr) represents 27-40% of the total nitrogen loading to the estuary.
b. Lake Okeechobee discharges account for 29% of the total nitrogen loading.
c. Runoff from agricultural land use is estimated as 56% of the total nitrogen loading.
5. For 2016, a year with high Lake Okeechobee discharges, FDEP’s estimated loading (0.23
million pounds) represents about 4% of the total estimated nitrogen loading to the estuary
(5.96 million pounds).
a. By contrast, Dr. LaPointe did not estimate septic tank loading for 2016.
b. Lake Okeechobee discharges account for 48% of the total nitrogen loading.
c. Runoff from agricultural land use is estimated as 41% of the total nitrogen loading.
6. Centralized sewer systems do not completely remove nitrogen from the watershed.
a. Land application of Class AA residuals – a practice which is neither regulated nor
monitored – may be a significant contributor to nitrogen loading in the St. Lucie
estuary watershed.
b. Periodically, leaks or breaks in the collection system result in spills of raw sewage
into local waterways.
7. Summary
a. The major pollutants of concern for the St. Lucie Estuary are toxic blue-green algae,
too much sediment (muck), too much water that lowers the salinity, too much
2
nitrogen, too much phosphorus, and pesticides. Of these, septic tanks are a non-
negligible source of only one of these pollutants – nitrogen.
b. Some – but not all – septic tanks in Martin and St. Lucie County should be
considered for conversion to centralized sewer systems (Ye and Sun 2013). Martin
County is aggressively pursuing septic-to-sewer conversion in areas identified as hot
spots. The City of Stuart and Port St. Lucie also have conversion programs
underway.
c. Land application of Class AA residuals in the St. Lucie estuary watershed should
evaluated for its impact on nitrogen loading to the estuary.
d. The most important project that can be implemented to reduce the damaging
discharges from Lake Okeechobee to the estuaries is completion of the EAA Storage
Reservoir.
Findings from a recent study by FDEP and Martin County (FDEP 2016):
• Martin County identified 14 tributary and river sites suspected as “hot spots” that may
contribute bacterial pollution to the St. Lucie River and Estuary.
• Acetaminophen, which is an indicator for untreated wastewater, was not detected at 12 of
14 sampling sites during the wet season, and was not detected at 12 of the 14 locations
during the dry season.
• During the wet season, 13 of 14 sites had Sucralose concentrations below 1.0 μg/L, a level
that may be indicative of human-derived wastewater; similarly, 11 of 14 sites had Sucralose
concentrations below that level during the dry season.
• The presence of a human fecal source marker (qPCR Bacteroidales HF183) was not detected
at 11 of the 14 sampling locations during the wet season, and was not detected at 9 of the
14 locations during the dry season.
References
FDEP 2016. South Fork St. Lucie Estuary and River Microbial Source Tracking Study. Division of
Environmental Assessment and Restoration Florida Department of Environmental Protection.
February 12, 2016.
Goforth, G. 2016. Estimates of Nitrogen Loading from Septic Systems in Martin County. DRAFT –
March 28, 2016
LaPointe, B. and L. Herren 2016. 2015 Martin County Watershed To Reef Septic Study Draft Final
Report. Marine Ecosystem Health Program, Harbor Branch Oceanographic Institute, Florida
Atlantic University. February 29, 2016.
3
Ye, Ming and Huaiwei Sun 2013. Estimation of Nitrogen Load from Removed Septic Systems to
Surface Water Bodies in the City of Port St. Lucie, the City of Stuart, and Martin County.
September 2013.
4
APPENDIX R:
NATIONAL ACADEMIES OF SCIENCE
EVERGLADES REPORT
Progress Toward Restoring the Everglades: The Fifth Biennial Review: 2014
Committee on Independent Scientific Review of Everglades Restoration Progress
Water Science and Technology Board
Board on Environmental Studies and Toxicology
Division on Earth and Life Studies
THE NATIONAL ACADEMIES PRESS

Washington, D.C.
www.nap.edu
Copyright National Academy of Sciences. All rights reserved.

THE NATIONAL ACADEMIES PRESS 500 Fifth Street, NW Washington, D.C. 20001
NOTICE: The project that is the subject of this report was approved by the Governing
Board of the National Research Council, whose members are drawn from the councils
of the National Academy of Sciences, the National Academy of Engineering, and the
Institute of Medicine. The members of the panel responsible for the report were chosen
for their special competences and with regard for appropriate balance.
Support for this study was provided by the Department of the Army under Cooperative
Agreement No. W912EP-04-2-0001. Support for this project was also provided by the
U.S. Department of the Interior and the South Florida Water Management District. Any
opinions, findings, conclusions, or recommendations expressed in this publication are
those of the author(s) and do not necessarily reflect the views of the organizations or
agencies that provided support for the project.
International Standard Book Number-13: 978-0-309-30576-1

International Standard Book Number-10: 0-309-30576-4
Cover credit: Cover image courtesy of David J. Policansky.
Additional copies of this report are available for sale from the National Academies
Press, 500 Fifth Street, NW, Keck 360, Washington, DC 20001; (800) 624-6242 or
(202) 334-3313; http://www.nap.edu/.
Copyright 2014 by the National Academy of Sciences. All rights reserved.
Printed in the United States of America

The National Academy of Sciences is a private, nonprofit, self-perpetuating society of

distinguished scholars engaged in scientific and engineering research, dedicated to the
furtherance of science and technology and to their use for the general welfare. Upon
the authority of the charter granted to it by the Congress in 1863, the Academy has a
mandate that requires it to advise the federal government on scientific and technical
matters. Dr. Ralph J. Cicerone is president of the National Academy of Sciences.
The National Academy of Engineering was established in 1964, under the charter of the
National Academy of Sciences, as a parallel organization of outstanding engineers. It is
autonomous in its administration and in the selection of its members, sharing with the
National Academy of Sciences the responsibility for advising the federal government.
The National Academy of Engineering also sponsors engineering programs aimed at
meeting national needs, encourages education and research, and recognizes the superior
achievements of engineers. Dr. C. D. Mote, Jr., is president of the National Academy of
Engineering.
The Institute of Medicine was established in 1970 by the National Academy of Sciences
to secure the services of eminent members of appropriate professions in the examina-
tion of policy matters pertaining to the health of the public. The Institute acts under the
responsibility given to the National Academy of Sciences by its congressional charter to
be an adviser to the federal government and, upon its own initiative, to identify issues
of medical care, research, and education. Dr. Victor J. Dzau is president of the Institute of
Medicine.
The National Research Council was organized by the National Academy of Sciences in
1916 to associate the broad community of science and technology with the Academy’s
purposes of furthering knowledge and advising the federal government. Functioning in
accordance with general policies determined by the Academy, the Council has become
the principal operating agency of both the National Academy of Sciences and the
National Academy of Engineering in providing services to the government, the public,
and the scientific and engineering communities. The Council is administered jointly by
both Academies and the Institute of Medicine. Dr. Ralph J. Cicerone and Dr. C. D. Mote,
Jr., are chair and vice chair, respectively, of the National Research Council.
www.national-academies.org


COMMITTEE ON INDEPENDENT SCIENTIFIC REVIEW OF

EVERGLADES RESTORATION PROGRESS1
JEFFREY R. WALTERS, Chair, Virginia Polytechnic Institute and State University,

Blacksburg
MARY JANE ANGELO, University of Florida, Gainesville
DAVID B. ASHLEY, University of Nevada, Las Vegas
LORETTA L. BATTAGLIA, Southern Illinois University, Carbondale
WILLIAM G. BOGGESS, Oregon State University, Corvallis
CHARLES T. DRISCOLL, Syracuse University, New York
PAUL H. GLASER, University of Minnesota, Minneapolis
WILLIAM L. GRAF, University of South Carolina, Columbia
STEPHEN G. MONISMITH, Stanford University, Stanford, California
DAVID H. MOREAU, University of North Carolina, Chapel Hill
K. RAMESH REDDY, University of Florida, Gainesville
HELEN REGAN, University of California, Riverside
JAMES E. SAIERS, Yale University, New Haven, Connecticut
DANIEL SIMBERLOFF, University of Tennessee, Knoxville
NRC Staff
STEPHANIE E. JOHNSON, Study Director, Water Science and Technology

Board
DAVID J. POLICANSKY, Scholar, Board on Environmental Studies and
Toxicology
MICHAEL J. STOEVER, Research Associate, Water Science and Technology
Board
SARAH E. BRENNAN, Program Assistant, Water Science and Technology
Board
The activities of this committee were overseen and supported by the National Research Council’s
1
Water Science and Technology Board and Board on Environmental Studies and Toxicology (see
Appendix C for listing). Biographical information on committee members and staff is contained in
Appendix D.
v

Acknowledgments
Many individuals assisted the committee and the National Research Council
staff in their task to create this report. We would like to express our appreciation
to the following people who have provided presentations to the committee and
served as guides during the field trips:
Nick Aumen, U.S. Geological Survey

Ernie Barnett, South Florida Water Management District
Drew Bartlett, Florida Department of Environmental Protection
Terrie Bates, South Florida Water Management District
Laura Brandt, U.S. Fish and Wildlife Service
Eric Bush, U.S. Army Corps of Engineers
Eric Cline, South Florida Water Management District
Allen Dray, U.S. Department of Agriculture
Jenny Eckles-Ketterlin, Florida Fish and Wildlife Conservation Commission
James Erskine, Miccosukee Tribe of Indians
Shannon Estenoz, U.S. Department of the Interior
Rory Feeney, Miccosukee Tribe of Indians
Jeremiah Foley, U.S. Department of Agriculture
Manley Fuller III, Florida Wildlife Federation
Rebekah Gibble, U.S. Fish and Wildlife Service
Howie Gonzales, U.S. Army Corps of Engineers
Susan Gray, South Florida Water Management District
David Hobbie, U.S. Army Corps of Engineers
Don Jodrey, U.S. Department of the Interior
Bob Johnson, U.S. National Park Service
Dan Kimball, U.S. National Park Service
Steve Kopecky, U.S. Army Corps of Engineers
Ellen Lake, U.S. Department of Agriculture
Glenn Landers, U.S. Army Corps of Engineers
vii
viii Acknowledgments
Jon Lane, U.S. Army Corps of Engineers

François Laroche, South Florida Water Management District
Tom MacVicar, MacVicar Consulting
Cherise Maples, Seminole Tribe of Indians
Susan Markley, Miami Dade County
Ernie Marks, Florida Department of Environmental Protection
Melissa Martin, U.S. Fish and Wildlife Service
Christen Mason, U.S. Fish and Wildlife Service
Frank Mazzotti, University of Florida
Agnes McLean, U.S. National Park Service
Melissa Meeker, South Florida Water Management District (formerly)
Gail Mitchell, U.S. Environmental Protection Agency
Temperince Morgan, South Florida Water Management District
Matt Morrison, South Florida Water Management District
Melissa Nasuti, U.S. Army Corps of Engineers
Jayantha Obeysekera, South Florida Water Management District
Bob Pace, U.S. Fish and Wildlife Service
Tony Pernas, Everglades National Park
Ellen Pokorny, U.S. Department of Agriculture
Paul Pratt, U.S. Department of Agriculture
Bob Progulske, U.S. Fish and Wildlife Service
Gina Ralph, U.S. Army Corps of Engineers
LeRoy Rodgers, South Florida Water Management District
David Rudnick, U.S. National Park Service
Larry Schwartz, South Florida Water Management District
Mark Shafer, U.S. Army Corps of Engineers
Lauren Sher, Office of U.S. Senator Bill Nelson
Dawn Shirreffs, Everglades Coalition
Fred Sklar, South Florida Water Management District
Melissa Smith, U.S. Department of Agriculture
Eric Summa, U.S. Army Corps of Engineers
Kim Taplin, U.S. Army Corps of Engineers
Tom Teets, South Florida Water Management District
Philip Tipping, U.S. Department of Agriculture
Tom Van Lent, Everglades Foundation
Robert Verrastro, South Florida Water Management District
Greg Wheeler, U.S. Department of Agriculture
David Wegner, U.S. House of Representatives Committee on Transportation and
Infrastructure
Walter Wilcox, South Florida Water Management District
Kevin Wittmann, U.S. Army Corps of Engineers

Preface
To much of the public, the Everglades is Marjory Stoneman Douglas’s

River of Grass—an immense, unique marsh teeming with life, represented
and protected in the form of Everglades National Park. As usual, reality is less
rosy, and more complicated and interesting, than the ideal. The South Florida
ecosystem is vast, stretching more than 200 miles from Orlando to Florida Bay,
and E verglades National Park is but a part located at the southern end. It is a
diverse and distinctive ecosystem that includes not only marshes, but also the
meandering Kissimmee River and associated floodplain and chain of small lakes,
the much larger Lake Okeechobee, sawgrass plains, ridge-and-slough wetlands,
tree islands, marl prairies, bays, and estuaries. During the 19th and 20th cen-
turies the ecosystem changed as the nation changed. The historical Everglades
has been reduced to half of its original size, and what remains is not the pristine
ecosystem many imagine it to be, but one that has been highly engineered and
otherwise heavily influenced, and is intensely managed by humans. Today the
Everglades is not only an iconic natural system, but also the source of water
for industry and the millions of residents of South Florida. To address the floods
that have occasionally devastated the region, water now moves through a maze
of canals, levees, pump stations, and hydraulic control structures, rather than
slowly flowing southward in a broad river of grass, and a substantial fraction
is diverted from the natural system (see NRC, 2010). The water that remains is
polluted by phosphorus and other contaminants originating from agriculture
and other human activities. Many components of the natural system are highly
degraded, and continue to degrade (see NRC, 2012a).
Recognizing the degradation of the South Florida ecosystem, and the depen-
dence of humans upon a functioning ecosystem, in 1999 the State of Florida and
the federal government agreed to a multidecadal, multi-billion-dollar Compre-
hensive Everglades Restoration Plan (CERP) to protect and restore the remaining
Everglades while addressing demands for water supply and flood control. In
authorizing the CERP, the U.S. Congress mandated periodic independent reviews
ix
x Preface
of progress toward restoration of the Everglades natural system. The National

Research Council’s (NRC’s) Committee on Independent Scientific Review of
Everglades Restoration Progress, or CISRERP, was formed for this purpose in
2004. This report, which is the fifth in a series of biennial evaluations that are
expected to continue for the duration of the CERP, reflects the concerted efforts
of 14 committee members and 4 NRC staff representing a wide range of scien-
tific and engineering expertise. Our committee met five times over a period of
16 months including three times in Florida and once in Washington, D.C. We
reviewed a large volume of written material and heard oral presentations from
state, federal, and tribal government personnel, academic researchers, interest
groups, and members of the public.
The CERP is a complex, multi-billion-dollar project managed by the U.S.
Army Corps of Engineers (USACE) and the South Florida Water Management
District (SFWMD) that was projected to require 40 years for completion. With 68
separate project components requiring sophisticated scientific knowledge of the
ecosystem and creation of new technologies for water management, the CERP
represents a research, planning, implementation, and construction challenge
unlike any other. At this writing, the CERP is nearly halfway through its second
decade, and in that time the ecosystem has continued to change as the nation
and indeed the planet changes. This report presents the committee’s consensus
view of restoration accomplishments and emerging challenges primarily during
the past 2 years but also over the 14 years since the CERP was authorized. In
discussing accomplishments, we focus on the progress made on the ground on
several CERP projects and supporting non-CERP projects that are producing the
first increments of restoration and learning progress to improve the restoration
plan through pilot projects and adaptive management. The emerging challenges
on which we focus are those posed by the ways in which the ecosystem is
changing. The Central Everglades Planning Project is an exciting accomplish-
ment that provides the means to accelerate the pace of restoration in the central
Everglades and thus address the ongoing degradation of that part of the eco-
system that was a focus of our last report (NRC, 2012a). However, this critical
project can only fulfill its potential if implemented in a timely way and to do so
will require finding creative solutions to overcome current constraints related to
authorization, funding, and water quality permitting. Climate change and sea-
level rise pose enormous challenges to a rainfall-driven system characterized by
a low elevational gradient, challenges that could perhaps be set aside for later
consideration in 1999, but not in 2014. Rather than compromising restoration,
climate change and sea-level rise provide even more incentive for restoring the
Everglades ecosystem. Indeed, in this context the CERP can be viewed as a water
sustainability plan for both the natural and human environments. Finally, South
Florida is now home to a plethora of species that were not present in the pre-

Preface xi
drainage ecosystem of the 19th century, and more are arriving every year. These
nonnative invasive species pose another important challenge to restoration.
It has been my privilege to serve on this committee with some of the nation’s
leading experts in biological, hydrologic, and geographic sciences, hydrologic
and systems engineering, project administration, law, and policy. I greatly appre-
ciate the time, attention, and thought each committee member invested in under-
standing the complexity of the Everglades ecosystem and corresponding scope
of the CERP. I also appreciate their careful, rigorous analyses, expert judgment,
constructive comments and reviews, and the professionalism, collegiality, and
good humor with which they conducted their business.
The committee is indebted to many individuals for their contributions of
information and resources. Specifically, we appreciate the efforts of the com-
mittee’s technical liaisons—David Tipple (USACE), Glenn Landers (USACE),
Larry Gerry (SFWMD), and Robert Johnson (Department of the Interior)—who
responded to numerous information requests and helped the committee utilize
the vast resources of agency expertise when needed. Many others educated the
committee on the complexities of Everglades restoration through their presenta-
tions, field trips, and public comments (see Acknowledgments).
The committee had the good fortune to be assisted by four dedicated and
talented NRC staff: Stephanie Johnson, David Policansky, Michael Stoever,
and Sarah Brennan. Stephanie Johnson has served as senior project officer
for all five CISRERP panels and is a true Everglades expert. Her encyclopedic
knowledge and understanding of the science, engineering, and administrative
aspects of the CERP, ability to identify and synthesize the complex interrelation-
ships among these aspects, deft management skills, and contacts were critical
to the committee’s success. NRC scholar David Policansky is also a veteran of
all five CISRERP panels and his experience, knowledge, understanding, sage
observations, and illuminating questions were instrumental to the committee’s
deliberations and understanding of the Everglades ecosystem and the CERP.
Michael Stoever attended to the complex logistical needs of the committee,
provided superb support during and between meetings, and, with assistance
from Sarah Brennan, was instrumental in producing the final report. I know I
speak for the entire committee in expressing our profound respect and apprecia-
tion for the NRC staff’s exceptional support and good humor.
This report was reviewed in draft form by individuals chosen for their
breadth of perspectives and technical expertise in accordance with the proce-
dures approved by the National Academies’ Report Review Committee. The
purpose of this independent review was to provide candid and critical com-
ments to assist the institution in ensuring that its published report is scientifically
credible and that it meets institutional standards for objectivity, evidence, and
responsiveness to the study charge. The reviewer comments and draft manuscript

xii Preface
remain confidential to protect the deliberative process. We thank the following

reviewers for their helpful suggestions, all of which were considered and many
of which were wholly or partly incorporated in the final report: G. Ronnie Best,
U.S. Geological Survey (retired); John R. White, Louisiana State University;
M. Siobhan Fennessy, Kenyon College; Evelyn Gaiser, Florida International
University; Julie Lockwood, Rutgers University; John C. Volin, University of
Connecticut; Wendy Graham, University of Florida; Ben Kirtman, University
of Miami; and W. Allen Marr, Jr., Geocomp Corporation.
Although these reviewers provided many constructive comments and sug-
gestions, they were not asked to endorse the conclusions and recommendations
nor did they see the final draft of the report before its release. The review of this
report was overseen by Kenneth W. Potter, University of Wisconsin, M adison,
and Bonnie McCay, Rutgers University. Appointed by the NRC, they were
responsible for making certain that an independent examination of this report
was carried out in accordance with institutional procedures and that all review
comments received full consideration. Responsibility for the final content of this
report rests entirely with the authoring committee and the NRC.
The CERP is a bold, challenging, and complex plan with great potential to
provide benefits to the ecosystem and the public, and the small increments of
restoration that have been achieved suggest that that potential can be realized.
But the time has come for equally bold action in implementing the CERP. Delays
in implementation make project costs higher, and the ecosystem degradation
that must be addressed larger. The challenges to implementation that exist can
be overcome, and in the case of climate change, make implementation more
urgent. We offer this report in support of that endeavor.
Jeffrey R. Walters, Chair

Committee on Independent Scientific Review
of Everglades Restoration Progress (CISRERP)

Contents
SUMMARY 1
1 INTRODUCTION 13
2 THE RESTORATION PLAN IN CONTEXT 21
3 CENTRAL EVERGLADES PLANNING PROJECT 35
4 IMPLEMENTATION PROGRESS 71
5 CLIMATE CHANGE AND SEA-LEVEL RISE: IMPLICATIONS

FOR EVERGLADES RESTORATION 131
6 BIOLOGICAL INVASIONS AND EVERGLADES RESTORATION 169
7 USE OF SCIENCE IN DECISION MAKING 213
REFERENCES 229
ACRONYMS 257
APPENDICES
A National Research Council Everglades Reports 261
B Additional Major Nonnative Plant and Animal Species in the
Everglades 269
C Water Science and Technology Board; Board on Environmental
Studies and Toxicology 279
D Biographical Sketches of Committee Members and Staff 281
xiii

Summary
The Florida Everglades, one of the world’s treasured ecosystems, has been
dramatically altered over the past century by an extensive water control infra-
structure, designed to increase regional economic productivity through improved
flood control, urban water supply, and agricultural production. The remnants of
the original Everglades now compete for vital water with urban and agricultural
interests and are impaired by contaminated runoff from these two activities. The
Comprehensive Everglades Restoration Plan (CERP), a joint effort led by the state
and the federal government and launched in 2000, seeks to reverse the decline
of the ecosystem. This $13.5 billion project was originally envisioned as a 30- to
40-year effort to achieve ecological restoration by restoring the hydrologic char-
acteristics of the Everglades, where feasible, and to create a water system that
serves the needs of both the natural and the human systems of South Florida
(Figure S-1).
The National Research Council established the Committee on Independent
Scientific Review of Everglades Restoration Progress in 2004 in response to a
request from the U.S. Army Corps of Engineers (USACE), with support from the
South Florida Water Management District (SFWMD) and the U.S. Department
of the Interior, based on Congress’s mandate in the Water Resources Develop-
ment Act of 2000 (WRDA 2000). The committee is charged to submit biennial
reports that review the CERP’s progress in restoring the natural system. This is
the committee’s fifth report in a series of biennial evaluations. Each biennial
report provides an update on natural system restoration progress over the previ-
ous 2 years, describes significant accomplishments (Chapter 4), and addresses
important developments in research, monitoring, and assessment that inform
restoration decision making (Chapter 7). In each new report, the committee also
identifies issues for in-depth evaluation in light of new CERP program develop-
ments, policy initiatives, or improvements in scientific knowledge that have
implications for restoration progress (see Chapter 1 for the committee’s full state-
ment of task). For this 2014 biennial review, the committee examined the Central
1
2 Progress Toward Restoring the Everglades
FIGURE S-1 The South Florida ecosystem, which shares the same boundaries as the South
Florida Water Management District.
SOURCE: © International Mapping Associates

Summary 3
Everglades Planning Project (Chapter 3) and the implications of climate change

(Chapter 5) and invasive species (Chapter 6) for Everglades restoration efforts.
OVERALL EVALUATION OF PROGRESS AND CHALLENGES
During the past 2 years, Everglades restoration has been defined by excep-
tional project planning accomplishments with substantial restoration poten-
tial on the one hand; and increasingly frustrating financial, procedural, and
policy constraints impeding project implementation on the other. All of this has
occurred against the backdrop of modest restoration progress focused along
the edges of the Everglades, considerable state efforts to improve water quality,
ongoing degradation of the core Everglades, and increasing restoration threats
posed by sea-level rise and invasive species.
The Central Everglades Planning Project is an impressive strategy to expedite
restoration and avert further degradation of the central Everglades. The Everglades
also appears on the threshold of other significant advances in natural system
restoration, particularly from several key non-CERP projects. However, project
authorization, funding levels, and cost-sharing complexities have impeded the
completion of important ongoing restoration projects, and water quality permit-
ting constraints represent serious impediments to near-term implementation of
the critical Central Everglades Planning Project. Timely authorization, adequate
funding levels, and creative policy and implementation strategies are essential
to realize important short-term restoration benefits, but more importantly to
expedite implementation of the Central Everglades Planning Project in order
to realize its substantial restoration potential.
Climate change provides additional incentives to expedite restoration. The
CERP provides important means to help mitigate the impacts of sea-level rise
and precipitation and temperature changes by enhancing ecosystem resilience,
promoting peat accretion, and reducing saltwater intrusion. Implementation
priorities should be revised to focus resources on those projects with the greatest
potential to avert ecosystem degradation and provide long-term benefits con-
sidering sea-level rise and potential changes in temperature and precipitation.
Restoration planners need to examine critical unknowns related to sea-level-rise
impacts on the Everglades, assess climate projections as they improve over time,
analyze their implications for restoration, and design for flexibility w herever
feasible.
Planners must also remain cognizant of the potential impacts of invasive
species on the success of the CERP. Additional strategic coordination is needed to
prioritize invasive species management and research resources while maintain-
ing an aggressive early detection and rapid response system. The report’s major
conclusions and recommendations are summarized below.

CENTRAL EVERGLADES PLANNING PROJECT
The Central Everglades Planning Project effort is responsive to the com-

mittee’s prior recommendations to expedite restoration in the central Ever-
glades via increments of restoration to avert further declines that could take
many decades or longer to recover. Overall, the project team did an impressive
job under a challenging time frame. The proposed plan seems reasonable and
thoughtfully developed with substantial stakeholder input. Implementation of the
plan would provide significant benefits to the remnant Everglades ecosystem,
including more than 200,000 acre-feet/yr of new water—a sizeable first incre-
ment of restoration for the central Everglades that represents approximately
two-thirds of the new water to northern Water Conservation Area 3 envisioned
in the CERP. A comprehensive adaptive management plan provides an important
mechanism to learn from project implementation to improve the operation of the
project and the design of future increments of restoration, although additional
attention to climate change uncertainties is needed.
If the Central Everglades Planning Project is to avert further ecosystem
degradation, CERP planners and policy makers need to expedite project imple-
mentation in the face of several hurdles. The best-laid plans will be of little
benefit if the project is not implemented in a timely way. Completion of the
Chief of Engineers’ Report for the Central Everglades Planning Project, con-
gressional authorization, and construction of project dependencies are key
near-term steps necessary to move forward. Project funding and water quality
permitting constraints currently appear to be the largest barriers to timely project
implementation. Creative solutions may be available to significantly expedite
restoration, such as finding permit mechanisms to move water that meets water
quality criteria into the Everglades prior to completion of the entire Restoration
Strategies project. Such approaches will require the agencies to recognize the
urgency and to work to find legal and engineering solutions to move increments
of water into the Everglades as soon as those increments have been adequately
treated to meet water quality standards. Without such solutions, redistribution of
existing water may not be feasible until 2035 or beyond, and at the envisioned
funding level of $100 million per year, construction would not be completed
for approximately four decades—exceedingly long for a system already in sig-
nificant decline.
Some important lessons were learned from the expedited planning process.
Although overall, participants and stakeholders thought the process led to a
useful outcome, the 22-month planning time frame was extremely challenging
for staff and stakeholders, alike. The process required large numbers of staff
and became the central focus of the restoration program. Such attention was
deserved for this high-priority initiative, but similar intensive efforts are unlikely

Summary 5
to be sustainable for future CERP planning. Furthermore, stakeholders with

technical expertise found it difficult to keep up with the pace of model output
presented and hence could not adequately evaluate the information provided.
Thus, 3 years might be a more reasonable time frame for such a complex effort.
Communication within and between agencies was a particular strength of the
expedited process; senior decision makers were involved at key decision points
and as needed to resolve issues and improve planning efficiency. However, the
existing USACE process for evaluating restoration benefits makes it difficult to
be transparent about tradeoffs in planning decisions.
The enhanced stakeholder and public engagement process was well exe-
cuted and should serve as a model for future planning processes. This level of
active and inclusive stakeholder engagement had not previously been imple-
mented as part of the CERP, and it provided a means for two-way dialogue
between stakeholders and agency staff that substantially influenced the planning
outcome. Although the abbreviated time frame led to concerns from partici-
pants, overall, the committee commends the efforts to educate and engage the
stakeholders and modify the project plan where feasible to address concerns.
RESTORATION PROGRESS
The infrequency of Water Resources Development Acts (WRDAs) has

impeded CERP progress over the past 2 years. Seven years have elapsed since
the last WRDA was passed, and four Generation 2 CERP projects with approved
project plans awaited congressional authorization between 2012 and June 2014
when the Water Resources Reform and Development Act of 2014 (WRRDA
2014) was signed into law. Additionally, two of the previously authorized CERP
projects require reauthorization due to cost escalations; thus, prior to WRRDA
2014, only one CERP project—Indian River Lagoon South—was eligible for size-
able (>$25 million) construction funding. With the passage of WRRDA 2014,
four additional projects are able to proceed with federal funding, although the
Central Everglades Planning Project was not completed in time to be included.
Lack of authorizations also had important implications for the cost-share bal-
ance, discussed below.
Availability of funding also impeded CERP progress in the past 2 years.
State CERP expenditures have declined substantially in recent years, because of
reduced SFWMD revenues and the need to fund non-CERP water quality projects
to meet a 2012 Consent Order. Even though the state has spent significantly
more than the federal government on the CERP since its inception, the state has
been precariously close to the mandated 50-50 cost-share requirement because,
prior to WRRDA 2014, land acquisition and construction expenditures could
only be credited for the four congressionally authorized Generation 1 projects.

Declining state funding for CERP projects over the past 2 years has contributed
to cost-sharing challenges, and as of September 2013, the state’s “creditable
expenditures” exceeded those of the federal government by only $98 million.
As a result, the federal government significantly reduced spending in FY 2014
so as not to exceed the 50-50 cost share. Passage of WRRDA 2014 could allow
the state to realize approximately $400 million in additional cost-sharing credits
for prior spending, thereby easing an impending constraint on federal contribu-
tions toward the CERP.
CERP planners need to revisit the Integrated Delivery Schedule with a
renewed urgency to advance projects with the greatest potential to avert
ongoing ecosystem degradation and those that promise the largest restoration
benefits. The current draft Integrated Delivery Schedule has not been updated
since 2011, and difficult decisions will need to be made to integrate the four
Generation 2 CERP projects and the Central Everglades Planning Project (and
related project dependencies) with existing CERP and non-CERP efforts. To
expedite Everglades restoration amid limited funding, all authorized projects
cannot be advanced equally. Some projects may be more beneficial in light of
climate change and sea-level rise and others less so, and these factors should
be considered in the prioritization of restoration funding.
The restoration progress made by CERP projects to date remains fairly
modest in scope. Ecosystem responses have been detected after phased imple-
mentation in the Picayune Strand, Biscayne Bay Coastal Wetlands, and C-111
Spreader Canal projects, although many of these improvements are limited. In
some cases, such as Biscayne Bay, the scope of the restoration increment to
date is simply so limited in area that ecological responses are equally small. In
other cases, such as Picayune Strand, additional time may be needed to achieve
full ecosystem responses to the restoration measures in place. Taylor Slough
has seen significant hydrologic improvements due to restoration efforts, but the
documented benefits to date are primarily derived from the C-111 South Dade
Project, a non-CERP project. For all three of these projects, ecological responses
would be expected to increase with construction and operation of additional
project increments as well as additional time for ecosystem recovery.
Several non-CERP projects have faced bureaucratic and policy issues that
hindered implementation progress. Agency disagreements about cost-sharing
arrangements and legal requirements affected progress on the Kissimmee River
Restoration and the C-111 South Dade project by delaying them for almost
2 years. However, the SFWMD and the USACE have made important progress
to resolve these differences and resume construction. Meanwhile, water quality
compliance issues and the lack of an operational plan are preventing realiza-
tion of restoration benefits in the Mod Waters project. Scientific knowledge is
adequate for success, and engineering problems in construction and opera-

Summary 7
tion appear not to be impeding restoration progress. These non-CERP founda-

tion projects offer large potential restoration benefits once fully implemented.
Renewed attention is needed to resolve the remaining bureaucratic challenges to
expedite restoration progress and realize the ecological returns from substantial
financial investments to date.
STA performance shows signs of improvement under recent management.
Long-term sustainable performance, however, will be directly influenced by
loading rates. Additional treatment-area and flow-equalization basins in the Res-
toration Strategies project are likely to further reduce loading rates and outflow
concentrations. Continued adaptive management, including implementation of
new strategies developed through ongoing research, is needed to meet water
quality standards and to sustain performance of these treatment systems.
CLIMATE CHANGE AND SEA-LEVEL RISE: IMPLICATIONS FOR RESTORATION
Climate change provides a strong incentive for accelerating restoration.

Current impacts of rising sea levels are a harbinger of future climate change
effects on the functioning and structure of the Everglades ecosystem and the eco-
system services on which South Florida depends. Sea-level rise in South Florida
is already increasing saltwater intrusion into Everglades freshwater habitats and
urban water supplies, and future climate changes are likely to be manifested
through changes in the timing, volume, and quality of freshwater; distributions
of species; and the extent of wetland habitats. Climate change is also expected
to increase agricultural water demands, which when paired with anticipated
population growth, highlights the potential regional water supply challenges in
South Florida under future scenarios. Everglades restoration enhances the ability
of the ecosystem to withstand and adapt to future changes and increases water
availability to the ecosystem and to urban and agricultural users. Improvements
in Everglades water depths promote higher rates of peat accretion that could
help mitigate the effects of sea-level rise and reduce the impacts of saltwater
intrusion on urban water supplies.
Although the projections are uncertain, significant changes in precipitation
and temperature coupled with increasing sea level have important implications
for the CERP. The Everglades landscape is especially sensitive to sea-level rise,
and rates of sea-level rise in South Florida are predicted to increase. A scenario
of 1.5-degree increase in temperature and a 10 percent decrease in precipitation
together with anticipated sea-level rise results in significant changes in coastal
ecosystems and insufficient freshwater to sustain the natural and built systems. To
decrease uncertainty associated with precipitation projections and clarify future
risk, global climate model projections of intra-annual, annual, and interannual
variability in precipitation and temperature need to be improved and refined.

These improved climate projections should, in turn, be used by CERP planners

as input to drive Everglades hydrologic models suitable for making inferences
on year-to-year and seasonal variations in freshwater availability.
Climate change is not adequately considered in the CERP planning process
and should be integrated into future ongoing analysis and monitoring. CERP
project designs are based on historical hydrology and have not been assessed in
the context of future precipitation and evapotranspiration scenarios. Currently,
only sea-level rise is considered in CERP planning and usually only as a cursory
analysis at the end of the process to assess loss of benefits through 2050 with
wetland inundation resulting from sea-level rise. The lack of consideration of the
effects of climate change paints an incomplete picture of hydrologic and ecosys-
tem response to the alternatives examined and ignores the potential benefits of
the projects to help mitigate the impacts of climate change. Additionally, hydro-
logic restoration goals are based on the natural systems model, which reflects
the past 50 years rather than any likely future. Depending on future climate
change, some hydrologic or ecological restoration goals may be unattainable
or prove to be not cost-effective. Urban and agricultural water demands unmet
under dire climate scenarios highlight the need for additional analysis of water
sustainability for the natural and built systems.
CERP planners should consider the implications of sea-level rise and poten-
tial hydrologic change in systemwide planning and project prioritization.
Likely sea-level-rise projections can be used to evaluate future project ben-
efits, considering uncertainties regarding the potential for accretion in coastal
and inland wetlands to mitigate these effects. Sea-level-rise scenarios should
also be coupled with hydrologic change scenarios to characterize systemwide
response to global change. The outcome of these analyses would inform future
systemwide decisions of project prioritization. Re-prioritization should include
consideration of both those rendered less important and less effective in light of
reduced benefits in the context of climate change and sea-level rise and those
projects that become more essential to enhance the ability of the ecosystem and
the built environment to adapt to changes and mitigate the effects of changing
climate.
Anticipating future changes in temperature, precipitation, and sea-level
rise, CERP planners should, where feasible, design for flexibility. Climate change
needs to be incorporated into adaptive management planning, at both project-
scale and when considering systemwide goals. It is likely that additional water
storage will be needed to address anticipated future increases in variability of
meteorological conditions. As new knowledge becomes available, it needs to
be incorporated into the CERP adaptive management framework so that man
agers can adjust future restoration efforts appropriately as the nature of changes
in climate become more evident. In addition, the current monitoring program

Summary 9
should be evaluated to ensure that important effects of climate change will be

characterized and quantified.
The committee identified several high-priority research needs related to
climate change and Everglades restoration:
• Assess the rates of peat/sediment accretion and subsidence in coastal and

inland freshwater wetlands in the context of sea-level rise;
• Improve modeling tools that can be used to assess the effects of projected
sea-level rise on groundwater supplies and coastal ecosystem functioning and
examine the potential for the CERP to mitigate these effects;
• Improve, refine, and evaluate downscaled climate model projections in
the context of South Florida water resources and Everglades restoration;
• Improve the understanding of factors that could help maintain the diverse
mosaic of Everglades habitats and increase their resilience amid changes in
climate and sea level; and
• With improved climate and sea-level projections, reevaluate the goals for
Everglades restoration and develop alternative goals as appropriate.
INVASIVE NONNATIVE SPECIES
Despite excellent progress in developing coordination of the management

of invasive species at the operational level, there is a lack of coordination at
a strategic level that includes a comprehensive view of all nonnative species
in all parts of the greater Everglades. Currently, plants and animals tend to
be considered separately. Management and restoration activities need to take
account of the entire biotic community and not be partitioned into different
taxa. For many invasive species, different agencies take on management activi-
ties in different areas, yet individuals of such species move between areas, so
that management in one area can impact other areas. These factors argue for
the creation of a high-level coordinative entity to oversee policy, management,
and budgets related to nonnative species. Prioritization of research needs and
control efforts across areas, species, habitats, and agencies would be a major
responsibility of this entity. The committee is optimistic that the Comprehensive
Invasive Species Strategic Action Framework being developed by the South
Florida Ecosystem Task Force will be a major step toward achieving these goals
of high-level coordination.
A strategic early detection and rapid response (EDRR) system that addresses
all areas, habitats, and species is needed. EDRR is an essential strategy if new
invasions of nonnative species in the Everglades are to be eradicated (or at least
contained) while it is still feasible and relatively inexpensive to do so. Several
EDRR efforts are under way, but the current level of monitoring is insufficient

to address the geographic extent and range of nonnative species threats in the
Everglades. In general, a rapid response requires quick access to resources, but
efforts to eradicate incipient invasions in the Everglades have more often been
stymied by the inability to obtain funds from federal, state, or local sources.
The costs of additional monitoring and response should be weighed against
the likely benefits of finding and acting on early invasions. Additional funding
would allow for greater public outreach, expanded operation of the reporting
hotline, increased early detection monitoring, and improved capacity for rapid
response to facilitate eradication. The committee recognizes that the goal of
this recommendation—addressing all areas, habitats, and species—likely is
beyond any reasonable expectation of resources, but keeping this goal in mind
emphasizes the value of prevention and clarifies the magnitude of the challenge.
There is no systemwide mechanism for prioritizing research on and man-
agement of invasive species. Many agencies participating in the Everglades
restoration already undertake research activities on certain nonnative species
and also implement management activities, but these efforts are limited by
insufficient resources and are typically driven by specific agency needs rather
than systemwide priorities. Effective prioritization requires a comprehensive
understanding of all nonnative species present in the Everglades, their impacts
and threats, as well as those of impending or likely new arrivals.
Research is lacking on nonnative species and their impacts to adequately
inform prioritization efforts. Many knowledge gaps exist about species consid-
ered to be priorities for management. Given the spatial extent of the problem
and the threats of future invasions, substantial research is needed to assess the
various impacts of nonnative species on ecosystem functioning and native spe-
cies and to develop or improve control mechanisms. This does not mean com-
prehensive research on all details of the biology and effects of every nonnative
species. Rather, enough basic information should be gathered systematically
to determine which species could reasonably be predicted to have consider-
able ecological impacts. Such knowledge is important in guiding decisions on
detailed research on possible impacts and management of particular threats and
would help inform priorities for management actions.
If eradication proves impossible, maintenance management and long-
term control at acceptable levels should be explicitly recognized as a goal in
some cases. Indeed, current practice seems to implicitly recognize this goal.
Maintenance management at low densities is sometimes possible by various
combinations of biological, chemical, mechanical, and physical controls. In
the E verglades, a striking example is the current management of melaleuca
(Melaleuca quinquenervia), once thought too widespread and dense to be
manageable. As a result of sustained intensive research, this species is currently
under substantial control in most regions through a combination of mechanical,

Summary 11
chemical, and biological control as well as prescribed burns. Maintenance man-

agement requires continued, diligent monitoring and flexible, but reliable fund-
ing that can be devoted strategically to achieve and maintain long-term control.
At every step of the CERP planning process, full consideration is needed
of the implications of restoration activities for nonnative species and their
impacts. Until very recently, invasive species have not been considered in CERP
project planning and implementation beyond simply removing any invasive
species encountered at construction sites. Ideally, hydrologic restoration should
favor the reestablishment and expansion of many native wetland species that
are better adapted to longer hydroperiods. However, aquatic and flood-tolerant
nonnative species may also benefit and replace native species. Removing levees
and filling in canals may, in certain circumstances, facilitate the spread of non-
native species by increasing their potential for dispersal. For each CERP project,
the potential to increase the spread of invasive species should be examined and
the effects on ecosystem functioning assessed. Based on this analysis, strategies
and technologies to lessen these impacts should be appropriately considered.
Recent CERP guidance and plans to implement national USACE invasive species
policy indicate that these considerations are increasingly being incorporated
into project planning and implementation, although it is too soon to evaluate
this new approach.
Long-term monitoring and research are needed to understand the poten-
tial impacts of climate change on Everglades nonnative species management.
Climate change has the potential to significantly impact the distributions and
abundances of nonnative species in the Everglades and their impacts on the
ecosystem as a whole. Thus, research and monitoring to understand long-term
changes in nonnative species distribution and behavior and the effectiveness
of maintenance control strategies in the context of climate change are needed.
SCIENCE AND DECISION MAKING
Useful long-term systemwide monitoring requires stable funding. If funding

cuts result in significant gaps in critical long-term monitoring data, important
changes and patterns could be missed, and data collected prior to or after the
funding gaps could lose their value. Given the substantial financial investment in
Everglades restoration by both the state and the federal government, a dedicated
source of funding could provide ongoing long-term systemwide monitoring and
assessment that are critical to meeting restoration objectives, ensuring that public
resources are spent wisely and adaptive management is supported.
A comprehensive reevaluation of restoration-related monitoring is needed
to determine its adequacy considering budget pressures, the extended CERP
implementation time frames, and the potential impacts of climate change and

sea-level rise. The dramatic 2011 cuts to Monitoring and Assessment Plan fund-
ing create a risk that adequate long-term data will not be available to assess
the effects of restoration projects in a systemwide context once they are imple-
mented. This reevalution should clearly articulate the value of the highest priority
monitoring to future restoration decision making and the risks of ceasing such
monitoring. Also, CERP planners should identify opportunities for improving the
efficiency of current monitoring and reducing the frequency of some monitoring
in the context of the current slow pace of CERP implementation.
Renewed attention to science coordination is warranted. Scientific research
and monitoring programs require coordination and communication to be effec-
tive and efficient, but science leadership and coordination appear to have waned
over the past few years. For the Science Coordination Group to contribute
significantly to better science coordination, it would need to have adequate
funding and staff and a clear charge to address critical science needs from a
restoration-wide perspective.

Introduction
The Florida Everglades, formerly a large and diverse aquatic ecosystem, has
been dramatically altered over the past century by an extensive water control
infrastructure designed to increase regional economic productivity through
improved flood control, urban water supply, and agricultural production (Davis
and Ogden, 1994; NRC, 2005). Shaped by the slow flow of water, its vast terrain
of sawgrass plains, ridges, sloughs, and tree islands supported a high diversity of
plant and animal habitats. This natural landscape also served as a sanctuary
for Native Americans. However, large-scale changes to the landscape have
diminished the natural resources, and by the mid- to late-20th century, many
of the area’s defining natural characteristics had been lost. The remnants of the
original Everglades (see Figure 1-1 and Box 1-1) now compete for vital water
with urban and agricultural interests, and contaminated runoff from these two
activities impairs the South Florida ecosystem.
Recognition of past declines in environmental quality, combined with con-
tinuing threats to the natural character of the remaining Everglades, led to
initiation of large-scale restoration planning in the 1990s and the launch of the
Comprehensive Everglades Restoration Plan (CERP) in 2000. This unprecedented
project envisioned the expenditure of billions of dollars in a multidecadal
effort to achieve ecological restoration by reestablishing the hydrologic char-
acteristics of the Everglades, where feasible, and to create a water system that
simultaneously serves the needs of both the natural and the human systems of
South Florida. Within the social, economic, and political latticework of the 21st
century, restoration of the South Florida ecosystem is now under way and rep-
resents one of the most ambitious ecosystem renewal projects ever conceived.
This report represents the fifth independent assessment of the CERP’s progress
by the Committee on Independent Scientific Review of Everglades Restoration
Progress (CISRERP) of the National Research Council (NRC).
13
FIGURE 1-1 Reconstructed (a) pre-drainage (circa 1850) and (b) current (1994) satellite images of the
Everglades ecosystem.
NOTE: The yellow line in (a) outlines the historical Everglades ecosystem, and the yellow line in (b) outlines
the remnant Everglades ecosystem as of 1994.
Figure 1-1
SOURCE: Courtesy of C. McVoy, J. Obeysekera,
R02233and W. Said, South
(Everglades 4) Florida Water Management District.
© International Mapping Associates
raster iamge
THE NATIONAL RESEARCH COUNCIL AND EVERGLADES RESTORATION
The NRC has been providing scientific and technical advice related to the
Everglades restoration since 1999. The NRC’s Committee on the Restoration of
the Greater Everglades Ecosystem (CROGEE), which operated from 1999 until
2004, was formed at the request of the South Florida Ecosystem Restoration
Task Force (Task Force), an intergovernmental body established to facilitate

Introduction 15
BOX 1-1
Geographic Terms
This box defines some key geographic terms used throughout this report.
• The Everglades, the Everglades ecosystem, or the remnant Everglades

ecosystem refers to the present areas of sawgrass, marl prairie, and other wetlands
and estuaries south of Lake Okeechobee (Figure 1-1b).
• The original, historical, or pre-drainage Everglades refers to the areas of
sawgrass, marl prairie, and other wetlands and estuaries south of Lake Okeechobee
that existed prior to the construction of drainage canals beginning in the late 1800s
(Figure 1-1a).
• The Everglades watershed is the drainage that encompasses the Everglades
ecosystem but also includes the Kissimmee River watershed and other smaller water
sheds north of Lake Okeechobee that ultimately supply water to the Everglades
ecosystem.
• The South Florida ecosystem (also known as the Greater Everglades Eco-
system; see Figure 1-2) extends from the headwaters of the Kissimmee River near
Orlando through Lake Okeechobee and the Everglades into Florida Bay and ultimately
the Florida Keys. The boundaries of the South Florida ecosystem are determined by
the boundaries of the South Florida Water Management District, the southernmost of
the state’s five water management districts, although they approximately delineate the
boundaries of the South Florida watershed. This designation is important and helpful to
the restoration effort because, as many publications have made clear, taking a water
shed approach to ecosystem restoration is likely to improve the results, especially when
the ecosystem under consideration is as water dependent as the Everglades (NRC,
1999, 2004).
The Water Conservation Areas (WCAs) include WCA-1 (the Arthur R. Marshall
Loxahatchee National Wildlife Refuge), WCA-2A and -2B, -3A, and -3B (see Figure 1-2).
The following represent legally defined geographic terms used in this report:
• The Everglades Protection Area is defined in the Everglades Forever Act as

comprising WCA-1, -2A, -2B, -3A, and -3B and Everglades National Park.
• The natural system is legally defined in the Water Resources Development Act of
2000 (WRDA 2000) as all land and water managed by the federal government or the state
within the South Florida ecosystem (see Figure 1-3). “The term ‘natural system’ includes
(i) water conservation areas; (ii) sovereign submerged land; (iii) Everglades National
Park; (iv) Biscayne National Park; (v) Big Cypress National Preserve; (vi) other Federal
or State (including a political subdivision of a State) land that is designated and managed
for conservation purposes; and (vii) any tribal land that is designated and managed for
conservation purposes, as approved by the tribe” (WRDA 2000).
Many maps in this report include shorthand designations that use letters and num-
bers for engineered additions to the South Florida ecosystem. For example, canals
are labeled C-#; levees and associated borrow canals as L-#; and structures, such as
culverts, locks, pumps, spillways, control gates, and weirs, as S-# or G-#.

FIGURE 1-2 The South Florida ecosystem.

Introduction 17
FIGURE 1-3 Land and waters managed by the State of Florida and the federal government
as of December 2005 for conservation purposes within the South Florida ecosystem.
SOURCE: Based on data compiled by Figure

Florida 1-2
State University’s Florida Natural Areas Inven-
R02233 (Everglades
tory (http://www.fnai.org/gisdata.cfm). 4)
© International Mapping Associates
raster iamge

coordination in the restoration effort, and the committee produced six reports
(NRC, 2001, 2002a,b, 2003a,b, 2005). The NRC’s Panel to Review the Critical
Ecosystem Studies Initiative produced an additional report in 2003 (NRC, 2003c;
see Appendix A). The Water Resources Development Act of 2000 (WRDA 2000)
mandated that the U.S. Department of the Army, the Department of the Inte-
rior, and the State of Florida, in consultation with the Task Force, establish an
independent scientific review panel to evaluate progress toward achieving the
natural system restoration goals of the CERP. The NRC’s CISRERP was therefore
established in 2004 under contract with the U.S. Army Corps of Engineers. After
publication of each of the first four biennial reviews (NRC, 2007, 2008, 2010,
2012a; see Appendix A for the report summaries), some members rotated off the
committee and some new members were added.
The committee is charged to submit biennial reports that address the fol-
lowing items:
1. An assessment of progress in restoring the natural system, which is defined

by section 601(a) of WRDA 2000 as all of the land and water managed by the
federal government and state within the South Florida ecosystem (see Figure 1-3
and Box 1-1);
2. A discussion of significant accomplishments of the restoration;
3. A discussion and evaluation of specific scientific and engineering issues
that may impact progress in achieving the natural system restoration goals of
the plan; and
4. An independent review of monitoring and assessment protocols to be
used for evaluation of CERP progress (e.g., CERP performance measures, annual
assessment reports, assessment strategies, etc.).
Given the broad charge, the complexity of the restoration, and the continu-
ally evolving circumstances, the committee did not presume it could cover all
issues that affect restoration progress in any single report. This report builds on
the past reports by this committee (NRC, 2007, 2008, 2010, 2012a) and empha-
sizes restoration progress since 2012, high-priority scientific and engineering
issues that the committee judged to be relevant to this time frame, and other
issues that have impacted the pace of progress. The committee focused particu-
larly on issues for which the “timing was right”—that is, where the committee’s
advice could be useful relative to the decision-making time frames—and on
topics that had not been fully addressed in past NRC Everglades reports. Inter-
ested readers should look to past reports by this committee (NRC, 2007, 2008,
2010, 2012a) to find detailed discussions of important topics, such as the human
context for the CERP, water quality and quantity challenges and trajectories,

Introduction 19
Lake Okeechobee, Modified Water Deliveries to Everglades National Park, and

incremental adaptive restoration, which are not repeated here.
The committee met five times during the course of this review; received
briefings at its public meetings from agencies, organizations, and individuals
involved in the restoration, as well as from the public; and took several field
trips to sites with restoration activities (see Acknowledgments) to help it evalu-
ate restoration progress. In addition to information received at the meetings, the
committee based its assessment of progress on information in relevant CERP
and non-CERP restoration documents. The committee’s conclusions and rec-
ommendations also were informed by a review of relevant scientific literature
and the experience and knowledge of the committee members in their fields of
expertise. The committee was unable to consider in any detail new materials
received after March 2014.
REPORT ORGANIZATION
In Chapter 2, the committee provides an overview of the CERP in the context

of other ongoing restoration activities and discusses the restoration goals that
guide the overall effort.
In Chapter 3, the committee discusses progress in the Central Everglades
Planning Project and presents its evaluation of the effort.
In Chapter 4, the committee analyzes the progress of CERP implementa-
tion, including recent developments at Picayune Strand, Biscayne Bay Coastal
Wetlands, the C-111 Spreader Canal, and Indian River Lagoon-South and several
pilot projects that are under way. Also discussed in the chapter are programmatic
progress and issues, including funding, authorization, and sequencing.
In Chapter 5, the committee discusses the implications of climate change
for Everglades restoration and recommends planning and research needs to
address this issue.
In Chapter 6, the committee examines the impacts of nonnative invasive
species on Everglades restoration, discusses current mechanisms to coordinate
monitoring and control efforts, and recommends additional steps to improve
invasive species control strategies.
In Chapter 7, the committee discusses the contributions and use of science
for CERP decision making. The chapter focuses on science coordination, adap-
tive management, and science support for water quality improvements.


The Restoration Plan in Context
This chapter sets the stage for the fifth of this committee’s biennial assess-
ments of restoration progress in the South Florida ecosystem. Background for
understanding the project is provided through descriptions of the ecosystem
decline, restoration goals, the needs of a restored ecosystem, and the specific
activities of the restoration project.
BACKGROUND
The Everglades once encompassed about 3 million acres of slow-moving

water and associated biota that stretched from Lake Okeechobee in the north to
Florida Bay in the south (Figures 1-1a and 2-1a). The conversion of the Everglades
wilderness into an area of high agricultural productivity and cities was a dream of
19th-century investors, and projects begun between 1881 and 1894 affected the
flow of water in the watershed north of Lake Okeechobee. These early projects
included dredging canals in the Kissimmee River Basin and constructing a channel
connecting Lake Okeechobee to the Caloosahatchee River and, ultimately, the
Gulf of Mexico. By the late 1800s, more than 50,000 acres north and west of the
lake had been drained and cleared for agriculture (Grunwald, 2006). In 1907,
Governor Napoleon Bonaparte Broward created the Everglades Drainage District
to construct a vast array of ditches, canals, dikes, and “improved” channels. By
the 1930s, Lake Okeechobee had a second outlet, through the St. Lucie Canal,
leading to the Atlantic Ocean, and 440 miles of other canals altered the hydrology
of the Everglades (Blake, 1980). After hurricanes in 1926 and 1928 resulted in
disastrous flooding from Lake Okeechobee, the U.S. Army Corps of Engineers
(USACE) replaced the small berm that bordered the southern edge of the lake with
the massive Herbert Hoover Dike, which was eventually expanded in the 1960s to
encircle the lake. The hydrologic end product of these drainage activities was the
drastic reduction of water storage within the system and an increased susceptibility
to drought and desiccation in the southern reaches of the Everglades (NRC, 2005).
21
FIGURE 2-1 Water flow in the Everglades under (a) historical conditions, (b) current conditions, and (c) con-
ditions envisioned upon completion of the Comprehensive Everglades Restoration Plan (CERP).
SOURCE: Graphics provided by USACE, Jacksonville District.
After further flooding inFigure

1947 2-1
and increasing demands for improved agri-
R02233 (Everglades
cultural production and flood control for 4) the expanding population centers
raster iamge
on the southeast Florida coast, the U.S. Congress authorized the Central and
shrink or landscape
Southern Florida (C&SF) Project. This project provided flood control and urban
and agricultural water supply by straightening 103 miles of the meandering
Kissimmee River, expanding the Herbert Hoover Dike, constructing a levee along
the eastern boundary of the Everglades to prevent flows into the southeastern
urban areas, establishing the 700,000-acre Everglades Agricultural Area (EAA)
south of Lake Okeechobee, and creating a series of Water Conservation Areas
(WCAs) in the remaining space between the lake and Everglades National Park
(Light and Dineen, 1994). The eastern levee isolated about 100,000 acres of
the Everglades ecosystem, making it available for development (Lord, 1993).
In total, urban and agricultural development have reduced the Everglades to
about one-half its pre-drainage size (see Figure 1-1b; Davis and Ogden, 1994)
and have contaminated its waters with chemicals such as phosphorus, nitrogen,
sulfur, mercury, and pesticides. Associated drainage and flood control structures,
including the C&SF Project, have diverted large quantities of water to the coastal

The Restoration Plan in Context 23
areas, thereby reducing the freshwater inflows and natural water storage that
defined the ecosystem (see Figure 2-1b).
The profound hydrologic alterations were accompanied by many changes in
the biotic communities in the ecosystem, including reductions and changes in the
composition, distribution, and abundance of the populations of wading birds.
Today, the federal government has listed 67 plant and animal species in South
Florida as threatened or endangered, with many more included on state lists.
Some distinctive Everglades habitats, such as custard apple forests and periph-
eral wet prairie, have disappeared altogether, while other habitats are severely
reduced in area (Davis and Ogden, 1994; Marshall et al., 2004). Approximately
1 million acres are contaminated with mercury (McPherson and Halley, 1996).
Phosphorus from agricultural runoff has impacted water quality in large portions
of the Everglades and has been particularly problematic in Lake Okeechobee
(Flaig and Reddy, 1995) (see Chapter 4 for a more detailed discussion of phos-
phorus enrichment in the Everglades). The Caloosahatchee and St. Lucie estuaries,
including parts of the Indian River Lagoon, have been greatly altered by high and
extremely variable freshwater discharges that bring nutrients and contaminants
and disrupt salinity regimes (Doering, 1996; Doering and Chamberlain, 1999).
At least as early as the 1920s, private citizens were calling attention to
the degradation of the Florida Everglades (Blake, 1980). However, by the time
Marjory Stoneman Douglas’s classic book The Everglades: River of Grass was
published in 1947 (the same year that Everglades National Park was dedicated),
the South Florida ecosystem had already been altered extensively. Beginning in
the 1970s, prompted by concerns about deteriorating conditions in Everglades
National Park and other parts of the South Florida ecosystem, the public, as well
as the federal and state governments, directed increased attention to the adverse
ecological effects of the flood control and irrigation projects (Kiker et al., 2001;
Perry, 2004). By the late 1980s it was clear that various minor corrective mea-
sures undertaken to remedy the situation were insufficient. As a result, a power-
ful political consensus developed among federal agencies, state agencies and
commissions, Native American tribes, county governments, and conservation
organizations that a large restoration effort was needed in the Everglades (Kiker
et al., 2001). This recognition culminated in the Comprehensive Everglades
Restoration Plan (CERP), which builds on other ongoing restoration activities
of the state and federal governments to create one of the most ambitious and
extensive restoration efforts in the nation’s history.
RESTORATION GOALS FOR THE EVERGLADES
Several goals have been articulated for the restoration of the South F lorida eco-
system, reflecting the various restoration programs. The South Florida Ecosystem

Restoration Task Force (hereafter, simply the Task Force), an intergovernmental

body established to facilitate coordination in the restoration effort, has three broad
strategic goals: (1) “get the water right,” (2) “restore, preserve, and protect natural
habitats and species,” and (3) “foster compatibility of the built and natural systems”
(SFERTF, 2000). These goals encompass, but are not limited to, the CERP. The
Task Force works to coordinate and build consensus among the many non-CERP
restoration initiatives that support these broad goals.
The goal of the CERP, as stated in the Water Resources Development Act of
2000 (WRDA 2000), is “restoration, preservation, and protection of the South
Florida Ecosystem while providing for other water-related needs of the region,
including water supply and flood protection.” The Programmatic Regulations
(33 CFR § 385.3) that guide implementation of the CERP further clarify this goal
by defining restoration as “the recovery and protection of the South Florida eco-
system so that it once again achieves and sustains the essential hydrological and
biological characteristics that defined the undisturbed South Florida ecosystem.”
These defining characteristics include a large areal extent of interconnected
wetlands, extremely low concentrations of nutrients in freshwater wetlands,
sheet flow, healthy and productive estuaries, resilient plant communities, and an
abundance of native wetland animals (DOI and USACE, 2005). Although devel-
opment has permanently reduced the areal extent of the Everglades ecosystem,
the CERP hopes to recover many of the Everglades’ original characteristics and
natural ecosystem processes. At the same time, the CERP is charged to maintain
levels of flood protection (as of 2000) and provide for other water-related needs,
including water supply, for a rapidly growing human population in South Florida
(DOI and USACE, 2005).
Although the CERP contributes to each of the Task Force’s three goals,
it focuses primarily on restoring the hydrologic features of the undeveloped
wetlands remaining in the South Florida ecosystem, on the assumption that
improvements in ecological conditions will follow. Originally, “getting the
water right” had four components—quality, quantity, timing, and distribution.
However, the hydrologic properties of flow, encompassing the concepts of direc-
tion, v elocity, and discharge, have been recognized as an important component
of getting the water right that had previously been overlooked (NRC, 2003c;
SCT, 2003). Numerous studies have supported the general approach to getting
the water right (Davis and Ogden, 1994; NRC, 2005; SSG, 1993), although it
is widely recognized that recovery of the native habitats and species in South
Florida may require restoration efforts in addition to getting the water right, such
as controlling exotic species and reversing the decline in the spatial extent and
compartmentalization of the natural landscape (SFERTF, 2000; SSG, 1993).
The goal of ecosystem restoration can seldom be the exact re-creation of
some historical or preexisting state because physical conditions, driving forces,

and boundary conditions usually have changed and are not fully recoverable.
Rather, restoration is better viewed as the process of assisting the recovery of a
degraded ecosystem to the point where it contains sufficient biotic and a biotic
resources to continue its functions without further assistance in the form of
energy or other resources from humans (NRC, 1996; Society for Ecological
Restoration International Science & Policy Working Group, 2004). The term eco-
system rehabilitation may be more appropriate when the objective is to improve
conditions in a part of the South Florida ecosystem to at least some minimally
acceptable level to allow the restoration of the larger ecosystem to advance.
However, flood control remains a critical aspect of the CERP design, and artificial
storage will be required to replace the lost natural storage in the system (NRC,
2005). For these and other reasons, even when the CERP is complete, it will
require large inputs of energy and human effort to operate and maintain pumps,
stormwater treatment areas, canals and levees, and reservoirs, and to continue
to manage exotic species. Thus, for the foreseeable future, the CERP does not
envision ecosystem restoration or rehabilitation that returns the ecosystem to a
state where it can “manage itself.”
Implicit in the understanding of ecosystem restoration is the recognition
that natural systems are self-designing and dynamic, and therefore, it is not
possible to know in advance exactly what can or will be achieved. Thus, eco-
system r estoration is an enterprise with some scientific uncertainty in methods
or outcomes that requires continual testing of assumptions and monitoring
and assessment of progress. This report discusses the challenges posed by
two major contributors to the dynamic circumstances in which restoration
is taking place, climate change (Chapter 5) and nonnative invasive species
(Chapter 6). Additional challenges in defining and implementing restoration
goals are discussed in the initial National Research Council (NRC) biennial
review (NRC, 2007).
What Natural System Restoration Requires
Restoring the South Florida ecosystem to a desired ecological landscape

requires reestablishment of the critical processes that sustained its historical func-
tions. Although getting the water right is the oft-stated and immediate goal, the
restoration will be considered successful if it restores the distinctive characteristics
of the historical ecosystem to the remnant Everglades (DOI and USACE, 2005).
Getting the water right is a means to an end, not the end in itself. The hydrologic
and ecologic characteristics of the historical Everglades serve as restoration goals
for a functional (albeit reduced in size) Everglades ecosystem. The first Commit-
tee on Independent Scientific Review of Everglades Restoration Progress review
identified five critical components of Everglades restoration (NRC, 2007):

1. Enough water storage capacity combined with operations that allow for
appropriate volumes of water to support healthy estuaries and the return of sheet
flow through the Everglades ecosystem while meeting other demands for water;
2. Mechanisms for delivering and distributing the water to the natural sys-
tem in a way that resembles historical flow patterns, affecting volume, depth,
velocity, direction, distribution, and timing of flows;
3. Barriers to eastward seepage of water so that higher water levels can be
maintained in parts of the Everglades ecosystem without compromising the cur-
rent levels of flood protection of developed areas as required by the CERP;
4. Methods for securing water quality conditions compatible with restora-
tion goals for a natural system that was inherently extremely nutrient poor,
particularly with respect to phosphorus; and
5. Retention, improvement, and expansion of the full range of habitats by
preventing further losses of critical wetland and estuarine habitats and by pro-
tecting lands that could usefully be part of the restored ecosystem.
If these five critical components of restoration are achieved and the dif-
ficult problem of invasive species can be managed (see Chapter 6), then the
basic physical, chemical, and biological processes that created the historical
Everglades can once again work to create a functional mosaic of biotic com-
munities that resemble what was distinctive about the historical Everglades.
The history of the Everglades likely will make replication of the historical
system impossible. Because of the historical changes that have occurred through
engineered structures, urban development, introduced species, and other factors,
the paths taken by the ecosystem and its components in response to restoration
efforts will not retrace the paths taken to reach current conditions. This means
that the paths toward restoration will pass through different intermediate condi-
tions from the ones they passed through on their way to the current status. This
phenomenon often is referred to as hysteresis (e.g., NRC, 2012c; Scheffer et al.,
2001; Tett et al., 2007) and is a complicating factor in any estimates of how long
restoration efforts are likely to take to achieve their goals (Chapter 4).
Even if the restored system does not exactly replicate the historical system, or
reach all of the biological, chemical, and physical targets, the reestablishment of
natural processes and dynamics should result in a viable and valuable Everglades
ecosystem. The central principle of ecosystem management is to provide for the
natural processes that historically shaped an ecosystem, because ecosystems are
characterized by the processes that regulate them. If the conditions necessary for
those processes to operate are met, then recovery of species and communities is
far more likely than if humans attempt to specify and manage every individual
constituent and element of the ecological system (NRC, 2007).

RESTORATION ACTIVITIES
Several restoration programs, including the largest of the initiatives, the

CERP, are now under way. The CERP often builds upon non-CERP activities (also
called “foundation projects”), many of which are essential to the effectiveness
of the CERP. The following section provides a brief overview of the CERP and
some of the major non-CERP activities.
Comprehensive Everglades Restoration Plan
WRDA 2000 authorized the CERP as the framework for modifying the
C&SF Project. Considered a blueprint for the restoration of the South Florida
ecosystem, the CERP is led by two organizations with considerable expertise
managing the water resources of South Florida—the USACE, which built most
of the canals and levees throughout the region, and the South Florida Water
Management District (SFWMD), the state agency with primary responsibility
for operating and maintaining this complicated water collection and distribu-
tion system.
The CERP conceptual plan (USACE and SFWMD, 1999; also called the
Yellow Book) proposes major alterations to the C&SF Project in an effort to
reverse decades of ecosystem decline. The Yellow Book includes approxi-
mately 50 major projects consisting of 68 project components to be con-
structed at a cost of approximately $13.5 billion (estimated in 2009 dollars;
DOI and USACE, 2011; Figure 2-2). Major components of the restoration plan
focus on restoring the quantity, quality, timing, and distribution of water for
the natural system. The Yellow Book outlines the major CERP components,
including the following:
• Conventional surface-water storage reservoirs. The Yellow Book includes

plans for approximately 1.5 million acre-feet of storage, located north of Lake
Okeechobee, in the St. Lucie and Caloosahatchee basins, in the EAA, and in
Palm Beach, Broward, and Miami-Dade counties.
• Aquifer storage and recovery (ASR). The Yellow Book proposes to pro-
vide substantial water storage through ASR, a highly engineered approach that
would use a large number of wells built around Lake Okeechobee, in Palm
Beach County, and in the Caloosahatchee Basin to store water approximately
1,000 feet belowground; the feasibility of this approach is currently being exam-
ined through pilot tests.
• In-ground reservoirs. The Yellow Book proposes additional water storage
in quarries created by rock mining.

FIGURE 2-2 Major project components of the CERP.
SOURCE: Courtesy of Laura Mahoney, USACE.

Figure 2-2
R02233 (Everglades 4)
raster iamge

• Stormwater treatment areas (STAs). The CERP contains plans for addi-
tional constructed wetlands that will treat agricultural and urban runoff water
before it enters natural wetlands.1
• Seepage management. The Yellow Book outlines seepage management
projects to prevent unwanted loss of water from the natural system through
levees and groundwater flow. The approaches include adding impermeable
barriers to the levees, installing pumps near levees to redirect lost water back
into the Everglades, and holding water levels higher in undeveloped areas
between the Everglades and the developed lands to the east.
• Removing barriers to sheet flow. The CERP includes plans for removing
240 miles of levees and canals, to reestablish shallow sheet flow of water through
the Everglades ecosystem.
• Rainfall-driven water management. The Yellow Book includes opera-
tional changes in the water delivery schedules to the WCAs and Everglades
National Park to mimic more natural patterns of water delivery and flow through
the system.
• Water reuse and conservation. To address shortfalls in water supply,
the Yellow Book proposes two advanced wastewater treatment plants so that the
reclaimed water could be discharged to wetlands along Biscayne Bay or used
to recharge the Biscayne aquifer.
The largest portion of the budget is devoted to storage and water conservation
projects and to acquiring the lands needed for them (see NRC, 2005).
The modifications to the C&SF Project embodied in the CERP were originally
expected to take more than three decades to complete (and will likely now take
much longer), and to be effective, they require a clear strategy for managing
and coordinating restoration efforts. The Everglades Programmatic Regulations
(33 CFR Part 385) state that decisions on CERP implementation are made by the
USACE and the SFWMD (or any other local project sponsors), in consultation
with the Department of the Interior, the Environmental Protection Agency (EPA),
1
Although some STAs are included among CERP projects, USACE has clarified its policy on fed-
eral cost-sharing for water quality features. A memo from the Assistant Secretary of the Army (Civil
Works) (USACE, 2007) states: “Before there can be a Federal interest to cost share a WQ [water
quality] improvement feature, the State must be in compliance with WQ standards for the current
use of the water to be affected and the work proposed must be deemed essential to the Everglades
restoration effort. . . . This determination must be based on some finding other than the project
is a part of CERP and generally will aid the restoration effort.” The memo goes on to state, “the
Yellow Book specifically envisioned that the State would be responsible for meeting water quality
standards.” Therefore, it appears that until the water flowing into the project features meets existing
water quality requirements or unless a special exemption is granted for projects deemed “essential
to Everglades restoration,” the state is responsible for 100 percent of the costs of CERP water quality
project features.

the Department of Commerce, the Miccosukee Tribe of Indians of Florida, the

Seminole Tribe of Florida, the Florida Department of Environmental Protection,
and other federal, state, and local agencies (33 CFR Part 385).
WRDA 2000 endorses the use of an adaptive management framework for
the restoration process, and the Programmatic Regulations formally establish an
adaptive management program that will “assess responses of the South Florida
ecosystem to implementation of the Plan; . . . [and] seek continuous improve-
ment of the Plan based upon new information resulting from changed or unfore-
seen circumstances, new scientific and technical information, new or updated
modeling; information developed through the assessment principles contained in
the Plan; and future authorized changes to the Plan.” An interagency body called
Restoration, Coordination, and Verification (RECOVER) has been established to
ensure that sound science is used in the restoration. The RECOVER leadership
group oversees the monitoring and assessment program that will evaluate the
progress of the CERP toward restoring the natural system and will assess the need
for changes to the plan through the adaptive management process.
Major Program-Level CERP-Related Developments Since 2000
Several major program-level developments have occurred since the CERP

was launched that have affected the pace and focus of CERP efforts. In 2004,
Florida launched Acceler8, a plan to hasten the pace of project implementation
that was bogged down by the slow federal planning process (for further discussion
of Acceler8, see NRC, 2007). Acceler8 originally included 11 CERP project com-
ponents and 1 non-CERP project, and although the state was unable to complete
all of the original tasks, the program led to increased state investment and expe-
dited project construction time lines for several CERP projects (see Chapter 4).
In 2008, Governor Charlie Crist announced the planned acquisition of
187,000 acres of agricultural land from the U.S. Sugar Corporation to maxi-
mize restoration opportunities for the South Florida ecosystem. The SFWMD
subsequently launched the River of Grass public planning process to facilitate
agency and stakeholder input on future uses of the new lands for restoration. In
October 2010, the SFWMD closed on the purchase of 26,800 acres of land for
approximately $197 million in cash and retained the option to acquire more than
153,000 additional acres over the next 10 years. Plans for use of the acquired
lands have not been finalized at this time.
In 2011, the USACE initiated a pilot program to improve the pace of its project
planning. As one of five pilot projects nationwide, the Central Everglades Planning
Process was launched in November 2011, with the objective of developing a
plan for restoration of the central Everglades that could be delivered for congres-
sional authorization within 2 years. This effort has focused attention on central

Everglades planning at all levels of the CERP partnering agencies and involves
extensive stakeholder engagement facilitated by the Task Force (see Chapter 3).
In 2010, EPA issued its court-ordered Amended Determination, which
directed the State of Florida to correct deficiencies in meeting the narrative
and numeric nutrient criteria in the Everglades Protection Area. In 2012, the
State of Florida launched its Restoration Strategies Regional Water Quality Plan,
which was approved by EPA and the Court as an alternative means to address
the Amended Determination. The State of Florida is currently in the process of
constructing approximately 6,500 acres of new STAs and 116,000 acres of flow
equalization basins (see Chapter 4). These water quality treatment improvements
are designed so that water leaving the STAs will meet a new water quality-based
effluent limit (WQBEL) to comply with the 10-ppb total phosphorus water quality
criterion for the Everglades Protection Area.
Non-CERP Restoration Activities
When Congress authorized the CERP in WRDA 2000, the SFWMD, the
USACE, the National Park Service, and the U.S. Fish and Wildlife Service
were already implementing several activities intended to restore key aspects
of the Everglades ecosystem. These non-CERP initiatives are critical to the
overall restoration progress. In fact, the CERP’s effectiveness was predicated
upon the completion of many of these projects, which include Modified Water
Deliveries to Everglades National Park (Mod Waters), C-111 (South Dade), and
the E verglades Construction Project (see Box 2-1). Several additional projects
are also under way to meet the broad restoration goals for the South Florida
ecosystem and associated legislative mandates. They include extensive water
quality initiatives, such as the Everglades Construction Project, and programs to
establish best management practices (BMPs) to reduce nutrient loading.
SUMMARY
The Everglades ecosystem is one of the world’s ecological treasures, but for
more than a century the installation of an extensive water control infrastructure has
changed the geography of South Florida and facilitated extensive agricultural and
urban development. These changes have had profound ancillary effects on regional
hydrology, vegetation, and wildlife populations. The CERP, a joint effort led by the
state and federal governments and launched in 2000, seeks to reverse the general
decline of the ecosystem. Since 2000, the CERP and other major Everglades restora-
tion efforts have adapted to changing budgets, refinements in scientific understand-
ing, and an evolving legal context, particularly as it relates to water quality. The
implications on implementation progress are discussed in more detail in Chapter 4.

BOX 2-1
Non-CERP Restoration Activities in South Florida
The following represent the major non-CERP initiatives currently under way in sup-
port of the South Florida ecosystem restoration (Figure 2-3).
Kissimmee River Restoration Project

This project, authorized by Congress in 1992, aims to reestablish the historical river-
floodplain system at the headwaters of the Everglades watershed and thereby restore
biological diversity and functionality. The project plans to backfill 22 miles of the 56-mile
C-38 Canal and carve new sections of the river channel to connect channel remnants,
thereby restoring over 40 miles of meandering river channel in the Kissimmee River.
The project includes a comprehensive evaluation program to track ecological responses
to restoration (Jones et al., 2014). See also Chapter 4.
State Water Quality Treatment Projects

The Everglades Forever Act (Fla. Stat. § 373.4592) required the State of Florida to
construct stormwater treatment areas (STAs) to reduce the loading of phosphorus into
the Arthur R. Marshall Loxahatchee National Wildlife Refuge (LNWR), the WCAs, and
Everglades National Park. As part of the state’s Everglades Construction Project and
long-term plan for meeting the total phosphorus criterion for the Everglades Protec-
tion Area of 10 parts per billion (ppb), the SFWMD constructed 57,000 acres of STAs
between 1993 and 2012. In 2012, after continued violations of water quality standards,
the state and the Environmental Protection Agency agreed upon a new Restoration
Strategies Regional Water Quality Plan that has been approved by the U.S. District
Court for the Southern District of Florida that requires an additional 6,500 acres of STAs
and 116,000 acres of flow equalization basins (see Chapter 4).
Modifications to the C&SF: C-111 (South Dade) Project

This project is designed to improve hydrologic conditions in Taylor Slough and the
Rocky Glades of the eastern panhandle of Everglades National Park and to increase
freshwater flows to northeast Florida Bay, while maintaining flood protection for urban
and agricultural development in south Miami-Dade County. The project plan includes a
tieback levee with pumps to capture groundwater seepage to the east, detention areas
to increase groundwater levels and thereby enhance flow into Everglades National Park,
and backfilling or plugging several canals in the area. A combined operational plan
(COP) will integrate the goals of the Mod Waters and C-111 projects and protect the
quality of water entering Everglades National Park (DOI and USACE, 2005).
Modified Water Deliveries to Everglades National Park Project (Mod Waters)

This federally funded project, authorized in 1989, is designed to restore more natural
hydrologic conditions in Everglades National Park. The project includes levee modifica-
tions and installation of a seepage control pump to increase water flow into WCA-3B
and northeastern portions of Everglades National Park. It also includes providing flood
mitigation to the 8.5-square-mile area (a low-lying but partially developed area on the
northeast corner of Everglades National Park) and raising portions of the Tamiami Trail.
continued

BOX 2-3 Continued
Mod Waters is a prerequisite for the first phase of decompartmentalization (i.e., remov-
ing some barriers to sheet flow), which is part of the CERP (DOI and USACE, 2005;
NRC, 2008). See also Chapter 4.
Northern Everglades and Estuaries Protection Program

In 2007, the Florida legislature expanded the Lake Okeechobee Protection Act
(LOPA) to include protection and restoration of the Lake Okeechobee watershed and
the Caloosahatchee and St. Lucie estuaries. The legislation, being implemented as the
Northern Everglades and Estuaries Protection Program, will focus resources on restora
tion efforts for Lake Okeechobee and the Caloosahatchee and St. Lucie estuaries. The
Lake Okeechobee Watershed Construction Project Phase II Technical Plan, issued
continued
FIGURE 2-3 Locations of major non-CERP initiatives.
Figure 2-3-1
R02233 (Everglades 4)
raster iamge

BOX 2-3 Continued
in February 2008 in accordance with LOPA, consolidated the numerous initiatives

already under way through Florida’s Lake Okeechobee Protection Plan (LOPP) and
Lake Okeechobee and Estuary Recovery (LOER) Plan.
Critical Projects
Congress gave programmatic authority for the Everglades and South Florida Eco-
system Restoration Critical Projects in Water Resources Development Act of 1996
(WRDA 1996), with modification in WRDA 1999 and WRDA 2007. These were small
projects that could be quickly implemented to provide immediate and substantial res-
toration benefits such as improved quality of water discharged into WCA-3A and Lake
Okeechobee and more natural water flows to estuaries. Examples of the Critical Projects
include the Florida Keys Carrying Capacity Study, Lake Okeechobee Water Retention
and Phosphorus Removal, Seminole Big Cypress Reservation Water Conservation
Plan, Tamiami Trail Culverts, Ten Mile Creek Water Preserve Area, and the Lake Trafford
Restoration (DOI and USACE, 2011).

Central Everglades Planning Project
Past NRC reports have emphasized the perils of slow restoration progress
(NRC, 2008) and documented the declining trajectories of many ecosystem
attributes that depend upon restoration of water flows to the central Everglades
(NRC, 2012a). NRC (2012a) noted that significant progress has been and con-
tinues to be made to reduce phosphorus in the central Everglades through the
state’s 20-year effort to construct and optimize stormwater treatment areas (STAs)
and to encourage best management practices for improved source control. The
$880 million Restoration Strategies project, launched in 2012, continues these
efforts to attain compliance with water quality criteria in the remnant Everglades
(see Chapter 4). However, little progress has been made in restoring flows in
the central Everglades because implementation of the Comprehensive Ever-
glades Restoration Plan (CERP) to date has been focused primarily on projects
at the edges of the historic Everglades. As a result, declines continue in many of
the characteristic features of the Everglades, such as ridge and slough and tree
islands, and oxidation of peat alters the slope of the land surface that shapes
water flow (as described in detail in NRC, 2012a). The reasons that CERP projects
at the periphery of the remnant Everglades have progressed faster than projects in
the central Everglades are complex, but include fewer stakeholder conflicts at the
periphery of the Everglades and/or strong local stakeholder support for specific
efforts. Additionally, the project planning process has been easily stalled by sci-
entific or technical uncertainties, particularly in complex or contentious projects
such as the Water Conservation Area 3 (WCA-3) Decompartmentalization and
Sheetflow project (Decomp; NRC, 2007). Unresolved water quality issues were
also significant barriers in the development of project plans to increase flow to
the central Everglades. To address these concerns and expedite restoration of the
central Everglades, key federal and state officials launched the Central Everglades
Planning Project in October 2011.
The primary purpose of the Central Everglades Planning Project is restora-
tion of more natural patterns of water flows in the central part of the Everglades
35
(WCA-3 and Everglades National Park). The U.S. Army Corps of Engineers
(USACE) described the project purpose:
to redirect water that is currently discharged to the east and west coast e stuaries
from Lake Okeechobee and restore water flow to the south, allowing for restora
tion of natural habitat conditions and water flow in the central Everglades and
reconnecting the ecosystem from Lake Okeechobee to Everglades National Park
and Florida Bay (76 Federal Register, [December 2, 2011], 75539).
The scope of the project includes increments (or components) of a number

of CERP projects described in the original restoration plan (USACE and SFWMD,
1999), such as the Everglades Agricultural Area Storage Reservoir, Decomp,
seepage management, and rain-driven operations. The Central Everglades Plan-
ning Project shifts the planning emphasis from multiple independent project
implementation reports (PIRs) to a regional integrated PIR for the first increment
of restoration. Thus, the Central Everglades Planning Project is a critical com-
ponent of the CERP. This chapter reviews the progress made on this effort and
prospects for restoration of the central Everglades.
THE EXPEDITED PLANNING PROCESS
The Central Everglades Planning Project was one of five USACE projects
chosen to pilot a new expedited planning process. The pilot process aimed to
reduce the typical 6-year preauthorization planning time frame to 18-24 months,
while still addressing all current legal and programmatic requirements (such as
the National Environmental Policy Act [NEPA] and independent external peer
review). The process required a shift toward more focused detailed analyses
and risk-based project planning prior to authorization—detailed engineering
plans would be deferred until after congressional authorization of a project.
The process also relied upon frequent engagement of senior leadership (known
as vertical team coordination) at key project phases or decision points. The
pilot projects are just one component of the USACE planning modernization
effort, which includes the “3×3×3 rule” that requires all feasibility studies to
be completed within a target of 18 months, but no more than 3 years, at a
cost of no more than $3 million, utilizing three levels of vertical team coor-
dination, and a “reasonable” report size (Walsh, 2012). The 3×3×3 rule was
applied to all feasibility studies that had not had a feasibility scoping meeting
by December 2011. The Central Everglades Planning Project represented an
extreme test of the expedited process, considering the complex nature of the
project, the extensive stakeholder involvement, and the diverse objectives of
various interest groups.

Central Everglades Planning Project 37
The Central Everglades Planning Project was launched in October 2011 with
initial targets of January 2013 to produce a draft PIR and December 2013 for final
approval by USACE headquarters (known as the Chief’s Report) and submittal to
Congress for authorization (K. Taplin, USACE, personal communication, 2012).
Staff from the USACE, South Florida Water Management District (SFWMD),
Department of the Interior, and other federal, state, local, and tribal govern-
ments worked within the project delivery team (PDT) process on all aspects of
technical planning. Integration of other stakeholder input occurred through the
South Florida Ecosystem Restoration Task Force’s (Task Force’s) Working Group
(see Stakeholder Engagement later in this chapter). The record of opportunities to
engage other federal and state agencies and the general public was impressive.
During the planning process there were
• 26 meetings of the PDT;

• 5 meetings of the full Task Force at which concepts and updates of the
Central Everglades Planning Process were presented;1
• 15 public workshops; and
• 12 public meetings as part of the NEPA process,
in addition to numerous briefings for the SFWMD Governing Board, the Water
Resources Advisory Committee, and the Task Force’s Working Group and
Science Coordination Group. Additional “in-progress review” meetings were
held at major decision points with the USACE Jacksonville District teams and
leadership from USACE regional and headquarters offices.
The process was only slightly behind its original aggressive schedule through
the first 15 months of effort, when the late introduction of new water supply
issues added approximately 4 months to the process. The draft PIR was publicly
released in August 2013. A longer-than-anticipated review process and unre-
solved policy issues further delayed the schedule. On May 23, 2014, the USACE
Civil Works Review Board approved release of the final Central Everglades
Planning Project PIR (contingent upon some revisions and concurrence by the
SFWMD) for state, agency, and administration review, and a Chief of Engineers’
Report is anticipated in summer 2014 (delayed approximately 8 months from the
original target). The report was not finished in time to be included in the Water
Resources Reform and Development Act (WRRDA) of 2014, which was signed
into law on June 10, 2014, authorizing 34 projects for which USACE Chief’s
Reports were complete. This section summarizes the development of the Central
Everglades Planning Project plan and key unresolved issues.
1
See www.sfrestore.org/tf_minutes.html.

Development of Alternatives
Alternative plans were formulated based on four spatially discrete subregions

of the planning area. The subregions, divided by color-referenced boundaries
shown in Figure 3-1 and listed here from north to south, are also characterized
by their functions:
A. North of the Red Line represents the location of storage and treatment
to reduce phosphorus loads and includes the Everglades Agricultural Area, Lake
Okeechobee, and surrounding areas;
B. South of the Red Line is the area characterized by management measures
for distribution and conveyance of water in northern WCA-3A;
C. The Green and Blue Lines represent the areas in southern WCA-3 and
into Everglades National Park that are characterized by management measures
for distribution and conveyance of water; and
D. The Yellow Line represents the location along the Lower East Coast Pro-
tective Area, which is the location of seepage management options to protect
the lower east coast urban areas from flooding.
The PDT and stakeholders via the public workshops formulated management
alternatives for each of the areas, starting from the north and working toward
the south. The PDT used a variety of analysis tools, including spreadsheet-style
screening models, an inverse model, and preliminary cost-effectiveness analyses,
to identify the most promising management options and optimize the combina-
tions of management measures within each region. In some areas, such as North
of the Red Line (storage and treatment), a single option emerged as best. South of
the Red Line (distribution and conveyance), two basic options emerged, while
four suites of options were considered worthy of further analysis in the Blue/
Green Line (distribution and conveyance in WCA-3) and Yellow Line (seep-
age management) areas (see Table 3-1). Some of the variation in proposed
seepage management measures (Yellow Line) was determined by the upstream
flow conditions created by the management options in the Blue/Green Line
scenarios. The management options were formulated into four alternatives (Alt-1,
Alt-2, Alt-3, and Alt-4) for further modeling analysis and evaluation. Detailed
descriptions of the rationale behind the formulation of these four alternatives
are documented in the draft PIR (USACE and SFWMD, 2013b).
Evaluation of Alternatives and Subsequent Refinements
Alternatives were evaluated using a variety of techniques, including hydro-

logic and ecological modeling to assess systemwide performance and detailed

FIGURE 3-1 Subregions of the Central Everglades Planning Project Study Area.
SOURCE: Adapted from USACE and SFWMD (2013b).

40
TABLE 3-1 Summary of the Major Features of the Four Alternatives

Zone Features Alt-1 Alt-2 Alt-3 Alt-4
North of Storage/treatment 28,000-acre FEB (A-2)

Red Line Lake Okeechobee Operation Refinements
South of Conveyance Diversion of L-6 flows and L-5 canal improvements, removal of ~3 miles of the L-4 levee, G404 and S-8
Red Line pump station modifications
Spreader Spreader canal ~3 mi Spreader canal ~3 mi west of S-8 (3,000 cfs), ~3 mi east of S-8 (800 cfs), and
west of S-8 (3,000 cfs) ~1.5 mi east of G-206 (400 cfs)
Backfill Backfill Miami Canal Backfill Miami Canal from S-8 to I-75
from 1.5 mi south of S-8
to I-75
Blue/Green L-67A Structures One 750-cfs gated One 750-cfs and two Four 500-cfs gated Three 500-cfs gated
Line structure 500-cfs gated structures structures structures
Blue Shanty Levee with No No No Yes

degrade of L-29 and
L-67C in flowway
6000-ft gaps in L-67C 1 3 4 1
Pumps and gravity NA One 500-cfs gravity Two 500-cfs pumps out NA
structures structure out of WCA-3B of WCA-3B
Yellow Line S-356 pump increase to 1,000 cfs
Seepage barriers S-335 NA Full depth Full depth NA
to S-334

Seepage barriers south NA 2-mile partial-depth 5-mile partial-depth seepage barrier
of Tamiami Trail, along seepage barrier
L-31N
L-31N pumps Two 250-cfs pumps One 250-cfs pump NA NA
ecological and physical responses in specific subareas. Systemwide evaluation

was based on multiple criteria as prescribed in the USACE’s Planning Guidance
(USACE, 2000), including
• Effectiveness—the extent to which an alternative plan alleviates the speci-

fied problems and achieves specified opportunities;
• Acceptability—the workability and viability of alternatives with respect
to acceptance by state and local entities and the public and compatibility with
existing laws, regulations, and public policies;
• Completeness—the extent to which a given alternative provides and
accounts for all necessary investments or other actions to ensure the realization
of the planned effects; and
• Efficiency—the extent to which the alternative maximizes environmental
benefits compared to costs, both overall and by individual project increments.
The effectiveness of each alternative was evaluated with a diverse set

of performance measures—hydrologic surrogates for physical, chemical, and
biological indicators of restoration progress. The study region was divided into
17 spatially distinct zones—two for the northern estuaries, six in the WCAs,
three in Everglades National Park, and six in Florida Bay—each with specified
performance measures that would be assessed relative to specific restoration
objectives.2 Some of the performance measures, such as statistics of high and
low flows in the northern estuaries and salinity in Florida Bay, were unique
to particular zones, while other performance measures (e.g., ridge and slough
inundation duration and sheet flow, drought intensity [a hydrologic surrogate
for soil oxidation]) were common to several zones. Hydrologic models were
used to evaluate the effectiveness of alternatives according to how well the
targets for each performance measure were met, on a scale of 0 to 100. Zero
represented a fully degraded ecosystem, and 100 represented full achievement
of the restoration target.
To assess the efficiency of the project alternatives, “habitat units” associ-
ated with each alternative were calculated. First, a habitat suitability index (HSI;
scaled from 0 to 1) was calculated for each of the 17 zones, based on an average
of the performance measure values. The benefits model included an option to
differentially weight the performance measures when calculating the HSI, but
that option was not used, and all performance measures were assigned equal
2
There were two performance measures for each of the northern estuaries, five for each zone in
the WCAs and Everglades National Park, and one for each Florida Bay zone. In several cases, per-
formance measures included submetrics that were then averaged to produce a single performance
measure score.

weights. The overall habitat units associated with each alternative were calcu-
lated by multiplying the area of each zone by the HSI and adding the habitat
units across the 17 zones. Habitat units generated by an alternative were then
compared against a “future without the project” (FWO) scenario to calculate
the “habitat unit lift” provided by that alternative (see Table 3-2). Habitat unit
contributions were also annualized by incorporating estimates of the time it
would take to realize the benefits.
After analysis and evaluation of the benefits of the four alternatives, several
refinements were made to the alternatives. Analysis of the additional spreader
and backfill features south of the Red Line in Alternatives 2, 3, and 4 (see
Table 3-1) showed no additional environmental benefit, despite $130,000 addi-
tional cost (observed differences in benefits in southern WCA-3 were assumed
to be dependent on other features). Thus, Alternatives 2, 3, and 4 were modified
by using the conveyance features south of the Red Line of Alt-1, and an “M”
was added to their labels.
The cost-effectiveness of alternatives, measured by the average cost per
habitat unit lift, is shown in Figure 3-2. Alt-1 generated slightly more habitat units
than Alt-2 and at a much lower cost. Alt-4 produced higher numbers of habitat
units than any of the alternatives at a unit cost only $3 higher than Alt-1. When
Alt-2, -3, and -4 were modified to include only the Alt-1 conveyance features
south of the Red Line, Alt-1 and Alt-4M were judged to be cost-effective. Alt-
4M became the most cost-effective option because it contributed substantially
greater habitat units at the lowest cost per habitat unit.
In addition to the efficiency criterion, Alt-1 and Alt-4 also scored as high
as or higher than Alt-2 and Alt-3 on USACE planning criteria of effective-
ness, acceptability, and completeness. Acceptability was judged on the basis
of documented concerns expressed by stakeholders (summarized in USACE
and SFWMD, 2013b). Completeness was evaluated primarily by noting project
TABLE 3-2 Systemwide Habitat Units for Future Without Project Condition and
Four Alternatives
FWO Alt-1 Alt-2 Alt-3 Alt-4
Habitat units 683,582 982,513 977,306 1,001,360 1,023,642

Habitat unit lift NA 298,931 293,724 317,778 340,060
Estimated cost (billion $) a
NA 1.93 2.24 2.35 2.22
a
Not including interest during construction.
SOURCE: USACE and SFWMD (2013b).

660
Average Cost, dollars per habitat unit
Alt-2
640
Alt-3
620
600 2
Alt-2M
580 Alt-3M
560
540 Alt-4
Alt-1
520
Alt-4M
500
200 210 220 230 240 250 260 270
Annual Average Incremental Benefits, 1,000s of Habitat Units
FIGURE 3-2 Cost-effectiveness of alternatives. Each alternative is located on the longitudinal axis by the
increment of average annual habitat units and on the vertical axis by the average annual cost per average
annual habitat unit.
SOURCE: Data from USACE and SFWMD (2013b).
dependencies of the Central Everglades Planning Project on other CERP and

non-CERP projects (discussed in more detail later in this chapter).
Subsequent operational refinements were made to Alt-4M to increase con-
fidence in satisfaction of the Savings Clause.3 The result, Alt-4R, was presented
to the SFWMD Governing Board in February 2013 as the initial version of the
Tentatively Selected Plan. In a presentation during the Governing Board meet-
ing, it was noted that next steps included a “sequencing strategy, saving clause/
project assurances analysis, draft preliminary operating manual, adaptive man-
The Savings Clause is a provision of WRDA 2000 related to CERP implementation that is
3
designed to ensure that an existing legal source of water (e.g., agricultural or urban water supply,
water supply for Everglades National Park, water supply for fish and wildlife) is not eliminated or
transferred until a replacement source of water of comparable quantity and quality—as was avail-
able on the date of enactment of WRDA 2000—is available and that levels of flood protection
existing as of 2000 are not reduced.

agement plan, and a monitoring plan” (Minutes of Governing Board Meeting,

Feb. 13, 20134). Issues regarding the Saving Clause and project assurances, cost
sharing, and water quality issues were still unresolved when the board met on
March 14, 2013, but the time required to resolve the issues was estimated to be
6-8 weeks. By the time the Governing Board met in April 2013, a new concern
was added that had not been included in the original purpose of the project—the
omission of “other project benefits to meet the additional water supply needs
especially for the Lower East Coast and agricultural users” (Minutes of Board
Meeting, April 11, 2013).
Alt-4R was subsequently refined to increase confidence in satisfaction of
the Savings Clause and to increase public water supply deliveries. Addressing
these issues to the satisfaction of stakeholders added about 4 months to the
schedule before a draft of the PIR (USACE and SFWMD, 2013b) was published
in August 2013. The further “refined” version of Alt-4R, referred to as Alt-4R2,
was selected as the Tentatively Selected Plan in the Draft PIR (Figure 3-3). None
of the refinements to Alt-4R involved changes to the major structural features
shown in Table 3-1. Instead, Alt-4R2 included updates and changes to operat-
ing policies for Lake Okeechobee and the CERP Indian River Lagoon-South and
Broward Water Preserve Areas projects, all within the limits of existing rules.
Those changes provided an additional 12 million gallons per day (MGD) for
municipal and industrial users in Lower East Coast Service Area 2 (Broward
County) and 5 MGD to Lower East Coast Service Area 3 (Miami-Dade County).
The refinements made to Alt-4M to formulate the Tentatively Selected Plan
came with a reduction in environmental benefits as measured by aggregate
habitat units (see Table 3-3). There was a net decrease of 54,371 habitat units
between Alt-4M and Alt-4R2, a decrease of 16 percent. Most of that loss was
attributable to Savings Clause modifications from Alt-4M to Alt-4R, with the larg-
est decreases in southern Everglades National Park, western Florida Bay, and east
central Florida Bay. In further modifications from Alt-4R to Alt-4R2, it is unclear
how the increase in water supply (not included in the original statement of pur-
pose for the project; 76 Federal Register [December 2, 2011], 75539) impacted
the overall benefits, because operational changes were made to provide more
water than otherwise could have been provided to the natural system.
Unresolved Issues
After release of the PIR, the agencies conducted required reviews and worked
to address several issues that were not fully resolved in the planning process,
See http://www.sfwmd.gov/portal/page/portal/xweb about us/gb application/ for all SFWMD

4
Governing Board meeting minutes.

FIGURE 3-3 The Central Everglades Planning Project Tentatively Selected Plan, Alt-4R2.
including phased implementation, revised systemwide operations, water quality,

and impacts on threatened and endangered species. Some of these issues were
resolved and others were deferred to more detailed post-authorization planning,
engineering, and design. Among the most significant issues are
• Effects of water levels on threatened and endangered species; and

• Water quality, including effects of increased flows once the Central Ever-
glades Planning Project is implemented.

TABLE 3-3 Habitat Unit Contributions of Alternatives and Changes Due to Refinements to
Alternative 4
Changes to Contributed HUs
Contributed HUs in Refinements to Alt-4
Future Alt-4 to Alt-4R to Alt-4 to
Region Without Alt-4 Alt-4R Alt-4R2 Alt-4R Alt-4R2 Alt-4R2
Caloosahatchee Estuary 34,070 4,968 4,968 4,968 0 0 0
St. Lucie Estuary 2,399 2,399 2,699 5,848 300 3,149 3,449
Total northern estuaries 36,469 7,367 7,667 10,816 300 3,149 3,449
Northeast WCA-3A 29,634 66,677 62,972 61,738 −3,705 −1,234 −4,939
WCA-3A Miami Canal 27,373 29,719 27,373 27,373 −2,346 0 −2,346
Northwest WCA-3A 30,266 23,228 23,932 23,932 704 0 704
Central WCA-3A 105,669 4,117 4,117 5,490 0 1,373 1,373
Southern WCA-3A 68,423 0 0 0 0 0 0
WCA-3B 48,842 5,998 9,426 10,283 3,428 857 4,285
Northern ENP 55,054 47,547 43,793 43,793 −3,754 0 −3,754
Southern ENP 126,454 62,034 42,946 42,946 −19,088 0 −19,088
Southeast ENP 81,062 2,702 4,054 2,702 1,352 −1,352 0
Total Greater Everglades 572,777 242,022 218,613 218,257 −23,409 −356 −23,765
Florida Bay West 20,534 31,590 18,954 20,534 −12,636 1,580 −11,056
Florida Bay Central 8,205 9,025 5,743 6,564 −3,282 821 −2,461
Florida Bay South 14,659 20,523 12,705 13,682 −7,818 977 −6,841
Florida Bay East Central 20,225 26,381 13,191 14,070 −13,190 879 −12,311
Florida Bay North Bay 2,028 887 506 633 −381 127 −254
Florida Bay East 8,685 2,265 1,133 1,133 −1,132 0 −1,132
Total Florida Bay 74,336 90,671 52,232 56,616 −38,439 4,384 −34,055
Total All Regions 683,582 340,060 278,512 285,689 −61,548 7,177 −54,371
Note: Shadings represent the areas of largest habitat unit declines between Alt-4 and Alt-4R2.
SOURCE: Data from USACE and SFWMD (2013b).
The U.S. Fish and Wildlife Service (FWS) issued a Biological Opinion for
the Central Everglades Planning Project in December 2013 (revised in April
2014) in which it came to the preliminary conclusion that “the proposed project
is not likely to jeopardize the continued existence” of the Cape Sable seaside
sparrow, snail kite, and wood stork and “is not likely to adversely modify criti-
cal habitat, where designated” (FWS, 2014). The Biological Opinion concluded
that the project would provide some benefits to all three species, but the FWS

was not able to predict the overall impacts to the three species because of the
many uncertainties related to project design, operations, and the time line for
project implementation. For the sparrow, which has a relatively narrow range of
hydrologic conditions that provide suitable habitat, the FWS noted that projected
adverse effects on one subpopulation (E) could outweigh the small benefits pro-
jected for subpopulation A. The single large subpopulation (B), which comprises
roughly 80 percent of the total population, is projected to be relatively unaffected
by the Central Everglades Planning Project (FWS, 2014). Providing suitable habi-
tat for sparrows and avoiding adverse effects on their population promise to be
ongoing challenges during implementation of the project. FWS (2014) concluded
that incidental take of all three species is likely but did not authorize incidental
take. This means that further consultation will be necessary as specific project
details are finalized and increments of the project are implemented.
Storage and treatment measures included in the Central Everglades Tentatively
Selected Plan are designed to keep phosphorus levels in compliance with water
quality-based effluent limits (WQBEL)5 for the STAs while allowing additional
flows of approximately 210,000 acre-feet per year (AF/yr). However, this plan
builds upon existing projects now under construction by the SFWMD to bring
current flows into compliance with the WQBEL, and there is some uncertainty
about the time line of meeting these water quality objectives (see also Chapter 4).
Also, there is uncertainty about how implementation of the Central Everglades
Planning Project will affect compliance with water quality criteria downstream
in Everglades National Park. The methodology within Appendix A of the 1991
Settlement Agreement that determines compliance with state phosphorus stan-
dards is a flow-weighted mean phosphorus limit that was developed based on the
observed relationship between water flow and phosphorus concentrations entering
the park. Both the quantity and the spatial pattern of flow will be modified by
implementation of the Central Everglades Planning Project. The Draft PIR (USACE
and SFWMD, 2013b) stated:
Over the long-term, distributing the flow over the northern WCA-3A marsh,
reducing short-circuiting down the canals to Everglades National Park, adding
more flow from the lake that is treated to the WQBEL, and distributing these
flows over the marsh should result in improvements by lowering the flow
weighted mean total phosphorus concentration entering the Park.
5
The WQBEL is a numeric discharge limit used to regulate permitted discharges from the STAs so
as not to exceed a long-term geometric mean of 10 μg/L within the Everglades Protection Area. This
numeric value is now translated into a flow-weighted mean (FWM) TP concentration and applied
to each STA discharge points, which now must meet the following: (1) the STAs are in compliance
with WQBEL when the TP concentration of STA discharge point does not exceed an annual FWM
of 13 μg/L in more than 3 out of 5 years, and (2) annual FWM of 19 μg/L in any water year (Leeds,
2014). See Chapter 4 for more details on STA performance.

The Everglades Technical Oversight Committee established by the 1991 Settle-

ment Agreement has been charged with reviewing the applicability of the cur-
rent version of Appendix A to the restored ecosystem and whether changes are
necessary in the context of the increased inflows to Northeast Shark River Slough.
Neither of these issues is likely to be resolved with confidence in the near
future. The ongoing Restoration Strategies project to bring concentrations in
existing outflows from STAs into compliance with WQBELs is not estimated to
be fully constructed until 2024, although the Central Flowpath projects, which
would receive inflow from the Central Everglades Planning Project FEB, are
expected to be completed by 20166 (see Chapter 4). The 2012 Consent Order
allows for up to a 5-year period for determining compliance with the WQBEL after
construction is completed. SFWMD staff have stated that the STAs are currently
permitted as a package, and therefore, the entire system must meet the WQBEL
before existing flows can be redistributed or additional flows can be initiated
(E. Barnett, SFWMD, personal communication, 2013)—potentially 2029 or later.7
Although water quality models have predicted compliance sufficiently well to
justify expenditures for corrective actions, actual post-construction water quality
will be a determining factor in how to proceed at that time. After an extensive
review period, the USACE Civil Works Review Board considered the remaining
unresolved issues and unanimously approved moving forward with the project. As
discussed previously, the Chief of Engineers has initiated the concluding adminis-
trative steps of preparing a final project report and recommendation to Congress.
ASSESSMENT OF THE CENTRAL EVERGLADES PLANNING PROJECT
This section includes the committee’s assessment of the Tentatively Selected

Plan, efficiency analyses used to select the plan, and the implementation plan.
The committee also assesses the adaptive management plan, the expedited
process, and stakeholder engagement.
The Plan
If implemented in a timely manner, the Tentatively Selected Plan for the

Central Everglades Planning Project would make substantial improvements toward
6
The 2014 South Florida Environmental Report (Leeds, 2014) reports that the Restoration Strate-
gies project features are anticipated to be completed for the Central Flowpath by July 2016, for the
Eastern Flowpath by December 2018, and for the Western Flowpath by December 2024.
7
After release of the report in prepublication form, it was pointed out that this sentence could
be read to imply that SFWMD is responsible for assessing permit compliance. FDEP, with oversight
from EPA, will interpret permit and Consent Order compliance, including any water quality issues
that may affect flow redistribution.

the goal of restoring the Everglades. It would make significant improvements to

the storage capacity and treatment performance of existing STAs by means of a
new 56,000-AF flow equalization basin. When combined with adjustments to the
regulation schedule in Lake Okeechobee operations, the flow equalization basin
will allow an additional 210,000 AF/yr of water to be treated by the STAs and
delivered to the remnant Everglades ecosystem that would otherwise be discharged
through the Caloosahatchee River and St. Lucie Canal. This represents a 21 percent
increase over existing flows (Figure 3-4) and approximately two-thirds of the new
water envisioned by the CERP to be delivered along the northern end of the WCAs.
The Central Everglades Planning Project will also shift the distribution of
water in the southern part of the system toward historical patterns, increasing
flows into Northeast Shark River Slough and reducing flows into Western Shark
River Slough. Excessively wet conditions in southwestern WCA-3A and adja-
FIGURE 3-4 Central Everglades Planning Project existing and future flows. The graphics represent the aver-
age annual overland flow from 1965 to 2005. The direction of the arrows represents the movement of water
across the landscape, and the colors represent the relative volume of water flow, with dark blue being the
highest and red being the lowest.

cent portions of Everglades National Park and excessively dry conditions in the
eastern portion of the park have contributed greatly to ecosystem degradation
in those areas (NRC, 2012a).
Systemwide, the plan improves ecosystem conditions (as measured in habi-
tat units) by more than 40 percent over the future without the project, and in
none of the 17 geographic areas would the condition be made worse overall.
Northern estuaries would improve by 30 percent, the Greater Everglades (con-
sisting of the WCAs and Everglades National Park) by 38 percent, and Florida
Bay by 76 percent. Northwestern portions of the WCA-3 that have been subject
to frequent dryouts would be rehydrated (USACE and SFWMD, 2013b). North-
east Shark River Slough on average would receive an additional 250,000 AF/yr
of overland flow compared with the future without the project (D. Crawford,
USACE, personal communication, 2014); however, an additional 500,000 to
700,000 AF/yr may be necessary to bring Florida Bay to full restoration. Eco-
logical and hydrologic connectivity of the WCAs and Everglades National Park
would be enhanced by partially degrading the L-67C and L-29 levees, and
protective water depths in the eastern parts of WCA-3B would be maintained
by the Blue Shanty Levee (USACE and SFWMD, 2013b).
Although improvements to estuaries of the Caloosahatchee River and
St. Lucie Canal were not primary objectives of the Central Everglades Planning
Project, water quality in those estuaries would be enhanced to some degree by
diversion of flows to the Everglades. Damaging high-flow events (>2,800 cfs) in
the Caloosahatchee would be reduced by 14 percent when compared with the
future without condition; high-flow events (> 2,000 cfs) in the St. Lucie would
be reduced by 34 percent. Low-flow events below desirable rates would be
reduced by 15 percent in the Caloosahatchee and 29 percent in the St. Lucie
(USACE and SFWMD, 2013b). Water available for existing users would be
maintained, and a modest increase of 17 MGD would be added for urban,
industrial, and agricultural water supply along the lower east coast.
In summary, the proposed plan offers significant benefits that should make
important strides toward reversing ongoing declines in the remnant Everglades
ecosystem and enhancing the condition of Everglades landscapes and species.
The project is one of the first CERP efforts to have successfully integrated water
flow restoration with water quality restoration. Overall, the plan is an impressive
response to the need to accelerate the pace of restoration of the central Everglades.
Efficiency Analysis
The use of habitat units for quantifying environmental restoration benefits

(discussed previously in this chapter) is included in Appendix E of the USACE
planning guidance (USACE, 2000). That document, however, does not provide

guidance as to how habitat units are to be estimated for complex multifunction,

multispecies projects such as the Everglades. The habitat unit approach evolved
out of the Habitat Evaluation Procedure (FWS, 1980) that was developed and
formally adopted for evaluation of effects on habitats of a single species. The
CERP applies the metric as an area-weighted performance measure, but there
are inherent flaws in its application to the optimization of environmental ben-
efits in such highly complex environments. First, it involves simple addition of
changes to very dissimilar environmental effects. For example, area-weighted
performance measures for changes in high- and low-flow patterns in estuaries are
added to area-weighted measures for changes to ridge-and-slough topography
when there is no obvious common effect between the two ecological settings.
Second, all performance measures are judged to be equally important, weighted
only by the size of zones to which they are applicable.
This second issue has proved especially problematic for integrating resto-
ration with recovery of threatened and endangered species, such as the Cape
Sable seaside sparrow and the snail kite. For example, take two areas: Zone A
with 100 acres of critical habitat for an endangered species, and Zone B with
1,000 acres of relatively abundant habitat not used by this species. If an alterna-
tive improves the condition of Zone A by 50 percent and Zone B by 25 percent
and the average cost of improvements are the same, the 50-habitat-unit lift
generated in Zone A represents only one-fifth of the 250-habitat-unit increase
in Zone B. If tradeoffs are necessary to achieve optimal environmental benefits
under a financial or water-budget constraint, investments to improve Zone B
would be far more highly valued than the same investment to improve critical
habitat for the endangered species in Zone A. Comparisons of highly aggregated
sums of habitat units across disparate environmental settings thus can obscure
the implicit tradeoffs that are being made between underlying ecosystem values.
Critics have argued that application of habitat suitability analysis to multiple
species is inappropriate, and in the context of the Central Everglades Planning
Project, it was applied to very different physical and biological processes that not
only impact multiple species but also a vast array of other environmental ben-
efits. Indeed Alt-4R2 is projected to have negative impacts on endangered bird
species in some locations (FWS, 2014) as was discussed previously. Protection
of endangered species will ultimately require additional detailed analysis beyond
the simple, habitat unit approach used to calculate environmental benefits, pos-
sibly resulting in some future modifications to project design.
The quest for a single benefit metric by which to compare alternatives is as
old as benefit-cost analysis, but, so far, a fully satisfactory answer remains elu-
sive. Promising advances have been made in the field of evaluating ecosystem
services, but much work remains before a comprehensive model could be
applied to the wide array of complex, interactive physical, chemical, and bio-

logical processes that make up the Everglades ecosystem. A recent report on an

ecosystem services approach to damages resulting from the Deepwater Horizon
oil spill in the Gulf of Mexico outlines a model for analysis based on fundamen-
tal concepts of microeconomics and discusses obstacles to its implementation
(NRC, 2013). Prior research and monitoring of the Everglades makes it a better
understood system than the impacted area of the Gulf. However, the lack of
well-defined ecosystem service production/response functions that relate specific
increases in ecosystem services in the Everglades to specific project attributes
(i.e., how much “lift” is achieved by the project for each of the identified ecosys-
tem services) remains an important obstacle. Another obstacle to implementation
common to both cases is the lack of a set of well-defined, relative values for the
diverse array of ecosystem services, many of which remain difficult to assess in
monetary terms. Evaluation of all services need not be in monetary terms, but, if
tradeoffs are necessary for optimization under constrained budgets, there must be
a well-defined set of preferences (or weights) among different kinds of services.
There were no such well-defined preferences for the Central Everglades Plan-
ning Project process,8 and it was not made clear what substitution of services
were being made as alternatives were being formulated, evaluated, modified, and
refined. In the absence of a well-defined set of preferences or weights for diverse
environmental benefits, a display of marginal differences of performance mea-
sures in the evaluation of effectiveness across the alternatives would be useful.
Such a display enhances transparency and understanding of the inherent tradeoffs
associated with alternative plans. Section 5 of the Draft PIR and supporting appen-
dices (USACE and SFWMD, 2013b) provide considerable detail of environmental
impacts of the final array of alternatives and the Tentatively Selected Plan. What
are not clearly shown and summarized in those displays are the principal tradeoffs
among alternatives as measured by performance measures and related ecosystem
effects that were made in moving from Alternative 4 to the Tentatively Selected
Plan. Changes (reductions) to aggregated habitat lifts were reported by zone, but
changes to the performance measures underlying aggregated measures are not
made clear. Those would be informative additions to the evaluation.
Implementation Plan
The strategy for implementing the Tentatively Selected Plan received con-
siderable attention in the planning process. Initial development of an imple-
mentation plan focused on recognizing constraints and dependencies, basic
8
The Central Everglades Planning Project attempted an evaluation of ecosystem services, but data
were lacking to support a comprehensive analysis (K. Whittmann, USACE, personal communication,
2013), and the results were not published in the August 2013 draft PIR.

sequencing of project features, reasonable estimates for cost and schedule, and
appropriate contingencies for the current level of design detail. The stated goal
was to realize restoration benefits as soon as practicable, while still adhering to
all water quality and other permitting requirements. The project team analyzed
variations of the implementation plan, ranging from a scenario with largely
unconstrained resources as a best-case to a more likely scenario governed by
moderate annual expenditures. The best case could provide full restoration ben-
efits 6 years after the Restoration Strategies water quality projects are permitted.
A more realistic schedule considering $100 million/year funding would have
those same benefits available by approximately 2053. To better understand these
timing differences, the committee looked further into the governing constraints,
sequencing of project components, and implementation uncertainties. Con-
straints and dependencies fell generally into four categories: physical (includ-
ing planning and design), operational (hydrology), operational (permitting and
compliance), and fiscal/funding (Table 3-4).
The dependencies, principles, and geography led to a simple, conceptual
organization of the Tentatively Selected Plan into three component projects, each
with a separate project partnership agreement (PPA): PPA North, encompassing
spreading and backfill projects in northern WCA-3A; PPA South, represent-
ing distribution and sheet flow features in southern WCA-3A and Everglades
National Park; and PPA New Water, representing the new storage and seepage
management features (see Box 3-1). The relationship between the implementa-
tion constraints and the three PPAs is also shown in Table 3-4.
From a construction/implementation point of view, the three PPAs were
configured to be independent from one another; the operational dependencies
were confined within each PPA. Sequencing of the three PPAs could thus be in
any order or even concurrent. However, funding realities are likely to limit the
potential to significantly overlap the PPAs. PPA North alone yielded approxi-
mately 17 percent of total project benefits at 33 percent of the total cost, with
the greatest benefit contribution to WCA-3A (Figure 3-5). PPA South alone pro-
vided about 21 percent of the total benefits at 22 percent of the total cost, with
Everglades National Park followed by Florida Bay as the greatest beneficiaries.
PPA New Water alone produced negligible benefits, but when PPA New Water
followed the other two, its total contribution was approximately 62 percent of
total project benefits at 45 percent of total cost. Figure 3-5 summarizes how the
project phases contribute individually and together to each of the five regions.
This benefits distribution analysis seems to show that moving forward with
PPA South first may have a slight edge in yielding early benefits, but the results
are essentially equivalent. Thus, the sequencing of PPAs should be a future
interagency decision that considers both existing constraints and potential for
delivering the most early benefits, particularly to regions in decline that require

TABLE 3-4 Central Everglades Planning Project Implementation Plan Constraints and
Dependencies
PPA Constrained
Constraint, PPA
Dependency, or PPA PPA New
Principle Rationale North South Water
Physical, design and planning
Availability of Timing the availability of fill for Miami Canal backfilling X X
borrow material and the Blue Shanty Levee minimizes costs
Adaptive Allowing steps and timing to test concepts will improve
management plan future success
Operational (hydrology)
A-1 FEB state Required prior to implementation of northern WCA-3A X X
restoration distribution features to ensure adequate water quality of
strategies inflows
C-358 and S-357N Construction of these seepage management features in X X
features Mod Waters is necessary to provide seepage mitigation
before increasing flows into Northeast Shark River
Slough
S-356 Operation Operation of the existing S-356 pump station is required X X
Plan to provide seepage management before increasing flows
into Northeast Shark River Slough
C-111 South Dade Completion of the detention areas required prior to X X
significantly increasing flows to Northeast Shark River
Slough to provide seepage management
BCWPA C-11 Required prior to increasing flow through S-333 or X X
Impoundment implementation of WCA-3B inflow structures along the
L-67A & C levees
Tamiami Trail Next Required prior to increasing capacities of S-333 and X X
Steps bridging and S-356 and implementation of WCA-3B inflow structures
road raising along the L-67A levee, gaps in L-67C levee and Blue
Shanty flowway
C-44 Reservoir Required prior to redirecting the maximum amount X
(IRL-S) and of water from Lake Okeechobee south to meet
connection to C-23 environmental performance
Canal
Modification of the Changes to the 2008 LORS needed prior to full utilization X
Lake Okeechobee of the A-2 FEB in order to achieve the complete
regulation schedule ecological benefits
Outlet capacity of Additional outlet capacity from WCA-3A must X
WCA-3A be provided before new project water from Lake
Okeechobee is released into the system
continued

TABLE 3-4 Continued

PPA Constrained
Constraint, PPA
Dependency, or PPA PPA New
Principle Rationale North South Water
Operational (permitting and compliance)
Restoration All features of the state’s Restoration Strategies must be X X X
Strategies Permit completed and meet state water quality standards prior
Compliance to initiating construction of most CEPP project features
Water Quality Appendix A water quality compliance issues need to be X X
Compliance in addressed for new project water entering Everglades
Everglades National National Park
Park
Compliance with Must first determine that feature will not violate state X X X
state water quality water quality standards, discharge limits, or permit
standards conditions and it will not adversely affect flora and fauna
in the area
Fiscal/Funding
Funding availability Pace of construction will be dependent on funding stream X X X
State-federal cost- Total federal creditable expenditures cannot exceed X X X

share balance state’s
SOURCES: Based on information from Bush (2013); Barnett, SFWMD, and Hobbie, USACE, personal communication, 2013;
E. Bush, USACE, personal communication, 2013.
long times to recover (NRC, 2012a). For the purposes of further analysis, the
draft Implementation Plan utilized the PPA North è PPA South è PPA New
Water sequence as a base.
Building on this base sequence and allowing for a small amount of overlap
between PPA phases, the project team produced a Central Everglades project
implementation schedule including recognized constraints and considering a
nominal $100 million/year funding level. This scenario estimated 19 years from
construction start to completion and was viewed as a realistic time line (although
it lacked risk-based schedule contingencies). As an extreme, best-case scenario,
executing all three PPAs concurrently with no resource constraints produced a
total construction duration of 6 years (E. Bush, personal communication, 2014).
The initiation of construction of the Central Everglades Planning Project will
be contingent upon numerous project dependencies (see Table 3-4), including
improvements in water quality. The draft PIR includes the statement that “all fea-
tures of the State’s Restoration Strategies must be completed and meet state water
quality standards prior to initiating construction of most CEPP p roject features”

BOX 3-1
Proposed Groupings of Central Everglades Project Features
PPA North
• L-6 Diversion
• S-8 Pump Modifications
• L-4 Levee Degrade and Pump Station
• L-5 Canal Improvements
• Miami Canal Backfill
PPA South
• L-67A Structure 1 North
• One L-67C Gap
• Increase S-356
• Increase S-333
• L-29 Divide Structure
• L-67A Structures 2 and 3 South
• L-67A Spoil Mound Removal
• Remove L-67C Levee Segment
• Remove L-67 Extension Levee (no backfill)
• 8.5-Mile Blue Shanty Levee
• Remove L-29 Levee Segment
• Backfill L-67 Canal Extension
• Remove Old Tamiami Traila
PPA New Water

• Seepage Barrier L-31N
• A-2 FEB
a
Removal of Old Tamiami Trail can be completed at any time during implementation,
but must precede backfilling of L-67 Extension Canal.
Source: E. Bush, USACE, personal communication, 2014.
(USACE and SFWMD, 2013b). Any project construction would require certifi-
cation by the state that the project would not lead to a violation of state water
quality standards. However, it does not necessarily follow that construction of
most Central Everglades Planning Project components must be deferred until all
components of the Restoration Strategies are implemented. For example, construc-
tion of the A-2 FEB and improved seepage management would only improve water
quality and could provide a means to expedite overall construction in the face of
other constraints. Combining the large contingencies resulting from funding with
the stated implementation and permitting constraints driven by the Restoration
Strategies project leads to a rather alarming picture of potential restoration prog-

180
160
Environmental Benefits, 1000 HU's
140
120 Florida Bay

100 ENP
80 WCA 3B
60 WCA 3A
Northern Estuaries
40
20
0
PPA North PPA South PPA New Water
FIGURE 3-5 PPA contribution to Central Everglades Planning Project benefits by region. PPA
New Water benefits are dependent upon the completion of both PPA North and PPA South.
SOURCE: Based on information from E. Bush, USACE, personal communication, 2014.
ress. Figure 3-6 shows the serial effect of constraints, permitting, construction phas-
ing, and contingency addition to the scheduled availability of project benefits—a
scenario in which the first new water benefits occur approximately four decades
into the future with funding of approximately $100 million/year in constant dol-
lars. Allocating schedule contingency from Appendix B in USACE and SFWMD
(2013a) proportionally to each PPA, the first benefits based on existing water could
be about 21 years from now, assuming that Central Everglades Planning Project
construction is not initiated until all of the Restoration Strategies projects have
been permitted. At three decades into the future, realized benefits could still only
be 38 percent or less of the full Central Everglades Planning Project. Different
funding streams could produce different scenarios, and more recent updates to the
cost engineering appendix of the PIR present even longer possible time frames.9
Since the August draft PIR was released, the Cost Engineering Appendix B has been revised
9
sing a project duration of 329 months with a contingency of 89 months, for a total of 34.8 years
u
at an 80 percent confidence level. This updated estimate reflects a scenario of $100 million/year,
not adjusted for inflation (Amro Habib, USACE, personal communication, 2014). In contrast, the
August draft PIR presented a schedule duration of 186 months with a contingency of 107 months,
for a total of 24.4 years at an 80 percent confidence level (USACE and SFWMD, 2013a).

FIGURE 3-6 Schedule implications of serial phasing of state water quality construction, operations, and
permitting of all STAs (as a unit), and Central Everglades project partnership agreement (PPA) construction.
The years shown reflect the risk-based schedule according to USACE and SFWMD (2013b, Appendix B) of
186 months with an additional 107 months of schedule contingency (at 80 percent confidence), based on
a $100 million/year constant dollar funding scenario. The 293-month total construction time (including
contingencies) was apportioned to the three PPAs according to time lines presented in a draft constrained
implementation schedule (E. Bush, USACE, personal communication, 2014).
The committee judges that this is an unreasonable and undesirable result.

To avert continued ecosystem declines in the central Everglades, including some
that would require decades to centuries to recover (NRC, 2012a), the Corps and
SFWMD need to look for creative implementation strategies to reduce exist-
ing constraints. To take advantage of the likelihood that the Central Flowpath
(STA-2 and -3/4) achieves compliance with the WQBEL much earlier than the
remainder of the STAs (see Chapter 4), the agencies should investigate design,
implementation, and permitting alternatives that would enable the C entral Ever-
glades P
lanning Project to move forward as quickly as possible with WQBEL-
compliant discharges, This will require a thorough evaluation of the risks, costs,
and benefits of such actions to the entire South Florida ecosystem and collabo-
ration among multiple agencies and stakeholders. Additionally, an increased
and consistent funding profile would have a major impact on achieving Central
Everglades restoration goals sooner.
Adaptive Management and Monitoring
The Central Everglades adaptive management plan (USACE and SFWMD,

2013b) was developed to identify key areas where restoration efforts would

benefit from monitoring and assessment and outline how proposed and existing
monitoring could be used to adapt project implementation to reduce uncer-
tainties and improve restoration outcomes. Additionally, the plan was to docu-
ment the monitoring necessary to meet project-specific regulatory and permit
objectives. It is intended to serve as a guide on the use of monitoring data to
maximize restoration benefits while reducing costs and to inform project design
and implementation to improve performance. Hence, the adaptive management
plan represents a highly ambitious balancing act, given the multiple objectives
and the scope of the Central Everglades Planning Project.
The Central Everglades Planning Project adaptive management team identi-
fied key project uncertainties, defined as planning questions “regarding the best
actions to achieve desired goals and objectives within constraints, which cannot
be fully answered with available data or modeling.” These uncertainties reflect
scientific understanding of the anticipated restoration responses to manage-
ment actions rather than administrative and funding uncertainties surrounding
scheduling and implementation. Based on criteria developed from RECOVER
(2011a), the team screened and prioritized uncertainties, selecting only those
uncertainties that
1. Are directly related to Central Everglades Planning Project goals, objec-

tives, or constraints;
2. Are focused at the project scale;
3. Are not already well understood;
4. Have at least one measurable attribute that can provide information to
resolve the uncertainty; and
5. Have adaptive management options (i.e., the ability to be addressed
through adjustments to restoration plans).
The uncertainties identified were further prioritized according to three

additional criteria:
1. The risk of not achieving restoration goals if the uncertainty is not resolved,
2. The level of knowledge about the uncertainty, and
3. The relevance or level of confidence that the uncertainty could be resolved.
Uncertainties that scored high on risk and relevance and low on knowledge
were ranked higher than those with low risk and relevance and high knowledge.
The highest priority uncertainties resulting from this process are listed in Box 3-2.
For each uncertainty listed in Box 3-2, the adaptive management plan
includes information on the drivers of uncertainty, CERP- and project-level tar-
gets for ecosystem attributes related to the uncertainty, and a plan for monitoring

BOX 3-2
Central Everglades Planning Project Uncertainties
The following uncertainties were identified in Annex D of the Central Everglades

Planning Project PIR (USACE and SFWMD, 2013b):
CEPP-wide: How will CEPP influence the introduction and growth of nonnative and
native nuisance species within the project area, and will the species influence the pre-
dicted landscape and performance of CEPP?
Lake Okeechobee: Will CEPP’s operational refinements for Lake Okeechobee affect its
littoral and nearshore vegetation coverage?
Flow Equalization Basins (FEBs): How can the operation of the FEBs be optimized to
maximize flows to the Everglades while serving the needs of Lake Okeechobee and
the northern Everglades?
Northern Everglades, St Lucie Estuary: Do reductions in high flows result in measurable

increases in submerged aquatic vegetation (SAV) coverage in the estuary? To what
extent will the reduction in the frequency and magnitude of high flows to the estuary
stabilize conditions sufficiently to improve benthic habitat in the south fork of the estu-
ary? To what extent will the reduction of frequency and magnitude of high flows to the
estuary help reestablish historic oyster beds on the south fork of the estuary?
Northern Everglades, Caloosahatchee Estuary: Do reductions in high flows result in

measurable increases in SAV coverage and oyster acreage and health in the estuary?
Will the reduction in low-flow violations in the estuary help reestablish persistent
Vallisneria beds in the upper estuary?
Greater Everglades: Are flow velocities, direction, volumes, and depth improvements
from CEPP sufficient to reestablish historic ridge-and-slough landscapes? Can CEPP
create hydrology favorable for tree island elevation requirements? Are inundation and
hydroperiod sufficient to reduce current high rates of soil oxidation and peat fires?
the attributes to track progress toward the targets. The plan also discusses the
time frame in which changes in the attributes will be measurable, and identifies
triggers or thresholds that would give early warning that project performance is
deviating from restoration goals. Management options are proposed that could
then be chosen based on monitoring results. The adaptive management plan
relies on data from hydrometeorological, ecological, water quality, and nuisance
and exotic vegetation monitoring and, as such, is tightly connected to monitoring
activities associated with project implementation within and beyond the Central
Everglades Planning Project. Although new monitoring activities are proposed,
the adaptive management plan also relies heavily on data from existing monitor-

How much will CEPP improve alligator relative density and body condition in northern
WCA-3A and -3B and northeast Shark River Slough? How much will hydrologic resto
ration and vegetation management result in increases in prey densities? How much
will hydrologic restoration and vegetation management result in increases in wading
bird foraging conditions and increased nest number and success of wood storks and
roseate spoonbills?
Greater Everglades/Lower East Coast: Will the full suite of CEPP TSP structures be
required in WCA-3B to create the Blue Shanty Flowway?
Everglades National Park (ENP)/Southern Coastal Systems: Will there be downstream

biogeochemical effects associated with modifying inflows and hydrologic conditions in
ENP, including effects on nutrient movement, availability, and ecological responses? Will
increased flows to northeastern Shark River Slough yield natural distribution of waters
toward the southeastern Everglades and northeast Florida Bay without operation of the
SFWMD Canal System east of L-30, L-31N, and L-31W? Will CEPP improve flows to
Florida Bay and the lower southwest coast, resulting in more natural salinity patterns,
and will responses be consistent with the expectations from the CEPP scenario model
predictions? Will predicted CEPP flows mitigate saltwater intrusion and associated
coastal wetland vegetation, soil stability, and nutrient retention or release? If salinity is
affected by overland flow increases through ENP to Florida Bay, how much benefit
is generated for SAV, prey, coastal wading birds, and crocodiles, and can operations
be adjusted to improve estuarine performance in Florida Bay?
Lower East Coast: Will the constructed and operational features of CEPP maintain flood
risk management level of service east of the L-30, L-31N, L-31W and C-111 without
reducing quantity or quality of groundwater in water supply well fields compared to exist-
ing conditions? Will the constructed and operational features of CEPP reduce surface
and/or groundwater base flows and wetland/groundwater recharge to the east of the
L-30 and L-31N in areas such as the Pennsuco Wetlands, south Miami-Dade wetlands,
and Biscayne Bay?
ing and ongoing research (such as the Science Plan for the Everglades Stormwater
Treatment Areas [SFWMD, 2013d] discussed in Chapter 7).
The Central Everglades Planning Project team has developed the most struc-
tured, complex, and comprehensive adaptive management plan within the CERP
to date. Because it has been developed in tandem with the Central Everglades
Planning Project process, rather than retrospectively, it exhibits greater integra-
tion, relevance, and coordination with implementation of the restoration plan.
Thus, monitoring data derived from the plan are more likely to be useful for
determining ecological responses to project implementation and the ability to
meet restoration goals. Although the adaptive management plan is intended

to provide flexibility and robustness in the face of unexpected or surprising

events (i.e., unknown uncertainties) within a highly dynamic system, aside from
operational changes, much of the flexibility would be derived from future incre-
ments of restoration in the central Everglades, rather than within the project itself.
Although the plan addresses system dynamics and variability, it does not
appear to take into account the strong likelihood of nonstationary processes
(i.e., system dynamics in which the trend and variability change through time).
This is particularly pertinent to the effects of climate change, for instance, where
not only the average temperature and precipitation are projected to change, but
temperatures and precipitation are projected to exhibit greater variance (Hansen
et al., 2012) (see Chapter 5). Nonstationary systems and processes are notori-
ously difficult to monitor and manage because they turn what are assumed to
be stationary baseline conditions, against which restoration responses are com-
pared, into a moving target. For the Central Everglades Planning Project, and
indeed much of the CERP, sea-level rise, climate change, and the introduction
of new invasive species are perhaps the most prominent threats on the horizon
that could impede the ability to measure restoration relative to starting condi-
tions. These are also highly uncertain in their timing and projected effects on
the system. In this context, flexibility to address these potentially disruptive
uncertainties should be explicitly built into adaptive management and monitor-
ing recommendations, and they should be made more explicit and expanded
upon in the adaptive management plan.
Stakeholder Engagement
CERP stakeholders range across a diverse set of actors including, but not
limited to, federal, state, and district agencies, tribes, environmental non
governmental organizations and community groups, recreation groups, agri-
cultural entities, and individual members of the general public. This is, in great
part, due to the multiobjective and multifaceted nature of restoration in this
region that cuts across social, economic, ecological, and hydrologic concerns
over a broad landscape. The types of stakeholder engagement elicited during
CERP project planning has typically focused on engaging other federal, state,
local, and tribal government agencies as part of the project development team
process. Under the Federal Advisory Committee Act (FACA), representatives from
nongovernmental organizations, recreation groups, and agricultural interests
are typically unable to actively and collaboratively participate in CERP project
planning. Public participation in these meetings is typically limited to brief (2- to
3-minute) comments with no mechanism for discussion. Recognizing that greater
acceptance is likely when stakeholders and the public are afforded opportuni-
ties to actively participate in the planning process, the USACE and the SFWMD

worked with the FACA-exempt Task Force to facilitate enhanced stakeholder

engagement for the Central Everglades Planning Project.
Building on experiences gained through previous stakeholder processes,
such as the state’s non-CERP River of Grass planning effort, the Task Force’s
Working Group, the Science Coordination Group, and the SFWMD’s Water
Resources Advisory Commission, moderated a total of 16 public workshops—
many that were day-long or longer—that emphasized enhanced public and
stakeholder engagement (Figure 3-7). This provided opportunities for two-way
dialogue at a more detailed level and accommodated input from a broader
representation of stakeholders and the public on the scope of the study; the
FIGURE 3-7 Public and stakeholder group workshops, showing feedback mechanism to the
Corps and the SFWMD.
SOURCE: http://www.sfrestore.org/wg/wgminutes/2011meetings/11172011/Draft_Protocol.
pdf.

development of goals, targets, and alternatives; and plan selection. One exam-
ple of active input from stakeholders and the public was the use of configura-
tion development exercises to elicit input on specific restoration strategies and
implementation sequencing. Although not aiming for consensus, the results
of the configuration development exercises allowed the Task Force to provide
informed input to the Corps and the SFWMD during planning. Feedback from
participants on the effectiveness of the enhanced public and stakeholder par-
ticipation process, and the outcomes that have arisen from it, has been mixed.
Feedback has been overwhelmingly positive about the extent to which the
Corps and the SFWMD have reached out to a broad range of stakeholder and
public entities to provide information and seek feedback, particularly with the
recreation community. The configuration development exercises were regarded
as valuable in soliciting input on some project components as they challenged
participants to offer justified preferences for plans. This exercise ultimately influ-
enced the recommendations, largely due to participation and feedback from the
recreation community. Other aspects of the enhanced public and stakeholder
participation process that were regarded as positive include a greater degree
of two-way dialogue not seen in other planning endeavors in the CERP, the
adoption of mechanisms to provide feedback and to change plans through more
active and inclusive engagement, greater effort to explain the plan to broader
stakeholder and public groups, and a more transparent planning process involv-
ing a more engaged community.
Opinion on the mechanism for feedback into the planning process (i.e.,
the process of using stakeholder comments, concerns, and suggestions to make
changes) was mixed. Some stakeholders point to the change in alternatives
based on concerns raised by the recreation community as evidence of a genuine
feedback mechanism. However, others assert that the process by which feed-
back would be used to change plans lacked transparency from the outset, that
input was accepted and acted upon on the basis of convenience and ease of
implementation, and that disparate views across the broader stakeholder body
were not addressed. It was the view of one stakeholder that most of the attention
and discussion arising during workshops focused on the northern portions of the
central Everglades because that is where most stakeholder and public interest
lay; however, this led to incomplete or superficial consideration of the southern
areas of the central Everglades and seepage issues.
Criticism has also been raised about the time frame of the Central Everglades
Planning Project. Although a broad representation of stakeholders and the public
was generally well informed by the process, those with knowledge of the CERP
and the hydrologic models used in CERP planning had insufficient time and
information to keep up with the dissemination of model output generated by
the modeling team. Hence, stakeholders were limited in their understanding of

the modeling results to what the agencies presented, rather than being able to
gain insights from their own analysis of the data.
Numerous challenges arose during the enhanced public and stakeholder
engagement process. First, to comply with FACA, multiple duplicate meetings
were necessary, causing meeting fatigue. The numerous day-long, information-
rich public workshops created a very intense process that was challenging
to keep up with. And although the process of stakeholder engagement was
regarded as an improvement over previous planning efforts, disappointment
was expressed that some key stakeholders did not actively participate in the
process, because they were either absent or reticent.
On balance, this committee judges that the enhanced stakeholder and
public engagement process was well executed and inclusive. Although the
expedited time frame was likely too fast to ensure rigorous consideration of all
components of such a complex project, this committee commends the efforts to
structure stakeholder and public engagement to allow more active participation
than in previous planning processes. The limitations incurred by an expedited
planning time frame aside, the efforts to educate stakeholders and the public
and to elicit comments, criticisms, and modifications to the Central Everglades
Planning Project have been exemplary and should serve as a model for future
planning processes.
The Process
Despite some delays in the schedule, the Central Everglades Planning Project
team produced a public version of a decision document within approximately
22 months and is anticipated to have a Chief’s Report in a total of 34 months.
Although there has been no formal survey of participants and stakeholders, it
appears that the project was well administered, stakeholders were generally
pleased with the opportunity to participate in the process directly, and it pro-
duced a useful product. Brief slippage in the schedule came in part because
of extra time required to develop confidence that the project could satisfy the
Savings Clause and provide project assurances as required in the authorizing
legislation, but another contributing factor in the delay was the tardy insertion
of concerns about additional public water supply that were not part of the early
scoping process. That is a reality for all planning processes. It is not uncommon
for some stakeholders to become engaged in a planning process after scoping
has been completed or after initial plans have been formulated and evaluated. At
that point, they are better informed as to what has or has not been included in
the alternatives. Although it is unfortunate that all interests are not always at the
table at the beginning of a process, concerns of late-arriving stakeholder groups
can be dismissed only at the risk of further delay in the review process. Only

time will tell as to whether all interests were sufficiently satisfied by address-
ing the stated need for additional water supply. The expedited process focused
more narrowly on environmental restoration than the broader goals of the 1999
Restudy (and WRDA 2000) and as a result, the Central Everglades Planning
Project may not enjoy the broad support of the original CERP. The project also
encountered some delays while unresolved issues were addressed during the
review process, particularly for policy issues associated with the FWS Biological
Opinion (FWS, 2014). Collectively, these delays meant that the project was not
authorized in WRRDA 2014, which proved to be a significant disappointment
for the staff and stakeholders who had committed extensive time and energy to
the project for more than 2 years. Whether this delay proves to be significant
to restoration progress remains to be seen. If the Central Everglades Planning
Project can be authorized through a future Water Resources Development Act
(or some other mechanism) within approximately the next 3-5 years, the impacts
of this delay are likely to be small, because numerous project dependencies
need to be constructed before the Central Everglades construction can proceed.
However, further authorization delays could lead to missed opportunities to
expedite restoration progress with the greatest benefits for the natural system.
The original target of 18 months to complete the planning process for such a
complex project proved to be overly ambitious. The staffing levels and intensity
were greater than anticipated, leading to total planning costs that were similar
to the prior 6-year planning model (M. Morrison, SFWMD, personal commu-
nication, 2013). Those involved in managing the expedited process suggested
that a 3-year time frame might be a better, more achievable goal. The agencies
supported the aggressive schedule and committed key personnel for the duration
of the study. However, not all agencies were able to make the same resource
commitments, and some agencies—particularly local governments—found it
challenging to keep up with the expedited process.
Communication within and between agencies, especially vertically into
senior management, was a particular strength of the expedited process. Leader-
ship was informed quickly of issues that required resolution and was involved
in developing solutions. This strategy significantly improved the efficiency of the
planning process compared with the prior USACE planning model.
It was recognized throughout the early alternatives analyses and plan selec-
tion that the level of site-specific information available and the features’ design
details were less than the normal, non-expedited planning process, posing pos-
sibly greater project risks. Probabilistic cost models at an 80 percent confidence
level yielded a construction cost contingency of 44 percent. Using escalation
adjustments and calculating on a constant-dollar basis, the estimated total
project cost was $1.9 billion (including $571 million for contingency; E. Bush,
USACE, personal communication, 2014). Similarly, the total project duration at

an 80 percent confidence level was originally reported as 293 months (includ-

ing a contingency of 107 months) (USACE and SFWMD, 2013b).10 A sensitivity
analysis provided a means of ranking the larger contributors to project risk, and
the uncertainty in the “funding profile” was the biggest contributor to both cost
and duration variation. Several feature design uncertainties also contributed to
the cost risk, but no other variables compared to funding as a driver of schedule
risk. This dominance of the funding uncertainty is particularly noteworthy in light
of the less-complete design and site data levels, and it reinforces the validity of
the expedited, more conceptual design planning process.
Although the number of unique alternatives specified in the expedited pro-
cess appears to be less than a typical USACE 6-year process, there were effec-
tively a large number of possibilities that were pared down through a rational,
defensible process. The four alternatives with the major variations described
earlier were actually a much larger number of possible combinations. Dividing
up the large geographic area into regions based on functional characteristics
(e.g., storage, conveyance) was an effective way of reducing the number of com-
binations and communicating both within and external to the team. Overall, the
planning organization, backed by the extensive technical modeling, gives strong
support to the conclusion that the range of alternatives evaluated was reasonable
and appropriate for this planning process. Given the time constraints, however,
there was limited ability to cycle back to earlier alternatives once assumptions
or options for a geographic area were defined. Simplistically, the expedited
process was a more linear approach than the typical process and provided less
opportunity for iteration. In the committee’s judgment, however, the weaknesses
of the expedited process did not materially diminish the quality of the final plan.
Some important analyses, however, were overlooked in the expedited plan-
ning process. Beyond a cursory examination of the effects of sea-level-rise
scenarios on project benefits due to land loss, there was a distinct lack of
consideration of the potential effects of climate change on restoration benefits
under each alternative. This is due, in part, to the expedited time frame driving
the project that limited such an in-depth analysis and the uncertainty inherent
in quantifying these effects with an acceptable degree of accuracy. However, the
lack of consideration of the effects of climate change paints an incomplete pic-
ture of hydrologic and ecosystem response to the alternatives examined. Explicit
consideration of how the performance measures are affected by increased
temperature, changes in precipitation (and increased variability in temperature
10
Since the August draft PIR was released, the Cost Engineering Appendix has been revised to
include an even-longer total project duration of 329 months with a contingency of 89 months, for
a total of 34.8 years at an 80 percent confidence level. This updated estimate reflects a scenario of
$100 million/year, not adjusted for inflation (Amro Habib, USACE, personal communication, 2014).

and precipitation), and sea-level rise would likely lead to revised habitat unit
estimates. Additionally, such an analysis would have enabled the quantification
of specific benefits associated with mitigating the impacts of climate change,
which might have elucidated additional alternatives pertinent to performance
measures explicitly related to climate change impacts, such as salinity enve-
lopes, slough vegetation, and extreme high and low water depths in the Greater
Everglades. Future planning efforts will need to consider the impacts of climate
change on performance measures and restoration goals more broadly, despite
the uncertainty associated with temperature and, particularly, precipitation pro-
jections (see Chapter 5). Scenario planning (Peterson et al., 2003; Polasky et al.,
2011), currently adopted by the National Park Service (NPS, 2013), provides a
structured framework for decision making under alternative futures in the face
of uncertainty in climate change projections and should be considered in future
planning efforts. A more complete analysis of the impacts of climate change and
quantification of the benefits associated with mitigating these could potentially
lead to increased public and political support for the Central Everglades Plan-
ning Project.
CONCLUSIONS AND RECOMMENDATIONS
The Central Everglades Planning Project effort is responsive to the need to

expedite restoration in the central Everglades via increments of restoration
to avert further declines that could take many decades or longer to recover. Over-
all, the project team did an impressive job under a challenging time frame. The
proposed plan seems reasonable and thoughtfully developed with substantial
stakeholder input. Implementation of the plan would provide significant benefits
to the remnant Everglades ecosystem, including more than 200,000 AF/yr of
new water—a sizeable first increment of restoration for the central Everglades
that represents approximately two-thirds of the new water to northern WCA-3
envisioned in the CERP. A comprehensive adaptive management plan provides
an important mechanism to learn from project implementation to improve the
operation of the project and the design of future increments of restoration,
although additional attention to climate change uncertainties is needed.
If the Central Everglades Planning Project is to avert further ecosystem
degradation, CERP planners and policy makers need to expedite project imple-
mentation in the face of several hurdles. The best-laid plans will be of little
benefit if the project is not implemented in a timely way. Completion of the
Chief’s Report for the Central Everglades Planning Project, congressional authori-
zation, and construction of project dependencies are key near-term steps neces-
sary to move forward. Project funding and water quality permitting constraints
currently appear to be the largest barriers to timely project implementation.

Creative solutions may be available to significantly expedite restoration, such as

finding permit mechanisms to move water that meets water quality criteria into
the Everglades prior to completion of the entire Restoration Strategies project.
Such approaches will require the agencies to recognize the urgency and to work
to find legal and engineering solutions to move increments of water into the
Everglades as soon as those increments have been adequately treated to meet
water quality standards. Without such solutions, redistribution of existing water
may not be feasible until 2035 or beyond and at the envisioned funding level of
$100 million per year, construction would not be completed for approximately
four decades—exceedingly long for a system already in significant decline.
Some important lessons were learned from the expedited planning process.
Although overall, participants and stakeholders thought the process led to a use-
ful outcome, the 22-month time frame to produce the PIR was extremely chal-
lenging for staff and stakeholders, alike. The process required large numbers of
staff and became the central focus of the restoration program. Such attention was
deserved for this high-priority initiative, but similar intensive efforts are unlikely
to be sustainable for future CERP planning. Furthermore, stakeholders with
technical expertise found it difficult to keep up with the pace of model output
presented and hence could not adequately evaluate the information provided.
Thus, 3 years might be a more reasonable time frame for such a complex effort.
Communication within and between agencies was a particular strength of the
expedited process; senior decision makers were involved at key decision points
and as needed to resolve issues and improve planning efficiency. However, the
existing Corps process for evaluating restoration benefits makes it difficult to be
transparent about tradeoffs in planning decisions.
The enhanced stakeholder and public engagement process was well exe-
cuted and should serve as a model for future planning processes. This level of
active and inclusive stakeholder engagement had not previously been imple-
mented as part of the CERP, and it provided a means for two-way dialogue
between stakeholders and agency staff that substantially influenced the planning
outcome. Although the abbreviated time frame led to concerns from partici-
pants, overall, the committee commends the efforts to educate and engage the
stakeholders and modify the project plan where feasible to address concerns.


Implementation Progress
This committee is charged with the task of discussing significant accom-

plishments of the restoration and assessing “the progress toward achieving
the natural system restoration goals of the Comprehensive Everglades Restora-
tion Plan [CERP]” (see Chapter 1). In this chapter, the committee updates the
National Research Council’s (NRC’s) previous assessments of CERP and related
non-CERP restoration projects (NRC, 2007, 2008, 2010, 2012a). This chapter
also addresses programmatic and implementation progress as well as analyzes
any natural system benefits resulting from the progress to date.
PROGRAMMATIC PROGRESS
To assess programmatic progress the committee reviewed a set of primary

issues that strongly influence the progress of the CERP toward its overall goals of
ecosystem restoration. These issues, described in the following sections, relate
to authorization, funding, and scheduling.
Project Authorization
Once project planning is complete, CERP projects with costs exceeding

$25 million1 must be individually authorized by Congress.2 Water Resources
Development Acts (WRDAs) have served as the mechanism to congressionally
1
Programmatic authority for smaller projects (less than $25 million each) was subject to a total
limit of $206 million (Water Resources Development Act of 2000 [WRDA 2000]).
2
WRDA 2000 included authorizations for 10 initial Everglades restoration projects (pending
congressional approval of the project implementation reports [PIRs]), and an adaptive management
and monitoring program. WDRA 2000 stipulated that the initial project authorizations are subject to
Section 902 of WRDA 1986, thereby requiring reauthorization if project costs increase by more than
20 percent of the original authorized cost (exclusive of inflation). As a result of the Section 902 limits
or other major project changes, all 10 conditionally authorized projects now require reauthorization
(S. Appelbaum, USACE, personal communication, 2012).
71
authorize U.S. Army Corps of Engineers (USACE) projects, and the CERP plan-
ning process was developed with the assumption that WRDAs would be passed
every 2 years. This, however, has not occurred. In the 13 years after the CERP
was launched in WRDA 2000, Congress passed only WRDA 2007, which
authorized Indian River Lagoon-South, Picayune Strand Restoration, and the Site
1 Impoundment projects. Additionally, the Melaleuca Eradication Project was
authorized under programmatic authority. These four projects are considered
Generation 1 projects (see Table 4-1; Figure 4-1), and for the past 7 years, they
have been the only projects that could receive federal appropriations to support
construction.
In May 2014, Congress passed the Water Resources Reform and Develop-
ment Act (WRRDA), which authorized four additional projects (C-43 Reservoir,
C-111 Spreader Canal, Biscayne Bay Coastal Wetlands, Broward County Water
Preserve Areas), termed Generation 2 projects. With the passage of WRRDA
2014, the federal government will be able to maintain progress on several state-
expedited projects now under way (e.g., C-111 Spreader Canal, Biscayne Bay
Coastal Wetlands) and initiate construction on two other new projects.
The expedited planning process for the Central Everglades Planning Project
was originally targeted for completion in December 2013 in anticipation of it
being included in a 2013 authorization. Delays in the planning process, how-
ever, pushed it beyond closure on WRRDA 2014 (see Chapter 3). Slippage in
the completion schedule for the Central Everglades Project beyond closure for
WRRDA 2014 delays authorization of that project to the next WRDA (or other
authorization mechanism). As discussed in Chapter 3, the implications of such
delays on restoration progress should be relatively minor if the next WRDA is
passed in the next 3-5 years.
Funding
Funding for restoration of the South Florida ecosystem comes from a variety
of federal and state sources.3 A combination of several factors has contributed
to deceleration in the rate of spending for restoration of the Everglades over
3
Federal agencies include the USACE, four Department of the Interior agencies (National Park
Service, U.S. Fish and Wildlife Service, Bureau of Indian Affairs, and U.S. Geological Survey),
two Department of Agriculture agencies (Natural Resources Conservation Service and Agricultural
Research Service), National Oceanic Atmospheric Administration, and U.S. Environmental Protec-
tion Agency. The South Florida Water Management District (SWFMD) is the nonfederal partner for
CERP and several non-CERP cost-shared projects, and the State of Florida directs monies from state
appropriations and several trust funds into SFWMD accounts. Two trust funds have been particularly
important to CERP projects—Florida Forever and Save Our Everglades. Other Florida agencies that
contribute include the Departments of Agriculture, Environmental Protection, and Transportation,
and the Fish and Wildlife Conservation Commission.

TABLE 4-1 CERP or CERP-Related Project Implementation Status as of May 2014

Yellow IDS Construction

Book (1999) (Aug. 2011) Status;
Estimated Estimated Installation and
Completion Completion PIR (or PPDR) Authorization Planning/ Testing Status for
Project or Component Name Date Date Status Status Design Pilots
PILOT PROJECTS
Hillsboro ASR Pilot 2002 Not specified PPDR Final Authorized in Completed Completed, 2013
(Fig. 4-1, No. 1) Oct. 2004 WRDA 1999
Kissimmee ASR Pilot 2001 Not specified PPDR Final Authorized in Completed Completed, 2013
(Fig. 4-1, No. 2) Oct. 2004 WRDA 1999
Regional ASR Study NA Not specified NA NA Completed Ongoing
LPA Seepage Management Pilot NA Not specified NA NA Completed Completed
(Fig. 4-1, No. 3)
C-111 Spreader Canal Design Test NA 2011 NA Programmatic Completed Testing completed
(Fig. 4-1, No. 9) authority WRDA
2000
Decomp Physical Model NA 2014 NA Programmatic Completed Ongoing
(Fig. 4-1, No. 4) authority WRDA

2000

continued

73
74
TABLE 4-1 Continued

Yellow IDS Construction
Book (1999) (Aug. 2011) Status;
Estimated Estimated Installation and
Completion Completion PIR (or PPDR) Authorization Planning/ Testing Status for
Project or Component Name Date Date Status Status Design Pilots
RESTORATION PROJECTS—Generation 1
Picayune Strand Restoration 2005 Merritt: 2012 Submitted to Construction Completed Prairie Canal
(Fig. 4-1, No. 5) Faka-Union: Congress in authorized in completed in 2007
2014 2005 WRDA 2007 Merritt, Faka Union
Miller: 2016 Reauthorization ongoing pending
required because Sec. 902 limits
of cost increases
Site 1 Impoundment* 2007 Submitted to Construction Completed
(Fig. 4-1, No. 6) Congress in Authorized in
2006 WRDA 2007
• Phase 1 2013 Ongoing

• Phase 2 TBD Phase 2 requires Not begun
reauthorization
Indian River Lagoon-South Not specified Submitted to Construction
(Fig. 4-1, No. 7) Congress in authorized in
2004 WRDA 2007
• C-44 Reservoir/STA* 2007 2018 Completed Ongoing
Melaleuca Eradication and Other 2011 2012 Final June 2010 Programmatic Completed Construction
Exotic Plants authority WRDA completed 2013
(Fig. 4-1, No. 7) 2000 Operations ongoin.


C-111 Spreader Canal* 2008 2012 Approved by Construction Completed Ongoing; expedited
USACE Chief of authorized in by FL
• Western Project (PIR #1) (Fig. 4-1,
Eng. in Jan. 2012 WRRDA 2014
No. 9)
Biscayne Bay Coastal Wetlands 2018 2016 Approved by Construction Completed Ongoing; expedited
(Phase 1) USACE Chief authorized in by FL
(Fig. 4-1, No. 10) of Eng. in May WRRDA 2014
2012
C-43 Basin Storage: West Basin 2012 TBD Approved by Construction Ongoing Not begun
Storage Reservoir USACE Chief of authorized in
(Fig. 4-1, No. 11) Eng. in Jan. 2011 WRRDA 2014
Broward County WPAs Approved by Construction
(Fig. 4-1, No. 12) USACE Chief authorized in
of Eng. in May WRRDA 2014
• C-9 Impoundment* 2007 Not specified Ongoing Not begun
2012
• Western C-11 Diversion 2008 2018 Ongoing Not begun
Impoundment*
• WCA-3A & -3B Levee Seepage 2008 2022 Ongoing Not begun
Management*
Central Everglades Planning Project NA NA Draft PIR Aug. Not authorized Ongoing Not begun
(Fig. 4-1, Nos. 13 and 14) 2013
Loxahatchee River Watershed (Fig. Not specified Not specified In development Not authorized Ongoing Ongoing; expedited
4-1, No. 15) by FL
* Projects that were conditionally authorized in WRDA 2000, subject to approval of the PIR.
NOTES: Projects in Table 4-1 reflect those CERP projects or pilot projects deemed by the committee to be relevant to CERP progress. This table does not include
non-CERP foundation projects. Gray shading of project names reflects projects being expedited and/or carried out entirely with state funding as of 2014. Gray
shading of construction cells indicates past or present aspects of projects that were expedited with state funding. NA = not applicable; TBD = to be determined.

SOURCE: www.evergladesplan.org.

75
FIGURE 4-1 Locations of CERP and CERP-related projects and pilots listed in Table 4-1. Projects under active
construction are noted with a green circle.

Implementation Progress 77
the past 3 to 4 years. Reductions in budgets for restoration by federal and state
agencies, court mandates requiring the SFWMD to make additional water quality
improvements, delays in congressional authorizations of additional projects, and
rules governing cost sharing have all had an impact. Those trends and potential
bottlenecks are explored in more detail in subsequent sections.
CERP Spending
Figures reported by the South Florida Ecosystem Restoration Task Force (Task
Force) indicate that federal and state governments have spent $4.25 billion on
CERP projects since passage of WRDA 2000 through FY 2013 (SFERTF, 2014).
Spending by the State of Florida, including the SFWMD, has substantially out-
paced federal spending—about $3.23 spent toward the CERP by the state for
every dollar spent by the federal government. Rates of CERP spending have been
highly variable, as shown in Figure 4-2. Florida’s spending grew at a fast pace
through FY 2007, remained high in FY 2008, and rapidly declined afterward.
Federal spending on the CERP accelerated after passage of WRDA 2007, peaking
in FY 2010, and declining since then.
FIGURE 4-2 Spending on CERP projects by federal and state governments.

NOTE:* Requested.
SOURCE: SFERTF (2014).

All South Florida Ecosystem Restoration Spending
The economic downturn and related fiscal constraints at both national and
state levels have slowed the rate of spending on most ecosystem restoration
projects in South Florida, not just the CERP. Some items that have been included
in the Task Force’s cross-cut budget may be indirectly related to restoration,
but those entries do not alter the general decline in expenditures since 2010,
as shown in Figure 4-3. Reported federal expenditures in FY 2013 were only
47 percent of FY 2010 values, and Florida’s spending on restoration in FY 2013
was only 41 percent of its FY 2010 numbers. If planned expenditures for FY 2014
are realized, they would represent a slight upturn from prior years at both the
federal and state levels. Among the larger items included in federal non-CERP
expenditures over the 5-year period FY 2010 to FY 2014 are $470 million by the
Natural Resources Conservation Service for agricultural programs; $149 million
by the National Park Service for Everglades National Park management; and over
$100 million for Kissimmee River Restoration.
FIGURE 4-3 Federal and state spending related to South Florida ecosystem restoration
activities, including CERP and non-CERP projects and related expenditures.
NOTE:* Requested.
SOURCE: SFERTF (2014).

Of particular concern, South Florida ecosystem restoration is losing ground

in USACE Civil Works construction budgets. As shown in Table 4-2, funding
requested by the USACE for South Florida ecosystem restoration is about 48 per-
cent of what it was 3 years earlier. The Corps’ construction budget has declined,
and the South Florida ecosystem restoration share of that budget is declining.
SFWMD Financial Resources
The SFWMD’s capacity to undertake new CERP projects and possibly com-
plete authorized projects is hampered by its obligations to fund the Restoration
Strategies, an $880 million project mandated under a consent order to remedy
phosphorus exceedances in the Everglades Protection Area (see Chapter 2; also
discussed later in this chapter). Expenditures of that magnitude are in competi-
tion with financial demands for the SFWMD’s ongoing operations, for other
non-CERP construction projects, and for completion of previously authorized
CERP projects.
If revenues are not enhanced, the SFWMD will be challenged to keep up
cost sharing as the nonfederal partner for CERP and some non-CERP p rojects and
to continue funding other restoration projects and programs. Annual SFWMD
revenues dropped sharply over the period FY 2008 through FY 2013 as shown in
Figure 4-4 from just over $900 million to $346 million. The largest portion of the
loss in revenue was due to a decline in ad valorem taxes, down from $549 million
in FY 2008 to $268 million in FY 2013 (Figure 4-4). Intergovernmental revenue
(state appropriations and trust fund transfers) also declined, from $286 million
in FY 2008 to $37 million in FY 2013.
Furthermore, in recent years, revenues have been supplemented with signifi-
cant funding from reserve fund balances: $228 million in FY 2013 and $299 mil-
lion in FY 2014 (SFWMD, 2013c, 2014). The SFWMD had accrued large fund
balances (approaching $900 million in FY 2009). Those funds were being held
in reserve to address future spending needs, and since 2009, the SFWMD has
TABLE 4-2 Budget Requests for the U.S. Corps of Engineers Civil Works Program
Requested Budgets FY 2011 FY 2012 FY 2013 FY 2014
All construction, $ million 1,690 1,480 1,471 1,350
South Florida Ecosystem Restoration, $ million 180 163 153 88
Percent of total 10.7 11.0 10.3 6.5
Note: These totals are different from appropriated budgets.
SOURCE: Data from USACE (2010, 2011b, 2012a, 2013c).

FIGURE 4-4 SFWMD revenues for FY 2008 through FY 2013. In addition to revenues, the
SFWMD drew upon fund balances, which are not reflected here.
SOURCES: SFWMD (2008, 2009b, 2010, 2011b, 2012a, 2013e).
been drawing on this balance to cover capital outlay and other expenses. By the
end of FY 2013, the fund balance was down to $448 million, and, as shown in
Figure 4-5, the FY 2014 budget projects that by FY 2018, the fund balance will
be only $37 million. Thus, unless other sources of revenue increase in the next
few years, there will be very limited funds available for new projects.
As SFWMD revenues have declined, so have expenditures, as shown in
Figure 4-6. Capital outlay was at $379 million in 2008, but by 2012 had been
reduced to only $67 million. In FY 2013, capital outlay decreased to $48 million,
with $21 million going to the CERP. The FY 2014 to FY 2018 Capital Improve-
ments Plan is based on a 5-year projection of $1,146 million in revenues to
fund items included in the plan; $770 million or 67 percent would be spent for
Everglades Restoration. Of this, $171 million is programmed for Generation 1
projects (Picayune Strand and Indian River Lagoon-South’s C-44 Reservoir and
stormwater treatment area [STA]) and $348 million for the Restoration Strate-
gies project and other water quality-related non-CERP projects. An additional
$30 million is budgeted for the Loxahatchee Watershed project (Generation 3),
and $176 million is for debt service (Table 4-3; Heater and Maytok, 2014).

FIGURE 4-5 Drawdown of SFWMD reserve fund balances.
SOURCE: SFWMD (2013b).
FIGURE 4-6 Overall SFWMD expenditures, including but not limited to South Florida eco
system restoration, FY 2008 through FY 2012.
SOURCES: SFWMD (2008, 2009b, 2010, 2011b, 2012a, 2013e).

TABLE 4-3 FY 2014 to FY 2017 SFWMD Capital Improvements Plan Expenditures

for Everglades Restoration
EXPENDITURES Total
East Flowway $ 2,000,000$
C-44 Reservoir and STA 156,439,234
Picayune Strand 14,601,218
C-111 South Contract 9,000,000
LTP EAA STA Compartment B Design Construction & Build Outs 1,345,536
Loxahatchee Watershed 29,812,479
MECCA FEB 19,836,468
A-1 FEB 67,824,964
STA-1W Expansion #1 161,793,068
STA-1W Expansion #2 21,554,370
L-8 Divide 5,042,000
G-716 5,309,000
S-5AS 3,032,000
Source Control 18,066,834
Science Plan 44,342,154
Debt Service - 2006 COPs 175,802,909
Future Restoration Projects (SOETF) 34,135,727
TOTAL 769,937,961
SOURCE: Heater and Maytok (2014).
For FY 2014, the SFWMD budgeted approximately $45 million for authorized
CERP projects, including $29 million for the C-44 Reservoir and $16 million for
Picayune Strand (SFWMD, 2014b).
Cost-Sharing Implications
Effects of 50-50 state-federal cost-sharing rules as outlined in WRDA 2000

and detailed in the 2009 Master Agreement have become a matter of significant
concern. Management of expenditures to satisfy cost-sharing rules has always
been a challenge, but it has become more challenging over the past 2 years
with reduced budgets, mandated expenditures for other projects, and limited
project authorizations. Even though the SFWMD has far outspent federal agen-
cies on CERP projects from land acquisition and expedited construction efforts
(Figure 4-2), cost-sharing rules dictate that the SFWMD can only apply its store-
house of potential credits toward the 50-50 cost-sharing requirement on those
projects that have been authorized by Congress, have signed project partnership
agreements, and have received federal appropriations. Non-planning-related
SFWMD expenditures on yet-to-be authorized CERP projects cannot be offi-

cially credited toward the 50-50 cost-sharing requirement until those projects
are authorized and funds have been appropriated. Cost-sharing agreements
are restrictive with respect to how credits can be used and how the federal-
nonfederal balance must be maintained. The 50 percent proportionate share
owed by the local sponsor (i.e., the SFWMD) must be brought into balance
annually across the CERP program. Several options are provided for eliminat-
ing an imbalance, including delaying construction schedules or requiring cash
payments from the local sponsor.4
As discussed previously, until late May 2014, only four projects had been
authorized. As of September 30, 2013, $1.88 billion had been spent on those
four projects: $891 million by the federal government and $989 million by the
SFWMD, leaving the state with only $98 million in excess creditable expendi-
tures above the minimum required 50-50 balance.
In the past few years, the CERP strategy to address this pending cost-sharing
issue has been to reduce federal spending where feasible, particularly for non-
construction-related activities. As of September 2013, the federal government
could spend no more than $49 million more than the state on the CERP through
September 2014 without necessitating additional expenditures by the state.
Recent federal Everglades restoration funding through the federal Consolidated
Appropriations Act of January 2014 was reduced to $46 million, down from prior
discussions of $80 million, because the USACE indicated it could not spend the
money (Scott, 2014) due to a combination of policy, authorization, and cost-
sharing issues (Tipple, USACE, personal communication, 2014).
WRRDA 2014, therefore, is an important achievement for Everglades resto
ration progress because it temporarily alleviates cost-sharing constraints that
have restricted federal spending over the past few years. The four Generation 2
project authorizations represent a critical step to enable the SFWMD to realize
an estimated $382 million in accumulated credits from its prior spending on
land acquisition and construction (G. Rogers, SFWMD personal communication,
2014), allowing federal appropriations to continue in the near future.
Project Scheduling and Prioritization
The CERP project construction schedule for the next decade is outlined in
the Integrated Delivery Schedule (IDS; Figure 4-7). The IDS was developed
in consultation with the Task Force and reflects the priorities of the CERP part-
ners as well as sequencing constraints and other project implementation issues.
Between 2008 and 2011, the IDS was revised several times per year (typically
See Master Agreement at http://www.evergladesplan.org/pm/pm_docs/master_agreement/081309_

4
master_agreement_cerp.pdf.

FIGURE 4-7 Integrated Delivery Schedule, August 2011 draft.

NOTE: Project costs cited represent October 2008 price levels and have been adjusted for inflation based
on construction start and finish dates for each contract.
SOURCE: K. Tippett, USACE, personal communication, 2011.
in draft) to reflect changing budgets and other developments that affect project
schedules, but with intense staff efforts devoted toward the development of
Central Everglades Planning Project, the IDS has not been revised since August
2011 draft (Figure 4-7). The IDS now is badly out of date.
Several factors will affect the next update of the IDS, including state and
federal CERP and non-CERP appropriations and the recent congressional autho-
rizations for the Generation 2 projects (Table 4-1). Much uncertainty remains
over how new authorizations and lack of authorization for the Central Everglades
Planning Project will affect the overall implementation schedule. For example,

should early authorized projects (Generation 1) be funded before later autho-

rized projects (Generation 2)? What is the priority of non-CERP projects, such
as the Modified Water Deliveries, Kissimmee River Restoration, C-111-South
Dade, or Tamiami Trail Next Steps projects, relative to the CERP Generation 1
and 2 projects? How can the Central Everglades Planning Project be expedited in
the schedule? Some planners have expressed a strong desire to finish what was
started in Generation 1 projects to show progress to the public and Congress.
However, expediting key non-CERP projects could provide large benefits and
complete important project dependencies for the Central Everglades Planning
Project, which was specifically intended to halt ecosystem declines and expedite
improvements in the condition of the remnant Everglades. Now that the four
Generation 2 projects are authorized in WRRDA 2014, they will be competing
for extremely limited funding under current fiscal pressures. With recent fed-
eral CERP expenditures of approximately $80 million/year (see Figure 4-2) and
projected state spending of approximately $40 million/year on authorized CERP
projects, the pace of progress on these projects will be slow if available funding
is spread equally across all projects (see Table 4-4).
The current budget situation necessitates clear priorities for spending. As
the state works to complete the Restoration Strategies project to address current
water quality issues, federal funds and available state funds should be targeted
toward CERP and non-CERP projects that will quickly avert current ecosystem
declines and/or promise the largest potential restoration benefits, considering
future climate change and sea-level-rise impacts. Until this point, CERP prioriti-
zation efforts have not explicitly considered the ecosystem condition and future
benefits. Previously, with more-plentiful construction funding and few autho-
rized projects, such difficult decisions were not necessary. But today’s grim bud-
getary landscape requires a new CERP prioritization framework to avert ongoing
ecosystem degradation and make the best use of currently available funding.
RESTORATION PROGRESS
In the following section the committee focuses on benefits emerging from

the implementation of CERP restoration projects as well as from non-CERP
foundation projects. The committee’s previous report (NRC, 2012a) contains
more extensive descriptions of the projects, and progress up to March 2012,
while only progress over the last 2 years is described here. The South Florida
Environmental Report (SFWMD, 2013a) and the 2012 Integrated Financial Plan
(SFERTF, 2012) also provide detailed information about implementation and
restoration progress. The 2014 System Status Report (RECOVER, 2014a; see
Chapter 7) provides additional information on changing ecosystem conditions
and discusses linkages to early project construction.

TABLE 4-4 Status of Project Expenditures and Estimated Funding Needs (in Millions) for
Generations 1, 2, and 3 and Key Non-CERP Projects as of December 2013
Estimated
Funding
Total State Federal Remaining
Estimated Funding Funding for
Project Name Costs to Date to Date Completion
NON-CERP FOUNDATION PROJECTS
Kissimmee River Restoration Project 780 345 315 120
C-111 South Dade Project 290 115 120 55
Modified Water Deliveries to Everglades National Park Project a 418 0.2 399 19
Restoration Strategies a
880 60 0 820
Non-CERP totals 2,368 520 834 1,014
1st GENERATION CERP PROJECTS

Picayune Strand Restoration 618 162 317 139
Indian River Lagoon-South (Phase 1) 1,450 365 55 1,030
Site 1 Impoundment (Phase 1) 85 8 68 9
Melaleuca Eradication Project and Other Exotic Plants 25b 0.212 4.3 20.5b
2nd GENERATION CERP PROJECTS
C-43 West Basin Storage Reservoir 570 90 35 445
Broward County WPAs 870 260 60 550
C-111 West Spreader Canal 85 64 12 9
Biscayne Bay Coastal Wetlands (Phase 1) 125 42 12 71
3rd GENERATION CERP PROJECTS
Central Everglades Planning Project 1,750 4 31 1,715
Loxahatchee River Watershed Restoration TBD 163 5 TBD
CERP Totals 5,578+ 1,158 599 3,989+
NOTES:
a
Not a 50-50 cost-shared project.
b
Includes operations and maintenance funding for mass rearing, release, and field monitoring of biocontrol agents.
SOURCE: T. Morgan, SFWMD, and H. Gonzales, USACE, personal communication, 2013; M. Collis, USACE, personal com-
munication, 2014; R. Johnson, DOI, personal communication, 2014; L. Gerry, SFWMD, personal communication, 2014.
Reportable restoration progress as a result of CERP and non-CERP projects

occurs in three sequential steps: completion of project construction, physical
system response (e.g., as a return of more natural hydrologic conditions or sheet
flow), and changes in the biological system, including changes to individual spe-
cies or ecosystem components. The physical and ecological responses are often

assessed by monitoring-project-specific and systemwide performance measures

(see RECOVER, 2007) and comparing the results to target values.
Construction for smaller projects takes place in a single phase, whereas more-
complex projects have multiple phases, each of which can be assessed in terms
of progress. Once construction is complete, most projects have their first effects
by introducing new quantities of water, direction of flow, hydroperiods that are
longer or shorter, or adjusted rates of change. The return of freshwater to some
landscapes and saltwater to others may be an objective, and the return of sheet
flow instead of highly concentrated confined flows is often essential to encour-
aging recovery of Everglades landscapes. Observations of restored hydrologic
conditions are therefore the initial indicator of success for restoration projects.
The return of hydrologic conditions more similar to pre-drainage condi-
tions sets the stage for the critical next step in observable restoration whereby
plant and animal communities develop that are more similar to pre-drainage
communities and less like those disrupted by water control infrastructure. After
many months to several years, these changes can become a matter of quantita-
tive record if there is routine monitoring using formal performance parameters.
Continuous monitoring in the post-project period is essential in judging the suc-
cess of the project, and the period of monitoring must be long enough to discern
whether changes in performance measures are linked to the project rather than
to other influences or to normal variability.
Most projects require several years to make the transition from the beginning
of construction to the quantitative observation of desired ecological changes.
However, this time frame, often of a few years to a decade, is fast relative to
the several decades of disruption that have created the altered ecosystem that
is the subject of Everglades restoration.
In the following sections, restoration progress is highlighted in four Genera-
tion 1, projects, two Generation 2 projects, and one Generation 3 project, for
which construction has begun. Progress in three pilot projects and three non-
CERP foundation projects is also discussed. The analysis that follows covers
only those projects with substantial new developments or information on natural
system restoration progress since the committee’s last report (NRC, 2012a). A
summary of all implementation progress as of May 2014, including develop-
ments in planning and authorization, is provided in Table 4-1. The location of
the various projects is shown in Figure 4-1.
Generation 1 CERP Projects
Generation 1 projects are those authorized by Congress in WRDA 2007

(Picayune Strand Restoration, Site 1 Impoundment, and Indian River Lagoon-
South) or by program authority (Melaleuca Eradication). Until June 2014, these

were the only projects eligible for federal funding, while other projects awaited
authorization.
Picayune Strand
The Picayune Strand, the first CERP project under construction, focuses on
an area in southwest Florida substantially disrupted by a real estate development
project that introduced 260 miles of roadway and 48 miles of major canals and
drained 55,000 acres of wetlands before being abandoned (Figure 4-1, No. 5).
The roads and drainage disrupted sheet flow into Ten Thousand Islands National
Wildlife Refuge and altered regional groundwater flows in surrounding natural
areas. The reduction of freshwater wetlands adversely affected habitat. These
disruptions were especially important because Picayune Strand is surrounded by
and contiguous with several other protected areas. Picayune Strand is particu-
larly important as a habitat for white-tailed deer and feral hogs that serve as prey
for the endangered Florida panther (USACE, 2011d; but see also Chapter 6 for
a discussion of feral hogs as a troublesome invasive species). As of May 2014,
the total anticipated cost of the project was $618 million (M. Collis, USACE,
personal communication, 2014).
There has been considerable progress in constructing the Picayune Strand
Restoration Project, with approximately $480 million expended to date on the
project by the state and federal governments. The project components are sum-
marized in Table 4-5 and the progress made to date in canal plugging, road
removal, construction of pump stations, and other project elements (Figure 4-8).
In late summer 2014, the USACE anticipates completing the Merritt Canal phase,
which included construction of a pump station with associated spreader canal
and levees, the plugging of 9 miles of canals, and removal of 95 miles of roads.
Construction is currently under way on the Faka Union and Miller Canal project
portions (see Table 4-5) (USACE, 2014c).
Despite this important progress, the Picayune Strand Restoration faces the
prospect of halting construction because the costs of the project are expected
to exceed the Section 902 limits.5 Thus, the USACE is required to submit a
limited reevaluation report to Congress for authorization (T. Morgan, SFWMD,
personal communication, 2013). The timeliness of congressional reauthorization
will determine when the project can be completed. There are no scientific or
engineering impediments to project completion.
5
WRDA 2007 authorized the project with a budget of $375.3 million, and Section 902 of WRDA
1986 requires that projects seek reauthorization if costs increase more than 20 percent above the
original authorized costs (exclusive of inflation).

TABLE 4-5 Phases and Progress of the Picayune Strand Project

Road Canals to
Lead Removal Be Plugged
Agency (mi) (mi) Other Project Phase Status
Prairie Canal State 65 7 Invasive vegetation Completed in 2007
expedited removed, 17 culverts
project constructed, >13,000 acres
of habitat enhanced
Merritt Canal Federal 95 9 Remove invasive Construction began
vegetation; construct in 2010; anticipated
Merritt pump station, completion in summer
~14,000 acres of enhanced 2014
habitat
Faka Union Federal 100 0 Construct Faka Union Construction began
Canal pump station in 2011; anticipated
completion in 2015
Miller Canal Federal 47 13 Construct Miller Canal Construction began in late
pump station and 2013; to be completed in
spreader canal 2017
Southwestern State 0 0 Construct 9-mile levee Construction to begin
protection for flood protection in 2016, pending
and manatee of adjacent lands and authorization of increased
mitigation excavate manatee project cost
features refugium to mitigate loss
Canal State 86 16.5 Remove roads north of Construction to begin
plugging and tie-back levee, plug Faka in 2017; canal plugging
road removal Union and east-west must follow completion of
canals manatee mitigation and
southwestern protection
features
SOURCE: L. Gerry, SFWMD, personal communication, 2013, 2014; USACE (2014c); M. Collis, USACE, personal communica-
tion, 2014.
Because one phase of the Picayune Strand Restoration Project (Prairie Canal)
was completed in 2007, there has been sufficient time to collect and assess envi-
ronmental data to determine what natural system benefits associated with the
completed work have been observed. In a general observation, USACE (2014c)
reports a resurgence of foraging wading birds (Figure 4-9) and native flora that
have been absent for many decades. In areas where canals have been plugged,
roads removed, and invasive plant species removed, freshwater wetlands have
returned (Figures 4-9 and 4-10). Researchers have also reported detailed—but
preliminary—results from multiyear monitoring efforts (Box 4-1). The most recent
data are for late 2011, with the assessments completed in late 2013 focused

FIGURE 4-8 Picayune Strand Restoration features.
SOURCE: L. Gerry, SFWMD, personal communication, 2014.
on the Prairie Canal area, which had 4 years of post-project monitoring data
(USACE, 2014c; RECOVER, 2014b).
These monitoring data show that the restoration of the Prairie Canal area
is having clear positive effects on the area by adjusting water levels and hydro-
periods toward pre-drainage conditions, although neighboring canals not yet
filled by the project continue to affect the area hydrology. The data also show,
however, that some biological components of the restored system are not yet
experiencing major changes. There are two likely reasons. First, more time may
be needed for the complex fauna and flora communities to adjust to new steady-
state conditions. Everglades ecosystems may have a reaction time during which
changes in controlling factors must work their way through the complexities of

FIGURE 4-9 Freshwater wetland in the Picayune Strand restoration area showing wading birds.
SOURCE: USACE (2014c).
FIGURE 4-10 Freshwater wetland in an area that was once a canal in the Picayune Strand restoration area.
SOURCE: USACE (2014c).

BOX 4-1
Analysis of Monitoring Results from 2007 to 2011
at Picayune Strand
The following outlines the natural system responses to the Prairie Canal features of
the CERP Picayune Strand project:
Water levels: Water-level data from observation wells measured from 1987 to 2012
demonstrate the effects of plugging Prairie Canal. Although water-level fluctuations in
the area have a complex history related to the management of the Picayune Strand
area, prior to the Prairie Canal component of the restoration, water levels were 5-6 feet
lower than the reference area in Fakahatchee Strand Preserve State Park. In the period
after the completion of the Prairie Canal component of the restoration project, water
levels have become similar to a relatively undisturbed reference area during the wet
period of the year, but are still about 3.5 feet lower than the reference during dry periods.
This shortcoming is probably related to the fact that the canal immediately to the west
of the Prairie Canal is not yet fully plugged, so it affects its neighbor’s water levels.
Hydroperiods: The number of days per year that water is at or above ground level
(called the hydroperiod), has lengthened in the restored area despite droughts in 2007
and 2012.
Plant communities: Comparisons between 25 vegetation transects in the Prairie Canal

restoration area with 11 transects in a relatively undisturbed reference area show that
the restoration has so far resulted in little change in dominant tree species. In the re-
stored area, cypress and pop ash growth rates are slower, and pine growth rates are
faster than in reference areas. There have been no changes in densities of cabbage
palm, considered a nuisance species in the Picayune area. Cypress plots in the resto
ration area also are developing plant assemblages that increasingly include wetland
species, showing that the hydrologic restoration is supporting a more wetland-like
community.
Aquatic macroinvertebrates: The restored areas had enough water to support aquatic
macroinvertebrates, and 45 percent of the aquatic species in the restored areas were
also found in the reference areas.
Fishes: Fish populations were generally similar between restored and reference sites,
except that two indicator species, Everglades pygmy sunfish (Elassoma evergladei) and
warmouth (Lepomis gulosus), were found only at reference sites.
Tree frogs: Indigenous species such as green and squirrel tree frogs were dominant at
reference sites, whereas the exotic Cuban tree frogs were dominant in restoration areas.
Water quality: Decades of water quality data are available, and they reveal no areas of
concern. Post-restoration data reflect no change and no water quality decline.
SOURCE: RECOVER (2014b).

the system which therefore does not respond immediately. Second, the restora-
tion of the Prairie Canal area is still affected by its neighboring Merritt Canal
area that was not yet restored during the 2007-2011 analysis period.
Site 1 Impoundment
The Site 1 impoundment project (also known as the Fran Reich Preserve;
No. 6 on Figure 4-1) is in Palm Beach County at the junction of the southern
tip of the Loxahatchee National Wildlife Refuge (LNWR, also known as Water
Conservation Area 1 [WCA-1]) with the Hillsboro Canal. The project was
originally cast as a single-phase effort to modify local hydrologic conditions
so that more water could be stored to alleviate demands on water in LNWR.
Without the project, during wet periods, runoff from LNWR is shunted to the
ocean, while during dry periods, water is taken from the refuge to meet user
demands elsewhere. With the Site 1 impoundment, water can be better man-
aged to s upply natural system demands within the LNWR. Project components
included construction of a reservoir to store 13,300 acre-feet (AF) of water, a
pump station, gated discharge culvert, spillway, and seepage control canal to
retain more flows within the LNWR (USACE, 2013g; Figure 4-11). In 2009 the
project was divided into two phases.
Construction of Phase 1 is under way, including the L-40 levee enhance-
ments and seepage management measures, and is scheduled for completion by
2015 (L. Gerry, SFWMD, personal communication, 2014). Once these features
are completed, designers anticipate a reduction in seepage loss from LNWR, but
no natural system benefits can be reported at this time. Phase 2 of the p
roject
awaits congressional reauthorization necessitated by increased costs (NRC,
2012a).
Indian River Lagoon-South
The Indian River Lagoon and St. Lucie Estuary are biologically diverse
e stuaries located on the east side of the Florida Peninsula, whose ecosystems
have been altered by polluted runoff from farmlands and urban areas and
surges of freshwater (USACE, 2013d). These changes have reduced the abun-
dances of many native species. The Indian River Lagoon-South project (Figure 4-1,
No. 7) is designed to reverse this damage through improved water management,
including the 56,000-AF C-44 storage reservoir, three additional reservoirs with
a total of 97,000 AF of storage, four new stormwater treatment areas (STAs),
dredging of the St. Lucie River to remove 7.9 million cubic yards of muck, and
restoring 53,000 acres of wetlands, among other features (Figure 4-12). The
project is anticipated to cost $1.45 billion (see Table 4-5).

94
FIGURE 4-11 Phase 1 and 2 project elements of the Site 1 impoundment.
SOURCE: USACE (2013g).

FIGURE 4-12 Components of the Indian River Lagoon-South restoration project.
SOURCE: USACE (2013d).

The Indian River Lagoon restoration project is just beginning. Preliminary

work for the C-44 reservoir and associated 6,300-acre STA is under way, includ-
ing an intake canal, access roads, culvert removal, and other improvements that
are anticipated to be completed in summer 2014 (L. Gerry, SFWMD, personal
communication, 2014). C-44 reservoir and STA construction is expected to
begin in 2015, with an estimated completion date of 2020 for the C-44 reservoir
and 2018 for the STA (H. Gonzales, USACE, personal communication, 2014).
Additionally, operational changes have been made using existing water control
structures to rehydrate the Allapattah Natural Storage Area (L. Gerry, SFWMD,
personal communication, 2013). However, it is too soon to expect substantial
restoration from this project based on the work completed to date.
Melaleuca Eradication and Other Exotic Plants
The Melaleuca Eradication and Other Exotic Plants Project is a CERP effort
to address the potential threat to restoration posed by nonnative invasive plant
species (see Chapter 6). Four invasive species that are particularly problematic
are the focus of major ongoing management efforts: Melaleuca ( Melaleuca
quinquenervia), Brazilian pepper (Schinus terebinthifolius), Australian pine
(Casuarina spp.), and Old World climbing fern (Lygodium microphyllum). A
crucial part of this work is centered at the U.S. Department of Agriculture’s
Invasive Plant Research Laboratory in Davie, Florida, where specific biological
control agents—mostly insects—are developed. With CERP funds, USDA has
constructed a 2,700-ft2 annex to the present laboratory to facilitate additional
mass rearing (Figure 4-1, No. 8). The $4.5 million annex was completed in
August 2013 and has been transferred to the local sponsor (USACE, 2014b;
T. Morgan, SFWMD, personal communication, 2013). The project includes
CERP operations and maintenance funding (estimated at $660,000/year) for
mass rearing, release, and field monitoring of biocontrol agents to manage the
spread of invasive nonnative plant species in the Everglades and South Florida
(USACE and SFWMD, 2010). This project enables a more aggressive approach
to biological control of invasive plants in the Everglades restoration area. It is
too soon to document specific natural system benefits resulting from this CERP
investment, although the expanded facility has increased the rearing capacity
by 3- to 10-fold (P. Tipping, USDA, personal communication, 2014). Invasive
species control in the Everglades is discussed in more detail in Chapter 6.
Generation 2 CERP Projects
Four second-generation CERP projects were authorized in June 2014 as part

of WRRDA 2014 (Table 4-1). Although they had received no federal funding for

construction at the time of authoring this report, in two cases—Biscayne Bay

Coastal Wetlands and the C-111 Spreader Canal projects—the State of Florida
has expedited construction in advance of federal authorization.
Biscayne Bay Coastal Wetlands
The Biscayne Bay Coastal Wetlands are along the southeastern edge of the
Florida Peninsula; the Miami-Dade County area at the western edge of Biscayne
Bay is a unit of the national park system (Figure 4-1, No. 10, and Figure 4-13).
The installation of canals has cut off a section of the wetlands from their source
of freshwater sheet flows resulting in a loss of wetland ecosystems and causing
an increase in salinity along the margin of the bay. The project seeks to reverse
these effects on 11,300 acres of the total 22,500 acres of wetlands by installing
pump stations, spreader canals, culverts, and canal plugs.
The project is in two phases: Phase 1 is a stand-alone project encompass-
ing three geographic areas (Deering Estates Flowway, Cutler Wetlands, and
L-31 Flowway), and Phase 2, which is not yet specifically planned. The three
components of the Deering Estates Flowway—a spur canal extension, spreader
canal, and pump station—were completed and became operational in Decem-
ber 2012; the two culverts in the L-31E canal designed to divert flows into
coastal wetlands were finished in June 2010. The work on Cutler Wetlands has
not yet begun. To date, $54 million has been expended on the $125 million
Phase 1 project (T. Morgan, SFWMD, personal communication, 2013), although
the project increments implemented so far have been rather small in the context
of the original project objectives.
The Biscayne Bay Coastal Wetlands Project has begun to show some eco
system restoration results (Figure 4-14). Deering Estates Flowway, which includes
170 acres of degraded freshwater wetlands that transition to tidal wetland (USACE
and SFWMD, 2012a), has been in operation for only a few months, so data are
preliminary and may show more about short-term variability than long-term
trends. Since the completion of the Deering Estates work, water flows have
elevated seasonable water surfaces by about 2.7 feet in wetland areas. There have
been some observed improvements of reduced salinity in the coastal wetlands
and in the establishment of more natural salinity gradients (from freshwater in the
wetlands to saline water in the bay). Although salinity of surface water in the near-
shore responds to freshwater infusions from the new S-700 pump s tation, there
are many other potential control mechanisms on salinity that must be sorted out
with the aid of longer monitoring records. Vegetation communities were surveyed
in 2009, and will be resurveyed in 2015 (RECOVER, 2014a). The L-31E culverts
have diverted approximately 3 percent of the canal flow in water years 2012
and 2013 into the adjacent coastal wetlands. Routine maintenance inspections

FIGURE 4-13 Biscayne Bay Coastal Wetlands project area in southeast Florida.
SOURCE: USACE (2014a).

FIGURE 4-14 Pump house and planted wetland vegetation at Deering Estate Flowway, Biscayne Bay Coastal
Wetlands. The hydrated surface with wetland vegetation replaces a dry surface without wetland vegetation.
SOURCE: Courtesy of the Deering Estate at Cutler and Brian F. Call Photography.
established that accumulation of aquatic vegetation, sediment, and silt hindered

the performance of the L-31E culverts. The SFWMD removed accumulated down-
stream sediment, realigned downstream sumps and pipe inverts, and installed
floating debris barriers across the mouth of each culvert inlet channel. This task
was completed in August 2012 and has shown local improvements to the system.
Monitoring data for nearshore salinity show no trends related to comple-
tion of the L-31E feature. Vegetation communities are, however, showing some
adjustment in the post-project period with a minor decrease in areal coverage
of red mangrove overall and an increase in sawgrass on the estuarine side of the
L-31E levee (RECOVER, 2014a). These minor or negligible results are consistent
with the low volume of diverted flows in 2012-2013.
C-111 Spreader Canal
The C-111 canal (Figure 4-1, No. 9) is the southernmost canal for the entire
Central and Southern Florida Project. Designed to provide drainage and an outlet
for confined flood flows, the C-111 also eliminated sheet flow from the Southern
Glades and drained water from Taylor Slough in Everglades National Park. The
C-111 Spreader Canal project promises increased flow volumes in Taylor Slough
through seepage control, return of sheet flow to wetlands, and improved salinity
regimes in western Florida Bay. The project complements the ongoing C-111
South Dade Project (see Box 2-1), which has related project objectives. The

C-111 Spreader Canal project is structured in two phases, with the first phase
(Western Project) to include two pumping stations, a 560-acre detention basin
(the Frog Pond), along with various canal modifications for the Aerojet, L-31, and
C-110 canals (Figure 4-15). The C-111 Spreader Canal Western Project is critical
to the restoration of Taylor Slough in Everglades National Park and Florida Bay
and was largely completed in February 2012 as described in NRC (2012a) and
USACE (2013b) and is now operational.
Some quantitative assessments of project effects on natural system hydrology
are available. For example, during the wet season of its first year in operation,
from June through mid-November, the project moved an average of 811 AF
per day (a total of about 140,000 AF) from the C-111 canal and transferred it
FIGURE 4-15 Project design features for C-111 Spreader Canal Western Project.
SOURCE: SFWMD (2013a).

to the Taylor Slough area of Everglades National Park (Audubon Florida, 2014).
Taylor Slough now carries more water than prior to the completion of the C-111
Spreader Canal project. Taylor Slough now is wet 10 months out of the year
rather than 3 months per year before the restoration efforts (Fleshler, 2014).
Recent research published in the refereed literature establishes some hydro-
logic improvements from the C-111 project. Kotun and Renshaw (2013) exam-
ined hydrologic data for the C-111 canal and for Taylor Slough and found that
specific improvements were observed in the Taylor Slough hydrology between
2000 and 2010 when C-111 South Dade projects were in operation. Water sur-
face level was maintained in the canal at heights above the surrounding terrain
for 6 months each year, contributing to a hydrologic seepage barrier effect. Their
data showed that a groundwater mound developed under the canal as expected,
an approach to keep water in the slough and prevent its migration eastward.
They also found that the hydroperiod in the Rocky Glades near Taylor Slough
had been lengthened 90 days.
Ecosystem benefits are difficult to assign to a single restoration project. An
important complication in interpreting the changes is that it is difficult to separate
and identify benefits from the C-111 Spreader Canal project from the benefits
of the closely related C-111 South Dade project. The C-111 South Dade deten-
tion areas, also designed to rehydrate Taylor Slough, were completed in 2009,
and improvements in flows in Taylor Slough are likely to be connected to that
project. Benefits from the ongoing C-111 Spreader Canal project are likely to be
smaller but result in some additional flow in Taylor Slough (R. Johnson, NPS,
personal communication, 2014).
Generation 3 Projects
Third-generation projects are near-term priorities, but they do not yet have
a USACE-headquarters-approved PIR. Until congressionally authorized, imple-
mentation of these projects could only occur if expedited with state fund-
ing. Restoration efforts are under way on one third-generation CERP project—
Loxahatchee River Watershed Restoration.
Loxahatchee River Watershed Restoration
The Loxahatchee River Watershed Restoration Project is a CERP project

that has been expedited by SFWMD investment (Figure 4-1, No. 15; USACE
and SFWMD, 2013c). The purpose of the project, located in the southern head-
waters of the Loxahatchee River and north of LNWR (WCA-1), is to rehydrate
several thousand acres of wetland habitat that has been desiccated by artificial
drainage, provide restoration flows to the Northwest Fork Loxahatchee River,

and address saltwater intrusion. Although development of the PIR is still ongo-
ing, the SFWMD has expedited installation of culverts and control structures
leading to Loxahatchee Slough (Flowway 1) to raise water levels and lengthen
periods of inundation, measures that affect about 5,000 acres (Figure 4-16). As
of 2013, Martin County, a local sponsor partner, was implementing invasive spe-
cies control efforts and construction of water control structures in the Cypress
Creek (Flowway 3).6 Other partners in addition to SFWMD include Palm Beach
County, the Loxahatchee River District, Florida Department of Environmental
Protection, and the Florida Park Service. The river is a federally designated
Wild and Scenic River, one of only two in Florida. Restoration related to the
Loxahatchee River Watershed includes attempts to improve downstream areas
all the way to the ocean, including the construction of 5.8 acres of new oyster
habitat and refined operations of control structure S-46. Since 2011, releases
of water to the northwest fork of the Loxahatchee River have reduced periods of
low flows, attenuated high flows in the southwest fork, and reduced damaging
high variability in salinity at the river mouth. The initial restoration steps under
the Loxahatchee River Watershed CERP Project have not yet led to significant
effects on the overall watershed behavior (Loxahatchee River District, 2013).
CERP Pilot Projects
Pilot projects are limited efforts designed to provide scientific or engineering

knowledge that can be applied to improve major projects that result in natural
system benefits. Additionally, pilot projects may inform larger projects to make
them more timely and cost-effective. Pilot projects provide the opportunity to
experiment with methods and approaches without the large expense of fully
developed restoration projects. Below, we briefly review the Aquifer Storage and
Recovery, Decomp Physical Model, and Seepage Management pilot projects.
Aquifer Storage and Recovery
Everglades restoration relies on increasing freshwater storage. In the CERP,

aquifer storage and recovery (ASR) was proposed as the largest contributor to
new storage, with more than 300 ASR wells providing up to 1.7 billion gallons
of freshwater storage per day (or 500,000 AF/yr; USACE and SFWMD, 1999).
Implementation of ASR in the Everglades would involve pumping excess surface
water into the Floridan Aquifer system and recovering this stored water during
dry periods to sustain freshwater flows (Figure 4-17). ASR is an established
See http://xportal.sfwmd.gov/paa_dad/docs/F1077773745/32_Loxahatchee%20Storage%20-%20
6
Barnett.pdf.

FIGURE 4-16 Restoration progress in the Loxahatchee Watershed from 2002 (top) to 2013 (bottom), show-
ing the transition from upland species to more wetland species due to increased periods of inundation.
SOURCE: Palm Beach County Environmental Resources Management.
technology, but it had not been widely tested in South Florida. Therefore, the
CERP ASR Pilot Project was conducted to address uncertainties and concerns
regarding the efficacy and feasibility of ASR, particularly given the high costs of
its implementation at the scale proposed by the CERP. The final report summa-
rizing the findings of this 14-year pilot project was released in December 2013
(USACE and SFWMD, 2013a).
The CERP ASR Pilot Project focused on construction, operation, and moni-
toring of two facilities. The Kissimmee River ASR was built along the K issimmee
River, near its confluence with Lake Okeechobee, while the Hillsboro ASR
was constructed along the Hillsboro Canal in southern Palm Beach County
(Figure 4-1). The objectives of these pilot studies were to (1) assess ASR feasibility
at two locations distinguished on the basis of surface-water chemistry, hydro-
geologic conditions, and surface-water distribution configurations; (2) evaluate
technical and regulatory compliance issues stemming from ASR operation; and
(3) quantify operation costs. The final technical data report describes, in con-
siderable detail, the various elements of these pilots, including planning and

FIGURE 4-17 Schematic diagram of the recharge and recovery phases of ASR for a typical South Florida
system.
SOURCE: NRC (2001).
permitting, system design and construction, groundwater hydrogeology, regula-

tory compliance, surface-water and groundwater quality testing, and ASR system
costs (USACE and SFWMD, 2013c).
Both ASR pilot systems were operated for multiple recharge-storage-recovery
cycles, with each cycle increasing in length and thus volume of water stored.
More than 1 billion gallons of water were stored for a 1-year period during the
fourth cycle test of the Kissimmee River ASR pilot, making this test one of
the largest single-well recharge events conducted in Florida.
Findings from the pilot projects are generally encouraging from the stand-
point of operational efficacy, and most problems that emerged during cycle
testing were resolved. Recoveries of stored freshwater were approximately
100 percent for the four cycle tests at Kissimmee River, but were lower (21 to
85 percent) during the three cycle tests of the Hillsboro ASR pilot, owing to mix-

ing with brackish formation water. Water quality analyses revealed that the
recharge waters at both facilities were in compliance with respect to all primary
constituents except total coliforms. Total coliforms were detected in both the
Kissimmee River and Hillsboro ASR wells in excess of the 4-CFU/100 mL crite-
rion, suggesting that the ultraviolet disinfection systems used to treat the surface
water prior to injection were insufficiently effective. Analysis of groundwater
collected from storage-zone monitoring wells indicated that arsenic concentra-
tions exceeded the 10-µg/L criterion at both facilities. Arsenic concentrations
peaked during the first cycle test, but, at both facilities, fell to permissible levels
prior to the onset of the next cycle test. Measurements of pressure made within
the ASR wells during freshwater recharge remained below 70 psi, which is less
than the pressure necessary to induce hydraulic fracturing.
The final construction costs differed significantly between the two ASR pilot
projects. The costs of the surface facility, ASR well, and monitoring wells at
Kissimmee River equaled $7.9 million, while the costs of these components
at the Hillsboro ASR facility equaled $4.3 million. Operational costs ranged from
$104 per million gallons of water during the recovery phase at the H illsboro
facility to $401 per million gallons of water during the recharge phase at the
Kissimmee River facility. Labor and electric power represented the greatest
portion of operational costs. The lessons learned through implementation of
these pilot projects should improve planning, design, and operational efficiency
which, in turn, should lower costs as more ASR systems are deployed.
Several issues deserving attention were identified through analysis of the ASR
pilots. For example, improved approaches must be adopted to disinfect recharge
water, and the supervisory control and data acquisition system requires further
refinement. Aside from these sorts of existing issues, additional challenges will
be encountered as new ASR facilities are constructed and operated on sites that
differ with respect to water quality, geological properties, and hydrologic con-
ditions. Nevertheless, these pilot studies have provided evidence that ASR is a
technically feasible approach for increasing freshwater storage in the Everglades,
and the report provides cost data that allow planners to compare ASR against
other available storage alternatives (NRC, 2005).
A report on the regional effects of multiple ASR wells in the Everglades is
anticipated in 2014. This regional-study report is likely to address several out-
standing issues that are needed to evaluate the feasibility and appropriateness of
large-scale implementation of ASR within South Florida. These issues include,
but are not limited to, overall costs for construction, operation, and monitoring;
systemwide energy demands associated with the conveyance and treatment of
large volumes of water; and the potential effects of large-scale ASR operations
on regional groundwater flow patterns, water supply, and water quality.

Decomp Physical Model
The Decomp Physical Model (DPM) is a large-scale field experiment

intended to inform project planning decisions by reducing uncertainty about
the ecological effects of various options for restoring sheet flow to the ridge-and-
slough landscape. The experiment involves measuring biophysical responses to
canal and levee modifications and is intended to address the following questions:
• Sheet flow questions: To what extent do entrainment, transport, and

settling of sediments differ in ridge-and-slough habitats under high- and low-
flow conditions? Does high flow cause changes in water chemistry and con-
sequently changes in sediment and periphyton metabolism and organic matter
decomposition?
• Canal backfill questions: Will canal backfill treatments act as sediment
traps, reducing overland transport of sediment? Will high flows entrain nutrient-
rich canal sediments and carry them into the water column downstream? To what
extent are these functions altered by the various canal backfill options, including
partial and full backfills?
The DPM experiment is being conducted between L-67A and L-67C, in

an area near the border of WCA-3A and WCA-3B known as the “the pocket”
(Figure 4-18). In preparation for the experiment, 10 gated culverts on the L-67A
canal (S-152, shown in Figure 4-18) were built. A 3,000-ft gap in the L-67C
levee and three backfill treatments in the adjacent canal were completed in
October 2013. The canal was left completely open for the northernmost treat-
ment, while the center and southernmost treatments have partial and complete
backfills, respectively.
A pulse-flow experiment was initiated on November 5, 2013, by opening
the 10 gated culverts that make up S-152. This allowed water from the L-67A
canal to enter the marsh and flow in a southerly direction toward the 3,000-ft
gap in the L-67 levee and portions of the adjacent canal that remain open or
have been partially or completed backfilled. Analysis of initial results from a dye-
tracer release suggests that surface-water velocities within the slough increased
several-fold, exceeding 3 cm/s in some locations (Figure 4-19). These hydrologic
changes are, in turn, expected to affect sediment entrainment, transport, and
deposition, which are processes believed to be instrumental in the maintenance
and formation of Everglades ridge-and-slough topography. The pulse flow was
maintained for 2 months, and two additional pulses are planned in 2014 and
2015 (F. Sklar, SFWMD, personal communication, 2014).
An extensive suite of measurements are being made during the experiment
to characterize the spatiotemporal variability in surface-water depth and velocity,

FIGURE 4-18 Map of the Decomp Physical Model located in “the pocket” between L-67A and L-67C.
SOURCE: Sklar (2013).
sediment transport in the marsh, sediment accumulation and entrainment rates

in the L-67C canal, surface-water chemistry, and suspended-sediment composi-
tion. Observations of fish, macroinvertebrate, and amphibian density are also
being taken in coordination with measurements of vegetation structure. The
physicochemical and ecological measurements being made during the experi-
ment will be compared with baseline data that have been collected since 2010,
prior to the beginning of DPM construction. Analysis of the data will require

FIGURE 4-19 Distribution of fluorescein dye within the ridge and slough at 1 hour (top) and
1,300 hours (bottom) after its release into the open water at a point adjacent to S-152.
SOURCE: F. Sklar, SFWMD, personal communication, 2014.

several months, but the project scientists should begin reporting preliminary
results in spring 2014.
Seepage Management
Seepage, when used in context of the CERP, generally refers to ground-

water movement. Seepage management involves regulating the exchange of
groundwater between compartmentalized areas of the Everglades that are sepa-
rated from one another by canals and levees. Groundwater and surface-water
reservoirs of the Everglades are hydraulically connected. Therefore, management
of groundwater seepage affects surface-water hydropatterns, and conversely,
manipulation of surface-water levels in canals and impoundments through
operation of control structures affects the magnitude and direction of ground-
water seepage.
Seepage management has focused most intensively on the north-south
boundary separating the remnant Everglades from isolated pockets of agricultural
land and large urban areas that have sprawled inland from the Atlantic Coast.
Plans to manage seepage across this boundary have not escaped controversy.
Restoration relies on holding water within the Everglades and thus lowering
groundwater seepage compared with current levels. However, too much seepage
control would lower the hydraulic heads within the region’s drinking-water well
fields, making them susceptible to saltwater intrusion. Others express concern
that restricting seepage from the Everglades would reduce freshwater flows to
Biscayne Bay. Thus, understanding seepage management strategies is critical to
meeting these multiple objectives as well as the Savings Clause in WRDA 2000,
which mandated that the CERP not impact existing water supplies.
The most progress on understanding seepage management can be traced
to a non-CERP project that is being privately funded by the Limestone Products
Association in exchange for wetland mitigation credits. This pilot project centers
around the construction of a 2-mile-long seepage barrier. The barrier extends
south of Tamiami Trail between the north-south trending L-31 Levee and L-31N
Canal (Figure 4-20). The hydrology of this area is dominated by the L-31N Canal,
which cuts through the exceedingly permeable bedrock of the Miami Limestone
Formation and into the top layers of the Fort Thompson Formation. During the
wet season in particular, the L-31N Canal diverts groundwater, drawn primarily
from the northeastern portion of Everglades National Park, to the C-111 basin
in south Miami-Dade County. The seepage barrier is intended to reduce this
groundwater discharge to the L-31N Canal, thereby increasing water levels and
promoting greater sheet flow in northeast Shark River Slough.
The 2-mile seepage barrier was completed on time in July 2012, only
5 months after excavation began. Construction of the barrier was an impressive

FIGURE 4-20 Position of the 2-mile-long, 35-foot-deep seepage barrier (shown in red), west of the L-31N
Canal. The orange line represents a possible 3-mile extension of the project.
SOURCE: MacVicar (2014).
technological feat and involved excavating a 32-inch-wide trench to a depth of

35 feet below ground surface using a bedrock trenching machine that resembled
a giant chain saw.7 The trench was filled with a concrete-bentonite slurry for-
mulated specifically for this application.
The performance of the L-31N seepage barrier is being evaluated through an
ongoing monitoring program including automated measurements of hydraulic
head in six pairs of groundwater monitoring wells. The four, northernmost pairs
consist of wells on the upgradient and downgradient sides of the seepage barrier.
The two, southernmost pairs of monitoring wells are positioned at control sites
beyond the terminus of the seepage barrier. The groundwater measurements are
7
See http://www.l31nseepage.org/.

supplemented by stage and velocity measurements made at five locations along

the L-31N Canal. Measurements of changes in discharge along the canal are
used to estimate groundwater seepage into L-31N.
Hydrologic conditions have been monitored for 20 months since comple-
tion of the construction of the seepage barrier. The hydrologic measurements
reveal that groundwater hydraulic heads have responded to installation of the
seepage barrier. Moreover, the data demonstrate that the subsurface barrier is
lowering rates of groundwater seepage from Everglades National Park into the
L-31N Canal (Figure 4-21). The observed reductions in seepage approximate
those predicted by a groundwater flow model developed to inform the design
of this pilot project. Based in part on these encouraging observations of seepage
reduction, Everglades National Park is supporting a proposal for a 3-mile south-
ward extension of the seepage barrier that is currently under review by the Lake
Belt Committee (Figure 4-20; R. Johnson, NPS, personal communication, 2014).
The LPA seepage management pilot provides a good example of incremen-
tal adaptive restoration, by providing tangible increments of restoration while
FIGURE 4-21 Reduction in groundwater seepage into L-31N Canal due to presence of seepage barrier.
SOURCE: MacVicar (2013).

working to resolve questions that prevent implementation of the full-scale project

(NRC, 2007). The project also appears to offer the potential for substantial seep-
age management at little to no CERP cost.
Non-CERP Projects
CERP projects are not the only restoration efforts ongoing in the Everglades
region. Several non-CERP projects are critical to the overall success of the
restoration program, and their progress directly affects CERP restoration prog-
ress. Four important non-CERP efforts with new information on their restoration
progress are reviewed in this section: the Modified Water Deliveries Project, the
Everglades Restoration Transition Plan, the Kissimmee River Restoration Project,
and the Seminole Big Cypress Water Conservation Plan. Progress on the C-111
South Dade project was discussed previously in the context of the related CERP
C-111 Spreader Canal project.
Modified Water Deliveries and the Tamiami Trail Bridge
Congress provided legislative authority in 1989 for the creation of a project

to improve water flows into Everglades National Park, where Everglades micro
topography and vegetation were in decline as a result of lack of sufficient inflows.
In 1992 the General Design Memorandum for the Modified Water Deliveries to
Everglades National Park Project (Mod Waters; USACE, 1992) envisioned several
features to increase the flow of water from WCA-3 into Everglades National Park
to accommodate flows up to 4,000 cubic feet per second (cfs). The memorandum
also provided mitigation of the effects of those flows for the 8.5-square mile
area (an area of private development and residences), and the establishment of
conveyance and seepage control measures (NPS, 2012; USACE, 2011c). Pro-
posed plans to reach these original project goals have varied over time (for more
details on the complex history of the Mod Waters project, see NRC, 2008), but
there now appears to be general agreement on the steps necessary to complete
the Mod Waters project.
As of December 2013, significant portions of the Mod Waters project have
been completed. Protection for the 8.5-square-mile area is substantially com-
plete, and many of the planned conveyance and seepage control features have
been constructed (USACE, 2013a), including installation of the S-355A and
B gated structures in the L-29 levee, S-333 modifications, four of nine planned
miles of reduction for the L-67 Extension levee, installation of the S-356 pump
station, and raising the Tigertail Camp (a tribal residential area). A central feature
of the Mod Waters project—improved conveyance across the Tamiami Trail—is
now partly complete with the construction of the 1-mile eastern bridge (Fig-

ure 4-22) and the completion of raising the roadbed to accommodate higher
canal stages.
The completion in March 2013 of the 1-mile bridge is a major step in
restoration of the hydrology and ecology of Shark River Slough in Everglades
National Park. The bridge and the raising of the remainder of the roadbed allows
for increased elevations in the L-29 canal, immediately north of Tamiami Trail.
These higher levels provide an increased head (height differential between the
canal surface in WCA-3B and Shark River Slough in Everglades National Park)
that results in peak flows of 1,848 cfs into the park, an increase of 47 percent over
FIGURE 4-22 The newly constructed 1-Mile Tamiami Trail bridge, showing conveyance area beneath the
newly elevated highway.
SOURCE: R. Johnson, NPS, personal communciation, 2014.

pre-bridge conditions, although still well below the original goal of 4,000 cfs
(NPS, 2012). The Next Steps project, directed by the National Park Service, has
completed an environmental impact statement with a plan to support up to a
total of 5.5 miles of additional bridging. The Park Service is presently working
on planning and design for the next bridge, about 2.6 miles long (DOI, 2013).
In August 2013, Governor Scott announced a commitment of $90 million in
state funds for a 2.6-mile bridge to match federal funding to reach the total cost
of $180 million (Scott, 2013).
There are several unfinished features of Mod Waters and unresolved issues,
however, that must be addressed before the project can be operated to deliver
ecosystem benefits. Some features in the protection of the 8.5-square-mile area
remain unfinished, and land acquisition and easement issues must be resolved
to complete the project. Additionally, an operations plan must be developed,
and operation of the project is dependent upon completion of an unfinished
contract of the C-111 South Dade project, which has been stalled over cost-
sharing disagreements. Finally, water quality concerns regarding compliance
under Appendix A of the Consent Decree must be resolved before more water
will be supplied to Everglades National Park (T. Morgan, SFWMD, personal
communication, 2013). Completing the Mod Waters project and overcoming
these final constraints deserves high priority to expedite restoration benefits
from substantial prior restoration investments at relatively low additional cost
(approximately $19 million; see Table 4-4).
Everglades Restoration Transition Plan
Water control for WCA-3 influences surrounding lands, including Everglades

National Park and Big Cypress National Preserve, as well as the distribution
of water within the 921-square-mile Water Conservation Area. Longstanding
challenges include balancing the right quantities of water and timing of flows
to h
abitats hosting the endangered Cape Sable seaside sparrow and other areas
providing food and nesting space for the endangered snail kite (Figure 4-23) and
wading bird species. Additionally, tree islands and ridge and slough topography in
the WCA-3 were being degraded by water levels that were too high in some places
and too low in others. An interim operating plan (IOP) was ineffective in dealing
with the various demands on water management, and in October 2012, the plan
was replaced by the Everglades Restoration Transition Plan (ERTP) (USACE, 2012c).
The ERTP was designed to provide a flexible multispecies approach to water
operations associated with WCA-3 by balancing the various demands for specific
water levels in specific places at designated times of the year. It establishes targets
for wet-season high water levels, recession rates, dry-season low water levels,
and ascension rates (Figure 4-24), and calls for increased operational fl exibility

FIGURE 4-23 Snail kite, an endangered species likely to be benefited by the Everglades
Restoration Transition Plan.
SOURCE: http://www.saj.usace.army.mil/SharedMedia/saj/2012/Nov/1/121022-A-CE999-001.
JPG.

FIGURE 4-24 ERTP targets versus observed performance, 2012-2013. The slanting horizontal boxes delineate
operational regulation zones, and the vertical red boxes represent desired water levels for snail kites, apple
snails, and wading birds at key points in the annual water cycle. The green line represents actual water levels.
The system operated under the interim operational plan (IOP) during January to October 2012 and ERTP
thereafter, although the figure plots ERTP targets throughout.
SOURCE: R. Johnson, NPS, personal communication, 2014.
in order to meet these objectives. The plan lowered wet season regulation stages
by 0.25 feet and dry season stages by 0.5 feet (USACE, 2012b), reducing stages in
northern and central WCA-3A relative to the IOP schedule. Under the new
arrangements, there will be no mandatory seasonal closure of the S-12C gate to
allow more water to flow southward into Everglades National Park in cases of
unusually high water in Southern WCA-3A (USACE, 2011a).
In the first year of ERTP operation, the dry season recession rate and the
wet season ascension rate were both somewhat faster than the target rate, and
both wet- and dry-season high water levels were higher than target levels, just

as they had been the previous year under the IOP (Figure 4-24). Deviations from
targets were actually larger in the first year of operation under ERTP, which water
managers attributed to a wet season that began early and was unusually wet,
along with a dry season (November 2012 to March 2013) with significantly less
rainfall than average (USACE, 2013b). Deviations in dry- and wet-season water
levels were sufficient to exceed levels of incidental take of endangered species
specified in the Biological Opinion of the ERTP (FWS, 2010). The threshold
deviation in short-term recession rate was also exceeded several times in water
year (WY) 2013, due to below-average rainfall in the first 3 months of 2013
(USACE, 2013a).
The ERTP has a number of ecological targets and performance measures for
individual species and ecological components (USACE, 2011a) derived from the
multispecies management strategy (FWS, 2010). The difficulties experienced in
meeting overall ERTP performance targets are reflected in failure to meet many of
these more specific targets (Table 4-6). In the spirit of adaptive management, water
managers have assessed the causes of performance failures and made adjustments
designed to improve performance in future years. However, under the ERTP, the
WCAs are to be managed to provide flood control and water s upply, as well as
TABLE 4-6 ERTP Performance Targets for Endangered Species and Other Ecological
Components of Special Concern
Measure 2011 2012 2013
Cape Sable Seaside Sparrow
>60 days dry conditions subpopulation A Yes Yes No
Water level subpopulation A < 7 ft by 12/31 Yes Yes Yes
Hydroperiod 90-210 days for six subpopulations — — 2 Yes, 6 No
Snail Kite
9.8-10.3 ft WCA-3A by 12/31 Yes Yes Yes
8.8-9.3 ft WCA-3A from 5/1 to 6/1 Yes No No
Apple Snails
Recession of 1 ft at rate of 0.05 ft per week No Yes Yes
Ascension rates < 0.25 ft per week February-September Yes, 8/8 months Yes, 8/8 months Yes, 7/8 months
Tree Islands
Peak < 10.8 ft WCA-3A Yes No No
< 60 days above 10.8 ft WCA-3A Yes Yes No
NOTES: “Yes” indicates achievement of target, “No” indicates failure to achieve target, and “—“ indicates not measured.
SOURCE: USACE (2013a).

to provide viable wetland habitat (USACE, 2011a). Thus, in many cases, failure to
meet a particular target is attributed to excessive or deficient rainfall at particular
times (USACE, 2013a), which imposes constraints related to flood control and
water supply on operations that impact the flexibility required to achieve eco-
logical goals. Also, the potential of ERTP to produce ecological benefits will be
limited until the Modified Waters Delivery Project is fully operational. Still, the
ERTP is providing opportunities for learning, and its ecological goals represent
informed management of the natural system for multiple species. It provides the
means to simultaneously address the needs of multiple species within the current
constraints imposed on water management and by the current condition of the
natural system. One cannot, however, expect the ERTP to produce significant
changes in those conditions, such as a shift in the distribution of water from
western to northeastern Shark River Slough or more flow into Everglades National
Park and Florida Bay, that the CERP is designed to achieve.
State Water Quality Treatment Projects
As part of its Long-Term Plan for Achieving Water Quality Goals, the state
has completed construction of STA Compartments B and C, and now has
approximately 57,000 acres of STAs that are permitted to operate (Figure 4-25).
Meanwhile, enhancements to maintain or improve the performance of existing
STAs continue, such as regrading some cells to address hydraulic short-circuiting
and converting or reestablishing vegetation as needed (Andreotta et al., 2014).
STA Performance. The STA performance, compliance, and optimization are

summarized in the annual South Florida Environmental Reports (Andreotta et
al., 2014; Ivanoff et al., 2013). Additionally, NRC (2010) summarizes key issues
and challenges regarding STA performance. This section reviews recent STA
performance in light of long-term goals.
During WY 2012 (May 1, 2011, to April 30, 2012), a relatively dry year
with low hydraulic loading rates, the six STAs reduced inflow total phospho-
rus (TP) flow-weighted mean concentrations from 111 to 19 μg/L (Table 4-7;
Ivanoff et al., 2013). With the exception of STA-5 and STA-6, all other STAs
produced outflow TP concentrations of <25 μg/L.8 During the wet WY 2013,
STAs received a higher average hydraulic loading rate and produced outflow
8
Both STA-5 and STA-6 receive highly phosphorus-enriched waters from the C-139 basin and
these STAs are subjected to frequent drying conditions as a result of inconsistent water availability.
Unlike other STAs, STA-5 and STA-6 are dominated by cells with emergent vegetation. These issues
have been addressed by adding additional treatment area and combining both STAs. In addition,
submerged aquatic vegetation cells were added to STA-5/6, which significantly improved the out-
flow TP concentrations (see Table 4-8).

FIGURE 4-25 Location of the Everglades stormwater treatment areas (STAs): STA-1E, STA-1W, STA-2, STA-3/4,
and STA-5/6 and the planned locations for additional STAs, STA earthwork, and flow equalization basins
(FEBs) associated with the Restoration Strategies plan.
SOURCE: http://www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_jpg/map_restoration_
strategies.jpg.
TABLE 4-7 STA Performance During WY 2012 (May 1, 2011, to April 30, 2012)
TP Inflow TP Retained
Hydraulic Load Rate Load Rate
Loading Rate Inflow TP Outflow TP (g/m2 per (g/m2 per % TP Removal
STAs (cm/day) (μg/L) (μg/L) year) year) Efficiency
STA-1E 1.4 109 21 0.56 0.46 83
STA-1W 1.2 143 22 0.63 0.54 85
STA-2 2.6 87 12 0.82 0.69 84
STA-3/4 1.4 109 19 0.54 0.44 82
STA-5 0.6 156 32 0.37 0.30 82
STA-6 1.7 126 75 0.78 0.53 68
All STAs 1.4 111 19 0.58 0.48 83
SOURCE: Data from Ivanoff et al. (2013).

flow-weighted TP concentrations ranging from 14 to 36 μg/L (Table 4-8). High

outflow TP in STA-1E and STA-1W was due to high phosphorus loading rate,
approximately two times higher than in STA-2 and STA-3/4. Overall, WY 2013
data show positive signs that STAs are performing well with an average outflow
TP concentration of 21 μg/L.
Compared with the period of record (Table 4-9), the past 2 years, includ-
ing both wet and dry conditions, demonstrate substantial improvement in STA
performance for most of the STAs when TP loading was maintained at <1 g P/m2
per year. Long-term data during the period of record show a direct relationship
between outflow TP concentrations and inflow TP concentrations (Table 4-9),
suggesting that maintaining low inflow TP levels may be needed to achieve low
outflow TP levels, although other biotic and abiotic factors also play important
roles. Overall, during the period of record, STAs have experienced variable
loadings, extreme weather conditions, and internal management of vegetation.
STA-2 and STA3/4 are the best-performing STAs over the period of record, due to
TABLE 4-8 STA Performance During WY 2013 (May 1, 2012 to April 30, 2013)
Hydraulic TP Inflow TP Retained
Loading Rate Inflow TP Outflow TP Load Rate Load Rate % TP Removal
STAs (cm/day) (μg/L) (μg/L) (g/m2/year) (g/m2/year) Efficiency
STA-1E 2.6 207 26 2.0 1.7 87
STA-1W 2.1 245 36 1.9 1.6 83
STA-2 2.6 106 22 1.0 0.8 78
STA-3/4 2.5 105 14 0.9 0.8 86
STA-5/6 0.6 131 17 0.3 0.3 90
All STAs 2.1 138 21 1.1 0.9 84
SOURCE: Data from Andreotta et al. (2014).
TABLE 4-9 Performance of STAs over the Entire Period of Record of Each STA’s Operation
Start Date Inflow TP Outflow TP TP Inflow TP Retained
[Years in μg/L μg/L Load Load % TP Removal
STAs operation] (SD) (SD) (mt) (mt) Efficiency
STA-1E 2004 [9] 179 (54) 52 (115) 173 125 72
STA-1W 1993 [20] 175 (56) 50 (31) 739 522 71
STA-2 1999 [14] 103 (38) 22 (9) 392 302 77
STA-3/4 2003 [10] 113 (29) 17 (4) 584 493 84
STA-5/6 1999 [16] 179 (59) 74 (40) 436 286 66
All STAs 1994-2012 140 (25) 37 (13) 2,323 1,727 74
SOURCE: Data from Andreotta et al. (2014).

long-term low TP loading rates (<1.0 g P/m2 per year) (Table 4-9). Rehabilitation
of STA-1W in 2007, combined with low loading rates in WY 2012, appears to
have improved its performance in WY 2012 and 2013 compared with the period
of record. It is expected that implementation of additional treatment area and
flow equalization basins (see Restoration Strategies in the next section) is likely
to reduce STA loading rates and therefore improve the outflow TP concentra-
tions. However, sustained performance of STAs depends on the effectiveness of
these restoration strategies and consistent hydraulic and TP loading to the STAs.
Restoration Strategies. In 2012, the State of Florida announced its Restora-

tion Strategies Regional Water Quality Plan to ensure that sufficient treatment
is provided for the approximately 1.4 million AF/yr currently flowing into the
Everglades Protection Area to meet the legally required water quality standard.
The plan was proposed as an alternative to the approach set forth by the U.S.
Environmental Protection Agency in its 2010 Amended Determination. The
plan includes six projects that create approximately 6,500 acres of new STAs
and 116,000 AF total capacity in three new flow equalization basins (FEBs),
which are intended to moderate inflows into existing STAs and thereby improve
their treatment performance (Figure 4-25). The Restoration Strategies plan also
includes some earthwork in STA-5/6 as well as additional source controls to
reduce nutrient loads on the STAs.9 The Restoration Strategies plan was formally
launched in September 2012, and the SFWMD FY 2014 budget includes nearly
$102 million toward the $880 million plan. The state anticipates that the entire
plan will not be constructed and fully implemented until 2024, but the A-1 FEB
in the central flow path could come online as soon as 2016 (Leeds, 2014). The
SFWMD plans to initiate construction of the A-1 FEB in June 2014 and already
has construction of the L-8 FEB under way.10 The plan’s focus on providing sig-
nificant additional flow equalization and water quality treatment is a significant
development with important implications for restoration of both water quality
and flow in the central Everglades (see Chapter 3).
As described in Chapter 3, water from the STAs cannot be redistributed in the
Central Everglades Planning Project unless it meets the water quality-based efflu-
ent limitation (WQBEL) set forth in the National Pollutant Discharge Elimination
System (NPDES) permit currently in effect.11 The NPDES permit covers discharges
9
See http://www.sfwmd.gov/portal/page/portal/xweb protecting and restoring/restoration
strategies#projects.
10
See http://www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/restoration_
strategies_update_2013_nov.pdf.
11
After release of the report in prepublication form, it was pointed out that this sentence was not
clear about which agencies are responsible for permit compliance decisions. As noted in Chapter
3, FDEP, with oversight from EPA, is the permitting authority that would interpret compliance as-
sociated with the redistribution of water.

from all STAs. However, as described above, the STAs currently operate at very
different levels of efficacy. Certain STAs, such as STA-2 and STA-3/4 (Figure 4-25)
in the central flow path, are the best performing and consistently have outputs
that are approaching the limitations prescribed by the WQBEL in the NPDES
permit (not to exceed an annual flow-weighted mean (FWM) of 13 μg/L total
phosphorus in more than 3 out of 5 years, and an annual FWM of 19 μg/L in any
water year). Other STAs are not performing as well and it could take more time
and work to bring the discharges from these STAs into compliance with NPDES
permit requirements. While the committee recognizes the need to comply with
all applicable law, including federally approved water quality standards and
the WQBEL set forth in the NPDES permit, the agencies may be unnecessarily
constraining themselves (and restoration progress) by concluding that all STA
discharges must comply with the WQBEL before any water from any STA can
be released to the Everglades. As discussed in Chapter 3, the agencies should
consider permitting mechanisms and, if necessary, design and implementation
alternatives that would allow discharge and redistribution of water from a flow
path that meets the WQBEL rather than postponing the redistribution of WQBEL-
compliant discharge until all STAs are WQBEL-compliant. If a revision to the
NPDES permit is necessary to authorize each STA (or subsets of the STAs, by
flow path) to discharge as soon as the WQBEL requirements are met, the agen-
cies should take the necessary steps to revise the permit to expedite restoration
in the central Everglades and avert ongoing ecosystem declines caused by a
reduced flow (NRC, 2012a).
Kissimmee River Restoration
The Kissimmee River drains the northern extremity of the entire Everglades
watershed, rising from lakes in the vicinity of Orlando, and flowing south
to empty into Lake Okeechobee. The Central and Southern Florida Project
caused widespread changes in the hydrology, geomorphology, and ecology of
the K
issimmee River. The previously meandering, 103-mile-long channel was
replaced by an artificial channel 55 miles long, control gates and pumps were
installed, and much of the floodplain was drained (USACE, 2012b). Unforeseen
consequences included a wholesale change in the bed and banks of the river
from sandy conditions to organic-rich fine materials with an accompanying loss
of native fishes. The replacement of freshwater wetlands with drained pasture
lands on the floodplain resulted in the loss of habitat for numerous waterfowl
and wading birds. Seventy-five percent of the historically active floodplain was
disconnected from the river; waterfowl populations declined by 90 percent
(Blake, 1980). Additionally, the river became a major source of phosphorus for
Lake Okeechobee, derived from drainage from agricultural lands.

FIGURE 4-26 Photos showing the natural historic Kissimmee River and floodplain in 1954 prior to channel-
ization of the river (left), and the same view after canal filling and river restoration (right). The filled channel
is circled in the right-hand figure.
SOURCE: T. Morgan, SFWMD, personal communication, 2013.
Two restoration projects were designed to reverse these impacts. The

issimmee Headwaters Revitalization Project refined operations in the head-
K
waters of the basin by canal refinements, construction of supplemental levees,
and improved management of control gates that regulated the outflow of the four
primary lakes in the region, and was completed in 2012. The more extensive
Kissimmee River Restoration Project was congressionally authorized in 1992 and
sought to replace the straight constructed channel with a 40-mile meandering
one more similar to the original geomorphology, backfilling of 22 miles of canal,
recarving 10 miles of river, and removing control structures that segmented the
lower reaches of the river (Figure 4-26). The first phase of its construction began
in 1999, with a 2001 completion date. Two additional phases of the project were
completed in 2007 and 2009 (Jones et al., 2014). The final project phase began
in 2012 but was recently delayed and is now expected to be completed in 2019.
Of all the projects described in this chapter, the Kissimmee River restoration
is probably the most advanced in demonstrating substantial restoration of the
natural system, and the long-term monitoring of restoration progress is a useful
example for many CERP projects. Jones et al. (2014) and USACE (2012b) report
recent insights on the project’s hydrologic and ecological benefits since the
committee’s last report:

• The project has met the target of connecting the floodplain to the channel
180 days during WY 2013 at four of the five observation sites.
• Wetland plants are thriving in the floodplain (see Figure 4-27), including
pickerelweed, arrowhead, Carolina willow, and buttonbush.
• Dissolved oxygen has met the targets for mean concentrations in the
2012-2013 year, and the dissolved oxygen concentration target of 2.0 mg/L was
met almost 90 percent of the time.
• The total phosphorus load into Lake Okeechobee remained virtually
unchanged.
• Native largemouth bass and various native sunfishes now make up 63 per-
cent of the fish community; prior to restoration, they represented only 38 percent.
• Organic deposits on the river bottom decreased by 71 percent, reestab-
lishing sand bars and providing new habitat for shorebirds and invertebrates,
including native clams.
FIGURE 4-27 A functional wetland restored to an area that once was a pasture on the floodplain of the
Kissimmee River.
SOURCE: Lawrence Glenn, SFWMD.

• Eight shorebird species, absent before restoration, have returned to the

river and floodplain, including breeding black-necked stilts.
• The 3-year running average for wading bird abundance was above the
target of 30.6 birds per square kilometer, although the target was missed during
WY 2013.
• Waterfowl abundance was more than double the target WY 2012 and
WY 2013.
The Kissimmee River restoration has made substantial construction progress

that has already resulted in measureable natural system benefits, and even more
benefits are anticipated once final control structures are finished and changes
to the water control schedule are implemented. Several project features that
remain to be completed include additional canal filling, removal of S-65C, and
construction of the S-69 weir. The agreement on a water control schedule for
the headwaters areas will also institute a more natural flow regime in the river
with additional natural system benefits.
Restoration progress on the Kissimmee River restoration (and the C-111
South Dade project) had been delayed for about 2 years by cost-sharing issues
between the SFWMD and the USACE. However, in April 2014 the agencies
reached agreement on the issues that separated them, and construction is now
set to move forward (USACE, 2014). These issues and their resolution continue
a record of solving management issues through interagency negotiation that
requires time (and delays), but that allows the project to eventually move forward
with full support of both partners.
Seminole Big Cypress Reservation Water Conservation Plan Critical Project
The Seminole Big Cypress Reservation Water Conservation Plan Critical

roject (Big Cypress Project, authorized by the 1996 WRDA) is a non-CERP
P
project focused on the Seminole Big Cypress Reservation located near the

northwest corner of WCA-3A. The project was intended to address water qual-
ity issues in agricultural runoff on the reservation, enhance water storage, and
thereby improve conditions for native vegetation on the reservation (USACE and
SFWMD, 2013d). The project included construction of new water storage areas
designed to capture stormwater discharges, a series of culverts and canals, and
24 pump stations. The Big Cypress Project addresses the quality of water flow-
ing eastward from the S eminole Reservation into the Everglades ecosystem in
WCA-3A and the M iccosukee lands (USACE, 2012d; Figure 4-28). The project
also provides additional water to rehydrate wetlands in the northern portion of
the Big Cypress National Preserve.

FIGURE 4-28 The Western Glades and the L-28 levee system at the northwest corner of WCA-3A.
SOURCE: Florida Gulf Coast University, http://www.fgcu.edu/bcw/Restore/History/History_L28.htm.

Construction of the Phase 1 conveyance canal system was completed in

2003. Phase 2 of this project has been divided into four basins north of the
West Feeder Canal (Figure 4-28). In 2008 the USACE completed construction
of the largest basin, Basin 1, which was transferred to the tribe for operations
and maintenance in 2010. The basin has failed to perform as designed, which
necessitated design modifications for the other three basins (SFERTF, 2012).
Construction of Basin 4 was completed in 2013, and Basin 2 is currently under
construction (USACE, 2014d). After tribal frustrations over the poor performance
of Basin 1, federal, state, and tribal leaders held multiple meetings in 2013 in
efforts to resolve these technical issues, and that process is ongoing.
The infrequency of Water Resources Development Acts (WRDAs) has

impeded CERP progress over the past 2 years. Seven years have elapsed since
the last WRDA was passed, and four Generation 2 CERP projects with approved
project plans awaited congressional authorization between 2012 and June 2014
when WRRDA 2014 was signed into law. Additionally, two of the previously
authorized CERP projects require reauthorization due to cost escalations; thus,
prior to WRRDA 2014, only one CERP project—Indian River Lagoon-South—was
eligible for sizeable (>$25 million) construction funding. With the passage of
WRRDA 2014, four additional projects are able to proceed with federal funding,
although the Central Everglades Planning Project was not completed in time
to be included. Lack of authorizations also had important implications for the
cost-sharing balance, discussed below.
Availability of funding also impeded CERP progress in the past 2 years.
State CERP expenditures have declined substantially in recent years, because of
reduced SFWMD revenues and the need to fund non-CERP water quality projects
to meet a 2012 Consent Order. Even though the state has spent significantly more
than the federal government on the CERP since its inception, the state has been
precariously close to the mandated 50-50 cost-sharing requirement because,
prior to WRRDA 2014, land acquisition and construction expenditures could
only be credited for the four congressionally authorized Generation 1 projects.
Declining state funding for CERP projects over the past 2 years has contributed
to cost-sharing challenges, and as of September 2013, the state’s “creditable
expenditures” exceeded those of the federal government by only $98 million.
As a result, the federal government significantly reduced spending in FY 2014
so as not to exceed the 50-50 cost share. Passage of WRRDA 2014 could allow
the state to realize approximately $400 million in additional cost-sharing credits
for prior spending, thereby easing an impending constraint on federal contribu-
tions toward the CERP.

CERP planners need to revisit the Integrated Delivery Schedule with a

renewed urgency to advance projects with the greatest potential to avert
ongoing ecosystem degradation and those that promise the largest restoration
benefits. The current draft Integrated Delivery Schedule has not been updated
since 2011, and difficult decisions will need to be made to integrate the four
Generation 2 CERP projects and the Central Everglades Planning Project (and
related project dependencies) with existing CERP and non-CERP efforts. To
expedite Everglades restoration amid limited funding, all authorized projects
cannot be advanced equally. Some projects may be more beneficial in light of
climate change and sea-level rise and others less so, and these factors should
be considered in the prioritization of restoration funding.
The restoration progress made by CERP projects to date remains fairly
modest in scope. Ecosystem responses have been detected after phased imple-
mentation in the Picayune Strand, Biscayne Bay Coastal Wetlands, and C-111
Spreader Canal projects, although many of these improvements are limited. In
some cases, such as Biscayne Bay, the scope of the restoration increment to
date is simply so limited in area that ecological responses are equally small. In
other cases, such as Picayune Strand, additional time may be needed to achieve
full ecosystem responses to the restoration measures in place. Taylor Slough
has seen significant hydrologic improvements due to restoration efforts, but the
documented benefits to date are primarily derived from the C-111 South Dade
Project, a non-CERP project. For all three of these projects, ecological responses
would be expected to increase with construction and operation of additional
project increments as well as additional time for ecosystem recovery.
Several non-CERP projects have faced bureaucratic and policy issues that
hindered implementation progress. Agency disagreements about cost-sharing
arrangements and legal requirements affected progress on the Kissimmee River
Restoration and the C-111 South Dade project by delaying them for almost
2 years. However, the SFWMD and the USACE have made important progress
to resolve these differences, and resume construction. Meanwhile, water quality
compliance issues and the lack of an operational plan are preventing realiza-
tion of restoration benefits in the Mod Waters project. Scientific knowledge is
adequate for success, and engineering problems in construction and opera-
tion appear not to be impeding restoration progress. These non-CERP founda-
tion projects offer large potential restoration benefits once fully implemented.
Renewed attention is needed to resolve the remaining bureaucratic challenges to
expedite restoration progress and realize the ecological returns from substantial
financial investments to date.
STA performance shows signs of improvement under recent management.
Long-term sustainable performance, however, will be directly influenced by
loading rates. Additional treatment area and flow equalization basins in the

Restoration Strategies project are likely to further reduce loading rates and out-
flow concentrations. Continued adaptive management, including implementa-
tion of new strategies developed through ongoing research, is needed to meet
water quality standards and maintain sustained performance of these treatment
systems.


Climate Change and Sea-Level Rise:

Implications for Everglades Restoration
Climate change is a major threat to the persistence and functioning of

ecosystems globally, including wetlands (IPCC, 2013; NCADAC, 2014; NRC,
2014). Warmer climates accompanied by changes in precipitation patterns and
increases in atmospheric carbon dioxide concentrations will affect wetland eco-
system functioning through changes in hydrologic conditions, biogeochemistry,
and primary productivity, and alter linkages with the built environment. Increases
in temperatures also will accelerate the rate of global sea-level rise, with median
projected global increases of 17 to 29 inches by 2100 for two s cenarios (IPCC,
2013; Figure 5-1). In this chapter the committee reviews the latest climate
change and sea-level-rise projections and discusses their implications for the
Everglades and restoration planning.
CLIMATE AND SEA-LEVEL CHANGE IN FLORIDA OVER THE PAST CENTURY
Global change effects on land surface temperature and precipitation are

manifested most clearly and strongly at northern latitudes, but in other regions,
patterns of global change are more complex and can be masked by other factors.
This is particularly true of the southeastern United States, which has generally
shown decreasing rather than increasing trends in land surface temperature in the
second half of the 20th century (DeGaetano and Allen, 2002; Portmann et al.,
2009; Trenberth et al., 2007). In Florida, Obeysekera et al. (2011b) investigated
trends in air temperature and precipitation at 32 meteorological stations (1950-
2008) and observed no consistent trends in either air temperature or precipitation.
There are several components of precipitation in South Florida that contrib-
ute to the complexity and variability of rainfall, including tropical cyclones and
less intense tropical storms, which can be a substantial and variable contributor
of precipitation. Adding to the complexity in precipitation patterns, sea surface
temperatures undergo slow oscillations between relative warm and cold con-
ditions (the Atlantic Multidecadal Oscillation [AMO] and the Pacific Decadal
131
FIGURE 5-1 Projected global rise in sea level for two emission scenarios in comparison with
historical records. Historical and paleorecords from salt marshes are shown in purple. The
green, blue, and red lines between 1900 and 2010 represent yearly average global mean
sea level reconstructed from tide gages using three different methods, while the light blue
line represents satellite altimetry data. The future projections show median estimates and
likely ranges for future sea-level rise for a low-emissions scenario (RCP2.6; blue) and a high-
emissions scenario (RCP8.5; red). The Intergovernmental Panel on Climate Change did not
assess the likelihood of the specific scenarios, but they should not be assumed to be equally
probable.
SOURCE: IPCC (2013).

Climate Change and Sea-Level Rise: Implications for Everglades Restoration 133
Oscillation [PDO]), which have been shown to influence precipitation quantity,

distribution, and interannual variability in South Florida (Enfield et al., 2011;
Moses et al., 2013; Shin and Lee, 2011). For example, for periods of two to three
decades, the AMO follows a warm-water phase of the North Atlantic, which is
characterized by more hurricanes and precipitation in South Florida, and then
shifts to a cold-water phase with fewer hurricanes and less rainfall (Enfield et
al., 2001; Kelly, 2004) although the PDO can interfere in ways that increase
or decrease these changes. The AMO has been in the warm-water phase since
the mid-1990s and will likely shift to the cold-water phase in the future, likely
decreasing precipitation inputs irrespective of the effects of greenhouse gases.
These oscillations may mask long-term trends in precipitation in Florida.
In contrast to temperature and precipitation, there is little uncertainty about
trends in sea level. Currently, sea level is rising almost an order of magnitude
faster than the long-term rate of 0.35 mm/yr that prevailed for the past 4,000 years
(Scholl and Stuiver, 1967; Scholl et al., 1969; Wanless et al., 1994). Using long-
term measurements at Key West, NOAA1 calculated the average sea-level rise to
be 8.8 inches (22 cm) over the past century (or 2.2 mm/yr). This value is more
than 30 percent higher than the global average of 6.7 inches (17 cm) for the
20th century (Figure 5-1) and is consistent with relatively rapid rates observed
along the Atlantic and Gulf coasts of North America (IPCC, 2013).
CLIMATE AND SEA-LEVEL PROJECTIONS FOR SOUTH FLORIDA
Given that the Comprehensive Everglades Restoration Plan (CERP) is a multi-

decadal restoration effort, it is important to understand how anticipated changes
in climate and sea level could impact restoration outcomes.
Climate Projections
Accurate projection of climate change and its effects is a major challenge

under the best of circumstances, but these challenges are amplified in the
complex meteorological environment and landscape of South Florida. Climate
change projections are derived through a complex, multistep process from
general circulation models (GCMs), which are large numerical models that
simulate land-ocean-atmosphere exchanges of energy, water, and other charac-
teristics within and across coarse grid cells. Dozens of different GCMs are used
in climate projections, which are driven by storylines that integrate economic,
demographic, and technological drivers to estimate potential future human-
caused emissions and land cover change.
1
See http://tidesandcurrents.noaa.gov/sltrends/sltrends_station.shtml?stnid=8724580.

There are several sources of uncertainty in projecting global climate changes

(Hawkins and Sutton, 2009; Kirtman et al., 2011), including uncertainty in initial
conditions, external forcings that drive model scenarios (e.g., changes in future
carbon dioxide emissions), and model uncertainty. The relative contribution of
these categories of uncertainty shift with the timescale of projections. For South
Florida, there are a number of specific issues that add to uncertainty of GCM
projections of changing temperature and precipitation. Peninsular Florida’s
proximity to the warm ocean and flat terrain create additional uncertainty in
GCM projections. South Florida is also positioned along a discontinuity in rain-
fall projections. Although the position of this discontinuity is uncertain, the
Intergovernmental Panel on Climate Change (IPCC) AR4 simulations suggest that
eastern North America will experience increases in precipitation, while for the
Caribbean there will be a marked decrease in precipitation (Enfield et al., 2011).
GCMs generally produce outputs on a relatively coarse grid scale (hundreds
of kilometers), which limits local-scale assessments of climate change. Two
broad downscaling approaches are used to translate coarse-scale GCM output
to local-scale conditions. Statistical downscaling uses empirical relationships
between past grid-based or station-based meteorological observations and com-
parable values from GCM hindcast simulations and relies on these relationships
to tune future GCM projections of climate output (e.g., surface air temperature,
precipitation) to local-scale grid or site conditions. In contrast, dynamical down-
scaling uses GCM meteorological output as input to a mesoscale climate model
to simulate potential future climate regionally or locally. Although investigations
have shown that both statistical and dynamically downscaled data are able to
reproduce historical temperature and precipitation patterns for Florida, there
are biases in these relationships which challenge the accuracy of future down-
scaled projections. For example, Obeysekera et al. (2011a) showed that various
GCMs typically underpredict historical wet-season precipitation in central and
southern Florida and do not represent the extremes in observed events. This bias
stems from an inability in the models to depict sea-breeze-driven convective
thunderstorms. There is considerable variability in projections across different
GCMs and under different future scenarios (Figure 5-2). However, ensembles
of GCMs that show similar results provide more confidence in outputs. More
consistent patterns are evident for changes in temperature than precipitation.
Obeysekera et al. (in press) summarized the general range of GCM down-
scaled climate change projections for South Florida for 2060 (Table 5-1). Results
suggest that South Florida will experience modest increases in temperature
(Figure 5-3, top). Precipitation projections are variable for different GCMs (Fig-
ure 5-2) and more uncertain than temperature projections (Figure 5-3, bottom).
Projections generally indicate increased precipitation in the fall and early winter
and decreases in late winter through early summer. Moreover, precipitation is

FIGURE 5-2 Spatial patterns of specific downscaled GCM projections of precipitation change for Florida.
SOURCE: J. Obeysekera, SFWMD, personal communication, 2014.
TABLE 5-1 Summary of Climate Change Projections for South Florida for 2060
Statistically Dynamically
Variable GCMs Downscaled Downscaled
Average temperature (°C) 1 to 1.5 1 to 2 1.8 to 2.1
Precipitation −10% to +10% −5% to +5% −76 to +50 mm
(−3 to +2 inches)
Reference crop evapotranspiration (in.) 76 to 15 mm
(3 to 6 inches)
SOURCE: Data from Obeysekera et al. (in press).

FIGURE 5-3 Box and whisker plots showing magnitude and variability of different downscaled
GCM-projected changes in (top) temperature and (bottom) precipitation from 1970-1999 to
2041-2070 for meteorological stations in Florida under the IPCC A2 scenario sorted by latitude
(after Obeysekera et al., 2014). GCM data are from the World Climate Research Programme
[WCRP] Coupled Model Intercomparison Project 3 [CMIP3] multimodel dataset. Note that the
A2 scenario depicts a world of independently operating, self-reliant nations, with continuously
increasing population, and regionally oriented economic development.

more likely to decrease with latitude through the Florida peninsula (Figures 5-2
and 5-3, bottom). Statistically downscaled projections for the Everglades also
show increases in annual temperature and decreases in annual precipitation
(Obeysekera et al., in press). As a result of this considerable uncertainty, rather
than evaluating specific projections, Obeysekera et al. (in press) developed
scenarios to probe the hydrologic response of the Everglades to hypothetical
changes in temperature, precipitation, and sea-level rise based on the results of
GCM projections (discussed below in Implications for Everglades Hydrology).
Potential changes in tropical cyclone activity also have important implica-
tions for the CERP. Although there is no evidence that climate change has altered
hurricane activity to date (Bender et al., 2010), the number of intense (category
4 and 5) hurricanes is projected to increase over the next century, while the total
number of hurricanes is expected to decrease (Bender et al., 2010; Enfield et al.,
2011). These projections are sensitive to the particular GCM models that are used
in the downscaling experiments, and hence should be interpreted cautiously.
Sea-Level Rise
Sea-level rise is already impacting South Florida. Sea level is certain to con-
tinue to rise, although the rate of the increase depends on global factors such
as future greenhouse gas emissions, thermal expansion of the ocean, and the
extent of melting from glaciers and ice sheets (IPCC, 2013). The vulnerability of
the Everglades to sea-level rise will depend on local factors, including isostatic
uplift rates, which are generally low in South Florida2 (Adams et al., 2010), and
accretion rates of peat and inorganic sediments (discussed later in this chapter).
The IPCC (2013) recently increased its estimates of global sea-level rise
(IPCC, 2007a) by 60 percent based on improved process models depicting
thermal expansion of the ocean, ice-sheet dynamics, and glacial melting. Model
simulations of future sea-level rise were run under four different scenarios for
greenhouse gas emissions called representative concentration pathways (RCPs).
The models project a likely rise in global sea level between 11 and 24 inches
by 2100 under the low-emissions scenario (RCP2.6, which requires technology
for CO2 capture that does not exist today) and a likely increase between 21 and
38 inches under a regime of continued high emissions (RCP8.5) (Figure 5-1). The
IPCC did not assess the likelihood of the RCP scenarios themselves, but these
scenarios should not be considered equally probable. Although the IPCC remains
2
South Florida rests on a relatively stable tectonic platform, located too far south to be affected
by glacial isostatic adjustment. However, Adams et al. (2010) suggested that Florida’s land surface
may be rising isostatically, driven by dissolution of the limestone bedrock. Their predicted uplift rate
of 0.047 mm/yr for northern Florida would be equivalent to a total rise of only 0.38 cm by 2100.

confident in these scenario-specific projections, some degree of uncertainty

remains with regard to (1) the climate models that are used to simulate thermal
expansion of the ocean; (2) modeling ice-sheet dynamics; and (3) modeling
the timing and magnitude of ice-sheet collapse. The stability of the Greenland
and Antarctic ice sheets has been a major element of uncertainty (IPCC, 2013),
and recent research in West Antarctica has reported more rapid rates of glacial
melting than previously anticipated (Rignot et al., 2014).
Following USACE guidance (USACE, 2011e), which was based on NRC
(1987), the USACE Jacksonville District Office developed projections for sea-level
rise in South Florida at low, intermediate, and high scenarios through 2100. These
local sea-level rise projections range from 4 to 26 inches in South Florida over
the next 50 years and between 9 and 78 inches over the next century (Figures 5-4
and 5-5; USACE and SFWMD, 2013b). As previously discussed in the context of
Annex I Water Quality and Sea Level Rise
Florida’s observed sea-level rise, ocean circulation patterns can cause local sea-
level changes to differ from global changes, creating more uncertainty in local sea-
level-rise projections compared with global projections. Thus, it is reasonable that
the local USACE projections fully encompass and, at the upper projections, exceed
U.S. Army Corps of Engineers EC 1165-2-212
Relative Sea Level Rise Scenarios for Key West, FL
8.00
Relative Rise (meters)

Dashed lines indicate the EC equations are being used past year 2100 which the guidance
7.00 documents do not address.
2.00
Relative Rise (feet)
6.00
5.00 1.50
4.00
1.00
3.00
2.00
0.50
1.00
0.00 0.00
1990
2000
2010
2020
2030
2040
2050
2060
2070
2080
2090
2100
2110
2120
Year
Figure I-7. Projected
FIGURE Sea rise
5-4 Sea-level Level Rise (1922-2113)
scenarios for Key West, Florida, based on USACE sea-level rise guidance EC
1165-2-212.

Impact of Sea Level Rise on Net Project Habitat Improvement
(Assumptions: TOR Land Elevation, Stage Adjustment)
350000
(Acres)
300000
FIGURE 5-5 Predicted land loss in Everglades National Park based on 2 feet of sea-level rise (the intermediate
scenario for 2100 in Figure 5-4), (a) assuming existing topography and (b) assuming complete loss of peat
soils, which leads to substantially greater land loss. Neither scenario considers new peat accretion.
the most recent global sea-level rise projections of the IPCC (2013; 11-38 inches
by 2100; see Figure 5-1) and NRC (2012b; 20-55 inches by 2100).3
IMPLICATIONS FOR THE EVERGLADES
The impacts of climate change on the Everglades will depend upon the
magnitude and rate of change in the physical environment (e.g., sea-level rise,
3
NRC (2012b) global sea-level rise estimates exceeded those of the IPCC (2013) because the NRC
assumed higher rates of loss from ice sheets and used a different extrapolation procedure based on
Meier et al. (2007).

temperature) and the ecosystem’s capacity to resist and/or be resilient to these

stressors. A warmer climate in South Florida accompanied by changes in pre-
cipitation patterns will affect hydrologic regimes, biogeochemical cycling, com-
munity composition and productivity, and, hence, wetland ecosystem structure
and function. Accelerated sea-level rise will likely submerge many areas, thereby
increasing the salinity of freshwater wetlands, altering biotic communities and
productivity, and changing the rates and decomposition pathways of organic
matter (Weston et al., 2006, 2011). Alterations to natural disturbance regimes,
such as fire or intense hurricanes, could also have significant ecosystem effects.
These issues were explored in a recent workshop on the ecological effects of
climate change in the Everglades.4 In this section, the committee describes the
implications of climate change and sea-level rise on Everglades hydrology,
landscapes, water quality, and biota.
Implications for Everglades Hydrology
The hydrologic responses to future climate conditions are particularly chal-

lenging to characterize and quantify in the rainfall-driven South Florida eco-
system. In addition to the uncertainties in climate projections discussed in the
preceding section, South Florida water management operations may also change
in response to changing climate. For example, the water level in coastal canals
could be maximized to buffer the coastal groundwater system against saltwater
intrusion (Obeysekera et al., 2011a). Future increases in the population of Florida
will increase the demand for water resources for urban areas, and under chang-
ing climate conditions, water demand is likely to change.
As a result of these important but uncertain drivers, projections of changes
in hydrologic conditions in response to a changing climate are highly uncer-
tain. From this perspective, Obeysekera et al. (2014) conducted a preliminary
(“screening level”) assessment to help understand the sensitivity of the water sys-
tem to climate change drivers and the potential implications for water resources
and management in South Florida. Using the South Florida Water Management
Model, Obeysekera et al. (2014) evaluated the hydrologic outcomes of a series
of hypothetical scenarios:
1. 2010 Baseline (2010 water demands and land use corresponding to and
simulated with 1965-2005 rainfall and evapotranspiration);
2. 2010 Baseline with a 10 percent decrease in rainfall;
3. 2010 Baseline with a 10 percent increase in rainfall;
4
See http://www.ces.fau.edu/climate_change/ecology-february-2013/.

4. 2010 Baseline with a 1.5oC increase in temperature and a 1.5-ft increase

in sea level with increases in coastal canal maintenance levels;
5. 2010 Baseline with 10 percent decrease in rainfall, 1.5oC increase in
temperature, and a 1.5-ft increase in sea level with increases in coastal canal
levels;
6. 2010 Baseline with 10 percent decrease in rainfall, 1.5oC increase in
temperature, and a 1.5-ft increase in sea level with no increases in coastal canal
levels; and
7. 2010 Baseline with 10 percent increase in rainfall, 1.5oC increase in tem-
perature, and a 1.5-ft increase in sea level with increases in coastal canal levels.
These hypothetical climate scenarios are reasonable changes that might be

anticipated based on statistically downscaled GCM projections for South Florida
for 2060 (Table 5-1). The analysis, however, was highly simplified, because sea-
sonal and extreme interannual variations in precipitation were not considered.
Instead, changes in precipitation were applied uniformly across the year, based
on 1965-2005 historical climate data, even though global climate models have
projected increasing precipitation extremes over many regions (Kharin et al.,
2007; O’Gorman and Schneider, 2009; Sun et al., 2007).
Results of this analysis show that water discharge and demand are sensi-
tive to hypothetical climate change projections. The hypothetical 10 percent
increases in precipitation results in increases in water stage and discharge
throughout the South Florida ecosystem (Figure 5-6). The scenarios of increases
in temperature (i.e., evapotranspiration) and decreases in rainfall are projected
to increase water demand and decrease runoff, which results in particularly
acute water shortages (Figure 5-6). Simulations show up to 1.7-ft decreases in
the stage of Lake Okeechobee with increasing temperature only (Scenario 4),
and simulations of 10 percent decrease in rainfall combined with increasing
temperature (Scenario 7) resulted in up to 6-ft decreases in lake stage (Table 5-2;
Obeysekera et al., 2014). This direst scenario resulted in unmet agricultural
water supply demand of 40 to 58 percent (up from 7 to 8 percent in the 2010
base), highlighting the potential water supply pressures under future scenarios.
Such decreases in precipitation would impact both surface-water and ground-
water levels, reducing freshwater flows to estuaries and increasing the extent of
saline intrusion of coastal wetlands and aquifers (Saha et al., 2011).
This analysis suggests that for conditions that are likely occur in the future,
water quantity challenges could become a critical issue in South Florida. Under
scenarios of increased precipitation, the CERP as currently designed could
produce desired hydrologic outcomes, but scenarios of decreased precipitation
or increased temperature (or both) result in large decreases in flow that would
undermine restoration as currently planned.

FIGURE 5-6 Hypothetical simulations showing mean annual changes in water stage for (a) 2010 baseline
with 10 percent decrease in rainfall, 1.5°C increase in temperature, and 1.5-ft increase in sea level with
increases in coastal canal levels (Scenario 5) and (b) 2010 baseline with a 10 percent increase in rainfall,
1.5°C increase in temperature, and 1.5-ft increase in sea level (Scenario 7).
Implications for Everglades Landscapes
The Everglades landscape is especially sensitive to rising sea level because it

has low topographic relief of porous limestone bedrock and is in close proximity
to the ocean. The topography of the Everglades is shaped by two components:
a dynamic surficial layer of wetland soil and the stable floor of the underlying
bedrock basin (Gleason and Stone, 1994; Parker and Cooke, 1944; Petuch and
Roberts, 2007). The bedrock rises less than 10 ft above mean sea level around

TABLE 5-2 Changes in Hydrologic Conditions Relative to 2010 Baseline with Three Climate
Change Scenarios
Scenario 4: Scenario 5: Scenario 7:
No change Precip., 10% Decrease Precip., 10% Increase Precip.,
+1.5°C, 1.5-ft SLR +1.5°C, 1.5-ft SLR +1.5°C, 1.5-ft SLR
Lake Okeechobee stage Up to 1.7-ft decrease Up to 6-ft decrease Minimal change
Structural inflow to WCA-3 −247 million m3/yr −704 million m3/yr +245 million m3/yr
(−15%) (-43%) (+15%)
Structural inflow to −337 million m3/yr −820 million m3/yr +314 million m3/yr
Everglades National Park (−24%) (−58%) (+22%)
NOTE: SLR = sea-level rise; WCA = Water Conservation Area.
SOURCE: Adapted from Obeysekera et al. (in press), Havens and Steinman (2013).
Lake Okeechobee, while the bedrock underlying the Shark River Slough rises
less than 3.3 ft above mean sea level (Parker and Cooke, 1944). Everglades
freshwater wetland soils, consisting mostly of organic-rich peat, are generally
less than 3.3 ft deep across large portions of the central and southern Everglades,
with thicker peats in some areas (e.g., northeastern Water Conservation Area 3
[WCA-3], localized depressions) (Richardson, 2008; Scheidt and Kalla, 2007).
Freshwater peat provides essential structure and slope that influence the direc-
tion and velocity of water flow in the Everglades as the peat itself is shaped by
the distribution and velocity of water across the landscape. These peat soils
also support the ridge-and-slough landscape and many tree islands (Box 5-1).
Soils within coastal wetlands (e.g., salt marshes, mangroves) contain substantial
organic matter along with varying amounts of inorganic sediment washed in by
tides, waves, or storm surges and trapped by plant structures (Castañeda-Moya
et al., 2013; Krauss et al., 2013).
Freshwater peat in the Everglades represents a dynamic surface that will
continue to change in the future through accretion and/or subsidence. Peat accu-
mulates when plant materials are only partially decomposed prior to burial and
compaction. Net accretion requires submerged, anaerobic conditions that allow
the accumulation of plant material to outpace decomposition and compaction
(see NRC, 2012a, for more detail on freshwater peat accretion rates). However,
peat is highly susceptible to subsidence under dry conditions. When the water
table falls, pore space collapses and oxygen penetrates more deeply into the
peat profile, driving more rapid decomposition (NRC, 2012a). A future scenario
of decreases or no change in precipitation coupled with increased temperature
and evapotranspiration would reduce hydroperiods, accelerating rates of peat
decomposition. Fire regimes are also likely to shift under such conditions. While

BOX 5-1
Potential Effects of Reduced Water Inflows on the
Ridge-and-Slough Landscape
In the central and southern Everglades, the ridge-and-slough landscape consists

of linear ridges that alternate with deeper sloughs; these patterns generally run north-
south, parallel to pre-drainage flows (Gaiser et al., 2012). The ridges have character-
istically short hydroperiods dominated by sawgrass, while the sloughs have longer
hydroperiods dominated by water lilies (McVoy et al., 2011). Tree islands are among
the highest and driest habitats in the Everglades (Wetzel et al., 2011), and irregularly
punctuate the ridge-and-slough habitat matrix. These islands are floristically diverse,
provide critical habitat for many wildlife species, and are sites where nutrients are con-
centrated (Ross et al., 2006; Wetzel et al., 2011) and sequestered by the dominant tree
species (Lejeune et al., 2004).
Flow patterns that initially built and maintained these features, through controls
on peat formation and sediment movement (Brandt et al., 2000), have been highly
modified, with a resultant compression of the once variable topography (Sklar et al.,
2001). Drainage and compartmentalization of the Everglades have led to peat sub
sidence and conversion to marl prairie habitat on the wet prairie ridges (Davis et al.,
2005b), and d egradation of tree island communities (NRC, 2012a). Compositional shifts
away from tree dominance on the islands have disrupted their capacity to concentrate
and store nutrients, with attendant release/leakage of nutrients into adjacent oligotrophic
habitats and displacement of sawgrass assemblages by cattails (Wetzel et al., 2009).
Without appropriate hydrologic restoration, the future of these features is in jeopardy
because water surface levels are currently inadequate to move sediment from slough to
ridge (Larsen et al., 2009). Continued disruption of flows, and potentially more severe
water deficits with climate change, will drive further deterioration of habitat heterogeneity
and increased homogenization of vegetation. In the face of climate change, implement-
ing hydrologic restoration in the central Everglades (see Chapter 3) would help protect
the remaining features of this iconic, patterned landscape that provides critical habitat
in support of Everglades diversity.
low-intensity surface fires generally have only ephemeral impacts on Everglades

vegetation, highly intense fires can result in large losses of inland peat over a
short period (Loveless, 1959; Sklar et al., 2001). With existing water manage-
ment, a scenario of reduced future precipitation would therefore increase rates
of freshwater peat loss, further altering the slope and microtopography of the
landscape and impacting water depth and flow (Nungesser et al., 2014; see
Box 5-1). However, increased precipitation in South Florida would increase
mean water depths in the freshwater Everglades wetlands (Figure 5-6), reducing
microbial decomposition rates (DeBusk and Reddy, 1998) and thereby promot-
ing peat accretion.
Rates of coastal peat and inorganic sediment accretion or subsidence will
directly influence the rate of coastal wetland retreat and other impacts of sea-

level rise on the Everglades landscape. Most coastal wetlands possess a limited
capacity to keep pace with rising sea level through accretion of organic matter
and storm-derived sediment. In coastal wetlands, accretion and subsidence rates
vary widely among different depositional settings and with the extent of human
impacts (Cahoon and Lynch, 1997; Kirwan and Megonigal, 2013). Sediment
cores indicate that the average accretion rate in mangroves is about 1 mm/yr
over millennial timescales, with a range of 1-3 mm/yr from Florida and adja-
cent regions (McKee et al., 2007; Parkinson et al., 1994). More rapid accretion
rates are possible over shorter time intervals—accretion rates of 6 mm/yr over
several years and even higher rates associated with single storm events have
been reported (see Box 5-2). However, the implications of these short-term,
local elevation changes in the context of sea-level rise remain poorly understood
(Kirwan and Megonigal, 2013). Continual monitoring of surface elevations is,
therefore, needed over extended time periods to determine the response of
wetland deposits to rising sea level.
BOX 5-2
Climate and Sea-Level Rise Effects on Mangrove Swamps
Mangrove swamps occupy the marine-terrestrial interface and are therefore among
the “first responders” to sea-level rise. These communities typically have distinct spatial
zonation patterns, which are governed by gradients in salinity and soil conditions (Chen
and Twilley, 1999; Egler, 1952). Marine forces are clearly important, but the timing
and quantity of freshwater flows from the upper parts of the watershed also influence
salinity levels. This interplay can influence water budgets and ecosystem productivity,
as elevated salinities during the dry season lead to decreased evapotranspiration and
carbon assimilation rates (Barr et al., 2014). Thus, mangrove community distribution on
the landscape is shaped bidirectionally through the interplay between freshwater flows
and tidal regimes (Davis et al., 2005a), making them excellent indicators of climate
change because they are highly vulnerable to marine forces and hydrologic changes
in the watershed.
With increasing sea-level rise and water management practices during the 20th cen-
tury, mangroves have been declining in coverage on the southern Everglades landscape
(Wanless et al., 2000), despite inland migration in many areas. A readily visible indicator
of this migration is the inland shift in the upper edge of the mangrove/marl prairie e
cotone,
also known as the “white zone” (Ross et al., 2002; Figure 5-7). These shifts often coincide
with displacement and sometimes concurrent inland movement of adjacent freshwater
sawgrass communities (Ross et al., 2000) and appear to be facilitated in some cases
by fire (Smith et al., 2013). With rising seas, potentially drier conditions that heighten
the likelihood of fire at the mangrove-marsh ecotone and increased salinity levels in the
estuaries are likely to continue.
continued

BOX 5-2 Continued

The seaward fringes of the mangrove landscape are maintained, in part, through
peat accretion, which occurs at the upper end of their tidal range (Scholl, 1964). The
ability of mangroves to keep pace with sea-level rise and persist in situ is also uncer-
tain because accretion rates are highly variable and dependent not only on sufficient
freshwater inflows to prevent oxidation of existing peat but also factors that control pro-
ductivity of the vegetation and rates of organic matter inputs that drive accretion rates.
In a recent review of mangrove adjustments to sea-level rise across the globe, Krauss
et al. (2013) reported soil surface elevation changes ranging from −3.7 to 6.2 mm/yr
over several years. Accretion rates as high as 20.8 mm/yr were reported, although
subsurface subsidence reduced the total surface elevation change. Storm events were
generally responsible for the upper limit of this range. Smoak et al. (2013) reported ac-
cretion rates of 5.9 and 6.5 mm/yr in Everglades mangrove forests produced by a single
storm-surge deposit, whereas long-term rates (averaged over a 130-year period) of 2.5
to 3.6 mm/yr were measured at the same sites. After Hurricane Wilma, Casteñeda-Moya
et al. (2010) reported 5 to 450 mm of sediment deposition in the Shark River mangrove
forests—up to 17 times greater than average annual accretion rates of approximately
3 mm/yr. Although storm surges can provide sizeable deposits of inorganic sediment,
part of this elevation gain will subsequently be lost through compaction and erosion
(Whelan et al., 2009). The challenge for interpreting these short-term accretion rates is
to determine their implications for accretion rates over multidecadal timescales or longer
in the context of projections of sea-level rise.
In some areas, these systems can keep pace with current rates of sea-level rise, but
in other places where accretion rates are low, saltwater encroaches and the swamps
succumb to the sea (Lodge, 2010). Continued acceleration of sea-level rise will increase
their vulnerability, as elevated salinity levels limit productivity and can lead to peat col-
lapse (Chambers et al., 2013a,b). The mangrove zone in Taylor Slough, for example, is
highly threatened due to low productivity and, hence, low accretion rates (Gaiser et al.,
2006). Future rates of sea-level rise that are sufficiently rapid to impede inland migration
may threaten their persistence in the broader landscape.
FIGURE 5-7 Images of the coastal gradient from 1940 (left) and 1994 (right)
between U.S. Highway 1 and Card Sound Road, illustrating shifts in the “white zone.”
SOURCE: Ross et al. (2000).

The USACE scenarios for Key West, Florida (Figure 5-4) describe rates
of sea-level rise that increase from historic rates of 2.24 mm/yr to between
8 and 27 mm/yr under low and high scenarios by the end of the 21st century
(G. Landers, USACE, personal communication, 2014). Thus, it remains highly
questionable whether accretion rates in coastal wetlands will be sufficient to
prevent inundation and retreating shorelines in the future or to what extent
accretion could at least mitigate the impacts. Assessing current accretion rates
in both the coastal and freshwater wetlands of the Everglades and understand-
ing the factors that contribute to their variability are high priorities for research.
Efforts are currently under way in the Everglades to monitor changes in surface
elevation across a network of control points using customized elevation gauges
to assess accretion rates in the context of sea-level rise (Box 5-3).
The phenomenon of “peat collapse” in coastal wetlands (Cahoon et al.,
2003; Day et al., 2011; DeLaune et al., 1994) poses significant concerns for
Everglades management and restoration in the face of climate change. Peat
collapse has been used to describe the conversion of coastal marshes to open
water as well as sudden land subsidence in salt marshes and mangroves
(Figure 5-10). The peat deterioration can release a large amount of sequestered
carbon (as carbon dioxide and methane) and nutrients, such as phosphorus,
stored in the soil profile (Bouillon et al., 2008; Nichols et al., 2007). An
important suspected mechanism for peat collapse is increasing saltwater intru-
sion from sea-level rise and tropical storm surges and associated high sulfate
concentrations that alter microbial organic matter decomposition pathways
and rates (Chambers et al., 2014; Erickson et al., 2007; Weston et al., 2006).
However, additional agents for lowering peat surface elevations could include
mechanical damage to the vegetation or peat skeleton by high winds or storm
surges (e.g., Doyle et al., 1995; Kirwan and Guntenspergen, 2010; Smith et
al., 1994, 2009), nitrogen inputs that enhance microbial decomposition of root
structures (Deegan et al., 2012; McKee et al., 2007), and loss of groundwater
inputs (Whelan et al., 2005).
Implications for Water Quality
Changes in climate can alter linkages between coupled hydrologic and

biogeochemical cycles that are critical to the functioning and persistence of
wetland ecosystems (Reddy and Delaune, 2008; Reddy et al., 2010; Rivera-
Monroy et al., 2007). Shifts in the frequency, timing, and intensity of rainfall
events can affect the transport of sediments, nutrients, and other constituents
from wetlands to downstream aquatic ecosystems. Perturbations in hydroperiod
and hydrologic and pollutant loading rates can significantly affect vegetation,
algae, microbial and animal communities in native and constructed wetlands

BOX 5-3
Measuring Peat Accretion and Subsidence
Wetlands have a dynamic land surface that continually rises and falls through the
interplay of physical, chemical, and biological processes. Key questions remain as to
what extent rates of peat and sediment accretion can keep pace with the rapid rise in
sea level projected for the 21st century.
Two different approaches have been used to measure accretion and subsidence
rates over contrasting timescales. The traditional method is based on the analysis of
sediment cores that can be dated into discrete time slices of 0-50 years by 137Cs, 0-150
years by 210Pb, and 500-40,000 years by 14C. Accretion rates can then be calculated by
dividing the length of each section by its total age, although much finer age resolution is
often possible for a 210Pb chronology. The alternative method directly measures shorter-
term changes in surface elevation by means of custom gauges (e.g., the sediment-
erosion table-marker horizon [SET]-MH system of Cahoon et al., 1995; Webb et al.,
2013; Figures 5-8 and 5-9). Although the different approaches are complementary, they
FIGURE 5-8 A sediment elevation table (SET) used to measure changes in the
elevation of the soil surface in a mangrove forest in Everglades National Park.
SOURCE: U.S. Geological Survey, http://fl.biology.usgs.gov/Science_Feature_

Archive/2010/monitoring_enp/monitoring_enp_gallery.html.

provide different measures of accretion rates that are specific to a discrete timescale.
Any comparison of accretion rates needs to consider the general tendency for these
rates to decline over longer time spans because of the continual loss of pore waters
(by compression) and organic matter (by decomposition). These processes are most
rapid in the upper portion of a sedimentary profile (e.g., Bemer, 1980; Glaser et al.,
2012), and therefore, caution should be exercised in extrapolating short-term rates to
longer timescales.
FIGURE 5-9 Measurement of soil accretion using the marker horizon method in
Everglades National Park. In this method, researchers place a layer of feldspar clay
(visible as a white layer) on the surface of the marsh and later return to measure the
soil that has accumulated. The marker horizon method is often used in conjunction
with a sediment elevation table (SET) to measure total soil accretion or erosion.
SOURCE: U.S. Geological Survey, http://fl.biology.usgs.gov/Science_Feature_

Archive/2010/monitoring_enp/monitoring_enp_gallery.html.

FIGURE 5-10 Peat collapse at northern Cape Sable, Everglades National Park.
SOURCE: Wanless and Vlaswinkel (2005).
(stormwater treatment areas), and associated biogeochemical processes that

ultimately have significant effects on water quality. Reduced precipitation and
increased evapotranspiration will decrease the water content of wetland soils.
Dry conditions promote the oxidation of soil organic matter, which results in
the mineralization of associated chemical elements (Holden et al., 2004; Reddy
et al., 2006). Also, oxidation of sulfides can occur, which can decrease soil
pH and facilitate the mobilization of phosphorus bound to calcium carbonate
(Reddy and DeLaune, 2008). When these dry areas are rehydrated, dissolved
and particulate forms of carbon, phosphorus, nitrogen, sulfur, and mercury are
released, increasing nutrient and contaminant loads to downstream habitats
(Bates et al., 2000; Strober et al., 1995). Increases in wet-dry cycles accelerate
biogeochemical cycling, and element availability and loss, which could lead to
exceedences of Everglades nutrient criteria.

Climate-change-induced temperature increases influence several biogeo-

chemical processes of wetlands and water quality. For example, increased
temperature can increase primary productivity, organic matter decomposition,
nutrient regeneration, and greenhouse gas emissions, and alter the composi-
tion and diversity of biotic communities (Carney et al., 2007; Watts et al., 2010).
An increase in rates of these biogeochemical processes is likely to increase over-
all export of nutrients, dissolved organic matter, and associated contaminants
and impact downstream water bodies (Qualls and Richardson, 2003; Reddy et
al., 1999).
Sea-level rise that exceeds the rate of vertical soil and sediment accretion
causes increased salinity stress in freshwater wetland communities and shifts
ecosystems from freshwater to brackish (Koch et al., 2012; Saha et al., 2011;
discussed in the next section). Increased sulfate inputs can potentially increase
organic carbon mineralization and carbon dioxide emissions while decreasing
methane emissions (Chambers et al., 2011, 2013a,b, 2014; Weston et al., 2011).
These biogeochemical changes can increase the release of bioavailable nutrients
(e.g., nitrogen, phosphorus), ultimately degrading water quality (Chambers et
al., 2013a, 2014). In coastal phosphorus-limited wetlands, additional inputs of
phosphorus from storm surge deposits can actually enhance the productivity
of mangrove forests (Castañeda-Moya et al., 2010).
Implications for Everglades Biota
The effects of climate change and human-driven alterations of freshwater

flows are already unfolding in the Everglades (Gaiser et al., 2012). Increased
rates of sea-level rise have decreased the areal extent of several Everglades habi-
tats, changed the distributions of many species, and driven inland migration of
coastal vegetation (Box 5-2; Willard and Bernhardt, 2011). The rate and nature
of future change remain unclear, however, because of uncertainty in downscaled
climate change forecasts for the Everglades (discussed previously in this chapter),
and the poor understanding of the capacity of ecological systems to respond
to these impacts. Additionally, the multiple, interacting factors (e.g., increases
in temperature, sea-level rise, changes in the quantity and distribution of pre-
cipitation, increases in atmospheric carbon dioxide and associated responses
in biogeochemistry and ecology) are likely to generate complex effects that are
difficult to fully predict.
Changes in precipitation, temperature, sea-level rise, and atmospheric car-
bon dioxide, in conjunction with anthropogenic alterations to hydrology, will
collectively dictate the future environmental templates to which species respond.
In climate change scenarios where sea-level rise is marked, temperature is
elevated, and precipitation is reduced (Figure 5-6a), shortening of hydroperiods

should occur, with concomitant shifts toward less flood-tolerant vegetation and
peat decomposition. The past century of Everglades water management offers
numerous lessons about the adverse ecological impacts of reduced water flows
(see Box 5-1 for one example). In scenarios with increasing rainfall (Figure 5-6b),
freshwater flows can continue to maintain the diverse array of habitats in the
Everglades and abate saltwater intrusion of coastal wetlands, essentially holding
the sea at bay (Gaiser et al., 2012, Saha et al., 2012).
Changes in the composition and structure of Everglades communities are
expected as species respond to changing climate. Species exposed to a warmer
and perhaps drier Everglades subject to sea-level rise are likely to shift in dis-
tribution across the landscape in accordance with their climatic envelopes. For
example, where salt marsh assemblages interface with mangroves, the northern
edge of the mangrove distribution is controlled by the lack of cold tolerance.
With increasing temperatures, mangroves are likely to advance northward with
the freeze line (Cavanaugh et al., 2014), perhaps at the expense of transitional,
brackish marsh assemblages (Stevens et al., 2006). Species with broad physi-
ological tolerances will be the slowest to respond to increases in temperature,
whereas those with narrow physiological ranges will be impacted more imme-
diately. Drier conditions in the Everglades are also likely to reduce the densities
of aquatic species that rely upon refugia during the dry season (e.g., fish and
invertebrates; Catano et al., 2014), with consequent negative impacts to wading
birds and other species that depend upon this prey base. With continued envi-
ronmental changes, species eventually reach tipping points, beyond which they
will either shift spatially on the landscape or gradually decline in abundance.
Low-lying coastal wetlands are sentinels of climate change impacts (Brinson
et al., 1995; Scavia et al., 2002). They may be initially capable of coping by
adjusting physiologically or vertically through biophysical processes to escape
submergence (Cherry et al., 2009; McKee and Cherry, 2009; Morris et al.,
2002). Where coastal species cannot keep pace with sea-level rise through ver-
tical adjustment, their distributions contract at the seaward edge, and upslope
expansion of species distributions must occur or their populations will gradu-
ally decline and disappear from the landscape (Brinson et al., 1995; Craft et
al., 2009; Donnelly and Bertness, 2001; Williams et al., 1999). Thus, upgradi-
ent freshwater wetlands may be gradually converted to brackish marshes and
finally to salt marshes in response to increased salinity. The rate and direction of
response will be determined by both stress tolerance at their seaward edge and
competitive ability at the inland edge of their distributions (Crain et al., 2004;
Ervin and W etzel, 2002; Kim et al., 2011). Among the biogeochemical processes
affected by saltwater intrusion are increased sulfate inputs, which can increase
the potential for sulfide toxicity to plants. Bidirectional compression could result
if species are increasingly limited by environmental stress at the lower or upper

ends of their distributions, leading to coastal “squeezing” (Shirley and Battaglia,

2006, 2008).
Increased hurricane intensity (Bender et al., 2010; Blake et al., 2011) could
also have important implications for Everglades biota. Intensified storms would
drive changes in light and water availability to plants (Bianchette et al., 2009;
Guntenspergen et al., 1995) and increase storm surges and associated wrack
deposition (Blake et al., 2011; Tate and Battaglia, 2013), salt burning (Cahoon,
2006; Lam et al., 2011), and wind-driven damage to forest canopies (Lam et al.,
2011; Rodgers et al., 2009).
A poorly understood but potentially important aspect of global change is
the fertilization effects of increases in atmospheric carbon dioxide concentra-
tions on wetland vegetation (Rasse et al., 2005). This process can enhance
primary productivity and peat accretion (Erickson et al., 2013; Krauss et al.,
2013) and lead to heightened sequestration of carbon, nitrogen, phosphorus,
and mercury. Carbon dioxide fertilization effects could ameliorate several of
the adverse consequences of climate change. Some plants, for example, reduce
the size of pores that allow CO2 to enter leaves for photosynthesis, while
still increasing carbon assimilation. This adjustment leads to reduced evapo
transpiration (de Boer et al., 2011) and increased water-use efficiency (Li et al.,
2010), potentially offsetting some effects of rising temperature. However, this
trend is unlikely to increase indefinitely because the responses of individual
plant species will be bounded by their genetic capacity to adapt structurally to
future atmospheric carbon dioxide levels (Lammertsma et al., 2011). Shifts in
plant community composition are also expected because CO2 fertilization can
influence the timing of life-cycle events (e.g., flowering) (Springer and Ward,
2007), germination patterns (Mohan et al., 2004), and salinity tolerance of
plants (Rozema et al., 1991). Grasses and sedges with a photosynthetic pathway
that can better utilize increased CO2 and photosynthesize faster (e.g., sawgrass)
may increase in abundance over similar species that use alternative pathways
(Drake et al., 1996; Pearlstine et al., 2010). The effects of elevated CO2 are
complicated, however, by temperature and precipitation regimes (Bjorkman et
al., 1974; Raven, 2001) and may be relatively short-lived in some species as
they plateau in their responses due to nutrient limitations (Reich et al., 2006).
An improved quantitative understanding of carbon dioxide fertilization effects
on wetland and marine ecosystems of South Florida would help refine predic-
tions of the impacts of changing climate.
IMPLICATIONS FOR THE CERP AS ORIGINALLY DEVELOPED
Rising sea level and changes in evapotranspiration and precipitation could

have significant effects on the success of the CERP.

Implications of Sea-Level Rise
Sea-level rise is already impacting shallow coastal marsh habitats (Figure 5-7),
altering the salinities of surface waters and groundwaters, and changing the struc-
tural and operational requirements of coastal water management infrastructure
(see Figure 5-11). To consider how future sea-level rise might affect the CERP,
the committee considered three projects or areas targeted for CERP restoration:
Picayune Strand, Biscayne Bay Coastal Wetlands, and Florida Bay. Each illustrates
a different aspect of how sea-level rise may affect restoration.
Picayune Strand
In the Picayune Strand area in southwest Florida, drainage for a failed

housing development caused the broad-scale conversion of freshwater wetland
forests and marshes to communities dominated by species better adapted to drier
conditions. The canal system increased the incidence of wildfires and oxidation
of peat, led to proliferation of invasive species, and caused some inland expan-
sion of mangroves (Chuirazzi and Duever, 2008). The objective of the Picayune
Strand CERP project is to plug canals, rehydrate the area, and restore freshwater
wetland habitat (see Chapter 4). The potential effects of sea-level rise stem from
the fact that the project is a low-lying freshwater wetland, with ground surface
elevations ranging from 3 to 10 ft NAVD, with several sloughs 0.5-2 ft lower in
elevation (USACE, 2013e). The groundwater table can be as low as 2 ft (0.6 m)
above sea level (Chuirazzi et al., 2012). USACE (2013e) determined that with
2 ft of sea-level rise (approximately the USACE intermediate local sea-level rise
scenario in 2100), 9 percent of the project area would be inundated. Thus, the
Picayune Strand project is likely to be minimally impacted by intermediate sea-
level rise projections, but the extent to which project goals are affected remains
unknown.
Shoreward portions of the soils in Picayune Strand will become increasingly
impacted by saline intrusions with sea-level rise. As sea level rises, the ground-
water salinity gradient would move inshore along with associated plant and thus
animal communities. These effects have not yet been assessed (USACE, 2013e),
although they could be determined using a coupled surface-water-flow variable-
density groundwater model (e.g., Langevin et al., 2005). However, elevated
groundwater stages resulting from the project will likely reduce the rate of salin-
ity intrusion (compared with a future scenario without the p roject). The Picayune
Strand project is therefore likely to delay ecological transitions from native
freshwater wetland vegetation (e.g., cypress forest, sawgrass marshes) to brack-
ish marshes and enhance the resilience capacity of coastal wetlands to cope
with sea-level rise.

FIGURE 5-11 Vulnerability of SFWMD coastal structures to sea-level rise. High-vulnerability structures

are red, medium-vulnerability structures are orange, and low-vulnerability structures are green. Those
that are vulnerable to sea-level rise may require the addition of pump stations in place of gravity-driven
control structures.
SOURCE: SFWMD (2009a).

Biscayne Bay Coastal Wetlands, Phase 1
The Biscayne Bay Coastal Wetlands Phase 1 project is designed to rehydrate

coastal wetlands impacted by canal drainage and thereby improve salinity dis-
tributions in nearshore regions of Biscayne Bay (see Chapter 4). The impacts of
various sea-level-rise scenarios on project benefits are shown in Table 5-3. At
2 ft of sea-level rise (the high scenario for 50 years and intermediate scenario for
100 years), less than 50 percent of the overall project benefits to freshwater and
saltwater benefits are projected, although 88 percent of the nearshore salinity
benefits remain (USACE and SFWMD, 2012b). On the basis of these analyses,
planners concluded that project benefits over the 50-year planning horizon
were sufficient to recommend the project, noting that the project would “delay
future degradation of coastal wetland habitat caused by increased sea level
conditions by redirecting freshwater flows into critical habitat” (USACE and
SFWMD, 2012b). However, at the highest levels of sea-level rise considered over
a 100-year time frame, all of the project benefits are lost (Table 5-3). USACE
and SFWMD (2012b) state:
The effects of SLR on project benefits that occur after the 50-year project lifes-
pan should be treated the same as benefits that occur after the project lifespan.
In other words, effects that occur after the 50 year project lifespan should not
be considered for plan selection or determination of project viability.
While consistency of planning constraints seems reasonable, the project high-

lights the limitations of 50-year planning horizons in the context of climate
change.
Compared to the Picayune Strand Project, which represents a large area
that is likely to gradually transition from freshwater to brackish wetlands, the
Biscayne Bay Coastal Wetlands, Phase 1 project represents a narrow strip of
coastal wetlands that are restricted from migrating landward by the L-31E
levee and existing development (Figure 4-13). Thus, unlike Picayune Strand,
all project benefits are likely to be lost at extreme levels of sea-level rise, and
significant benefits are lost at likely levels of sea-level rise over the 21st century
(2 ft; Table 5-3). However, these findings represent rather simplistic analysis of
increments of sea-level rise overlain upon geographic information system maps,
with no modeling of salinity changes expected in groundwater or nearshore
areas of Biscayne Bay. Assessing the value of the project in the context of sea-
level rise necessitates a rigorous analysis of existing ecosystem conditions and
trends, the impacts of various sea-level-rise scenarios on project performance
measures, and the extent to which the project could mitigate the impacts of
sea-level rise.

TABLE 5-3 Projected Reduction in Biscayne Bay Coastal Wetlands Benefits by Component and
Ecozone Under Several Sea-Level-Rise Scenarios
Percent Percent
Reduction Reduction Percent
Estimated in in Reduction in
Percent Benefit Freshwater Saltwater Nearshore
Reduction at 3" Wetland Wetland Salinity
of SLR Benefits Benefits Benefits
Estimated Percent Reduction in Benefits with 3" of SLR *
Deering 0% 2% 0%
Cutler 0% 2% 0%
L-31E 0% 10% 0%
Deering 0% 4% 0%
Cutler 0% 4% 0%
L-31E 0% 20% 0%
Deering 0% 5% 0%
Cutler 0% 5% 0%
L-31E 0% 30% 0%
Estimated Percent Reduction in Benefits with 24" of SLR
Deering 100% 10% 0%
Cutler 100% 10% 0%
L-31E 50% 100% 25%
Estimated Percent Reduction in Benefits with at 68" of SLR
Deering 100% 100% 100%
Cutler 100% 100% 100%
L-31E 100% 100% 100%
* Reduction in benefits for SLR less than 1 ft were estimated by interpolating between the estimated
losses at 0 ft of SLR and 1 ft of SLR.
Florida Bay
Florida Bay provides a large-scale example of the implications of sea-level

rise for restoration. Florida Bay is a unique estuarine system with salinity deter-
mined by evaporation and precipitation as well as freshwater inputs. Water
management changes in the South Florida ecosystem over the past 60 years have
reduced freshwater inflows to the bay such that it can be seasonally hypersaline
in the middle parts of the bay (Figure 5-12; Kelble et al., 2007; Nuttle et al.,

FIGURE 5-12 Mean salinity distributions in Florida Bay, showing conditions of hypersalinity (c).
SOURCE: Kelble et al. (2007).
2000). In 1987, a widespread collapse of seagrasses occurred, which is generally

attributed to hypersalinity (Deis, 2011).
In general, the large-scale increases in sea level will cause Florida Bay to
become deeper and incorporate portions of the southern Everglades. Increases
in sea level of 2 ft (roughly the intermediate USACE local sea-level rise projec-
tion for 2100) would change the average depth of Florida Bay from 3 to 5 ft,
presumably causing a significant change in salinity (e.g., Monismith et al., 2002).
Sea-level rise could increase salinities throughout much of present-day Florida
Bay during the wet season but could decrease occurrences of hypersalinity
in central and northern Florida Bay through dilution. Any potential increase in
salinity is of concern to the restoration, given that one of the objectives of the
CERP is to reduce salinities in Florida Bay through increased freshwater inflow

to more closely mimic pre-drainage hydrology. However, the effect of sea-level

rise on Florida Bay salinity is significant enough that restoration goals will need
to be revisited under various sea-level-rise scenarios.
To fully evaluate effects of sea-level rise on the conditions and restoration of
Florida Bay, a fully three-dimensional (3-D) circulation model of the bay, such as
that described by Zheng and Weisberg (2012), is needed, albeit one coupled with
a regional hydrologic model. Given the importance of precipitation and surface
flows to Florida Bay, ideally such a model would also include a regional atmo-
spheric model (e.g., Maxwell et al., 2011). Neither existing empirical models of
salinity in Florida Bay (FATHOM) nor 2-D models properly depict the physics
associated with sea-level rise (see also NRC, 2002a).
Water Budget Implications
Although future climate conditions are uncertain, the seven hypothetical

scenarios presented previously in the chapter (see Implications for Everglades
Hydrology; Obeysekera et al., 2014) highlight the range of challenges that
Everglades restoration could face. Increases in rainfall represent the best-case
scenario for the ecosystem, with increases in water flow (Table 5-2, Figure 5-6b).
The combination of increased coupled evapotranspiration, decreased precipita-
tion, and rising sea level over future decades represents the worst-case scenario
among those modeled. Under this scenario, water levels would decline through-
out the system (Figure 5-6a) and unmet water supply demands from agriculture
and urban population centers would intensify existing conflicts over water
supply. Declining groundwater levels combined with sea-level rise would further
exacerbate saltwater intrusion, compromising urban water supplies (Figure 5-13)
and impacting coastal ecosystems. Although no modeling has been done to
quantify the effects of the CERP as currently planned under such a scenario, it
is possible that the benefits of the CERP would be surpassed by the negative
impacts of reduced precipitation and increased evapotranspiration. However,
the scenario of potential future decreases in flow associated with decreases in
precipitation and increased evapotranspiration amplify the urgency to accelerate
projects that increase storage and move more water southward to enhance the
resilience of the ecosystem under future conditions.
Given the competing demands of water supply, environmental restoration,
and flood control, it is clear that providing as much flexibility as possible to water
managers will be critical to the success of the CERP. In many systems, flexibility
of operation is achieved through water storage (e.g., reservoirs, aquifer storage).
For example, in the western United States, large reservoirs such as Lake Mead or
Lake Shasta buffer water supplies against interannual variations in precipitation
as well as reducing high flows so as to prevent floods. In South Florida, existing

FIGURE 5-13 Saltwater intrusion interface in Miami-Dade County, and proximity to water
supply well fields.

operational flexibility is achieved by operation of the extensive network of chan-

nels and structures and through storage in Lake Okeechobee, although the lake
is highly constrained by ecological and dam safety considerations. The CERP, if
fully constructed, would add substantial storage via aquifer storage and recovery
and several large reservoirs, but it is unknown whether this storage would be
sufficient to sustain the ecosystem under the worst-case climate scenarios.
In the face of possible changes in hydrologic conditions and sea-level rise,
increasing water storage to provide more reliable flow to the Everglades as
well as to water users could provide useful operational flexibility. For example,
maximizing the ability to capture and store water in surface-water reservoirs
during wet periods that would otherwise be discharged through the northern
estuaries would also provide water for environmental and human uses during
dry periods and increase groundwater recharge to mitigate salinity intrusion into
South Florida aquifers.
Implications on CERP Goals
As discussed in Chapter 2, the CERP generally aims to restore the “essential

hydrological and biological characteristics that defined the undisturbed South
Florida ecosystem” (33 CFR § 385.3). The Natural Systems Model has played
an important role in shaping restoration goals. However, under the worst-case
scenarios of climate change and sea-level rise in the Everglades, some CERP
goals may not be attainable, and others may need to be revisited considering sub-
stantially changed conditions under sea-level rise (e.g., Florida Bay). Although
the CERP, as finalized in 2000, did not incorporate climate change effects on
restoration outcomes, there is broad recognition in the research and manage-
ment communities of the multiple facets of climate change and their impacts
(Aumen et al., in press). Estimates of sea-level rise and downscaled climate
projections and their ecological impacts will continue to be refined with future
research and incorporated into management planning. Restoration planners will
need to continuously revisit whether pre-drainage hydrologic and ecological
targets still provide useful restoration goals, and to develop alternative goals in
light of what is feasible and sustainable under future conditions. Meanwhile,
changes in temperature and precipitation could reduce regional water avail-
ability, necessitating additional attention to water sustainability for built and
natural systems to address potential water supply challenges unforeseen when
the CERP was originally developed.
Literature on climate change is replete with studies of probable and possible
impacts of climate change on water resources in various geographic regions.
Work of that kind in South Florida is impressive. Although there are many recom-
mendations to incorporate effects of climate change in water resource planning,

far fewer in-depth studies that suggest how that can be accomplished are available
than the assessments of impacts of climate change. Among the more complete
publications addressing adaptation is in the context of water management in
California, especially the Climate Change Handbook for Regional Water Plan-
ning (CDM, 2011) developed for the California Department of Water Resources,
USACE, EPA, and Resources Legacy Fund and the 2009 California Climate
Adaptation Strategy (California Natural Resources Agency, 2009). Hanak and
Lund (2012) and the National Research Council (NRC, 2012c) discussed progress
toward incorporating climate change in managing California’s water resources.
Those documents describe an array of management options to adapt to
climate change. Broad categories such as aggressive pursuit of water use effi-
ciency, enhancement and restoration of ecosystems, increased storage, improved
conveyance and transfers, management of land use, and optimization of system
operations are followed by more detailed measures. Many of those actions have
already been taken in South Florida, some of which are directly targeted by CERP
and non-CERP projects. Miami-Dade, West Palm Beach, and other urban areas
have adopted progressive increasing block-rate pricing and other conservation
measures to manage demand for public water supplies. Much of the degrada-
tion of the Everglades and demands on its services are due to external forces.
Although development of a comprehensive strategy to adapt to climate change
in South Florida is beyond the scope of the CERP, a more complete strategy for
restoration of the Everglades and adaptation to climate change would need to
address management of demand and supply for water and related land resources
within and external to the Everglades ecosystem. Ongoing research through the
5-year South Florida Water, Sustainability, and Climate Project5 led by Florida
International University and funded by the National Science Foundation may
help inform such planning.
PLANNING TO ADDRESS CLIMATE CHANGE
CERP planners are just beginning to address climate change impacts. USACE
project planning guidance (USACE, 2011e) specifically includes a method for
estimating sea-level rise in project design, and the USACE requires analysis of
three sea-level rise scenarios for all Civil Works projects (discussed earlier for
Biscayne Bay Coastal Wetlands and Picayune Strand). To date, however, such
analysis has been limited to the project level, where sea-level rise is used to
design coastal structures and adjust project benefits through simple analyses of
land lost due to inundation under different sea-level rise scenarios (Figure 5-14).
Early analyses primarily examined changes in benefits over 50 years (USACE
5
See http://sfwsc.fiu.edu/.

FIGURE 5-14 Projected impact of sea-level rise on overall habitat improvements (in habitat units) provided
by the Central Everglades Planning Project under different scenarios (see also Chapter 3 and Figure 5-4).
The analysis considers the reduction in overall project-derived benefits due to seawater inundation of
freshwater wetlands in the project area.
and SFWMD, 2012b), but more recent analyses considered changes in benefits
over 100 years—a more useful time horizon in the context of climate change,
sea-level rise, and restoration investments of the magnitude of the CERP (USACE,
2013e; USACE and SFWMD, 2013b). To the best of the committee’s knowledge,
models capable of computing salinity fields in surface waters and groundwater
have not been used in the CERP to assess the effects of sea-level rise on s alinities.
Nonetheless, suitable models (e.g., Langevin et al., 2005) currently exist that
could be coupled to surface-water models.
The USACE-required sea-level-rise analyses typically are performed at the
end of the planning process, after the desired project alternative has been
selected. However, CERP project planning would benefit from broader incor-
poration of climate and sea-level-rise scenarios during project development.
Despite uncertainty associated with climate projections, more rigorous scenario
planning (NPS, 2013; Peterson et al., 2003; Polasky et al., 2011) provides a
framework for project-level decision making. Such planning would consider

the effects of climate change on performance measures for various alternatives,

quantify the specific benefits associated with mitigating the impacts of climate
change, and identify the costs and benefits of additional flexibility to address
climate change uncertainties (see also Chapter 3).
System-Level Considerations
NRC (2008) emphasized that in light of climate change, restoration efforts

“are even more essential to improve the condition of the South Florida ecosystem
and strengthen its resiliency as it faces additional stresses in the future.” How-
ever, this perspective does not imply that restoration should continue unchanged.
To date, as far as the committee is aware, CERP planners have not made any
major adjustments at the systemwide level in light of climate change and sea-
level rise. Accurate precipitation or temperature projections are not available
that can serve as a basis for major changes in the CERP because of uncertainty
associated with future storylines driving greenhouse gas emissions and land
cover, GCM simulations, and regional downscaling. However, sea-level-rise
projections are available that should form the basis for project prioritization,
considering project benefits in the context of sea-level-rise and the potential for
the CERP to delay or mitigate sea-level-rise impacts. Existing scheduling and
prioritization as reflected in the Integrated Delivery Schedule are largely driven
by project authorizations (see Chapter 4), but wise expenditures of funds neces-
sitates periodic reassessment of the priority of previously authorized projects in
light of improved understanding of future conditions.
Considering the uncertain projections of future precipitation and tempera-
ture, systemwide analysis across an array of future scenarios is critical. Current
hydrologic and ecosystem modeling for the CERP is based solely on histori-
cal hydrology and does not address potential effects of long-term changes in
temperature, precipitation, evapotranspiration, and sea-level rise. Additional
Everglades climate-sensitivity analyses are needed to anticipate challenges, iden-
tify potential contingencies and system flexibilities to mitigate climate-related
changes, and target additional research efforts toward critical uncertainties. The
recent scenario analysis by Obeysekera et al. (2014) provides a useful first step,
although the scenarios selected were purposefully simplistic. More recently,
Swain et al. (2014) used a hydrodynamics model to examine the effects of
changes in sea level and precipitation on freshwater flows, surface-water salinity,
and inundation within Everglades National Park and coastal areas to the east.
Additional modeling of this kind is needed to analyze the sensitivity of the South
Florida ecosystem to seasonal and long-term precipitation and temperature vari-
ability to enhance the understanding of possible climate impacts. This modeling
should be used to examine the possible effects of climate change on Everglades

hydrology and ecology with and without the implementation of CERP. Scenario-
based modeling could also be used to explore the impacts of specific projects
to identify which projects are most resilient to climate change and/or which
projects mitigate the impacts of climate change under a range of scenarios. These
model simulations should be refined as new data become available that reduce
uncertainties in model forcings and hence model predictions.
Modeling Tools
CERP planning in the context of sea-level rise and climate change may
require the development of new modeling tools or the improvement and appli-
cation of existing tools. Improved salinity modeling tools are needed, such as
3-D circulation models for the major estuaries coupled to regional hydrologic
models, as discussed previously in this chapter. Understanding salinity intrusion
in coastal wetlands and aquifers used for urban water supply requires a surface-
water flow model coupled with a variable-density groundwater flow model
(e.g., Tides and Inflows in the Mangroves of the Everglades [TIME] v. 2.0). The
TIME model domain includes the terrestrial areas of Everglades National Park,
its coastal mangrove zones to the south and east of the park, and the northern
edge of Florida Bay, but it does not include Biscayne Bay. Currently, the model
is being refined for assessing sea-level rise and restoration alternatives in Ever-
glades National Park (Bahm and Fennema, 2013). For variable density modeling
of saltwater intrusion, the SFWMD uses a coupled MODFLOW/SEAWAT model
(Harbaugh et al., 2000; Langevin et al., 2003; Restrepo and Montoya, 2008).
Sea-level rise will lead to landward shifts of the interface between fresh
groundwater and saltwater, thereby increasing the potential of saltwater intrusion
(Figure 5-13). Sea-level rise will also increase groundwater levels, as the aquifer
system responds to new conditions along its seaward boundary. This water-table
adjustment will be greatest along the coast, but will propagate inland. Because
groundwater and surface-water systems are tightly connected in South Florida,
rising water tables will alter the rates and volumes at which excess water can be
conveyed through canals and natural waterways of the Everglades. The ability to
forecast these sea-level-induced changes in groundwater flow and groundwater/
surface-water interactions is requisite to informing the design and operation of
water control infrastructure under the CERP.
IMPORTANCE OF EVERGLADES RESTORATION

IN CONTEXT OF CLIMATE CHANGE
Systems such as the Everglades that are highly vulnerable to the impacts of
climate change are likely to benefit from management strategies that incorporate

maximizing ecological resilience (or the capacity to respond to environmental

changes) as a goal (Millar et al., 2007; Tompkins and Adger, 2004). The ability
of these systems to respond to the impacts of climate change depends upon
their capacity to rebound from disturbance as well as to respond to chronic and
gradual environmental changes (Gunderson, 2000). Long-term persistence of
ecological functioning depends upon maintenance of adequate source popula-
tions in the landscape. For example, unimpaired coastal wetland ecosystems
can respond to sea-level rise through a combination of biological (e.g., organic-
matter accumulation) and physical processes (e.g., sediment accretion from
storm surges). Where rate of environmental change exceeds the capacity for spe-
cies to remain in place, lateral migration, regulated by dispersal, and availability
of suitable habitat, is necessary. In the Everglades, restoration of hydrologic
regimes will support such processes, helping to perpetuate diversity, ecosystem
functioning, and fluidity of the landscape as climate envelopes of species shift
(Manning et al., 2009).
In the face of climate change, Everglades restoration will increase the
resilience of the ecosystem and the water management system and decrease
their vulnerability. From the perspective of water resources management, the
CERP may offer substantial benefits. In particular, increasing surface-water
flows through water conservation areas and into Everglades National Park
may help mitigate the sea-level-rise-induced salinization of the aquifers that
provide water supply for Dade, Broward, and adjoining counties. This issue
would benefit from modeling to better characterize and quantify the scope of
potential benefits.
Climate change provides a strong incentive for accelerating restoration.

Current impacts of rising sea levels are a harbinger of future climate change
effects on the functioning and structure of the Everglades ecosystem and the eco-
system services on which South Florida depends. Sea-level rise in South Florida
is already increasing saltwater intrusion into Everglades freshwater habitats and
urban water supplies, and future climate changes are likely to be manifested
through changes in the timing, volume, and quality of freshwaters; distributions
of species; and the extent of wetland habitats. Climate change is also expected
to increase agricultural water demands, which when paired with anticipated
population growth, highlights the potential regional water supply challenges in
South Florida under future scenarios. Everglades restoration enhances the ability
of the ecosystem to withstand and adapt to future changes, and increases water
availability to the ecosystem and to urban and agricultural users. Improvements
in Everglades water depths promote higher rates of peat accretion that could

help mitigate the effects of sea-level rise and reduce the impacts of saltwater
intrusion on urban water supplies.
Although the projections are uncertain, significant changes in precipitation
and temperature coupled with increasing sea level have important implications
for the CERP. The Everglades landscape is especially sensitive to sea-level rise,
and rates of sea-level rise in South Florida are predicted to increase. A scenario
of 1.5-degree increase in temperature and a 10 percent decrease in precipitation
together with anticipated sea-level rise results in significant changes in coastal
ecosystems and insufficient freshwater to sustain the natural and built systems. To
decrease uncertainty associated with precipitation projections and clarify future
risk, global climate model projections of intra-annual, annual, and interannual
variability in precipitation and temperature need to be improved and refined.
These improved climate projections should, in turn, be used by CERP planners
as input to drive Everglades hydrologic models suitable for making inferences
on year-to-year and seasonal variations in freshwater availability.
Climate change is not adequately considered in the CERP planning process
and should be integrated into future ongoing analysis and monitoring. CERP
projects are designed based on historical hydrology and have not been assessed
in the context of future precipitation and evapotranspiration scenarios. Currently,
only sea-level rise is considered in CERP planning and usually only as a cursory
analysis at the end of the process to assess loss of benefits through 2050 with
wetland inundation resulting from sea-level rise. The lack of consideration of the
effects of climate change paints an incomplete picture of hydrologic and eco
system response to the alternatives examined and ignores the potential benefits of
the projects to help mitigate the impacts of climate change. Additionally, hydro-
logic restoration goals are based on the natural systems model, which reflects
the past 50 years rather than any likely future. Depending on future climate
change, some hydrologic or ecological restoration goals may be unattainable
or prove to be not cost-effective. Urban and agricultural water demands unmet
under dire climate scenarios highlight the need for additional analysis of water
sustainability for the natural and built systems.
CERP planners should consider the implications of sea-level rise and poten-
tial hydrologic changes in systemwide planning and project prioritization.
Likely sea-level-rise projections can be used to evaluate future project benefits,
considering uncertainties regarding the potential for accretion in coastal and
inland wetlands to mitigate these effects. Sea-level-rise scenarios should also be
coupled with hydrologic change scenarios to characterize systemwide response
to global change. The outcome of these analyses would inform future system-
wide decisions of project prioritization. Re-prioritization should include consid-
eration of both those rendered less important and effective in light of reduced
benefits in the context of climate change and sea-level rise and those projects

that become more essential to enhance the ability of the ecosystem and the built
environment to adapt to changes and mitigate the effects of changing climate.
Anticipating future changes in temperature, precipitation, and sea-level
rise, CERP planners should, where feasible, design for flexibility. Climate change
needs to be incorporated into adaptive management planning, at the project
scale as well as when considering systemwide goals. It is likely that additional
water storage will be needed to address anticipated future increases in variability
of meteorological conditions. As new knowledge becomes available, it needs to
be incorporated into the CERP adaptive management framework so that man
agers can adjust future restoration efforts appropriately as the nature of changes
in climate become more evident. In addition, the current monitoring program
should be evaluated to ensure that important effects of climate change will be
characterized and quantified.
The committee identified several high-priority research needs related to
climate change and Everglades restoration:
• Assess the rates of peat/sediment accretion and subsidence in coastal and

inland freshwater wetlands in the context of sea-level rise;
• Improve modeling tools that can be used to assess the effects of projected
sea-level rise on groundwater supplies and coastal ecosystem functioning, and
examine the potential for the CERP to mitigate these effects;
• Improve, refine, and evaluate downscaled climate model projections in
the context of South Florida water resources and Everglades restoration;
• Improve the understanding of factors that could help maintain the diverse
mosaic of Everglades habitats and increase their resilience amid changes in
climate and sea level; and
• With improved climate and sea-level projections, reevaluate the goals for
Everglades restoration and develop alternate goals as appropriate.

Biological Invasions and

Everglades Restoration
Invasions by nonnative species can threaten Everglades restoration, displac-

ing native species and transforming large expanses into ecosystems that differ
radically from their historical structure, functioning, and provision of ecosystem
services. Melaleuca (Australian paperbark, Melaleuca quinquenervia), Brazilian
pepper (Schinus terebinthifolius), Australian pine (Casuarina spp.), and Old
World climbing fern (Lygodium microphyllum) together infest hundreds of thou-
sands of acres of the Everglades region and foster frequent, hot fires that destroy
native plants in the “River of Grass” and tree islands, facilitating dominance
by exotic vegetation (Schmitz et al., 1997). The Burmese python has quickly
become the top carnivore in the Everglades food web, eating not only alliga-
tors but virtually all vertebrates it can reach. Its invasion of the Everglades has
coincided with ~90 percent declines in populations of bobcats, raccoons, and
opossums and a 100 percent decline of rabbits (Dorcas et al., 2012).
The extent to which invasions hinder restoration depends on how restora-
tion is defined. As noted in Chapter 2, different agencies define “restoration”
differently, as is revealed in the specific goals. The South Florida Ecosystem
Restoration Task Force (Task Force) describes three goals for Everglades resto-
ration, of which the second is “Restore, preserve, and protect natural habitats
and species” (SFERTF, 2000). The Comprehensive Everglades Restoration Plan
(CERP) defines the goal of Everglades restoration as “restoration, preservation,
and protection of the South Florida Ecosystem while providing for other water-
related needs of the region, including water supply and flood protection.” The
Programmatic Regulations that guide implementation of the CERP state that
the desired hydrologic and biological characteristics include resilient plant
communities and an abundance of native wetland animals (DOI and USACE,
2005). Substantial establishment by nonnative species is certainly incompatible
with the Task Force goal, and establishment of at least some highly aggressive
nonnative species is incompatible with the CERP goal. A senior DOI official
succinctly summarized the problem: “an Everglades landscape teeming with
169
exotic species is not a restored Everglades” (S. Estenoz, DOI, personal com-
munication, 2013).
This chapter details the status and trends of nonnative species invasions of
the Everglades ecosystem, discusses some of the challenges for managing these
nonnative species in the context of Everglades restoration, and suggests ways of
addressing these challenges.
EVERGLADES INVASIVE SPECIES AND THEIR IMPACTS
Invasive species are increasingly common around the globe, and the impacts
of nonnative species are quite variable because they depend upon characteris-
tics of the species itself and the ecosystem it invades. Ecological changes induced
by these invaders range from no immediately discernible impacts to dramatic
effects limited to particular native species or specific groups of them, to broad-
scale habitat transformation with attendant changes in ecosystem structure and
functioning. This section reviews the effects of invasive species in the Everglades.
Some biological invasions affect particular native species or specific groups
of them. For instance, the newly invading redbay ambrosia beetle (Xyleborus
glabratus; Figure 6-1) attacks a native plant species, swamp bay (Persea palustris),
infecting trees with the deadly laurel wilt fungus (Raffaelea sp.) (Fraedrich et al.,
FIGURE 6-1 Laurel wilt damage to swampbay trees on an Everglades tree island (left), which is caused by
the laurel wilt fungus carried by the nonnative redbay ambrosia beetle (right). Note the scale: the adult
insect is less than 3 mm long.
SOURCES: Photographs courtesy E. Allen, SFWMD (left) and J. A. MacGowan, Mississippi Entomological
Museum (right).

Biological Invasions and Everglades Restoration 171
2008; Rodgers et al., 2014a). Ecological impacts may spread beyond swamp
bay to species that depend on it for food, habitat, or other services (Chupp and
Battaglia, 2014; Kendra et al., 2013). Other trees in the laurel family (Lauraceae),
such as avocado, are also at risk (Mayfield et al., 2008; FDACS, 2012).
Other biological invasions affect ecosystem processes (such as fire regimes
or nutrient or hydrologic cycles) or ecosystem structure, so they automatically
affect many native species. For instance, melaleuca and Old World climbing
fern both greatly affect the fire regime in parts of the Everglades. The altered fire
regime then adversely affects marshes, sawgrass prairies, and tree islands, with
follow-on effects on species that may occupy these habitats, such as epiphytes
and various rare plants (Serbesoff-King, 2003).
Further complicating both understanding and management of nonnative
species is the fact that they can interact synergistically with one another to
exacerbate impacts, a phenomenon known as invasional meltdown (Simberloff
and Von Holle, 1999). For example, nonnative fig trees were present for a cen-
tury in South Florida as ornamentals, restricted largely to residential areas and
not an invasive problem in the Everglades until the 1990s. This restriction was
caused by the fact that each fig species can be pollinated only by a particular
fig wasp species, and the nonnative fig species present in Florida lacked their fig
wasps. However, beginning in the 1980s, the nonnative fig wasps of three of
these fig species arrived (by unknown means) in Florida, and at least one of
the figs, Ficus microcarpa, is now an invasive pest in parts of the Everglades
(Kauffman et al., 1991).
The case of the figs and fig wasps also exemplifies another complication in
assessing invasion threats and prioritizing management activities—many intro-
duced species that ultimately become widespread, highly damaging invaders
can remain restricted and innocuous for an extended period, even for decades,
before spreading across the landscape (Crooks, 2011). In South Florida, Brazilian
pepper was introduced at least as early as the 1880s but did not explode across
the region until the 1950s (Ewel, 1986). Melaleuca was similarly present long
before it became a major vegetation element (Ewel, 1986). For Brazilian p epper
and melaleuca, the reason for the lag between introduction and widespread
invasion is not known with certainty. The rather sudden explosion of Brazilian
pepper in the 1950s might even have resulted from hybridization between two
separate earlier introductions that differed genetically (Mukherjee et al., 2012).
However, whether the reasons for a lag and its termination are known (as with
the fig) or not (as with Brazilian pepper and melaleuca), the frequency of such
invasion lags cannot be doubted (Crooks, 2011). These lags, in turn, challenge
attempts to classify a newly discovered introduction as likely innocuous or
potentially threatening.
Invasion of nonnative species is often accompanied by structural, func-

tional, and biogeochemical changes in the community. Invasion by melaleuca

has transformed prairies of sawgrass and muhly grass into forests almost wholly
composed of melaleuca (Bodle et al., 1994), while Australian pine (Casuarina
spp.) dominates formerly treeless beaches (Schmitz et al., 1997). Invasion by
these plants affects the native animal community; for example, Australian pine
interferes with nesting of endangered sea turtles (Wheeler et al., 2011). Similarly,
spread of the Burmese python (Python bivittatus) threatens the entire food chain,
challenging even the American alligator for the apex position. This top preda-
tor is spreading quickly, leading to dramatic declines in mammal populations
(Dorcas et al., 2012), and native bird populations may also be at risk (Dove et
al., 2011). Table 6-1 highlights several potential impacts of invasive species on
CERP performance measures in the Greater Everglades region.
Following their removal, some of those changes linger, leaving legacies of
unknown longevity and impacts. For example, substantial remains of melaleuca
trees persist on tree islands and other areas in the Everglades where trees were
girdled and died in place (Figure 6-2). Such legacies may negatively influence
subsequent regeneration of native species and/or facilitate reinvasion of non-
native species (D’Antonio and Meyerson, 2002). Areas with seedbanks of per-
TABLE 6-1 Examples of Potential Invasive Species Threats to CERP Performance Measures
Performance Measure Invasive Species/Guild Potential Threat or Risk
American Alligator Argentine black and white tegu Reduced reproduction due to egg and
Distribution, Size, Nesting Nile Monitor hatchling predation
and Condition Burmese python Direct competition for food resources
Direct predation by pythons
American Crocodile – Argentine black and white tegu Reduced reproduction due to egg and
Juvenile Growth and Survival Nile monitor hatchling predation
Burmese python Direct competition for food resources
Direct predation by pythons
Marl Prairie Cape Sable Melaleuca Degradation nesting habitat due to
Seaside Sparrow Habitat Australian pine changes in plant community structure and
fire regimes
Prey-Based Freshwater Fish Nonindigenous freshwater fish Reduced native small dish density due to
Density Performance Measure predation or competitive interactions
Ridge and Slough Melaleuca Alteration of plant community structure,
Community Sustainability Australian pine microtopography, and fire regimes
Old World climbing fern
Brazilian pepper
Wet Prairie Melaleuca Displacement of native plant community
Australian pine Alteration of fire regimes
Loss of wildlife habitat
SOURCE: RECOVER (2014a).

FIGURE 6-2 Stands of melaleuca in the Everglades, after control efforts using ground-based herbicide
application.
SOURCE: Photograph courtesy of D. Policansky, National Research Council.
sistent, viable propagules of nonnative species are particularly vulnerable to

reinvasion. Unfortunately, very little is known about these legacies and potential
impacts on restoration success in the Everglades. Sites where nonnative species
have been removed may remain altered by their legacies and require additional
rehabilitation to achieve long-term restoration goals.
In many cases, the effects of invaders are currently unknown. For example,
nonnative fishes are quite abundant in certain areas, exceeding 6 percent at
12 sites in Everglades National Park (RECOVER, 2014a). Yet no studies have
documented and quantified their effects on native fish species or on various

CERP performance measures (e.g., freshwater fish density, wading birds nesting
and foraging).1 Once established, removal of nonnative fishes is “problematic,”
although options may exist to prevent future introductions or limit their spread
(RECOVER, 2014a).
In this section, some of the more significant Everglades invaders are dis-
cussed, focusing on some of the most damaging ones or ones with unknown
potential to be damaging, to illustrate the types of impacts that can occur.
The types of treatments that have been used to manage them and the efficacy
of available treatments are summarized. Appendix B contains descriptions of
other invasive species in the Everglades, and extensive lists of invasive plants
and animals are provided in the 2014 System Status Report (RECOVER, 2014a).
Invasive Plants
Many nonnative plants are pervasive pests in the Everglades (reviewed

in Junk et al., 2006). There are approximately 250 nonnative plant species in
aquatic, wetland, and terrestrial habitats of the Everglades, constituting 16 per-
cent of the flora (Long, 1984). Currently, 75 species are listed as priorities for
control by the South Florida Water Management District (SFWMD), with a par-
ticular emphasis on species able to displace natives and transform ecosystem
structure and functioning (Rodgers et al., 2014a). Of the 75 priority species, 12
are considered particularly high priority because they are believed to threaten
the success of the mission of the SFWMD (Rodgers et al., 2014a), and four of
these are systemwide priorities, described in detail in the sections that follow.
Table 6-2 summarizes the status and trends of all 12 high-priority species, along
with two others that have been targeted for early detection and rapid response
(EDRR). Appendix B provides some additional details of invasion history and
impacts of these and other invasive species.
Melaleuca
Melaleuca is a highly invasive tree native to Australia, New Guinea, and

New Caledonia. It readily establishes in sawgrass prairies and tree islands (Davis
et al., 2005b), converting these communities into low-diversity forests with
highly altered structure and functioning (Schmitz et al. 1997, Serbesoff-King,
2003). Melaleuca can drive changes in soil chemistry, depth to the water table,
nutrient cycling, and perhaps most importantly, fire regime. Native sawgrass
communities are adapted to frequent, low-intensity fires, but when melaleuca-
dominated areas burn, the fires are much more intense as the essential oils in
1
See http://www.evergladesplan.org/pm/recover/perf_ge.aspx.

TABLE 6-2 High-Priority Invasive Plants in the South Florida Water Management District
Species Areal Extent Ecological Impacts Treatment Availability Systemwide Trends
Melaleuca 10,035 canopy Displaces native Integrated 36% decrease since
acres in ECISMA, vegetation, alters mechanical, 2010-2012
systemwide plant community herbicide,
structure, changes biocontrol—effective
soil chemistry, with continued
alters fire regime maintenance
management
Brazilian pepper 22,145 canopy Displaces native Mechanical, 16% increase since
acres in ECISMA, vegetation, reports herbicide, prescribed 2010-2012
systemwide of allelopathy burning (in low-
(chemicals density areas);
produced by expensive to
plant affect other manage successfully
vegetation), alters
fire regimes
Old World 5,927 canopy Displaces native Mechanical; 69% increase since
climbing fern acres in ECISMA, vegetation, alters herbicide effective 2010-2012
systemwide fire regime but affects nontarget
species; biocontrol
efforts promising
Australian pine 1,869 canopy Alters habitat Mechanical and Spatial distribution
acres in ECISMA, for nesting sea herbicide methods constant; 94%
systemwide turtles and small effective but require increase in canopy
mammals; limits repeat; biocontrol acres since 2010-2012
regeneration of agents under
native species development
Water hyacinth Acreage unknown; Displaces native Herbicide, biocontrol Unknown; monitored
significant aquatic species; in public waters;
infestations in clogs waterways treated as needed
Kissimmee Basin and and resources allow
Lake Okeechobee
Hydrilla Acreage unknown; Displaces native Herbicide effective Unknown; monitored
significant aquatic species; but recent detection in public waters;
infestations in clogs waterways of resistance; treated as needed
Kissimmee Basin and mechanical and as resources
Lake Okeechobee harvesting allow
Air potato Acreage unknown Shades and Herbicide; biocontrol Unknown
displaces native partially effective;
species other agents under
development
Shoebutton Acreage unknown; Shades and Mechanical cutting Unknown; difficult to
ardisia high densities in displaces native followed by detect aerially
southern Everglades species in a wide herbicide
and eastern part of range of habitats
ENP
continued

TABLE 6-2 Continued

Species Areal Extent Ecological Impacts Treatment Availability Systemwide Trends
Torpedograss 9,000 acres in Displaces native Herbicide; no Unknown; no
marshes of Lake species; readily biocontrol agents systemwide
Okeechobee invades disturbed approved monitoring
areas
Downy rose Unknown coverage; Displaces native Mechanical cutting, Unknown; difficult to
myrtle mostly coastal species herbicide; biocontrol assess aerially
counties agent under testing
and development
Cogongrass 6,900 acres in the Displaces native Herbicide, Unknown, appears to
SFWMD species in wide mechanical, have spread recently
range of habitats; prescribed fire; no
alters fire regimes biocontrol agents
and nutrient approved
cycling
Water lettuce Acreage unknown; Displaces native Repeat application Unknown; monitored
significant aquatic species; of herbicide; in public waters;
infestations in clogs waterways biocontrol agents treated as needed and
Kissimmee Basin and ineffective as resources allow
Lake Okeechobee.
Tropical American Restricted mainly to Displaces native Herbicide Unknown; some
water grass Lake Okeechobee aquatic species expansion on Lake
Okeechobee marsh
Black mangrove Limited and Not fully Mechanical EDRR reduced
contained understood; threat occurrence, likely to
distribution to diversity and eradicate
function of native
mangroves
Mile-a-minute Limited and Shades out and Herbicide EDRR reduced
contained blankets native occurrence,
distribution vegetation eradication may be
possible
NOTE: The acreage numbers presented here are compiled from multiple sources, some with sampling over the entire
South Florida Water Management District, which, at 4,662,000 ha, is much larger than the 728,000-ha Everglades Coop-
erative Invasive Species Management Area (ECISMA), representing all state and federal conservation lands within the
Everglades Protection Area, Miccosukee and Seminole lands, Broward County, Palm Beach County, and Miami-Dade
County. Sampling methods and time frames of measurement may also differ.
EDRR = early detection and rapid response; ENP = Everglades National Park.
SOURCE: Rodgers et al. (2014a,b); R. Johnson, NPS, personal communication, 2014.

the foliage become explosive, killing nearby sawgrass and many other native
members of the community (Center et al., 2012). Melaleuca seeds are liber-
ated by fire, and millions of seeds are released during the high-intensity fires.
Thus, melaleuca is a habitat transformer (Gordon, 1998). Rodgers et al. (2014b)
report that melaleuca infested over 40,000 acres of the Everglades Cooperative
Invasive Species Management Area (ECISMA; which consists of all state and
federal conservation lands within the Everglades Protection Area, Miccosukee
and Seminole lands, Broward County, Palm Beach County, and Miami-Dade
County) in 2010-2012, including more than 10,000 canopy acres.2
Melaleuca has been treated in many areas with mechanical removal, her-
bicide (Silvers et al., 2007), and most recently with several biocontrol agents—
natural enemies of melaleuca imported from its native range (Franks et al., 2006).
These insects reduce its seed production, growth, and density, and increase its
susceptibility to fire and herbicides, leading to greater native species diversity
in some targeted areas (Rayamajhi et al., 2009). Control efforts have been quite
successful, and abundance of melaleuca has been dramatically reduced to main-
tenance control levels in many areas of the Everglades (Figure 6-3; Center et al.,
2012; Rodgers et al., 2013), although not systemwide to date. With combined
control efforts, there was an estimated 36 percent decrease in canopy acres in
the ECISMA between 2010 and 2012 (Rodgers et al., 2014a). Untreated plants
can flower within a year of establishment, replenishing the seed bank, which
enables melaleuca to reinvade treated areas readily. Although biocontrol efforts
have slowed the rate of new invasions, frequent monitoring and retreatment are
necessary to achieve maintenance control of this species.
Brazilian Pepper
Brazilian pepper is highly invasive and widely distributed in the Everglades

(Ewe and Sternberg, 2002) with the highest spatial coverage of nonnative plant
species (see Figure 6-2). The growth form of Brazilian pepper is quite plastic, and
it can occur as a shrub, small tree, or even vine depending on environmental
conditions (Spector and Putz, 2006). On tree islands and other areas where it
dominates the canopy, understories support few if any native species (Rodgers
et al., 2013). It is highly fecund, producing thousands of seeds each year (Ewel et
al., 1982). Its rapid growth rates, vigorous sprouting capacity, and reported a bility
to produce chemicals that inhibit other plant species (Morgan and Overholt,
2005) enhance its capacity to displace native species and become dominant.
2
Canopy acres represent the area of ground covered by foliage of a particular invasive species.
Infested area is defined as the acreage encompassed after drawing a line around the perimeter of
the areas of infestation (the canopy cover of the plants) excluding areas not infested (Price, 2009).

FIGURE 6-3 Areas with melaleuca infestations, 1995-2010. Darker shades indicate denser
coverage of melaleuca.
SOURCE: J. Eckles, Florida Fish and Wildlife Conservation Commission, and L. Rodgers, SFWMD,
personal communication, 2013.

If allowed to continue to spread across the Everglades, Brazillian pepper could

significantly affect CERP performance measures, such as ridge-and-slough sus-
tainability (RECOVER, 2014a).
Fire can be used to manage low densities of Brazilian pepper. However,
in high-density stands, fire has much less impact, and Brazilian pepper can
drive a fire suppression feedback that leads to further invasion (Stevens and
Beckage, 2009). Brazilian pepper is particularly difficult to control in nutrient-
rich conditions. For example, the Hole-in-the-Donut within Everglades National
Park had perhaps the most expansive infestation of Brazilian pepper. This area
was heavily disturbed and had elevated soil nutrient availability from previous
farming practices (Li and Norland, 2001). Restoration required removing the
soil substrate, which had been highly modified by rock plowing (Ewel, 2013),
down to the bedrock to promote reestablishment of native vegetation (Smith et
al., 2011). Mechanical removal and herbicides can also control this species to
some degree. In the Picayune Strand CERP project, massive mortality of Brazilian
pepper occurred followed flooding of plot PC26 (RECOVER, 2014b). Biological
control agents have been tested, but to date, none have been released. Currently,
a gall-producing potential biocontrol agent is under testing and development.
Brazilian pepper is not under maintenance control and is still spreading. This
species infested 74,225 acres in the ECISMA area and there was an estimated
16 percent increase in canopy acres during the period 2010-2012 (Rodgers et
al., 2014a).
Australian Pine
Three species of Australian Casuarina are highly invasive in the Everglades.

Casuarina equiseitfolia is the most common and widely distributed of the three
and threatens coastal areas and beaches because it tolerates arid conditions
and saline soils with limited fertility. It is limited by long hydroperiods, occur-
ring mostly on better drained soils and in some short-hydroperiod sawgrass
habitats. C. glauca and C. cunninghamiana are often found on disturbed sites in
upland habitats adjacent to coastal communities. Left unchecked, C. equisetifolia
can dramatically alter the structure and functioning of many coastal areas of
the Everglades. For example, the root structures and fallen “needles” of the
Australian pine on invaded beaches inhibit nesting by loggerhead and green sea
turtles (Wheeler et al., 2011). Also, the heavy litter layer that develops under
dense Australian pine canopies can impede regeneration of other plant species.
Mechanical removal has proven difficult because these species have tremen-
dous sprouting capacity. Controlled burning is not effective (Doren and Jones,
1997), and herbicide applications are expensive. An effective biological control
agent is being sought (Wheeler et al., 2011); recent discovery that Australian

pine species are hybridizing in the Everglades may further complicate this
search. Australian pine is currently relatively low in abundance compared with
melaleuca and Brazilian pepper (Figure 6-4). Australian pine is considered to
be at maintenance control levels in most areas of the Everglades (Rodgers et al.,
2013). During 2010-2012, Australian pine infested 10,325 acres in the ECISMA
area. Although its spatial distribution has remained relatively constant, there was
an estimated 94 percent increase in canopy acres during this period (Rodgers et
al., 2014a). This dramatic increase in abundance indicates that monitoring and
control remain warranted.
Old World Climbing Fern
Old World climbing fern is native to Africa, southeast Asia, and Australia
and is highly invasive in the Everglades (Volin et al., 2004). Its tiny, numerous
spores disperse readily, and it poses a great risk to upland, marsh, and coastal
habitats alike. Rapid growth where light is not limiting, as in canopy gaps (Lynch
et al., 2011), and its twining growth form enable it to cover whole forest stands
rapidly (Figure 6-5). Once established vertically in the stand, it alters the fire
regime by extending “flame ladders” into the canopy not normally exposed dur-
ing low-intensity ground fires. Burning fern mats can be dislodged and carried
long distances to ignite new outbreaks. Between 1995 and 2010, Old World
climbing fern expanded from 1 percent to 10 percent of the ECISMA (F. Laroche,
SFWMD, personal communication, 2013). Figure 6-6 shows a lower rate of
continued expansion between 2003 and 2013.
Successful chemical control requires contact of the herbicide with all foliar
surfaces, and repeated applications of herbicide are often necessary. Because
the fern is most commonly found twining around other species, herbicide
application may have undesirable effects on surrounding nontarget native spe-
cies (Hutchinson and Langeland, 2012). Several biocontrol agents have been
introduced, and others are still in development. The most successful one to date
has been the brown Lygodium moth (Neomusotima conspurcatalis), which was
introduced in 2008 and is now established in the field. Its larvae can radically
reduce coverage of the fern. Despite these control efforts, this highly aggressive
invader remains widespread and is not currently under maintenance control
(Figure 6-5; Rodgers et al., 2013). Old World climbing fern infested 24,619
canopy acres in the ECISMA during 2010-2012 and has increased 69 percent
in canopy cover since that period (Rodgers et al., 2014a).
Invasive Animals
At least 192 nonnative animal species are established in the Greater Ever-

Figure 3. FIGURE 6-4

Distribution Aerial
and extent
abundance of melaleuca,
of Australian pine (a),Brazilian
Brazilian pepper,
peppertreeAustralian pine,
(b), melaleuca andOld
(c), and Lygodium in fern (d)
Word climbing
in the Florida
theEverglades. Values
Everglades. Therepresent
colors proportion
representofthegridpercentage
cell occupied of
by mapped
target species.
polygons within the grid
cell containing the species.
SOURCE: Rodgers et al. (2014b).
Rodgers et al.: Everglades invasive plant mapping N 367

FIGURE 6-5 Old World climbing fern completely blanketing a tree island in the Arthur R. Marshall Loxa-
hatchee Wildlife Refuge.
SOURCE: Photo courtesy of Tony Pernas, National Park Service.
glades. Particularly noteworthy species are discussed below, with information

on these and other species listed as high priority for management (Rodgers et al.,
2014a) given in Table 6-3. These and several other noteworthy animal invaders
are detailed in Appendix B. Those that have drawn particular attention (some
of which have been the focus of specific management efforts) tend to be either
large, flashy predators such as the Burmese python or Nile monitor (Varanus
niloticus) or insect species, such as the redbay ambrosia beetle, that attack or

FIGURE 6-6 The spread of Old World climbing fern between 2003 and 2011. Darker shades indicate more
dense coverage.
spread pathogens to plant species of special concern. Even for these nonnative
animal species, the impact on Everglades species and ecosystems cannot be
determined quantitatively without intensive research.
Whereas certain plant species (e.g., Brazilian pepper, melaleuca) overgrow
vast areas, so that at least some aspects of their impact are readily evident, ani-
mal impacts are generally not as obvious. Even if one sees a Burmese python
eating an alligator or a Mexican bromeliad weevil (Metamasius callizona) eat-
ing a bromeliad, the impact on the population of the prey or host species can
be determined only by substantial research. The great majority of established
nonnative species have not been studied in detail in the Everglades. Therefore,
great caution is warranted in determining which nonnative species pose threats

TABLE 6-3 Noteworthy Invasive Animals in South Florida

Treatment Priority
Species Areal Extent Ecological Impact Availability Trends Statusa
Burmese Spreading Depresses Licensed hunting, Unknown, Priority
python northward populations of many extremely limited population likely
beyond prey species success increasing within
Alligator Alley range
Argentine Established and Attacks many prey Traps Population Priority
black and spreading in species; population increasing within
white tegu Dade impact unknown range
Nile monitor Several areas of Likely predator but Snares, traps, Unknown Priority
South Florida impacts unknown hunting
Spectacled Dade and Likely predator but Hunting Unknown Priority
caiman Broward impacts unknown
Wild hog Throughout Greatly disturbs Hunting and Unknown Priority
region vegetation by trapping, but limited
rooting; may prey on because valued
accessible eggs and game animal
animals
Feral house Throughout Attacks many prey Trapping, but limited Unknown
cat region species, including because cannot use
birds, mammals, lethal means
reptiles, and
amphibians
Lionfish Entire Atlantic Attacks and locally Spearing on single Rapidly
and Gulf coasts eliminates reef fish corals; no regional increasing
of Florida management abundance
methods
Redbay Throughout Vectors laurel wilt, None Unknown Priority
ambrosia region which devastates
beetle redbay and swamp
bay
Gambian Grassy Key Unknown Trapping, but unable Stable Priority
pouched rat to eradicate because
cannot trap on
private property
Northern Small region of Unknown Intense hunting Unknown Priority
African Dade County
python
Oustalet’s Small region of Unknown Intense hunting Unknown Priority
chameleon Dade County
Veiled Lee and Dade Unknown Hunting Unknown Priority
chameleon
continued

TABLE 6-3 Continued

Treatment Priority
Species Areal Extent Ecological Impact Availability Trends Statusa
Cuban tree Throughout Unknown None Unknown Priority
frog region
Cane toad Throughout Unknown None Unknown
region
Purple Entire region Aggressive and eats None Unknown Priority
swamp hen except possibly eggs and young of
west coast waterfowl
Asian Miami and Unknown Electrofishing or Unknown, but Priority
swamp eel Tampa regions; toxicants possible in common
Everglades isolated areas
National Park
Mayan Southern part Unknown Electrofishing or Unknown
cichlid of region toxicants possible in
isolated areas
Pike killifish Much of region Unknown Electrofishing or Unknown
toxicants possible in
isolated areas
Black acara Southern part Unknown Electrofishing or Unknown
of region toxicants possible in
isolated areas
Island apple Throughout Believed to None Unknown Priority
snail region outcompete native
apple snail; may aid
snail kite
Giant Miami Eats wide variety Hand collecting; Slated for Priority
African land of cultivated and poison eradication
snail natural vegetation;
economic damage
Mexican Throughout Attacks native None Unknown Priority
bromeliad region bromeliads;
weevil threatens
populations of two
species
Rugose Dade County Attacks many plant None Unknown
spiraling and Florida species; population
whitefly Keys impact unknown
a
Included in a list of species prioritized for SFWMD management in Rodgers et al. (2014a).
SOURCE: Rodgers et al. (2013, 2014a).

and in predicting the nature and extent of those threats, particularly given the
subtlety and frequent delay of invasive impacts.
Several animal invaders that have attracted the most attention are presented
below; these are but a small fraction of recorded nonnative animals in South
Florida. In most cases, research has not been sufficiently extensive and detailed
to confirm the extent of the threats they pose, but in each case, existing observa-
tions and data suggest impact is likely great. Appendix B details the history and
potential impacts of several other notable invasive animals.
Burmese Python
The Burmese python is established in wide areas of the Everglades (Fig-

ures 6-7 and 6-8), although the population size, believed to be large, can be at
best estimated only with very wide confidence limits. It is believed that even
skilled herpetologists can detect at most 1 percent of those in areas they search,
FIGURE 6-7 A Burmese python in Everglades National Park.
SOURCE: Photograph courtesy of Catherine Puckett, U.S. Geological Survey.

FIGURE 6-8 Approximate distribution of Burmese pythons in South Florida from the 1990s to 2010, indicat-
ing rapid spread throughout the area.
SOURCE: Dorcas and Willson (2011).
and that the population in South Florida is in the tens of thousands (Dorcas and
Willson, 2013). By virtue of its massive size and position as top carnivore in
the food web, the python has attracted enormous attention in South Florida. A
precipitous decline in populations of many mammal species in the Everglades
was correlated with the arrival and spread of the Burmese python (Dorcas et al.,
2012), although population declines of nonprey species suggest other factors
may also have played roles (F. Mazzotti, University of Florida, personal com-
munication, 2013). The Burmese python likely affects several CERP performance

measures, such as juvenile crocodile survivorship, various aspects of alligator

population status, and wading bird survivorship (Dorcas and Willson, 2011).
The Burmese python is monitored along several defined routes that cover
a small fraction of the Everglades, as well as less systematically by reptile
enthusiasts (Rodgers et al., 2014a); however, no estimate of population size is
possible with existing data. Development of an attractant and means of detection
are recognized as critical needs, but limited resources have hamstrung control
efforts (Rodgers et al., 2013). Contracted research to develop an attractant pos-
sibly based on pheromones was terminated by the contractor (F. J. Mazzotti,
University of Florida, personal communication, 2013). Because this is a prom-
ising avenue for control, delays in this research are crippling. A promising trial
using highly trained dogs for detection (Romagosa et al., 2011) also has not been
followed up. The tremendous amount of press received by this invasion has led
to many unorthodox proposals for management, such as enlisting consultants
from the Irula tribe, a small group from southern India who traditionally hunt
snakes. The Florida Fish and Wildlife Conservation Commission (FWC) has a
permit program to allow trained hunters to capture Burmese pythons and other
nonnative reptiles.3
Argentine Black and White Tegu (Tupinambis merianae)
This predaceous lizard (Figure 6-9), which can reach 4 feet in length,
recently became established and is spreading from a small area in Dade County.
It poses a threat of unknown magnitude to ground-nesting birds and reptiles
(Rodgers et al., 2013). It also is established in Hillsborough and Polk counties
in Florida, and has reproduced in Miami-Dade County (Florida Fish and Wildlife
Conservation Commission, 2012; Pernas et al., 2012). They were first noticed in
Everglades National Park in 2009 (Tony Pernas, NPS, personal communication,
2014), and their presence in the wild is likely due to releases of unwanted pets.
ECISMA has coordinated a monitoring effort but lacks resources for an adequate
rapid response team (Rodgers et al., 2014a). As is often the case with nonnative
animals, no reliable information is available on the number of individuals in
the wild. According to the Florida Fish and Wildlife Conservation Commission
(2012), the current approach to reducing tegu numbers in the wild is targeted
trapping and removal, and additional trapping efforts are under way to contain
the invasion east of Everglades National Park (R. Johnson, DOI, personal com-
munication, 2014).
3
See http://myfwc.com/license/wildlife/nonnative-species/python-permit-program/.

FIGURE 6-9 Argentine black-and-white tegu.
SOURCE: Photograph courtesy of David Policansky, National Research Council.
Wild Hog (Sus scrofa)
Wild hogs (Figure 6-10) are damaging invaders worldwide, inflicting many
kinds of ecological and economic damage to varying degrees in different loca-
tions (Barrios-Garcia and Ballari, 2012). Furthermore, they can interact with
other introduced species, such as mutalistic mycorrhizal fungi and invasive
plants, to generate invasional meltdowns—that is, much greater impacts than
each species could have produced on its own (Nuñez et al., 2013). Wild hogs are

FIGURE 6-10 A wild hog in the Everglades.
SOURCE: National Aeronautics and Space Administration.

in all of Florida’s 67 counties, but information on local numbers and distribution

is difficult to obtain. Their impact is apparent in the Everglades, with large areas
disturbed by their rooting. Some hog control, mainly by trapping and hunting
through contracts given to trappers, is undertaken on particular SFWMD lands,
but there is no high-level coordination of such activities, and this is recognized
as a need (Rodgers et al., 2013, 2014a). Aggressive hog control is controversial
because hogs are a valued game species; the contribution of hunting to control
of hog populations is unstudied. Hogs are also a major prey item for the endan-
gered Florida panther (Maehr et al., 1990), and improved habitat for this prey
species has been offered as one benefit from the Picayune Strand restoration
project (see Chapter 4). These conflicting views on the desirability of hog control
greatly complicate an effective management response.
Feral House Cat (Felis catus)
Feral house cats are damaging invaders worldwide, killing approximately

2 billion birds and 12 billion mammals annually in the United States alone (Loss
et al., 2013). Their impact in Florida is similarly substantial (Feral Cat Issue Team,
2003), and they are a major predator of birds and mammals (and perhaps other
animals as well) in the Everglades, although the committee knows of no specific
estimate. Feral cats trapped in the Everglades may be neutered and released, but
they are not killed, as cat control generally is viewed through the lens of animal
welfare rather than as a conservation issue. Despite the damage feral cats can
cause, they are frequently overlooked in lists of priority invasive species (e.g.,
RECOVER, 2014a; Rodgers et al., 2014a).
Asian Swamp Eel (Monopterus albus complex)
Asian swamp eels are large, carnivorous eels first reported in the wild in
Florida in 1997 (Kline et al., 2013). They appear to have been introduced more
than once, and because they have some degree of salinity tolerance, apparently
variable across populations, swamp eels have the ability to invade estuaries as
well as freshwater. They are opportunistic predators and get quite large (up to at
least 4 feet; see Figure 6-11), and they also tolerate desiccation, pollution, and
low temperatures (Schofield and Nico, 2009). Therefore they have the potential
to be invasive as well as to affect ecosystem structure. Relatively little is known
about their distribution in South Florida.

FIGURE 6-11 Wood stork with swamp eel, which it ate; at Royal Palm, Everglades National
Park, 2013.
SOURCE: Photograph courtesy of Theron Mays.

Lionfish (Pterois volitans and P. miles)
Lionfish, native to the Indo-Pacific region, are now widely distributed in the
Caribbean and southeastern United States. They are highly predaceous, greatly
lower the population density of their prey, and outcompete native reef fish (Albins
and Hixon, 2008, 2011). Lionfish also invade estuaries, including that of the
Loxahatchee River (Jud and Layman, 2012), and a preliminary study shows a
major impact on estuarine invertebrates (Layman et al., 2014). In addition, their
venomous spines can cause extremely painful injuries to people who come in
contact with them. Lionfish are numerous in coral regions of South Florida (e.g.,
Florida Keys, Biscayne Bay). They are currently managed in South Florida on a
coral-head-by-coral-head basis as opposed to a regionwide basis, and the public
is encouraged to capture them by angling or spearfishing and to use them as food.
There is no evidence that this approach has hindered their spread or lowered
their density, except perhaps locally.4 They appear to be rapidly increasing in
abundance and impact (Ruttenberg et al., 2012).
Island Apple Snail (Pomacea insularum)
The island apple snail is much larger than the native apple snail, P. paludosa,
which is the main food of the endangered Everglades snail kite (Rostrhamus
sociabilis). Evidence strongly suggests that the island apple snail outcompetes
the native apple snail (Barnes et al., 2008), and also that the snail kite has lower
net energy balance when feeding on the island apple snail (Cattau et al., 2010).
However, Pias et al. (2012) report some initial behavioral adaptations by the kite
to the island apple snail. The island apple snail consumes a wide range of aquatic
plants as well as other food sources and is capable of completely defoliating lush
ecosystems (RECOVER, 2014a). In a single cell of STA-1E, a major increase in the
population of this snail in 2013 devastated submerged vegetation. The event was
correlated with large increases of total phosphorus in outflow concentrations,
such that the cell had to be taken offline for rehabilitation (Figure 6-12; L. Gerry,
SFWMD, personal communication, 2014). RECOVER (2014a) reported that as
of 2012, the island apple snail was well established in the northwestern littoral
zone of Lake Okeechobee and spreading southward. There is little coordina-
tion of monitoring and little research on impacts and possible control measures
(Rodgers et al., 2014a).
4
See http://www.reef.org/lionfish.

FIGURE 6-12 Island apple snail with egg cluster (left). Egg clusters of nonnative apple snails in STA-1E,
showing the extreme density of clusters and reproductive potential of this species. Native apple snails
have white eggs.
SOURCE: Photograph courtesy of Delia Ivanoff, SFWMD.
Redbay Ambrosia Beetle (Xyleborus glabratus)
The redbay ambrosia beetle (Figure 6-1) and its associated fungus, laurel
wilt (Raffaelea lauricola), have spread widely throughout the southeastern United
States since 2002 (Fraedrich et al., 2008; Kendra et al., 2013), including to
parts of the Everglades by 2010 (Rodgers et al., 2014a). It attacks some native
(e.g., swamp bay, redbay) and nonnative (e.g., avocado) members of the laurel
family and may also affect species that are restricted to feeding on such species,
including the Palamedes swallowtail butterfly (Papilio palamedes) (Lederhouse
et al., 1992). The fungal disease threatens tree island habitat where redbay is
a dominant species. There is also great potential for loss of cultural resources
because redbay is used extensively by local tribes. There is coordinated moni-
toring of the recent dramatic spread of the disease (see Figure 6-13) but little
research on impacts on native species (other than redbay) in the Everglades or
on management methods.

FIGURE 6-13 Distribution and abundance of laurel wilt-infected swamp bays in the central Everglades in
2011 and 2013.
MANAGING INVASIVE SPECIES IN THE EVERGLADES
Management approaches differ for species at different stages of the invasion

process. Biological invasions typically begin with the arrival of a small number
of individuals (propagules) that establish a population that grows slowly at first.
After this initial period, which varies in length depending on the species and
environmental conditions, population numbers increase rapidly until some envi-
ronmental limit is reached (e.g., fewer and fewer habitats are available as more

and more are occupied) and population growth slows, eventually leveling off.
The resulting sigmoid curve (Figure 6-14) can be used to determine management
costs and strategies.
Although eradication often is possible with early detection of new arrivals,
most of the species listed as priorities (Tables 6-2 and 6-3) are in advanced stages of
invasion. Eradication in such cases is always more expensive and difficult and may
not even be possible with current technology. In cases where eradication is highly
unlikely, containment and long-term management are the pragmatic strategies.
Thus, continued presence of some nonnative species on the landscape is a reality
in the modern Everglades. The following sections outline efforts in South Florida
with regard to prevention, early detection and rapid response, eradication, and
maintenance management (including containment and long-term management).
FIGURE 6-14 The invasive species invasion curve.
SOURCE: South Florida Ecosystem Restoration Task Force, 2013.

Prevention
A number of state and federal laws are designed to prevent the intro-
duction of potentially invasive nonnative species into the United States. Two
primary federal laws restrict the import of certain nonnative species into the
United States: (1) the Lacey Act (18 U.S.C. § 42 and 16 U.S.C. §§ 3371 et seq.)
and (2) the Plant Protection Act (7 U.S.C. §§ 7701 et seq.).
The Lacey Act prohibits the importation of certain specified fish and wildlife
that the Department of the Interior (DOI) designates as injurious to humans,
agriculture, wildlife, and “wildlife resources” of the United States. The law has
limited effect in that the only species that can be listed as injurious are certain
classes of animals. In practice, the Lacey Act has been applied only in a reac-
tive way (Fowler et al., 2007). For example, the DOI’s list of injurious species
includes only ones that have already become a problem. The DOI does not list
species that may, but have not yet, become problematic. To date, only about
239 species have been included on the list. Several of the invasive animal spe-
cies in the Everglades are on the DOI list of injurious species. For example, the
list currently includes four species of constrictors (Burmese python, northern
African python, southern African python, and yellow anaconda). The DOI has
proposed listing four additional species of constrictors. Many animal species that
are considered to be invasive in the Everglades, including the Argentine black
and white tegu, are not included on the list. Because the list is limited in scope
and because it contains only species that have already become problematic,
the Lacey Act as currently implemented is not particularly useful at ensuring
that new, potentially invasive species releases are prevented in the Everglades.
Under the Lacey Act, states are permitted to adopt and enforce laws related
to invasive animals, provided such laws are not inconsistent with federal law.
Florida law prohibits the importation for sale or use or release within Florida
of any wildlife not native to Florida unless specifically authorized by the FWC.
However, by regulation, the Commission has limited this prohibition only to the
import, sale, possession, or transport of any live specimens that it lists as “con-
ditional non-native species.” In recent years, the FWC has listed the Burmese
python, reticulated python, northern African python, southern African python,
scrub python, amethystine python, green anaconda, and Nile monitor as “con-
ditional non-native species,” and thus, the prohibition now applies to each of
these reptile species. The FWC regulations also encourage persons possessing
unwanted nonnative species, such as pet pythons, to turn over the animals to
the FWC by providing amnesty.
The Plant Protection Act authorizes the U.S. Department of Agriculture
(USDA) to prohibit or restrict the importation and interstate movement of cer-
tain organisms that USDA determines to be plant pests or noxious weeds. The

Act imposes significant restrictions on species listed as “noxious weeds.” USDA

can list species as noxious weeds if the species can directly or indirectly injure
agriculture or the natural resources of the United States, public health, or the
environment. As with the Lacey Act, species typically are listed as noxious weeds
only after they have become a problem. Many of the invasive plant species in
the Everglades are listed as noxious weeds, including melaleuca, hydrilla, water
hyacinth, and feathered mosquito fern. In addition, the Plant Protection Act
regulates the interstate movement of noxious weeds and authorizes emergency
action within states that are not taking adequate measure to eradicate the plant
pest or noxious weed. Under the Act, the USDA has broad authority to declare
quarantine and take remedial action to prevent the introduction of new, or not
widely distributed, plant pests or noxious weeds. In contrast to the Lacey Act,
the Plant Protection Act preempts state laws that are in conflict with or are more
stringent than the federal law except where a state can demonstrate a special
need for additional restrictions.
The Florida Division of Plant Industry (DPI) in the Department of Agricul-
tural and Consumer Affairs (DACS) does maintain its own list of noxious weeds,
which includes species on the federal list, as well as additional species that are
invasive in Florida but are not on the federal list. Similar to the Plant Protec-
tion Act, Florida law prohibits the introduction, possession, and movement of
noxious weed unless permitted by DPI for limited purposes, such as research.
Most of the invasive plants in the Everglades, including Brazilian pepper, air
potato, Australian pine, Old World climbing fern, cogongrass, water lettuce,
skunk vine, and downy rose myrtle, are on the Florida list (Florida Administrative
Code Annotated, Rule 5B-57.007). As with the federal noxious weed list, DPI
typically lists species only after they have become a problem.
Early Detection and Rapid Response
Eradication is far more likely and less costly early in the sigmoid invasion
curve (Figure 6-14), but this requires early detection associated with a rapid
response mechanism. Thus, investments in early monitoring can yield great
economic benefits by finding invasions while relatively inexpensive eradica-
tion or containment efforts are still feasible. Of course, monitoring has costs, so
the likely benefit of finding and acting early on invasions needs to be weighed
against the cost of a given degree of monitoring. Estimating these costs and
benefits involves many unknowns, but the principle is clear (Epanchin-Niell et
al., 2012).
In the Everglades, wild red rice (Oryza rufipogon) was detected and eradi-
cated before it could spread (Westbrooks and Eplee, 2011). Another success
of early detection and rapid response (EDRR) is the sacred ibis (Threskiornis

aethiopicus), a large African bird first discovered breeding in the Everglades in

2005 after many escapes from captivity (Herring and Gawlik, 2008). This spe-
cies, which is known to prey on eggs and young of several bird species in aquatic
habitats (Lefeuvre, 2013), appears to have been eliminated from the Everglades
before it could disperse widely (Rodgers et al., 2013). The quick effort to eradi-
cate exotic black mangrove (Lumnitzera racemosa) before it could spread from
a site near Fairchild Tropical Botanical Garden also appears to be nearing suc-
cess (Rodgers et al., 2013). These examples show that, if an invader is detected
before it is widespread and if action is quick, eradication is sometimes possible.
The Everglades Cooperative Invasive Species Management Area (ECISMA;
discussed later in this chapter), which coordinates nonnative species man-
agement in South Florida, has developed an EDRR system for the Everglades
(ECISMA, 2009). Such systems have proven to be possible and cost-effective
elsewhere (Westbrooks and Eplee, 2011). Several detection programs have been
established in the Everglades—largely through ECISMA—including a public
hotline and website (1-888-IVE-GOT-1; www.ivegot1.org) operated by FWC for
reporting nonnative animals (no analog exists yet for plants). The hotline was
used successfully to alert ECISMA agencies to Argentine black and white tegus
in residential areas of Dade County in time to remove them through targeted
trapping. Also, more than 30 tegus that had been abandoned in an outdoor
breeding facility in Panama City were discovered and removed after a hotline
report (J. Ketterlin-Eckles, FWC, personal communication, 2013).
Another ongoing ECISMA EDRR effort is the Everglades Invasive Reptile
and Amphibian Monitoring Program (EIRAMP), implemented by a team from
the University of Florida with support from FWC and the SFWMD. Under this
program, regular monthly monitoring is conducted on 20 routes (Rodgers et
al., 2014a). However, the degree of monitoring is insufficient, and the system
does not assign specific responsibilities for monitoring for many sorts of spe-
cies. EIRAMP’s routine monitoring routes are located along just a few roads and
trails (see Figure 6-15), a minuscule percentage of available area, and not all
reptiles and amphibians would be likely to occupy such habitats. Members of
the EIRAMP team are also contracted for a certain number of follow-up visits to
address hotline and website reports of reptiles and amphibians.
FWC has also mounted a Python Patrol program, started by the Nature
Conservancy and now operated by FWC, to limit the spread of pythons into new
areas. The program trains land managers to capture and remove large constric-
tors and also provides outreach to people who frequent natural areas, such as
hunters, local law enforcement agents, and state agency workers. These persons
are trained to identify and report pythons, as well as other nonnative animals.
Python Patrol trainees may also respond to hotline and website reports. This
program has resulted in the removal of the first Burmese python in Picayune

FIGURE 6-15 Everglades Invasive Reptile and Amphibian Monitoring Program (EIRAMP) routes.

Strand after it was reported on the hotline, and responders trained under the
Python Patrol Program were able to find and remove the snake. So far, over
400 persons have received capture training and over 1,400 have received train-
ing on detecting invasive animals.
ECISMA relies heavily on the Early Detection and Distribution Mapping
System (EDDMapS), a web-based mapping system and clearinghouse founded in
2005 by the Center for Invasive Species and Ecosystem Health at the University
of Georgia. EDDMapS accepts reports of nonnative species from the public at
large and forwards the information to South Florida invasive species managers
to determine the accuracy of the species determination. Panther chameleons
(Furcifer pardalis) were located and possibly eradicated from a Broward county
property after a report of the chameleons on EDDMapS. After surveys, FWC
and the University of Florida were able to remove more individuals from the
surrounding community, and continuing surveys have detected no panther
chameleons recently (J. Ketterlin-Eckles, FWC, personal communication, 2013).
It is unclear for certain types of species who decides what, if any, response
to a verified hotline, website, or EDDMapS report is required and what entity
should implement the response. In general, a rapid response requires quick
access to resources, which was generated by FWC for exotic black mangrove.
Other invasive species that were detected early and for which adequate resources
were mobilized quickly were the sacred ibis (now thought to be eradicated in
the Everglades) and the northern African rock python (currently contained to a
limited area). However, efforts to eradicate incipient invasions in the Everglades
have more often been stymied by the inability to obtain funds from federal, state,
or local sources. For example, the Argentine black and white tegu was confined
to a very small area when first discovered in Dade County in 2009. A plea for
quick action (Pernas et al., 2012), which would have yielded a high probability
of successful eradication, went unheeded because none of the agencies queried
could provide the necessary resources. This lizard is now almost surely too
widespread for eradication with currently available technologies.
Eradication
Time is of the essence for eradicating invasive species. Although eradication

was once viewed as impossible or unlikely in most cases (Simberloff, 2003),
technologies have improved greatly over the last few decades, particularly for
terrestrial vertebrates (Genovesi, 2011). In general, aquatic species, insects, and
plants have proven more difficult to eradicate than terrestrial species, especially
vertebrates. Nevertheless, there have been successful eradications of all classes
of organisms, especially in instances where the invasion was detected early
enough that the nonnative species had not spread widely. As noted in the pre-

vious section, several spatially restricted invaders have been eradicated in the
Everglades region, including plants. No widespread invader in the Everglades
has been targeted for eradication. However, invasions frequently pass through a
stage in which there are several spatially separated populations of a nonnative
species. In such a circumstance, it is often strategically desirable to eradicate
small, discrete populations in the hope of containing the invasion in a smaller
area and perhaps ultimately eradicating it (Moody and Mack, 1988). Nation-
ally, this strategy has been used successfully in a high-profile massive attempt
to stem the spread of the Asian long-horned beetle (Anaplophora glabripennis).
In the Everglades area, the strategy has been pursued with isolated populations
of the Argentine black and white tegu in residential areas.
Several invasions that became very widespread have nevertheless been
eradicated, such as a pasture weed (Kochia scoparia) in western Australia. In the
United States, a 50-year campaign to eradicate the parasitic plant witchweed
(Striga asiatica) from 400,000 acres of North and South Carolina is nearing
success (Simberloff, 2013a). However, in both instances, the locations of the
invasive individuals were well known and the invaders were agricultural pests,
so the high cost of the effort (particularly for witchweed) could be borne as an
agricultural expense. In the Everglades, some locations of many widespread
invaders are poorly known, and the cost of an invasion campaign against wide-
spread invaders would likely be prohibitive, at least with current technology.
As described previously in this section (see Prevention), a number of federal
and state laws are designed to reduce the introduction of potentially invasive
nonnative species into the United States. Once an invasive species is estab-
lished, however, federal statutes are of limited utility. Because large areas of
the E verglades are owned or managed by federal, state, or tribal governments,
however, these public land owners and managers typically have specific legal
authority to address invasive species concerns on their land. For example, the
National Park Service Management Policies (NPS, 2006) authorizes the destruc-
tion of species detrimental to Park Service resources on Park Service lands. In
addition, Presidential Executive Order 13112 directs federal agencies to take
actions to control invasive species, and the U.S. Army Corps of Engineers
(USACE) is currently working to develop a strategy to comply with this mandate.
When invasive species management depends on removing plants or animals
that may be on private land and may have the ability to spread to public lands,
it may be necessary to access private lands to remove invasive plants or ani-
mals. In some circumstances, private landowners are willing to provide access
and permission to remove the invasive species. When private landowners are
not willing to provide access and permission, however, there is limited legal
authority for government officials to control invasive species on private prop-
erty. In certain extreme circumstances, state or federal agencies may be able to

access private property to remove invasive species that pose significant risks.
For example, the State of Florida has destroyed citrus trees on private land when
citrus canker threatened a major economic interest in the state. However, in that
case, specific legislation authorized the state action and the state paid compen-
sation to private landowners whose trees were destroyed. It is not clear under
what circumstances and to what extent statutes such as the Plant Protection Act
would allow similar actions to be taken to private land to protect the natural
resources of the E verglades without running afoul of constitutional protections
against illegal search and seizure and taking of property without just compensa-
tion. The Gambian pouched rat on Grassy Key persists only because it is on six
private properties whose owners do not permit access to federal or state officials
(Witmer et al., 2010a). As long as this rat is present, there is the possibility that
it will spread to other areas of South Florida.
Maintenance Management
If eradication fails or is not attempted, several technologies can be used

to maintain invasive populations at low levels. Traditional approaches to such
maintenance management are
1. Physical control, such as pulling invasive weeds or catching snakes by

hand;
2. Mechanical control, entailing the use of machines;
3. Chemical control, using pesticides and herbicides; and
4, Biological control, importing natural enemies, such as predators and
parasites, from the native region of the targeted pest.
Each of these approaches has been successful in some cases, and each has failed
in other cases (Simberloff, 2009, 2014). The important point is that technologies
have evolved in all of these methods (e.g., Clout and Williams, 2009; DiTomaso,
2011; Van Driesche et al., 2008). All have been used in the Everglades. For
instance, several invasive plants, such as melaleuca and Australian pine, have
been targeted by specially adapted land-clearing machines (Anonymous, 2005).
Herbicides, both aerially dispersed and delivered by hand sprayers, have been
also been used (Laroche and McKim, 2004). The USDA’s Invasive Plant Research
Laboratory in Davie, Florida, seeks and tests biological agents, mostly insects, to
attack major invasive plants in the Everglades, and in 2013 the USDA completed
construction of a mass rearing facility as an annex to the laboratory as part of
the CERP (see Chapter 4). Melaleuca is one of several key invasive plants in the
Everglades that have been substantially reduced by biocontrol agents (Figures 6-2
and 6-16). Other methods of maintenance management are used less frequently

FIGURE 6-16 Melaeuca treatment via mechanical and chemical methods (left) and biological control by
means of the melaleuca snout beetle (Oxyops vitiosa).
SOURCE: Photograph courtesy of Tony Pernas, National Park Service, and Stephen Ausmus, Department of
Agriculture.
but have provided significant control of particular invaders. In the Everglades, for
example, prescribed fire applied to stands of melaleuca seedlings has contributed
to developing a successful maintenance management program.
New approaches to maintenance management have occasionally provided
control of previously refractory invaders, including the use of pheromones and
genetic manipulation (Simberloff, 2014). Invasive species whose control seems
hopeless today, as that of melaleuca did 20 years ago, may someday be managed
well by methods resulting from ongoing research. The melaleuca management
program, which evolved over 20 years and includes biological, chemical, and
mechanical control (Figure 6-16) as well as prescribed burns, is an example of
a program that developed gradually from several lines of research and is now
showing substantial success (Figure 6-3).
COORDINATION AND ORGANIZATION
The management of nonnative species in South Florida is distributed across

many federal, state, and local agencies and programs. Federal agencies that have
at least some jurisdiction over nonnative species include the FWS, the National
Marine Fisheries Service, the NPS, the USDA Animal and Plant Health Inspection
Service and Agricultural Research Service, the U.S. Forest Service, the USACE,
and the U.S. Customs and Border Patrol. State agencies include the FWC, the
SFWMD, the Florida Department of Agriculture and Consumer Services, and the
Florida Department of Transportation. Miami-Dade County and the Miccosukee

and Seminole Indian tribes also have strong interests and management roles, as
do a variety of nongovernmental and academic organizations.
Although official communication channels exist among many of these orga-
nizations and many individuals associated with them communicate as well, they
do not all share the same legislative and regulatory mandates, they have differing
budgetary and other constraints, and they have differing degrees of technical
expertise. Two notable attempts have been or are being made to coordinate
efforts and resources for managing nonnative species.
Everglades Cooperative Invasive Species Management Area
Federal, state, and local governments have been collaborating to address

Everglades nonnative species issues since the Everglades Forever Act was passed
in 1993, and the establishment of the ECISMA in 2008 formalized the collabora-
tion and expanded the partners involved. Like other CISMAs, ECISMA is a formal
partnership composed of federal, state, and local government agencies, tribes,
individuals, and various interested groups that manage invasive species in the
Everglades region (see Box 6-1).
ECISMA has fostered an increasing amount of coordination at the operational
level. Among ECISMA’s stated goals are to
• Formalize areas of coordination and cooperation among agencies;

• Define specific geographic areas and prioritize species for Everglades
restoration; and
• Integrate coordination, control, and management of invasive species at
regional, multijurisdictional levels.
ECISMA has improved coordination of the implementation of invasive species

management. Its website provides access to a great deal of pertinent information,
such as distribution maps (EDDMapS) of invasive plants and animals. ECISMA
has also had some notable successes with EDRR, as discussed previously in this
chapter. However, ECISMA does not coordinate and cross-calibrate sampling
methods.
However, there does not appear to be a formal process to determine system
wide priorities—which nonnative species are managed to what extent, what
monitoring is performed, and what monitoring or other observations trigger
a management response. Currently, nonnative species management appears
largely driven by the objectives of individual agencies, with limited leveraging of
funding across agencies to address the needs of multiple agencies. How system-
wide prioritization for management, coordination of management activities, and
funding sources are determined remains obscure. This committee could not iden-

BOX 6-1
ECISMA Partners
The Everglades Cooperative Invasive Species Management Area (ECISMA) coor-

dinates nonnative species management in South Florida through a formal agreement
under the Florida Invasive Species Partnership. The ECISMA partners include
Signatories
• Florida Fish and Wildlife Conservation Commission
• South Florida Water Management District
• U.S. Army Corps of Engineers
• U.S. Fish and Wildlife Service
• U.S. National Park Service
• Miami-Dade County
Cooperators
• Auburn University
• Broward County
• Friends of Everglades CISMA, Inc.
• The Everglades Foundation
• Fairchild Tropical Botanic Garden
• Florida Department of Agriculture and Consumer Affairs
• Florida Department of Transportation
• Florida Power and Light
• Miccosukee Tribe of Indians of Florida
• National Oceanic and Atmospheric Administration
• Seminole Tribe of Florida
• The Nature Conservancy
• University of Florida
• University of Georgia—Center for Invasive Species and Ecosystem Health
• USDA Agricultural Research Service
• USDA Wildlife Services
• U.S. Geological Survey
tify an algorithm or formal process by which nonnative species are prioritized

for management action or particular resources are allocated to nonnative species
management activities. Nor could the committee identify how a specific agency
comes to bear responsibility for dealing with particular nonnative species.
For instance, ECISMA lists 14 plant species and 16 animal species as the
highest priority for management. What process led to these designations? What
process led to specification of management activities targeting these species and
entities charged with carrying them out? For plants, there is no doubt that several
of the targeted species have great impacts, although it is not obvious that other

species do not have equally great impacts. Two of the 14 priority plant species
are not yet widespread and are perceived as eradicable. Nonnative animals
have been a management focus for a much shorter period than nonnative plants
(Rodgers et al., 2013), and it is even less clear how the 16 priority animal species
were chosen. Some priority species are already believed to have a substantial
impact, although others with suspected major impacts (e.g., feral housecat) are
unlisted. Other priority species are not currently having major effects and are
still geographically restricted and believed to be feasible targets for eradication,
even though possible management methods tend to be poorly developed or
unknown (Table 6-3).
Comprehensive Invasive Species Strategic Framework
The second coordinative effort recognizes and responds to a problem with

policy and management of biological invasions to date: namely, the absence
of sufficient coordination, particularly at the strategic level. DOI’s Office of
Everglades Restoration Initiatives, in coordination with the Task Force, is cur-
rently supporting development of a Comprehensive Invasive Species Strategic
Action Framework that includes greatly enhanced high-level coordination and
a crosscut budget.5 In December 2012, the Task Force established a working
group to conduct a comprehensive review of the coordination and nature of
efforts to combat invasive species in the Everglades. As of December 2013 a
strategic planning exercise was under way by the working group to fashion the
Strategic Action Framework. Efforts to devise a governance structure to address
the current gap in strategic and funding coordination could be particularly
useful. This activity is in its early stages, but it appears to be directed at a con-
cern expressed to this committee by many individuals—the lack of high-level
coordination in developing priorities for budgets and actions across agencies to
address invasive species.
INVASIVE NONNATIVE SPECIES IN THE CONTEXT OF RESTORATION GOALS
The catchphrase “get the water right” governing planning for Everglades
restoration assumes that restoring a semblance of the pre-development water
flow to the region will lead to restoration of ecosystems and species. This dictum
has dominated aquatic and wetland restoration since its inception and has been
termed the “field of dreams” hypothesis (Palmer et al., 1997). As discussed in
this chapter, many empirical examples of nonnative species invasions show that
this is not necessarily the case (Palmer et al., 2014). It is possible that getting
5
See http://www.sfrestore.org/ies.html.

the water right will, for certain nonnative species, at least help to lower their
populations and impacts, but for others, attempts to get the water right may
actually exacerbate impacts and even foster further invasions (Ogden et al.,
2005; RECOVER, 2014a).
Recently, the USACE and SFWMD issued guidance to incorporate invasive
species management into CERP project planning and implementation. CERP
Guidance Memorandum 062.00 (USACE and SFWMD, 2012d) required invasive
species management to be incorporated into all phases of CERP projects and
an invasive and nuisance species management plan to be developed as part
of the project implementation report (PIR) process. To date, plans have been
developed for C-111 Spreader Canal, Biscayne Bay Coastal Wetlands, and the
Central Everglades Planning Project (USACE and SFWMD, 2011, 2012a, 2013a),
although only the Central Everglades Planning Project addresses both plants and
animals (RECOVER, 2014a). Several projects that were developed prior to the
guidance memorandum have now developed vegetation management plans.
At a national level, a December 2013 draft Program Management Plan for the
Invasive Species Leadership Team (USACE, 2013f) provides a detailed vision for
management of invasives at all stages of project planning and implementation,
including considerations of design features to reduce the likelihood of enhancing
the spread of invasives. The document provides a strategic plan for e ducating
USACE staff and implementing new regulations, including Executive Order
13112 (1999), which directed federal agencies “to prevent the introduction of
invasive species and provide for their control and to minimize the economic,
ecological, and human health impacts that invasive species cause,” as well as
the 2009 USACE Invasive Species Policy Memorandum (Temple, 2009). If fully
implemented, these documents could help address major concerns of invasive
species management associated with CERP project planning and implementa-
tion, and the committee looks forward to evaluating their results.
Despite the enormous impacts of some nonnative species, invasive species
management has so far not been a major focus of the CERP, beyond treating
invasive plants that spread during construction. Communities composed of
mixtures of nonnative species with varying remains of native assemblages are
commonplace in the Everglades landscape. There is great urgency to detect and
eliminate new arrivals and manage those that have spread beyond the point
where eradication is still feasible. However, funding and manpower are limiting
and effective control techniques (e.g., biocontrol agents) for many nonnatives are
still in the development stage (see Tables 6-2 and 6-3). Prioritization decisions
leave some areas unmanaged and some nonnative species uncontrolled because
the particular system is not ecologically or economically feasible to restore
or because some nonnative species have not been shown to have substantial
harmful ecological impacts or possess seemingly desirable characteristics. Thus,

many parts of the greater Everglades landscape, particularly remote areas with
poor access, remain invaded by multiple nonnative species and will remain so
for the foreseeable future.
Thus, although restoration to a semblance of the way the Everglades looked
and functioned a century ago may, with sufficient effort, be possible for certain
areas, such a goal is impractical for other sites. The spatial extent of the problem
highlights the importance of understanding the effects on ecological functions
as well as potential ecosystem services of these altered assemblages to the
overall ecosystem. Although it is highly likely that the functioning is different
and services are reduced compared with uninvaded, native communities, some
research in this system has shown that certain nonnative species can provide
benefits (e.g., exotic apple snails as a potentially important food for native snail
kites [NPS, 2013]), even if they are simultaneously detrimental in other respects.
The need to prioritize management resources and decipher how vast sec-
tions of the “invaded” Everglades are functioning, however, should not obscure
the ideal goal of a functioning Everglades with its full complement of native
species. CERP partners will have to decide on the restoration goals for specific
areas of the Everglades, recognizing that hydrological restoration alone will not
necessarily achieve ecological restoration goals (Clewell and Aronson, 2013).
Discussions on this issue will need to consider that areas left unmanaged for
invasive species because full restoration is not a goal can serve as sources of
seeds, spores, and other propagules and thereby threaten other areas being man-
aged for more ambitious restoration goals.
In addition, as described earlier in the chapter, interactions between non
native and native species and between different nonnative species often are
complicated, and one invasion may exacerbate the spread of another (as exem-
plified by the case of the invasive figs and their fig wasp pollinators).
CLIMATE CHANGE AND INVASIVE SPECIES
With global warming, sea-level rise, and the water management activities
associated with them, distributions and abundances of nonnative species are
expected to shift across the landscape. New species will likely invade, while
distributions of some existing species will contract and others expand (Hellmann
et al., 2008). These changes will be driven in part by shifting climatic envelopes,
but also by changes in species interactions (Simberloff, 2012).
As discussed in Chapter 5, global climate change models developed at
coarse scales for the Everglades vary with respect to projected temperature and
precipitation regime changes, as well as rate of sea-level rise (Obeysekera et
al., 2011a). Responses of nonnative species to these scenarios, as well as the
anticipated changes in hydrologic regimes with restoration activities, present

an area of great uncertainty that has received relatively little attention. Given
the ever-growing number of nonnative species in the system, a multitude of
species-specific response patterns and interactions with native flora and fauna
is likely as the effects of these environmental changes unfold (Junk et al., 2006).
One possible climate change scenario includes increased temperatures
and potential evapotranspiration, along with reduced precipitation, resulting
in reduced water supplies to the entire system. Human-caused changes to the
Everglades ecosystem have already shortened hydroperiods, in some cases favor-
ing the spread of many nonnative plant and animal species (Jones and Doren,
1997; Olmstead and Loope, 1984). Further shortening of hydroperiods under
some climate change scenarios may promote their continued expansion (Davis
et al., 2005b). In addition, many of the species introduced into the Everglades
are native to tropical habitats and are thus likely to expand with warmer condi-
tions at the expense of resident native flora and fauna that are better adapted to
temperate and subtropical climates. For instance, Trexler et al. (2000) suggest
that the densities and range of several nonnative fish species in the Everglades
currently are limited by occasional low temperatures or severity of droughts. A
changed climate could relax some of those limiting constraints.
Biological control agents have been introduced to counter many invasive
nonnative species in the Everglades. These species themselves are generally non-
native as they have been selected and introduced from the home range of the
target species. As with other nonnative species, there is great uncertainty with
respect to effects of climate change on these agents. As temperature and rainfall
patterns change, geographic distributions of the agents and their targets are likely
to change as well. Such changes could promote increased contact between
agents and nontarget species. Further, the effectiveness of these agents could be
affected by changing environmental conditions and shifts in timing of the plants’
or animals’ life-cycle events (Parmesan, 2006; Simberloff, 2012). Agents that are
currently effective may become less useful and potentially problematic if they
become spatially or temporally decoupled from their targets.
Despite excellent progress in developing coordination of the management of

invasive species at the operational level, most notably through ECISMA, there is
a lack of coordination at a strategic level that includes a comprehensive view of
all nonnative species in all parts of the Greater Everglades. Currently, plants and
animals tend to be considered separately. Management and restoration activities
need to take account of the entire biotic community and not be partitioned into
different taxa. This indeed is consistent with the vision forEverglades restora-
tion. However, it can be difficult to take such a view at a project level. Further,

for many invasive species, different agencies take on management activities in

different areas, yet individuals of such species move between areas, so that man-
agement in one area can impact other areas. These factors argue for the creation
of a high-level coordinative entity to oversee policy, management, and budgets
related to nonnative species. Prioritization of research needs and control efforts
across areas, species, habitats, and agencies would be a major responsibility of
this entity. The committee is optimistic that the Comprehensive Invasive Species
Strategic Action Framework being developed by the Task Force will be a major
step toward achieving these goals of high-level coordination.
A strategic early detection and rapid response (EDRR) system that addresses
all areas, habitats, and species is needed. EDRR is an essential strategy if new
invasions of nonnative species in the Everglades are to be eradicated (or at least
contained) while it is still feasible and relatively inexpensive to do so. Currently
several EDRR efforts are under way, but the current level of monitoring is insuf-
ficient to address the geographic extent and range of nonnative species threats in
the Everglades. In general, a rapid response requires quick access to resources,
but efforts to eradicate incipient invasions in the Everglades have more often
been limited by the inability to obtain funds from federal, state, or local sources.
The costs of additional monitoring and response should be weighed against
the likely benefits of finding and acting on early invasions. Additional funding
would allow for greater public outreach, expanded operation of the reporting
hotline, increased early detection monitoring, and improved capacity for rapid
response to facilitate eradication. The committee recognizes that the goal of
this recommendation—addressing all areas, habitats, and species—likely is
beyond any reasonable expectation of resources, but keeping this goal in mind
emphasizes the value of prevention and clarifies the magnitude of the challenge.
There is no systemwide mechanism for prioritizing research on and man-
agement of invasive species. Many agencies participating in the Everglades res-
toration already undertake research activities on certain nonnative species and
also undertake management activities, but these efforts are limited by insufficient
resources and are typically driven by specific agency needs rather than system-
wide priorities. Effective prioritization requires a comprehensive understanding
of all nonnative species present in the Everglades, their impacts and threats, as
well as those of impending or likely new arrivals.
Research is lacking on nonnative species and their impacts to inform priori-
tization efforts adequately. Tables 6-2 and 6-3 highlight some of the many gaps
in knowledge about species considered to be priorities for management. Given
the spatial extent of the problem and the threats of future invasions, substantial
research is needed to assess the various impacts of nonnative species on ecosys-
tem functioning and native species and to develop or improve control mecha-
nisms. This does not mean comprehensive research of all details of the biology

and effects of every nonnative species. Rather, enough basic information should
be gathered systematically to determine which species could reasonably be pre-
dicted to have considerable ecological impacts. Such knowledge is important
in guiding decisions on detailed research on possible impacts and management
of particular threats and would help inform priorities for management actions.
If eradication proves impossible, maintenance management and long-
term control at acceptable levels should be explicitly recognized as a goal in
some cases. Indeed, current practice seems implicitly to recognize this goal.
Maintenance management at low densities is sometimes possible by various
combinations of biological, chemical, mechanical, and physical controls. In
the Everglades, a striking example is the current management of melaleuca,
once thought too widespread and dense to be manageable. As a result of sus-
tained intensive research, this species is currently under substantial control in
most regions through a combination of mechanical, chemical, and biological
control as well as prescribed burns. Maintenance management requires contin-
ued, diligent monitoring and flexible, but reliable funding that can be devoted
strategically to achieve and maintain long-term control.
At every step of the CERP planning process, full consideration is needed of
the implications of restoration activities for introduced species and their impacts.
Until very recently, invasive species have not been considered in CERP project
planning and implementation beyond simply removing any invasive species
encountered at construction sites. Ideally, hydrologic restoration should favor the
reestablishment and expansion of many native wetland species that are better
adapted to longer hydroperiods. However, aquatic and flood-tolerant nonnative
species may also benefit and replace native species. Removing levees and filling
in canals may, in certain circumstances, facilitate the spread of nonnative species
by increasing their potential for dispersal. For each CERP project, the potential
to increase the spread of invasive species should be examined and the effects on
ecosystem functioning assessed. Based on this analysis, strategies and technologies
to lessen these impacts should be appropriately considered. Recent CERP guid-
ance and plans to implement national USACE invasive species policy indicate that
these considerations are increasingly being incorporated into project planning and
implementation, although it is too soon to evaluate this new approach.
Long-term monitoring and research are needed to understand the poten-
tial impacts of climate change on Everglades nonnative species management.
Climate change has the potential to significantly impact the distributions and
abundances of nonnative species in the Everglades and their impacts on the
ecosystem as a whole. Thus, research and monitoring to understand long-term
changes in nonnative species distribution and behavior and the effectiveness
of maintenance control strategies in the context of climate change are needed.

Use of Science in Decision Making
Given the enormous scope and complexity of the restoration effort, strategic,
high-quality, responsive, and sustained science and adaptive management are
needed to ensure the effectiveness of the impressive Comprehensive E verglades
Restoration Plan (CERP) engineering efforts under way. In this chapter, the com
mittee reviews scientific support for Everglades restoration from several perspec-
tives. This chapter builds upon prior reviews of this topic by the National Research
Council (NRC, 2007, 2008, 2010, 2012a). First, science coordination and manage-
ment are discussed, with particular emphasis on what is needed for an effective,
sustainable systemwide monitoring program. Next, progress on the implementa-
tion of adaptive management is discussed. Finally, the recently released 2014
Draft System Status Report (RECOVER, 2014a) and the Science Plan for Everglades
Stormwater Treatment Areas (SFWMD, 2013d) are reviewed, because they repre-
sent important contributions to the adaptive management process.
SCIENCE COORDINATION AND MANAGEMENT
A comprehensive program of scientific research and systemwide monitoring

helps ensure that the substantial investment in Everglades restoration is being
directed effectively. Moreover, the concept of adaptive management depends on
monitoring data to assess whether restoration goals and targets are being met and
on lessons learned through scientific research to improve restoration outcomes.
Because of the complex nature of the Everglades ecosystem and the numerous
federal, state, and tribal government agencies and stakeholders involved with
multiple perspectives and objectives, science governance is a challenge. To
have a robust effective science and monitoring program for this complex resto-
ration program, a number of features are required, including stable funding and
effective science coordination and communication. The committee judges that
research support for Everglades decision making is robust; therefore, this section
focuses on monitoring and coordination.
213
The Need for Robust Science and Monitoring Programs
Substantial federal and state resources have been and continue to be invested
in Everglades restoration. To ensure that these resources are being used wisely to
achieve restoration objectives, a robust science and monitoring program is essen-
tial. Scientific research provides knowledge and tools that assist decision makers.
Monitoring involves the collection of data necessary to evaluate the success of
various restoration projects. Long-term data that describe the conditions, vari-
ability, trends, and patterns related to resources and processes in the Everglades
are fundamental to understanding whether and how projects, once implemented,
change conditions. Systemwide, long-term perspectives are all the more important
in the context of climate change, given that “baseline” (pre-project) conditions
are not anticipated to be stationary through time (see Chapter 5). Comprehensive
ongoing monitoring and assessment are also critical to adaptive management.
The importance of comprehensive monitoring and assessment to the suc-
cess of Everglades restoration has been recognized from the beginning by CERP
partners and by prior NRC committees (NRC, 2003a, 2007). Under RECOVER, a
systemwide Monitoring and Assessment Plan (MAP) was developed as “a single
integrated, system-wide monitoring and assessment plan that will be used and
supported by all participating agencies and tribal governments as the means
of tracking and measuring the performance of the CERP.”1 The most recent
is the 2009 MAP (RECOVER, 2009), which is a revised version of MAP 2004
(RECOVER, 2004). The RECOVER program is responsible for linking science with
CERP systemwide planning, evaluation, and assessment, and one key RECOVER
responsibility is to “ensure that a system-wide perspective is maintained through
the restoration process” (RECOVER, 2012a). The RECOVER program uses sci-
entific information developed pursuant to implementation of the MAP to assess
the performance of the CERP.
Beginning with a dedicated workshop in November 2001, the NRC has
reviewed the development of MAP and the selection of appropriate and practi-
cal performance measures by RECOVER (NRC, 2003, 2007, 2008, 2010). As
noted in NRC (2008), performance measures of both ecosystem condition and
critical ecosystem stressors (e.g., estuarine salinity, soil and water phosphorus
concentrations, hydropatterns) have been developed, which allows assessment
of cause-effect relationships. This is a great strength of the performance measure
system, because an understanding of ecosystem dynamics is crucial for imple-
menting an adaptive management approach. The MAP and its performance
measures were reviewed extensively in NRC (2008), which concluded that “[t]he
number of performance measures is not inherently problematic” but noted that
1
http://www.evergladesplan.org/pm/recover/recover_map_part2.aspx.

Use of Science in Decision Making 215
“the set of performance measures should be reviewed regularly to determine

whether . . . adequate data collection for each could be sustained over the
course of the restoration.”
Previous NRC reports (2003a, 2007, 2010) have judged that the RECOVER
MAP is a reasonable plan for monitoring and assessing systemwide conditions in
the Everglades. The team that developed the MAP seems to have taken a holistic
view and does not appear to have been overly influenced by parochial interests.
The MAP does not narrowly focus on specific projects, specific geographic
areas, or specific resources, but instead takes a systemwide approach. However,
there are several shortcomings with the MAP. From the beginning, the MAP was
intended to fill critical gaps in systemwide monitoring, rather than to control and
fund all restoration-related monitoring for the Everglades. The challenge with
this approach is that no entity oversees the entire monitoring effort or manages
monitoring priorities across a diverse array of agencies and institutions. Thus, it
would be helpful to have a process to evaluate and revise the MAP over time
as agencies’ funding changes.
A recurring, comprehensive review process would also help the MAP adapt
as new information becomes available and as ecosystem conditions change
due to climate change or other anthropogenic or natural circumstances. Given
the extremely long time lines currently being projected for completion of the
Central Everglades Planning Project (see Chapter 3) and the very real likelihood
that climate change and sea-level rise will cause significant changes to the
South Florida ecosystem during this time (see Chapter 5), the MAP should be
revisited to evaluate whether it is still appropriate. Key long-term monitoring to
understand the shifting ecological baselines in the context of climate change
may currently be overlooked, while other monitoring may be too frequent in
the context of the slow pace of CERP implementation.
The Need for Stable Funding
The most significant shortcoming with the MAP, however, is not necessarily
with the plan itself, but with the implementation of the plan, because of sub-
stantial funding cuts that have occurred in the past few years and the overall
funding structure. A dedicated, stable, and reliable funding source is essential
to obtain the long-term systemwide monitoring data necessary to evaluate the
success of restoration efforts. Such funding is necessary to conduct monitoring
for long enough to provide a scientifically sound understanding of the condi-
tions, trends, and patterns for each parameter of concern.
NRC (2012a) addressed the recent budgetary cuts and their impact on
monitoring and assessment. In that report, the committee concluded that the

large and sudden cuts to the RECOVER MAP pose a risk to systemwide assess-
ment, which is important to the success of Everglades restoration. The cost of the
RECOVER-funded monitoring through the MAP increased from about $0.7 mil-
lion in fiscal year (FY) 2000 to about $10 million in FY 2007, and MAP funding
has declined roughly 60 percent since 2007, with a sharp cut of 48 percent in
FY 2012 (NRC, 2012a). The funding level has remained flat throughout 2013.
These cuts were amplified by cuts in many other agencies’ monitoring budgets.
Agency staff voiced concerns that the monitoring cuts reduced the capacity to
understand systemwide ecosystem responses and to explain why changes may
have occurred.
Although the CERP is struggling with many budget uncertainties, the com-
mittee remains convinced of the vital importance of systemwide monitoring to
the success of Everglades restoration. Without a sufficient monitoring program,
the CERP cannot be accountable to federal or state sponsors and cannot sup-
port its adaptive management program. The committee recognizes the realities
of changed economic conditions, budgets cuts, and shifting priorities and thus
understands that funding cuts sometimes are unavoidable. But the long-term
costs of monitoring cuts are often overlooked. If funding cuts result in significant
gaps in critical long-term monitoring data, important changes and patterns could
be missed, and data collected prior to or after the gaps created by funding cuts
could lose their value. Therefore, to ensure that existing monitoring is cost-effec-
tive and provides adequate support for CERP planning, adaptive management,
and public communication, a comprehensive review of all monitoring programs
that were considered in the original design of the MAP is needed, considering
recent and projected reductions. The major MAP budget reductions for FY 2012
were implemented very quickly (NRC, 2012a), and time was not available to
reconsider the essential components of a monitoring program, particularly in
light of the slow pace of CERP implementation in a changing climate, or to
consider the shifting budgets of other agency monitoring programs.
The existing monitoring funding structure with its 50-50 state-federal cost-
sharing requirement (see Chapter 4) appears to be especially vulnerable to
changing economic and political conditions. The structure of the cost share
is such that if the state experiences cutbacks that result in funding cuts to the
MAP, the federal government may be constrained from making up the differ-
ences to fund the MAP because such funding will heighten imbalances in the
overall cost share. A different funding structure that not only provides a more
reliable, consistent long-term source of support could go a long way in ensur-
ing continuity in long-term monitoring. Although there may be a wide range of
possible mechanisms for providing long-term stable funding for ecosystem-wide
monitoring and assessment, one approach could be dedicated funding provided

to one federal agency without being tied to the 50-50 cost-sharing requirements
of the current system.
The Need for Effective Coordination and Communication
Scientific research and monitoring programs require coordination and com-

munication to be effective and efficient. Currently, there is no single entity that
is responsible for coordinating scientific study and scientific monitoring related
to restoration. Numerous federal and state agencies, as well as other entities
such as tribes, local governments, nongovernmental organizations, and univer-
sities carry out research projects and monitoring related to restoration. In fact,
in the last few years, at least 50 scientific studies have been conducted in the
Everglades. Many of these studies are either agency specific or project specific.
In other words, each agency carries out studies to support its own responsibili-
ties and objectives. For example, the National Park Service, the U.S. Geologi-
cal Survey, and the South Florida Water Management District (SFWMD) each
conduct research related to its agency missions and objectives. These research
efforts may focus on specific resources or specific geographic locations or may
have systemwide applications.
After the U.S. Government Accountability Office (GAO, 2003) recom-
mended improved science coordination, the South Florida Ecosystem Restora-
tion Task Force’s (Task Force’s) Science Coordination Group (SCG) developed
the 2006 Plan for Coordinating Science (SFERTF, 2006), last updated in 2010
(SFERTF, 2010). The Department of the Interior released its Science Plan in
Support of Ecosystem Restoration, Preservation, and Protection in South Florida
in 2005 (DOI, 2005). However, these plans are now dated, and no longer serve
to facilitate scientific coordination.
The SCG was specifically formed to coordinate the scientific aspects of
restoration to support the efforts of the Task Force. SCG members include both
scientists and senior managers from federal and state agencies, tribes, and local
governments. The purpose of including both scientists and senior managers
in the SCG is to “enhance the integration of science and management” (SCG,
2003). The SCG is charged with coordinating the scientific aspects of restoration
in general and thus is not limited to CERP projects or monitoring. Despite the
broad science coordination charge to the SCG, the group’s success in providing
coordination and oversight of science has been limited. Ideally, an organization
such as the SCG, broadly tasked with science coordination, would keep track
of ongoing scientific studies, identify gaps and redundancies, identify scientific
needs, and direct staff and financial resources to fill significant information
gaps. It does not appear that the SCG has played a significant role since 2006 in

evaluating the state of the science to identify gaps or overlaps. It is not entirely
clear why the SCG has not played a more significant role. The SCG’s lack of
dedicated funding and lack of authority to direct financial resources to pay for
needed science certainly is part of the limitation. For the SCG to significantly
contribute to better science coordination, it would need to have adequate
funding and staff and a clear charge to address critical science needs from a
restoration-wide perspective.
The history of the SCG suggests that its role and priorities have shifted over
time. These shifts may be contributing to the perceived diminution of science
coordination or may be the result of other actions such as cuts to science bud-
gets or personnel changes. In any event, a review of the role of the SCG over
the past 10 years may provide some insight that could inform efforts to improve
science coordination and communication.2 In the first few years after its forma-
tion in 2003, the SCG appears to have been intensely focused on developing
its comprehensive Plan for Coordinating Science and developing systemwide
indicators. Both were original efforts that required intense staff and SCG member
engagement and creative work. In 2007-2008, the SCG continued working on a
science coordination plan and systemwide indicators while beginning to tackle
some challenging questions in focused meetings or workshops (e.g., identifying
ecosystem features or areas with the largest rates of decline, potential impacts of
climate change). In 2009-2011, the SCG shifted its focus to new initiatives related
to climate change, invasive species, and new science. Although these efforts
were intended to be original syntheses to assist the Task Force in identifying next
steps, the actual impact of these efforts is not clear. During this time, the SCG
also held a workshop on science and decision making, which was well received.
Since 2011, the SCG meetings have discussed the Central Everglades Planning
Project and MAP budget issues. Although there was discussion of a workshop
to reevaluate the monitoring plan,3 such a workshop has not yet occurred. It
also appears that during the past few years, the number of SCG meetings has
tapered off and the meetings have been focused more on providing restoration
updates, rather than unique SCG initiatives. The extent to which recent budget
cuts and the recent intense focus on the Central Everglades Planning Project has
diverted attention from science coordination is unclear. In any event, it seems
clear to the committee that the SCG could and should reengage in its mission
of science coordination and leadership. An important task for the SCG would
be a comprehensive reevaluation of restoration-related monitoring in light of
current budget impacts, the extended CERP implementation time frames, and
2
Meeting agendas and minutes for the SCG can be found at http://www.sfrestore.org/scg_minutes.
html.
3
See http://www.sfrestore.org/wg/wgminutes/2013meetings/013113/minutes_092013.pdf.

climate change. This reevaluation should clearly articulate the value of the high-
est priority monitoring needs and the risks of ceasing such monitoring to future
restoration decision making. One important component of carrying out these
recommendations would be for the SCG to hold regular meetings focused on
science coordination planning and for the SCG to host occasional workshops
on important science-related issues.
Another issue that has been raised with regard to science governance is
the challenge of communication between scientists and upper-level managers
and policy makers. A communication structure that facilitates communication
between scientists and upper-level management is fundamental to sound deci-
sion making. Managers need a mechanism to communicate information needs
to researchers to meet policy objectives. Researchers also need a mechanism
that enables them to communicate science needs and results of research and
monitoring to upper-level management. One process that could be put into place
is a regular meeting between senior science staff and upper-level managers to
discuss the status of ongoing science and how it relates to decisions being made
at high levels. It does not appear that this type of meeting between scientists
and high-level managers occurs with any regularity. If the SCG revisited the
comprehensive 2006 Plan for Coordinating Science, the SCG could facilitate
discussions between policy makers and scientists to identify additional pressing
science needs.
ADAPTIVE MANAGEMENT
In the context of the CERP, adaptive management is defined as “a structured

management approach for addressing uncertainties by testing hypotheses, linking
science to decision making, and adjusting implementation as necessary to improve
the probability of restoration success” (USACE and SFWMD, 2011b). A major char-
acteristic of adaptive management is a feedback mechanism for refining project
planning and implementation based on new information gained from monitoring
results, thus reducing uncertainties that may prevent a project from proceeding
or achieving its intended outcomes. Adaptive management has been a core com-
ponent of the CERP since the year 2000 and remains an active and continually
evolving area of planning. Previous NRC reports have provided detailed reviews
and evaluations of the adaptive management principles and frameworks devel-
oped for the CERP in terms of their ability to meet adaptive management goals
and assess restoration outcomes (NRC, 2008, 2010). In this section the committee
reviews the progress made in activities to support adaptive management within
the CERP since 2010. A more detailed review of adaptive management progress
in the context of the Central Everglades Planning Project is provided in Chapter 3.

In the CERP Monitoring and Assessment Plan (RECOVER, 2004, 2006a, 2009)
monitoring and research needs are identified for measuring ecosystem responses
to CERP implementation, but the first authorized CERP projects did not include
formal adaptive management plans. Although the Water Resources Development
Act of 2000 acknowledged the adaptive management foundations of the CERP,
the 2003 Programmatic Regulations (33 CFR Part 385) required development
of an adaptive management program by CERP-implementing agencies, and the
2006 Comprehensive Everglades Restoration Plan Adaptive Management Strategy
(RECOVER, 2006b) laid a framework for adaptive management, it was not until
2009 that the U.S. Army Corps of Engineers (USACE) required adaptive man-
agement plans for all USACE ecosystem restoration p rojects (Convertino et al.,
2012; LoSchiavo et al., 2013). Since 2009, formal adaptive management plans
have been developed or revised for four CERP projects (Table 7-1): Decompart-
mentalization of Water Conservation Area 3 (WCA-3), the Biscayne Bay Coastal
Wetlands project, Broward County Water Preserve Area, and Central Everglades
Planning Project (USACE and SFWMD, 2011c, 2012c,e, 2013b). The C-111
Spreader Canal also has a monitoring and assessment plan (USACE and SFWMD,
2011d) that contains many components of an adaptive management plan despite
lack of formalization and approval as an adaptive management plan (Table 7-1).
These plans contain various levels of complexity, dictated in part by the scope
of the project and the suite of desired ecosystem responses with project imple-
mentation. Additionally, adaptive management options are limited if they are not
integrated into the project design from the outset—the initial phases of the CERP
did not explicitly integrate adaptive management into project implementation
plans because it was not mandated. Hence, since 2009, adaptive management
plans have become more integrated and sophisticated with time as guidance
has been developed and refined, with the Central Everglades Planning Project
adaptive management plan being the most complex and sophisticated to date
(see Chapter 3).
Two notable sets of guidelines have recently been finalized with the aim
of providing an explicit framework for developing consistent adaptive man-
agement plans for CERP projects: the Adaptive Management Integration Guide
(RECOVER, 2010), and the CERP Guidance Memorandum 56 (USACE and
SFWMD, 2011b). The CERP Guidance Memorandum 56 is the first guide to
merge the various adaptive management documents with other guidance mem-
oranda for development of project implementation reports (PIRs). It specifically
focuses on areas of intersection in adaptive management guidance across the
CERP 6-step planning process, the Adaptive Management Integration Guide,
the USACE Planning Guidance Notebook (USACE, 2000), engineering circulars
and regulations, and USACE Headquarters guidance memoranda. These two

TABLE 7-1 Time Line of Inclusion of Adaptive Management Plans in Active CERP Projects
Adaptive
Current USACE Life-Cycle Management
CERP Project Phase (year authorized) Plan Adaptive Management Features
Aquifer Storage and Pilot projects implemented No a
Testing pilot projects and sensitivity
Recovery (2000 authorization) modeling
Indian River Lagoon- Construction No
South (2007 authorization)
Picayune Strand Construction No Monitoring and assessment plan with
(2007 authorization) recommendations to use adaptive
management
Site 1 Impoundment Construction No
(2007 authorization)
Melaleuca eradication Implementation Noa Adaptive management
(2007 authorization) implementation strategy and some
monitoring
2009 USACE HQ policy requiring adaptive management for ecosystem restoration projects; 2011 CERP
Adaptive Management Integration Guide and 2011 CERP Memorandum Guide 56 released
C-111 Spreader Canal Pilot project and Planning Noa Design and operational tests, project
Chief’s report (2011), phasing
operations
Decompartmentalization Pilot project constructed Yes Decomp Physical Model adaptive
of WCA-3 (2013) management field test
Biscayne Bay Coastal Planning Chief’s report Yes Post-construction management
Wetlands (2012) options matrix and linked monitoring
Broward County Water Planning Chief’s report Yes Operational options linked to nutrient
Preserve Areas (2012), design and ecological monitoring, and design
improvements
Central Everglades Planning Yes Design tests, project phasing, post-
Planning Project construction contingency options, and
operations linked to monitoring
a
Indicates that the project had some components of adaptive management even though it did not have a formal adap-
tive management plan.
NOTE: Projects are listed chronologically by when they were authorized for construction or when the planning chief’s
report was approved for Congress.
SOURCE: Modified from LoSchiavo et al. (2013).

guides are the products of years of effort to develop a coherent, generalized, and
comprehensive structure for adaptive management planning for CERP p rojects
in response to repeated recommendations and mandates, and as such they
represent significant progress toward adaptive management planning.
Recommendations to develop decision analysis tools to support the adap-
tive management process have been a focus of two previous NRC reports (NRC,
2010, 2012a). Formal decision frameworks to integrate scientific information
from monitoring activities, stakeholder values, and costs, while addressing risk
and uncertainty, are crucial to providing transparent decision support to weigh
multiple objectives in highly complex and uncertain multiagent systems such
as the CERP. The development of multicriteria decision analysis tools to supple-
ment adaptive management for the CERP was under way during the prior com-
mittee’s review (NRC, 2012a). The year 2012 marked the Phase 1 completion
of a Bayesian network decision analysis tool intended to “provide managers
with a framework for evaluating and assessing multiple restoration objectives
(performance measures, constraints, costs, risk/uncertainty, and social values)
in order to understand how implementation of a program and/or project and its
adaptive management plan(s) should change based on a given state of informa-
tion” (Convertino et al., 2012, 2013). In a proof-of-concept case study, the tool
was applied to management alternatives related to the decompartmentalization
of WCA-3. The decision support tool characterizes linkages between the project
objectives, conceptual and predictive models, the direct and indirect effects of
project alternatives on project objectives, stakeholder values to weight objec-
tives, and the uncertainty associated with achieving competing objectives. A
distinctive feature of this tool is a global sensitivity analysis that allows for assess-
ment of the value of information each parameter in the decision tool contributes
to the decision. In this way the decision support tool can inform the monitoring
activities that can optimally reduce uncertainties while minimizing costs induced
in redirecting or increasing data acquisition efforts in the context of meeting
restoration objectives. If funding for the project is continued, this tool will be
improved in Phase 2 by broadening the stakeholder involvement and including
greater depth and breadth in spatial and ecological parameters and expanded
in Phase 3 to the larger ecosystem under the CERP (Convertino et al., 2012).
2014 SYSTEM STATUS REPORT
RECOVER System Status Reports (SSRs) provide periodic assessments of

monitoring data throughout the South Florida ecosystem to support adaptive
management and improve CERP planning and implementation. The 2014 Draft
SSR (RECOVER, 2014a), the fifth in the series, was released in late March

2014, and represents a synthesis of 5 years of monitoring data and scientific

research since the last comprehensive SSR was released in 2009. The document
is “intended to convey key scientific information to water managers, budget
directors, decision-makers, and the public about the status of the Everglades
ecosystem to support restoration and water management decisions” (RECOVER,
2014a).
The 2014 Draft SSR, like its predecessors, is a comprehensive document.
The committee’s review addresses the degree to which it provides information
to support adaptive management, and because the document became avail-
able late in the committee process, the committee focused on Chapters 1 (Key
Findings) and 4 (Systemwide Science). In brief, the committee concludes that
the SSR is well written and provides good information, including syntheses and
recommendations, that are helpful to management decisions about Everglades
restoration. Some specific comments are provided as examples below.
The Key Findings (Chapter 1) of the 2014 Draft SSR set the stage by reporting
recent hydrologic and climate conditions affecting the region between 2009 and
2013, compared with historical averages. Overall, it provides a succinct sum-
mary of major findings with regard to status and trends, projects and operations,
and new science covering scales from project level to systemwide. A strength
of the document is its synthesis of a huge array of monitoring data and recent
research into science-based recommendations for management. Rather than
simply reporting observed trends, the 2014 Draft SSR explains and documents
the causal mechanisms and provides recommendations for continued ecologi-
cal improvements. For example, on the basis of new research findings on oyster
survival in the St. Lucie Estuary, the SSR documents the adverse effects of back-
to-back dry years and proposes salinity targets that could be used in the opera-
tional plans for the Indian River Lagoon-South project. Research also determined
that oyster restoration in the St. Lucie Estuary is limited by suitable substrate
rather than the supply of larvae, and the SSR recommends substrate enhance-
ment just prior to spawning to improve restoration outcomes. Similar synthesis
of findings and recommendations are provided for the Greater E verglades, Lake
Okeechobee, and Florida Bay.
The 2014 Draft SSR also documents ecosystem improvements that can be
quantitatively linked to CERP and non-CERP projects:
• Hydrology improved due to the operational part of the Deering Estate

Biscayne Bay Coastal Wetlands expedited project;
• Picayune Strand showed higher water levels near the filled Prairie Canal
(1 to 2 feet higher) and vegetation is starting to show signs of improvement and
moving closer to reference conditions;

• Hydroperiods were 50 days longer (on an annual average basis) along the
centraleastern edge of Everglades National Park as a result of the C-111 South
Dade project; and
• Roseate spoonbill nesting improved, most likely due to favorable climatic
conditions and better real-time environmental coordination with water manage-
ment operational decisions.
The SSR does not overstate the ecological project responses (citing these as
“demonstrations of small restoration successes”) and points to steps necessary
to increase observable improvements in the Biscayne Bay Coastal Wetlands and
Picayne Strand projects. The document also highlights the continuing ecosystem
declines and “the need for and value of authorizing, constructing, and operating
more CERP restoration projects to achieve systemwide hydrologic (water quan-
tity, quality, timing, and distribution) and ecological (flora, fauna, and landscape)
goals and objectives.”
In Chapter 4, Systemwide Science, the draft SSR reflects a subject-matter
focus that is quite similar to the focus of this NRC report, with extensive informa-
tion and discussion of climate change and invasive species, including a substan-
tial appendix devoted to invasive species. The draft SSR includes discussions of
the implications for restoration of changing climate and rising sea level, and its
discussions and analyses of invasive species include regional and systemwide
status and trends. The 2014 SSR also provides a comprehensive review of recent
research and data on the role of fire in the Everglades (described as “one of the
first attempts to reconcile the historical data set of fire history in ENP [Everglades
National Park] and BCNP [Big Cypress National Preserve] with the current man-
agement”) and the implications for fire management. These summaries represent
important and useful synthesis efforts, building on other recent science synthesis
reports (RECOVER, 2011b; SERES Project Team, 2010; WG and SCG, 2010)
summarized in NRC (2012a).
The committee concludes that the Draft 2014 SSR reflects a comprehensive,
scientifically up-to-date and sound approach and execution. It is well organized
and illustrated, and for such a large document, it is easy to read. The document
is very clear and explicit in connecting the information presented with the needs
of managers as they make restoration decisions concerning project design, con-
struction, implementation, and operation. One area that could receive additional
attention is at the intersection between water quality and hydrology, including
recognition where conflicts exist between near-term restoration goals.

REVIEW OF THE SCIENCE PLAN FOR

EVERGLADES STORMWATER TREATMENT AREAS
The SFWMD in collaboration with the U.S. Environmental Protection Agency

and the Florida Department of Environmental Protection developed the Science
Plan for the Everglades Stormwater Treatment Areas (STAs; SFWMD, 2013d) to
investigate critical factors that regulate the sustainable removal of phosphorus by
STAs. The science plan is intended to support a $50 million water quality research
program over the next 10 years. The science plan identified several key questions
(Box 7-1) that need to be addressed to improve the understanding of various physi-
cal, chemical, and biological factors regulating the total phosphorus concentration
in STA outflows and research and monitoring efforts to address them. Examples of
proposed research include studies on the effects of inflow phosphorus concentra-
tions and loads, uptake of phosphorus by vegetation, microbial activity in soils and
the water column, and the stability of accreted phosphorus in soil compartments.
The SFWMD plans to use the results of these investigations to improve the design
and operations of STAs to achieve compliance with the total phosphorus water
quality-based effluent limit (WQBEL).4 Thus, the primary objective of this Science
Plan is to improve understanding of the external and internal drivers that regulate
the performance of STAs at low phosphorus concentration.
Overall, the Science Plan is comprehensive and well developed to meet
general operational goals of the STAs. Additional comments and suggestions
regarding the six key research questions are provided in Box 7-1. There are also
many interesting and useful science subquestions identified, but their usefulness
in developing improved STA management strategies needs additional consid-
eration. One overarching concern is the single-minded focus on phosphorus
cycling in the Science Plan, to the detriment of important analyses of the role
of other macroelements (carbon, nitrogen, and sulfur) on the regulation of total
phosphorus in STA outflows. It is critical to recognize the importance of coupled
biogeochemical cycles of these macroelements in regulating sustained perfor-
mance of STAs. Additionally, the Science Plan does not include any discussion
on the influence of extreme events such as hurricanes and severe droughts.
Currently, 60 percent of the STA treatment is in submerged aquatic vegetation,
which has been shown to be more prone to disturbances from extreme events.
4
The WQBEL is a numeric discharge limit used to regulate permitted discharges from the STAs
so as not to exceed a long-term geometric mean of 10 μg/L within the Everglades Protection Area.
This numeric value is now translated into a flow-weighted mean (FWM) total phosphorus (TP) con-
centration and applied to each STA discharge points, which now must meet the following: (1) the
STAs are in compliance with WQBEL when the TP concentration of STA discharge point does not
exceed an annual FWM of 13 μg/L in more than three out of five years, and (2) annual FWM of
19 μg/L in any water year (Leeds, 2014).

BOX 7-1
Reflections on Key Questions from the Science Plan for the
Everglades Stormwater Treatment Areas
Key Question 1: How can the flow equalization basins (FEBs) be designed and
operated to moderate and optimize phosphorus concentrations, phosphorus
loading rates, and hydraulic loading rates entering the STAs, possibly in combi-
nation with water treatment technologies, and/or inflow canal dredging/lining?
The Restoration Strategies program relies heavily on FEBs to improve the opera-
tion of STAs. Depending on water depth and residence time, FEBs can function both
as sources and sinks for nutrients, especially phosphorus. The FEBs may also respond
differently to low flows and high flows. The proposed research and monitoring plan will
provide new data that will be useful for implementing appropriate adaptive manage-
ment plans to support the design and operation of FEBs for maximum effectiveness.
Properly managed FEBs will potentially reduce inflow total phosphorus concentrations,
thus reducing loads to STAs.
Key Question 2: How can internal loading of phosphorus to the water column be
reduced or controlled, especially in the lower reaches of the treatment trains?
Microbial, periphyton, and vegetation communities are the major ecosystem biotic
components that respond to and exert reciprocal control on abiotic drivers and in doing
so generate biogeochemical cycles that may influence STA outflow total phosphorus
(TP) concentrations. For effective management of STAs, it is critical to understand the
internal biogeochemical dynamics of biotic and abiotic transformations in water, soils,
and periphyton that regulate low TP levels as proposed in the Science Plan.
For the past two decades, state and federal agencies and universities have accu
mulated a wealth of data on internal dynamics of microorganisms, periphyton, and
vegetation (SAV and EAV) in the Everglades Protection Area (WCAs and Everglades
National Park) and their role in regulating low TP concentrations (10 μg TP/L) in surface
waters (see Reddy et al., 2011, for a compilation of review papers). Although STAs are
operated at much higher flow rates and TP loading rates than the rates encountered in
the Everglades, contrasting these two ecosystems (STAs and the Everglades Protec-
tion Area) can provide insights to develop strategies to manage STA outflows for low
TP concentrations.
Key Question 3: What measures can be taken to enhance vegetation-based treat-

ment in STAs and FEBs?
The role of biotic communities in assimilating phosphorus from the soil and water
column is well known. In addition to assimilating phosphorus into their tissues, these
Useful long-term systemwide monitoring requires stable funding. If funding

cuts result in significant gaps in critical long-term monitoring data, important
changes and patterns could be missed, and data collected before or after the

biotic communities will also alter the micro- and macroenvironment in the water column
and soils and influence phosphorus retention and release. Several studies are proposed
in the Science Plan to understand the role of vegetation to reduce phosphorus concen-
tration of the water column, with primary focus on phosphorus assimilation and storage
in the vegetation. However, additional attention to nutrient balance (macro- and micro-
nutrients) and abiotic and biotic reactions that may be more important in regulating
phosphorus retention is merited.
Key Question 4: How can the biogeochemical and/or physical mechanisms be

managed to further reduce soluble reactive, particulate and dissolved organic
phosphorus concentrations at the outflow?
Very little is known on transformations of particulate phosphorus (PP) and dissolved
organic phosphorus (DOP) within various treatment cells. The challenge for STA opti-
mization is to develop innovative management strategies to reduce internal production
of PP and DOP. The Science Plan identifies various technologies to reduce PP and
DOP, and some of these technologies are currently being tested. Although this practi-
cal approach is important and useful, it is equally important to conduct some studies
that will determine the role of physical, chemical, and biological processes, vegetation
types, and hydraulic loading rates on internal production of PP and DOP. This informa-
tion will provide support to determine the type of technologies needed reduce outflow
PP and DOP.
Key Question 5: What operational and/or design refinements could be imple

mented at existing STAs and future features (i.e., STA expansions, flow equaliza-
tion basins) to improve and sustain treatment performance?
The Science Plan identifies the importance of some operational and/or design refine-
ments to STA to improve treatment performance. Examples of some operational/design
refinements may include: managing high flows and low flows by taking advantage of
FEBs and altering hydraulic retention times in treatment cells; sediment management in
inflows and outflows; minimizing short-circuiting and improving flow distribution; induc
ing downward flow in STAs to reduce upward flux of phosphorus. These strategies may
provide some operational flexibility to improve the overall performance of STAs to reduce
TP levels in outflow.
Key Question 6: What is the influence of wildlife and fisheries on the reduction
of phosphorus in the STAs?
Wildlife (birds, fish, alligators, macro-crustaceans, mollusks, and others) can be
a significant factor in phosphorus loading to STAs, especially in treatment cells near
outflows. It is important to determine direct and indirect effects of wildlife of the extent
of phosphorus loading and its ultimate impact on outflow TP concentrations.
funding gaps could lose their value. Given the substantial financial investment in
Everglades restoration by both the state and the federal governments, a dedicated
source of funding could provide ongoing long-term systemwide monitoring and
assessment that is critical to meeting restoration objectives, ensuring that public
resources are spent wisely, and adaptively managing restoration efforts.

A comprehensive reevaluation of restoration-related monitoring is needed

to determine its adequacy considering budget pressures, the extended CERP
implementation time frames, and the potential impacts of climate change and
sea-level rise. The dramatic 2011 cuts to MAP funding create a risk that adequate
long-term data will not be available to assess the effects of restoration projects
in a systemwide context once they are implemented. This reevalution should
clearly articulate the value of the highest priority monitoring to future restoration
decision making and the risks of ceasing such monitoring. Also, CERP planners
should identify opportunities for improving the efficiency of current monitoring
and reducing the frequency of some monitoring in the context of the current
slow pace of CERP implementation.
Renewed attention to science coordination is warranted. Scientific research
and monitoring programs require coordination and communication to be effec-
tive and efficient, but science leadership and coordination appear to have
waned over the past few years. For the SCG to significantly contribute to better
science coordination, the SCG would need to have adequate funding and staff
and a clear charge to address critical science needs from a restoration-wide
perspective.
In recent years, project-level adaptive management plans have become
more sophisticated and better integrated with project planning as guidance has
been developed and refined. After calls for adaptive management since 2000,
significant progress has been made toward adaptive management planning at
multiple scales. The Central Everglades Planning Project adaptive management
plan is the most complex and sophisticated to date.
The 2014 System Status Report is an effective synthesis of recent monitor-
ing and research and provides valuable science-based guidance to restoration
decision makers. Its key findings summarize ecosystem status and trends, moni-
toring related to implemented CERP and non-CERP projects, and new science
relevant at local and systemwide scales. A particular strength of the document
is its explanations of ecosystem trends and their causal mechanisms that lead
to recommendations for possible changes in project design or operations to
improve restoration outcomes.
Implementation of the Restoration Strategies Science Plan to develop strat-
egies to meet STA discharge criteria is a high priority for Everglades restoration.
The Science Plan and associated $50 million research program is an important
contribution that should improve STA management and effectiveness. However,
the single-minded focus on phosphorus in the Science Plan may overlook the
influence of other macroelements such as carbon, nitrogen, and sulfur on sus-
tained STA performance.

References
Adams, P. N., N. D. Opdyke, and J. M. Jaeger. 2010. Isostatic uplift driven by karstification and
sea-level oscillation: modeling landscape evolution in north Florida. Geology 38(6):531-534.
Available at http://geology.gsapubs.org/content/38/6/531.abstract [accessed May 21, 2014].
Albins, M. A., and M. A. Hixon. 2008. Invasive Indo-Pacific lionfish (Pterois volitans) reduce recruit-
ment of Atlantic coral-reef fishes. Marine Ecology Progress Series 367:233-238. Available at
http://www.vetofish.com/sites/vetofish.com/files/actu/attachment/invasive_indo-pacific_lionfish_
pterois_volitans_reduce_recruitment_of_atlantic_coral-reef_fishes.pdf [accessed May 21, 2014].
Albins, M. A., and M. A. Hixon. 2013. Worst case scenario: potential long-term effects of inva-
sive predatory lionfish (Pterois volitans) on Atlantic and Caribbean coral-reef communities.
Environmental Biology of Fishes 96:1151-1157. Available at http://link.springer.com/article/
10.1007%2Fs10641-011-9795-1 [accessed May 21, 2014].
Andreotta, H., M. Chimney, T. DeBusk, B. Garrett, G. Goforth, K. Grace, D. Ivanoff, M. Jerauld,
M. Kharbanda, N. Larson, S-L. Miao, D. Marois, W. Mitsch, T. Piccone, K. Pietro, N. Ralph,
A. Ramirez, L. Schwartz, L. Toth, S. Xue, Y. Yan, M. Zamorano, L. Zhang, and H. Zhao.
2014. Performance of the Everglades Stormwater Treatment Areas. Chapter 5B in 2014 South
Florida Environmental Report, Volume I: The South Florida Environment. West Palm Beach:
South Florida Water Management District. Available at http://my.sfwmd.gov/portal/page/portal/
pg_grp_sfwmd_sfer/portlet_prevreport/2014_sfer/v1/chapters/v1_ch5b.pdf.
Anonymous. 2005. Monster machines. Wildland Weeds 9(1):12-13.
Audubon Florida. 2014. The Southern Everglades and Florida Bay: Audubon Scientists Find Progress
One Year After C-111 Spreader Canal Western Component Ribbon-Cutting. Available at http://
fl.audubon.org/sites/default/files/documents/audubon_c111_factsheet_jan2014.pdf [accessed
May 21, 2014].
Aumen, N. G., K. E. Havens, G. R. Best, and L. Berry. In press. Predicting ecological responses of
the Florida Everglades to possible future climate scenarios: Introduction.
Bahm, K. E., and R. J. Fennema. 2013. Evaluation and Application of the TIME Model v. 2.0:
Restoration Alternatives and Sea Level Rise in Everglades National Park. Hydrologic Model
Report. SFNRC Technical Series 2013:1. South Florida Natural Resources Center, Everglades
National Park, Homestead, FL. Available at http://www.nps.gov/ever/naturescience/upload/
Time_complete_DOC.pdf [accessed May 21, 2014].
Barnes, M. A., R. K. Fordham, R. L. Burks, and J. J. Hand. 2008. Fecundity of the exotic apple snail,
Pomacea insularum. Journal of the North American Benthological Society 27:738-745. Avail-
able at http://www.bioone.org/doi/abs/10.1899/08-013.1 [accessed May 21, 2014].
Barr, J. G., M. S. DeLonge, and J. D. Fuentes. 2014. Seasonal evapotranspiration patterns in man-
grove forests. Journal of Geophysical Research: Atmospheres 119(7):3886-3899.
Barrios-Garcia, M. N., and S. A. Ballari. 2012. Impact of wild boar (Sus scrofa) in its introduced and
native range: a review. Biological Invasions 14:2283-2300. Available at http://link.springer.com/
article/10.1007%2Fs10530-012-0229-6 [accessed May 21, 2014].
229
Bates, A. L., W. H. Orem, J. W. Harvey, and E. C. Spiker. 2000. Tracing sources of sulfur in the
Florida Everglades. Journal of Environmental Quality 31:287-299.
Bemer, R. A., 1980. Early Diagenesis: A Theoretical Approach. Princeton, NJ: Princeton University
Press. Available at http://press.princeton.edu/titles/121.html [accessed May 21, 2014].
Bender, M. A., T. R. Knutson, R. E. Tuleya, J. J. Sirutis, G. A. Vecchi, S. T. Garner, and I. M. Held. 2010.
Modeled impact of anthropogenic warming on the frequency of intense Atlantic hurricanes.
Science 327:454-458. Available at http://www.sciencemag.org/content/327/5964/454.full [ac-
cessed May 21, 2014].
Bennett, D. 1998. Monitor Lizards: Natural History, Biology & Husbandry. Frankfurt, Germany:
Chimaira.
Bianchette, T. A., K. B. Lu, N. S. Lam, and L. M. Kiage. 2009. Effects of Hurricanes Ivan and Katrina
on the coastal ecosystems of Weeks Bay, Alabama, U.S.A. Presentation at 10th International
Coastal Symposium, Lisbon, Portugal.
Bjorkman, O., B. Mahall, M. Nobs, W. Ward, F. Nicholson, and H. Mooney. 1974. An analysis of
the temperature dependence of growth under controlled conditions. Pp. 757-767 in Carnegie
Institution of Washington, Yearbook 73, S. A. McGough, ed. Baltimore: J. D. Lucas.
Blake, E. S., C. W. Landsea, and E. J. Gibney. 2011. The Deadliest, Costliest, and Most Intense United
States Tropical Cyclones from 1851 to 2006 (and Other Frequently Requested Hurricane Facts).
National Oceanic and Atmospheric Administration Technical Memorandum. Available at
http://www.nhc.noaa.gov/pdf/nws-nhc-6.pdf [accessed May 21, 2014].
Blake, N. 1980. Land into Water—Water into Land: A History of Water Management in Florida.
Tallahassee: University Presses of Florida.
Bodle, M. J., A. P. Ferriter, and D. D. Thayer. 1994. The biology, distribution, and ecological conse-
quences of Melaleuca quinquenervia in the Everglades. Pp. 341-356 in Everglades: The Eco-
system and Its Restoration, S. M. Davis and J. C. Ogden, eds. Boca Raton, FL: St. Lucie Press.
Bouillon, S., A. V. Borges, E. Castañeda-Moya, K. Diele, T. Dittmar, N. C. Duke, E. Kristensen, S. Y.
Lee, C. Marchand, J. J. Middelburg, V. H. Rivera Monroy, T. J. Smith III, and R. R. Twilley.
2008. Mangrove production and carbon sinks: a revision of global budget estimates. Global
Biogeochemical Cycles 22(2). doi:10.1029/2007GB003052. Available at http://onlinelibrary.
wiley.com/doi/10.1029/2007GB003052/abstract [accessed May 21, 2014].
Brandt, L. A., K. M. Portier, and W. M. Kitchens. 2000. Patterns of change in tree islands in Arthur R.
Marshall Loxahatchee National Wildlife Refuge from 1950 to 1991. Wetlands 20:1-14. Available
at http://link.springer.com/article/10.1672%2F0277-5212(2000)020%5B0001%3APOCITI%
5D2.0.CO%3B2 [accessed May 21, 2014].
Brewer, S. 2008. Declines in plant species richness and endemic plant species in longleaf pine
savannas invaded by Imperata cylindrica. Biological Invasions 10:1257-1264. Available at
http://link.springer.com/article/10.1007%2Fs10530-007-9200-3 [accessed May 21, 2014].
Brinson, M. M., R. R. Christian, and L. K. Blum. 1995. Multiple states in the sea-level induced tran-
sition from terrestrial forest to estuary. Estuaries 18:648-659. Available at http://link.springer.
com/article/10.2307%2F1352383 [accessed May 21, 2014].
Bush, E. 2013. Central Everglades Planning Project Update to the South Florida Ecosystem Restora-
tion Task Force, December 17, 2013. Available at http://www.sfrestore.org/tf/minutes/2013_
meetings/121713/Central_Everglades_Planning_Project_Update.pdf [accessed May 21, 2014].
Byrd, J. D., Jr., and C. T. Bryson. 1999. Biology, Ecology, and Control of Cogongrass [Imperata
cylindrica (L.) Beauv.]. Mississippi Department of Agriculture and Commerce. Bureau of Plant
Industry Fact Sheet 1999-01.
Cahoon, D. R. 2006. A review of major storm impacts on coastal wetland elevations. Estuaries
and Coasts 29(6A):889-898. Available at http://www.pwrc.usgs.gov/prodabs/pubpdfs/6702_
cahoon.pdf [accessed May 21, 2014].
Cahoon, D. R., and J. C. Lynch. 1997. Vertical accretion and shallow subsidence in a mangrove
forest of southwestern Florida, U.S.A. Mangroves and Salt Marshes 1:173-186. Available at
http://link.springer.com/article/10.1023%2FA%3A1009904816246 [accessed May 21, 2014].

References 231
Cahoon, D. R., D. J. Reed, and J. W. Day. 1995. Estimating shallow subsidence in microtidal salt
marshes of the southeastern United States: Kaye and Barghoorn revisited. Marine Geology
128:1-9. Available at http://www.sciencedirect.com/science/article/pii/002532279500087F
[accessed May 21, 2014].
Cahoon, D. R., P. Hensel, J. Rybczyk, K. L. McKee, E. Proffitt, and B. C. Perez. 2003. Mass tree
mortality leads to mangrove peat collapse at Bay Islands, Honduras after Hurricane Mitch.
Journal of Ecology 91:1093-1105. Available at http://onlinelibrary.wiley.com/doi/10.1046/
j.1365-2745.2003.00841.x/abstract [accessed May 21, 2014].
California Natural Resources Agency. 2009. 2009 California Climate Adaptation Strategy. Avail-
able at http://resources.ca.gov/climate_adaptation/docs/Statewide_Adaptation_Strategy.pdf
Carney, K. M., B. A. Hungate, B. G. Drake, and J. P. Megonigal. 2007. Altered soil microbial com-
munity at elevated CO2 leads to loss of soil carbon. Proceedings of the National Academy of
Sciences of the United States of America 104:4990-4995.
Castañeda-Moya, E., R. R. Twilley, and V. H. Rivera-Monroy. 2013. Allocation of biomass and net
primary productivity of mangrove forests along environmental gradients in the Florida coastal
Everglades, USA. Forest Ecology and Management 307:226-241. Available at http://www.
sciencedirect.com/science/article/pii/S037811271300443X [accessed May 21, 2014].
Catano, C. P., S. S. Romanach, J. M. Beerens, L. G. Pearlstine, L. A. Brandt, K. M. Hart, F. J. M
azzotti,
and J. C. Trexler. In press. Using Scenario Planning to evaluate the impacts of climate change on
wildlife populations and communities in the Florida Everglades. Environmental M anagement.
Cattau, C. E., J. Martin, and W. M. Kitchens. 2010. Effects of an exotic prey species on a native
specialist: example of the snail kite. Biological Conservation 143:513-520. Available at http://
www.sciencedirect.com/science/article/pii/S000632070900487X [accessed May 21, 2014].
Cavanaugh, K. C., J. R. Kellner, A. J. Forde, D. S. Gruner, J. D. Parker, W. Rodriguez, and I. C. Feller.
2014. Poleward expansion of mangroves is a threshold response to decreased frequency of
extreme cold events. Proceedings of the National Academy of Sciences of the United States of
America 111:723-727. Available at http://www.pnas.org/content/early/2013/12/26/1315800111.
abstract [accessed May 21, 2014].
CDM. 2011. Climate Change Handbook for Regional Water Planning. Available at http://www.
water.ca.gov/climatechange/docs/Climate_Change_Handbook_Regional_Water_Planning.pdf
Center, T. D., M. F. Purcell, P. D. Pratt, M. B. Rayamajhi, P. W. Tipping, S. A. Wright, and F. A.
Dray. 2012. Biological control of Melaleuca quinquenervia: an Everglades invader. Biocontrol
57:151-165. Available at http://link.springer.com/article/10.1007%2Fs10526-011-9390-6 [ac-
Center, T. D., M. Rayamajhi, F. A. Dray, P. M. Madeira, G. Witkus, E. Rohrig, E. Mattison, E. Lake,
M. Smith, J. L. Zhang, M. Purcell, A. Konstantinov, and D. Schmitz. 2013. Host range valida-
tion, molecular identification and release and establishment of a Chinese biotype of the Asian
leaf beetle Lilioceris cheni (Coleoptera: Chrysomelidae: Criocerinae) for control of Dioscorea
bulbifera L. in the southern United States. Biocontrol Science and Technology 23:735-755.
Available at http://www.tandfonline.com/doi/abs/10.1080/09583157.2013.790931 [accessed
May 21, 2014].
Chambers, L. G., K. R. Reddy, and T. Z. Osborne. 2011. Short-term response of carbon cycling to
salinity pulses in a freshwater wetland. Soil Science Society of America Journal 75:2000-2007.
Chambers, L. G., S. E. Davis, T. Troxler, J. N. Boyer, A. Downey-wall, and L. J. Scinto. 2013a. Bio-
geochemical effects of simulated sea level rise on carbon loss in Everglades mangrove peat
soil. Hydrobiologia. doi:10.1007/s10750-013-1764-6. Available at http://www.researchgate.
net/publication/259195457_Biogeochemical_effects_of_simulated_sea_level_rise_on_carbon_
loss_in_an_Everglades_mangrove_peat_soil [accessed May 21, 2014].
Chambers, L. G., T. Z. Osborne, and K. R. Reddy. 2013b. Effect of salinity pulsing events on soil
organic carbon loss across an intertidal wetland gradient: a laboratory experiment. Biogeo-
chemistry 115:363-383.

Chambers, L. G., S. E. Davis, T. Troxler, A. Downey-Wall, J. N. Boyer, and L. J. Scinto. 2014.

Biogeochemical effects of simulated sea level rise on carbon loss in an Everglades mangrove
peat soil. Hydrobiologia 726:195-211. Available at http://link.springer.com/article/10.1007%
2Fs10750-013-1764-6 [accessed May 21, 2014].
Chen, R. H., and R. R. Twilley. 1999. Patterns of mangrove forest structure and soil nutrient
dynamics along the Shark River estuary, Florida. Estuaries 22:955-970. Available at http://link.
springer.com/article/10.2307%2F1353075 [accessed May 21, 2014].
Cherry, J. A., K. McKee, and J. B. Grace. 2009. Elevated CO2 enhances biological contributions
to elevation change in coastal wetlands by offsetting stressors associated with sea-level rise.
Journal of Ecology 97:67-77. Available at http://onlinelibrary.wiley.com/doi/10.1111/j.1365-
2745.2008.01449.x/abstract [accessed May 21, 2014].
Chuirazzi, K., and M. Duever. 2008. Picayune Strand Restoration Project baseline. Appendix 7A-2 in
2008 South Florida Environmental Report. West Palm Beach: South Florida Water Management
District. Available at http://www.sfwmd.gov/portal/page/portal/pg_grp_sfwmd_sfer/portlet_sfer/
tab2236041/volume1/appendices/v1_app_7a-2.pdf.
Chuirazzi, K., M. Duever, and W. Abtew. 2012. Annual Permit Report for the Picayune Strand Resto-
ration Project, Phase 1—Prairie Canal Backfill and Road Removal Component. Appendix 2-1 in
2012 South Florida Environmental Report. South Florida Water Management District. Available
at http://www.sfwmd.gov/portal/page/portal/pg_grp_sfwmd_sfer/portlet_prevreport/2012_sfer/
v3/appendices/v3_app2-1.pdf.
Chupp, A. D. and L. L. Battaglia. In press. Potential for host-shifting in Papilio palamedes follow-
ing invasion of Laurel Wilt Disease. Biological Invasions. DOI 10.1007/s10530-014-0693-2.
Clewell, A. C., and J. Aronson. 2013. The SER Primer and climate change. Ecological Manage-
ment and Restoration 14:182-186. Available at http://onlinelibrary.wiley.com/doi/10.1111/
emr.12062/abstract [accessed May 21, 2014].
Clout, M. N., and P. A. Williams. 2009. Invasive Species Management. A Handbook of Principles
and Techniques. New York: Oxford University Press.
Convertino, M. C., M. Foran, and I. Linkov. 2012. Enhanced Adaptive Management: Decision
Analysis Tool to Support CERP Restoration and Monitoring Decisions. Risk and Decision Sci-
ence Team, Environmental Lab, Engineering Research and Development Center, U.S. Army
Corps of Engineers.
Convertino, M., C. M. Foran, J. M. Keisler, L. Scarlett, A. LoSchiavo, G. A. Kiker, and I. Linkov.
2013. Enhanced adaptive management: Integrating decision analysis, scenario analysis and
environmental modeling for the Everglades. Scientific Reports 3:2045-2322. Available at http://
www.nature.com/srep/2013/131011/srep02922/full/srep02922.html.
Craft, C. B., J. Clough, J. Ehman, S. B. Joye, R. Park, S. C. Pennings, H. Guo, and M. Machmuller.
2009. Forecasting the effects of accelerated sea level rise on tidal marsh ecosystem services.
Frontiers in Ecology and the Environment 7(2):73-78. Available at http://www.esajournals.org/
doi/abs/10.1890/070219 [accessed May 21, 2014].
Crain, C. M., B. R. Silliman, S. L. Bertness, and M. D. Bertness. 2004. Physical and biotic d rivers
of plant distribution across estuarine salinity gradients. Ecology 85:2539-2549. Available at
http://bertnesslab.com/docs/labpublications/Crain%20et%20al%202004.pdf [accessed May 21,
2014].
Crooks, J. A. 2011. Lag times. Pp. 404-410 in Encyclopedia of Biological Invasions, D. Simberloff
and M. Rejmánek, eds. Berkeley: University of California Press.
D’Antonio, C. D., and L. A. Meyerson. 2002. Exotic plant species as problems and solutions in
ecological restoration: a synthesis. Restoration Ecology 10:703-713. Available at http://planet.
uwc.ac.za/nisl/Invasives/Assignment1/DAntonioandMeyerson.pdf [accessed May 21, 2014].
Davis, S. M., and J. C. Ogden, eds. 1994. Everglades: The Ecosystem and Its Restoration. Delray
Beach, FL: St. Lucie Press.

References 233
Davis, S. M., D. L. Childers, J. J. Lorenz, H. R. Wanless, and T. E. Hopkins. 2005a. A conceptual model
of ecological interactions in the mangrove estuaries of the Florida Everglades. Wetlands 25:832-
842. Available at http://link.springer.com/article/10.1672%2F0277-5212(2005)025%5B0832%
3AACMOEI%5D2.0.CO%3B2 [accessed May 21, 2014].
Davis, S. M., E. E. Gaiser, W. F. Loftus, and A. E. Huffman. 2005b. Southern marl prairies con-
ceptual ecological model. Wetlands 25:821-831. Available at http://link.springer.com/
article/10.1672%2F0277-5212(2005)025%5B0821%3ASMPCEM%5D2.0.CO%3B2 [accessed
May 21, 2014].
Day, J. W., G. P. Kemp, D. J. Reed, D. R. Cahoon, R. M. Boumans, J. M. Suhayda, and R. G ambrell.
2011. Vegetation death and rapid loss of surface elevation in two contrasting Mississippi
delta salt marshes: the role of sedimentation, autocompaction and sea-level rise. Ecologi-
cal Engineering 37(2):229-240. Available at http://www.sciencedirect.com/science/article/pii/
S0925857410003150 [accessed May 21, 2014].
de Boer, H. J., E. I. Lammertsma, F. Wagner-Cremer, D. L. Dilcher, M. J. Wassen, and S. C. Dekker.
2011. Climate forcing due to optimization of maximal leaf conductance in subtropical vegeta-
tion under rising CO2. Proceedings of the National Academy of Sciences of the United States
of America, doi:10.1073/pnas.1100555108.
DeBusk, W. F., and K. R. Reddy. 1998. Turnover of detrital organic carbon in a nutrient-impacted
Everglades marsh. Soil Science Society of American Journal 62:1460-1468. Available at https://
www.soils.org/publications/sssaj/abstracts/62/5/SS0620051460 [accessed May 21, 2014].
Deegan, L. A., D. S. Johnson, R. S. Warren, B. J. Peterson, J. W. Fleeger, S. Fagherazzi, and W. M.
Wollheim. 2012. Coastal eutrophication as a driver of salt marsh loss. Nature 490:388-392.
Available at http://www.nature.com/nature/journal/v490/n7420/full/nature11533.html [ac-
DeGaetano, A. T., and R. J. Allen. 2002. Trends in twentieth-century temperature extremes across
the United States. Journal of Climate 15:3188-3205. Available at http://faculty.ucr.edu/~rjallen/
Publications_files/DeGaetanoAlJClim02.pdf [accessed May 21, 2014].
Deis, D. 2011. Pre-drainage flows and salinities in coastal systems: effects of flow and rainfall on
salinity and seagrass in Florida Bay and estimates of freshwater flow requirements to achieve
restoration. Pp. 11-20 in Scientific and Technical Knowledge Gained in Everglades Restoration
(1999-2009). Restoration Coordination and Verification. Available at http://www.evergladesplan.
org/shared-definition/shared_def_docs/sd_2010/081811_skd/081811_skd_complete.pdf.
DeLaune, R. D., J. A. Nyman, and W. H. Patrick, Jr. 1994. Peat collapse, ponding, and wetland loss
in a rapidly submerging coastal marsh. Journal of Coastal Research 10:1021-1030. Available
at http://www.rnr.lsu.edu/people/nyman/Journal%20of%20Coastal%20Research%2010.4%20
(1994).pdf [accessed May 21, 2014].
DiTomaso, J. M. 2011. Herbicides. Pp. 323-330 in Encyclopedia of Biological Invasions, D.
Simberloff and M. Rejmánek, eds. Berkeley: University of California Press.
Doering, P. H. 1996. Temporal variability of water quality in the St. Lucie Estuary, South Florida.
Journal of the American Water Resources Association 32:1293-1306. Available at http://
onlinelibrary.wiley.com/doi/10.1111/j.1752-1688.1996.tb03498.x/abstract [accessed May 21,
2014].
Doering, P. H., and R. H. Chamberlain. 1999. Water quality and source of freshwater discharge
to the Caloosahatchee Estuary, Florida. Journal of the American Water Resource Associa-
tion 35:793-806. Available at http://onlinelibrary.wiley.com/doi/10.1111/j.1752-1688.1999.
tb04175.x/abstract [accessed May 21, 2014].
DOI (U.S. Department of the Interior). 2005. Science Plan in Support of Ecosystem Restoration,
Preservation, and Protection in South Florida. Available at http://sofia.usgs.gov/publications/
reports/doi-science-plan/2005-DOI-Science-Plan-final.pdf.
DOI. 2013. Secretary Jewell Commends Florida Governor Rick Scott’s Commitment to Support Next
Phase of Bridging for Tamiami Trail. Press release, August 28.

DOI and USACE (U.S. Department of the Interior and U.S. Army Corps of Engineers). 2005. Central
and Southern Florida Project Comprehensive Everglades Restoration Plan: 2005 Report to
Congress. Available at http://www.evergladesplan.org/pm/program_docscerp_report_congress_
2005.cfm [accessed May 21, 2014].
DOI and USACE. 2011. Comprehensive Everglades Restoration Plan Central and Southern Florida
Project: 2010 Report to Congress. Available at http://www.evergladesplan.org/pm/program_
docs/cerp_reports_congress.aspx [accessed January 13, 2012].
Donnelly, J., and M. D. Bertness. 2001. Rapid shoreward encroachment of salt marsh vegetation
in response to sea-level rise. Proceedings of the National Academy of Sciences of the United
States of America 98:14218-14223. Available at http://www.pnas.org/content/98/25/14218.
long [accessed May 21, 2014].
Dorcas, M. E., and J. D. Willson. 2011. Invasive Pythons in the United States. Athens: University
of Georgia Press.
Dorcas, M. E., and J. D. Willson. 2013. Hidden giants: problems associated with studying secretive
invasive pythons. Pp. 367-385 in Reptiles in Research, W. I. Lutterschmidt, ed. New York:
Nova Science.
Dorcas, M. E., J. D. Willson, R. N. Reed, R. W. Snow, M. R. Rochford, M. A. Miller, W. E. Meshaka,
P. T. Andreadis, F. J. Mazzotti, C. M. Romagosa, and K. M. Hart. 2012. Severe mammal declines
coincide with proliferation of Burmese pythons in Everglades National Park. Proceedings of the
National Academy of Sciences of the United States of America 109:2418-2422. Available at
http://www.pnas.org/content/early/2012/01/23/1115226109 [accessed May 21, 2014].
Doren, R. F., and D. T. Jones. 1997. Plant management in Everglades National Park. Pp. 275-286
in Strangers in Paradise: Impact and Management of Nonindigenous Species in Florida, D.
Simberloff, D. C. Schmitz, and T. C. Brown, eds. Washington, DC: Island Press.
Dove, C. J., R. W. Snow, M. R. Rochford, and F. J. Mazzotti. 2011. Birds consumed by the invasive
Burmese python (Python molurus bivittatus) in Everglades National Park, Florida, USA. W ilson
Journal of Ornithology 123:126-131. Available at http://userwww.sfsu.edu/parker/bio840/
pdfs/2012StudentPapers/DaveTannaci/Dove-BirdsConsumed.pdf [accessed May 21, 2014].
Doyle, T. W., T. J. Smith III, and M. B. Robblee. 1995. Wind damage effects of Hurricane Andrew
on mangrove communities along the southwest coast of Florida, USA. Journal of Coastal
Research 21(Special Issue):159-168. Available at http://sflwww.er.usgs.gov/publications/papers/
wind_damage_mang/Wind_Damage_Mangroves.pdf [accessed May 21, 2014].
Drake, B. G., M. S. Muehe, G. Peresta, M. A. Gonzalez-Meler, and R. Matamala. 1996. Acclimation
of photosynthesis, respiration and ecosystem carbon flux of a wetland on Chesapeake Bay,
Maryland to elevated atmospheric CO2 concentration. Plant and Soil 187:111-118.
ECISMA (Everglades Cooperative Invasive Species Management Area). 2009. Early Detection and
Rapid Response Plan, 2009-2011. Available at http://www.evergladescisma.org/ECISMA_
EDRRPlan_2009-2011.pdf [accessed May 21, 2014].
Egler, F. E. 1952. Southeast saline Everglades vegetation, Florida, and its management. Vegatation
3:213-265. Available at http://link.springer.com/article/10.1007%2FBF00539820 [accessed
May 21, 2014].
Enfield, D. B., A. M. Mestas-Nunez, and P. J. Trimble. 2001. The Atlantic Multidecadal Oscillation
and its relationship to rainfall and river flows in the continental U.S. Geophysical Research
Letters 28:2077-2080.
Enfield, D., S. K. Lee, F. Marks, and M. Powell. 2011. Mid-century expectations for tropical cyclone
activity and Florida rainfall. Pp. 11-18 in Climate Scenarios: A Florida-Centric View, V. Misra,
E. Carlson, R. K. Craig, and D. Enfield, eds. Florida Climate Change Task Force. Available at
http://floridaclimateorg/whitepapers/.
Epanchin-Niell, R. S., R. G. Haight, L. Berec, J. M. Kean, and A. M. Liebhold. 2012. Optimal surveil-
lance and eradication of invasive species in heterogeneous landscapes. Ecology Letters 15:803-
812. Available at http://www.ncbi.nlm.nih.gov/pubmed/22642613 [accessed May 21, 2014].

References 235
Erickson, J. E., J .P. Megonigal, G. Peresta, and B. G. Drake. 2007. Salinity and sea level mediate
elevated CO2 effects on C3–C4 plant interactions and tissue nitrogen in a Chesapeake Bay
tidal wetland. Global Change Biology 13:202-215. Available at http://onlinelibrary.wiley.com/
doi/10.1111/j.1365-2486.2006.01285.x/abstract [accessed May 21, 2014].
Erickson, J. E., G. Peresta, K. J. Montovan, and B. G. Drake. 2013. Direct and indirect effects of
elevated atmospheric CO2 on net ecosystem production in a Chesapeake Bay tidal wetland.
Global Change Biology 19:3368-3378. Available at http://onlinelibrary.wiley.com/doi/10.1111/
gcb.12316/abstract [accessed May 21, 2014].
Ervin, G. N., and R. G. Wetzel. 2002. Effects of sodium hypochlorite sterilization and cold dry
storage on germination of Juncus effuses L. Wetlands 22:191-195. Available at http://link.
springer.com/article/10.1672%2F0277-5212(2002)022%5B0191%3AEOSHSA%5D2.0.CO%
3B2 [accessed May 21, 2014].
Evans, C. W., D. J. Moorhead, C. T. Bargeron, and G. K. Douce. 2007. Implementation of control
and prevention strategies for managing cogongrass (Imperata cylindrica) by the Georgia Inva-
sive Species Task Force. Natural Areas Journal 27:226-231. Available at http://www.bioone.org/
doi/abs/10.3375/0885-8608(2007)27%5B226:IOCAPS%5D2.0.CO%3B2 [accessed May 21,
2014].
Evans, J. M. 2013. Pistia stratiotes L. in the Florida Peninsula: biogeographic evidence and con-
servation implications of native tenure for an “invasive” aquatic plant. Conservation and
Society 11:233-246. Available at http://www.conservationandsociety.org/article.asp?issn=0972-
4923;year=2013;volume=11;issue=3;spage=233;epage=246;aulast=Evans [accessed May 21,
2014].
Ewe, S. M. L., and da L. Sternberg. 2002. Seasonal water-use by the invasive exotic, Schinus
terebinthifolius, in native and disturbed communities. Oecologia 133:441-448. Available
at http://link.springer.com/article/10.1007%2Fs00442-002-1047-9 [accessed May 21, 2014].
Ewe, S. M. L., W. A. Overholt, L. G. Kirton, E. M. Lai, I. Ahmad, and S. Ulaganathan. 2006. Foreign
exploration for biological control agents of three invasive plant species from Asia. Wildland
Weeds (Fall):19-21.
Ewel, J. J. 1986. Invasibility: lessons from south Florida. Pp. 214-230 in Ecology of Biological Inva-
sions of North America and Hawaii, H. Mooney and J. Drake, eds. New York: Springer.
Ewel, J. J. 2013. Case study: Hole-in-the-Donut, Everglades. Pp. 11-15 in Novel Ecosystems: Inter-
vening in the New Ecological World Order, R. J. Hobbs, E. S. Higgs, and C. M. Hall, eds. West
Sussex, UK: Wiley-Blackwell.
Ewel, J. J., D. Ojima, D. Karl, and W. Debusk. 1982. Schinus in Successional Ecosystems of Ever-
glades National Park. South Florida Research Center Report T-676. Homestead, FL: Everglades
National Park, National Park Service.
FDACS (Florida Department of Agriculture and Consumer Services). 2012. Florida Department of
Agriculture and Consumer Services identifies laurel wilt disease in avocado production area
of Miami-Dade County: Disease threatens health of avocado trees, a $13 million industry in
South Florida. May 1, 2012. Available at http://www.freshfromflorida.com/News-Events/Press-
Releases/2012-Press-Releases/ [accessed June 23, 2014].
Feral Cat Issue Team. 2003. Issue Assessment: Impacts of Feral and Free-Ranging Domestic Cats on
Wildlife in Florida. Florida Wildlife and Conservation Commission.
Flaig, E. G., and K. R. Reddy. 1995. Fate of phosphorus in the Lake Okeechobee watershed, Florida,
USA: Overview and recommendations. Ecological Engineering 5:127-142. Available at http://
www.sciencedirect.com/science/article/pii/0925857495000216 [accessed May 21, 2014].
Fleshler, D. 2014. Everglades project produces quick success. Ft. Lauderdale Sun Sentinel.
January 12. Available at http://www.sun-sentinel.com/news/palm-beach/fl-everglades-canal-
success-20140112,0,7309093.story [accessed March 14, 2014].
Florida Fish and Wildlife Conservation Commission. 2012. Tegus in Florida: How You Can Help Stop
the Spread of an Invasive Lizard. Available at http://myfwc.com/media/2380549/tegu-brochure.
pdf [accessed May 2014].

Florida Fish and Wildlife Conservation Commission. 2014. Nonnative species: Reptiles: Spectacled
Caiman. Available at http://myfwc.com/wildlifehabitats/nonnatives/reptiles/spectacled-caiman/
Fowler, A. J., D. M. Lodge, and J. F. Hsia. 2007. Failure of the Lacey Act to protect US ecosystems
against animal invasion. Frontiers in Ecology and the Environment 5(7):353-359. Available at
http://www.jstor.org/discover/10.2307/20440700?uid=30374&uid=3739584&uid=2129&uid=
2&uid=70&uid=3&uid=30373&uid=67&uid=62&uid=3739256&sid=21103788356721 [ac-
Fraedrich, S. W., T. C. Harrington, R. J. Rabaglia, M. D. Ulyshen, A. E. Mayfield, J. L. Hanula, J. M.
Eickwort, and D. R. Miller. 2008. A fungal symbiont of the redbay ambrosia beetle causes a
lethal wilt in redbay and other Lauraceae in the southeastern United States. Plant Disease
92:215-224. Available at http://www.treesearch.fs.fed.us/pubs/29373 [accessed May 21, 2014].
Franks, S. J., A. M. Kral, and P. D. Pratt. 2006. Herbivory by introduced insects reduces growth and
survival of Melaleuca quinquenervia seedlings. Environmental Entomology 35:366-372. Avail-
able at http://www.bioone.org/doi/pdf/10.1603/0046-225X-35.2.366 [accessed May 21, 2014].
FWS (U.S. Fish and Wildlife Service). 1980. Habitat Evaluation Procedures Handbook. Available at
http://www.fws.gov/policy/esmindex.html [accessed April 4, 2014].
FWS. 2010. Biological Opinion on Everglades Restoration Transition Plan, Phase I, November 17.
Vero Beach: South Florida Ecological Services Office.
FWS. 2014. Programmatic Biological Opinion and Select Concurrence for the Central Everglades
Planning Project on Effects to Threatened or Endangered Species and Critical Habitat, April.
Vero Beach:: South Florida Ecological Services Office.
Gaiser, E. E., A. Zafiris, P. L. Ruiz, F. A. C. Tobias, and M. S. Ross. 2006. Tracking rates of ecotone
migration due to salt-water encroachment using fossil mollusks in coastal South Florida.
Hydrobiologia 569:237-257. Available at http://link.springer.com/article/10.1007%2Fs10750-
006-0135-y [accessed May 21, 2014].
Gaiser, E., J. Trexler, and P. Wetzel. 2012. The Everglades. Pp. 231-252 in Wetland Habitats of
North America: Ecology and Conservation Concerns, D. Batzer and A. Baldwin, eds. Berkeley:
University of California Press.
GAO (U.S. Government Accountability Office). 2003. South Florida Ecosystem Restoration: Task
Force Needs to Improve Science Coordination to Increase the Likelihood of Success. Wash-
ington, DC: GAO.
Genovesi, P. 2011. Eradication. Pp. 198-203 in Encyclopedia of Biological Invasions, D. Simberloff
and M. Rejmánek, eds. Berkeley: University of California Press.
Glaser, P. H., J. C. Volin, T. J. Givnish, B. C. S. Hansen, and C. A. Stricker. 2012. Carbon and sediment
accumulation in the Everglades (USA) during the past 4000 years: rates, drivers, and sources of error.
Journal of Geophysical Research—Biogeosciences 117(G03026); doi:0.1029/2011JG001821.
Available at http://onlinelibrary.wiley.com/doi/10.1029/2011JG001821/abstract [accessed
May 21, 2014].
Gleason, P. J., and P. Stone. 1994. Age, origin and landscape evolution of the Everglades peatland.
Pp. 149-198 in Everglades: The Ecosystem and Its Restoration, S. M. Davis and J. C. Ogden,
eds. Delray Beach, FL: St. Lucie Press.
Gordon, D. R. 1998. Effects of invasive, non-indigenous plant species on ecosystem processes:
lessons from Florida. Ecological Applications 8:975-989. Available at http://www.esf.edu/efb/
schulz/seminars/gordon.pdf [accessed May 21, 2014].
Gordon, D. R., and K. P. Thomas. 1997. Florida’s invasion by nonindigenous plants: history, screen-
ing, and regulation. Pp. 21-37 in Strangers in Paradise: Impact and Management of Non
indigenous Species in Florida, D. Simberloff, D. C. Schmitz, and T. C. Brown, eds. W ashington,
DC: Island Press.
Grunwald, M. 2006. The Swamp: The Everglades, Florida, and the Politics of Paradise. New York:
Simon and Schuster.

References 237
Gunderson, L. H. 2000. Ecological resilience—in theory and application. Annual Review of Ecol-
ogy and Systematics 31:425-439. Available at http://www.annualreviews.org/doi/abs/10.1146/
annurev.ecolsys.31.1.425?journalCode=ecolsys.1 [accessed May 21, 2014].
Guntenspergen, G. R., D. R. Cahoon, J. B. Grace, G. D. Steyer, S. Fournet, M. A. Townson, and A. L.
Foote. 1995. Disturbance and recovery of the Louisiana coastal marsh landscape from the im-
pacts of Hurricane Andrew. Journal of Coastal Research SI(21):324-339. Available at http://www.
jstor.org/discover/10.2307/25736018?uid=30374&uid=3739584&uid=2129&uid=2&uid=70&
uid=3&uid=30373&uid=67&uid=62&uid=3739256&sid=21103788356721 [accessed May 21,
2014].
Hanak, E., and J. R. Lund. 2012. Adapting California’s water management to climate change.
Climatic Change 111(1):17-44. Available at http://dx.doi.org/10.1007/s10584-011-0241-3 [ac-
Hansen, J., M. Sato, and R. Ruedy. 2012. Perception of climate change. Proceedings of the National
Academy of Sciences of the United States of America 109(37):E2415-E2423. Available at http://
www.pnas.org/content/early/2012/07/30/1205276109.abstract [accessed May 21, 2014].
Harbaugh, A. W., E. R. Banta, M. C. Hill, and M. G. McDonald. 2000. MODFLOW-2000, the U.S.
Geological Survey Modular Ground-Water Model–User Guide to Modularization Concepts
and the Ground-Water Flow Process. Open-File Report 00-92. Reston, VA: U.S. Geological
Survey.
Havens, K. E., and A. D. Steinman. 2013. Ecological responses of a large shallow lake (Okeechobee,
Florida) to climate change and potential future hydrologic regimes. Environmental Manage-
ment; doi:10.1007/s00267-013-0189-3.
Hawkins, E., and R. Sutton. 2009. The potential to narrow uncertainty in regional climate predic-
tions. Bulletin of the American Meteorological Society 90:1095-1107. Available at http://
journals.ametsoc.org/doi/abs/10.1175/2009BAMS2607.1 [accessed May 21, 2014].
Heater, C., and J. Maytok. 2014. Five-year capital improvements plan. Chapter 4 in 2014 South
Florida Environmental Report, Volume II: District Annual Plans and Reports. West Palm Beach:
pg_grp_sfwmd_sfer/portlet_prevreport/2014_sfer/v2/chapters/v2_ch4.pdf.
Hellmann, J. J., J. E. Byers, B. G. Bierwagen, and J. S. Dukes. 2008. Five potential consequences of
climate change for invasive species. Conservation Biology 22(3):534-543. Available at http://
onlinelibrary.wiley.com/doi/10.1111/j.1523-1739.2008.00951.x/pdf [accessed May 21, 2014].
Herring, G., and D. E. Gawlik. 2008. Potential for successful population establishment of the non-
indigenous sacred ibis in the Florida Everglades. Biological Invasions 10:969-976. Available
at http://link.springer.com/article/10.1007%2Fs10530-007-9174-1[accessed May 21, 2014].
Holden, J., P. J. Chapman, and J. C. Labadz. 2004. Artificial drainage of peatlands: hydrological and
hydrochemical process and wetland restoration. Progress in Physical Geography 28:95-123.
Hutchinson, J. T., and K. A. Langeland. 2012. Repeated herbicide application for control of Old
World climbing fern (Lygodium microphyllum) and the effects on nontarget vegetation on Ever-
glade tree islands. Invasive Plant Science and Management 5:477-486. Available at http://www.
bioone.org/doi/abs/10.1614/IPSM-D-12-00015.1?journalCode=ipsm [accessed May 21, 2014].
IPCC (Intergovernmental Panel on Climate Change). 2007a. Contribution of Working Group II to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change, M. L. Parry, O. F.
Canziani, J. P. Palutikof, P. J. van der Linden, and C. E. Hanson, eds. Cambridge, UK and New
York: Cambridge University Press. Available at http://www.ipcc.ch/publications_and_data/ar4/
wg2/en/contents.html.
IPCC. 2007b. Summary for policy makers. In Climate Change 2007. The Physical Science Basis.
Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental
Panel on Climate Change, S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt,
M.Tignor, and H. L. Miller, eds. Cambridge, UK and New York: Cambridge University Press.
Available at http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-spm.pdf.

IPCC. 2013. Changes to the underlying scientific/technical assessment. In Climate Change 2013:
The Physical Science Basis. A Special Report of Working Group I Contribution to the IPCC 5th
Assessment Report. Available at https://www.ipcc.ch/report/ar5/wg1/docs/review/P36Doc4_
WGI-12_Changes-Underlying-Assessment.pdf.
Ivanoff, D., K. Pietro, H. Chen, and L. Gerry. 2013. Performance and optimization of the Everglades
Stormwater Treatment Areas. Chapter 5 in 2013 South Florida Environmental Report, Volume I:
The South Florida Environment. West Palm Beach: South Florida Water Management District.
Jones, B. L., D. J. Colangelo, D. H. Anderson, S. G. Bousquin, M. D. Cheek, C. Carlson, and R. T.
James. 2014. Kissimmee River restoration and basin initiatives. Chapter 9 in 2014 South
Florida Environmental Report, Volume I: The South Florida Environment. West Palm Beach:
pg_grp_sfwmd_sfer/portlet_prevreport/2014_sfer/v1/chapters/v1_ch9.pdf.
Jones, D. T., and R. F. Doren. 1997. The distribution, biology and control of Schinus terebinthifolius
in southern Florida, with special reference to Everglades National Park. Pp 81-93 in Plant
Invasions: Studies from North America and Europe, J. H. Brock, M. Wade, P. Pysek, and D.
Green, eds. Leiden, The Netherlands: Backhuys.
Jud, Z. R., and C. A. Layman. 2012. Site fidelity and movement patterns of invasive lionfish, Pterois
spp., in a Florida estuary. Journal of Experimental Marine Biology and Ecology 414-415:69-74.
Available at http://www.sciencedirect.com/science/article/pii/S0022098112000305 [accessed
May 21, 2014].
Junk, W. J., M. Brown, I. C. Campbell, M. Finlayson, B. Gopals, L. Ramberg, and B. G. Warner.
2006. The comparative biodiversity of seven globally important wetlands: a synthesis. Aquatic
Sciences 68:400-414. Available at http://www.d.umn.edu/~vbrady/WE_website/wetlands101/
WE-readings/Junk2006.pdf [accessed May 21, 2014].
Kauffman, S., D. B. McKey, M. Hossaert-McKey, and C. C. Horvitz. 1991. Adaptations for a two-
phase seed dispersal system involving vertebrates and ants in a hemiepiphytic fig (Ficus
microcarpa: Moraceae). American Journal of Botany 78:971-977. Available at http://www.
jstor.org/discover/10.2307/2445176?uid=30374&uid=3739584&uid=2129&uid=2&uid=70&
uid=3&uid=30373&uid=67&uid=62&uid=3739256&sid=21103788356721 [accessed May 21,
2014].
Kelble, C. R., E. M. Johns, W. K. Nuttle, T. N. Lee, R. H. Smith, and P. B. Ortner. 2007. Salinity pat-
terns of Florida Bay. Estuarine, Coastal, and Shelf Science 71(1):318-334. Available at http://
www.sciencedirect.com/science/article/pii/S0272771406003477 [accessed May 21, 2014].
Kelly, M. 2004. Florida River Flow Patterns and the Atlantic Multidecadal Oscillation. Ecologic
Evaluation Section Report, Southwest Florida Water Management District.
Kendra, P. E., W. S. Montgomery, J. Niogret, and N. D. Epsky. 2013. An uncertain future for
American Lauraceae: A lethal threat from redbay ambrosia beetle and laurel wilt disease (a
review). American Journal of Plant Sciences 4:727-738. Available at http://www.scirp.org/
journal/PaperInformation.aspx?paperID=29498 [accessed May 21, 2014].
Kharin, V. V., F. W. Zwiers, X. Zhang, and G. Hegerl. 2007. Changes in temperature and precipita-
tion extremes in the IPCC ensemble of global coupled model simulations. Journal of Climate
20:1419-1444. Available at http://journals.ametsoc.org/doi/abs/10.1175/JCLI4066.1 [accessed
May 21, 2014].
Kiker, C. F., J. W. Milon, and A. W. Hodges. 2001. Adaptive learning for science based policy: The
Everglades Restoration. Ecological Economics 37:403-416.
Kim, D., D. M. Cairns, and J. Bartholdy. 2011. Winddriven sea-level variation influences dynamics of
salt marsh vegetation. Annals of the Association of American Geographers 101:231-248. Avail-
able at http://www.tandfonline.com/doi/abs/10.1080/00045608.2010.544933?journalCode=
raag20#preview [accessed May 21, 2014].
Kirtman, B., V. Misra, and D. Letson. 2011. Uncertainty of climate projections. Pp. 5-10 in Climate
Scenarios: A Florida-Centric View, V. Misra, E. Carlson, R. K. Craig, and D. Enfield, eds. Florida
Climate Change Task Force. Available at http://floridaclimateorg/whitepapers/.

References 239
Kirwan, M. L., and G. R. Guntenspergen. 2010. The influence of tidal range on the sta-
bility of coastal marshland. Journal of Geophysical Research: Earth Surface
115(F02009); doi:10.1029/2009JF001400. Available at http://onlinelibrary.wiley.com/
doi/10.1029/2009JF001400/abstract [accessed May 21, 2014].
Kirwan, M. L., and J. P Megonigal. 2013. Tidal wetland stability in the face of human impacts
and sea-level rise. Nature 504:53-60. Available at http://www.mangroverestoration.com/pdfs/
Kirwan2013nature%20marsh%20stability.pdf [accessed May 21, 2014].
Kline, J. L., W. F. Loftus, K. Kotun, J. C. Trexler, J. S. Rehage, J. J. Lorenz, and M. Robinson. 2013.
Recent fish introductions into Everglades National Park: An unforeseen consequence of water
management? Wetlands, doi:10.1007/s13157-012-0362-0. Available at http://link.springer.
com/article/10.1007%2Fs13157-012-0362-0 [accessed May 21, 2014].
Koch, G. R., D. L. Childers, P. A. Staehr, R. M. Price, S. E. Davis, and E. E. Gaiser. 2012. Hydrological
conditions control P loading and aquatic metabolism in an oligotrophic, subtropical estuary.
Estuaries and Coasts 35(1):292-307.
Koger, C. H., and C. T. Bryson. 2004. Effect of cogongrass (Imperata cylindrica) extracts on ger-
mination and seedling growth of selected grass and broadleaf species. Weed Technology
18:236-242. Available at http://www.bioone.org/doi/abs/10.1614/WT-03-022R1 [accessed
May 21, 2014].
Kotun, T., and A. Renshaw. 2013. Taylor Slough hydrology: Fifty years of water management.
Wetlands, doi:10.1007/s13157-013-0441-x.
Krauss, K. W., K. L. McKee, C. E. Lovelock, D. R. Cahoon, N. Saintilan, R. Reef, and L. Chen. 2013.
How mangrove forests adjust to rising sea level. New Phytologist 202:19-34. Available at http://
onlinelibrary.wiley.com/doi/10.1111/nph.12605/pdf [accessed May 21, 2014].
Lam, N., K. B. Liu, W. Liang, and T. A. Bianchette. 2011. Effects of hurricanes on the Gulf Coast
ecosystems: a remote sensing study of land cover change around Weeks Bay, Alabama. Journal
of Coastal Research 64:1707-1711
Lammertsma, E., H. J. de Boer, S. C. Dekker, D. L. Dilcher, A. F. Lotter, and F. Wagner-Cremer. 2011.
Reduced maximum stomatal conductance in Florida vegetation under anthropogenic CO2
increase. Proceedings of the National Academy of Sciences of the United States of America
1108(1):4035-4040.
Langevin, C., E. Swain, and M. Wolfert. 2005. Simulation of integrated surface-water/ground-water
flow and salinity for a coastal wetland and adjacent estuary. Journal of Hydrology 314(1):212-
234. Available at http://www.sciencedirect.com/science/article/pii/S0022169405001575 [ac-
Langevin, C. D., W. B. Shoemaker, and W. Guo. 2003. SEAWAT-2000, the U.S. Geological Sur-
vey Modular Ground-Water Model–Documentation of the SEAWAT-2000 Version with the
Variable-Density Flow Process (VDF) and the Integrated MT3DMS Transport Process (IMT).
U.S. Geological Survey Open-File Report 03-426.
Laroche, F. 2013. Update: FWC Funding for Exotic Plant Control in Loxahatchee NWR. Presentation
to the South Florida Water Management District Governing Board, December 12.
Laroche, F. B., and J. McKim. 2004. Cost comparison of melaleuca treatment methods. Wildland
Weeds 7(2):12-15.
Larsen, L. G., J. W. Harvey, G. B. Noe, and J. P. Crimaldi, 2009. Predicting organic floc transport
dynamics in shallow aquatic ecosystems: Insights from the field, the laboratory, and numeri-
cal modeling. Water Resources Research 45(W01411), doi: 10.1029/2008WR007221. Avail-
able at http://water.usgs.gov/nrp/jharvey/pdf/Larsen_etal_WRR_2008WR007221.pdf [accessed
May 21, 2014].
Layman, C. A., Z. R. Jud, and P. Nichols. 2014. Lionfish alter benthic invertebrate assemblage.
Submitted to Marine Biology.

Leeds, J. 2014. Restoration strategies–design and construction status of water quality improvement
projects. Chapter 5A in 2014 South Florida Environmental Report, Volume I: The South Florida
Environment. West Palm Beach: South Florida Water Management District. Available at http://
my.sfwmd.gov/portal/page/portal/pg_grp_sfwmd_sfer/portlet_prevreport/2014_sfer/v1/chapters/
v1_ch5a.pdf [accessed May 21, 2014].
Lederhouse, R. C., M. P. Ayers, J. K. Nitao, and J. M. Scriber. 1992. Differential use of lauraceous
hosts by swallowtail butterflies, Papilio troilus and P. palamedes (Papilionidae). Oikos
63:244-252. Available at http://www.jstor.org/discover/10.2307/3545384?uid=30374&uid=
3739584&uid=2129&uid=2&uid=70&uid=3&uid=30373&uid=67&uid=62&uid=3739256&
sid=21103788356721 [accessed May 21, 2014].
Lefeuvre, J. C. 2013. Les Invasions Biologiques, un Danger pour la Biodiversité. Paris: Buchet-Chastel.
Lejeune, O., M. Tlidi, and R. Lefever. 2004. Vegetation spots and stripes: dissipative structures in
arid landscapes. International Journal of Quantum Chemistry 98:261-271. Available at http://
onlinelibrary.wiley.com/doi/10.1002/qua.10878/pdf [accessed May 21, 2014].
Lever, C. 2001. The Cane Toad. Otley, UK: Westbury.
Li, J. H., J. E. Erickson, G. Peresta, and B. G. Drake. 2010. Evapotranspiration and water use
efficiency in a Chesapeake Bay wetland under carbon dioxide enrichment. Global Change
Biology 16:234-245. Available at http://onlinelibrary.wiley.com/doi/10.1111/j.1365-
2486.2009.01941.x/abstract [accessed May 21, 2014].
Li, Y., and M. Norland. 2001. The role of soil fertility in invasion of Brazilian pepper (Schinus
terebinthifolius) in Everglades National Park, Florida. Soil Science 166:400-405.
Light, S. S., and J. W. Dineen. 1994. Water control in the Everglades: A historical perspective. Pp.
47-84 in Everglades: The Ecosystem and Its Restoration, S. M. Davis and J. C. Ogden, eds.
Boca Raton, FL: St. Lucie Press.
Lippincott, C. L. 2000. Effects of Imperata cylindrica (L.) Beauv. (cogongrass) invasion on fire regime
in Florida sandhill (USA). Natural Areas Journal 20:140-149.
Lodge, T. E. 2010. The Everglades Handbook: Understanding the Ecosystem, 3rd edition. Boca
Raton, FL: CRC Press.
Long, R. W. 1984. Origin of the vascular flora of southern Florida. Pp. 118-126 in Environments of
South Florida Present and Past II, P. J. Gleason, ed. Coral Gables, FL: Miami Geological Society.
Lord, L. A. 1993. Guide to Florida Environmental Issues and Information. Winter Park: Florida
Conservation Foundation.
LoSchiavo, A. J., R. G. Best, R. E. Burns, S. Gray, M. C. Harwell, E. B. Hines, A. R. McLean, T.
St. Clair, S. Traxler, and J. W. Vearil. 2013. Lessons learned from the first decade of adaptive
management in comprehensive Everglades restoration. Ecology and Society 18(4): article 70,
doi:10.5751/ES-06065-180470.
Loss, S. R., T. Will, and P. P. Marra. 2013. The impact of free-ranging domestic cats on wildlife
of the United States. Nature Communications 4: article 1396, doi:10.1038/ncomms2380.
Available at http://www.nature.com/ncomms/journal/v4/n1/full/ncomms2380.html [accessed
May 21, 2014].
Loveless, C. M. 1959. A study of the vegetation of the Florida Everglades. Ecology 40:1-9. Available
at http://www.jstor.org/stable/1929916 [accessed May 21, 2014].
Loxahatchee River District. 2013. 2014 System Status Report, November 18, 2013 draft ver-
sion. Jupiter, FL: Loxahatchee River District. Available at http://www.loxahatcheeriver.org/
pdf/2014%20SSR%20Loxahatchee.pdf [accessed March 12, 2014].
Lynch, R. L., L. A. Brandt, H. J. Chen, D. Ogurcak, I. Fujisaki, and F. J. Mazzotti. 2011. Recruit-
ment and growth of Old World climbing fern in hurricane-caused canopy gaps. Journal
of Fish and Wildlife Management 2:199-206. Available at http://www.fwspubs.org/doi/
abs/10.3996/062011-JFWM-040 [accessed May 21, 2014].
MacVicar, T. 2013. L-31N Seepage Barrier Project, Monitoring Results Update. Presentation to the
Lake Belt Mitigation Committee. November 20, 2013. Available at http://www.l31nseepage.
org/DOCUMENTS/Mit Committee meeting L-31N monitoring 11-20-13.pdf.

References 241
MacVicar, T. 2014. L-31N Seepage Barrier Project, Next Steps. Presentation to the Lake Belt Mitiga-
tion Committee, March 5, 2014. Available at http://www.l31nseepage.org/DOCUMENTS2/Mit
Committee meeting L-31N next steps 3-5-14.pdf.
Madeira, P. T., T. D. Center, J. A. Coetzee, R. W. Pemberton, M. F. Purcell, and M. P. Hill. 2013. Iden-
tity and origins of introduced and native Azolla species in Florida. Aquatic Botany 111:9-15.
Maehr, D. S., R. C. Belden, E. D. Land, and L. Wilkins. 1990. Food habits of panthers in south-
west Florida. Journal of Wildlife Management 54:420-423. Available at http://www.jstor.org/
discover/10.2307/3809651?uid=30374&uid=3739584&uid=2129&uid=2134&uid=2&uid=
70&uid=3&uid=30373&uid=67&uid=62&uid=3739256&sid=21103788779311[accessed
May 21, 2014].
Manning, A. D., J. Fischer, A. Felton, B. Newell, W. Steffen, and D. B. Lindenmayer. 2009. Land-
scape fluidity—a unifying perspective for understanding and adapting to global change. Journal
of Biogeography 36:193-199. Available at http://onlinelibrary.wiley.com/doi/10.1111/j.1365-
2699.2008.02026.x/pdf [accessed May 21, 2014].
Marshall, C., Jr., R. Pielke, Sr., L. Steyaert, and D. Willard. 2004. The impact of anthropogenic land
cover change on the Florida peninsula sea breezes and warm season sensible weather. Monthly
Weather Review 132:28-52. Available at http://journals.ametsoc.org/doi/abs/10.1175/1520-
0493(2004)132%3C0028%3ATIOALC%3E2.0.CO%3B2.
Martin, P. 2014. Five-Year Water Resource Development Work Program. Chapter 5A in 2014 South
Florida Environmental Report, Volume II: District Annual Plans and Reports. West Palm Beach:
pg_grp_sfwmd_sfer/portlet_prevreport/2014_sfer/v2/chapters/v2_ch5a.pdf.
Maxwell, R. M., J. K. Lundquist, J. D. Mirocha, S. G. Smith, C. S. Woodward, and A. F. B. Tompson.
2011. Development of a coupled groundwater–atmosphere model. Monthly Weather Review
139:96-116.
Mayfield, A. E., III, J. A. Smith, M. Hughes, and T. J. Dreaden. 2008. First report of laurel wilt dis-
ease caused by Raffaelea sp. on avocado in Florida. Plant Disease 92:976. Available at http://
apsjournals.apsnet.org/doi/abs/10.1094/PDIS-92-6-0976A [accessed May 21, 2014].
McKee, K. L., and J. A. Cherry. 2009. Hurricane Katrina sediment slowed elevation loss in subsiding
brackish marshes of the Mississippi River Delta. Wetlands 29:2-15. Available at http://link.
springer.com/article/10.1672%2F08-32.1 [accessed May 21, 2014].
McKee, K. L., D. R. Cahoon, and I. C. Feller. 2007. Caribbean mangroves adjust to rising sea level
through biotic controls on change in soil elevation. Global Ecology and Biogeography 16:545-
556. Available at http://www.serc.si.edu/labs/animal_plant_interaction/pubs/McKee%20et%20
al%202007.pdf [accessed May 21, 2014].
McPherson, B. F., and R. Halley. 1996. The South Florida Environment: A Region Under Stress.
USGS Circular 1134. Washington, DC: U.S. Government Printing Office.
McVoy, C. W., W. Park Said, J. Obeysekera, J. A. VanArman, and T. W. Dreschel. 2011. Landscapes
and Hydrology of the Predrainage Everglades. Gainesville: University Press of Florida.
Meier, M., M. B. Dyurgerov, U. K. Rick, S. O’Neel, W. T. Pfeffer, R. S. Anderson, S. P. Anderson,
and A. F. Glazovsky. 2007. Glaciers dominate eustatic sea-level rise in the 21st century. Sci-
ence 317:1064-1067. Available at http://www.sciencemag.org/content/317/5841/1064.abstract
Millar, C. I., N. L. Stephenson, and S. L. Stephens. 2007. Climate change and forests of the future:
Managing in the face of uncertainty. Ecological Applications 17:2145-2151. Available at http://
www.esajournals.org/doi/abs/10.1890/06-1715.1[accessed May 21, 2014].
Miller, J. E. 2007. The context of the South’s cogongrass crisis. Pp. 6-9 in Proceedings of the Regional
Cogongrass Conference: Confronting the Cogongrass Crisis Across the South, N. J. Loewenstein
and J. H. Miller, eds. Mobile. Alabama Cooperative Extension System and Auburn University.
Available at http://www.cogongrass.org/conference-old.cfm.

Mitchell, D. S. 1976. The growth and management of Eichhornia crassipes and Salvinia spp. in
their native environment and in alien situations. Pp. 167-176 in Aquatic Weeds in South-
east Asia: Proceedings of a Regional Seminar on Noxious Aquatic Vegetation, New Delhi,
December 12-17, 1973, C. K. Varshney and J. Rzoska, eds. Dordrecht: Springer Netherlands.
Mohan, J. E., J. S. Clark, and W. H. Schlesinger. 2004. Genetic variation in germination, growth,
and survivorship of red maple in response to subambient through elevated atmospheric CO2.
Global Change Biology 10:233-247.
Monismith, S. G., W. Kimmerer, J. R. Burau, and M. T. Stacey. 2002. Structure and flow-induced
variability of the subtidal salinity field in northern San Francisco Bay. Journal of Physical
Oceanography 32(11); doi:10.1175/1520-0485(2002)032<3003:SAFIVO>2.0.CO;2. Available
at http://journals.ametsoc.org/doi/abs/10.1175/1520-0485(2002)032%3C3003%3ASAFIVO%
3E2.0.CO%3B2 [accessed May 21, 2014].
Moody, M. E., and R. N. Mack. 1988. Controlling the spread of plant invasions: The importance of
nascent foci. Journal of Applied Ecology 25:1009-1021. Available at http://www.bio.miami.edu/
horvitz/bil235/biocontrol_spread_review/MOODYANDMACK_1988.PDF [accessed May 21,
2014].
Morgan, E. C., and W. A. Overholt. 2005. New records of invasive exotic plant species in St. Lucie
County, Florida. Castanea 70:59-61. Available at http://www.jstor.org/discover/10.2307/4034338?
uid=30374&uid=3739584&uid=2129&uid=2134&uid=2&uid=70&uid=3&uid=30373&uid=
67&uid=62&uid=3739256&sid=21103788779311 [accessed May 21, 2014].
Morris, J. T., P. V. Sundareshwar, C. T. Nietch, B. Kjerfve, and D. R. Cahoon. 2002. Responses of
coastal wetlands to rising sea level. Ecology 83:2869-2877.
Moses, C., W. T. Anderson, C. J. Saunders, and F. H. Sklar. 2013. Regional climate gradients in
precipitation and temperature in response to climate teleconnections in the Greater Everglades
ecosystem of South Florida. Journal of Paleolimnology 49(1):5-14.
Mukherjee, A., D. A. Williams, G. S. Wheeler, J. P. Cuda, S. Pal, and W. A. Overholt. 2012. B razilian
peppertree (Schinus terebinthifolius) in Florida and South America: evidence of a possible
niche shift driven by hybridization. Biological Invasions 14:1415-1430. Available at http://link.
springer.com/article/10.1007%2Fs10530-011-0168-7 [accessed May 21, 2014].
NCADAC (National Climate Assessment Development Advisory Committee). 2014. Draft Third
National Climate Assessment Report. U. S. Global Change Research Program.
Nichols, R. J., P. P. Wong, V. R. Burkett, J. O. Codignotto, J. E. Hay, R. F. McLean, S. Ragoonaden,
and C. D. Woodroffe. 2007. Coastal systems and low-lying areas. Pp. 315-356 in Climate
Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to
the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, M. L. Parry,
O. F. Canziani, J. P. Palutikof, P. J. van der Linden, and C. E. Hanson, eds. Cambridge, UK:
Cambridge University Press.
NPS (National Park Service). 2006. Management Policies 2006. Available at http://www.nps.gov/
policy/mp2006.pdf [accessed June 23, 2014].
NPS. 2012. Tamiami Trail Modifications: Modified Waters Deliveries & Next Steps Projects.
Homestead, FL: South Florida Natural Resources Center. Available at http://www.nps.gov/
ever/naturescience/upload/Hi-ResTTModsFact-SheetMay2012.pdf [accessed August 16, 2013].
NPS. 2013. Using Scenarios to Explore Climate Change: A Handbook for Practitioners. Fort Collins,
CO: National Park Service Climate Change Response Program.
NRC (National Research Council). 1987. Responding to Changes in Sea Level: Engineering Implica-
tions. Washington, DC: National Academy Press.
NRC. 1996. Upstream: Salmon and Society in the Pacific Northwest. Washington, DC: National
Academy Press.
NRC. 1999. New Directions for Water Resources Planning for the U.S. Army Corps of Engineers.
Washington, DC: National Academy Press.
NRC. 2001. Aquifer Storage and Recovery in the Comprehensive Everglades Restoration Plan: A
Critique of the Pilot Projects and Related Plans for ASR in the Lake Okeechobee and Western
Hillsboro Areas. Washington, DC: National Academy Press.

References 243
NRC. 2002a. Florida Bay Research Programs and Their Relation to the Comprehensive Everglades
Restoration Plan. Washington, DC: The National Academies Press.
NRC. 2002b. Regional Issues in Aquifer Storage and Recovery for Everglades Restoration. Washing-
ton, DC: The National Academies Press.
NRC. 2003a. Adaptive Monitoring and Assessment for the Comprehensive Everglades Restoration
Plan. Washington, DC: The National Academies Press.
NRC. 2003b. Does Water Flow Influence Everglades Landscape Patterns? Washington, DC: The
National Academies Press.
NRC. 2003c. Science and the Greater Everglades Ecosystem Restoration: An Assessment of the Criti-
cal Ecosystems Initiative. Washington, DC: The National Academies Press.
NRC. 2004a. River Basins and Coastal Systems Planning Within the U.S. Army Corps of Engineers.
Washington, DC: The National Academies Press.
NRC. 2005. Re-Engineering Water Storage in the Everglades: Risks and Opportunities. Washington,
DC: The National Academies Press.
NRC. 2007. Progress Toward Restoring the Everglades: The First Biennial Review–2006. Washington,
DC: The National Academies Press.
NRC. 2008. Progress Toward Restoring the Everglades: The Second Biennial Review—2008. Wash-
ington, DC: The National Academies Press.
NRC. 2010. Progress Toward Restoring the Everglades: The Third Biennial Review—2010. Wash-
NRC. 2012a. Progress Toward Restoring the Everglades: The Fourth Biennial Review—2012. Wash-
NRC. 2012b. Sea-Level Rise for the Coasts of California, Oregon, and Washington: Past, Present,
and Future. Washington, DC: The National Academies Press.
NRC. 2012c. Sustainable Water and Environmental Management in the California Bay-Delta. Wash-
NRC. 2013. An Ecosystem Services Approach to Assessing the Impacts of the Deepwater Horizon
Oil Spill in the Gulf of Mexico. Washington, DC: The National Academies Press.
NRC. 2014. Climate Change: Evidence, Impacts, and Choices. Washington, DC: The National
Academies Press.
Nuñez, M. A., J. Hayward, T. R. Horton, G. C. Amico, R. D. Dimarco, M. N. Barrios-Garcia, and D.
Simberloff. 2013. Exotic mammals disperse exotic fungi that promote invasion by exotic trees.
PLoS ONE 8(6):e66832; doi:10.1371/journal.pone.0066832.
Nungesser, M., C. Saunders, C. Coronado-Molina, J. Obeysekera, J. Johnson, C. McVoy, and
B. Benscoter. 2014. Potential effects of climate change on Florida’s Everglades peatlands.
Manuscript.
Nuttle, W. K., J. W. Fourqurean, B. J. Cosby, J. C. Zieman, and M. B. Robblee. 2000. Influence of
net freshwater supply on salinity in Florida Bay. Water Resources Research 36:1805-1822.
Obeysekera, J., M. Irizarry, J. Park, J. Barnes, and T. Dessalegne. 2011a. Climate change and its
implications for water resources management in South Florida. Stochastic Environmental
Research and Risk Assessment 25(4):495-516.
Obeysekera, J., J. Park, M. Irizarry-Ortiz, P. Trimble, J. M. Barnes, J. VanArman, W. Said, and E.
Gadzinski. 2011b. Past and Projected Trends in Climate and Sea Level for South Florida:
Hydrologic and Environmental Systems Modeling Technical Report. Interdepartmental Cli-
mate Change Group. West Palm Beach: South Florida Water Management District. Avail-
able at http://my.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/ccireport_
publicationversion_14jul11.pdf [accessed May 23, 2014].
Obeysekera, J., J. Barnes, and M. Nungesser. 2014. Predicting response of the greater Florida Ever
glades to climate change and future hydrologic regimes: Climate sensitivity runs and regional
hydrologic modeling. Environmental Management. doi: 10.1007/s00267-014-0315-X.

Ogden, J. C., S. M. Davis, K. J. Jacobs, T. Barnes, and H. E. Fling. 2005. The use of conceptual
ecological models to guide ecosystem restoration in South Florida. Wetlands 25:795-809.
Available at http://www.evergladesplan.org/pm/recover/recover_docs/cems/cem_use_of_cems.
pdf [accessed May 21, 2014].
O’Gorman, P. A., and T. Schneider. 2009. The physical basis for increases in precipitation extremes
in simulations of 21st-century climate change. Proceedings of the National Academy of Sci-
ences of the United States of America 106:14773-14777. Available at http://www.pnas.org/
content/early/2009/08/20/0907610106 [accessed May 21, 2014].
Olmstead, I. C., and L. L. Loope. 1984. Plant communities of Everglades National Park. In Envi-
ronments of South Florida: Past and Present II, P. J. Gleason, ed. Coral Gables, FL: Miami
Geological Society.
Palmer, M. A., R. F. Ambrose, and N. L. Poff. 1997. Ecological theory and community resto-
ration ecology. Restoration Ecology 5:291-300. Available at http://onlinelibrary.wiley.com/
doi/10.1046/j.1526-100X.1997.00543.x/abstract [accessed May 21, 2014].
Palmer, M. A., K. L. Hondula, and B. J. Koch. 2014. Ecological restoration of streams and rivers:
shifting baselines and shifting goals. Annual Review of Ecology, Evolution, and Systematics,
in press.
Parker, G. G., and C. W. Cooke. 1944. Late Cenozoic geology of southern Florida with a discussion
of the groundwater. Florida Geological Survey Bulletin 27:1-119.
Parkinson, R. W., R. D. DeLaune, and J. R. White. 1994. Holocene sea-level rise and the fate of
mangrove forests within the wider Caribbean region. Journal of Coastal Research 10(4):1077-
1086. Available at http://www.jstor.org/discover/10.2307/4298297?uid=30374&uid=3739584
&uid=2129&uid=2134&uid=2&uid=70&uid=3&uid=30373&uid=67&uid=62&uid=3739256
&sid=21103788779311[accessed May 21, 2014].
Parmesan, C. 2006. Ecological and evolutionary responses to recent climate change. Annual Re-
view of Ecology, Evolution, and Systematics 37:637-669. Available at http://www.ent.orst.edu/
insect_ecology/PDF/Parmesan%202006%20Ann%20Rev.pdf [accessed May 21, 2014].
Pearlstine, L. G., E. V. Pearlstine, and N. G. Aumen. 2010. A review of the ecological consequences
and management implications of climate change for the Everglades. Journal of the North
American Benthological Society 29:1510-1526.
Pemberton, R. W., and J. M. Bodle. 2009. Native North American Azolla weevil, Stenopelmus
rufinasus (Coleoptera: Curculionidae), uses the invasive Old World Azolla pinnataas as a host
plant. Florida Entomologist 92:153-155. Available at http://www.jstor.org/discover/10.2307/
40205342?uid=30374&uid=3739584&uid=2129&uid=2134&uid=2&uid=70&uid=3&uid=
30373&uid=67&uid=62&uid=3739256&sid=21103788779311 [accessed May 21, 2014].
Pernas, T., D. J. Giardina, A. McKinley, A. Parns, and F. J. Mazzotti. 2012. First observations of nest-
ing by the Argentine black and white tegu, Tupinambis merianae, in south Florida. Southeastern
Naturalist 11(4):765-770.
Perry, W. 2004. Elements of South Florida’s Comprehensive Everglades Restoration Plan. Ecotoxicol-
ogy 13:185-193. Available at http://download.springer.com/static/pdf/262/art%253A10.1023%
252FB%253AECTX.0000023564.10311.4a.pdf?auth66=1400868578_8e2412eee59f31807b9
52ec76bddab41&ext=.pdf [accessed May 21, 2014].
Peterson, G. D., G. S. Cumming, and S. R. Carpenter. 2003. Scenario planning: a tool for con-
servation in an uncertain world. Conservation Biology 17(2):358-366. Available at http://
onlinelibrary.wiley.com/doi/10.1046/j.1523-1739.2003.01491.x/full [accessed May 21, 2014].
Petuch, E. J., and C. E. Roberts. 2007. The Geology of the Everglades and Adjacent Areas. Boca
Raton, FL: CRC Press.
Pias, K., Z. Welch, and W. M. Kitchens. 2012. An artificial perch to help snail kites handle an exotic
apple snail. Waterbirds 35:347-351.
Polasky, S., S. R. Carpenter, C. Folke, and B. Keeler. 2011. Decision-making under great un-
certainty: environmental management in an era of global change. Trends in Ecology and
Evolution 26(8):398-404. Available at http://www.sciencedirect.com/science/article/pii/
S0169534711001157 [accessed May 21, 2014].

References 245
Portmann, R. W., S. Solomon, and G. C. Hegerl. 2009. Spatial and seasonal patterns in climate
change, temperatures, and precipitation across the United States. Proceedings of the National
Academy of Sciences of the United States of America 106:7324-7329. Available at http://www.
pnas.org/content/106/18/7324.short [accessed May 21, 2014].
Possley, J. Update on Lumnitzera racemosa: a non-native mangrove at Fairchild Tropical Botanic
Garden and Matheson Hammock Park. Center for Tropical Plant Conservation. Coral Gables,
FL. Available at http://www.evergladescisma.org/summit11/3_LumnitzUpdate2011.pdf [ac-
cessed June 10, 2014].
Price, F. 2009. When is an acre “infested”? Using the FNAI implementation of NAWMA standards
to describe invasive plant occurrences. Wildland Weeds (Spring):4-6.
Punzo, F., and L. Lindstrom. 2001. The toxicity of eggs of the giant toad, Bufo marinus, to aquatic
predators in a Florida retention pond. Journal of Herpetology 35:693-697.
Puri, A., G. E. MacDonald, and W. T. Haller. 2007. Stability of fluridone-resistant hydrilla (Hydrilla
verticillata) biotypes over time. Weed Science 55:12-15. Available at http://www.bioone.org/
doi/abs/10.1614/WS-06-150.1 [accessed May 21, 2014].
Qualls, R. G., and C. J. Richardson. 2003. Factors controlling concentrations, export, and decompo-
sition of dissolved organic nutrients in the Everglades of Florida. Biogeochemistry 62:197-229.
Rasse, D. P., G. Peresta, and B. G. Drake. 2005. Seventeen years of elevated CO2 exposure in
a Chesapeake Bay wetland: sustained but contrasting responses of plant growth and CO2
uptake. Global Change Biology 11(3):369-377. Available at http://onlinelibrary.wiley.com/
doi/10.1111/j.1365-2486.2005.00913.x/pdf [accessed May 21, 2014].
Raven, J. A. 2001. Roots: evolutionary origins and biogeochemical significance. Journal of Experi-
mental Botany 52:381-401.
Rayamajhi, M. B., P. D. Pratt, T. D. Center, P. W. Tipping, and T. K. Van. 2009. Decline in exotic
tree density facilitates increased plant diversity: the experience from Melaleuca quinquenervia
invaded wetlands. Wetland Ecology and Management 17:455-467. Available at http://link.
RECOVER (Restoration, Coordination, and Verification). 2004. 2004 RECOVER Monitoring and
Assessment Plan. Available at http://www.evergladesplan.org/pm/recover/recover_map.aspx
[accessed April 4, 2014].
RECOVER. 2006a. 2006 RECOVER Monitoring and Assessment Plan. Available at http://www.
evergladesplan.org/pm/recover/recover_map.aspx [accessed April 4, 2014].
RECOVER. 2006b. Comprehensive Everglades Restoration Plan Adaptive Management Strategy.
Restoration Coordination and Verification. Jacksonville, FL: U.S. Army Corps of Engineers
and West Palm Beach: South Florida Water Management District. Available at http://www.
evergladesplan.org/pm/recover/recover_docs/am/rec_am_stategy_brochure.pdf [accessed
May 23, 2014].
RECOVER. 2007. Development and Application of Comprehensive Everglades Restoration Plan
System-wide Performance Measures. Jacksonville, FL: U.S. Army Corps of Engineers and West
Palm Beach: South Florida Water Management District. Available at http://www.evergladesplan.
org/pm/recover/perf_systemwide.aspx [accessed May 14, 2014].
RECOVER. 2009. 2009 RECOVER Monitoring and Assessment Plan. Jacksonville, FL: U.S. Army
Corps of Engineers and West Palm Beach: South Florida Water Management District. Avail-
able at http://www.evergladesplan.org/pm/recover/recover_map.aspx [accessed April 4, 2014].
RECOVER. 2010. CERP Adaptive Management Integration Guide. Restoration Coordination and
Verification. Jacksonville, FL: U.S. Army Corps of Engineers and West Palm Beach: South
Florida Water Management District.
RECOVER. 2011a. Monitoring and Assessment Plan (MAP) Regional and System-wide Prioritization
Process: Instructions. August. Jacksonville, FL: U.S. Army Corps of Engineers and West Palm
Beach, Florida: South Florida Water Management District.
RECOVER. 2011b. Scientific and Technical Knowledge Gained in Everglades Restoration (1999-
2009). Restoration Coordination and Verification. Jacksonville, FL: U.S. Army Corps of Engi-
neers and West Palm Beach, Florida: South Florida Water Management District.

RECOVER. 2012. RECOVER—Restoration, Coordination, & Verification. Comprehensive Everglades

Restoration Plan. Fact Sheet. Jacksonville, FL: U.S. Army Corps of Engineers and West Palm
Beach: South Florida Water Management District.
RECOVER, 2014a. External Draft, 2014 System Status Report. Comprehensive Everglades Restoration
Plan. Available at http://evergladesplan.org/pm/ssr_2014/ssr_2014_pdfs/2014_exec_summ_
ssr_022714.pdf [accessed May 23, 2014].
RECOVER. 2014b. Picayune Strand Restoration Project. 2014 Systems Status Report Draft August,
2013. West Palm Beach: South Florida Water Management District.
Reddy, K. R., and R. D. DeLaune. 2008. Biogeochemistry of Wetlands: Science and Applications.
Boca Raton, FL: CRC Press.
Reddy, K. R., J. R. White, A. Wright, and T. Chua. 1999. Influence of phosphorus loading on
microbial processes in soil and water column of wetlands. Pp. 249-273 in Phosphorus Biogeo
chemistry in Subtropical Ecosystems: Florida as a Case Example, K. R. Reddy, G. A. O’Connor,
and C. L. Schelske, eds. CRC/Lewis Pub.
Reddy, K. R., T. Z. Osborne, K. S. Inglett, and R. Corstanje. 2006. Influence of Water Levels on
Subsidence of Organic Soils in the Upper St. John’s River Basin. Final Report. Palatka, FL: St.
Johns River Water Management District.
Reddy, K. R., R. DeLaune, and C. B. Craft. 2010. Nutrients in Wetlands: Implications to Water
Quality Under Changing Climatic Conditions. Final Report. Washington, DC: U.S. Environ-
mental Protection Agency.
Reddy, K. R., G. R. Best, and F. Sklar. 2011. Biogeochemistry and water quality of the Everglades:
symposium overview. Critical Reviews in Environmental Science and Technology 41(6):1-3.
Reich, P. G., S. E. Hobbie, T. Lee, D. S. Ellsworth, J. B. West, D. Tilman, J. M. H. Knos, S. Naeem,
and J. Trost. 2006. Nitrogen limitation constrains sustainability of ecosystem response to CO2.
Nature 440:922-925.
Restrepo, J., and A. Montoya. 2008. MODFLOW/SEAWAT-2000WMD Report—Version 4. Pre-
pared by Florida Atlantic University for South Florida Water Management District, Water
Supply Department. Available at http://www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_
repository_pdf/sfwmd_input_inst_modflow-seawat_v4_2000_wmd.pdf.
Richard, A., and V. Ramey. 2007. Invasive and Non-native Plants You Should Know, Recognition
Cards. SP 431. University of Florida Institute of Food and Agricultural Sciences Center for
Aquatic and Invasive Plants.
Richardson, C. J. 2008. The Everglades Experiments. New York: Springer.
Rignot, E., J. Mouginot, M. Morlighem, H. Seroussi, and B. Scheuchi. 2014. Widespread, rapid
grounding line retreat of Pine Island, Thwaites, Smith, and Kohler glaciers, West Antartica, from
1992 to 2011. Geophysical Research Letters 41; doi:10.1002/2014GL060140. Available at
http://onlinelibrary.wiley.com/doi/10.1002/2014GL060140/abstract [accessed May 27, 2014].
Rivera-Monroy, V. H., K. de Mutsert, R. R. Twilley, E. Castaneda-Moya, M. M. Romigh, and S. E.
Davis. 2007. Patterns of nutrient exchange in a riverine mangrove forest in the Shark River
Estuary, Florida. Hydrobiologia 17:169-178.
Rödder, D., and F. Weinsheimer. 2009. Will future anthropogenic climate change increase the
potential distribution of the alien invasive Cuban treefrog (Anura: Hylidae)? Journal of Natural
History 43:1207-1217. Available at http://www.tandfonline.com/doi/abs/10.1080/002229309
02783752#preview [accessed May 21, 2014].
Rodgers, J. C., III, A. W. Murrah, and W. H. Cooke. 2009. The impact of Hurricane Katrina on the
coastal vegetation of the Weeks Bay Reserve, Alabama from NDVI data. Estuaries and Coasts
32:496-507. Available at http://link.springer.com/article/10.1007%2Fs12237-009-9138-z.
Rodgers, L., M. Bodle, D. Black, and F. Laroche. 2013. Chapter 7: Status of nonindigenous species.
2013 South Florida Environmental Report. South Florida Water Management District.
Rodgers, L., M. Bodle, D. Black, and F. Laroche. 2014a. Status of nonindigenous species. Chapter 7
in 2014 South Florida Environmental Report, Volume I: The South Florida Environment. West
Palm Beach: South Florida Water Management District.

References 247
Rodgers, L., T. Pernas, and S. Hill. 2014b. Mapping invasive plant distribution in the Florida Ever-
glades using digital aerial sketch mapping technique. Invasive Plant Science and Management,
doi:10.1614/IPSM-D-12-00092.1.
Romagosa, C. R., T. D. Steury, M. A. Miller, C. Guyer, B. Rogers, T. C. Angle, and R. L. Gillette. 2011.
Assessment of Detector Dogs as a Potential Tool for Python Detection Efforts. Final Report.
National Park Service and South Florida Water Management District.
Ross, M. S., J. F. Meeder, J. P. Sah, P. L. Ruiz, and G. J. Telesnicki. 2000. The Southeast Saline
Everglades revisited: a half century of coastal vegetation change. Journal of Vegetation Sci-
ence 11:101-112. Available at http://www2.fiu.edu/~geology/Content/Files/Ross/627-(Ross%20
et%20al.%202000)%20-%20Southeast%20saline%20Everglades%20revisited.pdf [accessed
May 21, 2014].
Ross, M. S., E. E. Gaiser, J. F. Meeder, and M. T. Lewin. 2002. Multi-taxon analysis of the “white
zone”: a common ecotonal feature of the South Florida coastal wetlands. Pp. 205-238 in The
Everglades, Florida Bay and Coral Reefs of the Florida Keys: An Ecosystem Sourcebook, J. W.
Porter and K. G. Porter, eds. Boca Raton, FL: CRC Press.
Ross, M. S., S. Mitchell-Bruker, J. P. Sah, S. Stothoff, P. L. Ruiz, D. L. Reed, K. Jayachandran, and C. L.
Coultas. 2006. Interaction of hydrology and nutrient limitation in the ridge and slough land-
scape of the southern Everglades. Hydrobiologia 569:37-59. Available at http://link.springer.
com/article/10.1007%2Fs10750-006-0121-4 [accessed May 21, 2014].
Rozema, J., F. Dorel, R. Janissen, G. Lenssen, R. Broekman, W. Arp, and B. G. Drake. 1991. Effect
of elevated atmospheric CO2 on growth, photosynthesis and water relations of salt-marsh grass
species. Aquatic Botany 39:45-55.
Ruttenberg, B. I., P. J. Schofield, J. L. Atkins, A. Acosta, M. W. Feeley, J. Blondeau, S. G. Smith,
and J. S. Ault. 2012. Rapid invasion of the Indo-Pacific lionfishes (Pterois volitans and Pterois
miles) in the Florida Keys, USA: evidence from multiple pre- and post-invasion data sets.
Bulletin of Marine Science 88:1051-1059. Available at http://www.reef.org/reef_files/BMS_
lionfishinvasion2012_ruttenbergetal.pdf [accessed May 21, 2014].
Saha, A. K., S. Saha, J. Sadle, J. Jiang, M. S. Ross, R. M. Price, L. Sternberg, and K. S. Wendelberger.
2011. Sea level rise and South Florida coastal forests. Climatic Change 107:81-108.
Saha, A. K., C. S. Moses, R. M. Price, V. Engel, T. J. Smith III, and G. Anderson. 2012. A hydrologi-
cal budget (2002-2008) for a large subtropical wetland ecosystem indicates marine ground
water discharge accompanies diminished freshwater flow. Estuaries and Coasts 35(2):459-474.
Available at http://link.springer.com/article/10.1007%2Fs12237-011-9454-y [accessed May 23,
2014].
Scavia, D., J. C. Field, D. F. Boesch, R. W. Buddemeier, V. Burkett, D. R. Cayan, M. Fogarty, M. A.
Harwell, R. W. Howarth, C. Mason, D. J. Reed, T. C. Royer, A. H. Sallenger, and J. G. Titus.
2002. Climate change impacts on U.S. Coastal and Marine Ecosystems. Estuaries 25:149-
164. Available at http://tenaya.ucsd.edu/~cayan/Pubs/66_Scavia_Estuaries_2002.pdf [accessed
May 21, 2014].
SCG (Science Coordination Group). 2003. Charter, South Florida Ecosystem Restoration Science
Coordination Group. Available at http://www.sfrestore.org/scg/scg_charter.pdf [accessed May 21,
2014].
Schardt, J. D. 1997. Maintenance control. Pp. 229-243 in Strangers in Paradise: Impact and Manage
ment of Nonindigenous Species in Florida, D. Simberloff, D. C. Schmitz, and T. C. Brown, eds.
Washington, DC: Island Press.
Scheffer, M., S. Carpenter, J. A. Foley, C. Folke, and B. Walker. 2001. Catastrophic shifts in
ecosystems. Nature 413:591-596. Available at http://www.gatsby.ucl.ac.uk/~pel/environment/
catastrophe.pdf [accessed May 21, 2014].
Scheidt, D. J., and P. I. Kalla. 2007. Everglades Ecosystem Assessment: Water Management And
Quality, Eutrophication, Mercury Contamination, Soils and Habitat. Monitoring for Adap-
tive Management: A R-EMAP Status Report. EPA 904-R-07-001. Athens, GA: U.S. Environ-
mental Protection Agency Region 4. Available at http://www.epa.gov/region4/sesd/reports/
epa904r07001/epa904r07001.pdf [accessed May 23, 2014].

Schmitz, D., D. Simberloff, R. Hoffstetter, W. Haller, and D. Sutton. 1997. The ecological impact
of non-indigenous plants. Pp. 29-74 in Strangers in Paradise: Impact and Management of
Nonindigenous Species in Florida, D. Simberloff, D. Schmitz, and T. Brown, eds. Washington,
DC: Island Press.
Schofield, P. J., and L. G. Nico. 2009. Salinity tolerance of non-native Asian swamp eels (Teleostei:
Synbranchidae) in Florida, USA: comparison of three populations and implications for disper-
sal. Environmental Biology of Fishes, doi:10.1007/s10641-009-9456-9. Available at http://link.
Scholl, D. W. 1964. Recent sedimentary record in mangrove swamps and rise in sea level over the
southwestern coast of Florida. Marine Geology 1:344-366; doi:10.1016/0025-3227(64)90020-9.
Scholl, D. W., and M. Stuiver. 1967. Recent submergence of Southern Florida: A comparison with
adjacent coasts and other eustatic data. Geological Society of America Bulletin 78:437-454.
Available at http://gsabulletin.gsapubs.org/content/78/4/437 [accessed May 21, 2014].
Scholl, D. W., F. C. Craighead, and M. Stuiver. 1969. Florida submergence curve revised: Its relation
to coastal sedimentation rates. Science 163:562-564. Available at http://www.sciencemag.org/
content/163/3867/562.abstract [accessed May 21, 2014].
Scott, R. 2013. Gov. Scott Announces $90 M Commitment for Tamiami Trail Project. Available at
http://www.flgov.com/gov-scott-announces-90-m-commitment-for-tamiami-trail-project-2/ [ac-
cessed January 15, 2014].
Scott, R. 2014. Governor Rick Scott Announces $130 Million Investment in South Florida’s Ever
glades. Available at http://www.flgov.com/2014/01/22/governor-rick-scott-announces-130-
million-investment-in-south-floridas-everglades/.
SCT (Science Coordination Team). 2003. The Role of Flow in the Everglades Ridge and Slough
Landscape. Available at http://www.sfrestore.org/sct/docs/ [accessed November 12, 2009].
Serbesoff-King, K. 2003. Melaleuca in Florida: A literature review on the taxonomy, distribu-
tion, biology, ecology, economic importance and control measures. Journal of Aquatic Plant
Management 41:98-112. Available at http://www.apms.org/japm/vol41/v41p98.pdf [accessed
May 21, 2014].
SERES Project Team. 2010. Key Questions at the Interface of Science and Management. Available
at http://www.everglades-seres.org/Products_files/SERESKeyScienceManagementQuestions_1.
SFERTF (South Florida Ecosystem Restoration Task Force). 2000. Coordinating Success: Strategy
for Restoration of the South Florida Ecosystem. July. Available at http://www.sfrestore.org/
documents/work_products/coordinating_success_2000.pdf [accessed on August 20, 2010].
SFERTF. 2006. Plan for Coordinating Science. Available at http://www.sfrestore.org/scg/documents/
PCS_Draft_for_Task_Force_Oct_29.pdf [accessed May 21, 2014].
SFERTF. 2010. Plan for Coordinating Science Update: July 2008-June 2010. Available at http://
www.sfrestore.org/scg/documents/Plan_for_Coordinating_Science_2010_Update.pdf [accessed
May 21, 2014].
SFERTF. 2012. South Florida Integrated Financial Plan. Available at http://www.sfrestore.org/
documents/2012_IFP/2012_Integrated_Financial_Plan.pdf [accessed November 8, 2013].
SFERTF. 2014. Cross-Cut Budget. Available at http://www.sfrestore.org/documents/2014_Cross_Cut_
Budget.pdf [accessed April 7, 2014].
SFWMD (South Florida Water Management District). 2008. Comprehensive Annual Financial Report
(CAFR)—FY 2008. Available at http://my.sfwmd.gov/portal/page/portal/xrepository/sfwmd_
repository_pdf/2008_sfwmd_cafr.pdf [accessed March 29, 2012].
SFWMD. 2009a. Climate Change and Water Management in South Florida. Available at http://www.
sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/climate_change_and_water_
management_in_sflorida_12nov2009.pdf [accessed May 21, 2014].
SFWMD. 2009b. Comprehensive Annual Financial Report (CAFR)—FY 2009. Available at http://
my.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/2009_sfwmd_cafr.pdf [ac-
cessed March 29, 2012].

References 249
SFWMD. 2010. Comprehensive Annual Financial Report (CAFR)—FY 2010. Available at http://
cessed March 29, 2012].
SFWMD. 2011. Comprehensive Annual Financial Report (CAFR)—FY 2011. Available at http://
www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/sfwmd_2011_cafr.pdf
SFWMD. 2012a. Comprehensive Annual Financial Report (CAFR)—FY 2012. Available at http://
www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/2012_sfwmd_cafr.pdf
SFWMD. 2013a. 2013 South Florida Environmental Report. Available at http://www.sfwmd.gov/
portal/page/portal/pg_grp_sfwmd_sfer/portlet_prevreport/2013_sfer/v1/vol1_table_of_contents.
html [accessed May 5, 2014].
SFWMD. 2013b. FY2014 Tentative Budget Submission. August 1, 2013. Available at http://www.
sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/aug_1_2013_fy14_sfwmd_
complete_report.pdf [accessed April 7, 2014].
SFWMD. 2013c. Monthly Financial Statement—September 2013. Available at http://www.sfwmd.
gov/portal/page/portal/xrepository/sfwmd_repository_pdf/monthly_financial_statement_2013_
september.pdf [accessed April 7, 2014].
SFWMD. 2013d. Science Plan for the Everglades Stormwater Treatment Areas. Restoration Strategies
Regional Water Quality Plan. June 2013.
SFWMD. 2013e. Comprehensive Annual Financial Report (CAFR)—FY 2013. Available at http://
cessed June 6, 2014].
SFWMD. 2014a. 2014 Budget in Brief. Available at http://www.sfwmd.gov/portal/page/portal/
xrepository/sfwmd_repository_pdf/2014_budget_brief.pdf [accessed May 21, 2014].
SFWMD. 2014b. FY2014 Programmatic Line Item Report (Functional Area Level). Available at http://
www.sfwmd.gov/portal/page/portal/xrepository/sfwmd_repository_pdf/fy14_adopted_budget_
functional_area_line_item_detail_complete.pdf.
Shin, S. I., and S. K. Lee. 2011. Changing characteristics of El Niño and the Southern Oscillation
and the Pacific Decadal Oscillation. In Climate Scenarios: A Florida-Centric View, V. Misra,
E. Carlson, R. K. Craig, and D. Enfield, eds. Florida Climate Change Task Force. Available at
http://floridaclimateorg/whitepapers/ [accessed May 21, 2014].
Shirley, L. J., and L. L. Battaglia. 2006. Assessing vegetation change in coastal landscapes of
the northern Gulf of Mexico. Wetlands 26:1057-1070. Available at http://www.bioone.org/
doi/pdf/10.1672/0277-5212(2006)26%5B1057:AVCICL%5D2.0.CO%3B2 [accessed May 21,
2014].
Shirley, L. J., and L. L. Battaglia. 2008. Projecting fine resolution land-cover dynamics for a rapidly
changing terrestrial–aquatic transition in Terrebonne Basin, Louisiana, U.S.A. Journal of Coastal
Research 246:1545-1554. Available at http://www.jcronline.org/doi/abs/10.2112/07-0903.1
Silvers, C. S., P. D. Pratt, A. P. Ferriter, and T. D. Center. 2007. T.A.M.E. Melaleuca: a regional ap-
proach for suppressing one of Florida’s worst weeds. Journal of Aquatic Plant Management
45:1-8. Available at http://www.apms.org/japm/vol45/v45p01.pdf [accessed May 21, 2014].
Simberloff, D. 2003. Eradication—preventing invasions at the outset. Weed Science 51:247-253.
Available at http://www.tsusinvasives.com/dotAsset/3a05e4d0-bb25-40ff-a72a-c25a486bb90f.
Simberloff, D. 2009. We can eliminate invasions or live with them! High-tech and low-tech success
stories. Biological Invasions 11:149-157. Available at http://izt.ciens.ucv.ve/ecologia/Archivos/
ECO_POB%202009/ECOPO4_2009/Simberloff%202009.pdf [accessed May 21, 2014].
Simberloff, D. 2012. Risks of biological control for conservation purposes. BioControl 57(2):
263-276. Available at http://link.springer.com/article/10.1007/s10526-011-9392-4 [accessed
May 21, 2014].

Simberloff, D. 2013. Eradication: pipe dream or real option? Pp. 549-560 in Plant Invasions in Pro-
tected Areas: Patterns, Problems and Challenges, L. C. Foxcroft, P. Pyšek, D. M. Richardson,
and P. Genovesi, eds. Dordrecht: Springer Netherlands.
Simberloff, D. 2014. Biological invasions: What’s worth fighting and what can be won? Ecologi-
cal Engineering 65:112-121. Available at http://www.sciencedirect.com/science/article/pii/
S0925857413003431 [accessed May 23, 2014].
Simberloff, D., and B. Von Holle. 2011. Positive interactions of nonindigenous species: invasional
meltdown? Biological Invasions 1:21-32.
Sklar, F. 2013. An Update of the Decomp. Physical Model (DPM): The largest AM Assessment in
USACOE History. Joint Working Group and Science Coordination Group Meeting Nov. 19. Avail-
able at http://www.sfrestore.org/wg/wgminutes/2013meetings/111913/Decompartmentalization_
Physical_Model_Update.pdf.
Sklar, F., C. McVoy, R. VanZee, D. E. Gawlik, K. Tarboton, D. Rudnick, S. L. Miao, and T. Armentano.
2001. The effects of altered hydrology on the ecology of the Everglades. Pp. 39-82 in The
Everglades, Florida Bay, and Coral Reefs of the Florida Keys, an Ecosystem Source Book, J. W.
Porter and K. G. Porter, eds. Boca Raton, FL: CRC Press.
Smith, C. S., L. Serra, Y. Li, P. Inglett, and K. Inglett. 2011. Restoration of disturbed lands: the Hole-
in-the-Donut restoration in the Everglades. Critical Reviews in Environmental Science and
Technology 41(S1):723-739. Available at http://www.tandfonline.com/doi/abs/10.1080/10643
389.2010.530913#preview [accessed May 21, 2014].
Smith, K. G. 2005. Effects of nonindigenous tadpoles on native tadpoles in Florida: evidence
of competition. Biological Conservation 123:433-441. Available at http://sites.davidson.edu/
kevinsmith/wp-content/uploads/2014/03/Smith2005.pdf [accessed May 21, 2014].
Smith, T. J., III, M. B. Robblee, H. R. Wanless, and T. W. Doyle.1994. Mangroves, hurricanes, and
lightning strikes. BioScience 44:256-262. Available at http://sofia.usgs.gov/publications/papers/
mang_hurr_lightning/mang_hurr_lightning.pdf [accessed May 21, 2014].
Smith, T. J., III, G. H. Anderson, K. Balentine, G. Tiling, G. A. Ward, and K. R. T. Whelan. 2009.
Cumulative impact of hurricanes on Florida mangrove ecosystems: sediment deposition,
storm surges and vegetation. Wetlands 29:24-34. Available at http://sofia.usgs.gov/publications/
papers/hurricane_sed/ [accessed May 21, 2014].
Smith, T. J., III, A. M. Foster, G. Tiling-Range, and J. W. Jones. 2013. Dynamics of mangrove–marsh
ecotones in subtropical coastal wetlands: fire, sea-level rise, and water levels. Fire Ecology 9.
doi:10.4996/fireecology.0901066.
Smoak, J. M., J. L. Breithaupt, T. J. Smith III, and C. J. Sanders. 2013. Sediment accretion and organic
carbon burial relative to sea-level rise and storm events in two mangrove forests in Everglades
National Park. Catena 104(May), doi:10.1016/j.catena.2012.10.009.
Society for Ecological Restoration International Science & Policy Working Group. 2004. The
SER International Primer on Ecological Restoration. Available at http://www.ser.org/content/
ecological_restoration_primer.asp [accessed April 3, 2012].
Spector, T., and F. E. Putz. 2006. Biomechanical plasticity facilitates invasion of maritime forests in
the southern USA by Brazilian pepper (Schinus terebinthifolius). Biological Invasions 8:255-
260. Available at http://link.springer.com/article/10.1007%2Fs10530-004-5571-x [accessed
May 21, 2014].
Spencer, W., and G. Bowes. 1986. Photosynthesis and growth of water hyacinth uncer CO2 enrich-
ment. Plant Physiology 82:528-533. Available at http://www.ncbi.nlm.nih.gov/pmc/articles/
PMC1056153/ [accessed May 21, 2014].
Springer, C. J., and J. K. Ward. 2007. Flowering time and elevated atmospheric CO2. New Phytologist
176:243-255.
SSG (Science Sub-Group). 1993. Federal Objectives for the South Florida Restoration by the Science
Sub-Group of the South Florida Management and Coordination Working Group. Miami, FL.
Staletovich, J. 2014. U.S. and Florida resolve money disputes over two stalled Everglades projects.
Miami Herald, April 30. Available at http://www.miamiherald.com/2014/04/30/4090228/us-
and-florida-resolve-money-disputes.html, accessed May 19, 2014.

References 251
Stevens, J. T., and B. Beckage. 2009. Fire feedbacks facilitate invasion of pine savannas by Brazilian
pepper (Schinus terebinthifolius). New Phytologist 184:365-375. Available at http://www.uvm.
edu/~bbeckage/Manuscripts/Stevens&Beckage.NewPhytologist.2009.pdf [accessed May 21,
2014].
Stevens, P. W., S. L. Fox, and C. L. Montague. 2006. The interplay between mangroves and salt-
marshes at the transition between temperate and subtropical climate in Florida. Wetlands Ecol-
ogy and Management 14:435-444. Available at http://download.springer.com/static/pdf/586/ar
t%253A10.1007%252Fs11273-006-0006-3.pdf?auth66=1400871287_b84424fc19361a003d8
27308c2d88b89&ext=.pdf [accessed May 21, 2014].
Stocks, I. C., and G. Hodges. 2012. The Rugose Spiraling Whitefly, Aleurodicus rugioperculatus
Martin, a New Exotic Whitefly In South Florida (Hemiptera: Aleyrodidae). Pest Alert DACS-
P-01745. Gainesville: Florida Department of Agriculture and Consumer Services, Division of
Plant Industry.
Strober, Q. J., R. D. Jones, and D. J. Scheidt. 1995. Ultra trace level mercury in the Everglades eco-
system: a multi-media canal pilot study. Water, Air, and Soil Pollution 80:991-1001.
Sun, Y., S. Solomon, A. Dai, and R. W. Portmann. 2007. How often will it rain? Journal of Climate
20:4801-4818.
Swain, E., L. Stefanova, and T. Smith. 2014. Applying downscaled global climate model data to a
hydrodynamic surface-water and groundwater model. American Journal of Climate Change
3:33-49.
Tate, A. S., and L. L. Battaglia. 2013. Community disassembly and reassembly following experi
mental storm surge and wrack application. Journal of Vegetation Science 24:46-57. Available
at http://onlinelibrary.wiley.com/doi/10.1111/j.1654-1103.2012.01447.x/abstract [accessed
May 21, 2014].
Temple, M. W. B. 2009. Memorandum: U.S. Army Corps of Engineers Invasive Species Policy.
June 2, 2009. Available online at http://www.nae.usace.army.mil/Portals/74/docs/regulatory/
InvasiveSpecies/policy.pdf.
Tett, P., R. Gowen, D. Mills, T. Fernandes, L. Gilpin, M. Huxham, K. Kennington, P. Read, M. S ervice,
M. Wilkinson, and S. Malcolm. 2007. Defining and detecting undesirable disturbance in the
context of eutrophication. Marine Pollution Bulletin 53:282-297. Available at http://www.
sciencedirect.com/science/article/pii/S0025326X06003213 [accessed May 21, 2014].
Tompkins, E. L., and W. N. Adger. 2004. Does adaptive management of natural resources enhance
resilience to climate change? Ecology and Society 9(2): article 10. Available at http://www.
cetesb.sp.gov.br/userfiles/file/mudancasclimaticas/proclima/file/publicacoes/adaptacao/ingles/
managementofnaturalresourcesandclimatechange.pdf [accessed May 21, 2014].
Trenberth, K. E., P. D. Jones, P. Ambenje, R. Bojariu, D. Easterling, A. K. Tank, D. Parker, F.
Rahimzadeh, J. A. Renwick, M. Rusticucci, B. Soden, and P. Zhai. 2007. Observations: surface
and atmospheric climate change. Pp. 235-336 in Climate Change 2007: The Physical Science
Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovern-
mental Panel on Climate Change, S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.
B. Averyt, M. Tignor and H. L. Miller, eds. Cambridge, UK and New York: Cambridge Univer-
sity Press. Available at http://www.ipcc.ch/publications_and_data/publications_ipcc_fourth_
assessment_report_wg1_report_the_physical_science_basis.htm [accessed May 23, 2014].
Trexler, J. C., W. F. Loftus, F. Jordan, J. J. Lorenz, J. H. Chick, and R. M. Kobza. 2000. Empirical
assessment of fish introductions in a subtropical wetland: an evaluation of contrasting views.
Biological Invasions 2:265-277. Available at http://link.springer.com/article/10.1023%2FA%
3A1011488118444 [accessed May 21, 2014].
USACE (U.S. Army Corps of Engineers). 1992. General Design Memorandum and Environmental
Impact Statement: Modified Water Deliveries to Everglades National Park. Atlanta, GA: USACE.
USACE. 2000. Planning Guidance Notebook. ER 1105-2-100 (April 2000). Washington, DC.
USACE. 2007. Memorandum for Director of Civil Works on Comprehensive Everglades Restoration
Plan, Water Quality Improvements. Washington, DC: USACE.

USACE. 2010. Fiscal Year 2011: Civil Works Budget of the U.S. Army Corps of Engineers. Depart-
ment of the Army, Office of Assistant Secretary of the Army (Civil Works). Wahsington, DC.
USACE. 2011a. Everglades Restoration Transition Plan: Final Environmental Impact Statement.
Jacksonville, Florida. Available at http://www.evergladesplan.org/pm/pm_docs/ertp/final_

dec_2011/feis/102612_ertp_feis_vol_1_dec_2011_main_report.pdf [accessed August 14, 2013].
USACE. 2011b. Fiscal Year 2012: Civil Works Budget of the U.S. Army Corps of Engineers. Depart-
ment of the Army, Office of Assistant Secretary of the Army (Civil Works). Washington, DC.
USACE. 2011c. Modified Water Deliveries, Everglades National Park & Tamiami Trail. Jacksonville,
Florida. Fact Sheet. Available at http://www.saj.usace.army.mil/Media/FactSheets/
FactSheetArticleView/tabid/6106/Article/479992/modified-water-deliveries-everglades-
national-park-tamiami-trail.aspx.
USACE. 2011d. Picayune Strand Restoration Project: Facts and Information, March. Available at
http://www.evergladesplan.org/docs/fs_picayune_mar_2011.pdf [accessed May 29, 2012].
USACE. 2011e. Sea Level Change Considerations for Civil Works Programs. EC 1165-2-212. Avail-
able at http://planning.usace.army.mil/toolbox/library/ECs/EC11652212Nov2011.pdf [accessed
May 21, 2014].
USACE. 2012a. Fiscal Year 2013: Civil Works Budget of the U.S. Army Corps of Engineers. Depart-
USACE. 2012b. Kissimmee River Restoration Project: Facts and Information. Jacksonville, FL. Avail-
able at http://www.saj.usace.army.mil/Portals/44/docs/FactSheets/Kissimmee_FS_July2012_508.
pdf [accessed August 16, 2013].
USACE. 2012c. Record of Decision Signed for Everglades Restoration Transition Plan. USACE News
Story. Available at http://www.saj.usace.army.mil/Media/NewsStories/tabid/6070/Article/5627/
record-of-decision-signed-for-everglades-restoration-transition-plan.aspx [accessed August 14,
2013].
USACE. 2012d. Review Plan for Interim Water Control Plan and Environmental Assessment
for Seminole Big Cypress Reservation Water Conservation Plan Critical Project Basin 4.
Jacksonville, FL. Available at http://www.saj.usace.army.mil/Portals/44/docs/Planning/

ReviewPlans/SeminoleCypressResWCPbasin4.pdf [accessed August 18, 2013].
USACE. 2013a. Annual Assessment Report: Water Year 2013 (October 1, 2012-September 30,
2013). Everglades Restoration Transition Plan.
USACE. 2013b. C-111 Spreader Canal Western Project: Facts and Information. Jacksonville, FL. Avail-
able at http://www.evergladesplan.org/docs/fs_c111_july_2013_508.pdf [accessed August 14,
2013].
USACE. 2013c. Fiscal Year 2014: Civil Works Budget of the U.S. Army Corps of Engineers. Depart-
USACE. 2013d. Indian River Lagoon-South. Facts and Information. Jacksonville, FL. Available at
http://www.evergladesplan.org/docs/fs_irls_aug_2013.pdf [accessed December 8, 2013].
USACE. 2013e. Picayune Strand Restoration Project: Draft Limited Reevaluation Report and Environ-
mental Assessment. Available at http://evergladesplan.org/pm/projects/project_docs/pdp_30_
sgge/lrr/052813_lrr_30_lrr_ea_main_report.pdf [accessed May 21, 2014].
USACE. 2013f. Program Management Plan for the Invasives Species Leadership Team and Invasive
Species CoP [Communities of Practice] to the Environmental CoP, December 2013 draft.
USACE. 2013g. Site 1 Impoundment, Fran Reich Preserve. Facts and Information, November.
Jacksonville, FL. Available at http://www.evergladesplan.org/docs/fs_site1_jan_2013_508.pdf
[accessed November 10, 2013].
USACE. 2014a. Biscayne Bay Coastal Wetlands Project Fact Sheet. Jacksonville, Florida. Available
at http://www.evergladesplan.org/docs/fs_bbcw_jan_2014_508.pdf
USACE. 2014b. Melaleuca Eradication and Other Exotic Plants: Implement Biological Controls Fact
Sheet, January. Available at http://www.evergladesplan.org/docs/fs_melaleuca_jan_2014.pdf

References 253
USACE. 2014c. Picayune Strand Restoration Project: Facts and Information. Jacksonville, FL. Avail-
able at http://www.evergladesplan.org/docs/fs_picayune_jan_2014.pdf [accessed March 1,
2014].
USACE. 2014d. Seminole Big Cypress Reservation Water Conservation Plan Fact Sheet, February.
Available at http://www.saj.usace.army.mil/Portals/44/docs/FactSheets/SeminoleBigCypress_
FS_February2014.pdf [accessed May 21, 2014].
USACE. 2014e. Press Release: Everglades Restoration Projects Move Forward to Restore River of
Grass. April 29, 2014. Office of the Assistant Secretary of the Army. Washington, D.C. Avail-
able at http://www.army.mil/article/124933/Everglades_restoration_projects_move_forward_
to_restore__river_of_grass_/ [accessed May 19, 2014].
USACE and SFWMD (U.S. Army Corps of Engineers and South Florida Water Management District).
1999. Central and Southern Florida Project Comprehensive Review Study, Final Integrated
Feasibility Report and Programmatic Environmental Impact Statement. Available at http://www.
evergladesplan.org/pub/restudy_eis.cfm#mainreport [accessed August 13, 2010].
USACE and SFWMD. 2010. Central and Southern Florida Project Comprehensive Everglades Restora-
tion Plan: Melaleuca Eradication and Other Exotic Plants, Implement Biological Controls. Final
Integrated Project implementation Report and Environmental Impact Statement, June. Available
at http://www.evergladesplan.org/pm/projects/project_docs/pdp_95_melaleuca/100710_mela_
final_pir_main.pdf [accessed May 21, 2014].
USACE and SFWMD. 2011a. Central and Southern Florida Project Comprehensive Everglades
Restoration Plan C-111 Spreader Canal Western Project Final Integrated Project Implementa-
tion R eport and Environmental Impact Statement. United States Army Corps of Engineers,
Jacksonville, FL, and South Florida Water Management District, West Palm Beach, FL.
USACE and SFWMD. 2011b. CERP Guidance Memorandum 56, Appendix A. CERP Guidance
Memorandum 56: Integration of Adaptive Management into Program and Project Management.
CGM Number-Revision: 056.00. C/O U.S. Army Corps of Engineers, Jacksonville District,
Jacksonville, FL and South Florida Water Management District, West Palm Beach, FL.
USACE and SFWMD. 2011c. Project monitoring plan, Part IV: Adaptive management plan. Annex
E in Biscayne Bay Coastal Wetlands Phase 1 Final Integrated PIR and EIS. Available at http://
www.evergladesplan.org/pm/projects/project_docs/pdp_28_biscayne/010612_fpir/010612_
vol_3_annex_e.pdf.
USACE and SFWMD. 2011d. Project monitoring plan. Annex E in C-111 Spreader Canal Western
Project Final Integrated PIR and EIS.
USACE and SFWMD. 2012a. Central and Southern Florida Project Comprehensive Everglades
Restoration Plan Biscayne Bay Coastal Wetlands Phase 1. Final Integrated Project implemen-
tation Report and Environmental Impact Statement. Jacksonville, FL. Available at http://www.
evergladesplan.org/pm/projects/docs_28_biscayne_bay_pir.aspx [accessed August 13, 2013].
USACE and SFWMD. 2012b. Central and Southern Florida Project Comprehensive Everglades
Restoration Plan Biscayne Bay Coastal Wetlands Phase 1: Final Integrated Project Implementa-
tion Report And Environmental Impact Statement. July 2011 – Revised March 2012. Available
at http://evergladesplan.org/pm/projects/project_docs/pdp_28_biscayne/010612_fpir/010612_
bbcw_vol_1_main_report_rev_mar_2012.pdf [accessed May 21, 2014].
USACE and SFWMD. 2012c. Preliminary integrated adaptive management plan for Decomp project.
Appendix D in Project Documentation Report for the Decompartmentalization of Water Con-
servation Area 3 and Sheet Flow Enhancement Project Implementation Report 1. Available at
http://www.evergladesplan.org/pm/projects/project_docs/pdp_12_decomp/092512_decomp_
doc_report/092512_decomp_doc_report_app_d.pdf.
USACE and SFWMD. 2012d. CERP Guidance Memorandum 062.00: Invasive and Native Nuisance
Species Management. Available online at http://corpslakes.usace.army.mil/employees/invasive/
pdfs/cgm_62-00.pdf.
USACE and SFWMD. 2012e. Project monitoring plan, Part IV: Adaptive management plan. Annex E
in Revised Final Broward County Water Preserve Areas PIR and EIS.

USACE and SFWMD. 2013a. Central and Southern Florida Project Comprehensive Everglades Res-
toration Plan: Final Technical Data Report, Comprehensive Everglades Restoration Plan Aquifer
Storage and Recovery Pilot Project. December. Available at http://www.evergladesplan.org/
pm/projects/project_docs/pdp_32_lake_o_asr_pilot/TDR_Final.pdf [accessed May 21, 2014].
USACE and SFWMD. 2013b. Central Everglades Planning Project Draft PIR/EIS. Available at http://
www.evergladesplan.org/pm/projects/docs_51_cepp_draft_pir.aspx [accessed April 4, 2014].
USACE and SFWMD. 2013c. CERP Project: Loxahatchee River Watershed Restoration Project. Avail-
able at http://www.evergladesplan.org/pm/projects/proj_17_lox_river.aspx [accessed August 14,
2013].
USACE and SFWMD. 2013d. Non-CERP South Florida Ecosystem Restoration Projects: Seminole
Big Cypress Reservation Water Conservation Plan. Available at http://www.saj.usace.army.mil/
Portals/44/docs/FactSheets/SeminoleBigCypress_FS_August2013_508.pdf [accessed Decem-
ber 8, 2013].
Van Driesche, R., M. Hoddle, and T. Center. 2008. Control of Pests and Weeds by Natural Enemies.
An Introduction to Biological Control. Malden, MA: Blackwell.
Volin, J. C., M. S. Lott, J. D. Muss, and D. Owen. 2004. Predicting rapid invasion of the Florida
Everglades by Old World Climbing Fern (Lygodium microphyllum). Diversity and Distributions
10:439-446.
Walsh, M. J. 2012. U.S. Army Corps of Engineers Civil Works Feasibility Study Program Execution
and Delivery Memorandum dated February 8, 2012. Washington, DC: USACE.
Wanless, H. R., and B. Vlaswinkel. 2005. Coastal Landscapes and Channel Evolution Affecting
Critical Habitats at Cape Sable, Everglades National Park, Florida. Final Report to Everglades
National Park. National Park Service, Department of the Interior.
Wanless, H. R., R. W. Parkinson, and L. P. Tedesco. 1994. Sea level control on stability of Everglades
wetlands. Pp. 199-222 in Everglades: The Ecosystem and Its Restoration, S. M. Davis and J. C.
Ogden, eds. Boca Raton, FL: St. Lucie Press.
Wanless, H. R., P. Oleck, L. P. Tedesco, and B. E. Hall. 2000. Next 100 years of evolution at the
Greater Everglades ecosystem in response to anticipated sea level rise: nature, extent and
causes. Presentation at Greater Everglades Ecosystem Restoration Science Conference, Naples,
FL.
Watts, D., M. J. Cohen, J. Heffernan, T. Z. Osborne, and M. W. Clark. 2010. Hydrologic modifica-
tion and the loss of self-organized patterning in the ridge slough mosaic of the Everglades.
Ecosystems 13:813-827.
Webb, E. L., D. A. Friess, K. W. Krauss, D. R. Cahoon, G. R. Guntenspergen, and J. Phelps. 2013.
A global standard for monitoring coastal wetland vulnerability to accelerated sea level rise.
Nature Climate Change 3:458-465. Available at http://www.nature.com/nclimate/journal/v3/
n5/full/nclimate1756.html [accessed May 21, 2014].
Westbrooks, R. G., and R. E. Eplee. 2011. Early detection and rapid response. Pp. 169-177 in
Encyclopedia of Biological Invasions, D. Simberloff and M. Rejmánek, eds. Berkeley: Univer-
sity of California Press.
Weston, N. B., R. E. Dixon, and S. B. Joye. 2006. Ramifications of salinity intrusion in tidal fresh
water sediments: Geochemistry and microbial pathways of organic matter mineralization.
Journal of Geophysical Research: Biogeosciences 111:G0100, doi:10.1029/2005JG000071.
Weston, N. B., M. A. Vile, D. C. Neubauer, and D. J. Velinsky. 2011. Accelerated microbial organic
matter mineralization following salt-water intrusion into tidal freshwater marsh soils. Biogeo-
chemistry 102:135-151.
Wetzel, P. R., A. G. van der Valk, S. Newman, C. A. Coronado, T. G. Troxler-Gann, D. L. Childers,
W. H. Orem, and F. H. Sklar. 2009. Heterogeneity of phosphorus distribution in a patterned
landscape, the Florida Everglades. Plant Ecology 200:83-90.
Wetzel, P. R., F. H. Sklar, C. A. Coronado, T. G. Troxler, S. L. Krupa, P. L. Sullivan, S. Ewe, R. M.
Price, S. Newman, and W. H. Orem. 2011. Biogeochemical processes on tree islands in the
greater Everglades: initiating a new paradigm. Critical Reviews in Environmental Science and
Technology 41(S1):670-701.

References 255
WG and SCG (Working Group and Science Coordination Group). 2010. New Science: Advancing
Understanding of the South Florida Ecosystem. Available at http://www.sfrestore.org/information_
brief/New_Science_May_20_2010.pdf [accessed May 25, 2012].
Whelan, K. R., T. J. Smith III, D. R. Cahoon, J. C. Lynch, and G. H. Anderson. 2005. Groundwater
control of mangrove surface elevation: shrink and swell varies with soil depth. Estuaries
26(6):833-843.
Whelan, K. R. T., T. J. Smith III, G. H. Anderson, and M. L. Ouellette. 2009. Hurricane Wilma’s im-
pact on overall soil elevation and zones within the soil profile in a mangrove forest. Wetlands
29:16-23.
Wheeler, G. S., G. S. Taylor, J. F. Gaskin, and M. F. Purcell. 2011. Ecology and management of
sheoak (Casuarina spp.), an invader of coastal Florida, U.S.A. Journal of Coastal Research
27:485-492.
Willard, D. A., and C. E. Bernhardt. 2011. Impacts of past climate and sea level change on Ever-
glades wetlands: placing a century of anthropogenic change into a Late Holocene context.
Climatic Change 107:59-80.
Williams, K., K. C. Ewel, R. P. Stumpf, F. E. Putz, and T. W. Workman. 1999. Sea-level rise and
coastal forest retreat on the west coast of Florida, USA. Ecology 80:2045-2063.
Witmer, G. W., J. Eisemann, P. Hall, M. L. Avery, and A. Duffiney. 2010a. Challenges and unique
solutions to rodent eradication in Florida. Pp. 23-28 in Proceedings of the 24th Vertebrate Pest
Conference, R. M. Timm and K. A. Fagerstone, eds. Davis: University of California.
Witmer, G. W., N. P. Snow, and P. W. Burke. 2010b. Potential attractants for detecting and removing
invading Gambian pouched rats (Cricetomys gambianus). Pest Management Science 66:412-416.
Wyatt, J. L., and E. A. Forys. 2004. Conservation implications of predation by Cuban treefrogs
(Osteopilus septentrionalis) on native hylids in Florida. Southeastern Naturalist 3:695-700.
Zheng, L. Y., and R. H. Weisberg. 2012. Modeling the west Florida coastal ocean by downscaling
from the deep ocean, across the continental shelf and into the estuaries. Ocean Modelling
48:10-29.


Acronyms
AF acre-feet
AMO Atlantic Multidecadal Oscillation
ARS Agricultural Research Service
ASR aquifer storage and recovery
BCNP Big Cypress National Preserve

BCWPA Broward County Water Preserve Areas
BIA Bureau of Indian Affairs
BMP best management practice
CEPP Central Everglades Planning Project

CERP Comprehensive Everglades Restoration Plan
CESI Critical Ecosystem Studies Initiative
CFU colony-forming unit
CISRERP Committee on Independent Scientific Review of Everglades
Restoration Progress
COP combined operational plan
CROGEE Committee on the Restoration of the Greater Everglades Ecosystem
C&SF Central and Southern Florida
DACS Florida Department of Agriculture and Consumer Services

DOI U.S. Department of the Interior
DOP dissolved organic phosphorus
DPI Florida Department of Plant Industry
DPM Decomp Physical Model
EAA Everglades Agricultural Area

EAV emergent aquatic vegetation
ECISMA Everglades Cooperative Invasive Species Management Act
257
EDDMapS Early Detection and Distribution Management System

EDRR early detection and rapid response
EIRAMP Everglades Invasive Reptile and Amphibian Monitoring Program
ENP Everglades National Park
EPA U.S. Environmental Protection Agency
ERTP Everglades Restoration Transition Plan
FACA Federal Advisory Committee Act

FEB flow equalization basin
FWC Florida Fish and Wildlife Conservation Commission
FWM flow-weighted mean
FWO future without the project
FWS U.S. Fish and Wildlife Service
FY fiscal year
GCM general circulation model

GIS geographic information system
HASR Hillsboro Aquifer Storage and Recovery

HSI habitat suitability index
IDS Integrated Delivery Schedule

IOP Interim Operational Plan
IPCC Intergovernmental Panel on Climate Change
IPR in-progress review
IRL-S Indian River Lagoon-South
KRASR Kissimmee River Aquifer Storage and Recovery
LNWR Loxahatchee National Wildlife Refuge

LOER Lake Okeechobee and Estuary Recovery
LOPA Lake Okeechobee Protection Act
LOPP Lake Okeechobee Protection Plan
LORS Lake Okeechobee Regulation Schedule
LPA Limestone Products Association
MAP monitoring and assessment plan

MCDA Multi Criteria Decision Analysis
MGD million gallons per day
NAVD North American Vertical Datum

Acronyms 259
NCRS Natural Resources Conservation Service

NEPA National Environmental Policy Act
NRC National Research Council
OERI Office of Everglades Restoration Initiatives
PDT project delivery team

PIRs project implementation reports
PP particulate phosphorus
PPA project partnership agreement
ppb parts per billion
PPDR Pilot Project Design Report
RCP Representative Concentration Pathway

RECOVER Restoration, Coordination, and Verification
SAV submerged aquatic vegetation

SCG Science Coordination Group
SFERTF South Florida Ecosystem Restoration Task Force
SFWMM South Florida Water Management Model
SSR System Status Report
STA Stormwater Treatment Area
TIME Tides and Inflows in the Mangroves of the Everglades

TP total phosphorus
TSP Tentatively Selected Plan
USACE U.S. Army Corps of Engineers

USDA U.S. Department of Agriculture
WCA Water Conservation Area

WQ Water Quality
WQBEL water quality-based effluent limit
WRDA Water Resources Development Act
WRRDA Water Resources Reform and Development Act
WY water year


Appendix A
National Research Council

Everglades Reports
Progress Toward Restoring the Everglades: The Fourth Biennial Review, 2012
(2012)
This report is the fourth biennial evaluation of progress being made in the
Comprehensive Everglades Restoration Plan (CERP), a multi-billion-dollar effort
to restore historical water flows to the Everglades and return the ecosystem
closer to its natural state. The report finds that 12 years into the Comprehensive
Everglades Restoration Project, little progress has been made in restoring the core
of the remaining Everglades ecosystem; instead, most project construction so far
has occurred along its periphery. To reverse ongoing ecosystem declines, it will
be necessary to expedite restoration projects that target the central Everglades,
and to improve both the quality and quantity of the water in the ecosystem.
The new Central Everglades Planning Project offers an innovative approach to
this challenge, although additional analyses are needed at the interface of water
quality and water quantity to maximize restoration benefits within existing legal
constraints.
Progress Toward Restoring the Everglades: The Third Biennial Review, 2010
(2010)
This report is the third biennial evaluation of progress being made in the
to restore historical water flows to the Everglades and return the ecosystem closer
to its natural state. The report finds that while natural system restoration prog-
ress from CERP remains slow, in the past 2 years, there have been noteworthy
improvements in the pace of implementation and in the relationship between the
federal and state partners. Continued public support and political commitment
to long-term funding will be needed for the restoration plan to be completed.
The science program continues to address important issues, but more transparent
261
262 Appendix A
mechanisms for integrating science into decision making are needed. Despite
such progress, several important challenges related to water quality and water
quantity have become increasingly clear, highlighting the difficulty of achiev-
ing restoration goals simultaneously for all ecosystem components. Achieving
these goals will be enormously costly and will take decades at least. Rigorous
scientific analyses of potential conflicts among the hydrologic requirements of
Everglades landscape features and species, and the tradeoffs between water
quality and quantity, considering timescales of reversibility, are needed to inform
future prioritization and funding decisions. Understanding and communicating
these tradeoffs to stakeholders are critical.
Progress Toward Restoring the Everglades: The Second Biennial Review, 2008
(2008)
This report is the second biennial evaluation of progress being made in the
to restore historical water flows to the Everglades and return the ecosystem closer
to its natural state. Launched in 2000 by the U.S. Army Corps of Engineers and
the South Florida Water Management District, the CERP is a multiorganization
planning process that includes approximately 50 major projects to be completed
over the next several decades. The report concludes that budgeting, planning,
and procedural matters are hindering a federal and state effort to restore the
Florida Everglades ecosystem, which is making only scant progress toward
achieving its goals. Good science has been developed to support restoration
efforts, but future progress is likely to be limited by the availability of funding
and current authorization mechanisms. Despite the accomplishments that lay
the foundation for CERP construction, no CERP projects have been completed to
date. To begin reversing decades of decline, managers should address complex
planning issues and move forward with projects that have the most potential to
restore the natural ecosystem.
Progress Toward Restoring the Everglades: The First Biennial Review, 2006
(2007)
This report is the first in a congressionally mandated series of biennial

evaluations of the progress being made by the CERP, a multi-billion-dollar effort
to restore historical water flows to the Everglades and return the ecosystem
closer to its natural state, before it was transformed by drainage and by urban
and agricultural development. The report finds that progress has been made in
developing the scientific basis and management structures needed to support
a massive effort to restore the Florida Everglades ecosystem. However, some

Appendix A 263
important projects have been delayed due to several factors including budget-
ary restrictions and a project planning process that can be stalled by unresolved
scientific uncertainties. The report outlines an alternative approach that can help
the initiative move forward even as it resolves remaining scientific uncertainties.
The report calls for a boost in the rate of federal spending if the restoration of
Everglades National Park and other projects are to be completed on schedule.
Re-engineering Water Storage in the Everglades: Risks and Opportunities

(2005)
Human settlements and flood control structures have significantly reduced

the Everglades, which once encompassed more than 3 million acres of slow-
moving water enriched by a diverse biota. The CERP was formulated in 1999
with the goal of restoring the original hydrologic conditions of the remaining
Everglades. A major feature of this plan is providing enough storage capacity to
meet human and ecological needs. This report reviews and evaluates not only
storage options included in the plan, but also other options not considered in
the plan. Along with providing hydrologic and ecological analyses of the size,
location, and functioning of water storage components, the report also discusses
and makes recommendations on related critical factors, such as timing of land
acquisition, intermediate states of restoration, and tradeoffs among competing
goals and ecosystem objectives.
The CERP imposes some constraints on sequencing of its components.
The report concludes that two criteria are most important in deciding how to
sequence components of such a restoration project: (1) protecting against addi-
tional habitat loss by acquiring or protecting critical lands in and around the
Everglades and (2) providing ecological benefits as early as possible.
There is a considerable range in the degree to which various proposed
storage components involve complex design and construction measures, rely
on active controls and frequent equipment maintenance, and require fossil
fuels or other energy sources for operation. The report recommends that, to the
extent possible, the CERP should develop storage components that have fewer
of those requirements, and are thus less vulnerable to failure and more likely to
be sustainable in the long term.
Further, as new information becomes available and as the effectiveness
and feasibility of various restoration components become clearer, some of the
earlier adaptation and compromises might need to be revisited. The report
recommends that methods be developed to allow tradeoffs to be assessed over
broad spatial and long temporal scales, especially for the entire ecosystem, and
gives an example of what an overall performance indicator for the Everglades
system might look like.

264 Appendix A
Adaptive Monitoring and Assessment for the Comprehensive Everglades

Restoration Plan (2003)
A key premise of the CERP is that restoring the historical hydrologic regime
in the remaining wetlands will reverse declines in many native species and bio-
logical communities. Given the uncertainties that will attend future responses
of Everglades ecosystems to restored water regimes, a research, monitoring,
and adaptive management program is planned. This report assessed the extent
to which the restoration effort’s “monitoring and assessment plan” included
the following elements crucial to any adaptive management scheme: (1) clear
restoration goals and targets, (2) a sound baseline description and conceptu-
alization of the system, (3) an effective process for learning from management
actions, and (4) feedback mechanisms for improving management based on the
learning process.
The report concludes that monitoring needs must be prioritized, because
many goals and targets that have been agreed to may not be achievable or inter-
nally consistent. Priorities could be established based on the degree of flexibility
or reversibility of a component and its potential impact on future management
decisions. Such a prioritization should be used for scheduling and sequencing
of projects, for example. Monitoring that meets multiple objectives (e.g., adap-
tive management, regulatory compliance, and a “report card”) should be given
priority.
Ecosystem-level, systemwide indicators should be developed, such as land-
cover and land-use measures, an index of biotic integrity, and diversity measures.
Regionwide monitoring of human and environmental drivers of the ecosystem,
especially population growth, land-use change, water demand, and sea-level
rise are recommended. Monitoring, modeling, and research should be well
integrated, especially with respect to defining the restoration reference state and
using “active” adaptive management.
Does Water Flow Influence Everglades Landscape Patterns? (2003)
A commonly stated goal of the CERP is to “get the water right.” This has
largely meant restoring the timing and duration of water levels and the water
quality in the Everglades. Water flow (speed, discharge, direction) has been con-
sidered mainly in the coastal and estuarine system, but not elsewhere. Should
the restoration plan be setting targets for flows in other parts of the Everglades
as well?
There are legitimate reasons why flow velocities and discharges have thus
far not received greater emphasis in the plan. These include a relative lack of
field information and poor resolution of numerical models for flows. There are,

Appendix A 265
however, compelling reasons to believe that flow has important influences in

the central Everglades ecosystem. The most important reason is the existence
of major, ecologically important landforms—parallel ridges, sloughs, and “tree
islands”—are aligned with present and inferred past flow directions. There are
difficulties in interpreting this evidence, however, as it is essentially circumstan-
tial and not quantitative.
Alternative mechanisms by which flow may influence this landscape can
to some extent be evaluated from short-term research on underlying bedrock
topography, detailed surface topographic mapping, and accumulation rates of
suspended organic matter. Nonetheless, more extensive and long-term research
will also be necessary, beginning with the development of alternative con-
ceptual models of the formation and maintenance of the landscape to guide a
research program. Research on maintenance rather than evolution of the land-
scape should have higher priority because of its direct impact on restoration.
Monitoring should be designed for the full range of flow conditions, including
extreme events.
Overall, flows approximating historical discharges, velocities, timing, and
distribution should be considered in restoration design, but quantitative flow-
related performance measures are not appropriate until there is a better scientific
understanding of the underlying science. At present, neither a minimum nor a
maximum flow to preserve the landscape can be established.
Florida Bay Research Programs and Their Relation to the Comprehensive

Everglades Restoration Plan (2002)
This report of the Committee on Restoration of the Greater Everglades Eco-

system (CROGEE) evaluated Florida Bay studies and restoration activities that
potentially affect the success of the CERP. Florida Bay is a large, shallow marine
system immediately south of the Everglades, bounded by the Florida Keys and the
Gulf of Mexico. Some of the water draining from the Everglades flows directly
into northeast Florida Bay. Other freshwater drainage reaches the bay indirectly
from the northwest.
For several decades until the late 1980s, clear water and dense seagrass
meadows characterized most of Florida Bay. However, beginning around 1987,
the seagrass beds began dying in the western and central bay. It is often
assumed that increased flows to restore freshwater Everglades habitats will
also help restoration of Florida Bay. However, the CERP may actually result in
higher salinities in central Florida Bay than exist presently, and thus exacerbate
the ecological problems. Further, some percentage of the proposed increase
in fresh surface-water flow discharging northwest of the bay will eventually
reach the central bay, where its dissolved organic nitrogen may lead to algal

266 Appendix A
blooms. Complicating the analysis of such issues is the lack of an operational

bay circulation model.
The report notes the importance of additional research in the following
areas: estimates of groundwater discharge to the bay; full characterization and
quantification of surface runoff in major basins; transport and total loads of nitro-
gen and phosphorous from freshwater sources, especially in their organic forms;
effects on nutrient fluxes of decreasing freshwater flows into the northeastern
bay, and of increasing flows northwest of the bay; and the development of an
operational Florida Bay circulation model to support a bay water quality model
and facilitate analysis of CERP effects on the bay.
Science and the Greater Everglades Ecosystem Restoration: An Assessment of

the Critical Ecosystems Study Initiative (2003)
The Everglades represents a unique ecological treasure, and a diverse group

of organizations is currently working to reverse the effects of nearly a century
of wetland drainage and impoundment. The path to restoration will not be
easy, but sound scientific information will increase the reliability of the restora-
tion, help enable solutions for unanticipated problems, and potentially reduce
long-term costs. The investment in scientific research relevant to restoration,
however, decreased substantially within some agencies, including one major
Department of the Interior (DOI) science program, the Critical Ecosystem Studies
Initiative (CESI). In response to concerns regarding declining levels of funding
for scientific research and the adequacy of science-based support for restora-
tion decision making, the U.S. Congress instructed the DOI to commission the
National Academy of Sciences to review the scientific component of the CESI
and provide recommendations for program management, strategic planning, and
information dissemination.
Although improvements should be made, this report notes that the CESI has
contributed useful science in support of the DOI’s resource stewardship interests
and restoration responsibilities in South Florida. It recommends that the funda-
mental objectives of the CESI research program remain intact, with continued
commitment to ecosystem research. Several improvements in CESI management
are suggested, including broadening the distribution of requests for proposals
and improving review standards for proposals and research products. The report
asserts that funding for CESI science has been inconsistent and as of 2002 was
less than that needed to support the DOI’s interests in and responsibilities for
restoration. The development of a mechanism for comprehensive restoration-
wide science coordination and synthesis is recommended to enable improved
integration of scientific findings into restoration planning.

Appendix A 267
Regional Issues in Aquifer Storage and Recovery for Everglades Restoration:

A Review of the ASR Regional Study Project Management Plan of the
Comprehensive Everglades Restoration Plan (2002)
The report reviews a comprehensive research plan on Everglades restora-

tion drafted by federal and Florida officials that assesses a central feature of the
restoration: a proposal to drill more than 300 wells funneling up to 1.7 billion
gallons of water a day into underground aquifers, where it would be stored and
then pumped back to the surface to replenish the Everglades during dry periods.
The report says that the research plan goes a long way to providing information
needed to settle remaining technical questions and clearly responds to sugges-
tions offered by scientists in Florida and in a previous report by the NRC.
Aquifer Storage and Recovery in the Comprehensive Everglades Restoration

Plan: A Critique of the Pilot Projects and Related Plans for ASR in the Lake
Okeechobee and Western Hillsboro Areas (2001)
Aquifer storage and recovery (ASR) is a major component in the CERP,

which was developed by the U.S. Army Corps of Engineers (USACE) and the
South Florida Water Management District (SFWMD). The plan would use the
upper Floridian aquifer to store large quantities of surface water and shallow
groundwater during wet periods for recovery during droughts.
ASR may limit evaporation losses and permit recovery of large volumes of
water during multiyear droughts. However, the proposed scale is unprecedented
and little subsurface information has been compiled. Key unknowns include
impacts on existing aquifer uses, suitability of source waters for recharge, and
environmental and/or human health impacts due to water quality changes dur-
ing subsurface storage.
To address these issues, the USACE and SFWMD proposed aquifer storage
recharge pilot projects in two key areas. The CROGEE charge was to examine
a draft of their plans from a perspective of adaptive management. The report
concludes that regional hydrogeologic assessment should include development
of a regional-scale groundwater flow model, extensive well drilling and water
quality sampling, and a multiobjective approach to ASR facility siting. It also rec-
ommends that water quality studies include laboratory and field bioassays and
ecotoxicological studies, studies to characterize organic carbon of the source
water and anticipate its effects on subsurface biogeochemical processes, and
laboratory studies. Finally, it recommends that pilot projects be part of adaptive
assessment.


Appendix B
Additional Major Nonnative Plant and

Animal Species in the Everglades
This appendix provides additional information about the occurrence and

threats posed by invasive plants and animals in the Everglades, as well as avail-
able management strategies.
PLANTS
Water Hyacinth (Eichhornia crassipes)
South American water hyacinth is one of the world’s worst invasive s pecies
(Mitchell, 1976). It can double its biomass in 6 days (Mitchell, 1976) and has
average annual biomass production of up to 250 tons of dry weight per hectare
(Spencer and Bowes, 1986). In the Everglades, water hyacinth readily invades
wetlands, freshwater lakes, and other open-water habitats, where it effectively
outcompetes other floating species, and its dense mats shade out submerged
aquatic vegetation. This highly aggressive species is treated using a combina-
tion of herbicide and biocontrol agents, including the latest release in 2010, the
water hyacinth plant hopper (Megamelus scutellaris). Some biocontrol agents
have reduced biomass (> 50 percent) and seed production. Other potential
agents are undergoing trials and are in development. However, chemical control,
primarily with 2,4-D, is currently the main factor in controlling water hyacinth
(Schardt, 1997).
Hydrilla (Hydrilla verticillata)
This Asian species, introduced in Florida in the 1950s, is a submerged

aquatic that has become a widespread invader. It can displace aquatic species,
affecting the composition and functioning of communities. Heavy infestations
clog waterways, interfering with recreation and navigation. It is readily dispersed
by boat traffic and has greatly expanded its distribution in recent years. Herbicide
269
270 Appendix B
application is the main form of management, but recently detected resistance

to the most commonly used herbicide, fluridone, leads to concerns that it will
cease to be effective (Puri et al., 2007). Mechanical harvesting is sometimes used
for particularly dense infestations.
Air Potato (Dioscorea bulbifera)
Air potato is an Asian vine. It was introduced to the United States as an

ornamental, but in many areas of the world (e.g., Africa) its starchy bulbils are
used as a food (Ewe et al., 2006). In its introduced range, it spreads vegetatively
through dispersal of its bulbils (Rodgers et al., 2013), and it can quickly blanket
an area, shading out native vegetation. Chemical control has been used in the
past. Recently, a foliage-consuming beetle from China (Lilioceris cheni), released
as a biocontrol agent, has been successfully established in the field and is impos-
ing substantial defoliation. A second beetle from China (L. egeria) is under testing
and shows promise for its bulbil consumption (Center et al., 2013).
Shoebutton Ardisia (Ardisia elliptica)
This ornamental shrub was introduced from Southeast Asia to Florida in

the early 1900s (Gordon and Thomas, 1997). It is bird-dispersed, which facili-
tated its escape into surrounding natural areas (Ewe et al., 2006), including tree
islands, hammocks, and short-hydroperiod wetlands that it readily invades in
the Everglades. Highly shade-tolerant, this shrub creates thickets that cast dense
shade, excluding regeneration of other species. It is difficult to detect and con-
trol. Cutting, followed by herbicide application to the stumps, has been used
effectively in some invasions. However, where the population is exceptionally
dense, a two-step process of shredding and herbicide application has been the
most effective (Rodgers et al., 2013).
Torpedograss (Panicum repens)
This invasive exotic grass is native to Africa and Eurasia and has invaded a
broad range of wetland and successional habitats in Florida. Despite its wide-
spread distribution in South Florida, it is not known to reproduce via seeds but
spreads rapidly and aggressively via vegetative means. It readily displaces native
species and (along with Melaleuca) has expanded rapidly into the marshes
around Lake Okeechobee (Ogden et al., 2005). If not controlled, torpedograss
is likely to impact CERP performance measures in Lake Okeechobee, such as
increased native fish recruitment or the recovery of native vegetation (RECOVER,
2014a). Control efforts are largely through aerial and ground-level herbicide

Appendix B 271
application. Because it is a grass and the Everglades system has many native
grass species, this species is not likely to be a target for biocontrol development
Downy Rose Myrtle (Rhodomyrtus tomentosa)
This Asian shrub was introduced as an ornamental and has now spread into
pine flatwoods and some cypress stands, where it can displace native s pecies.
It occurs in many counties of central and South Florida, particularly along
the coast. The extent of its infestation is unclear, however, and on-the-ground
surveys/observations are required to detect it unambiguously. Herbicide appli-
cation and shredding have been the most effective methods for controlling it. A
potential biocontrol agent is currently under testing and development (Rodgers
et al., 2014a).
Cogongrass (Imperata cylindrica)
Cogongrass is native to southeastern Asia and is a highly aggressive, wide-

spread exotic grass in the southeastern United States (Evans et al., 2007). It has
a deep, dense rhizomatous mat (Byrd and Bryson, 1999), allowing it to resprout
readily, even after fire. The chemical residue it exudes into the soil can hinder
establishment of some species (Koger and Bryson, 2004), and it can alter fire
regimes (Lippincott, 2000). This species negatively affects native biodiversity
and is one of the few that has been definitively shown to result in native species
extirpations, particularly of low-stature forbs in pine savannas of the southeastern
United States (Brewer, 2008). Its coverage is estimated at ~1 million acres in
Florida (Miller, 2007), including pine flatwood and freshwater marsh commu-
nities. Control efforts include mechanical removal (often repeated), herbicide
application, and repeated prescribed fire. No biocontrol agents have been iden-
tified (Rodgers et al., 2014a), and it is an unlikely candidate for that program
because of its close phylogenetic relationships with many native species in the
Everglades (P. Tipping, USDA, personal communication, 2013).
Water Lettuce (Pistia stratiotes)
Water lettuce is a native of temperate to tropical regions of South America.

Its status in Florida is uncertain; it may be another “native invader” (Evans,
2013). In any event, it is a highly invasive floating macrophyte that has invaded
many open-water and wetland habitats in the Everglades (Rodgers et al., 2014a).
This species can take advantage of elevated nutrients and quickly expand its
distribution (along with Eichhornia) (Ogden et al., 2005), clogging waterways

272 Appendix B
and shading out submerged aquatic species. Herbicide application is the most
effective control method, and it has been used quite successfully in many canals
in South Florida. Biocontrol agents released to control this species have had
minimal impact on its abundance.
Tropical American Water Grass (Luziola subintegra)
This nonnative aquatic grass species was first reported in 2007 from Lake
Okeechobee. Plants with aquatic and terrestrial morphologies were documented.
Preliminary observations suggest that this species could become a highly aggres-
sive invader capable of displacing native species and altering structure and func-
tion of these aquatic habitats. Herbicide application has been used to control
this species.
Black Mangrove (Lumnitzera racemosa)
This nonnative mangrove was planted in Fairchild Tropical Botanic Garden

in 1966-1971 (14 plants) and an infestation was discovered in nearby Miami-
Dade County in 2008 (Possley, no date). It is listed as a priority species and
focused eradication efforts are under way and expected to be successful (Possley,
no date).
Mile-a-Minute (Mikania micrantha)
This invasive vine is also known as climbing hempweed, Chinese creeper,

and bittervine. It has recently been detected in the Redlands area of Miami-Dade
County. It is fast-growing and potentially invasive; it is listed as a priority species.
Because of its currently limited distribution efforts are focused on containment
and eradication.1
Skunk Vine (Paederia foetida)
This Asian vine was introduced to the United States in the late 1800s as a
potential fiber plant. It has a woody root stock and can produce trailing aboveg-
round vines that extend ~10 m in length. It is widely distributed in Florida. It
can reproduce vegetatively and via seed that is dispersed by frugivorous birds.
It can invade a wide range of habitats where it can displace native species and
See http://www.freshfromflorida.com/Divisions-Offices/Plant-Industry/Pests-Diseases/Mikania-
1
Micrantha-Mile-a-minute.

Appendix B 273
blanket stands of trees (University of Florida, Institute of Food and Agricultural

Sciences Extension).
Climbing Cassia (Senna pendula)
This evergreen South American shrub is an ornamental widely cultivated in

Florida. It has escaped into the wild and is naturalized in South Florida. Some of
its characteristics suggest that it may become highly invasive, as it is known to
invade hammocks and cypress strands (Richard and Ramey, 2007). Further, it is
a legume, and therefore its capacity for nitrogen fixation should be considered
when evaluating its potential invasiveness.
Feathered Mosquitofern (Azolla pinnata)
Feathered mosquitofern is an aquatic floating fern that is establishing in

canals and some open-water habitat in South Florida (Pemberton and Bodle,
2009). This species is native to parts of Africa, Asia, and Australia, but recent
molecular evidence suggests that the subspecies invading the Everglades is of
Australian origin, suggesting that biocontrol efforts that focus on Australian
insects may be fruitful (Madeira et al., 2013). Earlier research indicates that
the native herbivorous weevil Stenopelmus rufinasus, which uses native Azolla
spp., has been found in association with A. pinnata and is a potential “native”
biocontrol agent (Pemberton and Bodle, 2009) that has not, to our knowledge,
been investigated further. Currently, herbicide application is used for control.
ANIMALS
Gambian Pouched Rat (Cricetomys gambianus)
Gambian pouched rats are established on Grassy Key in the Florida Keys,
despite a long-term eradication effort. Although they are currently restricted to
Grassy Key, the concern is that they could be inadvertently or deliberately car-
ried to mainland South Florida. They are difficult to trap (Witmer et al., 2010b),
as witness the persistence of the Grassy Key population. As they are the largest
muroid rodent in the world, their impact in the Everglades could be enormous,
although there is no substantial research on the magnitude of this threat should
they become established. They also carry monkeypox. The Florida Fish and
Wildlife Conservation Commission (FWC) and U.S. Department of Agriculture
plan to continue trapping on Grassy Key to the extent that funding permits
(Rodgers et al., 2014a), though they are hindered by the fact that they cannot
gain access to six private properties (Witmer et al., 2010a).

274 Appendix B
Northern African Python (Python sebae)
The northern African python is apparently still established in Dade County—

two individuals were captured and a third was photographed in March 2013 (C.
Romagosa, Auburn University, perconal communication, 2013). Were they to
become widespread, their impacts might be similar to those of the Burmese python.
Other than monitoring, there is no long-term management plan at this time.
Oustalet’s Chameleon (Furcifer oustaleti)
A population of this large chameleon was discovered in rural Dade County

in 2011; FWC removed a large number of them, and an interagency team is
periodically monitoring the population (Rodgers et al., 2013). It is unclear what
observation would trigger management activity or further research.
Veiled Chameleon (Chamaeleo calyptratus)
This large chameleon is native to the Middle East. A breeding population

was discovered in Lee County in 2002, and recently, another population was
discovered in an agricultural area in south Miami-Dade County less than 4 miles
from the boundary of Everglades National Park (Rodgers et al., 2014a). Their
presence is suspected to have resulted from “intentional releases by reptile
enthusiasts” (Rodgers et al., 2014a).
Spectacled Caiman (Caiman crocodilus)
This crocodilian is native to southern Mexico and southward to Argentina. It

can attain 8 feet in its native habitat but usually is less than 6 feet long in Florida,
where its distribution results from escapes or releases from the pet trade (Florida
Fish and Wildlife Conservation Commission, 2014). Breeding populations have
been reported from Miami-Dade and Broward counties; its northern expansion
is limited by occasional freezes (Florida Fish and Wildlife Conservation Com-
mission, 2014). Control methods include egg collection and hunting.
Nile Monitor (Varanus niloticus)
This large predatory lizard, native to Africa, is said to be adaptable and intel-
ligent (Bennett, 1998). It is a generalist feeder, and impacts on South Florida’s
fauna are unknown, although it has a high potential for predation and competi-
tion (Rodgers et al., 2014a). It is established in several South Florida locations,
and control methods—currently applied in piecemeal fashion—include snares,

Appendix B 275
traps, and hunting (Rodgers et al., 2014a). It is monitored in Palm Beach County
Cuban Tree Frog (Osteopilus septentrionalis)
This tree frog is ubiquitous and common throughout the Everglades, and there
is substantial evidence that it depresses native tree frog populations (Rodgers et
al., 2014a; Smith, 2005; Wyatt and Forys, 2004). In the Picayune Strand Resto-
ration Project, the Cuban tree frog colonized and dominates the restored areas,
rather than native tree frogs that are found in the reference sites (RECOVER,
2014b). Predicted climate change is likely to be favorable for the Cuban tree
frog (Rödder and Weinsheimer, 2009). The Everglades Invasive R eptile and
Amphibian Monitoring Program records Cuban tree frogs on its routes (Rodgers
et al., 2014a), but there is no coordination or management effort for this tree
frog, and basic research on its impacts and possible control methods is needed
Cane Toad (Rhinella marina, formerly Bufo marinus)
Despite the fact that the cane toad is a legendary invader with major impacts
in Australia and elsewhere (Lever, 2001), there does not appear to be concern
over its possible impacts in the Everglades. They are uncommon in the heart
of the Everglades but very common in suburban areas (C. Romagosa, Auburn
University, personal communication, 2013). There has been no substantial study
of status or impacts throughout the Everglades, despite the fact that Punzo and
Lindstrom (2001) found that ingestion of its eggs causes massive mortality among
several native Everglades vertebrates.
Purple Swamp Hen (Porphyrio porphyrio)
This species established in the Everglades ca. 1996 and has spread widely
from an initial location in Pembroke Pines despite sustained efforts (which ter-
minated in 2009) to limit the population (Rodgers et al., 2013). This is a highly
aggressive, territorial species that is omnivorous, feeding on, among other things,
eggs and young of waterfowl. There is no current coordinated monitoring or
control effort (Rodgers et al., 2014a).
Asian Swamp Eel (Monopterus albus)
This is a large, generalized predator that has now spread substantially into
the Everglades (Rodgers et al., 2014a). It appears to be expanding northward and

276 Appendix B
is commonly found in some canals (Kline et al., 2013). There is no coordinated

monitoring or management (Rodgers et al., 2014a).
Mayan Cichlid (Cichlasoma urophthalmus)
This fish became abundant in the estuarine zone of northern Florida Bay,
and density of native fishes varies inversely with density of the Mayan cichlid
(Trexler et al., 2000). It has now become intermittently abundant even in the
northern reaches of Everglades National Park (Trexler et al., 2000). Observational
data show interference competition with and predation on several native fish
species (Trexler et al., 2000), although no research has linked the Mayan cichlid
to population-level impacts on native species. The Mayan cichlid and other
introduced fish potentially affect various CERP performance measures, such
as regional population sizes of fishes, crayfish, grass shrimp, and amphibians.
Pike Killifish (Belonesox belizanus)
The pike killifish persisted in small populations in canals east of the Ever-
glades for over 20 years before dramatically spreading across much of the
Everglades, where its density fluctuates greatly locally (Trexler et al., 2000).
There is no definitive evidence of impact or lack of impact on any native species.
Black Acara (Cichlasoma bimaculatum)
Black acara populations fluctuate locally and the species is common in solu-
tion holes (Trexler et al., 2000). There is no evidence of impact on native species.
Giant African Land Snail (Lissachatina fulica)
This snail, slated for eradication, was discovered in Miami in 2011. It eats
a wide variety of vegetation, including agricultural, horticultural, and native
species (Rodgers et al., 2014a). In addition, it is an intermediate host of rat lung-
worm, which can infect humans and cause meningitis (Rodgers et al., 2014a). A
prior infestation occurred in 1966; eradication took 10 years and cost $1 million.
Eradication in this case again seems likely (Rodgers et al., 2014).
Mexican Bromeliad Weevil (Metamasius callizona)
This species, introduced in the late 1980s, attacks bromeliads, including

species of conservation concern. It has spread widely in the Everglades. There
is no coordinated regional monitoring because of lack of funding. A parasitic

Appendix B 277
fly was approved for release as a biological control but has not established a
population. Research is ongoing by University of Florida scientists on a biologi-
cal control agent (Rodgers et al., 2014a).
Rugose Spiraling Whitefly (Aleurodicus rugioperculatus)
First discovered in South Florida in 2009 and rapidly spreading throughout

the region, this whitefly attacks and kills many host plants, both native and non-
native, and achieves massive densities (Stocks and Hodges, 2012). There is no
research on its ecological impact. A parasitic fly was approved for release as a
biological control but has not established a population (Rodgers et al., 2014a).
Research on biological or chemical control methods is under way at the Uni-
versity of Florida.


Appendix C
Water Science and Technology Board;

Board on Environmental
Studies and Toxicology
WATER SCIENCE AND TECHNOLOGY BOARD
GEORGE M. HORNBERGER, Chair, Vanderbilt University, Nashville, Tennessee

EDWARD J. BOUWER, Johns Hopkins University, Baltimore, Maryland
YU-PING CHIN, Ohio State University, Columbus
M. SIOBHAN FENNESSY, Kenyon College, Gambier, Ohio
BEN GRUMBLES, Clean Water America Alliance, Washington, D.C.
GEORGE R. HALLBERG, The Cadmus Group, Watertown, Massachusetts
CATHERINE L. KLING, Iowa State University, Ames
DEBRA S. KNOPMAN, RAND Corporation, Arlington, Virginia
LARRY LARSON, Association of State Floodplain Managers, Madison, Wisconsin
RITA P. MAGUIRE, Maguire & Pearce PLLC, Phoenix, Arizona
DAVID I. MAURSTAD, OST, Inc., McLean, Virginia
ROBERT SIMONDS, The Robert Simonds Company, Culver City, California
FRANK H. STILLINGER, Princeton University, Princeton, New Jersey
MARYLYNN V. YATES, University of California, Riverside
JAMES W. ZIGLAR, SR., Van Ness Feldman, Washington, D.C.
Staff
JEFFREY JACOBS, Director

LAURA J. EHLERS, Senior Program Officer
STEPHANIE E. JOHNSON, Senior Program Officer
M. JEANNE AQUILINO, Financial and Administrative Associate
MICHAEL J. STOEVER, Research Associate
ANITA A. HALL, Senior Program Associate
SARAH E. BRENNAN, Senior Program Assistant
BRENDAM R. McGOVERN, Senior Program Assistant
279
280 Appendix C
BOARD ON ENVIRONMENTAL STUDIES AND TOXICOLOGY
ROGENE F. HENDERSON, Chair, Lovelace Respiratory Research Institute,

Albuquerque, New Mexico
PRAVEEN AMAR, Private Consultant, Boston, Massachusetts
RICHARD A. BECKER, American Chemistry Council, Washington, D.C.
MICHAEL J. BRADLEY, M.J. Bradley & Associates, Concord, Massachusetts
JONATHAN Z. CANNON, University of Virginia, Charlottesville
GAIL CHARNLEY-ELLIOT, HealthRisk Strategies, Washington, D.C.
DOMINIC M. DI TORO, University of Delaware, Newark, Delaware
DAVID C. DORMAN, North Carolina State University, Raleigh, North Carolina
CHARLES T. DRISCOLL, JR., Syracuse University, New York
WILLIAM H. FAIRLAND, Colorado State University, Fort Collins, Colorado
LYNN R. GOLDMAN, George Washington University, Washington, D.C.
LINDA E. GREER, Natural Resources Defense Council, Washington, D.C.
WILLIAM E. HALPERIN, University of Medicine and Dentistry of New Jersey,
Newark
STEVEN P. HAMBURG, Environmental Defense Fund, New York
ROBERT A. HIATT, University of California, San Francisco
PHILIP K. HOPKE, Clarkson University, Potsdam, New York
SAMUEL KACEW, University of Ottawa, Ontario
H. SCOTT MATTHEWS, Carnegie Mellon University, Pittsburgh, Pennsylvania
THOMAS E. McKONE, University of California, Berkeley
TERRY L. MEDLEY, E.I. du Pont de Nemours & Company, Wilmington, Delaware
JANA MILFORD, University of Colorado, Boulder
MARK A. RATNER, Northwestern University, Evanston, Illinois
JOAN B. ROSE, Michigan State University, East Lansing, Michigan
GINA M. SOLOMON, California Environmental Protection Agency,
Sacramento, California
PETER S. THORNE, University of Iowa, Iowa City, Iowa
JOYCE S. TSUJI, Exponent Environmental Group, Bellevue, Washington
Senior Staff
JAMES J. REISA, Director

DAVID J. POLICANSKY, Scholar
RAYMOND A. WASSEL, Senior Program Officer for Environmental Studies
SUSAN N. J. MARTEL, Senior Program Officer for Toxicology
ELLEN K. MANTUS, Senior Program Officer for Risk Analysis
MISADA KARALIC-LONCAREVIC, Manager, Technical Information Center
RADIAH ROSE, Manager, Editorial Projects

Appendix D
Biographical Sketches of
Committee Members and Staff
Jeff Walters, Chair, is the Harold Bailey Professor of Biology at Virginia Tech, a
position he has held since 1994. His professional experience includes assistant,
associate, and full professorships at North Carolina State University from 1980
until 1994. Dr. Walters has done extensive research and published many articles
on the red-cockaded woodpeckers in North Carolina and Florida, and he chaired
an American Ornithologists’ Union Conservation Committee Review that looked
at the biology, status, and management of the Cape Sable seaside sparrow, a bird
endemic to the Everglades. His research interests are in the behavioral ecology,
population biology, and conservation of birds, and his recent work has focused
on cooperative breeding, dispersal behavior, and endangered species issues.
Dr. Walters served in two panels of the Sustainable Ecosystems Institute that
addressed issues with endangered birds in the Everglades restoration in addi-
tion to previously serving as a member of the NRC’s Committee on Restoration
of the Greater Everglades Ecosystem and the first and fourth Committees on
Independent Scientific Review of Everglades Restoration Progress. He holds a
B.A. from West Virginia University and a Ph.D. from the University of Chicago.
Mary Jane Angelo is professor of law at the University of Florida’s Levin C ollege
of Law and Director of the Environmental and Land Use Law Program. Her
research areas focus on environmental law, water law, administrative law, bio-
technology law, dispute resolution, pesticides law, law and science, and legal
ethics. Prior to joining the faculty, Ms. Angelo served as an attorney in the U.S.
Environmental Protection Agency’s Office of General Counsel and as senior
assistant general counsel for the St. Johns River Water Management District. She
received her B.S. in biological sciences from Rutgers University and her M.S.
and J.D. from the University of Florida.
David B. Ashley is professor of civil engineering at the University of Nevada,

Las Vegas (UNLV). Dr. Ashley also served as the eighth president at the school
281
282 Appendix D
from 2006 to 2009. Prior to joining UNLV, President Ashley served as executive
vice chancellor and provost at the University of California, Merced, and held the
Shaffer-George Chair in Engineering. He has also served as dean of engineering
at The Ohio State University and has held civil engineering faculty positions at
the University of California, Berkeley, the University of Texas at Austin, and the
Massachusetts Institute of Technology. Dr. Ashley’s principal research and teach-
ing activities are in the area of construction project planning, focusing primarily
on risk analysis and management of large-scale, complex projects. His recent
studies have addressed innovative project financing and new project procure-
ment approaches. He has served on several NRC committees, including the
Committee on Assessing the Results of External Independent Reviews for U.S.
Department of Energy Projects. Dr. Ashley received a B.S. in civil engineering
and an M.S. in civil engineering–project management from the Massachusetts
Institute of Technology, an M.S. in engineering–economic systems, and a Ph.D.
in civil engineering–constructing, engineering, and management from Stanford
University.
Loretta L. Battaglia is an associate professor of plant biology at Southern Illinois

University, Carbondale. Her research interests focus on the dynamics of wet-
land plant communities and the ecological processes that link them with the
surrounding landscape. Specifically, her research is focused on the effects of
climate change and large-scale phenomena, such as exotic species invasions, on
community structure and function, as well as development of restoration targets
for coastal wetlands undergoing rapid climate change. She received her B.S. in
zoology and her M.S. in biological sciences from the University of Louisiana,
Monroe, and her Ph.D. in ecology from the University of Georgia.
William G. Boggess is professor and executive associate dean of the College of

Agricultural Sciences at Oregon State University (OSU). Prior to joining OSU,
Dr. Boggess spent 16 years on the faculty at the University of Florida in the Food
and Resource Economics Department. His research interests include interactions
between agriculture and the environment (e.g., water allocation, groundwater
contamination, surface-water pollution, sustainable systems); economic dimen-
sions and indicators of ecosystem health; and applications of real options to
environmental and natural resources. Dr. Boggess previously served on the
Oregon Governor’s Council of Economic Advisors and the Board of Directors
of the American Agricultural Economics Association, and he currently serves
on the Board of the Oregon Environmental Council. He served on the State of
Oregon Environment Report Science Panel and has been active in the design
and assessment of the Oregon Conservation Reserve Enhancement Program.
Dr. Boggess served as a member of the National Research Council Commit-

Appendix D 283
tee on the Use of Treated Municipal Wastewater Effluents and Sludge in the
Production of Crops for Human Consumption, and on the second, third, and
fourth Committees on Independent Scientific Review of Everglades Restoration
Progress serving as chair of the fourth committee. He received his Ph.D. from
Iowa State University in 1979.
Charles T. Driscoll (NAE) is university professor in the Department of Civil and

Environmental Engineering at Syracuse University where he also serves as the
director of the Center for Environmental Systems Engineering. His teaching and
research interests are in the area of environmental chemistry, biogeochemistry,
and environmental quality modeling. A principal research focus has been the
response of forest, aquatic, and coastal ecosystems to disturbance, including air
pollution, land-use change, climate change, and elevated inputs of nutrients and
mercury. Dr. Driscoll is currently a principal investigator of the National Science
Foundation’s Long Term Ecological Research Network’s project at the Hubbard
Brook Experimental Forest in New Hampshire. He is a member of the National
Academy of Engineering and was a member of the National Research Council’s
(NRC’s) Panel on Process of Lake Acidification, the Committees on Air Quality
Management in the U.S. and the Collaborative Large-Scale Engineering Analysis
Network for Environmental Research (CLEANER), and the second, third, and
fourth Committees on Independent Scientific Review of Everglades Restoration
Progress. He is a member of the NRC Board on Environmental Studies and Toxi-
cology. Dr. Driscoll received his B.S. in civil engineering from the University
of Maine and his M.S. and Ph.D. in environmental engineering from Cornell
University.
Paul H. Glaser is a research professor in the Department of Earth Sciences at the

University of Minnesota (Twin Cities campus) with appointments to the Graduate
Faculty in Earth Sciences and Conservation Biology. He is a fellow of the Geo-
logical Society of America and a member-at-large of the Geology and Geography
Section of the American Association for the Advancement of Science. His current
research interests are focused on wetland-groundwater interactions in peatlands
with special reference to carbon cycling and greenhouse gases. However, his
research interests are cross-disciplinary, spanning the fields of wetland ecology,
hydrology, biogeochemistry, and paleoecology. Dr. Glaser earned his Ph.D. from
the University of Minnesota in 1978.
William L. Graf is Foundation University Distinguished Professor, Emeritus, at

the University of South Carolina. His expertise is in fluvial geomorphology and
hydrology, as well as policy for public land and water. Dr. Graf’s research and
teaching have focused on river-channel change, human impacts on river pro-

284 Appendix D
cesses, morphology, and ecology, along with contaminant transport and storage
in river systems. His present work emphasizes the downstream effects of dams on
rivers. In the arena of public policy, he has emphasized the interaction of science
and decision making, and the resolution of conflicts among economic develop-
ment, historical preservation, and environmental restoration for rivers. Dr. Graf
has served as a member of the National Research Council’s (NRC’s) Water
Science and Technology Board and Board on Earth Sciences and Resources,
the Panel to Review the Critical Ecosystem Studies Initiative, the Committee on
Restoration of the Greater Everglades Ecosystem, and the first three Committees
on Independent Scientific Review of Everglades Restoration Progress, serving as
chair of the second committee. He is chair of the NRC’s Geographical Sciences
Committee. He is also a national associate of the National Academies and an
American Association for the Advancement of Science fellow. Dr. Graf earned
a certificate of water resources management and his Ph.D. from the University
of Wisconsin, Madison.
Stephen G. Monismith is chair of the Civil and Environmental Engineering

Department and Obayashi Professor in the School of Engineering at Stanford
University. His research in environmental and geophysical fluid dynamics is
focused on the application of fluid mechanics principles to the analysis of flow
processes operating in rivers, lakes, estuaries, and the oceans. Flows that involve
physical-biological interactions are of particular interest to him. Dr. Monismith
has previous National Research Council experience, having served on the Panel
to Review California’s Draft Bay Delta Conservation Plan and the Committee on
Sustainable Water and Environmental Management in the California Bay-Delta.
He earned his B.A., M.S., and Ph.D. degrees from the University of California,
Berkeley.
David H. Moreau is research professor, Department of City and Regional Plan-

ning, at the University of North Carolina at Chapel Hill. He recently completed
a term as chair of the Curriculum for the Environment and Ecology. His research
interests include analysis, planning, financing, and evaluation of water resource,
water quality, and related environmental programs. Dr. Moreau is engaged in
water resources planning at the local, state, and national levels. He has served
on several National Research Council committees, including the Committee on
New Orleans Regional Hurricane Protection Projects Review, the Committee
on the Mississippi River and Hypoxia in the Gulf of Mexico, and the second,
third, and fourth Committees on Independent Scientific Review of Everglades
Restoration Progress. Dr. Moreau recently completed 19 years as a member
and 16 years as chairman of the North Carolina Environmental Management
Commission, the state’s regulatory commission for water quality, air quality, and

Appendix D 285
water allocation. For his service to North Carolina he was awarded the Order of
the Long Leaf Pine, the highest civilian award offered by the state. He received
his B.S. and M.S. from Mississippi State University and North Carolina State
University, respectively, and his Ph.D. degree from Harvard University.
K. Ramesh Reddy is graduate research professor and chair of the Department of

Soil and Water Science at the University of Florida. His research areas include
biogeochemistry, soil and water quality, ecological indicators, and restoration
of wetlands and aquatic systems. Dr. Reddy investigates biogeochemical cycling
of macronutrients in natural ecosystems, including wetlands, shallow lakes,
estuaries, and constructed wetlands, as related to soil and water quality, carbon
sequestration, and greenhouse gas emissions. He served as a member of the
U.S. National Committee for Soil Sciences in the National Academy’s Policy
and Global Affairs Division. He served on the U.S. Environmental Protection
Agency’s Science Advisory Board Panel. Dr. Reddy served as a member of the
second and third Committees on Independent Scientific Review of Everglades
Restoration Progress. Dr. Reddy earned his Ph.D. in agronomy and soil science
from Louisiana State University.
Helen Regan is an associate professor of biology at the University of California,

Riverside. Her research areas span quantitative conservation ecology and proba-
bilistic risk assessment. Dr. Regan has applied population models, uncertainty
analyses, and decision-making techniques to address a variety of conservation
and wildlife management issues. She focuses on methodological issues of these
techniques, the practicalities of their application and their interpretation for
management. Projects include ecological risk assessment of chemical contami-
nants, population viability of species impacted by a range threats, monitoring of
multiple species habitat conservation plans, population-level effects of habitat
fragmentation, and fire and disease on plants in fire-prone ecosystems. Current
research includes examination of the impact of uncertainty on potential adapta-
tion strategies for threatened species impacted by climate change. She currently
serves on the Standards and Petitions Subcommittee of the International Union
for the Conservation of Nature Species Survival Commission and on the scientific
advisory committee for the Australian Centre of Excellence for Risk Analysis.
Dr. Regan received her B.S. from LaTrobe University and her Ph.D. from the
University of New England in Armidale, both in Australia.
James E. Saiers is professor of hydrology, associate dean of Academic Affairs,

and professor of chemical engineering at the Yale School of Forestry and Envi-
ronmental Studies. Dr. Saiers studies the circulation of water and the movement
of waterborne chemicals in surface and subsurface environments. One element

286 Appendix D
of his research centers on quantifying the effects that interactions between

hydrologic and geochemical processes have on the migration of contaminants
in groundwater. Another focus is on the dynamics of surface-water and ground-
water flow in wetlands and the response of fluid flow characteristics to changes
in climate and water management practices. His work couples field observations
and laboratory-scale experimentation with mathematical modeling. He earned
his B.S. in geology from the Indiana University of Pennsylvania and his M.S.
and Ph.D. degrees in environmental sciences from the University of Virginia.
Daniel Simberloff (NAS) is the Nancy Gore Hunger Professor of Environmental

Science in the Department of Ecology and Evolutionary Biology at the University
of Tennessee, Knoxville. His research centers on ecology, statistical ecology, bio-
geography, evolution, and conservation biology, and addresses plants, insects,
birds, and mammals. Specifically, his research focuses on invasion biology,
community composition and structure, and community morphological structure.
He maintains an extensive world-wide field research program focused on issues
of biological invasions and global change and is a leading innovator in the
application of statistical methods to large ecological data sets. Dr. Simberloff is
a member of the National Academy of Sciences and is the recipient of numerous
awards, including the Ecological Society of America’s Eminent Ecologist Award
and the Ramon Margalef Award for Ecology. He has served on multiple National
Research Council (NRC) committees and was a member of the NRC Board on
Life Sciences. He received his A.B. and Ph.D. from Harvard University.
STAFF
Stephanie E. Johnson, study director, is a senior program officer with the Water
Science and Technology Board. Since joining the National Research Council
in 2002, she has worked on a wide range of water-related studies, on topics
such as desalination, wastewater reuse, contaminant source remediation, coal
and uranium mining, coastal risk reduction, and ecosystem restoration. She has
served as study director for 15 committees, including the Panel to Review the
Critical Ecosystem Studies Initiative and all five Committees on Independent
Scientific Review of Everglades Restoration Progress. Dr. Johnson received her
B.A. from Vanderbilt University in chemistry and geology, and her M.S. and
Ph.D. in environmental sciences from the University of Virginia.
David J. Policansky is a scholar of the Board on Environmental Studies and

Toxicology. He earned a Ph.D. in biology from the University of Oregon.
Dr. Policansky has directed approximately 35 National Research Council studies,

Appendix D 287
and his areas of expertise include genetics; evolution; ecology, including fishery
biology; natural resource management; and the use of science in policy making.
Michael J. Stoever is a research associate with the Water Science and Technology
Board. He has worked on a number of studies including Desalination: A National
Perspective, the Water Implications of Biofuels Production in the United States,
and the Committee on Louisiana Coastal Protection and Restoration. He has also
worked on National Research Council studies on the National Flood Insurance
Program, the effect of water withdrawals on the St. Johns River, and Chesapeake
Bay restoration. Mr. Stoever received his B.A. in political science from The
Richard Stockton College of New Jersey in Pomona.
Sarah E. Brennan is a senior program assistant with the Water Science and Tech-
nology Board (WSTB). Since joining the NRC in 2010, she has worked on six
projects including Everglades restoration progress, U.S. Army Corps of Engineers’
water resources, and water and environmental management in the California
bay delta. Before joining WSTB, Ms. Brennan was a Peace Corps Volunteer in
Ghana, West Africa. She Received her B.S. in international development from
Susquehanna University.


APPENDIX S:
INDIAN RIVER LAGOON ECONOMIC VALUATION
UPDATE
INDIAN
RIVER LAGOON
ECONOMIC VALUATION UPDATE
2016
Indian River Lagoon
Economic Valuation Update
Prepared by
East Central Florida Regional Planning Council
Treasure Coast Regional Planning Council
Supported by
Florida Department of Economic Opportunity
Community Planning Technical Assistance Grant P0169
August 26, 2016

East Central Florida Regional Planning Council Florida Department of Economic Opportunity
Treasure Coast Regional Planning Council Community Planning Technical Assistance Grant P0169
Contents
Page
Preface..........................................................................................................................................ii
Foreword ......................................................................................................................................iii
Acknowledgments........................................................................................................................iv
Acronyms .....................................................................................................................................v
Executive Summary .....................................................................................................................vi
Introduction ..................................................................................................................................1
Methodology ................................................................................................................................6
Economic Contribution of Living Resources ..............................................................................10
Economic Contribution of Marine Industries ..............................................................................11
Economic Contribution of Recreation & Visitor-related .............................................................13
Economic Contribution of Resource Management ......................................................................14
Economic Contribution of Defense & Aerospace .......................................................................15
Economic Contribution of Industry Groups ................................................................................17
Total Economic Contribution of Industry Groups .......................................................................19
Total Economic Contribution of Volusia County ........................................................................21
Total Economic Contribution of Brevard County .......................................................................23
Total Economic Contribution of Indian River County ................................................................25
Total Economic Contribution of St. Lucie County ......................................................................27
Total Economic Contribution of Martin County .........................................................................29
Total Economic Contribution of Indian River Lagoon Counties.................................................31
Employment in Indian River Lagoon Counties ...........................................................................33
County Contribution to Employment ..........................................................................................35
Industry Group Contribution to Employment ..............................................................................37
Total Economic Contribution by Industry Group ........................................................................39
Total Economic Contribution by County.....................................................................................41
Cost of a Sustainable Indian River Lagoon Economy .................................................................43
Return on Investment from a Sustainable Indian River Lagoon ..................................................46
An Indian River Lagoon-Dependent Industry in Decline ............................................................47
Boat Registrations as an Indicator of the Indian River Lagoon’s Environmental
and Economic Health ............................................................................................................48
Missing and Incomplete Data ......................................................................................................49
References and Data Sources .......................................................................................................52
Appendices (provided in a separate document)
1. Final Data & Analysis
2. Summaries of Teleconferences
Indian River Lagoon Economic Valuation Update i Final Report 08-26-2016

Preface
Florida’s regional planning councils (RPCs) are dedicated to increasing awareness and education
of their member local governments statewide to foster better decision-making and bring together
necessary public and private sector constituencies to solve regional problems. Their members
include a wide range of local elected officials, governor appointees from the private business
sector and academia, and ex-officios from economic development, transportation, environmental
and water management organizations.
The East Central Florida Regional Planning Council (ECFRPC) was formed by its local
governments in 1962. The Treasure Coast Regional Planning Council (TCRPC) was established
in the same manner in 1976. Together the two regions include 11 counties and 126
municipalities, occupying a territory of over 9,729 square miles with a combined population of
5,467,614. These regions also include the six counties that contain the 156-mile long Indian
River Lagoon – a national ecologic and economic treasure.
These regional planning councils have been long-focused on regional ecosystem restoration
efforts and serve as clearinghouses for information, education, legislation, and policy related to
critical coastal and environmental resource issues facing the region. The RPCs stand with their
congressional and state legislative delegations, responsible governmental jurisdictions, and the
dozens of local, regional, and statewide grass-roots organizations focused on restoration of the
Everglades, Indian River and Lake Worth lagoons and St. Lucie and Caloosahatchee river
ecosystems.
In 2015, the Florida Department of Economic Opportunity (FDEO) funded the ECFRPC to
produce an updated economic contributions analysis for the Indian River Lagoon (IRL).
ECFRPC worked closely with TCRPC to accomplish the update. The effort represents a
tremendous opportunity for the region to quantify, not only the annualized economic
contribution of this nationally significant estuary, but more importantly, what is at stake
economically if the ecosystem collapses. The authors and contributors of the study recognize the
vital role of a healthy IRL, not just to the region’s economic vitality, but as the foundation of its
extraordinary quality of life. Copies of this report are available for viewing and download on the
RPC’s websites: www.ecfrpc.org and www.tcrpc.org.
Indian River Lagoon Economic Valuation Update ii Final Report 08-26-2016

Foreword
In May of 2015, the ECFRPC and TCRPC entered into discussions with FDEO and subject
matter experts regarding a methodology to update and refine economic contributions data and
analyses related to the IRL. Previous studies were evaluated, which included the 1996 IRL
National Estuary Program’s Economic Assessment and Analysis, Hazen and Sawyer’s 2008
update of the Indian River Lagoon Economic Assessment and Analysis, and Florida Ocean
Alliance’s 2013 Florida’s Oceans and Coasts Report. Additional discussions to further refine the
scope were held among the group, including staff members from the IRL National Estuary
Program, Florida Inland Navigation District, Water Management District, Tampa Bay Regional
Planning Council, and Florida Fish and Wildlife Conservation Commission, along with private
sector economic consultants. The budget was limited and the goal was to find a relatively
inexpensive, reliable, and repeatable method for valuing the regional economic contributions of
the IRL.
Ultimately, the group decided on a methodology similar to what was used to produce the
statewide economic contributions analysis published in the 2013 Florida’s Oceans and Coasts
report by the Florida Ocean Alliance. This report used a reliable and repeatable set of
employment and market data for quantifying the value of economic activity attributable to
productive estuarine and nearshore coastal systems in Florida. Unlike the Florida’s Oceans and
Coasts report, the IRL Economic Valuation Update focuses only on the IRL’s economic
contributions to the IRL region, which includes all of coastal Brevard, Indian River, St. Lucie,
and Martin counties and portions of Volusia and Palm Beach counties.
The team of twelve subject matter experts from various environmental and economic disciplines
were also assembled to review the drafts of the report (Appendix 2). This report is the result of
that collaborative effort.
The intent of this report is to establish a repeatable and cost-efficient method for estimating the
direct, indirect, and induced economic values attributable to the IRL. Quantifying these values
and the IRL’s total annual economic contributions, serves to emphasize the region’s economic
interdependence with the IRL and the importance of restoration efforts to sustain its health and
productivity.
Indian River Lagoon Economic Valuation Update iii Final Report 08-26-2016
Acknowledgments
The ECFRPC and TCRPC gratefully acknowledge the support of FDEO for funding this
important effort documenting the value of annual economic contributions of the IRL. Special
thanks to Beth Frost, FDEO contract administrator, and her reminders about deadlines. Also,
special thanks to Barbara Lenczewski, James Stansbury and George Foster for their expertise,
constant words of encouragement, and their assistance in obtaining the release of proprietary
North American Industry Classification System (NAICS) code data. A personal thank you to
Kathie Hughes, Program Manager for the Bureau of Labor Statistics at FDEO, for preventing the
authors from breaking state laws related to the public release of confidential NAICS code data.
Special thanks to Tara McCue of ECFRPC, contract manager, for her timely administration of all
the contract modifications necessary to complete the project on time and within budget. The
RPCs would like to extend special appreciation to the subject matter experts for volunteering
their time and for their constructive suggestions to improve the scope, methodology, and final
report. They include:
Duane De Freese IRL Council

Tara McCue East Central Florida Regional Planning Council
Hugh Harling East Central Florida Regional Planning Council
Luis Nieves-Ruiz East Central Florida Regional Planning Council
Barbara Lenczewski Florida Department of Economic Opportunity
George Foster Florida Department of Economic Opportunity
James Stansbury Florida Department of Economic Opportunity
Jerri Lindsey Florida Fish & Wildlife Conservation Commission
Julian Hwang Florida Fish & Wildlife Research Institute
W. Thomas Lavash WTL+a
Tom Moriarity Retail Development Strategies, LLC
Kathy LaMartina South Florida Water Management District
A special thanks to our economic consulting team of Tom Lavash, Tom Moriarity, and James
Stover of the firm WTL+a for their assistance in refining the scope, their professionalism, and
their personal commitment to a successful project. Finally, thank you to the authors of the
Florida’s Oceans and Coasts report whose finished work inspired the methodology for this
report. The project managers and primary authors of this report are Michael J. Busha and Peter
G. Merritt of TCRPC.
Indian River Lagoon Economic Valuation Update iv Final Report 08-26-2016

Acronyms
BEA Bureau of Economic Analysis
BLS Bureau of Labor Statistics
BMAP Basin Management Action Plan
ECFRPC East Central Florida Regional Planning Council
FDEO Florida Department of Economic Opportunity
FFWCC Florida Fish & Wildlife Conservation Commission
IMPLAN IMpact Analysis for PLANing Model
IRL Indian River Lagoon
LMS Bureau of Labor Market Statistics
NAICS North American Industry Classification System
NOAA National Oceanographic and Atmospheric Administration
QCEW Quarterly Census of Employment and Wages Program
RIMS II Regional Input/Output Modeling System
ROI Return on Investment
RPC Regional Planning Council
TCRPC Treasure Coast Regional Planning Council
TDT Tourism Development Tax
TMDL Total Maximum Daily Load
VFR Visiting Friend and Relatives
Indian River Lagoon Economic Valuation Update v Final Report 08-26-2016

Executive Summary
This report illustrates the link between the IRL region’s economy and the IRL. Completing this
economic analysis expands and updates various sections of three earlier economic studies:
Economic Assessment and Analysis of the Indian River Lagoon, January 1996; Indian River
Lagoon Economic Assessment and Analysis Update, August 2008; and Florida’s Oceans and
Coasts: An Economic and Cluster Analysis, May 2013. Other key objectives of the update are to:
1) estimate Total Annual Economic Output attributable to the five main IRL-related coastal
industry groups (Living Resources; Marine Industry; Recreation and Visitor-related; Resource
Management; and Defense & Aerospace); and 2) establish a repeatable and valid process for
analyzing and reporting the economic value of these five main coastal industry groups for the
IRL region.
This economic analysis was accomplished primarily by using the latest 2014, six-digit NAICS
code data and measuring economic impacts through the IMPLAN Regional Economic
Input/Output Model. The IMPLAN model estimates direct impacts (employment and spending),
indirect impacts (secondary effects of spending through incomes generated by relevant
industries), and induced impacts (indirect spending by employees). For example, a boat dealer
would be compensated (direct impact), might own a house (indirect impact) and may spend
money dining out, which would generate economic activity through income for cooks and
servers in restaurants (induced impacts). IMPLAN incorporates a series of algorithms to
calculate direct, indirect, and induced impacts that ripple through the economy.
Study Area
The study area for this report extends from the Ponce de Leon Inlet in Volusia County to the
Jupiter Inlet in northern Palm Beach County, completely traversing Brevard, Indian River, St.
Lucie, and Martin counties. Economic data sets used in this analysis represent the year 2014.
These include countywide information on IRL-related economic drivers for Brevard, Indian
River, St. Lucie and Martin counties. Even though the Ponce de Leon Inlet divides the north-
south distance of Volusia County in approximately half, only about 20 percent of Volusia
County’s coastal population is adjacent to the IRL system. Approximately 80 percent of Volusia
County’s coastal population is adjacent to the Halifax River Lagoon, which extends north from
the Ponce de Leon Inlet. Therefore, this report relies on the assumption that only 20 percent of
the total economic output of Volusia County are attributable to the IRL. Also, the IRL’s
economic contribution to Palm Beach County was not included in this report. This was because
broad countywide data could not be disaggregated, and only a small fraction of countywide
economic output totals are attributed this portion of the study area.
Just How Valuable is the Indian River Lagoon?
Overall the total annual economic output or value received from the IRL in 2014 is about $7.6
billion. This does not include the estimated $934 million in annualized real estate value added
for property located on or near the IRL (Hazen and Sawyer 2008). Also not included are the
economic contributions from estuarine-related resources in Volusia County north of the Ponce de
Indian River Lagoon Economic Valuation Update vi Final Report 08-26-2016

Leon Inlet. If both of these economic contributions are considered, the total annual economic
output is valued at about $9.9 billion.
Total Annual Economic Output by County

With 20% Volusia County
$609,128,748 $330,353,935
4.3%
$549,660,385 8.0%
7.2%
$797,562,586
10.4%
$5,353,605,910
Total $7,640,311,564 70.1%
Volusia County Brevard County Indian River County St. Lucie County Martin County
Total Annual Economic Output by County

$609,128,748
$549,660,385 6.8% $1,651,769,673
6.1% 18.4%
$797,562,586
8.9%
Total $8,961,727,302
$5,353,605,910
59.7%
Volusia County Brevard County Indian River County St. Lucie County Martin County
Indian River Lagoon Economic Valuation Update vii Final Report 08-26-2016
Total Annual Economic Output by Industry Group

$48,230,566 $767,440,497
0.6% 10.0%
$1,574,078,178
20.6%
$101,410,945
$5,149,151,378 1.3%
67.4%
Total $7,640,311,564
Living Resources Marine Industries Recreation and Visitor-related

Resource Management Defense & Aerospace
Total Annual Economic Output by Industry Group

$54,659,107 $1,049,735,424
0.6% 11.7%
$2,388,370,522
26.7%
$5,361,562,245
59.8%
$107,400,004
Total $8,961,727,302 1.2%

Indian River Lagoon Economic Valuation Update viii Final Report 08-26-2016
The five main IRL-related industry groups are categorized as Living Resources, Marine
Industries, Recreation and Visitor-related, Resource Management and Defense & Aerospace.
These industry groups, which are directly influenced by the IRL, support nearly 72,000 jobs
annually with annual wages of more than $1.2 billion. Total annual income (labor and
proprietor’s) for IRL-related industry groups is around $2.6 billion.
Recreation and visitor-related activity related to the IRL contributes around $1.57 billion
annually to the IRL region’s economy. In 2014, the IRL region attracted over 7.4 million
visitors. Previous studies have estimated that annually, between 31 to 46 percent or 2.3 - 3.5
million of the region’s visitors participate in IRL-related recreation. By 2025, the IRL region is
expected to receive over 11 million visitors annually. All things being equal, this would equate
to between 3.4 - 5.1 million visitors annually
looking to participate in IRL-related recreation.
Today, each visitor to the IRL region spends
about $162 a day. For every 85 visitors to the
IRL-region, one job can be expected to be
created.
Indian River Lagoon Economic Valuation Update ix Final Report 08-26-2016

Cost of a Sustainable Indian River Lagoon-Based Economy and the Region’s Return on
Investment
Enhancing water quality, increasing habitat, providing additional points of access, and
implementing other best management practices to improve the IRL can generate thousands of
new jobs and increase the value and output of the IRL-based economy. This presumes state and
local plans to improve the IRL’s health and productivity are implemented and continue to be
successful in the future. Implementing these plans will require a long-term commitment of time
and funding.
One way to estimate how much it will cost annually and cumulatively over time to achieve a
sustainable IRL, is to estimate the cost of implementing the adopted Basin Management Action
Plans (BMAPs) for the IRL region. There are four BMAPs covering the IRL region. Each
contains nutrient load reduction goals intended to establish a sustainable IRL in the future.
TCRPC and ECFRPC (2015) estimated it would cost

$4.6 billion to accomplish the required nutrient load
reductions in all four of the BMAPs. By this measure,
and with efforts extended over a 20 year period, it
would require an annual investment of $230 million
to sustain an IRL-based economy. When comparing
this average annual cost to the IRL’s total average
annual economic output of $7.6 billion, the Return on
Investment (ROI) from a sustainable IRL is 33 to 1.
Or, for every dollar spent on achieving a sustainable
IRL, the lagoon returns $33 in total economic value.
Over the years, as the IRL improves in health and
productivity, annual ROI in the future can be expected
to increase.
Based on the estimated 2015 population of the IRL counties, the annual cost per person to
accomplish the nutrient load reduction goals in the four IRL-related BMAPs equates to about
$140. If this is divided equally among IRL-county households, each would pay around $334 a
year, or about $28 a month. The 1996 Economic Assessment and Analysis of the IRL found that
1000 IRL households polled said they would be willing to pay an additional $60 a year to
support IRL improvement programs. The 2008 update of this study found that 50 percent of the
985 IRL households polled would be willing to pay a one-time tax of $20 to improve the lagoon,
with about 27 percent of the respondents willing to pay a one-time tax of $400.
The cost calculations outlined above assume that IRL residents will bear the full cost of BMAP
implementation. This is not likely to be the case. Floridians outside the IRL region will
contribute to state water quality improvement programs and water management district projects
through state sales taxes and federal income taxes redistributed to the IRL region in the form of
state and federal assistance. American and foreign tourists will also offset some local
implementation costs through sales and other special taxes levied on local goods and services
Indian River Lagoon Economic Valuation Update x Final Report 08-26-2016

that may flow to state water quality improvement programs. In addition, use of Land Acquisition
Trust Fund monies may also offset the cost of BMAP-related water quality improvement projects
for the lagoon.
An Indian River Lagoon-Dependent Industry in Decline
Commercially harvested clams, oysters, crabs and shrimp were worth $12.6 million at the docks
in 1994. Adjusted for inflation, this 1994 amount is $20.1 million in 2015 dollars. The overall
value of the commercial clam, oyster, crab, and shrimp harvest for 2015 is $4.3 million – a
decline of nearly 80 percent. Pounds of shellfish harvested have also declined during about that
same period, from 7.1 million pounds to 2 million pounds, or almost 72 percent. IRL counties
showing the sharpest decline in value and pounds harvested were Volusia, Brevard and Martin.
The commercial fin fishery fared a little better, but still showed significant declines in value and
pounds landed. In 1990, the estimated value of commercial fin fish landed was worth $13
million. Adjusted for inflation, this 1990 amount is $23.5 million in 2015 dollars. The overall
value of the commercial fin fish harvest for 2015 is estimated at $14.8 million – a decline of 37
percent. Pounds of fin fish landed also declined during that same period, from 17.3 million
pounds to 8 million pounds or almost 54 percent. IRL counties showing the sharpest declines in
value and pounds harvested are Brevard, Indian River, and Martin counties.
Indian River Lagoon Economic Valuation Update xi Final Report 08-26-2016

Indian River Lagoon Region Boat Registrations Declining
One indicator of the environmental health and economic productivity of the IRL is the increase
or decrease in boat registrations over time. Increased access and use of the lagoon by boats for
fishing and other outdoor lagoon-based activities has an economic impact on retail, lodging,
marine industry and food and beverage sectors, and also on sales tax, fuel tax, boat registration
fees and fishing license revenues.
A declining trend in boat registrations for the IRL region has been observed in the last decade.
While the IRL region’s population has increased by 12 percent, boat registration has decreased
by 13,045 or by 11 percent since 2005. Much of the decline has been for boats in the 12-39-foot
range with boat registrations in the 40 foot and larger class remaining relatively flat. The sharpest
declines in boat registrations have occurred in Brevard, Volusia and St. Lucie counties.
IRL REGION BOATER

REGISTRATIONS
2005 2015
Conclusion
This economic valuation update is not a status report on the IRL’s environmental health or how
to address human-related impacts affecting its long-term sustainability. The report simply serves
to measure the economic values the IRL region stands to lose if it does not eliminate regional
and local environmental inputs that are degrading the health and productivity of this estuary of
national significance.
The IRL and nearshore coastal resources are significant contributors to the region’s strong
economy and extraordinary quality of life. Sustaining the IRL is important to the region’s
economic recovery, expanding opportunities within the five main industry groups, recruiting new
industry sectors and talent, and sustainable growth. The health and sustainability of the IRL will
be a key determinant in the future of the region’s economy, public health and other natural,
cultural, and societal values that are collectively worth far more than what has been quantified in
this report.
Indian River Lagoon Economic Valuation Update xii Final Report 08-26-2016
IRL-RELATED INDUSTRY GROUPS
Indian River Lagoon Economic Valuation Update xiii Final Report 08-26-2016
Introduction
The IRL is a shallow, navigable estuary where saltwater from the Atlantic Ocean mixes with
freshwater from surrounding lands, canals, and natural tributaries. The IRL is contained within
six counties along Florida’s east coast between a chain of barrier islands and the mainland. It
extends over 156 miles from the Ponce de Leon Inlet in Volusia County, to the Jupiter Inlet in
Palm Beach County. This seagrass-based ecosystem supports an abundant collection of plants
and animals and is considered to be the most biologically diverse estuary in North America.
The importance of sustaining the biologic and economic productivity of the lagoon is nationally
recognized. In setting national priorities, which advocate balancing economic contributions of
the nation’s estuaries with the protection of these environments, Congress established the
National Estuaries Program in 1987 and designated the IRL an estuary of national significance in
1990. For the counties that make up the IRL region, the regional connection and balance between
a healthy lagoon and a strong economy is equally important.
Over 7.4 million tourists were attracted to visit the IRL region last year. This number is expected
to grow to 11 million by 2025. The IRL region’s resident population in 2015 is estimated to be
around 1.5 million. By 2025, that population is expected to grow to nearly 1.9 million. In 2015,
more than 1.1 million residents or just over 76 percent of the IRL region’s population lived
within a few miles of the lagoon. This estuarine environment not only fuels a coastal industry-
based economy worth an estimated $7.6 billion annually, but contributes directly to the region’s
appeal and extraordinary quality of life.
Purpose of the Study
The purpose of the study is to: 1) establish the economic value and contribution of the IRL to the
economies of the coastal counties that comprise the IRL region; 2) highlight the connection
between a healthy IRL ecosystem and the region’s economy; and 3) identify missing or
incomplete data sets that could be used to improve future economic valuation studies of the IRL
system.
The study achieves this purpose by providing a regional update of the statewide economic
contributions analysis published in the 2013 Florida’s Oceans and Coasts report by the Florida
Ocean Alliance. This report used a reliable and repeatable set of employment and market data
for quantifying the general value of economic activity statewide attributable to productive
estuarine and nearshore coastal systems in Florida. Unlike Florida’s Oceans and Coasts report,
this economic analysis focuses only on the IRL’s economic contributions to the IRL region,
which includes all of coastal Brevard, Indian River, St. Lucie, and Martin counties and portions
of Volusia and Palm Beach counties.
The Florida’s Oceans and Coasts report primarily used 2010 and 2011 data. The year 2014 was
identified as the “target year” for data and analysis for this update as some employment
information categories were not yet available for all of 2015. Year 2014 data provides the most
recent and accurate year-long information for the economic contributions analysis.
Indian River Lagoon Economic Valuation Update 1 Final Report 08-26-2016


Key objectives of the study are to: 1) estimate Total Annual Economic Output attributable to the
IRL by measuring the direct, indirect, and induced impacts through the IMPLAN Regional
Economic Input/Output Model; and 2) establish a repeatable and valid process for analyzing and
reporting economic value and contributions of the main traditional coastal industry groups for
the IRL using the best available data.
The IMPLAN model includes direct impacts (employment and spending), indirect impacts
(secondary effects of spending through incomes generated by relevant industries for housing,
etc.), and induced impacts (indirect spending by employees). For example, a boat dealer would
be compensated (direct impact), might own a house (indirect impact) and may spend money on
dining out, which would generate economic activity through income for cooks and servers in
restaurants (induced impacts). IMPLAN incorporate a series of algorithms to calculate direct,
indirect and induced impacts that ripple through the economy.
IMPLAN is a computer model that is the national standard for measuring a region’s or county’s
economic activity in terms of sales, income, employment, and tax revenue. In comparison, the
Florida’s Oceans and Coasts report used the US Bureau of Economic Analysis (BEA) Regional
Input/Output Modeling System (RIMS II) to measure broad statewide economic values. It was
determined that IMPLAN provided the most detailed and robust econometric model for the
regional study. As such, it may not be possible to make direct comparisons of the values and
results provided by category and industry group between these two studies.
Similar to the Florida’s Oceans and Coasts report, the focus of the economic contributions
analysis is based on main traditional coastal industry groups: 1) Living Resources; 2) Marine
Industries; 3) Recreation and Visitor-related; and 4) Resource Management. Relevant economic
data was collected and geo-coded for these industry sectors by the BEA which are organized
under the NAICS. This is the standard classification system used by business and government to
categorize economic activity. The 2014 six-digit NAICS code data were used in this effort to
further assure that economic contribution data is directly related to the IRL system.
In addition, economic data and analyses were provided for the industry group known as Defense
& Aerospace. This industry group was included as an economic contributor because the study
considered the economic contribution of this group, particularly around the Cape Canaveral/Cape
Kennedy Space Center, to be directly dependent on the geography, climate, and exclusive
waterfront locations provided by the IRL. The total annual regional economic contribution of this
group alone is estimated to be $5.1 billion.
One of the major economic values related to IRL, which is not analyzed in the Florida’s Oceans
and Coasts report or this study, is the value premium for waterfront real estate, both commercial
and residential, attributable to the IRL. The scope and budget for this economic impact study did
not allow for collection and analysis of the thousands of properties along the 156-mile length of
the lagoon.
Previous studies that examined real estate values and economic contributions attributable to real
property located on or near the IRL include the Apogee Research Inc. (1996) and Hazen and
Sawyer (2008). Hazen and Sawyer’s study estimated the total increase in property values

attributed to being on or near the IRL is about $47 billion. The study estimated the $47 billion
represented 22 percent of the 2006 market value of all real property in the five-county IRL
region. It further estimated the regional annualized contribution to real estate values attributable
to the IRL is about $934 million.
Further emphasizing the direct link between a clean and healthy IRL and a strong IRL regional
economy is the Florida Realtor’s 2015 study on the Impact of Water Quality on Florida Home
Values. It estimated for Martin County, that an increase in water clarity (i.e. a one-foot increase
in average annual Secchi disk depth) raises aggregate property values by 10.32 percent or $428
million. The study further suggests the inverse -- that a one-foot decline in average annual
Secchi disk depth results may be associated with similar size losses in aggregate property values.
The overall outputs of this economic valuation update study are as follows:
1. Economic Contribution by Industry

2. Economic Contribution by County
3. Employment in the Indian River Lagoon Regional Economy
4. Total Economic Contribution, Including Indirect and Induced Effects
In addition, this update examines the following elements of the IRL economy:
1. Cost of a Sustainable Indian River Lagoon Economy

2. Return on Investment from a Sustainable Indian River Lagoon
3. Commercial Shellfish and Finfish Industry Contribution and Trends
4. Boat Registration Trends
5. Missing or Incomplete Data Sets
Study Area
The geographic limits of the study area and the extent of data to be collected and used for the
economic valuation analysis extend from the Ponce de Leon Inlet in Volusia County to the
Jupiter Inlet in northern Palm Beach County, completely traversing Brevard, Indian River, St.
Lucie, and Martin counties. As a result, economic data sets used in this analysis included
countywide information on water-related economic drivers for Brevard, Indian River, St. Lucie,
and Martin counties. However, this report relies on different assumptions for Volusia and Palm
Beach counties. Even though the Ponce de Leon Inlet divides the north-south distance of Volusia
County in approximately half, analysis of the population of Volusia coastal municipalities
indicates only about 20 percent of this population is adjacent to the IRL system (Exhibit 1).
Eighty percent of the population of Volusia County coastal municipalities is adjacent to the
Halifax River Lagoon, which extends north from the Ponce de Leon Inlet. Therefore, this report
relies on the assumption that only 20 percent of the economic contribution of Volusia County is
attributed to the economy of the IRL. Also, the IRL’s economic contribution to Palm Beach
County was not included in this report. This was because broad countywide data could not be

A Secchi Disk measures visible depth into the water.
disaggregated, and only a small fraction of the countywide economic output totals are attributed
to this portion of the study area.
Exhibit 1
Population of Coastal Municipalities In Volusia County
Municpalities Associated with Indian River Lagoon Population Percent
Edgewater 20,958
New Smyrna Beach 24,285
Oak Hill 1,869
Subtotal 47,112 20%
Municpalities Associated with Halifax River Lagoon
Daytona Beach 63,534
Daytona Beach Shores 4,263
Flagler Beach (part) 60
Holly Hill 11,712
Ormond Beach 40,013
Ponce Inlet 3,047
Port Orange 58,656
South Daytona 12,538
Subtotal 193,823 80%
Total 240,935 100%
Source: Bureau of Economic and Buisiness Research, Univerity of Florida, Florida Estimates of Population 2015
One of the key objectives of the study was to establish a repeatable and valid process for
analyzing and reporting economic contributions of the main traditional coastal industries for the
IRL region. The structure of the report is intended to be standardized enough to be easily
replicable in the future. The concept was that these types of regional industry-based impact
analyses might be repeated regularly in order to monitor the positive and negative economic
impacts associated with different lagoon-related economic drivers, and subsequently compare
patterns, differences and causal effects of outside forces such as water quality, commercial and
recreational fishing productivity, boat sales and registration, real estate values, etc.

Methodology
This study methodology is intended to instruct an economic valuation and contributions analysis
for the IRL region. Using a reliable and repeatable set of employment and market data, the
methodology defines a process for quantifying the value of economic activity attributable to the
IRL which includes all of coastal Brevard, Indian River, St. Lucie, and Martin counties and a
portion of Volusia and Palm Beach counties.
Many of the findings and conclusions of the analysis rely on data that are only defined on a
countywide basis, and are not disaggregated for sub-areas of the counties. Because it was not
possible to disaggregate the Palm Beach County portion of the study area, and only a small
fraction of the countywide economic output totals are attributed to the IRL, Palm Beach County
information was not included. In addition, only a portion of Volusia County, south of the Ponce
de Leon Inlet, was included in the value and contributions analysis (Exhibit 1).
Key objectives of the methodology are to: 1) establish a repeatable and valid process for
analyzing and measuring the economic contributions of the main traditional coastal industry
groups for the IRL region using the best available data; and 2) estimate Total Annual Economic
Output for the IRL region by measuring the direct, indirect, and induced economic impacts
through the IMPLAN Regional Economic Input Output Model. IMPLAN stands for IMpact
Analysis for PLANing. It is a computer model which allows the user to measure a region’s or
county’s economic activity terms of sales, income, employment, and revenues. IMPLAN is the
national standard and most detailed econometric model for regional analyses of this type.
Five main industry groups were selected for economic analysis. These included: 1) Living
Resources; 2) Marine Industries; 3) Recreation and Visitor-related; 4) Resource Management;
and 5) Defense & Aerospace. Relevant economic data was collected and geo-coded for these
industry groups by the BEA, which are organized under the NAICS codes. This is the standard
classification system used by business and government to generate statistical data and categorize
the economic activity related to the U.S. business economy. A number of customary data sources
were consulted as part of the demographic and economic analysis for this project. However, only
NAICS data on 2014 employment in relevant industry sectors for each county were utilized in
the IMPLAN economic impact model. NAICS data were provided by the FDEO, Bureau of
Labor Market Statistics (LMS), Quarterly Census of Employment and Wages (QCEW) Program.
The NAICS code data provided by FDEO, LMS, QCEW included confidential information and
is not provided as part of this report. All other data and analysis is included in Appendix 1.
The methodology for developing current estimates of the Total Annual Economic Output for the
IRL system are described in the procedure outlined below:
1. Estimate number of covered wage and salary workers based on datasets from FDEO,
LMS, QCEW, which reports number of establishments (reporting units), total
employment and total (average annual) wages in the selected NAICS codes.
2. Convert estimates of average annual wages and salaries into estimates of employee
compensation by multiplying the ratio of employee compensation to wages and salaries

countywide based on BEA wage and salary data. This conversion accounts for
allowances related to employer contributions to social security, employee pension and
insurance (i.e., fringe benefits).
3. Convert estimates of employee compensation to Total (Labor) Earnings, for each IRL
county by selected industry (NAICS codes), using the BEA estimates of labor earnings
above.
4. Apply ratio of Total (Labor) Earnings to employee compensation in selected NAICS

codes, by industry for each county. This adjustment will reflect earnings of independent
contractors or self-employed to employee compensation in these industries.
5. Apply ratios of Total Annual Economic Output to Total Annual (Labor) Earnings for
each industry group to labor earnings in the broad coastal industries to develop estimates
of contributions to Total Annual Economic Output, by the IRLs five main traditional
coastal industry groups.
6. Estimate total contributions to Total Annual Economic Output of the IRLs five main
traditional coastal industry groups through the IMPLAN Regional Economic Input
Output Model using the following methodology:
o Use Final Demand multipliers for each industry

o Use Direct Effect multipliers to derive direct impact from employment estimates
o Calculate indirect and induced (“ripple”) effects for the IRL counties
Other Assumptions and Information
 The year 2014 was identified as the “target year” for data and analysis, as some
employment information categories were not yet available for all of 2015. Year 2014
provides the most recent and accurate year-long information for the economic
contributions analysis.
 The IMPLAN model includes direct impacts (employment and spending), indirect
impacts (secondary effects of spending through incomes generated by relevant industries
for housing, etc.), and induced impacts (indirect spending by employees). For example, a
boat dealer would be compensated (direct impact), might own a house (indirect impact)
and may spend money on dining out, which would generate economic activity through
income for cooks and servers in restaurants (induced impacts). IMPLAN incorporates a
series of complex algorithms that calculate direct, indirect, and induced impacts.
 The Florida’s Oceans and Coasts report includes some limited economic data points and
contributions related to nearshore Atlantic coastal resources and waters. For the IRL
region, there is a strong economic interdependence derived from the close physical and
ecological relationships and interconnections between the IRL and nearshore Atlantic
coastal ecosystems. For example, 75 percent of all commercially landed saltwater fish
and 80 to 90 percent of all recreationally caught saltwater fish spend part of their lives in

the Indian River Lagoon (pers.comm. Dr. Zack Jud, Director of Education and Exhibits,
Florida Oceanographic Society, March 2016. Therefore, the methodology does not
exclude these data points and economic values from the IRL economic contributions
analysis.
 One of the major economic values related to the IRL is the value premium for waterfront
real estate, both commercial and residential. The scope and budget for this economic
impact study did not allow for collection and analysis of the thousands of properties
along the 156-mile length of the lagoon, but there is significant potential that additional
economic value attributable to the lagoon is represented. These values would include
property tax revenues from incremental waterfront property, real estate sales activity for
brokers and realtors, premiums for direct access to private piers and other activities.
 When calculating economic outputs and valuations related to the IRL, the industry group
known as Defense & Aerospace was included in the study. In the 1950’s, the defense and
aerospace industry began scouting ideal locations in the United States where space
exploration operations could be best established. Aerospace industry representatives, the
national space program, and the military were especially attracted to the northern IRL
region for its undeveloped open spaces, location abutting wide-water bodies of the IRL
and Atlantic Ocean, and its close proximity to the equator. Over time the industry grew,
diversified, and flourished in the area and became the international model for how and
where to operate a successful aerospace program. Much of the competition in today’s
aerospace industry sector considers the IRL-region as a geographic paradigm when
looking to expand to other states. In addition, the quality of life and place offered by the
IRL is important to the industry for recruiting and retaining a new generation of talented
engineers and technicians who can be particularly discriminating about who they work
for and where they work (pers.comm. Dale Ketcham, Chief of Strategic Alliances, Space
Florida. June 2016). Including the Defense & Aerospace industry was the subject of
significant debate by the subject matter experts. For this reason, this report presents
economic output figures for the IRL region with and without contributions from this
industry group.
 NAICS code descriptions only include those employees who pay into the unemployment
tax system. Self-employed, seasonal, part-time and contract employees are not included
in the NAICS code totals; this factor suggests that employment and compensation values
for each employment category do not reflect total employment and compensation for the
IRL geographic area. As an example, the NAICS codes lists very few commercial
fishermen employed in the IRL counties, but there are many self-employed, part-time and
seasonal fishing employees in the IRL region.
 Although tourism is Florida’s largest industry, using the NAICS code employment
approach is not structured to capture all visitor-related expenditures and jobs for IRL
region, especially in the Food & Beverage and Accommodation categories. For example,
some tourists are defined as overnight guests who are “visiting friends and relatives”
(VFRs). This definition is important because expenditures made by the VFR market are
significant, but are separate from documented hotel, condominium and seasonal rental
occupants who pay for lodging. County-specific, visitor-related economic data in the IRL
study area is also limited and inconsistent. Florida’s Tourism Development Tax offers
one measure, but the amount tourism tax charged varies by county, and includes a state
and local tax rate charged on hotels, condominium and residential rentals in aggregate.
In order to disaggregate and supplement data gaps in the economic impact of all “visitor-
related” or tourism expenditures, the study utilized 2015 economic reporting documents
from STR Global, the industry leader in hotel performance data; Visit Florida’s 2014
Florida Visitor Study which includes IRL region-specific visitor expenditure and
economic impact data; and employment generation data contained in a 2013 study by the
Center for Competitive Florida and Florida Tax Watch titled, Investing in Tourism:
Analyzing the Economic Impact of Expanding Florida Tourism. Appendix 1 provides
further explanation of how employment and expenditures in the Food & Beverage and
Accommodation categories were adjusted to more accurately measure these economic
impacts for the IRL region.
 The Florida Department of Revenue receives a flat six percent share of the Tourism
Development Tax (TDT); individual Florida counties are also allowed to set their local
TDT rates, which range from four percent to six percent of all hotel/motel, condominium
and seasonal rental properties. The state’s six percent share has not been included in the
total TDT calculations, because it is used to fund statewide tourism marketing and other
programs.
 The scope of work for this study was to document as many water-related economic
activities associated with the IRL, the amount of employment, sales tax and other public
revenues generated, and the aggregated economic effects of these sources on the regional
economy within the designated geographic area defined as the IRL.
 The structure of the report is intended to be standardized enough to be easily replicable in

the future. These types of regional industry-based economic impact analyses might be
repeated regularly in order to monitor the positive and negative economic impacts
associated with different water-related economic drivers, and subsequently compare
patterns, differences and causal effects of outside forces. These forces may include water
quality, commercial fishing volumes, boat sales and registrations, real estate values, etc.
 Confidential Labor Statistics
Total annual economic output information for the IRL region was calculated using raw
NAICS code data related to wages, income, employment and sales. Some of the detailed
industry data had to be suppressed in order to protect information about individual
employers. This is based on confidentiality requirements written in State
Law. Confidential data is noted with an asterisk in each table or figure where this
confidentiality applies.

Economic Contribution of Living Resources

The Living Resources industry group included three industry sectors represented in the NAICS
code data (Table 1). The total annual wages of Living Resources was $1,277,712 in 2014. The
dominant sector in Living Resources was *Shellfishing, followed by Other Marine Fishing
(29.0%) and *Finfishing.
Table 1
Living Resources
Contribution to the Indian River Lagoon Economy
Based on 2014 NAICS Data from FDEO, LMS, QCEW
Industry Sector Total Annual Wages Percent
Finfishing * *
Shellfishing * *
Other Marine Fishing $370,689 29.0%
Total $1,277,712 100.0%
Adjusted to include 20% of the total annual wages in Volusia County
*Confidential data

Economic Contribution of Marine Industries

The Marine Industries industry group included 16 industry sectors represented in the NAICS
code data (Table 2; Figure 2). The total annual wages of Marine Industries was $130,669,582 in
2014. The dominant sectors in Marine Industries were Boat Building (39.2%), followed by Boat
Dealers (17.7%), and Marinas (12.3%).
Table 2
Marine Industries
Ship Building & Repair $5,846,642 4.5%
Boat Building $51,229,290 39.2%
Other Transportation Goods Merchant Wholesalers $14,544,765 11.1%
Sporting Goods Merchant & Marine Pleasure Craft Wholesalers $8,474,043 6.5%
Boat Dealers $23,123,547 17.7%
Deep Sea Freight Transportation * *
Deep Sea Passenger Transportation NA NA
Coastal & Great Lakes Freight Transportation NA NA
Coastal & Great Lakes Passenger Transportation * *
Inland Water Freight Transportation $1,050,178 0.8%
Inland Water Passenger Transportation NA NA
Ports and Harbor Operations * *
Marine Cargo Handling $5,338,879 4.1%
Navigational Services to Shipping $2,667,622 2.0%
Other Support Activities for Water Transportation $1,463,244 1.1%
Marinas $16,103,453 12.3%
Total $130,669,582 100.0%
*Confidential data

Figure 2
Marine Industries
Total Annual Wages
$51,229,290
$23,123,547
$14,544,765 $16,103,453
$5,846,642 $8,474,043
$5,338,879

Economic Contribution of Recreation & Visitor-related

The Recreation and Visitor-related industry group included four industry sectors represented in
the NAICS code data (Table 3; Figure 3). The total annual wages of Recreation & Visitor-related
was $129,799,711 in 2014. The dominant sectors in Recreation & Visitor-related were Hotels &
Motels, Except Casino Hotels (76.7%), followed by All Other Amusement & Recreational
Industries (20.2%).
Table 3
Recreation & Visitor-related
Scenic & Sightseeing Water Transportation $2,939,215 2.3%
Recreational Goods Rentals $1,058,932 0.8%
All Other Amusement & Recreation Industries $26,191,764 20.2%
Hotels & Motels, Except Casino Hotels $99,609,799 76.7%
Total $129,799,711 100.0%
Figure 3
Recreation & Visitor-related
Total Annual Wages
$99,609,799
$26,191,764
$2,939,215 $1,058,932
Scenic & Recreational Goods All Other Hotels & Motels,

Sightseeing Water Rentals Amusement & Except Casino
Transportation Recreation Hotels
Industries

Economic Contribution of Resource Management

The Resource Management industry group included two industry sectors represented in the
NAICS code data (Table 4; Figure 4). The total annual wages of Resource Management was
$18,704,651 in 2014. The dominant sectors in Resource Management were Administration of
Conservation Programs (59.8%), followed by Air/Water & Waste Program Administration
($40.2%).
Table 4
Resource Management
Air/Water & Waste Program Administration $7,514,967 40.2%
Administration of Conservation Programs $11,189,684 59.8%
Total $18,704,651 100.0%
Figure 4
Resource Management
Total Annual Wages
$11,189,684
$7,514,967
Air/Water & Waste Program Administration of Conservation

Administration Programs

Economic Contribution of Defense & Aerospace

The Defense & Aerospace industry group included nine industry sectors represented in the
NAICS code data (Table 5; Figure 5). The total annual wages of Defense & Aerospace was
$969,319,495 in 2014. The dominant sectors in Defense & Aerospace were Search/Detection &
Navigation Instruments (31.3%), followed by Space Research & Technology (22.4%), Guided
Missile & Space Vehicle Manufacturing (16.9%) and National Security (16.1%).
Table 5
Defense & Aerospace
Search/Detection & Navigation Instruments $303,666,364 31.3%
Aircraft Manufacturing $50,213,704 5.2%
Aircraft Engine & Parts Manufacturing $27,997,049 2.9%
Other Aircraft Parts & Equipment $46,373,604 4.8%
Guided Missile & Space Vehicle Manufacturing $163,545,661 16.9%
Space Vehicle Propulsion Units & Parts Manufacturing * *
Other Guided Missile & Space Vehicle Parts * *
Space Research & Technology $217,054,441 22.4%
National Security $155,711,192 16.1%
Total $969,319,495 100.0%
*Confidential data

Figure 5
Defense & Aerospace
Total Annual Wages
$303,666,364
$217,054,441
$163,545,661 $155,711,192
$50,213,704 $46,373,604
$27,997,049

Economic Contribution of Industry Groups

The industry group contribution to the economy of the IRL based on NAICS data is presented in
Table 6 and Figures 6A and 6B. Defense & Aerospace is considered an important component of
the economy of the IRL study area. However, because the economic activity associated with
Defense & Aerospace is not considered to be completely dependent on water-related lagoon
activities, the total economic contribution of the industry groups are compared with and without
Defense & Aerospace. The total annual wages of the industry groups in the IRL study area was
$1,249,771,151 in 2014 including Defense & Aerospace, and $280,451,656 without Defense &
Aerospace. The total annual wages of the Defense & Aerospace industry group was 7.4 times as
large as the next largest industry group, Marine Industries. Without Defense & Aerospace, the
dominant industry groups were Marine Industries (46.6%) and Recreation and Visitor-related
($46.3%), followed by Resource Management (6.7%), and Living Resources (0.5%).
Table 6
Industry Group
With Defense & Aerospace
Industry Group Total Annual Wages Percent
Living Resources $1,277,712 0.1%
Marine Industries $130,669,582 10.5%
Recreation and Visitor-related $129,799,711 10.4%
Resource Management $18,704,651 1.5%
Defense & Aerospace $969,319,495 77.6%
Total $1,249,771,151 100.0%
Without Defense & Aerospace
Industry Group Total Annual Wages Percent
Total $280,451,656 100.0%

Figure 6A
Total Annual Wages
$1,277,712 $130,669,582
0.1% 10.5%
$129,799,711
10.4%
$18,704,651
1.5%
$969,319,495
77.6%

Figure 6B
Total Annual Wages
$18,704,651 $1,277,712
6.7% 0.5%
$130,669,582
46.6%
$129,799,711
46.3%
Living Resources Marine Industries Recreation and Visitor-related Resource Management

Total Economic Contribution of Industry Groups

The IMPLAN model was run to provide an estimate of the total economic activity related to each
industry group associated with the IRL study area (Table 7; Figures 7A and 7B). Defense &
Aerospace is considered an important component of the economy of the IRL study area.
However, because the economic activity associated with Defense & Aerospace is not considered
to be completely dependent on water-related lagoon activities, the total economic contribution of
the industry groups are compared with and without Defense & Aerospace. The total economic
activity of the industry groups was $7,640,311,564 in 2014 including Defense & Aerospace, and
$2,491,160,186 without Defense & Aerospace. Excluding Defense & Aerospace, the dominant
industry groups were Recreation and Visitor-related (63.2%), followed by Marine Industries
(30.8%), Resource Management (4.1%), and Living Resources (1.9%).
Table 7
Industry Group
Total Output Amount Based on IMPLAN Analysis
Industry Group Amount Percent
Recreation and Visitor-related $1,574,078,178 20.6%
Defense & Aerospace $5,149,151,378 67.4%
Total $7,640,311,564 100.0%
Recreation and Visitor-related $1,574,078,178 63.2%
Total $2,491,160,186 100.0%
Adjusted to include 20% of the amount in Volusia County
Adjusted to include estimated total jobs in Accommodations and Food Services within Recreation & Visitor-related industry group
Total output amount equals all sales associated with that economic activity or jobs classification

Figure 7A
Industry Group Total Economic Contribution
$48,230,566 $767,440,497
0.6% 10.0%
$1,574,078,178
20.6%
$101,410,945
$5,149,151,378 1.3%
67.4%

Figure 7B
Industry Group Total Economic Contribution
$101,410,945 $48,230,566
4.1% 1.9%
$767,440,497
30.8%
$1,574,078,178
63.2%

Total Economic Contribution of Volusia County

industry group associated with the Volusia County portion of the IRL study area (Table 8;
Figures 8A and 8B). Because 80 percent of the population of Volusia County coastal
municipalities is adjacent to the Halifax River Lagoon rather than the IRL, only 20 percent of the
Volusia County economic activity is included in this analysis. The total economic activity of the
industry groups in Volusia County was $330,353,935 in 2014 including Defense & Aerospace,
and $277,251,218 without Defense & Aerospace. Excluding Defense & Aerospace, the dominant
industry groups were Recreation and Visitor-related (73.4%), followed by Marine Industries
(25.5%), Living Resources (0.6%), and Resource Management (0.5%).
Table 8
Volusia County
Total $330,353,935 100.0%
Total $277,251,218 100.0%

Figure 8A
Volusia County Total Economic Contribution
$53,102,717 $1,607,135
16.1% 0.5% $70,573,732
$1,497,265 21.4%
0.5%
$203,573,086
61.6%

Figure 8B
Volusia County Total Economic Contribution
$1,497,265 $1,607,135
0.5% 0.6%
$70,573,732
25.5%
$203,573,086
73.4%

Total Economic Contribution of Brevard County

industry group associated with the Brevard County portion of the IRL study area (Table 9;
Figures 9A and 9B). The total economic activity of the industry groups in Brevard County was
$5,353,605,910 in 2014 including Defense & Aerospace, and $1,111,737,960 without Defense &
Aerospace. Excluding Defense & Aerospace, the dominant industry groups were Recreation and
Visitor-related (72.0%), followed by Marine Industries (22.5%), Resource Management (3.9%),
and Living Resources (1.6%).
Table 9
Brevard County
Defense & Aerospace $4,241,867,950 79.2%
Total $5,353,605,910 100.0%
Total $1,111,737,960 100.0%

Figure 9A
Brevard County Total Economic Contribution
$18,250,109 $249,892,104
0.3% 4.7%
$800,266,645
14.9%
$43,329,102
0.8%
$4,241,867,950
79.2%

Figure 9B
Brevard County Total Economic Contribution
$43,329,102 $18,250,109
3.9% 1.6%
$249,892,104
22.5%
$800,266,645
72.0%

Total Economic Contribution of Indian River County

industry group associated with the Indian River County portion of the IRL study area (Table 10;
Figures 10A and 10B). The total economic activity of the industry groups in Indian River County
was $797,562,586 in 2014 including Defense & Aerospace, and $228,797,246 without Defense
& Aerospace. Excluding Defense & Aerospace, the dominant industry groups were Recreation
and Visitor-related (65.7%), followed by Marine Industries (19.7%), Resource Management
(12.7%), and Living Resources (1.8%).
Table 10
Indian River County
Total $797,562,586 100.0%
Total $228,797,246 100.0%

Figure 10A
Indian River County Total Economic Contribution
$4,220,849 $45,070,653
0.5% 5.7%
$150,414,030
18.9%
$29,091,714
3.6%
$568,765,340
71.3%

Figure 10B
Indian River County Total Economic Contribution
$29,091,714 $4,220,849
12.7% 1.8%
$45,070,653
19.7%
$150,414,030
65.7%

Total Economic Contribution of St. Lucie County

industry group associated with the St. Lucie County portion of the IRL study area (Table 11;
Figures 11A and 11B). The total economic activity of the industry groups in St. Lucie County
was $549,660,385 in 2014 including Defense & Aerospace, and $549,244,295 without Defense
& Aerospace. Excluding Defense & Aerospace, the dominant industry groups were Recreation
and Visitor-related (55.0%), followed by Marine Industries (39.2%), Resource Management
(4.5%), and Living Resources (1.3%).
Table 11
St. Lucie County
Defense & Aerospace $416,090 0.1%
Total $549,660,385 100.0%
Total $549,244,295 100.0%

Figure 11A
St. Lucie County Total Economic Contribution
$416,090 $7,072,422
$24,954,338
0.1% 1.3%
4.5%
$215,246,003
39.2%
$301,971,532
54.9%

Figure 11B
St. Lucie County Total Economic Contribution
$24,954,338 $7,072,422
4.5% 1.3%
$215,246,003
39.2%
$301,971,532
55.0%

Total Economic Contribution of Martin County

industry group associated with the Martin County portion of the IRL study area (Table 12;
Figures 12A and 12B). The total economic activity of the industry groups in Martin County was
$609,128,748 in 2014 including Defense & Aerospace, and $324,129,467 without Defense &
Aerospace. Excluding Defense & Aerospace, the dominant industry groups were Marine
Industries (57.6%), followed by Recreation and Visitor-related (36.4%), Living Resources
(5.3%), and Resource Management (0.8%).
Table 12
Martin County
Total $609,128,748 100.0%
Total $324,129,467 100.0%

Figure 12A
Martin County Total Economic Contribution
$17,080,051
2.8%
$186,658,005
30.6%
$284,999,281
46.8%
$2,538,526 $117,852,885
0.4% 19.3%

Figure 12B
Martin County Total Economic Contribution
$2,538,526 $17,080,051
0.8% 5.3%
$117,852,885
36.4%
$186,658,005
57.6%

Total Economic Contribution of Indian River Lagoon Counties

The IMPLAN model was run to provide an estimate of the total economic activity associated
with lagoon-related industries in each of the counties within the IRL study area (Table 13;
Figures 13A and 13B). Because 80 percent of the population of Volusia County coastal
municipalities is adjacent to the Halifax River Lagoon rather than the IRL, only 20 percent of the
Volusia County economic activity is included in this analysis. The total economic activity of the
counties in the IRL study area was $7,640,311,564 in 2014 with Defense & Aerospace, and
$2,491,160,186 without Defense & Aerospace. Brevard County had the highest total economic
contribution with or without Defense & Aerospace. Excluding Defense & Aerospace, Brevard
County had the greatest economic contribution (44.6%), followed by St. Lucie County (22.0%),
Martin County (13.0%), Volusia County (11.1%), and Indian River County (9.2%).
Table 13
County Contribution to the Indian River Lagoon Economy
County Amount Percent
Volusia County $330,353,935 4.3%
Brevard County $5,353,605,910 70.1%
Indian River County $797,562,586 10.4%
St. Lucie County $549,660,385 7.2%
Martin County $609,128,748 8.0%
Total $7,640,311,564 100.0%
County Amount Percent
Volusia County $277,251,218 11.1%
Brevard County $1,111,737,960 44.6%
Indian River County $228,797,246 9.2%
St. Lucie County $549,244,295 22.0%
Martin County $324,129,467 13.0%
Total $2,491,160,186 100.0%

Figure 13A
County Total Economic Contribution
$5,353,605,910
$797,562,586 $549,660,385 $609,128,748

$330,353,935
Volusia County Brevard County Indian River St. Lucie County Martin County
County
Figure 13B
County Total Economic Contribution
$1,111,737,960
$549,244,295
$324,129,467
$277,251,218
$228,797,246
County

Employment in Indian River Lagoon Counties

Total employment in the industry groups impacting the IRL economy based on NAICS data was
20,050 with Defense & Aerospace and 9,231 without Defense & Aerospace (Table 14; Figures
14A and 14B). Defense & Aerospace employment accounted for more than half (54%) of the
total employment in all industry groups. Brevard County had the largest employment both with
(13,308) and without (4,208) Defense & Aerospace. Without Defense & Aerospace, Recreation
and Visitor-related employment accounted for 60.7% of all employment, followed by Marine
Industries (34.8%), Resource Management (3.9%), and Living Resources (0.5%).
Table 14
Living Marine Recreation and Resource Defense &
County Resources Industries Visitor-related Management Aerospace Total Percent
Volusia County * 237 498 * * 880 4.4%
Brevard County 34 1,189 2,780 205 9,100 13,308 66.4%
Indian River County * 196 966 * 865 2,119 10.6%
St. Lucie County * 815 726 * * 1,580 7.9%
Martin County * 780 631 * 720 2,163 10.8%
Total 50 3,217 5,601 363 10,819 20,050 100.0%
Percent 0.2% 16.0% 27.9% 1.8% 54.0% 100.0%
County Resources Industries Visitor-related Management Aerospace Total Percent
Volusia County * 237 498 * 747 8.1%
Brevard County 34 1,189 2,780 205 4,208 45.6%
Indian River County * 196 966 * 1,254 13.6%
St. Lucie County * 815 726 * 1,579 17.1%
Martin County * 780 631 * 1,443 15.6%
Total 50 3,217 5,601 363 9,231 100.0%
Percent 0.5% 34.8% 60.7% 3.9% 100.0%
Adjusted to include 20% of the employees in Volusia County
Reflects employees covered under Unemployment Insurance and Reemployment Assistance
*Confidential data

Figure 14A
13,308
2,119 2,163
1,580
880
County
Figure 14B
4,208
1,579
1,443
1,254
747
County

County Contribution to Employment

The IMPLAN model was run to provide an estimate of the indirect, induced, and total
employment in each county of the IRL study area (Table 15; Figures 15A and 15B). With
Defense & Aerospace, Brevard County had the highest total employment (65.4%), followed by
St. Lucie County (11.5%), Martin County (8.1%), Indian River County (7.8%), and Volusia
County (7.3%). Without Defense & Aerospace, Brevard County had still the highest total
employment (49.1%), followed by St. Lucie County (20.0%), Volusia County (11.9%), Martin
County (10.3%), and Indian River County (8.8%).
Table 15
Based on IMPLAN Analysis
Direct Indirect Induced
County Employment Employment Employment Total Percent
Volusia County 3,121 336 615 4,072 7.3%
Brevard County 21,485 6,200 8,715 36,400 65.4%
Indian River County 3,057 593 697 4,347 7.8%
St. Lucie County 4,802 688 902 6,392 11.5%
Martin County 3,173 572 740 4,485 8.1%
Total 35,638 8,389 11,669 55,696 100.0%
County Employment Employment Employment Total Percent
Volusia County 2,988 269 549 3,806 11.9%
Brevard County 12,385 1,182 2,145 15,712 49.1%
Indian River County 2,192 254 372 2,818 8.8%
St. Lucie County 4,801 687 902 6,390 20.0%
Martin County 2,453 378 459 3,290 10.3%
Total 24,819 2,770 4,427 32,016 100.0%

Figure 15A
40,000
35,000
30,000
Employees
25,000
20,000
15,000
10,000
5,000
0
Volusia County Brevard County Indian River St. Lucie Martin County
County County
Direct Employment Indirect Employment Induced Employment
Figure 15B
16,000
14,000
12,000
Employees
10,000
8,000
6,000
4,000
2,000
0
Volusia County Brevard County Indian River St. Lucie Martin County
County County

Industry Group Contribution to Employment

The IMPLAN model was run to provide an estimate of the indirect, induced, and total
employment in each industry group the IRL study area (Table 16; Figures 16A and 16B). With
Defense & Aerospace included, Recreation and Visitor-related had the highest total employment
(46.4%), followed by Defense & Aerospace (42.5%), Marine Industries (9.5%), Resource
Management (1.3%), and Living Resources (0.3%). Without Defense & Aerospace, Recreation
and Visitor-related had the highest total employment (80.7%), followed by Marine Industries
(16.6%), Resource Management (2.3%), and Living Resources (0.4%).
Table 16
Industry Group Employment Employment Employment Total Percent
Living Resources 50 42 52 144 0.3%
Marine Industries 3217 1082 1000 5299 9.5%
Recreation and Visitor-related 21189 1451 3212 25852 46.4%
Resource Management 363 196 162 721 1.3%
Defense & Aerospace 10819 5619 7242 23680 42.5%
Total 35638 8390 11668 55696 100.0%
Industry Group Employment Employment Employment Total Percent
Living Resources 50 42 52 144 0.4%
Marine Industries 3217 1082 1000 5299 16.6%
Recreation and Visitor-related 21189 1451 3212 25852 80.7%
Resource Management 363 196 162 721 2.3%
Total 24819 2771 4426 32016 100.0%

Figure 16A
30,000
25,000
Employees
20,000
15,000
10,000
5,000
0
Resources Industries Visitor-related Management Aerospace
Figure 16B
30,000
25,000
Employees
20,000
15,000
10,000
5,000
0
Living Resources Marine Industries Recreation and Resource
Visitor-related Management

Total Economic Contribution by Industry Group

The IMPLAN model was run to provide an estimate of the direct effect, indirect effect, induced
effect, and total economic contribution of each industry group the IRL study area (Table 17;
Figures 17A and 17B). The total economic contribution of the industry groups analyzed was
$7,640,311,564 with Defense & Aerospace, and $2,491,160,186 without Defense & Aerospace.
Considering all five industry groups, Defense & Aerospace had the highest economic
contribution (67.4%), followed by Recreation and Visitor-related (20.6%), Marine Industries
(10.0%), Resource Management (1.3%), and Living Resources (0.6%). Without Defense &
Aerospace, Recreation and Visitor-related had the highest economic contribution (63.2%),
followed by Marine Industries (30.8%), Resource Management (4.1%), and Living Resources
(1.9%).
Table 17
Industry Group Effect Effect Effect Total Percent
Living Resources $36,352,575 $5,898,065 $5,979,927 $48,230,566 0.6%
Marine Industries $524,134,087 $128,143,592 $115,162,818 $767,440,497 10.0%
Recreation and Visitor-related $1,031,020,914 $170,340,248 $372,717,015 $1,574,078,178 20.6%
Resource Management $58,757,539 $23,595,112 $19,058,293 $101,410,945 1.3%
Defense & Aerospace $3,580,678,909 $729,949,742 $838,522,727 $5,149,151,378 67.4%
Total $5,230,944,025 $1,057,926,759 $1,351,440,780 $7,640,311,564 100.0%
Industry Group Effect Effect Effect Total Percent
Living Resources $36,352,575 $5,898,065 $5,979,927 $48,230,566 1.9%
Marine Industries $524,134,087 $128,143,592 $115,162,818 $767,440,497 30.8%
Recreation and Visitor-related $1,031,020,914 $170,340,248 $372,717,015 $1,574,078,178 63.2%
Resource Management $58,757,539 $23,595,112 $19,058,293 $101,410,945 4.1%
Total $1,650,265,115 $327,977,017 $512,918,053 $2,491,160,186 100.0%
Adjusted to include 20% of the amounts in Volusia County

Figure 17A
$6
$5
$4
Billions
$3
$2
$1
$0
Living Resources Marine Recreation and Resource Defense &
Industries Visitor-related Management Aerospace
Direct Effect Indirect Effect Induced Effect
Figure 17B
$1.6
$1.4
$1.2
$1.0
Billions
$0.8
$0.6
$0.4
$0.2
$0.0
Living Resources Marine Industries Recreation and Resource
Visitor-related Management

Total Economic Contribution by County

The IMPLAN model was run to provide an estimate of the direct effect, indirect effect, induced
effect, and total economic contribution of each county in the IRL study area (Table 18; Figures
18A and 18B). The total economic contribution of the counties analyzed was $7,640,311,564
with Defense & Aerospace, and $2,491,160,186 without Defense & Aerospace. With Defense &
Aerospace, Brevard County had the highest total economic contribution (70.1%), followed by
Indian River County (10.4%), Martin County (8.0%), St. Lucie County (7.2%), and Volusia
County (4.3%). Without Defense & Aerospace, Brevard County had still the highest total
economic contribution (44.6%), followed by St. Lucie County (22.0%), Martin County (13.0%),
Volusia County (11.1%), and Indian River County (9.2%).
Table 18
County Effect Effect Effect Total Percent
Volusia County $214,178,839 $43,375,814 $72,799,282 $330,353,935 4.3%
Brevard County $3,559,409,878 $786,540,410 $1,007,655,622 $5,353,605,910 70.1%
Indian River County $630,482,916 $82,211,375 $84,868,295 $797,562,586 10.4%
St. Lucie County $366,010,100 $80,314,689 $103,335,596 $549,660,385 7.2%
Martin County $460,862,292 $65,484,471 $82,781,985 $609,128,748 8.0%
Total $5,230,944,025 $1,057,926,759 $1,351,440,780 $7,640,311,564 100.0%
County Effect Effect Effect Total Percent
Volusia County $177,578,323 $34,697,695 $64,975,199 $277,251,218 11.1%
Brevard County $724,837,337 $138,916,024 $247,984,599 $1,111,737,960 44.6%
Indian River County $152,437,370 $31,062,306 $45,297,570 $228,797,246 9.2%
St. Lucie County $365,692,976 $80,247,706 $103,303,613 $549,244,295 22.0%
Martin County $229,719,109 $43,053,286 $51,357,072 $324,129,467 13.0%
Total $1,650,265,115 $327,977,017 $512,918,053 $2,491,160,186 100.0%
Adjusted to include 20% of the amounts in Volusia County

Figure 18A
$6
$5
$4
Billions
$3
$2
$1
$0
County
Figure 18B
$1.2
$1.0
$0.8
Billions
$0.6
$0.4
$0.2
$0.0
County

Cost of a Sustainable Indian River Lagoon Economy

Enhancing water quality, increasing habitat, providing additional points of access, and
implementing best management practices to improve the IRL can generate thousands of new jobs
and increase the value and output of the IRL-based economy. This presumes state and local plans
to improve the lagoon’s health and productivity are implemented and continue to be successful in
the future. Implementing these plans will require a long-term commitment of time and funding.
One way to estimate how much it will cost annually and cumulatively over time to achieve a
sustainable IRL, is to estimate the cost of implementing the adopted Basin Management Action
Plans (BMAPs) for the IRL region.
The Florida Department of Environmental Protection BMAP initiative is based on their Total
Maximum Daily Load (TMDL) program, which is a statewide watershed-based management
approach to restore and protect water quality in Florida. TMDLs are developed, allocated, and
implemented through phased approach, which includes the development of a BMAP. Each
BMAP contains a comprehensive set of strategies/projects for restoring impaired waters by
reducing pollutant loadings to meet the allowable loadings established in a TMDL. TMDLs are
water quality targets based on state water quality standards for specific pollutants such as
excessive nitrogen and phosphorus. BMAPs are broad-based plans that are developed with local
stakeholders and rely on local input and local commitment for implementation.
Most of the IRL is addressed by four adopted BMAPs, which cover an area of over 990,000
acres. These BMAPs are spread across boundaries of five IRL counties and two regional
planning councils (Exhibit 2). TMDL reduction targets focus on improving water clarity and
quality by reducing total nitrogen and phosphorus loads to the lagoon (Exhibit 3).
Exhibit 2
Indian River Lagoon BMAPs
BMAP County Regional Planning Council
North Indian River Lagoon Brevard, Volusia ECFRPC
Banana River Lagoon Brevard ECFRPC
Central Indian River Lagoon Brevard, Indian River, St. Lucie ECFRPC and TCRPC
St. Lucie River and Estuary* Martin, St. Lucie TCRPC
* A portion of eastern Okeechobee County also is included in the St. Lucie River and Estuary BMAP
Exhibit 3
TMDL Total Required Reductions (Pounds/Year)
BMAP Total Nitrogen Total Phosphorus
North Indian River Lagoon 221,863.9 52,057.2
Banana River Lagoon 192,708.0 38,647.0
Central Indian River Lagoon* 836,713.0 144,453.0
St. Lucie River and Estuary 1,053,414.0 404,166.0
Total 2,304,698.9 639,323.2
* Reduction allocations were not calculated when this BMAP was developed, but TMDL’s have been set.


TCRPC and ECFRPC (2015) estimated it would cost $4.6 billion to accomplish the required
nutrient load reductions in all four of the BMAPs which is intended to establish a sustainable
IRL. By this measure, and extended over a 20-year period, it would require an annual investment
of $230 million to sustain an IRL-based economy. Based on the estimated 2015 population of the
IRL counties, the annual cost per person equates to about $140. If this is divided equally among
IRL-county households, each would pay around $334 a year, or about $28 a month. The 1996
Economic Assessment and Analysis of the IRL found that 1000 IRL households polled said they
would be willing to pay an additional $60 a year to support IRL improvement programs. The
2008 update of this study found that 50 percent of the 985 IRL households polled would be
willing to pay a one-time tax of $20 to improve the lagoon, with about 27 percent of the
respondents willing to pay a one-time tax of $400.
The cost calculations outlined above assume that IRL residents will bear the full cost of BMAP
implementation. This is not likely to be the case. Floridians outside the IRL region will
contribute to state water quality improvement programs and water management district projects
through state sales taxes and federal income taxes redistributed to the IRL region in the form of
state and federal assistance. American and foreign tourists will also offset some local
implementation costs through sales and other special taxes levied on local goods and services
that may flow to state water quality improvement programs. In addition, use of Land Acquisition
Trust Fund monies may also offset the cost of BMAP-related water quality improvement projects
for the lagoon.

Return on Investment from a Sustainable Indian River Lagoon

One way to calculate the current Return on Investment (ROI) from the Indian River Lagoon is to
establish a ratio of annual costs for providing a productive and sustainable IRL, compared to the
lagoon’s total annual economic output. Achieving water quality targets called for in the IRL-
region BMAPs should result in a sustainable IRL. It is estimated that meeting these targets will
cost the region $230 million a year over the next 20 years. This report estimates the 2014 Total
Annual Economic Output is valued at $7.6 billion. This value should remain stable or increase
over time as water quality improves from implementing the IRL BMAPs. When comparing this
average annual cost to the IRL’s total average annual economic output of $7.6 billion, the ROI
from a sustainable IRL is 33 to 1. Or, for every dollar spent on achieving a sustainable IRL, the
lagoon returns $33 in total economic value. Over the years, as the IRL improves in health and
productivity, annual ROI in the future can be expected to increase.

An Indian River Lagoon-Dependent Industry in Decline

Commercially harvested clams, oysters, crabs and shrimp were worth $12.6 million at the docks
in 1994 (Apogee Research Inc. 1996). Adjusted for inflation, this 1994 amount is $20.1 million
in 2015 dollars. The overall value of the commercial clam, oyster, crab, and shrimp harvest for
2015 is $4.3 million (NOAA 2015) – a decline of nearly 80 percent. Pounds of shellfish
harvested have also declined during about that same period, from 7.1 million pounds to 2 million
pounds, or almost 72 percent. IRL counties showing the sharpest decline in value and pounds
harvested were Volusia, Brevard and Martin.
The commercial fin fishery fared a little better, but still showed significant declines in value and
pounds landed. In 1990, the estimated value of commercial fin fish landed was worth $13
million (FFWCC 2016). Adjusted for inflation, this 1990 amount is $23.5 million in 2015
dollars. The overall value of the commercial fin fish harvest for 2015 is estimated at $14.8
million – a decline of 37 percent. Pounds of fin fish landed also declined during that same
period, from 17.3 million pounds to 8 million pounds or almost 54 percent. Indian River Lagoon
counties showing the sharpest declines in value and pounds harvested are Brevard, Indian River,
and Martin counties.

Boat Registrations as an Indicator of the Indian River Lagoon’s

Environmental and Economic Health
One indicator of the environmental health and economic productivity of the IRL is the increase
or decrease in boat registrations over time. Increased access and use of the IRL by boats for
fishing and other outdoor lagoon-based activities clearly has an economic impact on retail,
lodging, marine industry and food and beverage sectors, and also on sales tax, fuel tax, boat
registration fees and fishing license revenues. Recreational fishing represents significant
spending for lodging, transportation, food, fishing equipment, etc. Average per capita spending
for resident anglers is estimated to total approximately $186 per person per year, and spending
by non-resident anglers is estimated to average approximately $440 per person per year.
Recreational boating, fishing and other water-related activities dependent on the IRL accounts
for around $1.2 billion in total annual spending. This level of economic impact is also a clear
indication of the willingness of residents and visitors to pay for access to lagoon fishing grounds
and other lagoon-based activities.
Information provided by the Florida Department of Highway Safety and Motor Vehicles (2005-
2015) indicate a declining trend in boat registrations for the IRL region has been observed in the
last decade. Boat registration has decreased by 13,045 or by 11 percent since 2005, while the
IRL region’s population has increased by 12 percent or by 115,585 residents. Much of the
decline has been for boats in the 12 to 39-foot range with boat registrations in the 40-foot and
larger class remaining relatively flat. The sharpest declines in boat registrations have occurred in
Brevard, Volusia and St. Lucie counties.
IRL REGION BOATER

REGISTRATIONS
2005 2015

Missing and Incomplete Data

An economic impact analysis of this type is highly dependent on both the quantity and the
quality of the available data from which conclusions can be drawn. This study is based on
comprehensive information from a number of sources, including the FDEO Labor Market
Statistics Center; VisitFlorida, the state’s tourism development agency; and other federal, state,
and local agencies.
While there is a large amount of data available that is central to completion of this analysis, there
are data gaps, inconsistent methodologies across counties in documenting tourism and
tourist/visitor expenditure differences in the level of detail available between jurisdictions, and
differences in the manner and frequency with which data is collected and distributed. The
following list of missing and incomplete data is provided as a potential tool to both increase the
amount, breadth, and consistency of information to be used in future economic valuation updates
for the IRL.
The environmental sustainability of the IRL and of all of Florida’s waterways and ocean areas is
inextricably linked to sustaining future economic activity as well. Improving the following data
sources and opportunities to increase information about IRL economic activity will also improve
the analytical process and the accuracy of the conclusions. A list of missing and incomplete data
includes the following:
1. Many data categories and NAICS code information are not available at the block level or zip
code level. Breaking down demographic and employment information at a more detailed
level would make it possible to include economic activity information for the portions of
Palm Beach and Volusia counties that are part of the IRL study area. It is suggested that the
State of Florida take actions necessary to record NAICS code data by zip code.
2. While an assumed incremental value of waterfront real estate was included in a previous
study, the hedonic analysis of assumed values and distances were not fully documented, and
the scale and scope of this analysis was not expansive enough to document comparative
values along the 300 plus miles of shoreline along the IRL. The IRL adds value to both
commercial and residential real estate adjacent to and near waterfronts, but a larger study
would be required to meaningfully document the full measure of that increment, or multiple
location-based increments.
3. NAICS code data from FDEO, LMS, QCEW were used as the basis of the economic
activities incorporated into the IMPLAN model. These data are considered the most
comprehensive employment measure in economic analysis, but it is also possible that some
employment categories that support IRL economic activities may not be reflected in the
NAICS categories.
4. The data are not all inclusive. State labor statistics are based on employees who pay into the
unemployment tax system, but do not include a portion of self-employed, part-time and
seasonal workers.

5. Geography may also affect how data is captured. For example, within the commercial fishing
industry it is possible that fishing boats may come to the IRL and ocean fishing grounds
nearby from other Florida counties or out of state, but their employment and commercial haul
information would not be captured as part of the IRL study area.
6. The significant amount of economic activity for the Defense and Aerospace industry group
was considered dependent on its proximity to IRL and the Atlantic Ocean. Clearly the
Kennedy Space Center and Cape Canaveral have had a profound effect on related corporate
office locations, government facilities and tourism, particularly in Brevard County.
However, the dependency level of these industries on an environmentally-sustained IRL was
a more complex correlation to document. For this reason, the analysis includes economic
impacts with and without the defense and aerospace industries.
7. At a statewide impact level of $67 Billion (in 2015), tourism is Florida’s largest single
industry, both in terms of direct expenditures and indirect impacts. VisitFlorida, the official
state tourism marketing and documentation agency, provides both regional and statewide
data on the multiple categories of tourism impact. Within the IRL study area, however, there
is a significant inconsistency in the degree of detail available on annual tourism and visitor
volumes, expenditures and trends. Palm Beach County has very comprehensive data, but the
bulk of the tourism activity is not in the Jupiter area, and totals were available only at the
countywide level. Within the IRL study area, most of the tourism development funding
made possible by the tourist tax is used for marketing purposes, not for documentation of
tourist and visitor volumes, so it is not possible to understand trends and direct/indirect
effects on a county-by-county basis. It is suggested that a statewide effort be created to
determine a core of tourism information to be collected at the county level annually
paid for through the tourism taxes collected through VisitFlorida. The core data should
include:
a. total visitor volume (including both hotel-related stays and an estimate of VFRs,
which are not included in hotel-based visitor documentation;
b. leisure and business-based visitor volumes;
c. average length of stay;
d. average party size;
e. primary activities motivating the stay in Florida; and
f. average per person and per visitor party expenditures by category
8. Over time, the cruise terminals and port authorities in the IRL study area will have data on
shipping volumes, employment and other economic activities. At the time of this study,
there was not sufficient information available from port authorities and terminals to
document volume and trends. Because Florida has undertaken a statewide initiative to
upgrade its ports in response to widening of the Panama Canal (and the larger ships and
cargo volumes it will accommodate), this economic sector is likely to grow over time.
9. Recreational fishing is another major industry in Florida with permits for freshwater and
saltwater fishing held by both state residents and out-of-state recreational fishermen. In order
to consider expenditures for recreational fishing, the study used a report on spending released

in 2008 by NOAA and the National Marine Fisheries Agency and based on a state-by-state
national survey. The size of this recreational industry in Florida suggests the impact should
be included, but the data collected in 2006 is out-of-date. NOAA is currently conducting the
national survey again, and the updated expenditures report is expected to be released
sometime over the next one-to-two years. The future update of this analysis should include
the updated information on recreational fishing expenditures in Florida.

References and Data Sources

Apogee Research, Inc. 1996. Economic Assessment and Analysis of the Indian River Lagoon.
Natural Resource Valuation of the Lagoon. Submitted to the Indian River Lagoon
National Estuary Program, Melbourne, Florida.
Bureau of Economic and Business Research. 2015. University of Florida 2015, Florida Estimates
of Population 2015.
Bureau of Economic and Business Research. 2015. University of Florida 2016, Projections of
Florida Population by County 2020-2045.
Bureau of Labor Market Statistics. 2015. Quarterly Census of Employment and Wages Program
2015.
Center for Competitive Florida and Florida Tax Watch. 2013. Investing in Tourism: Analyzing
the Economic Impact of Expanding Florida Tourism. Tallahassee, Florida.
Conservation Alliance of St. Lucie County, St. Lucie Waterfront Council, and Audubon Society
of St. Lucie County. 1997. Death of the Indian River Lagoon? Submitted to The John D.
and Catherine T. MacArthur Foundation, Fort Pierce, Florida.
D.K. Shifflet & Associates, 2014. Florida Visitor Survey Data.
Dun & Bradstreet, Inc. 2015. ESRI Business Analyst, County Business Statistics
ESRI Business Analyst Forecasts, 2000-2020. Various Demographic Characteristics.
Federal Aviation Administration Commercial Airport Data and Statistics, Calendar Years 2009
through 2014. Enplanements at All Commercial Service Airports (by Rank).
Florida Department of Economic Opportunity & Bureau of Labor Statistics, 2014-3Q/2015.

Quarterly Census of Employment & Wages Program, NAICS Employment Data.
Florida Department of Highway Safety and Motor Vehicles, Data for 2005, 2010, 2012, 2014,
2015. Annual Vessel Statistics by County, Count of Vehicles Registered, Vehicle and
Vessel Reports and Statistics.
Florida Department of Revenue, 2013-2014. Annual Hotel Bed Tax Receipts.
Florida Department of Revenue, 2015-2015. Annual Sales Tax Collections.
Florida Fish and Wildlife Conservation Commission, Data for 2014. Recreational Hunting and
Fishing License Statistics by County (Saltwater and Freshwater Permits).

Florida Fish and Wildlife Conservation Commission. Retrieved on March 25, 2016 from:
https://public.myfwc.com/FWRI/PFDM/ReportCreator.aspx.
Florida Ocean Alliance. 2013. Florida’s Oceans and Coasts: A Economic and Cluster Analysis.
Florida Ocean Alliance. Fort Lauderdale, Florida.
Florida Regional Councils Association, 2015. Annual Report and Directory, 2014-2015.
Tallahassee, Florida.
Genter, Brad and Scott Steinback. 2008. The Economic Contribution of Marine Angler
Expenditures in the United States, 2006. U.S. Department of Commerce, NOAA,
National Marine Fisheries Service, NOAA Technical Memorandum NMFS – F/SPO – 94,
November, 2008
Hazen and Sawyer, P.C. 2008. Indian River Lagoon Economic Assessment and Analysis Update.
Prepared for the Indian River Lagoon National Estuary Program in cooperation with the
St. Johns River Water Management District and South Florida Water Management
District. Indian River Lagoon National Estuary Program. Palm Bay, Florida.
IMPLAN Group LLC. 2014. IMPLAN Pro 3.1, Economic Impact Multiplier Data,
Hendersonville, NC
Indian River Lagoon National Estuary Program. 1996. Indian River Lagoon Comprehensive
Conservation and Management Plan. Melbourne, Florida.
Indian River Lagoon National Estuary Program. 2008. Indian River Lagoon Comprehensive
Conservation and Management Plan Update. Palm Bay, Florida.
Jud, Zack, PhD., Director of Education and Exhibits, Florida Oceanographic Society, Personal
Communication, March 2016.
Ketcham, Dale, Chief of Strategic Alliances, Space Florida, Personal Communication, June
2016.
National Oceanographic and Atmospheric Administration (NOAA) Office of Science and

Technology, National Marine Fisheries Service, Data for 2014. Annual Commercial
Fishing Landings by County, Commercial Fisheries Statistics.
National Oceanographic and Atmospheric Administration (NOAA) Office of Science and

Technology. 2015. Commercial Fishing Statistics.
South Florida Water Management District, Florida Department of Environmental Protection, and
Florida Department of Agriculture and Consumer Services. 2009. St. Lucie River
Watershed Protection Plan. South Florida Water Management District, West Palm Beach,
Florida.

STR Global, 2010-2015. Hotel Performance Metrics.
Treasure Coast Regional Planning Council and East Central Florida Regional Planning Council.
2015. Impediments to Implementation of the Indian River Lagoon Basin Management
Action Plans. Stuart and Altamonte Springs, Florida.
U.S. Census Bureau, 2014. American Community Survey.
U.S. Bureau of Census, Zip Code Tabulation Areas, US Census 2010. USBoundary.com.
Retrieved on May 3, 2016 from:
http://www.usboundary.com/Areas/5-
Digit%20ZIP%T20Code%20Tabulation%20Area/Florida
Visit Florida. 2015. Profile of Domestic Visitors to Florida. 2014. Tallahassee, Florida.
Visit Florida. 2015. Estimate of Visitors to Florida 2014. Tallahassee, Florida.
Visit Florida. 2015. Travel Patterns of Florida Residents 2014. Tallahassee, Florida.
Visit Florida. 2015. Visit Florida Marketing Plan 2016-17
Visit Florida. 2016. E-mail communication, Sandy Johnson, Research Assistant, Visit Florida, to
Michael Busha, Executive Director, TCRPC on Visitor Spending, June 17, 2016.
Visit Florida Research Department. 2015. Profile of Domestic Visitors to Florida.
Visit Florida Research Department. 2015. Profile of Domestic Visitors to Florida By Vacation
Region.
Visit Florida Research Department. 2015. Profile of Domestic Visitors to Florida Traveling for
Leisure.
Visit Florida Research Department. 2015. Tourism’s Impact on the Florida Economy.

APPENDIX T:
FLORIDA REALTORS ASSOCIATION STUDY
The Impact of Water Quality
on Florida’s Home Values
March 2015
Final Report
Table of Contents
List of Tables .................................................................................................................................................................... ii

List of Figures..................................................................................................................................................................iii
Executive Summary ...................................................................................................................................................... iv
1. INTRODUCTION.................................................................................................................................................... 1
2. BACKGROUND: LAKE OKEECHOBEE, THE EVERGLADES, AND THE ESTUARIES..................... 3
3. METHODOLOGY AND DATA COLLECTION................................................................................................ 6
3.1 Isolating Price Effects in Real Estate Market Analyses ............................................................. 6
3.2 Hedonic Pricing Models for Lee and Martin Counties............................................................... 7
3.2.1 Sale Data ....................................................................................................................................... 9
3.2.2 Control Characteristics ........................................................................................................... 9
3.2.3 Measuring Ambient Water Quality ................................................................................. 19
4. MODEL RESULTS .............................................................................................................................................. 27
4.1 Overview of Hedonic Pricing Model Performance .................................................................. 27
4.2 The Effect of Recent Water Quality on Home Sale Prices ..................................................... 28
4.3 The Effect of Long-Term Water Quality on Home Sale Prices ............................................ 34
5. APPLYING THE RESULTS: AGGREGATE COUNTYWIDE EFFECTS ................................................ 38
5.1 A Simple Approach ............................................................................................................................... 38
5.2 Example: A Uniform Change in Water Quality .......................................................................... 40
5.3 Example: Martin County in the Summer of 2013..................................................................... 41
6. CONCLUDING REMARKS................................................................................................................................ 44
Appendix A ..................................................................................................................................................................... 45
Appendix B ..................................................................................................................................................................... 46
References ...................................................................................................................................................................... 49
Acknowledgements .................................................................................................................................................... 50
List of Tables
ES.1 Marginal price effects of changes in 1-month average Secchi disk depth at select
distances ............................................................................................................................................................... vii
ES.2 Marginal price effects of changes in 1-year average Secchi disk depth at select
distances .............................................................................................................................................................. viii
3.1 Summary statistics for control characteristics and sale prices ......................................................... 8
3.2 Summary statistics for water quality variables .................................................................................... 25
4.1 Marginal price effects of changes in 1-month water quality measures at select
distances ............................................................................................................................................................... 29
4.2 Marginal effects of a 20% improvement over the mean value of the 1-month water
quality measures ............................................................................................................................................... 32
4.3 Marginal price effects of changes in 1-year water quality measures at select distances..... 35
4.4 Marginal effects of a 20% improvement over the mean value of the 1-year water
quality measures ............................................................................................................................................... 36
A.1 Description of Lee County flood insurance risk map zone types................................................... 45
A.2 Description of Martin County flood insurance risk map zone types ............................................ 45
B.1 Results of 1-month and 1-year regressions for Martin County ...................................................... 46
B.2 Results of 1-month regressions for Lee County ................................................................................... 47
B.3 Results of 1-year regressions for Lee County ........................................................................................ 48
ii
List of Figures
2.1 Side-by-side maps of water flow through the Everglades ecosystem before human
intervention (left) and currently (right)..................................................................................................... 4
3.1 A monthly repeat-sales price index for single family homes in Florida. The base month
is January of 2000. ............................................................................................................................................ 12
3.2 Map of Lee County water bodies included in the study to control for water amenity
effects. .................................................................................................................................................................... 14
3.3 A graph illustrating the modeled relationship between home values and distances
from a waterfront when raw distance measurements are used in a hedonic pricing
model. .................................................................................................................................................................... 15
3.4 Graphs of exponential functions mapping raw distance measurements into relative
impact measurements. The blue curve represents the particular function used in this
study, RelativeImpact = exp(–2 ⨯ Distance). .......................................................................................... 16
3.5 A map depicting various distance-from-waterfront intervals for the Caloosahatchee

Estuary. Corresponding transformed values are displayed in the legend. ............................... 18
3.6 Map of Martin County water bodies included in the study to control for water amenity
effects. .................................................................................................................................................................. 20
3.7 Locations of Caloosahatchee Estuary water quality monitoring points used in the
study. Inset table identifies which types of measurements were used from each site........ 23
3.8 A map of St. Lucie River and Indian River Lagoon zones for which the Florida
Oceanographic Society reports water quality measurements. The map depicts only
those zones from which we used measurements; FOS currently monitors three other
nearby zones, as well....................................................................................................................................... 26
5.1 Map of Martin County single family home parcels, color-coded according to distance
from the “Estuary” class of water bodies as defined in Section 3.2. Each color accounts
for one quarter of the aggregate change in Martin County property values resulting
from a one-foot change in average Secchi disk depth in the St. Lucie Estuary,
Loxahatchee Estuary, and the Indian River Lagoon (north of the St. Lucie Inlet). ................. 42
iii
Executive Summary
Overview
There has long been a belief that there is a connection between home values and the quality
and clarity of Florida waterways. The objective of this study was to determine whether that “belief”
is in fact true.
We examined the impact of water quality and clarity on the sale prices of homes in Martin
and Lee counties over a four-year period, from 2010 through 2013. What was clearly found was
that the ongoing problem of polluted water in the Caloosahatchee and St. Lucie rivers and estuaries
has indeed resulted in a negative impact on home values.
In addition, the study found a significant economic impact resulting from improved water
quality and clarity - Lee County’s aggregate property values increase by an estimated $541 million,
while Martin County’s aggregate property values increase by an estimated $428 million. These
increased property values also provide additional revenue for city and county governments.
How the Study Was Done
We used hedonic pricing models to control for the effects of other factors which are known
to affect home prices, thereby allowing for the true impact of water quality to be estimated. The
controls we used in the models included structural characteristics for each home that were
recorded either in the county property tax rolls or local multiple listing service (MLS) listings.
Examples of these structural characteristics include the effective age of the home, lot acreage,
heated square footage, and whether the home has a swimming pool.
We also factored in characteristics of the neighborhoods surrounding each home, utilizing
school attendance zone maps and corresponding school quality scores, as well as U.S. Census block
group-level estimates of median household income and the percentages of residents who are age 65
or older (to identify areas popular for retirees) or less than age 18 (to identify areas popular for
families with children). Other controls were put in place to account for whether homes were on a
waterfront, as well as their distance to the ocean and other major water bodies, including the
estuaries of concern. We also accounted for macroeconomic and seasonal economic effects using
standard techniques for these types of models.
iv
For the Lee County models, we chose four different types of ambient water quality
measurements from the STORET database maintained by the Florida Department of Environmental
Protection (FDEP), each collected roughly once per month at up to 16 separate monitoring sites in
the Caloosahatchee River between the W. P. Franklin Lock and the mouth of the river at San Carlos
Bay. These measures were not chosen based on whether they are harmful to the ecosystem so
much as how well they might represent visible characteristics of the water—something which
might impact the opinion of a potential homebuyer viewing the home. In particular, an emphasis
was placed on measurements of water clarity.
The first type of measure we used was micrograms per liter (μg/L) of chlorophyll a in the
water, which captures the extent to which the water is populated by microscopic photosynthetic
organisms that can cause discoloration and murkiness in the water. Some of these organisms are
harmful to both aquatic life and humans, such as cyanobacteria (i.e. blue-green algae).
The second type of measure we used was milligrams per liter (mg/L) of dissolved oxygen. A
lack of oxygen in the water is associated with the presence of algae-friendly nutrients. Moreover, if
dissolved oxygen levels remain too low for too long, most aquatic life will begin to die off, which of
course can result in unpleasant sights and odors that may impact the perceptions of potential
homebuyers.
The third measure, turbidity, is a measure of the cloudiness of water due to any sort of tiny
suspended materials, living or non-living. Turbidity is measured by an instrument called a
nephelometer, which captures its readings with a deflector that measures the way in which a light
beam is deflected by particles suspended in the water. Nephelometer measurements are reported
in Nephelometric Turbidity Units (NTUs). Higher NTUs mean the water is more turbid.
The final metric is an older and less sophisticated—yet tried-and-true—method of
measuring water clarity. For this measurement, an 8-inch wide disk, called a Secchi disk, is slowly
lowered into the water until it is no longer visible to the naked eye, at which point the depth of the
disk is recorded. Secchi disk depth is most often measured in meters; we chose to convert to feet
for explanatory purposes.
Using GIS software, each Lee County home sale record in our data set was assigned to the
nearest monitoring point for each of the four metrics. Using these assignments, we produced for
each home sale record the average value of each metric’s measurements over both the full month
and full year leading up to the sale’s contract date. For example, a home sale with a contract date of
July 15, 2013 would first have been assigned the average of all chlorophyll a readings occurring
between June 16, 2013 and July 15, 2013, and then the average of all chlorophyll a readings
v
occurring between July 16, 2012 and July 15, 2013. The process would then be repeated for
dissolved oxygen, turbidity, and Secchi disk depth. It should be noted that because each metric in
our Lee County data was measured only about once per month anyway, the one-month averages
were often simply just the most recently recorded value of a metric. In the few cases where there
were no measurements in the month prior to a home sale’s contract date, the most recent
measurement was substituted, which was never any more than 37 days prior to the contract date.
As for our Martin County models, the St. Lucie River data in FDEP’s STORET database was
inadequate for our purposes, so we instead used weekly dissolved oxygen and Secchi disk
measurements compiled by the Florida Oceanographic Society (FOS). While this prevented us from
being able to estimate the Martin County models for chlorophyll a and NTUs of turbidity, the FOS
data was an upgrade in terms of how frequently it was collected (i.e. weekly instead of monthly).
FOS data was also collected over the full four-year time period for the wide portion of the Indian
River Lagoon in Martin County extending north from the mouth of the St. Lucie River. FDEP data
did turn out to be available for the Loxahatchee River, located in the southern part of the county, so
the full coverage area for our models included all estuarine waters of any significant width located
in and around Martin County.
Model Results and Estimated Effects
The results of the one-month models and one-year models were statistically significant for
all water quality metrics except for dissolved oxygen, which was not significant in either the Lee or
Martin County models. These results are discussed at length in Section 4 of this report.
Key findings of note from the one-month models:
• Three types of water quality measures for Lee County were found to have positive (negative)
impacts on home values when water quality increases (decreases). The robust statistical
significance of these results strongly supports the notion that water quality plays a role in the
determination of nearby home prices.
• Statistically significant results were found for two completely separate counties, another strong
indicator that the water quality does indeed affect home prices in the hypothesized manner.
vi
• Because of the lack of readings occurring more often than about once per month for the Lee
County water data, there is the possibility of attenuation bias. What this means is that it is likely
that the true effects are even stronger than what the model predicted.
Selected estimates from the one-month models are displayed in Table ES.1.
Table ES.1
Marginal price effects of changes in 1-month average Secchi disk depth at select distances
Resulting
Water Quality Change in Standard
Location of Property change in
Measure Water Quality Error
Property Value
Lee County
Caloosahatchee Estuary waterfront Secchi disk depth +1 foot +2.47% 0.41%

1/8 of a mile from the waterfront: Secchi disk depth +1 foot +1.93% 0.32%
1 mile from the waterfront: Secchi disk depth +1 foot +0.34% 0.055%
2 miles from the waterfront: Secchi disk depth +1 foot +0.045% 0.0075%
Martin County
Waterfront on the St. Lucie Estuary,

Loxahatchee Estuary, or Indian River Secchi disk depth +1 foot +5.41% 0.86%
Lagoon north of St. Lucie Inlet
Key findings of note from the one-year models:
• The effects are much larger when we use the one-year averages rather than the one-month
averages. This result indicates while the algal blooms and water discharge events have caused
distress to home prices, for the most part, individual events have not affected homebuyers’
opinions of homes. The stronger results for the one-year averages means that rather, homebuyers
take into account the quality of water over the long term when making their offers. What is
happening is that, while one algal bloom is not alarming in isolation, the recurrence of the algal
vii
blooms on a regular basis is showing up in the one-year models. This regular recurrence is what is
concerning homebuyers and sellers. That is, a one-time event may not have a detrimental effect,
but multiple times is a big problem.
• Secchi disk depth proves to be superior to chlorophyll a, turbidity, and dissolved oxygen in terms of
capturing homebuyers’ and sellers’ perceptions of water quality.
Selected estimates from the one-year models are displayed in Table ES.2.
Table ES.2
Marginal price effects of changes in 1-year average Secchi disk depth at select distances
Resulting
Property Value
Lee County

Martin County

In Section 5 of this report, we apply the above results to countywide property data, yielding the
following estimates:
• Other things equal, we find that a one-foot increase in average Secchi disk depth throughout
the Caloosahatchee Estuary raises Lee County’s aggregate property value by an
estimated $541 million.
viii
the St. Lucie Estuary, Loxahatchee Estuary, and the portion of the Indian River Lagoon north
of the St. Lucie Inlet raises Martin County’s aggregate property value by an estimated
$428 million.
• We find that changes in the water quality of the St. Lucie Estuary, Loxahatchee Estuary, and
the portion of the Indian River Lagoon north of the St. Lucie Inlet—as measured by changes to
one-year average Secchi disk depth at each monitoring point—resulted in an estimated $488
million reduction in Martin County’s aggregate property value between May 1, 2013
and September 1, 2013.
ix
Section 1. Introduction
Much was made of the water releases from Lake Okeechobee into the Caloosahatchee
and St. Lucie rivers in the summer of 2013. These discharges contributed to extremely
unpleasant conditions for anyone on or near either of the two rivers. In Lee County, the
Caloosahatchee River turned into a sort of reddish-brown hue, variously described by locals as
a “tea” or “cola” sort of color. 1 This murky water flowed out from the mouth of the river into
San Carlos Bay, around Sanibel Island, and into the Gulf of Mexico, wildly contrasting with the
clear, vibrant greens and blues of the nearby beach waters. 2 At about the same time, parts of
the St. Lucie River in Martin County were green as well—not a clear, vibrant green, but rather a
thick, slimy green—due to toxic algal blooms.
While these were significant events that garnered a great deal of attention, it was not as
though these were first-time incidents in either river. Episodes such as these have occurred
many times over the last several years. Naturally, there is concern among residents of both
counties not only for the health of their rivers, but also for the local economies that depend on
them.
The waters and beaches of Lee and Martin counties have long supported the local
marine industries and have attracted recreation and tourism dollars. These are clearly the
most vulnerable sectors of each county’s economy when it comes to deteriorating ambient
water quality. 3 But it is not difficult to imagine other sectors of these local economies suffering
ill effects from unsightly and unclean waters, as well.
Many have speculated that real estate is one such sector, and there is some evidence to
support this notion. Many Realtors, for instance, have reported lost sales due to poor water
quality related to discharges from Lake Okeechobee. 4 Stories have also circulated of outraged
tourists vowing never to return after witnessing how bad the water can get, or after being
prevented from even entering the water for the duration of their vacations due to health
1 Spiewak (2013)
2 Linette (2013)
3 The term water quality can be used in reference to the quality of local drinking water, the quality of water used for
industrial purposes, or simply the environmental quality of water bodies present in the area. The term ambient
water quality refers to the latter.
4 See, for example, Gordon (2012).
1
warnings. 5 Yet, while such evidence is compelling, it is also overwhelmingly anecdotal in
nature. Little is truly known about the actual extent to which these discharge-related events
have impacted—and will continue to impact—the real estate markets in Lee and Martin
counties.
It is well established, however, that bodies of water such as lakes, streams, rivers, bays,
and oceans generally have a positive effect on the demand for nearby residential properties,
thereby increasing their value. Economists typically refer to this type of effect as an amenity
effect. In theory, poor water quality will dampen the amenity effect generated by a water body,
resulting in less demand for nearby properties. This reduced demand will hypothetically
manifest in the form of some combination of lower home prices, fewer home sales, and
lengthier times on market for nearby properties.
The purpose of this study is to test the above theory by analyzing the impact of short-
term and long-term ambient estuarine water quality on the sale prices of single family homes
in Martin and Lee counties over a four-year period, from 2010 through 2013. We use hedonic
pricing models to control for the effects of other factors which are known to affect home prices,
thereby allowing for the true impact of water quality to be estimated.
In Section 2, we offer a brief synopsis of the interdependent relationship between Lake
Okeechobee, the Everglades, and the Caloosahatchee and St. Lucie estuaries. In Section 3, we
describe the data and methodology we use in our analysis. We report our results in Section 4,
and in Section 5, we present simple examples of how our results might be applied in the
measurement of countywide impacts. Concluding remarks are offered in Section 6.
5 See, for example, Hecker (2013). Tourists are, of course, a significant source of vacation home sales and sales to
retirees in both counties, as well as elsewhere in Florida.
2
Section 2. Background: Lake Okeechobee, the
Everglades, and the Estuaries
Lake Okeechobee is one of the largest freshwater lakes in the United States, covering
over 660 square miles of south central Florida. Twelve times the size of Florida’s next largest
lake, it appears on even the most basic maps of Florida, and is easily visible from orbit. It is not,
however, easily visible from the ground. To get a glimpse of “Lake O,” one must first ascend to
the top of the 30-foot Herbert Hoover Dike which almost completely encloses it. The dike was
originally built by the U.S. Army Corps of Engineers in the 1930s to replace smaller earthen
dikes that had failed to contain the lake during devastating hurricanes in 1926 and 1928—
catastrophes that resulted in over 2,500 deaths. Following another particularly powerful
hurricane in 1947, it was decided that the dike system should be expanded further.
Construction was completed in the 1960s, and the dike has not been expanded since. 6
For several thousand years leading up to this containment, Lake Okeechobee regularly
spilled over its banks during rainy seasons, sending sheet after sheet of fresh water rolling
south into the Everglades. Today, as Figure 2.1 illustrates, this water is largely rerouted by
canals through the surrounding agricultural lands toward the South Florida metroplex. Some
of these canals were put in place by the Corps of Engineers in order to regulate flows from the
lake, but others had been in place for several decades already, built during early efforts to drain
the northern Everglades to make it suitable for agriculture. Especially of note was a canal built
in the 1890s which linked Lake Okeechobee to the Gulf of Mexico via the previously isolated
Caloosahatchee River. 7 This man-made connection was further solidified by the Corps of
Engineers’ construction of the Okeechobee Waterway in 1937, which also connected the lake to
the St. Lucie River and Atlantic Ocean to the east.
Since the construction of the Okeechobee Waterway, the Army Corps of Engineers has
used discharges into the Caloosahatchee and St. Lucie rivers to lower the water level in Lake
Okeechobee when its height poses a threat to the structural integrity of the aging dike.
Historically, a major problem associated with these releases has been their effect on the
Caloosahatchee and St. Lucie estuaries. Estuaries are, by definition, meeting places between
6 U.S. Army Corps of Engineers (n.d.)

7 Florida Department of Environmental Protection (2011)
3
Image source: evergladesplan.org; modified from original (labels added)
Figure 2.1
Side-by-side maps of water flow through the Everglades ecosystem before human intervention (left) and currently (right)
inland freshwater sources and saltwater from the oceans, and the aquatic life that are suited to
living in these conditions are sensitive to abrupt changes in salinity. Normally, the salinity of
the water in an estuary ebbs and flows along with the tides and any fluctuations in the amount
of freshwater flowing into them (due to, say, heavy rainfall events). When enough water from
Lake Okeechobee is released over a short period of time, it can effectively push out all of the
saline water, causing distress to the aquatic life living within.
More recently, this damage to the estuaries has been amplified by the increasingly high
levels of pollutants that have been collecting in Lake Okeechobee for a number of years. Most
of the water in Lake Okeechobee comes from the Kissimmee River and its watershed, which
begins in the southern suburbs of the Orlando metropolitan area and moves south through a
system of canals surrounded by agricultural lands, collecting residential and agricultural runoff
as it goes. In 2007, the South Florida Water Management District (SFWMD) extracted several
4
layers of muck from the bottom of Lake Okeechobee in order to expose the natural lakebed.
Samples of the extracted muck tested high for a number of pesticides and other contaminants;
levels of arsenic were four times higher than what is allowable for residential land. 8
The high levels of phosphorus and nitrogen found in Lake Okeechobee’s water have
fueled harmful algal blooms throughout the Everglades ecosystem, including the
Caloosahatchee and St. Lucie rivers. Both rivers are already impacted by the phosphorus and
nitrogen from local fertilizer runoff, and the Lake Okeechobee discharges only serve to
intensify the problem. 9
8Reid (2007)
9Interestingly enough, many have also argued that the Caloosahatchee does not get enough water from Lake
Okeechobee during dry seasons, which also damages the local ecosystem. See Lollar (2010, Nov.)
5
Section 3. Methodology and Data Collection
In this section, we outline the design of our hedonic pricing models and describe the
data set we analyze with them. In Subsection 3.1, we introduce the concept of hedonic pricing
models, and demonstrate why they are useful in the analysis of housing markets. In Subsection
3.2, we lay out the methodology and structure of our hedonic pricing models, and provide a
detailed overview of the data set we constructed to test the models.
3.1 Isolating Price Effects in Real Estate Market Analyses
Markets for commodities such as cattle, sugar, copper, or oil are in many ways easier for
economists to analyze than housing markets. What can make housing market analysis tricky is
the fact that no two homes are exactly alike. With commodities, the lowest price usually
determines what sells because price is the only major differentiating factor. By contrast,
potential homebuyers may be willing to pay much more for one home versus another based on
differences in the characteristics of each home. Even identical models of homes cannot be
exactly the same because they cannot occupy the exact same space—any Realtor can tell you
just how important location is when it comes to pricing a home.
To illustrate how this aspect of housing markets is problematic for an analyst, let us
imagine that we wanted to know how much value that homebuyers and sellers place on
waterfront homes in a particular seaside community. A fairly simple approach would be to
calculate the average sale price of recently-sold waterfront homes and compare it to the
average sale price of recently-sold non-waterfront homes. Now, let us suppose we find these
averages to be $300,000 and $200,000, respectively. This result tells us that waterfront homes
are selling for a 50 percent premium over non-waterfront homes.
Yet, while that may be true, we cannot necessarily conclude that the waterfront location
itself is entirely responsible for the 50 percent sale price premium. What if the schools serving
the waterfront areas are more highly rated? What if the waterfront homes are, on average,
much larger than the non-waterfront homes? If so, then part of that 50 percent premium may
be attributable to higher school quality or greater square footage. Based on what we know so
far, we cannot yet discern the true value of the waterfront characteristic of these homes.
6
An economist’s go-to tool for solving this sort of problem is called a hedonic pricing
model. 10 This type of econometric regression model allows us to control for other influential
factors in order to isolate the effects that we want to analyze. To generate a hedonic pricing
model for the simple example above, we would need to collect data on each home sale in the
seaside town over a given period of time. This data would need to include, at a minimum, the
sale price for each home and a “yes/no” indicator of whether each home is on the water. 11
Data must also, however, be collected for the factors for which we would like to control.
In this case, we would therefore need the measured square footage of each home and some sort
of measure that quantifies nearby school quality for each home. Assuming we have this data,
we could then run a hedonic pricing model and obtain an estimate of the value of waterfront vs.
non-waterfront property, holding the other factors constant.
There are, of course, a lot more factors that go into the determination of housing prices
than we are controlling for above. As a consequence, accurate hedonic pricing models typically
require the collection of large data sets containing information about numerous characteristics
of each home and its surroundings.
3.2 Hedonic Pricing Models for Lee and Martin Counties
For the present study, our focus is on identifying the effect of ambient water quality on
the sale prices of nearby single family homes in Lee and Martin counties. In this subsection of
the report, we describe the data set that we constructed for this purpose, beginning with the
sale data and control variables we include in the models. Much of the data we collected or
constructed was geographic in nature, and we relied heavily on geographic information
systems (GIS) software. Because Lee County and Martin County are separate housing markets,
and because there are some differences in the type of data we were able to collect for each
county, we built our data sets (and estimated our models) for the two counties separately.
10Hedonic pricing models were first conceptualized by Sherwin Rosen (1974).

11In econometric models, these “yes/no” indicators are typically quantified as “yes”=1 and “no”=0. There are many
names for these types of indicators, e.g. binary variables, dummy variables, indicator variables, etc. All refer to the
same thing: a numerical representation of a “yes/no” type of characteristic.
7
Table 3.1
Summary statistics for control characteristics and sale prices
Lee County Stats: N = 48,572 Martin County Stats: N = 7,875

Units Mean Std. Dev. Source Mean Std. Dev. Source
Structural Characteristics
Effective age of home years 15.6 (11.9) LCPA 19.0 (11.5) MCPA
Lot acreage acres 0.297 (0.744) LCPA 0.382 (0.851) MCPA
Heated area sq. ft. 1,827 (676) LCPA — — —
Total finished area sq. ft. — — — 2,131 (1,018) MCPA
Bedrooms — 3.07 (0.641) LCPA 3.06 (0.837) MCPA
Bathrooms — 2.18 (0.570) LCPA — — —
Number of stories — 1.12 (0.318) LCPA — — —
Garage — 0.900 (binary) LCPA — — —
Carport — 0.0588 (binary) LCPA — — —
Sea wall — 0.122 (binary) LCPA — — —
Boat dock — 0.0874 (binary) LCPA 0.0801 (binary) MCPA
Swimming pool — 0.390 (binary) LCPA 0.405 (binary) MCPA
Located on golf course — 0.0443 (binary) LCPA 0.106 (binary) MCPA
Sale Characteristics
Short sale — 0.173 (binary) MLS 0.136 (binary) MLS
Foreclosure/REO — 0.270 (binary) MLS 0.0903 (binary) MLS
Neighborhood Characteristics
Pct. of residents under age 18 % 20.9 (10.6) ACS 16.5 (7.33) ACS
Pct. of residents age 65+ % 20.6 (15.7) ACS 30.7 (16.4) ACS
Median household income $10,000s 5.66 (2.00) ACS 6.67 (2.45) ACS
Jurisdictional Characteristics
Millage rate mills 18.9 (2.45) LCPA; LCTC 16.4 (0.472) MCPA; MCTC
Primary School Score score 279 (33.6) MAPX; FDOE 293 (41.7) MAPX; FDOE
Middle School Score score 260 (31.5) MAPX; FDOE 293 (34.9) MAPX; FDOE
High School Score score 246 (24.5) MAPX; FDOE 292 (16.3) MAPX; FDOE
Regional Characteristics
Statewide home price index — 139 (8.88) FDOR 140 (9.39) FDOR
Flood risk characteristics

Risk type X02 — 0.121 (binary) LC; FEMA — — —
Risk type X500 — — — — 0.627 (binary) MC; FEMA
Risk type A — 0.00237 (binary) LC; FEMA 0.00577 (binary) MC; FEMA
Risk type AH — — — — 0.0162 (binary) MC; FEMA
Risk type AE — 0.330 (binary) LC; FEMA 0.141 (binary) MC; FEMA
AE elevation feet 3.18 (5.16) LC; FEMA 1.20 (3.20) MC; FEMA
Risk type VE — 0.00523 (binary) LC; FEMA 0.00715 (binary) MC; FEMA
VE elevation feet 0.0730 (1.02) LC; FEMA 0.0750 (0.886) MC; FEMA
Water Proximity
Bay waterfront — 0.00294 (binary) LCPA — — —
Distance to bay miles 4.39 (2.50) LCPA; SFWMD — — —
Gulf waterfront — 0.00214 (binary) LCPA — — —
Distance to Gulf miles 6.48 (2.47) LCPA; SFWMD — — —
Canal waterfront — 0.116 (binary) LCPA — — —
River (non-estuary) waterfront — 0.00212 (binary) LCPA; SFWMD — — —
Lake waterfront — 0.106 (binary) LCPA — — —
Intracoastal waterway
— — — — 0.0124 (binary) MCPA; SFWMD
waterfront
Distance to I.C.W. miles — — — 4.54 (3.24) MCPA; SFWMD
Ocean waterfront — — — — 0.00426 (binary) MCPA; SFWMD
Distance to ocean miles — — — 4.71 (3.12) MCPA; SFWMD
Estuary waterfront — 0.00334 (binary) LCPA 0.0557 (binary) MCPA; SFWMD
Distance to estuary miles 4.31 (2.88) LCPA 1.46 (1.88) MCPA; SFWMD
Sale price $ 187,226 (245,839) MLS 320,120 (576,771) MLS

Source abbreviations: ACS – 2012 American Community Survey 5-year estimates (U.S. Bureau of the Census); FDOE – Florida Department of
Education; FDOR – Florida Department of Revenue; FEMA – Federal Emergency Management Agency; LC – Lee County Government; LCPA – Lee
County Property Appraiser; LCTC – Lee County Tax Collector; MAPX – School zone boundary GIS shapefiles from Maponics, LLC; MC – Martin
County Government; MCPA – Martin County Property Appraiser; MCTC – Martin County Tax Collector; MLS – Local Multiple Listing Services (see
full text); SFWMD – South Florida Water Management District
8
3.2.1 Sale data
For the Lee County models, we began with a data set comprised of all single family
home listings found in the three multiple listing services (MLSs) serving the area. We did the
same for the three MLSs with listings in Martin County. 12 From there, we narrowed our focus
to the set of all listings which both (a) successfully resulted in a closed sale, and (b) went under
contract for this sale on any date from 2010 through 2013. We then removed any duplicate
listings to ensure that each sale is represented only once in the databases. 13
We primarily chose 2010 as the first year of our four-year analysis period because it
was the earliest year for which we had complete MLS data for either county. Similarly, we
made 2013 the final year of the analysis period because some of the supplemental database
sources we are using were not updated past the end of 2013 at the time we constructed our
core data sets in mid-2014. Our focus on the contract (i.e. “pending”) date rather than the final
closing date for each home is notable because it is the date at which the decision was made to
purchase the home, which is much more relevant to the model than the date upon which the
sale was closed.
3.2.2 Control characteristics
Summary statistics for the control characteristics (and for sale prices) are provided in
Table 3.1. Our list of control variables for the models is rather extensive. We therefore found it
helpful to categorize them as follows:
• Structural characteristics – representing physical features of the property itself
• Sale characteristics – representing circumstances of the home’s sale
• Neighborhood characteristics – representing demographics of residents of the

area immediately surrounding the home
• Jurisdictional characteristics – representing impact of local government services

and public schools
12 The MLSs in Florida are each run by one or more of the state’s 58 local REALTOR® boards and associations. In
partnership with these local boards and associations, Florida REALTORS® receives data feeds from the state’s
MLSs for use in the production of research and statistics.
13 In areas served by multiple MLSs, it is not uncommon for homes to be listed in more than one of them.
9
• Regional characteristics – representing “macro” housing market conditions at a
regional/state level
• Flood risk characteristics – representing risk posed by flooding
• Water proximity – representing the location of the home with respect to various
water bodies
• Time fixed effects – sets of indicator variables—one group for months, another for
years—that respectively control for seasonal and annual effects not accounted for
by the other control variables included in the models
All of the structural characteristic data that we use comes from the offices of the Lee
County Property Appraiser (LCPA) and the Martin County Property Appraiser (MCPA). The
golf course location indicator was only found in the Lee County data we obtained, but we were
able to construct the same indicator for Martin County using a GIS parcel map and property use
code data from MCPA. The Lee County data set contains several additional structural
characteristics that were not available from the Martin County data. We would note that this is
more of a luxury for our Lee County model than it is a detriment to the Martin County model.
The most important structural characteristics—those that typically have the greatest impact on
housing prices in hedonic pricing models—are all present in the Martin County data, including
effective age, lot size, square footage, and presence of a swimming pool.
The sale characteristics we include as controls in the model are drawn from the MLS
listings. They are indicators of whether a home was advertised as a short sale and whether it
was marketed as an REO. These sales were arms-length to the extent that they were openly
marketed in a multiple listing service.
For the neighborhood characteristics, we used GIS software to identify the census block
group in which each home is located by overlapping the latest census block group GIS maps
from the U.S. Bureau of the Census with GIS parcel maps from the county property appraisers’
offices. The block group-level demographic data are from the 2012 American Community
Survey’s 5-year estimates data set. We used the percentage of block group residents under the
age of 18 and age 65 and older to account for the idea that families with young children tend to
prefer to live near other families with young children, and likewise that retirees generally
prefer to live among other retirees. Similarly, block group median household income is
10
included to capture the notion that higher income households are attracted to higher income
neighborhoods. 14
We included each home’s effective ad valorem millage rate (as of the last full year
before its sale) as the lone local government characteristic in our jurisdictional characteristic
category. The school quality scores in this category were constructed using school
performance metrics published each year by the Florida Department of Education. These
metrics include performance scores, by school, for the reading, math, writing, and science
components of the Florida Comprehensive Assessment Test (FCAT). The score for each FCAT
component represents the percentage of students who scored “satisfactory” or higher on the
exams the prior school year. We summed these four scores to arrive at a total score for each
school and assigned these values to each home sale according to school attendance zone
boundary GIS maps compiled a third-party provider, Maponics, LLC. 15 The scores assigned to
each home sale were those of the academic year prior to the sale.
The lone regional characteristic we used is a monthly statewide repeat-sales single
family home price index, which we constructed from home sales reported in the Florida
Department of Revenue’s annual certified county property tax rolls from 1995 through 2014. A
time series graph of the index is displayed in Figure 3.1. We followed the same methodology
that S&P Dow Jones Indices uses to produce its popular S&P/Case-Shiller Home Price Indices. 16
As a control variable, the index captures the ebb and flow of the general Florida housing market
over the four years covered in the study. One example of why we include this variable is that in
2013, the Florida housing market really took off in terms of both sale prices and the overall
volume of sales, to such an extent that the positive effect could likely completely wash out the
detrimental impact that the poor water quality of 2013 had on the home sales included in our
study. Hence, this is a very important control variable in the model.
Flood insurance costs certainly play a role in the determination of Florida home prices,
but unfortunately there is little available data that accurately represents these costs. As an
approximation, we obtained FEMA flood insurance risk GIS maps from both counties, and
14 We included median household income in terms of tens of thousands of dollars in order to make the model
results easier to read; whether dollars are used or tens of thousands of dollars are used in the model has no bearing
on the actual results.
15 The Lee County School District uses a school choice system that allows residents of each attendance zone to
choose from a set list of schools, depending upon availability, rather than requiring attendance at a single school.
To deal with this minor complication, we simply averaged the scores for each eligible school for each home.
16 S&P Dow Jones Indices (2015, Feb.) provides a very detailed description of the methodology.
11
Figure 3.1
A monthly repeat-sales price index for single family homes in Florida. The base month is January of 2000.
attached the risk data to each home under the assumption that these maps play a significant
role in determining flood insurance rates. Each risk category was given its own “yes/no”
indicator variable, and for homes located in zones with higher risk types AE or VE, the
associated estimated flood elevations were included as well. Definitions for the risk type
categories are presented in Appendix A.
Proximity to water plays a major role in the determination of home values. In our
models, we expect there to be an inverse relationship between property values and the
distance from various water bodies. That is, the shorter the distance to a body of water, the
greater the property value, other things constant. For the Lee County model, we examined the
GIS map of water bodies from the South Florida Water Management District (SFWMD) found
within its Arc Hydro Enhanced Database (AHED), and we identified six distinct groupings of
water bodies for which to produce proximity measurements:
12
• Bay – includes Estero Bay, San Carlos Bay, Matlacha Pass, Pine Island Sound, and
other saltwater bodies separated from the Gulf of Mexico by Lee County’s barrier
islands
• Gulf – the Gulf of Mexico
• Canal – predominantly comprised of the canals of Cape Coral
• River – all rivers and large streams except for the portion of the Caloosahatchee
River located downstream from the W.P. Franklin Lock
• Lake – a number of small to mid-size lakes mapped by SFWMD in the AHED water
body map
• Estuary – the portion of the Caloosahatchee River located downstream from the
W.P. Franklin Lock
The above water body groups are color-coded and displayed in Figure 3.2. The group
names for the water bodies are consistent with their classification in AHED. For each single
family home in our Lee County data set, we calculated the direct distance, in miles, to the
nearest location of each of these water body types.
If we were to include these raw distances in the hedonic pricing model, we would be
making an implicit assumption that the positive impact of a water body on nearby property
values declines at a constant rate as we move away from the water body, as depicted in Figure
3.3. This assumption is a bit problematic, because it implies that as we move further away from
the water body, there will eventually be a point at which the water body will begin to negatively
affect the value of the property. In reality, we would just expect the magnitude of the effect to
decline toward zero as we move a greater distance away. In light of this issue, we apply a
mathematical transformation to all of the raw distance measurements before including them in
the model:
𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒𝐼𝑚𝑝𝑎𝑐𝑡 = 𝑒 −2 × 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 . (3.1)
This type of mathematical function is sometimes called an exponential decay function.17

The function appears as the blue curve on the graph shown in Figure 3.4, where the horizontal
17 Equation 3.1 can alternatively be written as RelativeImpact = exp(–2 ⨯ Distance). The operator function exp()
simply means “take e to the power of (value in parentheses).” The number e is a mathematical constant; it is the
base of the natural logarithm and its approximate value is 2.72.
13
14
Original water body map file source: SFWMD AHED
Figure 3.2
Map of Lee County water bodies included in the study to control for water amenity effects.
Figure 3.3
A graph illustrating the modeled relationship between home values and distances from a waterfront when raw distance measurements
are used in a hedonic pricing model.
axis represents the input value (distance in miles) and the vertical axis shows the
corresponding transformed value. 18 The important result from using this function is that we
have converted a raw distance measurement into a measure of relative impact that will
necessarily be greater than 0, but no greater than 1. This value represents the impact the water
body will have on a property value at a given distance relative to when the distance is zero. For
example, a relative impact value of 0.5 means that the impact is about half of what it would be
for a property located on the waterfront. The shape of the curve illustrates that more weight is
given to properties in the immediate vicinity of the water body, whereas a few miles out, the
water body’s influence is negligible.
18 We also considered using –1 instead of –2 as the parameter value in the function (shown as the green curve in
Figure 3.4), but we ultimately judged the rate of decay over distance was not significant enough to capture the true
effect.
15
Figure 3.4
Graphs of exponential functions mapping raw distance measurements into relative impact measurements. The blue curve represents
the particular function used in this study, RelativeImpact = exp(–2 ⨯ Distance).
16
Figure 3.5 shows how the relative value impact of proximity to the Caloosahatchee
Estuary changes over various distance intervals in our model. These intervals are represented
as different colored bands surrounding the estuary. In the legend, the distances characterizing
each interval are displayed alongside their transformed values (which are converted to
percentages for easier interpretation).
To better understand how to interpret the map, let us examine the orange band (i.e. the
third band out from the estuary) as an example. This band covers properties located between a
quarter mile and a half mile from the Caloosahatchee Estuary. The corresponding range of
relative impact values, as shown in the legend, goes from about 37 percent to about 61 percent.
Essentially what this means is that the water’s impact on the value of a property half a mile
away is expected to be about 37 percent of what it would be if the property were located at the
river’s edge. Likewise, the impact on a property located a quarter of a mile away is about 61
percent of what it would be if the property were on the water. Anything located between a
quarter mile and a half mile away will have an intermediate percentage value somewhere
between 37 percent and 61 percent.
There is one other important takeaway from an examination of Figure 3.5: the relative
impact becomes quite negligible as we move beyond a distance of one mile from the river. Only
properties that are within easy walking distance of the river enjoy significant relative impacts
due to water proximity. Keep in mind, also, that these relative impacts by themselves tell us no
more about the actual dollar value associated with proximity to water than the original raw
distance measurements do. They are simply another input into our hedonic pricing model.
One downside to using a distance-based measure (regardless of whether transformed
or not) to capture the effect of water proximity is that it fails to fully capture the added value of
being right on the water. That is, waterfront homes have benefits not shared by non-
waterfront homes located even right across the street. It is very reasonable to assume that
there should be an additional jump in value associated with having markedly easier access to
the water and a great waterfront view. As a result, in addition to our distance-based measure,
we include a waterfront indicator variable in the model for each of the six water body groups.
These indicators were already present in the data we obtained from the Lee County Property
Appraiser’s office. 19
19 The LCPA data did not distinguish between an Estuary waterfront and a River waterfront as we have defined
them, so we used the water body and parcel GIS maps to identify them.
17
18
Figure 3.5
A map depicting various distance-from-waterfront intervals for the Caloosahatchee Estuary. Corresponding transformed values are displayed in the legend.
For our Martin County models, we constructed our distance-based measures in the
same fashion as those for the Lee County models. Our data sources did not include waterfront
indicators, so we constructed them using GIS parcel maps from MCPA and the water body GIS
map from SFWMD’s AHED geodatabase. We then formed three groups of water bodies:
• Intracoastal Waterway – includes all of the Indian River Lagoon in Martin County
that is located south of the St. Lucie Inlet 20
• Ocean – the Atlantic Ocean 21
• Estuary – includes the St. Lucie Estuary, the Loxahatchee Estuary, and all of the
Indian River Lagoon located north of the St. Lucie Inlet
These water body groups are color-coded and displayed in Figure 3.6. There are, of
course, noticeably fewer groups here than we used in the Lee County model. A comparison
between the AHED water body maps and aerial images of Martin County revealed that
numerous lakes and streams were not included in the AHED maps, which prevents us from
accurately representing these categories in the model. 22 We did not include a canal group
simply because of the insignificant number of residential canals located in the county.
The final group of controls we include with the models are commonly referred to as
time-fixed effects in econometrics. We include one set of fixed effects for months, and another
for years, that serve to account for any seasonal or long-term macroeconomic effects that are
not picked up by the other control characteristics.
3.2.3 Measuring Ambient Water Quality
In terms of constructing our models, there are still two key questions that we need to
answer. First, how exactly should we measure water quality? And second, how do we account
for the influence of proximity to the water on the impact of water quality?
20 This portion of the Indian River Lagoon was excluded from the Estuary category because of a lack of consistent
water quality monitoring over all four years of the study.

21 We omit the indicator variable (but retained the transformed distance measure) for the Ocean category from our
model because it is highly collinear with the indicator variable for flood risk type VE. In other words, it provides
what was essentially redundant information and its inclusion would cause complications in the model. Because it
is not our variable of interest, we are not missing anything by excluding it.
22 The AHED coverage for these water body types in Lee County was much more complete.
19
Figure 3.6
Map of Martin County water bodies included in the study to control for water amenity effects.
20
The answer to the first question is more complicated than one might think. There is a
wealth of ambient water quality data out there, but at the same time, the water quality data we
can use in our model must meet some very specific criteria. We face the difficult task of
quantifying the ambient water quality of the estuaries as perceived by potential homebuyers and
sellers. The first criterion, therefore, is that we need to choose types of water quality measures
that represent aspects of the water that are readily apparent to a person inspecting a nearby
property. In Section 2, we noted that nitrogen and phosphorus runoff are key contributors to
algal blooms in the estuaries, and there happens to be a reasonable amount of nitrogen and
phosphorous measurements available in our study areas from various sources. However,
nitrogen and phosphorus counts are actually fairly poor candidates for our analysis, because
their presence does not directly influence the decisions of homebuyers and sellers.
Certainly, nitrogen and phosphorus frequently play a role in causing undesirable
conditions in the estuaries, but a measurement which shows elevated levels of either element
does not guarantee that an algal bloom is present. Whether an algal bloom occurs depends on
more than just the presence of nitrogen and phosphorus—the water also needs to meet other
criteria related to temperature, salinity, and other factors. So, for our purposes, the more
appropriate indicators will directly measure the visible characteristics of the water, because we
aim to represent, as best we can, factors which might impact the opinion of a potential
homebuyer viewing a nearby home. The water quality characteristics in our model are,
therefore, not chosen based solely on the potential damage they can cause to the ecosystem.
Once we find some water quality metrics that meet this standard, the next step is to
ensure that measurements are taken from multiple locations across each estuary. The
Caloosahatchee and St. Lucie estuaries are quite large, so it is unreasonable to assume that
readings from a single location will be representative of conditions throughout the entirety of
each estuary.
Our final criterion is similar to the previous one. In addition to requiring good
geographic coverage for our measurement data, we also require good measurement coverage
over time for each of the monitoring points chosen for inclusion in our models. We are looking
for water quality metrics that are measured somewhat frequently and at regular intervals. We
are, after all, trying to represent constantly evolving conditions across continuous periods of
time with measurements that only give a glimpse of conditions at specific points in time. Hence,
we desire frequent measurements so that we have a better chance of accurately representing
21
perceived water quality in the models. Similarly, we need data that is measured at regular
intervals because ambient water quality are effected by seasonal factors. For example, 20
water quality measurements spread out evenly throughout the year will be much more
representative than 20 water quality measurements clustered together in one part of the year
or another.
With the above criteria in mind, we have selected four different types of ambient water
quality measurements for our Lee County models. These measurements are from the STORET
database maintained by the Florida Department of Environmental Protection (FDEP), with
each type of measurement collected roughly once per month at up to 16 separate monitoring
sites in the Caloosahatchee River between the W. P. Franklin Lock and the mouth of the river at
San Carlos Bay. The locations of the 16 monitoring points are displayed on the map in Figure
3.7, which also includes an inset table denoting which types of measurements were used at
each monitoring point.
The first type of measure we use is micrograms per liter (μg/L) of chlorophyll a in the
water, which captures the extent to which the water is populated by microscopic
photosynthetic organisms that cause discoloration and murkiness in the water. As small as
these organisms may be on an individual basis, their presence is certainly visible in large
amounts. Some of these organisms are harmful to both aquatic life and humans, such as
cyanobacteria (i.e. blue-green algae). 23
The second type of measure we use is milligrams per liter (mg/L) of dissolved oxygen.
A lack of oxygen in the water is associated with the presence of algae-friendly nutrients.
Moreover, if dissolved oxygen levels remain too low for too long, most aquatic life will begin to
die off, which of course can result in unpleasant sights and odors that may impact the
perceptions of potential homebuyers. Like nitrogen and phosphorus, we consider dissolved
oxygen to be a weak approximation of actual visible conditions in the water, but we
nevertheless include it here because it was easily the most frequently measured characteristic
that we came across in the various databases we looked at.
The third measure, turbidity, is a measure of the cloudiness of water due to any sort of
tiny suspended materials, living or non-living. Turbidity is often measured by an instrument
23 We actually analyzed two different chlorophyll a data series from the STORET database: “Chlorophyll a
(corrected for pheophytin)” and “Chlorophyll a (uncorrected for pheophytin).” In terms of the results of our
models, the difference between the two was negligible. Hence, we only report the results for one of them in this
report—Chlorophyll a (corrected for pheophytin).
22
23
Original monitoring point map file source: FDEP
Figure 3.7
Locations of Caloosahatchee Estuary water quality monitoring points used in the study. Inset table identifies which types of measurements were used from each site.
called a nephelometer, which captures its readings with a deflector that measures the way in
which a light beam is deflected by particles suspended in the water. Nephelometer
measurements are reported in Nephelometric Turbidity Units (NTUs). Higher NTUs mean that
the water is more turbid.
The final metric is an older and seemingly less sophisticated—yet tried-and-true—
method of measuring water clarity. For this measurement, an 8-inch wide disk, called a Secchi
disk, is slowly lowered into the water until it is no longer visible to the naked eye, at which
point the depth of the disk is recorded. 24 Secchi disk depth is most often measured in meters;
we chose to convert to feet for explanatory purposes. 25
Using GIS software, each Lee County home sale record in our data set was assigned to
the nearest monitoring point for each of the four metrics. Using these assignments, we
produced for each home sale record the average value of each metric’s measurements over
both the full month and the full year leading up to the sale’s contract date. 26 For example, a
home sale with a contract date of July 15, 2013 would first have been assigned the average of
all chlorophyll a readings occurring between June 16, 2013 and July 15, 2013, and then the
average of all chlorophyll a readings occurring between July 16, 2012 and July 15, 2013. The
process would then be repeated for dissolved oxygen, turbidity, and Secchi disk depth. Because
each metric in our Lee County data was measured only about once per month, the one-month
averages were often simply the most recently recorded value of a metric. In the very few cases
where there were no measurements in the month prior to the contract date for a particular
home sale, the most recent measurement was substituted, which was never any more than 37
days prior to the contract date.
As for our Martin County models, the St. Lucie River data in FDEP’s STORET database
was inadequate for our purposes, so we instead used weekly dissolved oxygen and Secchi disk
24 Secchi disk depth is the sole water quality measurement used in most of the existing published research wherein
hedonic pricing models are used to calculate the effect of water quality on home prices. See, for example, Boyle et
al. (1999), Michael et al. (2000), Boyle and Taylor (2001), and Walsh et al. (2010). Two exceptions are Leggett and
Bockstael (2000), who use fecal coliform bacteria counts, and Poor et al. (2007), who use dissolved inorganic
nitrogen and total suspended solids.
25 In rare instances, the water was clear enough that the Secchi disk hit the bottom of the estuary without
disappearing from view. To resolve these cases, we simply used the reported estuary depth at the monitoring site.
While this approach underestimates the true clarity of the water, its impact is either minimal or result in slightly
more conservative estimates.
26 For this purpose, we collected water quality data for 2009, as well as 2010-2013.
24
measurements compiled by the Florida Oceanographic Society (FOS). 27 Summary statistics for
both the Lee County and the Martin County water quality measures are shown in Table 3.2.
While our use of FOS data prevented us from being able to estimate Martin County
models for chlorophyll a and NTUs of turbidity, the FOS data was an upgrade in terms of how
frequently it was collected (i.e. weekly instead of monthly). FOS data was also collected over
the full five-year time period for the wide portion of the Indian River Lagoon in Martin County
extending north from the mouth of the St. Lucie River. FOS reports its measurements by
numbered zones. The zones used in our model are displayed in Figure 3.8.
At the beginning of this passage we mentioned the need to answer a second question
regarding water proximity and its relation to our water quality measurements. Recall that the
amenity effect generated by a water body declines as the distance from the water body
increases, which is represented in our model by the transformed distance measurements.
Because our theory is that ambient water quality impacts this amenity effect, we multiply the
water quality measurements for each home by the home’s transformed distance measurement
(which still enters into the model separately, as well). This effectively applies a distance-based
discount to the impact of water quality; if we do not apply this discount, we are implicitly
making the erroneous assumption that the effect of water quality is the same across all homes,
regardless of how far away they are from the water body.
Table 3.2
Summary statistics for water quality variables*
Lee County Stats: N = 48,572 Martin County Stats: N = 7,876

Units Mean Std. Dev. Source Mean Std. Dev. Source
Water Quality
Chlorophyll a (1-month avg.) μg/L 8.73 (12.2) FDEP — — —
Chlorophyll a (1-year avg.) μg/L 8.30 (6.57) FDEP — — —
Dissolved O2 (1-month avg.) mg/L 6.99 (1.92) FDEP 5.85 (0.960) FOS; FDEP
Dissolved O2 (1-year avg.) mg/L 6.81 (0.541) FDEP 5.94 (0.550) FOS; FDEP
Turbidity (1-month avg.) NTU 2.55 (2.88) FDEP — — —
Turbidity (1-year avg.) NTU 2.44 (0.901) FDEP — — —
Secchi depth (1-month avg.) feet 4.23 (1.86) FDEP 3.39 (1.20) FOS; FDEP
Secchi depth (1-year avg.) feet 4.28 (0.890) FDEP 3.40 (0.933) FOS; FDEP
*Summary statistics calculated over all home sales and their associated water quality values
Source abbreviations: FDEP – Florida Department of Environmental Protection STORET database, including measurements originally submitted by
the City of Cape Coral, Lee County, the Loxahatchee River District, and the South Florida Water Management District ; FOS – Florida Oceanographic
Society
27 FDEP data did turn out to be available for the Loxahatchee River, located in the southern part of the county, so
the full coverage area for our models included all estuarine waters of any significant width located in and around
Martin County.
25
Figure 3.8
A map of St. Lucie River and Indian River Lagoon zones for which the Florida Oceanographic Society reports water quality
measurements. The map depicts only those zones from which we used measurements; FOS currently monitors three other nearby
zones, as well.
26
Section 4. Model Results
In this section, we reveal the results of our model estimations. In Subsection 4.1, we
give a brief overview of the overall performance of the models. Subsections 4.2 and 4.3 go into
detail about the results of the models which respectively use the one-month and one-year
water quality measures. Finally, in Subsection 4.4, we identify and demonstrate a useful
application for our results: an estimation of water quality’s effect on aggregate countywide
property values.
4.1 Overview of Hedonic Pricing Model Performance
In all, we produced estimates for eight separate Lee County hedonic pricing models.
The only difference between each of these models was the type of water quality measure we
included (i.e. one-month and one-year averages for chlorophyll a, dissolved oxygen, turbidity,
and Secchi disk depth). Because we did not have adequate chlorophyll a or turbidity data for
the St. Lucie Estuary, we ran only four models for Martin County (i.e. one-month and one-year
averages for both dissolved oxygen and Secchi disk depth). The raw estimates and other
output from the models can be found in Appendix B.
All twelve models performed very well in terms of explanatory power. Each of the Lee
County models explained about 88 percent of the variation in sale prices, while the Martin
County models explained about 86 percent of sale price variation. 28 In addition, the estimated
price effects of both the water quality measures and the control characteristics were, for the
most part, strongly significant. 29 Among the water quality measures, only the estimated effects
of dissolved oxygen on home prices failed to exhibit an acceptable level of statistical
significance.
The dissolved oxygen result is not all that surprising since, as we noted in Section 3,
dissolved oxygen levels are not a direct measurement of the appearance of a water body. As a
28 Explanatory power in regression models such as these is measured by a statistic called the coefficient of
determination (usually denoted R2). It measures how well the actual data fit the model.
29 In the context of statistical estimation, the word significance does not refer to importance; rather, it refers to how
reasonably sure we can be (based on the model results) that a characteristic we included in the model does indeed
have a real effect, regardless of how large or small that effect actually is. Our results might yield, for instance, a
larger estimated effect on home sale prices for Characteristic A than it does for Characteristic B—but if the margin
of error associated with our estimate for Characteristic A is huge, then we can’t really make a valid comparison
with Characteristic B and can’t even be sure whether Characteristic A has any effect on home prices at all.
27
predictor or indicator of algal growth, dissolved oxygen can be somewhat fickle. It is sensitive
to changes in pressure and salinity, and it actually increases during the first stage of an algal
bloom during the day when the algae are engaged in photosynthesis. The algae also, however,
respire using oxygen, and even more oxygen is consumed when the short-lived algae organisms
start to die off and begin to decay.
4.2 The Effect of Recent Water Quality on Home Sale Prices
The actual estimates produced by hedonic pricing models are a form of what
economists call marginal effects. A marginal effect is simply the change in one value (in our
case, sale price) in response to a small change in another value (the water quality
measurements). 30 The marginal price effects that we estimated for the one-month averages of
chlorophyll a, turbidity, and Secchi disk depth are displayed in Table 4.1. Because these effects
decline as distance from the estuaries increases, we include estimates for various distances
from the shoreline: on the waterfront, an eighth of a mile, a quarter of a mile, a half mile, one
mile, two miles, and four miles. 31,32 Note that measured improvements in water quality are
indicated by either increases to Secchi disk depth or decreases to chlorophyll a or NTUs of
turbidity. As a result, in Table 4.1, where our intent is to show the marginal price effects of
increasing water quality, we display the estimated effect of a positive change in Secchi disk
depth and a negative change in both chlorophyll a and turbidity.
Note also that the design of our models is such that the results can be inverted; that is,
the change in water quality and change in price associated with each marginal effect listed in
Table 4.1 can both be multiplied by –1 and the result will still be true. For example, in Table 4.1
we indicate that a one-foot increase in Secchi disk depth is associated with an estimated 2.48
percent increase in the sale price of a home on the waterfront of the Caloosahatchee Estuary.
But the inverse also works: if there is a one-foot decrease in Secchi disk depth, then there is an
estimated 2.48 percent decrease in sale price.
30 In our hedonic pricing models, we followed the standard procedure of using the natural logarithm of sale prices
rather than the raw sale prices. When the natural logarithm of sale price is used, the interpretation of the marginal
effects changes from a set dollar amount to a percentage change in sale price.
31 These distances, in feet, are respectively 0 ft; 660 ft; 1,320 ft; 2,640 ft; 5,280 ft; 10,560 ft; and 21,120 ft.
32 Figure 3.5 from the previous section provides a good geographic frame of reference for the Lee County results.
The boundaries of the mapped distance intervals in Figure 3.5 are identical to the distances reported in Table 4.1.
28
Table 4.1
Marginal price effects of changes in 1-month water quality measures at select distances
Resulting
Property Value
Lee County
Caloosahatchee Estuary waterfront Chlorophyll a –1 μg/L +0.46% 0.092%

1/8 of a mile from the waterfront: Chlorophyll a –1 μg/L +0.36% 0.072%
1 mile from the waterfront: Chlorophyll a –1 μg/L +0.062% 0.012%
2 miles from the waterfront: Chlorophyll a –1 μg/L +0.0084% 0.0017%
4 miles from the waterfront: Chlorophyll a –1 μg/L +0.00015% 0.000031%
Caloosahatchee Estuary waterfront Turbidity –1 NTU +1.07% 0.36%

1/8 of a mile from the waterfront: Turbidity –1 NTU +0.84% 0.28%
1 mile from the waterfront: Turbidity –1 NTU +0.15% 0.049%
2 miles from the waterfront: Turbidity –1 NTU +0.020% 0.0067%
4 miles from the waterfront: Turbidity –1 NTU +0.00036% 0.00012%

Martin County

Now, what does Table 4.1 actually tell us? First and foremost, all of the results in the table
support the hypothesis that higher water quality in the month leading up to a sale has a
positive impact on the sale price, regardless of the county or the measurement type. And as
mentioned above, we can likewise infer from these same results that lower water quality
means lower sale prices.
29
The differences between the two counties’ estimated marginal price effects for Secchi
disk depth are also interesting, in that the magnitudes of the estimated effects for Martin
County are more than twice that of their Lee County counterparts. As interesting as this
outcome is, however, it is difficult to draw a solid conclusion from it. This discrepancy could be
due to differences in the innate supply or demand for estuarine water quality in each county,
but there are many other potential reasons for it, including the fact that the Martin County
model has different data sources and includes fewer control characteristics than the Lee
County model.
The Secchi disk measurements themselves serve as an excellent example of how
differences between data sources may give rise to the discrepancy between the results for the
two counties. For each home sale, what we would ideally like to know is what the average
Secchi disk depth at the nearest monitoring point would be over the course of the entire
preceding month, as if it were being measured continuously over time. But Secchi disk depth is
not measured continuously over time, so we have to settle with averaging whatever
measurements were taken at single points in time over the course of the month.
Recall that the Secchi disk measurements for Lee County come from FDEP’s STORET
database, whereas most of the Martin County measurements are derived from data compiled
by FOS. The FDEP measurements are taken approximately once per month in each location,
whereas the measurements compiled by FOS for the St. Lucie Estuary and Indian River Lagoon
are reported on a weekly basis. Therefore, in the month leading up to the sale contract date for
a home in Lee County, we are typically only observing a single Secchi disk measurement, while
in Martin County we are seeing about four measurements over the same amount of time. The
greater number of measurements over the course of a month that we get from the Martin
County data is akin to having a larger sample size in a survey. That is, the one-month Secchi
disk averages that we calculated for the various monitoring locations in Martin County are
statistically more likely to be representative of the typical clarity level exhibited by the water
throughout the entire month.
When one of the variables of interest in a hedonic model is imperfectly measured, it can
cause the estimated magnitude of the variable’s effect to be lower than it really is. Among
statisticians and econometricians, this phenomenon is known as attenuation bias. Perhaps,
then, Lee County’s Secchi disk effect is in reality closer to its counterpart in the Martin County
model than our results indicate. Unfortunately, we cannot know for certain.
30
Be that as it may, the difference between the magnitudes of the two estimates is not an
important result here. What is important, on the other hand, is that Secchi disk depth’s impact
on the sale prices of nearby homes was found to be positive and strongly significant in the
models for both counties, just as we had hypothesized it would be. The fact that this result was
found in models of two different housing markets that were built using different data sources
further verifies the robustness of this result. Simply put, our model clearly shows that ambient
water quality affects housing prices.
Now, what about the differences between the estimated effects for the three
measurement types used in the Lee County models? Can we make valid comparisons between
these results? The answer is “yes,” although this task is difficult to accomplish based on how
the data are presented in Table 4.1 because a one-unit change in NTUs of turbidity, for example,
is not the same thing as a one-foot change in Secchi disk depth. We can get closer to making a
meaningful comparison, however, if we think in terms of percentage changes rather than single
unit changes in the water quality measurements.
In order to do this, we first need to acknowledge that our model results explicitly
describe relationships between unit changes in water quality measurements and percentage
changes in sale prices. These results hold regardless of the initial (i.e. pre-change) values for
the water quality measurements. That is, it doesn’t matter if the current Secchi disk depth is 2
feet or 8 feet; our model results simply indicate that an additional foot of Secchi disk depth will
result in a price increase of 2.48 percent for waterfront homes.
But what if Secchi disk depth increases by 20 percent instead of by one foot? If the
current depth is 2 feet, this will be an increase of 0.4 feet, but if the current depth is 8 feet, a 20
percent increase is a change of 1.6 feet. For waterfront homes, then, the resulting price
changes would be 0.4 × 2.48% = 0.992% and 1.6 × 2.48% = 3.968%, respectively. If we are
going to analyze percentage changes in the water quality measurements, then we need to be
aware that the resulting price effects will change depending on the initial level of water quality.
Our best option, then, is to compare the effects of a percentage change where we
assume the initial value is the average value (across all home sales in a county over all four
years of the study) of each of the one-month water quality measurements. 33 Table 4.2 shows
33 These average measurement values are displayed in Table 3.2 in the previous section.
31
Table 4.2
Marginal effects of a 20% improvement over the mean value of the 1-month water quality measures
Change in
Equivalent Resulting
Water Quality Water Quality
Location of Property Change in Change in
Measure from Average
Units Property Value
Value
Lee County
Caloosahatchee Estuary waterfront Chlorophyll a –20% –1.75 μg/L +0.80%

1/8 of a mile from the waterfront: Chlorophyll a –20% –1.75 μg/L +0.62%
1 mile from the waterfront: Chlorophyll a –20% –1.75 μg/L +0.11%
2 miles from the waterfront: Chlorophyll a –20% –1.75 μg/L +0.015%
Caloosahatchee Estuary waterfront Turbidity –20% –0.51 NTU +0.55%

1/8 of a mile from the waterfront: Turbidity –20% –0.51 NTU +0.43%
1 mile from the waterfront: Turbidity –20% –0.51 NTU +0.074%
2 miles from the waterfront: Turbidity –20% –0.51 NTU +0.010%
Caloosahatchee Estuary waterfront Secchi disk depth +20% +0.85 feet +2.09%
1/8 of a mile from the waterfront: Secchi disk depth +20% +0.85 feet +1.63%
1 mile from the waterfront: Secchi disk depth +20% +0.85 feet +0.28%
2 miles from the waterfront: Secchi disk depth +20% +0.85 feet +0.038%
Martin County

Loxahatchee Estuary, or Indian River Secchi disk depth +20% +0.68 feet +3.69%
the estimated impact of a 20 percent change in these average values. 34 Clearly, the strongest
effects are associated with increases in Secchi disk depth. Why is this the case?
For one, the diagnostic tests we ran on our models tell us as much. The Secchi disk
measurements were easily the most statistically significant water quality measures in our
34 Our choice of 20 percent here is arbitrary; the choice of percentage is actually irrelevant for this type of
comparison—what is important is that the same percentage is used for all three measurement types.
32
models. 35 This could very well be due to the ability of Secchi disk depth to capture the essence
of perceived water quality in ways that chlorophyll a and turbidity cannot.
The relative weakness of chlorophyll a as a measure is that it ignores inorganic
impediments to water clarity, whereas Secchi disks and turbidity meters more directly
measure clarity itself. The fact that chlorophyll a is still statistically significant in our models is
possibly a testament to the very unpleasant nature—and frequency—of algal blooms occurring
in the Caloosahatchee in recent years.
As for turbidity, the source of its underperformance in our model might actually be
something that is more generally considered to be one of its advantages as a measure of water
quality, relative to Secchi disks. For limnologists and other water quality experts, Secchi disks
have their drawbacks. The readings can be affected by a number of factors such as choppy
water or the amount of sunlight hitting the water. If we imagine two rivers that are identical in
every way except that one is measured on a windy, cloudy day and the other is measured on a
calm, sunny day, we might not be surprised to find a difference in the recorded Secchi disk
depths. Nephelometers, on the other hand, produce their own consistent source of light. For
many scientific applications, this might be considered an advantage. 36 But in our case, we are
trying to measure water clarity as perceived by someone who is looking at it from above—and
this person is subject to the same environmental effects as the person taking the Secchi disk
measurements.
There are other potential explanations for the greater impact of Secchi disk depth
versus chlorophyll a and turbidity, as well. As we show in Figure 3.7 in the previous section,
our Lee County data set includes turbidity measurements taken from 4 different monitoring
sites and chlorophyll a measurements from 6 different sites. In contrast, our Secchi disk depths
are drawn from 15 different sites within the Caloosahatchee Estuary, so the Secchi disk
measurement that we attach to any particular home sale is much more likely to come from a
nearby station, thus better representing water conditions in the home’s immediate vicinity.
This leaves our estimated Secchi disk depth effects less susceptible to attenuation bias
compared to the other two measurement types.
35See the t-statistics reported in the model output tables in Appendix B.

36There are several disadvantages associated with nephelometry, as well. First, nephelometers are not the only
instruments used to measure turbidity; other instruments are sometimes used and their results are converted—
imperfectly—into NTUs. Additionally, some turbidity meters use infrared light while others use white light, which
can lead to different readings. Lastly, nephelometers and other turbidity meters produced by different
manufacturers often fail to produce the identical results.
33
As before, though, we would point out that the exact differences between the
magnitudes of the estimated effects of the three measurement types is not as important as the
notion that they all were found to be highly significant with the hypothesized signs (i.e. positive
for Secchi disk depth, negative for chlorophyll a and turbidity).
4.3 The Effect of Long-Term Water Quality on Home Sale Prices
We turn now to the results for the long-term models, which feature nearby water
quality measurements averaged over a full year leading up to each home sale. The marginal
price effects for unit changes in one-year average water quality measurements are reported in
Table 4.3. In Table 4.4, we present the estimated price changes associated with a 20 percent
increase in the value of a home facing average water quality conditions for the county in which
it is located. What is immediately quite evident from these tables is that the marginal effects
for these models are appreciably greater in magnitude than those estimated using only the
average over the most recent month—especially in the case of Secchi disk depth in Lee
County. 37 Unlike most of the one-month measures for Lee County, the one-year measures are
constructed from multiple observations over time, which means they are not nearly as prone to
measurement error and potential attenuation bias. More importantly, the results strongly
suggest that long-run ambient water quality plays a more significant role in determining home
sale prices than the water quality conditions immediately prior to the contract date.
This is not to say that the events surrounding, for instance, the Lake Okeechobee
releases in the summer of 2013 did not impact the local housing markets in Lee and Martin
counties. They likely did—but they did so as part of a long-term, ongoing sequence that was
already established and well known to most market participants. In other words, if the
summer of 2013 had been the first time in recent history that either estuary experienced an
acute decline in ambient water quality following discharges from Lake Okeechobee, then the
level of public outrage and enduring concern for the health of the local economies would likely
have been considerably less. But this was not the case. The events of 2013 were the latest
incidents in a chain that goes back for decades.
37 The t-statistics for the one-year models indicate a substantial improvement in statistical significance, as well. See
Appendix B.
34
Table 4.3
Marginal price effects of changes in 1-year water quality measures at select distances
Resulting
Property Value
Lee County
Caloosahatchee Estuary waterfront Chlorophyll a -1 μg/L +1.03% 0.16%

1/8 of a mile from the waterfront: Chlorophyll a -1 μg/L +0.80% 0.12%
1 mile from the waterfront: Chlorophyll a -1 μg/L +0.14% 0.021%
2 miles from the waterfront: Chlorophyll a -1 μg/L +0.019% 0.0029%
4 miles from the waterfront: Chlorophyll a -1 μg/L +0.00034% 0.000053%
Caloosahatchee Estuary waterfront Turbidity -1 NTU +7.94% 0.91%

1/8 of a mile from the waterfront: Turbidity -1 NTU +6.18% 0.71%
1 mile from the waterfront: Turbidity -1 NTU +1.07% 0.12%
2 miles from the waterfront: Turbidity -1 NTU +1.45% 0.017%
4 miles from the waterfront: Turbidity -1 NTU +0.0027% 0.0030%

Martin County

Prospective homebuyers who already live in these counties or who have spent a lot of
time there are probably well aware of the history behind these events. Their willingness to pay
might be somewhat influenced by the current water conditions, but if they have lived in the
area long enough they probably know how the water looks under better conditions. Their
35
Table 4.4
Marginal effects of a 20% improvement over the mean value of the 1-year water quality measures
Change in
Equivalent Resulting
Water Quality Water Quality
Location of Property Change in Change in
Measure from Average
Units Property Value
Value
Lee County
Caloosahatchee Estuary waterfront Chlorophyll a –20% –1.66 μg/L +1.70%

1 mile from the waterfront: Chlorophyll a –20% –1.66 μg/L +0.23%
Caloosahatchee Estuary waterfront Turbidity –20% –0.49 NTU +3.88%

1 mile from the waterfront: Turbidity –20% –0.49 NTU +0.52%
Caloosahatchee Estuary waterfront Secchi disk depth +20% +0.86 feet +12.55%
Martin County

Loxahatchee Estuary, or Indian River Secchi disk depth +20% +0.68 feet 7.02%
1/8 of a mile from the waterfront: Secchi disk depth +20% +0.68 feet 5.47%
1 mile from the waterfront: Secchi disk depth +20% +0.68 feet 0.95%
2 miles from the waterfront: Secchi disk depth +20% +0.68 feet 0.13%
4 miles from the waterfront: Secchi disk depth +20% +0.68 feet 0.0024%
home purchase decisions are going to be based less, then, on whether an event is currently
happening, and more on their knowledge of how often these sorts of events occur. The one-
month models we estimated have short memories compared to the one-year models, so they
fail to capture any aspect of this important part of the decision process.
The important result here is that the true danger to the housing markets in both
counties is not from the immediate effects of algal blooms or other sorts of unpleasant
36
conditions caused by releases from Lake Okeechobee and local runoff. Certainly, there is the
potential for prospective homebuyers to back out of sales or for sellers of near- or on-
waterfront homes to keep holding on to their properties for a few extra weeks. For these
people, the delay in the sale of their home could cause economic hardship or undue stress. But
the fact remains that these homes will eventually sell, and our one-month models indicate that
these short-term price effects subside as soon as these events clear up. 38
The greater direct danger to home values in these markets is the continuing recurrence
of these types of events, and the increasing frequency with which they are occurring. This
effect is longer lasting and does not go away when the algal blooms and murky waters clear up
after each incident. In other words, while the short-term price effect might subside, the larger,
long-term effect does not.
38 We attempted to estimate the effects of water quality on the length of time homes were on the market, but our
models yielded highly inconclusive results. According to Johnson et al. (2007), this is not an unusual result in the
economics literature. Economists have had considerable trouble building models for market duration, which
relative to sale price is much more dependent on unobservable characteristics of buyers and sellers. We also found
little evidence that the number of sale contracts for homes near the water is dependent on nearby water quality.
37
Section 5. Applying the Results:
Aggregate Countywide Effects
One of the motivating factors behind our decision to go forward with this study was that
we anticipated our models would provide concrete results that could supplement future
research on the water quality’s aggregate impact on property values in an area. Such research
would be of great benefit to both policymakers and the general public.
We suspect that very little research currently exists on this matter simply because the
value placed on water quality at the micro (i.e. individual) level has not been well quantified.
As far as we can tell, our study is the first to provide such quantitative measurements for the
estuaries of Lee and Martin counties, and only the second to use hedonic methods to provide
quantitative measurements of the impact of water quality on real estate throughout any part of
Florida. 39
In this section we offer a simple, intuitive approach to measuring the aggregate
property value impact of changes in average estuarine water quality in Lee and Martin
counties, based on our estimates from Section 4. We follow our description of this approach
with two separate examples of its application.
5.1 A Simple Approach
Our approach relies on the just values of single family homes, as reported in the 2013
FDOR-certified tax rolls for each county. The just value of a property is simply the county
property appraiser’s estimate of the market value for the property. It is estimated and tracked
by Florida county property appraisers because it forms the basis for determining a property’s
assessed and taxable values.
Apart from just values, the only other property-level data our approach requires is the
distance of each single family home in each county to the nearest estuary shoreline. As before,
we use GIS parcel maps of the two counties in conjunction with the AHED water body maps
from SFWMD to calculate these distances.
39 Walsh, Milon, and Scrogin (2011) use hedonic pricing models to determine the effect of lake water quality on
nearby homes in the Orlando area.
38
The basic premise of our approach is that our previously estimated marginal effects of
water quality on price can also be interpreted as estimated marginal effects of water quality on
just values, since just value is itself an estimate of the market price a property would fetch if it
were on the market. The just values of homes rise when they, or similar nearby homes, sell for
higher prices. Most anything that affects the sale prices of homes essentially affects the
property value of nearby homes that are not currently on the market.
Using this logic in conjunction with our model results from Section 4, we would
estimate that the change in a home’s value resulting from a change in the nearest monitoring
point’s average measured Secchi disk depth over the past year is:
∆𝑉𝑎𝑙𝑢𝑒 = 𝐽𝑢𝑠𝑡𝑉𝑎𝑙𝑢𝑒 × ∆𝑆𝑒𝑐𝑐ℎ𝑖 × 𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡 (5.1)
where ΔValue is the change (in USD) in the home’s market value, JustValue is the pre-change
market value of the home as determined by the county property appraiser’s office, ΔSecchi is
the change (in feet) in the average Secchi disk depth over the past year, and MarginalEffect is
the estimated marginal effect of a one-foot change in the average Secchi disk depth over the
past year.
The value of MarginalEffect varies based on the distance of the home to the nearest
estuary waterfront. It is equivalent to the marginal effect for a waterfront property multiplied
by the distance-based RelativeImpact value we described in Section 3. That is,
𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡 = 𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡0 × 𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒𝐼𝑚𝑝𝑎𝑐𝑡, (5.2)
where MarginalEffect0 is the estimated marginal effect of a one-foot change in average Secchi
disk depth over the past year on the value of a waterfront home. MarginalEffect0 is estimated
by our models and can be found in Table 4.3 in the previous section. 40 We know how to
calculate RelativeImpact from Equation 3.1 in Section 3. This yields the following equation:
𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡 = 𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡0 × 𝑒 −2 × 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 . (5.3)
40 In Table 4.3, these values have been rounded to the nearest hundredth of a percentage point. For this analysis,
we use the raw full output from our statistical software package since it offers more precision: 14.66204% for Lee
County and 10.32006% for Martin County.
39
If we substitute the right-hand side of Equation 5.3 for MarginalEffect in Equation 5.1, we get
∆𝑉𝑎𝑙𝑢𝑒 = 𝐽𝑢𝑠𝑡𝑉𝑎𝑙𝑢𝑒 × ∆𝑆𝑒𝑐𝑐ℎ𝑖 × 𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡0 × 𝑒 −2 × 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 . (5.4)
In Equation 5.4, only the values JustValue and Distance are specific to each property, and
they are predetermined by the tax rolls and GIS measurements, respectively. MarginalEffect0
only differs by county, not by property. To calculate an aggregate countywide effect, we assign
a value for ΔSecchi to each Secchi disk monitoring point in the county, use Equation 5.4 to
calculate ΔValue for each individual single family home in the county, and then simply sum the
results.
5.2 Example: A Uniform Change in Water Quality
For our first example, we assume each water quality monitoring point in each county
has experienced a one-foot increase in its average Secchi disk depth measurements over the
past year. 41 Under this scenario, ∆𝑆𝑒𝑐𝑐ℎ𝑖 = 1 across all monitoring points, which simplifies
Equation 5.4 to
∆𝑉𝑎𝑙𝑢𝑒 = 𝐽𝑢𝑠𝑡𝑉𝑎𝑙𝑢𝑒 × 𝑀𝑎𝑟𝑔𝑖𝑛𝑎𝑙𝐸𝑓𝑓𝑒𝑐𝑡0 × 𝑒 −2 × 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒 . (5.5)
For our calculations, we use the single family home just values reported in the 2013 FDOR-
certified tax rolls for Lee and Martin counties. The results are as follows:
the Caloosahatchee Estuary raises Lee County’s aggregate property value by an
estimated $541 million.
the St. Lucie Estuary, Loxahatchee Estuary, and the portion of the Indian River Lagoon north
41 We use Secchi disk depth for our analysis here because it is available for both counties and we have established
that it is a superior measure of perceived water quality relative to chlorophyll a and turbidity. Additionally, it is
much easier for to visualize a one-foot change in Secchi disk depth than it is a one-μg/L change in chlorophyll a or a
one-NTU change in turbidity.
40
of the St. Lucie Inlet raises Martin County’s aggregate property value by an estimated
$428 million.
Figure 5.1 serves to illustrate that the impact of water quality on property values is not
limited to the waterfront. It shows Martin County’s single family home parcels, color-coded
according to their distances from estuary shorelines. Red parcels include waterfront and near-
waterfront properties, up to the distance at which 25 percent of the $428 million of estimated
added value has been accounted for. Yellow parcels account for the next quarter of added
value, so combined, the red and yellow parcels account for half of the $428 million estimate.
Cyan-colored parcels account for the third quarter of generated aggregate value, and the
remaining parcels, colored purple, account for the final 25 percent.
We should note that the estimated changes to each county’s aggregate property value
are entirely due to estimated increases in the value of each county’s stock of single family home
properties. Since we made no attempts in our study to estimate the effects of improved water
quality on other types of properties, we made no assumption that the values of these properties
would increase. In reality, given that the single family home effects we estimated were so
strong and significant, it is probably safe to assume that improved water quality will at the very
least have similar effects on values of condominiums and other residential properties. So in
this respect, our estimates may be a bit conservative. That said, in the 2013 tax roll data we are
using, single family homes account for 55 percent of total just value in Lee County and 61
percent of total just value in Martin County.
As we mentioned previously in Section 4, our model results are applicable to both
increases and decreases in measured water quality. Just as a one-foot increase in the average
annual Secchi disk depth at each monitoring station respectively leads to $541 million and
$428 million improvements in the aggregate market value of single family homes in Lee and
Martin counties, a one-foot decrease in average annual Secchi disk depth results would—other
things constant—be associated with estimated losses of $541 million and $428 million.
5.3 Example: Martin County in the Summer of 2013
In the summer of 2013, major water releases from Lake Okeechobee into the St. Lucie
River resulted in a significant decline in observed water quality throughout the St. Lucie
41
Original parcel map source: MCPA
Figure 5.1
Map of Martin County single family home parcels, color-coded according to distance from the “Estuary” class of water bodies as defined
in Section 3.2. Each color accounts for one quarter of the aggregate change in Martin County property values resulting from a one-foot
change in average Secchi disk depth in the St. Lucie Estuary, Loxahatchee Estuary, and the Indian River Lagoon (north of the St. Lucie
Inlet).
42
Estuary and adjoining portions of the Indian River Lagoon. Because water quality data was
collected over this time period, we can produce a more precise estimate of the impact to Martin
County’s single family home values. For this estimate, we use Equation 5.4 because we know
how much the one-year average Secchi disk depth changed at each Martin County monitoring
point over the summer. That is, we do not need to make any assumptions about the values of
ΔSecchi in Equation 5.4 (as we did in the previous example, where we assumed a uniform one-
foot change across all monitoring points) because in this case, these values were actually
measured.
In this example, we specifically define ΔSecchi for each monitoring point as the change
(in feet) in the one-year average Secchi disk depth between May 1, 2013 and September 1,
2013. After calculating Equation 5.4 for each single family home in Martin County and
summing the results, we arrive at the following estimate:
• We find that changes in the water quality of the St. Lucie Estuary, Loxahatchee Estuary, and
the portion of the Indian River Lagoon north of the St. Lucie Inlet—as measured by changes to
one-year average Secchi disk depth at each monitoring point—resulted in an estimated $488
million reduction in Martin County’s aggregate property value between May 1, 2013
and September 1, 2013.
As in the previous example, this assessment is derived entirely from estimated changes
in the value of Martin County’s stock of single family homes; we did not estimate changes in the
values of other property types. We would also emphasize that our result does not imply that
there was an overall reduction of $488 million in Martin County’s aggregate property value
between May 1 and September 1. Water quality is not, after all, the only factor that affected
home prices over that time span. In fact, 2013 was actually a banner year for real estate in
Florida, marked by a significant statewide recovery in home prices, so it is quite possible that
Martin County’s aggregate property value increased during the summer. 42 Our result simply
reflects the notion that any overall recovery of property values was significantly dampened by
poor water quality, and would have been much stronger otherwise.
42 In the price index shown in Figure 3.1 in Section 3, this increase in the statewide price level for single family
homes is clearly evident.
43
Section 6. Concluding Remarks
In this study, we compiled an extensive database of water quality measurements and

control characteristics, which we used to produce robust estimates of the effect of water
quality on single family home prices in two metropolitan Florida counties. The statistical
significance, direction, and magnitude of these estimates leave little doubt that water quality
plays an important role in the determination of home sale prices. The results also corroborate
those of the only similar Florida-based study, published by Walsh, Milon, and Scrogin in 2011.
Our methodology and our results for these models (or similar models) can be used as a
building block for future studies on the effect of water quality on housing prices elsewhere in
Florida, although this is largely dependent on whether the appropriate water quality data are
available. The lack of water quality metrics that are tracked or reported consistently and
frequently over time and across multiple locations in various markets is the major limiting
factor here. That is, in order to further our understanding of water quality’s effect on real
estate markets elsewhere in the state, more comprehensive monitoring will be required.
The examples we provide in Section 5 are rather simple; future research may provide
more precise aggregate estimates through more complex, nuanced approaches. As we noted,
our models and our approach focuses solely on single family homes. Effects on other property
types should be explored, as well. Our own aggregate estimates show, at the very least, that
these effects are significant in both size and scope.
Policy makers and the public would also benefit from research into the possible effects
of Everglades restoration on water quality in the estuaries of Martin and Lee counties. If such
estimates could be obtained, the future benefit to the real estate markets in these counties
could be estimated through our methodology, as well.
Finally, we should note that our estimates merely reflect the impact of water quality on
home prices near the water, and we suspect that the estimates we have produced are merely a
drop in the bucket compared to the overall economic impact of water quality on the leisure,
recreation, and marine industries in these counties.
44
Appendix A
Table A.1
Description of Lee County flood insurance risk map zone types
Zone Name Zone Description
An area of minimal flood hazard that is determined to be outside the Special Flood Hazard Area and higher than the
X
elevation of the 0.2-percent-annual-chance (or 500-year) flood.
An area of minimal flood hazard that is determined to be outside the Special Flood Hazard Area between the limits of
X02
the base flood and the 0.2-percent-annual-chance (or 500-year) flood.
Areas subject to inundation by the 1-percent-annual-chance flood event generally determined using approximate
methodologies. Because detailed hydraulic analyses have not been performed, no Base Flood Elevations (BFEs) or flood
A
depths are shown. Mandatory flood insurance purchase requirements and floodplain management standards apply. (In
unincorporated Lee County, the county has estimated BFEs in these areas.)
Areas subject to inundation by the 1-percent-annual chance-flood event determined by detailed methods. Base Flood
AE Elevations (BFEs) are shown. Mandatory flood insurance purchase requirements and floodplain management standards
apply.
Areas subject to inundation by the 1-percent-annual-chance flood event with additional hazards due to storm-induced
VE velocity wave action. Base Flood Elevations (BFEs) derived from detailed hydraulic analyses are shown. Mandatory
flood insurance purchase requirements and floodplain management standards apply.
Source: Lee County

All properties in the study fell into one of these categories, so Zone X was omitted as an indicator variable in the models
Table A.2
Description of Martin County flood insurance risk map zone types
Zone Name Zone Description
Area of minimal flood hazard, usually depicted on FIRMs as above the 500-year flood level. Zone C may have ponding
X and local drainage problems that don't warrant a detailed study or designation as base floodplain. Zone X is the area
determined to be outside the 500-year flood and protected by levee from 100-year flood.
Area of moderate flood hazard, usually the area between the limits of the 100-year and 500-year floods. B Zones are also
X500 used to designate base floodplains of lesser hazards, such as areas protected by levees from 100-year flood, or shallow
flooding areas with average depths of less than one foot or drainage areas less than 1 square mile.
Areas with a 1% annual chance of flooding and a 26% chance of flooding over the life of a 30-year mortgage. Because
A
detailed analyses are not performed for such areas; no depths or base flood elevations are shown within these zones.
Areas with a 1% or greater annual annual chance of flooding and a 26% chance of flooding over the life of a 30-year
AE
mortgage. The base floodplain where base flood elevations are provided.
Areas with a 1% annual chance of shallow flooding, usually in the form of a pond, with an average depth ranging from 1
AH to 3 feet. These areas have a 26% chance of flooding over the life of a 30-year mortgage. Base flood elevations derived
from detailed analyses are shown at selected intervals within these zones.
Coastal areas with a 1% or greater chance of flooding and an additional hazard associated with storm waves. These
VE areas have a 26% chance of flooding over the life of a 30-year mortgage. Base flood elevations derived from detailed
analyses are shown at selected intervals within these zones.
Source: Martin County

All properties in the study fell into one of these categories, so Zone X was omitted as an indicator variable in the models
45
Appendix B
Table B.1
Results of 1-month and 1-year regressions for Martin County
(1) (2) (3) (4)

Effective age of home –0.0194** [–34.30] –0.0194** [–34.28] –0.0195** [–34.23] –0.0196** [–34.13]
ln(lot acreage) 0.177** [28.95] 0.177** [28.97] 0.178** [29.11] 0.178** [29.20]
Total finished area 2.51E–04** [21.67] 2.51E–04** [21.68] 2.51E–04** [21.71] 2.51E–04** [21.91]
Bedrooms 0.0522** [6.52] 0.0524** [6.53] 0.0516** [6.47] 0.0505** [6.37]
Boat dock 0.343** [12.79] 0.344** [12.81] 0.348** [12.84] 0.351** [13.08]
Swimming pool 0.191** [20.18] 0.191** [20.06] 0.190** [20.07] 0.187** [19.89]
Located on golf course 0.114** [9.62] 0.113** [9.61] 0.115** [9.78] 0.117** [9.97]
Short sale –0.197** [–20.54] –0.198** [–20.63] –0.196** [–20.37] –0.195** [–20.32]
Foreclosure/REO –0.349** [–23.40] –0.350** [–23.44] –0.349** [–23.43] –0.349** [–23.32]
Pct. of residents 18 & under –0.0124** [–9.81] –0.0123** [–9.72] –0.0125** [–9.89] –0.0122** [–9.68]
Pct. of residents age 65+ 7.99E–04 [1.59] 7.92E–04 [1.57] 7.11E–04 [1.41] 6.95E–04 [1.37]
Median household income 0.0254** [12.01] 0.0252** [11.78] 0.0250** [11.86] 0.0245** [11.51]
Millage rate 0.0474** [3.39] 0.0470** [3.36] 0.0660** [4.76] 0.0835** [6.01]
Primary School Score 1.96E–04 [1.10] 1.84E–04 [1.03] 2.07E–04 [1.17] 2.31E–04 [1.31]
Middle School Score 3.54E–04 [1.68] 3.69E–04 [1.74] 0.00506* [2.37] 5.91E–04** [2.75]
High School Score 0.00112** [3.40] 0.00111** [3.37] 6.73E–04* [2.05] 2.50E–04 [0.76]
Statewide home price index 0.00609** [5.57] 0.00632** [5.80] 0.00721** [6.66] 0.00659** [6.12]

Risk type X500 –0.00632 [–0.49] –0.00667 [–0.52] 0.00680 [0.53] 0.0222 [1.72]
Risk type A 0.144** [5.27] 0.145** [5.29] 0.159** [5.78] 0.173** [6.29]
Risk type AH 0.0441* [2.00] 0.0449* [2.04] 0.0646** [2.92] 0.0873** [3.94]
Risk type AE –0.0143 [–0.47] –0.0171 [–0.56] –0.0175 [–0.58] –0.0106 [–0.35]
Risk type AE × AE elevation 0.00104 [0.40] 0.00126 [0.49] 0.00181 [0.71] 0.00207 [0.81]
Risk type VE –2.261 [–1.48] –2.247 [–1.49] –2.33 [–1.70] –2.537 [–1.89]
Risk type VE × VE elevation 0.238 [1.58] 0.237 [1.59] 0.243 [1.79] 0.262 [1.96]
Water Proximity and Quality

Intracoastal waterway
waterfront 0.222** [4.03] 0.224** [4.05] 0.217** [3.95] 0.207** [3.79]
exp(–2 × distance to I.C.W.) 0.245** [6.31] 0.244** [6.28] 0.266** [6.82] 0.284** [7.29]
exp(–2 × distance to ocean) 0.763** [6.88] 0.766** [6.93] 0.737** [6.68] 0.714** [6.52]
Estuary waterfront 0.350** [10.57] 0.348** [10.52] 0.341** [10.18] 0.332** [10.02]
exp(–2 × distance to estuary) 0.199** [3.11] 0.311* [2.53] –0.0978** [–2.72] –0.267** [–5.96]
exp(–2 × distance to estuary)
× dissolved O2 (1–month) –0.0200 [–1.80] — — — — — —
× dissolved O2 (1–year) — — –0.0387 [–1.85] — — — —
× Secchi depth (1–month) — — — — 0.0541** [6.26] — —
× Secchi depth (1–year) — — — — — — 0.103** [9.06]
Constant 9.994** [34.56] 9.974** [34.68] 9.615** [33.92] 9.510** [33.65]

N 7,975 7,975 7,975 7,975
R-squared 0.8566 0.8566 0.8576 0.8587
Dependent variable in all models is ln(sale price); estimation by ordinary least squares with heteroskedasticity-robust clustered standard errors
Month and year fixed effects omitted from table; available upon request
t-statistics in brackets
* p < 0.05, ** p < 0.01
46
Table B.2
Results of 1-month regressions for Lee County
(1) (2) (3) (4)

ln(lot acreage) 0.0475** [10.59] 0.0471** [10.50] 0.0472** [10.52] 0.0471** [10.51]
Heated area 3.88E–04** [32.68] 3.88E–04** [32.79] 3.88E–04** [32.71] 3.88E–04** [32.74]
Bedrooms –0.0264** [–5.14] –0.0264** [–5.14] –0.0264** [–5.14] –0.0263** [–5.12]
Bathrooms 0.0753** [13.85] 0.0757** [13.92] 0.0756** [13.89] 0.0755** [13.91]
Number of stories –0.0823** [–10.25] –0.0826** [–10.30] –0.0826** [–10.28] –0.0827** [–10.31]
Garage 0.268** [32.40] 0.268** [32.40] 0.268** [32.40] 0.267** [32.28]
Carport 0.0222* [2.19] 0.0219* [2.16] 0.0220* [2.17] 0.0224* [2.22]
Sea wall 0.158** [10.65] 0.159** [10.76] 0.159** [10.70] 0.158** [10.60]
Boat dock 0.115** [12.45] 0.115** [12.45] 0.115** [12.44] 0.115** [12.44]
Swimming pool 0.258** [60.80] 0.258** [60.87] 0.258** [60.83] 0.258** [60.83]
Short sale –0.270** [–72.83] –0.270** [–72.80] –0.270** [–72.84] –0.270** [–72.79]
Foreclosure/REO –0.321** [–94.00] –0.321** [–93.94] –0.321** [–93.97] –0.321** [–93.99]
Pct. of residents age 65+ 0.00363** [20.09] 0.00365** [20.22] 0.00364** [20.18] 0.00363** [20.11]
Millage rate –0.0230** [–21.88] –0.0230** [–21.89] –0.0230** [–21.92] –0.0229** [–21.79]
Primary School Score 0.00211** [10.45] 0.00213** [10.52] 0.00211** [10.45] 0.00210** [10.38]
Middle School Score 0.00739** [60.61] 0.00740** [60.63] 0.00740** [60.68] 0.00740** [60.70]
High School Score –0.00396** [–21.33] –0.00393** [–21.10] –0.00393** [–21.18] –0.00394** [–21.24]

Risk type X02 0.0597** [12.88] 0.0589** [12.70] 0.0588** [12.69] 0.0590** [12.72]
Risk type A –0.208** [–6.25] –0.208** [–6.27] –0.208** [–6.28] –0.208** [–6.26]
Risk type AE 0.0361** [4.23] 0.0354** [4.14] 0.0353** [4.13] 0.0345** [4.04]
Risk type AE × AE elevation 0.00671** [8.90] 0.00678** [8.99] 0.00678** [8.99] 0.00680** [9.03]
Risk type VE –0.201 [–0.69] –0.205 [–0.71] –0.203 [–0.70] –0.196 [–0.67]

Bay waterfront 0.378** [6.92] 0.377** [6.90] 0.378** [6.90] 0.378** [6.90]
exp(–2 × distance to bay) 0.358** [17.36] 0.357** [17.30] 0.357** [17.34] 0.358** [17.40]
Gulf waterfront 0.696** [7.54] 0.696** [7.53] 0.696** [7.53] 0.696** [7.54]
exp(–2 × distance to Gulf) 0.877** [25.82] 0.876** [25.78] 0.876** [25.81] 0.877** [25.84]
Canal waterfront 0.224** [15.64] 0.223** [15.59] 0.224** [15.62] 0.224** [15.62]
River (non–estuary) waterfront 0.319** [4.71] 0.310** [4.58] 0.312** [4.60] 0.311** [4.60]
Lake waterfront 0.248** [44.01] 0.247** [43.89] 0.248** [43.93] 0.248** [43.97]
exp(–2 × distance to estuary) 0.256** [18.14] 0.161 [4.90] 0.248** [16.89] 0.112** [4.96]
× chlorophyll a (1–month) –0.00459** [–5.01] — — — — — —
× dissolved O2 (1–month) — — 0.00818 [1.89] — — — —
× turbidity (1–month) — — — — –0.0107** [–2.95] — —
× Secchi depth (1–month) — — — — — — 0.0248** [6.04]
Constant 8.170** [126.63] 8.154** [126.53] 8.159** [126.63] 8.140** [126.42]

N 48,572 48,572 48,572 48,572
R-squared 0.8767 0.8766 0.8766 0.8767
* p < 0.05, ** p < 0.01
47
Table B.3
Results of 1-year regressions for Lee County
(1) (2) (3) (4)

ln(lot acreage) 0.0484** [10.76] 0.0472** [10.50] 0.0478** [10.66] 0.0478** [10.66]
Heated area 3.87E–04** [32.65] 3.88E–04** [32.72] 3.88E–04** [32.70] 3.88E–04** [32.76]
Bedrooms –0.0264** [–5.12] –0.0264** [–5.13] –0.0264** [–5.14] –0.0260** [–5.07]
Bathrooms 0.0747** [13.73] 0.0757** [13.92] 0.0747** [13.75] 0.0731** [13.49]
Number of stories –0.0816** [–10.17] –0.0826** [–10.27] –0.0819** [–10.21] –0.0813** [–10.14]
Garage 0.268** [32.38] 0.268** [32.43] 0.265** [32.09] 0.260** [31.67]
Carport 0.0223* [2.20] 0.0217* [2.14] 0.0231* [2.28] 0.0249* [2.47]
Sea wall 0.157** [10.59] 0.160** [10.75] 0.156** [10.54] 0.153** [10.31]
Boat dock 0.115** [12.42] 0.115** [12.46] 0.114** [12.33] 0.114** [12.42]
Swimming pool 0.258** [60.79] 0.258** [60.80] 0.258** [60.75] 0.257** [60.77]
Short sale –0.270** [–72.84] –0.270** [–72.81] –0.270** [–72.91] –0.270** [–73.01]
Foreclosure/REO –0.321** [–93.94] –0.321** [–93.97] –0.321** [–94.08] –0.321** [–94.20]
Pct. of residents age 65+ 0.00360** [19.92] 0.00365** [20.19] 0.00364** [20.15] 0.00355** [19.65]
Millage rate –0.0229** [–21.82] –0.0230** [–21.88] –0.0229** [–21.86] –0.0221** [–21.20]
Primary School Score 0.00210** [10.39] 0.00213** [10.54] 0.00205** [10.16] 0.00193** [9.57]
Middle School Score 0.00739** [60.57] 0.00740** [60.74] 0.00740** [60.73] 0.00743** [61.09]
High School Score –0.00400** [–21.55] –0.00393** [–21.15] –0.00394** [–21.25] –0.00397** [–21.49]

Risk type X02 0.0601** [12.96] 0.0589** [12.70] 0.0590** [12.74] 0.0587** [12.72]
Risk type A –0.207** [–6.23] –0.208** [–6.28] –0.208** [–6.25] –0.206** [–6.18]
Risk type AE 0.0373** [4.36] 0.0354** [4.14] 0.0351** [4.11] 0.0316** [3.70]
Risk type AE × AE elevation 0.00661** [8.77] 0.00678** [8.99] 0.00676** [8.96] 0.00689** [9.14]
Risk type VE –0.196 [–0.68] –0.207 [–0.71] –0.170 [–0.59] –0.152 [–0.53]

Bay waterfront 0.378** [6.93] 0.377** [6.90] 0.378** [6.93] 0.377** [6.93]
exp(–2 × distance to bay) 0.359** [17.43] 0.356** [17.29] 0.360** [17.54] 0.368** [17.88]
Gulf waterfront 0.696** [7.55] 0.696** [7.53] 0.697** [7.56] 0.698** [7.61]
exp(–2 × distance to Gulf) 0.878** [25.86] 0.876** [25.78] 0.878** [25.94] 0.881** [26.08]
Canal waterfront 0.225** [15.66] 0.223** [15.57] 0.226** [15.83] 0.227** [15.78]
River (non–estuary) waterfront 0.320** [4.71] 0.311** [4.60] 0.316** [4.66] 0.322** [4.76]
Lake waterfront 0.248** [44.08] 0.247** [43.92] 0.249** [44.21] 0.249** [44.29]
exp(–2 × distance to estuary) 0.298** [17.34] 0.170 [1.29] 0.420** [16.85] –0.421** [–8.87]
× chlorophyll a (1–year) –0.0103** [–6.46] — — — — — —
× dissolved O2 (1–year) — — 0.00719 [0.38] — — — —
× turbidity (1–year) — — — — –0.0794** [–8.77] — —
× Secchi depth (1–year) — — — — — — 0.147** [14.32]
Constant 8.189** [126.58] 8.158** [126.41] 8.213** [126.56] 8.236** [127.70]

N 48,572 48,572 48,572 48,572
R-squared 0.8767 0.8766 0.8769 0.8774
* p < 0.05, ** p < 0.01
48
References
Boyle, Kevin J., P. Joan Poor, and Laura O. Taylor. (1999) “Estimating the Demand for Protecting Freshwater Lakes from
Eutrophication,” American Journal of Agricultural Economics, 81(5), pp. 1118-1122.
Boyle, Kevin J. and Laura O. Taylor. (2001) “Does the Measurement of Property and Structural Characteristics Affect
Estimated Implicit Prices for Environmental Amenities in a Hedonic Model?” Journal of Real Estate Finance and
Economics, 22(2), pp. 303-318.
Florida Department of Environmental Protection. (2011) Brief History of Lake Okeechobee. Web page, retrieved on 28
Feb. 2015 from http://www.dep.state.fl.us/evergladesforever/about/lakeo_history.htm.
Gordon, Sade M. (2012) “When St. Lucie River Suffers, Martin’s Economy, Quality of Life, Residents’ Health Suffer,”
TCPalm, 6 Dec.
Hecker, Jennifer. (2013) “Poor Water Condition Has a Ripple Effect,” The News-Press, 5 Sep.
Johnson, Ken H., Justin D. Benefield, and Jonathan A. Wiley. (2007) “The Probability of Sale for Residential Real Estate,”
Journal of Housing Research, 16(2), pp. 131-142.
Leggett, Christopher G. and Nancy E. Bockstael. (2000) “Evidence of the Effects of Water Quality on Residential Land
Prices,” Journal of Environmental Economics and Management, 39, pp. 121-144.
Linette, Jim. (2013) “Murky Waters,” Cape Coral Daily Breeze, 9 Aug.
Lollar, Kevin. (2010) “More Fresh Water on Way to Caloosahatchee,” The News-Press, 24 Nov.
Michael, Holly J., Kevin J. Boyle, and Roy Bouchard. (2000) “Does the Measurement of Environmental Quality Affect
Implicit Prices Estimated from Hedonic Models?” Land Economics, 76(2), pp. 283-298.
Poor, P. Joan, Keri L. Pessagno, and Robert W. Paul. (2007) “Exploring the Hedonic Value of Ambient Water Quality: A
Local Watershed-Based Study,” Ecological Economics, 60, pp. 797-806.
Reid, Andy. (2007) “Lake Okeechobee Muck Tests High for Arsenic,” Sun-Sentinel, 8 Jul.
Rosen, Sherwin. (1974) “Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition,” Journal of
Political Economy, 82(1), pp. 797-806.
S&P Dow Jones Indices. (2015, February) S&P/Case-Shiller Home Price Indices Methodology. Retrieved on 1 Mar. 2015
from http://us.spindices.com/documents/methodologies/methodology-sp-cs-home-price-indices.pdf
Spiewak, Jim. (2013) “Is the Dark Water Dangerous to Humans?” NBC-2, 26 Jul. Web page, retrieved on 28 Feb. 2015
from http://www.nbc-2.com/story/22943481/is-the-dark-water-dangerous-to-humans#.VPSMM02BFaQ
United States Army Corps of Engineers. (n.d.) Jacksonville District – Herbert Hoover Dike. Web page. Retrieved on
February 27, 2015 from
http://www.saj.usace.army.mil/Missions/CivilWorks/LakeOkeechobee/HerbertHooverDike.aspx
Walsh, Patrick J., J. Walter Milon, and David O. Scrogin. (2011) “The Spatial Extent of Water Quality Benefits in Urban
Housing Markets,” Land Economics, 87(4), pp. 628-644.
49
Acknowledgements
This research was conducted by Brad O’Connor, Ph.D., Director of Economic Research for Florida
REALTORS®i with assistance from Stephanie Lichtenstein, Research Assistant, and additional
contributions from Erica Cross, M.S.F., Research Analyst. The study was funded in part by a
generous grant from the Everglades Foundation. ii
i Address: 7025 Augusta National Drive, Orlando, Florida 32822; Phone: (407) 438-1400
ii Address: 18001 Old Cutler Road, Suite 625, Palmetto Bay, Florida 33157; Phone: (305) 251-0001
50
APPENDIX U:
SOUTH FLORIDA WATER MANAGEMENT
DISTRICT FACT SHEET
North Fork of St. Lucie River, page 1
April 2015
North Fork of the St. Lucie River
Water Reservation
The joint state-federal Comprehensive Everglades Restoration Plan (CERP) identifies

restoration of the Indian River Lagoon – South as an integral step in achieving
systemwide benefits in the south Florida ecosystem. Subject to extreme salinity
variations, the St. Lucie Estuary and Indian River Lagoon together are home to more
just the than 50 endangered or threatened species. Restoring a more natural volume, timing and
FACTs distribution of flows to the river, floodplain and estuary will give native plant and
animal life a better opportunity for recovery.
This fact sheet is provided as a
The Water Resources Development Act (WRDA) of 2000 required the South Florida
reference to encourage a greater
Water Management District to legally protect water intended for the natural system
understanding of the various
before any federal funding could be authorized to construct the CERP Indian River
issues related to managing
Lagoon – South project. The District adopted a water reservation rule for the North
water in South Florida. Fork of the St. Lucie River, and construction is underway on the C-44 Reservoir and
Stormwater Treatment Area components.
Defining water reservations

• A water reservation is a legal mechanism to set aside water for the
protection of fish and wildlife or public health and safety. When a water
reservation is in place, quantities and timing of water flows at specific
locations are protected for the natural system. The necessary quantities and
timing are determined using data which link local hydrology to the needs of
fish and wildlife.
• All presently existing legal uses of water are protected so long as the use is
not contrary to the public interest. If a project develops water above the
amount needed to protect fish and wildlife, the Governing Board may
certify the volume available for allocation to consumptive uses.
Reservations benefit the environment

Long-term success of ecosystem restoration is measured, in part, by the ability
of native fish and wildlife to thrive in the habitats restored. Key facts
considered in the development of the North Fork of the St. Lucie water
For more information
on this subject, scan
reservation in support of CERP restoration efforts included:
this QR code using a • The St. Lucie Estuary and Indian River Lagoon together form the most
barcode reader app on diverse estuarine environments in North America. The lagoon contains the
your smartphone.
St. Lucie Estuary, a designated Estuary of National Significance.
• Land and marine animals include shortnose sturgeon, smalltooth sawfish,
the Florida scrub jay, peregrine falcon, sandhill crane, manatee and five
species of sea turtles – hawksbill, green, loggerhead, leatherback and
Kemp’s ridley. Plants include Lakela’s mint, Johnson’s seagrass and bay
cedar.
(more)
• The St. Lucie River is designated an Outstanding Florida Water. The river has
more species of fish than any other river in Florida, and more than 500 fish
species have been documented within a 2-mile radius of the St. Lucie River
where it meets the Indian River Lagoon.
• The North Fork of the St. Lucie River is a state aquatic preserve and part of
Florida’s “Save Our Rivers” program. More than 650 native species, including
fish, amphibians, reptiles, birds, mammals, invertebrates, plants and
phytoplankton, have been located and identified within the preserve and
adjacent floodplain, including 33 species of special concern and 20 additional
rare species. The preserve also contains rookeries for the endangered wood
stork.
Background and conditions

• Considered the most diverse estuarine environment in North America, the
St. Lucie Estuary and Indian River Lagoon support multimillion-dollar
fishing, tourism, agricultural and recreational industries. In 2007, the
economic benefits of the Indian River Lagoon were estimated at $3.7 billion.
• Altered hydrology of the system results in highly variable freshwater outflow
to the St. Lucie Estuary, often either too much or too little, creating abnormal
downstream salinity levels throughout the year.
• Extreme salinity variations and ever-increasing inflows have contributed to
major changes in the natural communities of the estuary, as seen by seagrass
and oyster losses.
• Key elements of the CERP Indian River Lagoon – South Project are reducing
high-volume discharges of fresh water to the St. Lucie Estuary, providing a
more natural quantity, quality, timing and distribution of inflows to the
estuary and restoring the North Fork of the St. Lucie River and its floodplain.
• Construction of the CERP project was authorized by WRDA 2007.
Reservation development and public input

• The District Governing Board initiated water reservation rule development in
April 2008. The public process included:
o Review by a panel of nationally recognized, independent scientific
experts.
o Public workshops (between April 2009 and October 2009).
• The North Fork of the St. Lucie water reservation was adopted by the
Governing Board on Sept. 10, 2009, and became effective on March 18, 2010
(Rule 40E-10.051, Florida Administrative Code).
o The rule identifies and reserves from consumptive use the water needed
for the CERP project.
o The water reservation ensures that the St. Lucie River, as well as the
estuary and southern Indian River Lagoon, fully benefit from the
Everglades restoration effort.
Location Map
For more information about this water reservation rule or other previously
adopted reservations, please visit www.sfwmd.gov/reservations.
APPENDIX V:
FLORIDA FISH AND WILDLIFE MANATEE
MORTALITY STATISTICS
07 March 2019
FLORIDA FISH AND WILDLIFE CONSERVATION COMMISSION
MARINE MAMMAL PATHOBIOLOGY LABORATORY
2018 Preliminary Red Tide Manatee Mortalities, Jan 01 – December 31
Manatees Carcasses Collected Within the Known Red Tide Bloom Boundary
SIZE
COUNTY DATE FIELD ID SEX (cm) WATERWAY LOCATION CAUSE
Charlotte 4-Jan MSW18002 F 228 Myakka River Port Charlotte Natural; Other (Red Tide)
Lee 4-Jan MSW18003 M 225 Caloosahatchee River Ft. Myers Natural; Other (Red Tide)
Collier 4-Jan MSW18004 M 141 Tigertail Lagoon Marco Island Perinatal; Undetermined (+)
Lee 5-Jan MSW18005 M 298 Charlotte Harbor Cape Coral Natural; Other (Red Tide)
Lee 6-Jan MSW18006 M 292 Matlacha Pass Cape Coral Natural; Other (Red Tide)
Charlotte 7-Jan MSW18007 F 270 Chalotte Harbor Placida Natural; Other (Red Tide)
Lee 11-Jan MSW18012 F 235 Matlacha Pass Matlacha Watercraft (+)
Lee 13-Jan MSW18015 F F 223 Imperial River Bonita Springs Natural; Cold Stress (%)
Charlotte 17-Jan MSW18020 F M 210 Myakka River Port Charlotte Undetermined; Too Decomposed (%)
Manatee 26-Jan MNW18018 M 302 Anna Maria Sound Bradenton Undetermined; Too Decomposed (+)
Charlotte 6-Feb MSW18035 F M 217 Peace River Punta Gorda Natural; Other (%)
Charlotte 11-Feb MSW18038 F F 267 Charlotte Harbor Punta Gorda Natural; Other (Red Tide)
Lee 17-Feb MSW18039 F F 235 Boca Grande Bayou Boca Grande Natural; Other (Red Tide)
Sarasota 18-Feb MSW18041 F F 239 Myakka River North Port Natural; Other (Red Tide)
Lee 18-Feb MSW18042 F 279 Boca Grande Bayou Boca Grande Natural; Other (Red Tide)
Lee 19-Feb MSW18043 F F 267 Pine Island Sound Captiva Natural; Other (Red Tide)
Lee 24-Feb MSW18044 M 221 Pine Island Sound Captiva Natural; Other (Red Tide)
Lee 26-Feb MSW18045 M 237 Caloosahatchee River Ft. Myers Natural; Other (Red Tide)
Lee 26-Feb MSW18046 F M 193 Pejuan Cove Cayo Costa Natural; Other (Red Tide)
Lee 27-Feb MSW18047 F F 225 Boca Grande Bayou Boca Grande Undetermined; Too Decomposed (%)
Lee 2-Mar MSW18048 F F 268 Dinken Bayou Sanibel Undetermined; Too Decomposed (%)
Lee 3-Mar MSW18050 F 326 Caloosahatchee River Alva Undetermined; Too Decomposed (%)
Lee 4-Mar MSW18051 M 285 Henley Canal Saint James City Natural; Other (Red Tide)
Charlotte 4-Mar MSW18052 F 225 Gasparilla Sound Placida Natural; Other (Red Tide)
Sarasota 5-Mar MSW18053 M 250 Blackburn Bay Nokomis Natural; Other (Red Tide)
Charlotte 6-Mar MSW18054 F F 226 Bull Bay Placida Natural; Other (Red Tide)
Lee 6-Mar MSW18055 F F 236 Pine Island Sound St. James City Undetermined; Too Decomposed (+)
Lee 11-Mar MSW18057 M 173 Imperial River Bonita Springs Natural; Other (%)
Lee 13-Mar MSW18059 M 245 San Carlos Bay Sanibel Natural; Other (Red Tide)
Lee 13-Mar MSW18060 F M 236 Pine Island Sound Sanibel Natural; Other (Red Tide)
Lee 15-Mar MSW18061 M 217 Pine Island Sound Captiva Natural; Other (Red Tide)
Lee 19-Mar MSW18064 M 204 Matlacha Pass Ft. Myers Beach Natural; Other (%)
Lee 19-Mar MSW18065 F M 158 Caloosahatchee River Cape Coral Natural; Other (%)
Lee 26-Mar MSW18067 F M 296 Roosevelt Channel Captiva Natural; Other (Red Tide)
Lee 26-Mar MSW18068 M 284 Iona Cove Ft. Myers Natural; Other (Red Tide)
Lee 29-Mar MSW18070 F 237 Gulf of Mexico North Captiva Island Natural; Other (Red Tide)
Lee 30-Mar MSW18071 M 246 Charlotte Harbor Bokeelia Natural; Other (Red Tide)
Lee 2-Apr MSW18073 F F 271 Roosevelt Channel Captiva Natural; Other (Red Tide)
Lee 4-Apr MSW18074 F 292 Roosevelt Channel Captiva Natural; Other (Red Tide)
Charlotte 4-Apr MSW18077 M 233 Gasparilla Sound Placida Undetermined; Too Decomposed (%)
Lee 4-Apr MSW18078 M 312 Pine Island Sound St. James City Natural; Other (Red Tide)
Lee 5-Apr MSW18079 M 302 Charlotte Harbor Boca Grande Natural; Other (Red Tide)
Lee 6-Apr MSW18080 M 254 Gasparilla Sound Boca Grande Natural; Other (Red Tide)
Lee 6-Apr MSW18081 M 331 Roosevelt Channel Captiva Natural; Other (Red Tide)
Page 1 of 5
07 March 2019
Lee 6-Apr MSW18082 U 266 Roosevelt Channel Captiva Natural; Other (Red Tide)
Lee 7-Apr MSW18083 F 264 Roosevelt Channel Captiva Natural; Other (Red Tide)
Lee 9-Apr MSW18084 F 350 Caloosahatchee River Ft. Myers Natural; Other (Red Tide)
Lee 9-Apr MSW18086 M 266 Charlotte Harbor Bokeelia Natural; Other (Red Tide)
Lee 11-Apr MSW18087 M 307 Charlotte Harbor Boca Grande Natural; Other (Red Tide)
Charlotte 10-Apr SWFTm1826B M 205 Charlotte Harbor Punta Gorda Natural; Other (%)
Lee 10-Apr SWFTm1827B M 221 Blind Pass Captiva Natural; Other (%)
Lee 13-Apr MSW18090 M 260 Kesson Bayou Sanibel Natural; Other (Red Tide)
Lee 14-Apr MSW18092 F F 245 Boca Grande Bayou Boca Grande Natural; Other (Red Tide)
Lee 14-Apr MSW18093 M 239 Pine Island Sound Captiva Natural; Other (Red Tide)
Lee 16-Apr MSW18095 F F 221 Pine Island Sound Captiva Natural; Other (Red Tide)
Lee 18-Apr MSW18099 M 265 Pine Island Sound St. James City Natural; Other (Red Tide)
Collier 17-Apr MSW18100 F 315 Faka Union Canal Ports of the Islands Natural; Other (Red Tide)
Collier 19-Apr MSW18101 F F 240 Gullivan Bay Cape Romano Natural; Other (Red Tide)
Lee 21-Apr MSW18105 F F 255 Clam Bayou Sanibel Natural; Other (Red Tide)
Lee 23-Apr MSW18106 M 335 Kesson Bayou Sanibel Natural; Other (Red Tide)
Lee 23-Apr MSW18107 M 331 Dinken Bayou Sanibel Natural; Other (Red Tide)
Lee 24-Apr MSW18108 F 220 Big Carlos Pass Ft. Myers Beach Natural; Other (Red Tide)
Collier 26-Apr MSW18109 F M 295 Rookery Bay Naples Natural; Other (Red Tide)
Collier 26-Apr MSW18110 F 347 Gulf of Mexico Naples Natural; Other (Red Tide)
Lee 28-Apr MSW18112 M 272 San Carlos Bay Saint James City Natural; Other (Red Tide)
Lee 28-Apr MSW18113 F 152 San Carlos Bay Ft. Myers Natural; Other (%)
Lee 29-Apr MSW18115 F 263 Imperial River Bonita Springs Natural; Other (Red Tide)
Lee 30-Apr MSW18114 F 189 Imperial River Estero Natural; Other (%)
Collier 1-May MSW18116 M 285 Gulf of Mexico Helen Key Natural; Other (Red Tide)
Collier 2-May MSW18118 F M 270 Caxambas Pass Marco Island Natural; Other (Red Tide)
Lee 2-May MSW18119 F 283 Buccaneer Bay Ft. Myers Beach Natural; Other (Red Tide)
Charlotte 7-May MSW18123 M 251 Charlotte Harbor Boca Grande Natural; Other (Red Tide)
Lee 7-May MSW18124 M 98 Hurricane Bay Ft. Myers Beach Perinatal; Natural; Other (%)
Collier 9-May MSW18126 F M 160 Cocohatchee River Naples Undetermined; Too Decomposed (%)
Lee 7-May MSW18128 F F 246 Boca Grande Bayou Boca Grande Natural; Other (Red Tide)
Collier 10-May MSW18130 F F 273 Faka Union Canal Port of the Islands Undetermined; Too Decomposed (%)
Collier 11-May MSW18131 F 263 Gulf of Mexico Cape Romano Natural; Other (Red Tide)
Lee 11-May MSW18133 F 303 Ten Mile Canal Ft. Myers Natural; Other (Red Tide)
Collier 16-May MSW18137 F 302 Venetian Bay Naples Natural; Other (Red Tide)
Lee 24-May MSW18141 F 248 San Carlos Bay St. James City Natural; Other (Red Tide)
Lee 23-May MSW18144 M 285 Thunderbird Lake Cape Coral Verified; Not Recovered (+)
Lee 2-Jun MSW18149 F 344 Pine Island Sound North Captiva Island Natural; Other (Red Tide)
Charlotte 2-Jun MSW18151 M 207 Gasparilla Sound Placida Undetermined; Too Decomposed (%)
Collier 10-Jun MSW18155 M 318 Little Hickory Bay Bonita Springs Watercraft (+)
Lee 11-Jun MSW18156 F M 228 Jug Creek Bokeelia Undetermined; Too Decomposed (%)
Lee 12-Jun MSW18157 M 204 Boca Grande Bayou Boca Grande Natural; Other (Red Tide)
Lee 16-Jun MSW18159 M 242 Pine Island Sound Bokeelia Natural; Other (Red Tide)
Lee 19-Jun MSW18160 F 290 Pine Island Sound St. James City Natural; Other (Red Tide)
Charlotte 19-Jun MSW18161 M 310 Gulf of Mexico Little Gasparilla Island Undetermined; Too Decomposed (+)
Collier 22-Jun MSW18162 F F 339 Dollar Bay Naples Natural; Other (Red Tide)
Lee 26-Jun MSW18166 F 318 Pine Island Sound Cayo Costa Natural; Other (Red Tide)
Lee 26-Jun MSW18167 M 254 Pine Island Sound Cayo Costa Natural; Other (Red Tide)
Collier 28-Jun MSW18170 M 332 Sanctuary Sound Marco Island Undetermined; Too Decomposed (+)
Charlotte 30-Jun MSW18172 M 302 Gasparilla Sound Boca Grande Natural; Other (Red Tide)
Sarasota 4-Jul MSW18174 M 303 South Creek Osprey Undetermined; Too Decomposed (+)
Charlotte 6-Jul MSW18176 F 187 Lemon Bay Englewood Natural; Other (%)
Collier 7-Jul MSW18177 M 298 Sanctuary Sound Marco Island Undetermined; Too Decomposed (+)
Page 2 of 5
07 March 2019
Charlotte 10-Jul MSW18178 M 332 Peace River Port Charlotte Natural; Other (Red Tide)
Charlotte 10-Jul MSW18179 F 323 Charlotte Harbor Cape Haze Natural; Other (Red Tide)
Charlotte 13-Jul MSW18182 F 298 Lemon Bay Englewood Natural; Other (Red Tide)
Collier 13-Jul MSW18183 F 235 Gullivan Bay Kice Island Natural; Other (Red Tide)
Sarasota 14-Jul MSW18184 F 223 Forked Creek Englewood Natural; Other (Red Tide)
Collier 19-Jul MSW18186 M 186 Rookery Bay Naples Natural; Other (Red Tide)
Lee 29-Jul MSW18189 M 192 Caloosahatchee River Cape Coral Natural; Other (Red Tide)
Sarasota 30-Jul MSW18190 F 313 Lemon Bay Englewood Natural; Other (Red Tide)
Lee 31-Jul MSW18191 F 355 Caloosahatchee River Cape Coral Watercraft (+)
Sarasota 31-Jul MSW18192 F 290 Lemon Bay Englewood Natural; Other (Red Tide)
Sarasota 1-Aug SWFTm1844B M 279 Lemon Bay Placia Natural; Other (Red Tide)
Lee 2-Aug MSW18194 M 217 Fish Trap Bay Bonita Springs Natural; Other (Red Tide)
Sarasota 2-Aug MSW18195 M 292 Lemon Bay Venice Watercraft (+)
Lee 3-Aug MSW18197 F 232 Pine Island Sound Upper Captiva Undetermined; Too Decomposed (+)
Sarasota 3-Aug MSW18198 F 290 Little Sarasota Bay Sarasota Natural; Other (Red Tide)
Charlotte 4-Aug MSW18200 M 268 Gasparilla Sound Placida Undetermined; Too Decomposed (+)
Sarasota 4-Aug MSW18201 M 296 Lemon Bay Englewood Natural; Other (Red Tide)
Sarasota 5-Aug MSW18203 M 214 Dona Bay Venice Natural; Other (Red Tide)
Sarasota 5-Aug MSW18205 M 260 Little Sarasota Bay Sarasota Undetermined; Too Decomposed (+)
Charlotte 5-Aug MSW18206 F M 215 Placida Harbor Placida Natural; Other (Red Tide)
Lee 6-Aug MSW18207 F 285 Matlacha Pass Bokeelia Natural; Other (Red Tide)
Sarasota 7-Aug MSW18208 M 315 Intracoastal Waterway Manasota Key Natural; Other (Red Tide)
Sarasota 7-Aug MSW18209 M 239 Dona Bay Nokomis Natural; Other (Red Tide)
Collier 8-Aug MSW18211 F 197 Johnson Bay Naples Natural; Other (Red Tide)
Manatee 9-Aug MNW18095 F 317 Manatee River Parrish Undetermined; Too Decomposed (+)
Lee 13-Aug MSW18216 M 131 San Carlos Bay St. James City Perinatal: Natural (Red Tide)
Charlotte 13-Aug MSW18217 F 139 Ainger Creek Englewood Perinatal; Natural (%)
Sarasota 13-Aug MSW18218 M 327 Blackburn Bay Nokomis Natural; Other (Red Tide)
Sarasota 15-Aug MSW18220 M 176 Phillippi Creek Sarasota Undetermined; Too Decomposed (+)
Manatee 16-Aug MNW18098 F 278 Terra Ceia Bay Palmetto Undetermined; Too Decomposed (+)
Sarasota 16-Aug MSW18221 M 298 Little Sarasota Bay Osprey Natural; Other (Red Tide)
Lee 17-Aug MSW18222 F 258 Pine Island Sound Cayo Costa Natural; Other (Red Tide)
Sarasota 18-Aug MSW18223 M 280 Phillipe Creek Sarasota Natural; Other (Red Tide)
Lee 18-Aug MSW18224 F 278 San Carlos Bay Sanibel Natural; Other (Red Tide)
Collier 19-Aug MSW18227 F 243 Tigertail Lagoon Marco Island Natural; Other (Red Tide)
Lee 19-Aug MSW18228 M 229 Caloosahatchee River Cape Coral Natural; Other (Red Tide)
Sarasota 19-Aug MSW18229 M 198 Blackburn Bay Nokomis Watercraft (+)
Sarasota 20-Aug MSW18230 M 226 Shakett Creek Nokomis Natural; Other (Red Tide)
Lee 20-Aug MSW18231 F F 366 San Carlos Bay St. James City Natural; Other (Red Tide)
Lee 20-Aug MSW18232 F M 126 Caloosahatchee River Cape Coral Perinatal; Natural (Red Tide)
Collier 20-Aug MSW18233 F M 231 Tarpon Bay Marco Island Natural; Other (Red Tide)
Sarasota 21-Aug MSW18234 F 290 Little Sarasota Bay Sarasota Natural; Other (Red Tide)
Collier 21-Aug MSW18235 F M 309 Smokehouse Bay Marco Island Natural; Other (Red Tide)
Collier 21-Aug MSW18236 F M 306 Sanctuary Sound Marco Island Natural; Other (Red Tide)
Sarasota 22-Aug MSW18237 F 286 South Creek Osprey Natural; Other (Red Tide)
Collier 22-Aug MSW18238 F F 256 Sanctuary Sound Marco Island Natural; Other (Red Tide)
Sarasota 23-Aug MSW18239 F 319 Sarasota Bay Longboat Key Natural; Other (Red Tide)
Sarasota 26-Aug MSW18245 M 308 Dryman Bay Osprey Natural; Other (Red Tide)
Manatee 26-Aug MNW18102 M 325 Anna Maria Sound Holmes Beach Natural; Other (Red Tide)
Sarasota 27-Aug MSW18246 M 313 Dryman Bay Osprey Natural; Other (Red Tide)
Lee 27-Aug MSW18247 F F 356 Pine Island Sound St. James City Undetermined; Too Decomposed (+)
Lee 27-Aug MSW18248 M 248 Tarpon Bay Sanibel Natural; Other (Red Tide)
Sarasota 28-Aug MSW18249 M 328 Sarasota Bay Sarasota Natural; Other (Red Tide)
Page 3 of 5
07 March 2019
Manatee 28-Aug MNW18103 M 302 Anna Maria Sound Bradenton Natural; Other (Red Tide)
Lee 28-Aug MSW18251 F 308 Charlotte Harbor Bokeelia Natural; Other (Red Tide)
Sarasota 30-Aug MSW18252 M 255 Little Sarasota Bay Sarasota Natural; Other (Red Tide)
Sarasota 30-Aug MSW18253 M 308 Blackburn Bay Osprey Natural; Other (Red Tide)
Charlotte 30-Aug MSW18254 F 211 Lemon Bay Englewood Natural; Other (Red Tide)
Manatee 31-Aug MNW18104 M 274 Bowlees Creek Bradenton Natural; Other (Red Tide)
Sarasota 1-Sep MSW18257 M 261 Intracoastal Waterway Venice Natural; Other (Red Tide)
Sarasota 2-Sep MSW18258 M 213 Turtle Bay Sarasota Natural; Other (Red Tide)
Sarasota 2-Sep MSW18259 F 263 Roberts Bay Venice Natural; Other (Red Tide)
Sarasota 2-Sep MSW18260 F 140 Dona Bay Nokomis Perinatal; Natural (+)
Lee 2-Sep MSW18261 F 258 Tarpon Bay Sanibel Natural; Other (Red Tide)
Manatee 2-Sep MNW18106 F 301 Sarasota Bay Bradenton Natural; Other (Red Tide)
Collier 2-Sep MSW18263 M 214 Addison Bay Marco Island Natural; Other (Red Tide)
Sarasota 4-Sep MSW18266 M 230 Sarasota Bay Sarasota Natural; Other (Red Tide)
Manatee 6-Sep MNW18107 M 247 Sarasota Bay Longboat Key Natural; Other (Red Tide)
Manatee 8-Sep MNW18109 M 302 Sarasota Bay Longboat Key Natural; Other (Red Tide)
Collier 9-Sep MSW18272 F 261 Collier Bay Marco Island Natural; Other (~)
Lee 10-Sep MSW18273 F F 273 Roosevelt Channel Captiva Undetermined; Too Decomposed (~)
Lee 11-Sep MSW18274 F 270 San Carlos Bay Ft. Myers Undetermined; Too Decomposed (*)
Collier 12-Sep MSW18275 F 296 Tigertail Lagoon Marco Island Natural; Other (Red Tide)
Collier 14-Sep MSW18278 F 298 Collier Bay Marco Island Natural; Other (Red Tide)
Charlotte 18-Sep MSW18279 M 323 Gulf of Mexico Placida Natural; Other (Red Tide)
Sarasota 18-Sep MSW18280 M 335 Gulf of Mexico Venice Natural; Other (Red Tide)
Charlotte 18-Sep MSW18281 M 337 Gulf of Mexico Englewood Natural; Other (Red Tide)
Sarasota 18-Sep MSW18283 M 288 Gulf of Mexico Englewood Natural; Other (Red Tide)
Pinellas 19-Sep MNW18117 F 299 Clearwater Pass Clearwater Natural; Other (Red Tide)
Charlotte 19-Sep MSW18284 M 222 Lemon Bay Englewood Natural; Other (Red Tide)
Charlotte 20-Sep MSW18286 M 260 Gulf of Mexico Palm Island Natural; Other (Red Tide)
Charlotte 22-Sep MSW18289 M 305 Lemon Bay Englewood Natural; Other (Red Tide)
Sarasota 22-Sep MSW18290 M 290 Gulf of Mexico Osprey Natural; Other (Red Tide)
Sarasota 23-Sep MSW18291 M 261 Sarasota Bay Sarasota Natural; Other (Red Tide)
Lee 25-Sep MSW18294 F 281 Estero Bay Bonita Springs Natural; Other (Red Tide)
Lee 25-Sep MSW18296 M 295 Estero Bay Bonita Springs Natural; Other (Red Tide)
Lee 26-Sep MSW18298 F 218 Caloosahatchee River Ft. Myers Natural; Other (Red Tide)
Lee 27-Sep MSW18300 M 346 Charlotte Harbor Placida Watercraft (+)
Sarasota 27-Sep MSW18301 F 260 Blackburn Bay Osprey Natural; Other (Red Tide)
Pinellas 28-Sep MNW18118 M 145 Riviera Bay St. Petersburg Perinatal; Undetermined (%)
Sarasota 29-Sep MSW18303 M 288 Dona Bay Nokomis Natural; Other (Red Tide)
Sarasota 29-Sep MSW18305 M 296 South Creek Osprey Natural; Other (Red Tide)
Sarasota 29-Sep MSW18306 M 304 Roberts Bay Venice Natural; Other (Red Tide)
Pinellas 29-Sep MNW18119 F 290 Clearwater Harbor Clearwater Natural; Other (Red Tide)
Pinellas 1-Oct MNW18120 F 222 Intracoastal Waterway Indian Shores Natural; Other (Red Tide)
Lee 2-Oct MSW18307 M 312 San Carlos Bay Sanibel Natural; Other (Red Tide)
Pinellas 2-Oct MNW18121 M 286 Boca Ciega Bay Treasure Island Natural; Other (Red Tide)
Martin 5-Oct MSE18073 M 284 Indian River Stuart Natural; Other (Red Tide)
Sarasota 5-Oct MSW18311 F 312 Sarasota Bay Sarasota Natural; Other (Red Tide)
Pinellas 6-Oct MNW18123 M 277 Clearwater Harbor Largo Natural; Other (Red Tide)
Lee 6-Oct MSW18312 F M 220 San Carlos Bay St. James City Natural; Other (Red Tide)
Pinellas 13-Oct MNW18127 M 272 Boca Ciega Bay St. Petersburg Natural; Other (Red Tide)
Pinellas 14-Oct MNW18128 F 273 McKay Creek Largo Natural; Other (Red Tide)
Pinellas 13-Oct MNW18129 M 338 Soldier Hole St. Petersburg Undetermined; Too Decomposed (+)
Pinellas 14-Oct MNW18130 F 261 Long Bayou St. Petersburg Natural; Other (Red Tide)
Page 4 of 5
07 March 2019
Pinellas 19-Oct MNW18133 M 216 Long Bayou St. Petersburg Natural; Other (Red Tide)
Lee 20-Oct MSW18318 F F 255 Pine Island Sound St. James City Undetermined; Too Decomposed (~)
Pinellas 29-Oct MNW18136 M 284 Boca Ciega Bay St. Pete Beach Natural; Other (Red Tide)
Pinellas 29-Oct MNW18138 M 238 Mullet Key Bayou St. Petersburg Natural; Other (Red Tide)
St. Lucie 30-Oct MSE18077 M 269 Indian River Ft. Pierce Natural; Other (Red Tide)
Pinellas 31-Oct MNW18140 M 166 Stevenson Creek Clearwater Undetermined; Too Decomposed (+)
Indian River 1-Nov MEC18131 F 235 Indian River Vero Beach Natural; Other (Red Tide)
Hillsborough 3-Nov MNW18142 M 117 Old Tampa Bay Tampa Perinatal; Natural (%)
Manatee 5-Nov MNW18143 F 229 Palma Sola Bay Bradenton Natural; Other (Red Tide)
Pinellas 14-Nov MNW18150 M 280 St. Joseph Sound Dunedin Natural; Other (Red Tide)
St. Lucie 15-Nov MSE18081 M 249 Indian River Stuart Natural; Other (~)
Charlotte 16-Nov MSW18324 M 298 Stump Pass Englewood Natural; Other (Red Tide)
Pinellas 17-Nov MNW18151 M 290 Tampa Bay St. Petersburg Natural; Other (Red Tide)
Lee 18-Nov MSW18325 M 212 San Carlos Bay Sanibel Natural; Other (Red Tide)
Pinellas 29-Nov MNW18155 M 284 Boca Ciega Bay St. Petersburg Natural; Other (~)
Pinellas 8-Dec MNW18158 M 232 Clearwater Harbor Belleair Beach Natural; Other (~)
Collier 19-Dec MSW18338 F 320 Sanctuary Sound Marco Island Natural; Other (~)
F Indicates a Field Necropsy was Performed. (+) (%) (#) (*) (~) Red Tide Suspect
Red Tide Positive = 169

Red Tide Suspect = 55
Red Tide Total = 224
(Red Tide) and (+) Brevetoxin detected by ELISA analysis in Stomach Contents, Liver, Urine, or other collected sample.
(%) Red Tide Suspect: Low level of brevetoxin detected.
(#) Red Tide Suspect: Tissue samples being tested at FWRI.
(~) Red Tide Suspect: Tissue samples at
MMPL.
(*) Carcass recovered within the known bloom boundary and no other cause of death was apparent. Tissues were not tested.
Numbers are preliminary. More manatee mortalities may be added after review of
brevetoxin lab results and spatial/temporal analysis of carcasses that were verified
but not recovered. For a list of all 2019 carcasses to date, see:
https://myfwc.com/research/manatee/rescue-mortality-
response/statistics/mortality/2018/.
Page 5 of 5
APPENDIX W:
BMAA STUDIES
Food and Chemical Toxicology 70 (2014) 26–32
Contents lists available at ScienceDirect
Food and Chemical Toxicology

journal homepage: www.elsevier.com/locate/foodchemtox
Environmental neurotoxins b-N-methylamino-L-alanine (BMAA)

and mercury in shark cartilage dietary supplements
Kiyo Mondo a, W. Broc Glover b, Susan J. Murch b, Guangliang Liu c, Yong Cai c, David A. Davis a,
Deborah C. Mash a,⇑
a
Department of Neurology, Miller School of Medicine, University of Miami, Miami, FL 33136, USA
b
Department of Chemistry, University of British Columbia, Kelowna, British Columbia V1V 1V7, Canada
c
Bioinorganic and Environmental Analytical Facility, Florida International University, 11200 SW 8th St., Miami, FL 33199, USA
a r t i c l e i n f o a b s t r a c t
Article history: Shark cartilage products are marketed as dietary supplements with claimed health benefits for animal
Received 25 February 2014 and human use. Shark fin and cartilage products sold as extracts, dry powders and in capsules are
Accepted 8 April 2014 marketed based on traditional Chinese medicine claims that it nourishes the blood, enhances appetite,
Available online 19 April 2014
and energizes multiple internal organs. Shark cartilage contains a mixture of chondroitin and glucosamine,
a popular nutritional supplement ingested to improve cartilage function. Sharks are long-lived apex
Keywords: predators, that bioaccumulate environmental marine toxins and methylmercury from dietary exposures.
b-N-methylamino-L-alanine
We recently reported detection of the cyanobacterial toxin b-N-methylamino-L-alanine (BMAA) in the fins
Mercury
Shark cartilage
of seven different species of sharks from South Florida coastal waters. Since BMAA has been linked to
Dietary supplements degenerative brain diseases, the consumption of shark products may pose a human risk for BMAA
Cyanotoxin exposures. In this report, we tested sixteen commercial shark cartilage supplements for BMAA by high
performance liquid chromatography (HPLC-FD) with fluorescence detection and ultra performance liquid
chromatography/mass spectrometry/mass spectrometry (UPLC–MS/MS). Total mercury (Hg) levels were
measured in the same shark cartilage products by cold vapor atomic fluorescence spectrometry (CVAFS).
We report here that BMAA was detected in fifteen out of sixteen products with concentrations ranging from
86 to 265 lg/g (dry weight). All of the shark fin products contained low concentrations of Hg. While Hg
contamination is a known risk, the results of the present study demonstrate that shark cartilage products
also may contain the neurotoxin BMAA. Although the neurotoxic potential of dietary exposure to BMAA is
currently unknown, the results demonstrate that shark cartilage products may contain two environmental
neurotoxins that have synergistic toxicities.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction of glycosaminoglycans, a popular nutritional supplement source of

chondroitin sulfate that is used to improve cartilage function. More
An estimated 100 million sharks are traded on the fin market than 40 different brand names of shark cartilage are available on
annually to support human demand for dietary and medicinal the market today and they are widely sold and remain popular
use (Ritter, 2004). Although the U.S. FDA has not confirmed any with consumers (Holt, 1995; NCI, 2013).
health benefits, Traditional Chinese Medicine claims that shark Sharks are long-lived apex predators known to accumulate
fin has health benefits for multiple internal organs. Shark cartilage mercury and other heavy metals over their lifespan (Dolan et al.,
has been investigated as a treatment for cancer, arthritis, 2003; Kim, 2004; Leblond et al., 2008). Recently, the cyanobacterial
osteoarthritis, and other medical conditions (Himmel and toxin b-N-methylamino-L-alanine (BMAA) was detected and quan-
Seligman, 1999; Loprinzi et al., 2005; Lu et al., 2010; Milner, tified in the fins of seven different species of South Florida sharks
1999; Patra and Sandell, 2012). Shark cartilage contains a mixture (Mondo et al., 2012). BMAA has been detected and quantified in
the marine food web from South Florida coastal waters (Brand
⇑ Corresponding author. Tel.: +1 305 243 5888; fax: +1 305 243 3649. et al., 2010) and in the temperate aquatic ecosystem of the Baltic
E-mail addresses: kmondo@med.miami.edu (K. Mondo), williambrocglover@ Sea (Jonasson et al., 2010). BMAA is produced by diverse species
gmail.com (W. Broc Glover), susan.murch@ubc.ca (S.J. Murch), liug@fiu.edu (G. Liu), of free-living cyanobacteria found in terrestrial and aquatic
caiy@fiu.edu (Y. Cai), d.davis12@med.miami.edu (D.A. Davis), dmash@med.miami. environments (Cox et al., 2005) and cyanobacterial symbionts
edu (D.C. Mash).
http://dx.doi.org/10.1016/j.fct.2014.04.015
0278-6915/Ó 2014 Elsevier Ltd. All rights reserved.
K. Mondo et al. / Food and Chemical Toxicology 70 (2014) 26–32 27
(Cox et al., 2003). Production of BMAA by marine cyanobacteria Commercial shark cartilage powders were compared with shark matrix nega-
tive for BMAA spiked with known amounts of a reference BMAA standard (Sigma
may be a route of human exposure to this neurotoxin (Brand
B-107; >95% purity, St. Louis, MO, USA). The limits of detection (LOD) and the limit
et al., 2010; Jonasson et al., 2010). A potential link between BMAA of quantification (LOQ) were based on the standard deviation (SD) of response and
and neurodegenerative disease was originally discovered on the slope (S), calculated from linearity of the response of BMAA. The LOD and LOQ were
island of Guam in the Western Pacific Ocean, where there was an obtained by using formula (3.3 SD)/S and (10 SD)/S, respectively. The efficiency
exceptionally high incidence of amyotrophic lateral sclerosis of the analyte recovery was estimated adding known amounts of a BMAA standard
spiked into a reference sample that was below the LOD.
(ALS) of late onset with features of Parkinsonism and dementia
(Bell, 2009). BMAA has been measured in the brain from North
2.4. UPLC–MS/MS detection of BMAA, AEG, and DAB
American ALS and Alzheimer’s disease patients (Pablo et al.,
2009), although a causal link to either of these neurodegenerative Identification of a BMAA peak detected by reverse-phase HPLC was verified by
diseases has not been established. While the exposure risk to an orthogonal method with UPLC separation and tandem mass spectrometry.
humans is currently unknown, BMAA is misincorporated into brain Approximately 50 mg of shark cartilage powder samples were hydrolyzed for
proteins (Dunlop et al., 2013; Xie et al., 2013), where it may serve 16 h in 6 N HCl at 110 °C. The sample was centrifuge filtered (0.22 lm Ultrafree-
MC, Millipore), and 100 lL of the filtrate was lyophilized overnight. The dried sam-
as a reservoir for slow release neurotoxicity with repeated dietary
ple was reconstituted to 1000 lL in 20 mM HCl. An aliquot of 20 lL of each sample
exposures over the lifespan (Murch et al., 2004b). was derivatized with 60 lL of borate buffer and 20 lL of AccQ-Tag. The derivatized
Since BMAA has been detected in shark fins, we investigated the sample was analyzed using a triple quadruple tandem mass spectrometer after sep-
concentrations of BMAA and total Hg in commercial shark cartilage aration by a Waters Acquity I-Class UPLC system, BEH C18 column (Waters,
150 2.1 mm, 1.7 lm) at 52 °C. BMAA, AEG, and DAB separation were achieved
dietary supplements. Shark cartilage dietary supplements were
using gradient elution at 0.7 mL/min in 20 mM ammonium formate with 0.2% for-
analyzed for BMAA using high performance liquid chromatography mic acid (Eluent A) and 0.1% formic acid in acetonitrile (Eluent B): 0.0 min 95% A;
with fluorescence detection (HPLC-FD) according to standard pro- 1.0 min 95% A; 7.0 min 85% A; 7.5 min 78% A; 8.0 min 15% A; 8.5 min 15% A;
tocols (Banack and Cox, 2003; Banack et al., 2007). Independent lab- 8.6 min 95% A; 10.0 min 95% A. The Waters Xevo TQ-S triple quadrupole mass spec-
oratory confirmation of BMAA and its isomers 2,4-diaminobutyric trometer was run in ES+ mode, with a cone voltage of 16 V, a capillary voltage of
2500 V, and a source offset of 50 V. The desolvation temperature was set to
acid (DAB) and N-(2-aminoethyl) glycine (AEG) was determined
550 °C, with corresponding gas flow of 800 L/h and a cone gas flow of 150 L/h. Col-
for each sample by ultra performance liquid chromatography/mass lision-induced-dissociation (CID) was performed with 99.999% pure argon and reg-
spectrometry/mass spectrometry (UPLC–MS/MS). Total Hg concen- ulated to 7.0 psi. The first quadrupole was set to exclude all masses except m/z 459,
trations were determined by cold vapor atomic fluorescence the AQC double-derivatized BMAA. The transitions and collision energies (CE) were
spectrometry (CVAFS). used to monitor for BMAA: 459 > 119.0 (CE 17); 459 > 289.1 (CE 12); 459 > 171.0
(CE 28). Additional specific transitions, 459 > 258.0 (CE 18) for BMAA, 459 > 188.0
(CE 20) for DAB, and 459.0 > 214.0 (CE 20) for AEG were monitored to obtain struc-
tural information for each isomer. The LOD and LOQ were determined as described
2. Materials and methods above according to the USP/EPA methods (Betz et al., 2011).
2.1. Chemicals
2.5. Determination of total mercury
Acetonitrile (ACN), HPLC grade water (J.T. Baker; Phillipsburg, NJ), and hydro-
chloric acid (HCl) were obtained from VWR (Suwanee, GA, USA). AccQ Tag Fluor Total Hg concentrations in shark cartilage powder samples were determined by
(AQC) Reagent Kit was purchased from the Waters Corporation (Milford, MA). the cold vapor atomic fluorescence spectrometry (CVAFS) (Millennium Merlin
L-BMAA hydrochloride and DL-2,4-DAB dihydrochloride were purchased from Sigma 10.035, PS Analytical, Florida, USA) following the standard operating procedure
Aldrich (St. Louis, MO). N-2(amino)ethylglycine was obtained from TCI America modified after EPA method 1631 (USEPA, 2002). Total Hg was isolated and oxidized
(Tokyo, Japan). to mercuric ion using acid digestion and reduced to elemental Hg by stannous chlo-
ride, purged from the liquid by argon, and then introduced into CVAFS system. The
method limit of detection (MLD) for the instrument was 0.002 mg/kg.
2.2. Sample preparation
Sixteen commercially available shark cartilage powders were assayed for

3. Results
BMAA, AEG and DAB levels in total acid hydrolysates. Each powder sample
(50 mg) was hydrolyzed in 6 N HCl (1000 lL) for 18 h at 110 °C. The hydrolysates A total of 16 samples were tested for BMAA and total Hg con-
were filtered at 13,800g for 3 min (0.22 lm Ultrafree-MC, Millipore, Bedford, MA) centrations from 7 different commercial manufacturers. We
and concentrated in a speed-vacuum (Thermo-Savant SC250DDA Speed Vac Plus
obtained different batch samples for testing of reference shark car-
with a Savant refrigerator trap RVT 4104). The dried extract was resuspended to
1000 lL in 20 mM HCl. An aliquot of the sample extract was derivatized with tilage products supplied to us from two top-rated supplement
6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) using the AccQ-Fluor brands and merchants (ConsumerLab.com). We also tested pow-
reagent (Waters Crop, Milford, MA). The derivatized samples (20 lL resuspended ders from capsules purchased from an additional 5 commercial
HCl extract, 60 lL of borate buffer, and 20 lL AccQ-Tag) were run in parallel with nutritional supplement suppliers for BMAA, its structural isomers
buffer and AQC blanks and BMAA, AEG, DAB and reference amino acid standards.
The sample matrix was spiked with known amounts of BMAA to confirm the effi-
DAB and AEG, and total Hg. The samples were analyzed in replicate
ciency of the analyte extraction. runs and BMAA was analyzed and quantified in two independent
laboratories by HPLC-FD and UPLC–MS/MS methods to ensure
accuracy of detection. The identification and quantification of
2.3. Fluorescence HPLC methods for analysis of total BMAA BMAA in complex matrices has been challenging until the recent
BMAA was detected and quantified using a validated HPLC method (Banack and
introduction of reliable methods (Banack et al., 2010, 2012; Jiang
Cox, 2003; Banack et al., 2007) with minor modifications (Mondo et al., 2012; et al., 2012, 2013). The potential interference from structural iso-
Murch et al., 2004b). The non-protein amino acid BMAA was separated from its mers of BMAA on the accuracy of detection was analyzed using a
structural isomers and protein amino acids (Banack and Cox, 2003; Banack et al., validated LC–MS/MS method to resolve the identification of BMAA
2007) by reverse-phase elution (Waters Nova-Pak C18 column, 3.9 mm 300 mm)
in the shark fin cartilage products tested in this report.
on a Waters 1525 Binary HPLC pump and a Waters 717 autosampler. The mobile
phase consisted of 140 mM sodium acetate, 5.6 mM triethylamine, pH 5.7 (Eluent The HPLC-FD method validation for total BMAA in shark carti-
A), and 52% (v/v) aqueous acetonitrile (Eluent B) using a flow rate of 1.0 mL/min, lage supplements is shown in Table 1. Linearity was determined
and 10 lL sample injection volume. The samples were eluted using a 60 min gradi- for BMAA standard concentrations ranging from 0.05 to 100
ent: 0.0 min 100% A; 2.0 min 90% A; 5.0 min 86% A; 10.0 min 86% A; 18.0 min 73% A; ng/lL. These results demonstrate linearity over this range with a
30.0 min 57% A; 35.0 min 40% A; 37.5 min 100% B; 47.5 min 100% B; 50.0 min 100%
A; 60.0 min 100% A. Detection of the AQC fluorescent tag was achieved using a
correlation coefficient R2 = 0.998. The LOD and LOQ were 3.6 and
Waters 2475 Multi k-Fluorescence Detector with excitation at 250 nm and emission 10 ng, respectively. The percentage recovery was determined from
at 395 nm. analytical runs of post hydrolysis spike-in BMAA standards assayed
28 K. Mondo et al. / Food and Chemical Toxicology 70 (2014) 26–32
Table 1
HPLC-FD assay validation parameters.
HPLC-FD
Linearity
Lowest concentration 0.025 ng
Highest concentration 50 ng
Correlation coefficient (R2) 0.998
Detection and quantification limits
LOD 2.8 ng
LOQ 8.4 ng
Selectivity RT (min)
AEG 29.131
BMAA 30.575
DAB 32.663
Precision RSD %
Intraday (n = 4), response 1.4
Intraday (n = 4), RT 0.04
Interday (n = 4), response 2.3
Interday (n = 4), RT 0.1
Recovery % 96 ± 6
BMAA, b-N-methylamino-L-alanine; AEG, N-(2-aminoethyl) glycine; DAB, 2,4-

diaminobutyric acid; LOD, limit of detection; LOQ, limit of quantification, RT,
retention time: RSD, relative standard deviation.
in a reference matrix of shark cartilage. The estimated percentage

of BMAA recovery was 96% (6.2% RSD) for the AQC-tagged HPLC-FD
method. BMAA was visibly separated by HPLC-FD and distin-
guished from reference standard amino acids and its structural iso-
mers AEG and DAB (Fig. 1). The relative retention time of BMAA,
AEG, and DAB was 29.1, 30.6 and 32.7 min, respectively. Relative
retention times of BMAA and its related diamino acids are in close
agreement with previous reports (Banack et al., 2010; Spacil et al., Fig. 1. HPLC-FD chromatograms after 6-aminoquinolyl-N-hydroxysuccinimidyl
2010). carbamate (ACQ) derivatization of (A) diamino acids: N-2(amino)ethylglycine
The unambiguous detection, identification, and quantification (AEG), b-N-methylamino-L-alanine (BMAA), and 2,4-diaminobutyric acid (DAB);
(B) amino and diamino acids: valine (Val), methionine (Met), AEG, BMAA, DAB; (C
of BMAA in complex biological samples requires mass spectrome- and D) powder sample number 9 spiked with AEG, BMAA, and DAB reference
try validation and AQC derivatization to distinguish BMAA from its standards.
positional isomers DAB and AEG (Cohen, 2012; Duncan, 2012;
Jiang et al., 2012, 2013). Therefore, we conducted analysis of all
shark cartilage samples analyzed by UPLC–MS/MS and indepen- to cells (Weiss et al., 1989), that is produced by cyanobacteria
dent laboratory confirmation of the HPLC-FD results. Linearity of (Li et al., 2012; Rosen and Hellenas, 2008; Spacil et al., 2010).
the UPLC–MS/MS system was determined by injecting standards AEG has been detected in diverse strains of marine cyanobacteria
ranging from 0.06 pg/lL to 1.0 ng/lL. The lower limits of detection (Banack et al., 2010, 2012; Jiang et al., 2013), which may account
were 1.1 pg/lL for BMAA, 1.2 pg/lL for AEG and 0.8 pg/lL for for the levels measured in these commercial products.
DAB. The linear range for quantification was 0.2 pg/lL (lower All sixteen shark cartilage dietary supplements tested in this
LOQ) to 0.6 ng/lL (upper LOQ) for BMAA (R2 = 0.9697) and AEG study contained low levels of total Hg (Table 4). The mean concen-
(R2 = 0.9788), and 0.2 pg/lL (lower LOQ) to 0.2 ng/lL (upper LOQ) tration of total Hg found in the samples was 0.046 lg/g, ranging
for DAB (R2 = 0.999). Retention time of BMAA, AEG, and DAB were from 0.003 to 0.135 lg/g (dry weight). The commercial samples
6.55 min (0.23% RSD), 6.66 min (0.22% RSD), and 6.82 min (0.26% tested had total Hg concentrations that were in agreement with
RSD) respectively. Discrimination between isomers was achieved prior reports albeit some samples were significantly lower (Dolan
using product ions m/z 459 (AQC double-labeled BMAA, AEG and et al., 2003; Kim, 2004; Leblond et al., 2008). The manufacturer
DAB), m/z 289 (AQC singled-labeled BMAA, AEG and DAB), m/z recommended average intake of shark cartilage supplement based
171 (AQC tag), m/z 119 (native BMAA, AEG and DAB). To increase on the product label is 4.4 g. Thus, the total Hg levels reported here
specificity and selectivity, the diagnostic ions were selected for are below the maximum allowable level (MAL) of 1.5 lg/g (JECFA,
BMAA (m/z 258), AEG (m/z 214), and DAB (m/z 188) as previously 2011; USP, 2013).
described (Banack et al., 2010; Jiang et al., 2012) for analysis of ref-
erence compound standards and commercial sample number 15
(Fig. 2 and Table 2). 4. Discussion
BMAA was detected and quantified in fifteen out of sixteen
samples analyzed by HPLC-FD at concentrations ranging from 86 Dietary supplements of marine origin are experiencing a large
to 265 lg/g dry weight (Table 3). These results were confirmed increase in consumption as a result of perceived health benefits
for each sample using UPLC–MS/MS (Table 3). BMAA concentra- and their direct marketing to consumers. Shark cartilage is not a
tions in the shark cartilage powders were comparable to the values defined single product, but contains proteins, proteoglycans and
reported for shark fins sampled from South Florida coastal waters rough elastic cartilage, carbohydrates and lipids (Institute of
(Mondo et al., 2012). All of the shark cartilage powder sampled in Medicine and National Research Council, 2005). Shark cartilage
this study had detectable levels of the BMAA isomers AEG and DAB extracts and powder formulations may include extensive washing,
(Table 3). DAB is a structural isomer of BMAA known to be neurotoxic soaking and bleaching, which removes water-soluble components
Fig. 2. UPLC–MS/MS chromatograms of (A) 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (ACQ) derivatized BMAA, AEG, and DAB standards, (B) shark cartilage sample
number 15. Parent ion m/z 459 and daughter ions from m/z 459 to 289, 258, 214, 188, 171, and 119 are shown. The diagnostic selected reaction monitoring (SRM) transitions
of BMAA, AEG and DAB were 459 > 258, 459 > 214, 459 > 188, respectively.
Table 2 However, our study demonstrates that shark fin products may con-
Selected reaction monitoring (SRM) ion ratios of BMAA and its isomers in shark tain neurotoxic compounds, including BMAA, DAB, and Hg.
cartilage. The bioaccumulation properties of algae, coral, krill and shark
Standard Sample number 15 can potentially lead to high concentrations of health-damaging
marine contaminants. BMAA is a neurotoxic amino acid produced
SRM ion ratio RT (min) SRM ion ratio RT (min)
by almost all blue-green algae that are ubiquitous to lakes and
BMAA
oceans (Cox et al., 2005). BMAA is biomagnified with increasing
119:258 7.6 6.57 6.9 6.55
AEG accumulation through higher trophic levels of the food chain (Cox
119:214 19.2 6.69 14.3 6.66 et al., 2003). Our objective was not to evaluate the risk to purported
DAB benefits of human use of shark cartilage products, but to provide
119:188 5.1 6.84 4.0 6.82 the first measures of BMAA, its structural isomers, and total Hg
within a selected sample of available commercial products. At pres-
ent, no human risk assessment of BMAA exposures can be made
that may include marine neurotoxicants. However, there is a because of the extremely limited information on this non-protein
substantial lack of safety data for shark cartilage both clinically amino acid in mammals. There is a lack of standard toxicological
and in animal toxicology studies, including a deficiency of any seri- tests using the oral route of BMAA exposure, which is most relevant
ous adverse event reports (Institute of Medicine and National to environmental exposure health risk assessments.
Research Council, 2005; NCI, 2013). At present, no active ingredient BMAA misincorporation into proteins induces apoptosis in neu-
of shark cartilage has been defined, despite a number of scientific ronal cells (Dunlop et al., 2013) and endoplasmic reticulum stress
studies, which have demonstrated angiogenesis inhibition that leads to cell death (Okle et al., 2013). BMAA has been mea-
in vitro (Lu et al., 2010; Patra and Sandell, 2012). Based on limited sured in proteins from cyanobacteria and other organisms (Cox
animal and in vitro data, currently there is no evidence that et al., 2003). BMAA and other non-protein amino acids can mimic
consumption of shark cartilage products produce neurotoxicity. one of the 20 canonical amino acids and become misincorporated
Table 3
Comparison of HPLC-FD (left) and UPLC–MS/MS (right) quantification of BMAA and its structural isomers DAB and AEG in shark cartilage. ND, below limits of detection.
Sample no. BMAA (lg/g) ± SE DAB (lg/g) ± SE AEG (lg/g) ± SE BMAA (lg/g) ± SE DAB (lg/g) ± SE AEG (lg/g) ± SE
1 180.9 ± 21 164.1 ± 12 959.5 ± 52 148.4 ± 6.5 139.8 ± 13 1885.1 ± 25
2 210.8 ± 30 125.0 ± 10 847.8 ± 58 113.8 ± 5.9 115.5 ± 1.3 1663.2 ± 21
3 161.6 ± 16 139.8 ± 9 817.3 ± 38 203.3 ± 9.1 118.9 ± 7.3 1729.3 ± 28
4 264.9 ± 19 128.5 ± 5 994.6 ± 24 212.2 ± 42 335.3 ± 167 1578.1 ± 3.8
5 246.3 ± 24 123.5 ± 5 894.7 ± 22 173.5 ± 12 158.7 ± 14 1654.1 ± 66
6 155.0 ± 15 122.3 ± 9 916.6 ± 31 93.0 ± 6.8 150.9 ± 8.9 1666.8 ± 85
7 92.4 ± 4 123.7 ± 6 889.1 ± 23 75.1 ± 12 495.5 ± 27 1710.8 ± 118
8 211.8 ± 7 113.3 ± 1 729.3 ± 2 352.2 ± 59 320.5 ± 60 1765.1 ± 82
9 108.2 ± 9 207.7 ± 17 1226.8 ± 72 329.3 ± 12 233.3 ± 50 1565.7 ± 9.2
10 115.9 ± 5 147.6 ± 6 1338.7 ± 43 103.2 ± 6.9 175.2 ± 17 1590.2 ± 57
11 83.6 ± 2 114.8 ± 7 1266.6 ± 18 74.8 ± 5.5 567.9 ± 33 1696.5 ± 74
12 93.2 ± 9 106.2 ± 3 831.9 ± 39 186.2 ± 5.0 1483.4 ± 118 1611.6 ± 30
13 211.2 ± 15 118.6 ± 9 766.81 ± 31 168.6 ± 15 206.6 ± 4 1298.4 ± 76
14 ND 53.1 ± 2 703.4 ± 30 ND 69.2 ± 4.3 1512.1 ± 103
15 86 ± 5 65.3 ± 2 841.2 ± 45 111.5 ± 4.0 382.0 ± 33 1496.2 ± 34
16 138.3 ± 5 111.4 ± 2 1080.6 ± 25 192.0 ± 5.7 134.8 ± 7.1 1609.6 ± 44
Mean 157.5 122.8 944.1 169.1 318.0 1627.1
Range 86–265 53–207 703–1339 75–352 69–1483 1298–1729
Table 4
Total mercury concentration in shark cartilage supplements.
No. Hga (lg/g) Max. daily intake b

(lg/day) PDI c
(lg/kg bw/day) % PTDId
1 0.050 0.220 0.004 0.64
2 0.135 0.594 0.010 1.74
3 0.096 0.422 0.007 1.24
4 0.018 0.079 0.001 0.23
5 0.018 0.079 0.001 0.23
6 0.060 0.264 0.004 0.77
7 0.018 0.079 0.001 0.23
8 0.072 0.317 0.005 0.93
9 0.089 0.392 0.007 1.15
10 0.029 0.128 0.002 0.37
11 0.018 0.079 0.001 0.23
12 0.027 0.119 0.002 0.35
13 0.012 0.053 0.001 0.15
14 0.003 0.013 0.0002 0.04
15 0.040 0.176 0.003 0.51
16 0.044 0.194 0.003 0.57
a
Total Hg concentrations detected in commercial shark cartilage supplements.
b
The maximum daily intake of total Hg from of shark cartilage supplements based upon manufacturer’s label.
c
The probable daily intake (PDI) of total Hg for a 60 kg consumer if the maximum daily direction of is followed.
d
The percent provisional tolerable daily intake (PTDI) of total Hg for a 60 kg consumer if the maximum daily direction is followed. The calculation is based the on
provisional tolerable weekly intake (PTWI) for Hg (4 lg/kg bw/week) and a PTDI of 0.57 lg/kg bw/day for metals (WHO, 2011).
into proteins (Dunlop et al., 2013; Xie et al., 2013; for review The major dietary sources contributing to the ingestion of
Hendrickson et al., 2004). Expression of mutant proteins disrupts BMAA and Hg are fish and shellfish of the aquatic food webs
protein folding, a mechanism characteristic of many neurodegen- (Brand et al., 2010; Groth, 2010; Jonasson et al., 2010; Mondo
erative diseases, including Alzheimer’s disease and ALS (Jucker et al., 2012; Pethybridge et al., 2010). Sharks like humans are
and Walker, 2011; Trojsi et al., 2013). Misincorporation of BMAA long-lived apex predators that bioaccumulate Hg (Delshad et al.,
in neural proteins with chronic dietary exposure over the lifespan 2012; Escobar-Sanchez et al., 2010; Pethybridge et al., 2010). Com-
is a possible environmental pathway for recurrent neuronal insults mercial marketing of shark cartilage supplements are often not
(Bradley and Mash, 2009; Murch et al., 2004a). clearly defined by the manufacturers and may include tough elastic
The estimated total Hg exposure through consumption of shark cartilage and exoskeleton from various species of shark harvested
cartilage supplements analyzed in this report is below the tolerable across different geographical locales and combined into a single
weekly intake estimated for humans who consume contaminated product. Shark cartilage is a dietary supplement sold with high
fish. However, a recent study by Rush et al., 2012 showed that healing benefits for diseases like cancer, arthritis, and osteoarthri-
when BMAA and methyl Hg are combined in vitro, they have syn- tis that contain two angiogenesis inhibitors (Hassan et al., 2005;
ergistic neurotoxic effects that include decreases in the levels of Patra and Sandell, 2012; Ritter, 2004). Because these supplements
cellular glutathione, an important neuroprotective brain antioxi- are added to the normal daily food ingestion, their use should be
dant. Methyl Hg concentrations in fish muscle are directly propor- made with some caution to avoid inadvertent neurotoxin
tional to total Hg concentrations (Kannan et al., 1998) and exposures. The consumer should pay special attention to the good
estimates suggest that 95% of the total Hg in shark muscle is the manufacturing product quality control of the supplier and be
more toxic compound methyl Hg (Krystek and Ritsema, 2005). aware that shark cartilage products carry a risk for exposure to
Although not directly measured in our report, concentrations of BMAA and its structural isomers. Recent experimental findings
methyl Hg in sharks collected along the Florida coast often suggest that exposure to environmental toxicants like BMAA and
exceeded the U.S. FDA action level of 1 lg/g (Hueter et al., 1995). Hg may trigger a neurodegenerative process in genetically
predisposed people (Bradley and Mash, 2009 for review, Trojsi Hassan, Z.M., Feyzi, R., Sheikhian, A., Bargahi, A., Mostafaie, A., Mansouri, K.,
Shahrokhi, S., Ghazanfari, T., Shahabi, S., 2005. Low molecular weight fraction of
et al., 2013). Although the potential health risk for BMAA expo-
shark cartilage can modulate immune responses and abolish angiogenesis. Int.
sures is currently unknown, shark cartilage products may contain Immunopharmacol. 5, 961–970.
two marine contaminants that have known synergistic CNS Hendrickson, T.L., de Crecy-Lagard, V., Schimmel, P., 2004. Incorporation of
toxicities. nonnatural amino acids into proteins. Ann. Rev. Biochem. 73, 147–176.
Himmel, P.B., Seligman, T.M., 1999. Treatment of systemic sclerosis with shark
cartilage extract. J. Orthomol. Med. 14, 73–77.
Holt, S., 1995. Shark cartilage and nutriceutical update. Altern. Complem. Ther. 1,
Conflict of Interest
414–416.
Hueter, R.E., Fong, W.G., Henderson, G., French, M.F., Manire, C.A., 995.
The authors declare that there are no conflicts of interest. Methylmercury concentration in shark muscle by species, size and
distribution of sharks in Florida coastal waters. Air Soil Pollut. 80, 893–899.
Institute of Medicine and National Research Council, 2005. Dietary Supplements: A
Transparency Document Framework for Evaluating Safety. The National Academies Press, Washington,
DC.
JECFA, 2011. Evaluation of Certain Food Additives and Contaminants, 72nd Report
The Transparency document associated with this article can be of Joint FAO/WHO Expert Committee on Food Additives.
found in the online version. Jiang, L., Aigret, B., De Borggraeve, W.M., Spacil, Z., Ilag, L.L., 2012. Selective LC–MS/
MS method for the identification of BMAA from its isomers in biological
samples. Anal. Bioanal. Chem. 403, 1719–1730.
Jiang, L., Johnston, E., Aberg, K.M., Nilsson, U., Ilag, L.L., 2013. Strategy for
Acknowledgements
quantifying trace levels of BMAA in cyanobacteria by LC/MS/MS. Anal.
Bioanal. Chem. 405, 1283–1292.
The Herbert W. Hoover Foundation (USA) provided the funding Jonasson, S., Eriksson, J., Berntzon, L., Spacil, Z., Ilag, L.L., Ronnevi, L.O., Rasmussen,
for this research study. We are grateful to the commercial suppliers U., Bergman, B., 2010. Transfer of a cyanobacterial neurotoxin within a
temperate aquatic ecosystem suggests pathways for human exposure. Proc.
who provided shark cartilage products for analytical tests for Natl. Acad. Sci. USA 107, 9252–9257.
BMAA and its structural isomers. Jucker, M., Walker, L.C., 2011. Pathogenic protein seeding in Alzheimer disease and
other neurodegenerative disorders. Ann. Neurol. 70, 532–540.
Kannan, K., Smith Jr., R.G., Lee, R.F., Windom, H.L., Heitmuller, P.T., Macauley, J.M.,
References Summers, J.K., 1998. Distribution of total mercury and methyl mercury in
water, sediment, and fish from south Florida estuaries. Arch. Environ. Contam.
Banack, S.A., Cox, P.A., 2003. Biomagnification of cycad neurotoxins in flying foxes: Toxicol. 34, 109–118.
implications for ALS-PDC in Guam. Neurology 61, 387–389. Kim, M., 2004. Mercury, cadmium and arsenic contents of calcium dietary
Banack, S.A., Downing, T.G., Spacil, Z., Purdie, E.L., Metcalf, J.S., Downing, S., supplements. Food Addit. Contam. 21, 763–767.
Esterhuizen, M., Codd, G.A., Cox, P.A., 2010. Distinguishing the cyanobacterial Krystek, P., Ritsema, R., 2005. Mercury speciation in thawed out and refrozen fish
neurotoxin beta-N-methylamino-L-alanine (BMAA) from its structural isomer samples by gas chromatography coupled to inductively coupled plasma mass
2,4-diaminobutyric acid (2,4-DAB). Toxicon 56, 868–879. spectrometry and atomic fluorescence spectroscopy. Anal. Bioanal. Chem. 381,
Banack, S.A., Johnson, H.E., Cheng, R., Cox, P.A., 2007. Production of the neurotoxin 354–359.
BMAA by a marine cyanobacterium. Mar. Drugs 5, 180–196. Leblond, C., Mephara, J., Sauvé, S., 2008. Trace metals (Cd, Co, Cr, Cu, Hg, Ni, Pb, and
Banack, S.A., Metcalf, J.S., Jiang, L., Craighead, D., Ilag, L.L., Cox, P.A., 2012. Zn) in food supplements of marine origin. Hum. Ecol. Risk Assess. 14, 408–420.
Cyanobacteria produce N-(2-aminoethyl)glycine, a backbone for peptide Li, A., Fan, H., Ma, F., McCarron, P., Thomas, K., Tang, X., Quilliam, M.A., 2012.
nucleic acids which may have been the first genetic molecules for life on Elucidation of matrix effects and performance of solid-phase extraction for LC–
Earth. PLoS ONE 7, e49043. MS/MS analysis of beta-N-methylamino-L-alanine (BMAA) and 2,4-
Bell, E.A., 2009. The discovery of BMAA, and examples of biomagnification and diaminobutyric acid (DAB) neurotoxins in cyanobacteria. Analyst 137, 1210–
protein incorporation involving other non-protein amino acids. Amyotrop. 1219.
Later. Scleros. 10 (Suppl. 2), 21–25. Loprinzi, C.L., Levitt, R., Barton, D.L., Sloan, J.A., Atherton, P.J., Smith, D.J., Dakhil, S.R.,
Betz, J.M., Brown, P.N., Roman, M.C., 2011. Accuracy, precision, and reliability Moore Jr., D.F., Krook, J.E., Rowland Jr., K.M., et al., 2005. Evaluation of shark
of chemical measurements in natural products research. Fitoterapia 82, cartilage in patients with advanced cancer: a North Central Cancer Treatment
44–52. Group trial. Cancer 104, 176–182.
Bradley, W.G., Mash, D.C., 2009. Beyond Guam: the cyanobacteria/BMAA hypothesis Lu, C., Lee, J.J., Komaki, R., Herbst, R.S., Feng, L., Evans, W.K., Choy, H., Desjardins, P.,
of the cause of ALS and other neurodegenerative diseases. Amyotrop. Later. Esparaz, B.T., Truong, M.T., et al., 2010. Chemoradiotherapy with or without AE-
Scleros. 10 (Suppl. 2), 7–20. 941 in stage III non-small cell lung cancer: a randomized phase III trial. J. Natl.
Brand, L.E., Pablo, J., Compton, A., Hammerschlag, N., Mash, D.C., 2010. Cancer Inst. 102, 859–865.
Cyanobacterial blooms and the occurrence of the neurotoxin beta-N- Milner, M., 1999. A guide to the use of shark cartilage in the treatment of arthritis
methylamino-L-alanine (BMAA) in South Florida aquatic food webs. Harmf. and other inflammatory joint disease. Am. Chiropractor 21, 40–42.
Algae 9, 620–635. Mondo, K., Hammerschlag, N., Basile, M., Pablo, J., Banack, S.A., Mash, D.C., 2012.
Cohen, S.A., 2012. Analytical techniques for the detection of alpha-amino-beta- Cyanobacterial neurotoxin beta-N-methylamino-L-alanine (BMAA) in shark
methylaminopropionic acid. Analyst 137, 1991–2005. fins. Mar. Drugs 10, 509–520.
Cox, P.A., Banack, S.A., Murch, S.J., 2003. Biomagnification of cyanobacterial Murch, S.J., Cox, P.A., Banack, S.A., 2004a. A mechanism for slow release of
neurotoxins and neurodegenerative disease among the Chamorro people of biomagnified cyanobacterial neurotoxins and neurodegenerative disease in
Guam. Proc. Natl. Acad. Sci. USA 100, 13380–13383. Guam. Proc. Natl. Acad. Sci. USA 101, 12228–12231.
Cox, P.A., Banack, S.A., Murch, S.J., Rasmussen, U., Tien, G., Bidigare, R.R., Metcalf, J.S., Murch, S.J., Cox, P.A., Banack, S.A., Steele, J.C., Sacks, O.W., 2004b. Occurrence of
Morrison, L.F., Codd, G.A., Bergman, B., 2005. Diverse taxa of cyanobacteria beta-methylamino-L-alanine (BMAA) in ALS/PDC patients from Guam. Acta
produce beta-N-methylamino-L-alanine, a neurotoxic amino acid. Proc. Natl. Neurol. Scand. 110, 267–269.
Acad. Sci. USA 102, 5074–5078. NCI, 2013. National Cancer Institute. PDQÒ Cartilage (Bovine and Shark). National
Delshad, S.T., Mousavi, S.A., Islami, H.R., Pazira, A., 2012. Mercury concentration of Cancer Institute 03/28/2013.
the whitecheek shark, Carcharhinus dussumieri (Elasmobranchii, Okle, O., Stemmer, K., Deschl, U., Dietrich, D.R., 2013. L-BMAA induced ER stress and
Chondrichthyes), and its relation with length and sex. Pan. Am. J. Aquat. Sci. enhanced caspase 12 cleavage in human neuroblastoma SH-SY5Y cells at low
7, 135–142. nonexcitotoxic concentrations. Toxicol. Sci. 131, 217–224.
Dolan, S.P., Nortrup, D.A., Bolger, P.M., Capar, S.G., 2003. Analysis of dietary Pablo, J., Banack, S.A., Cox, P.A., Johnson, T.E., Papapetropoulos, S., Bradley, W.G.,
supplements for arsenic, cadmium, mercury, and lead using inductively coupled Buck, A., Mash, D.C., 2009. Cyanobacterial neurotoxin BMAA in ALS and
plasma mass spectrometry. J. Agric. Food Chem. 51, 1307–1312. Alzheimer’s disease. Acta Neurol. Scand. 120, 216–225.
Duncan, M.W., 2012. Good mass spectrometry and its place in good science. J. Mass Patra, D., Sandell, L.J., 2012. Antiangiogenic and anticancer molecules in cartilage.
Spectrom. 47, 795–809. Expert Rev. Mol. Med. 14, e10.
Dunlop, R.A., Cox, P.A., Banack, S.A., Rodgers, K.J., 2013. The non-protein amino acid Pethybridge, H., Cossa, D., Butler, E.C., 2010. Mercury in 16 demersal sharks from
BMAA is misincorporated into human proteins in place of L-serine causing southeast Australia: biotic and abiotic sources of variation and consumer health
protein misfolding and aggregation. PLoS ONE 8, e75376. implications. Mar. Environ. Res. 69, 18–26.
Escobar-Sanchez, O., Galvan-Magana, F., Rosiles-Martinez, R., 2010. Mercury and Ritter, S., 2004. A Fishy Tale. Chemical & Engineering News 20:80 December 20.
selenium bioaccumulation in the smooth hammerhead shark, Sphyrna zygaena Rosen, J., Hellenas, K.E., 2008. Determination of the neurotoxin BMAA (beta-N-
Linnaeus, from the Mexican Pacific Ocean. Bull. Environ. Contam. Toxicol. 84, methylamino-L-alanine) in cycad seed and cyanobacteria by LC–MS/MS (liquid
488–491. chromatography tandem mass spectrometry). Analyst 133, 1785–1789.
Groth 3rd, E., 2010. Ranking the contributions of commercial fish and shellfish Rush, T., Liu, X., Lobner, D., 2012. Synergistic toxicity of the environmental
varieties to mercury exposure in the United States: implications for risk neurotoxins methylmercury and beta-N-methylamino-L-alanine. NeuroReport
communication. Environ. Res. 110, 226–236. 23, 216–219.
Spacil, Z., Eriksson, J., Jonasson, S., Rasmussen, U., Ilag, L.L., Bergman, B., 2010. Weiss, J.H., Christine, C.W., Choi, D.W., 1989. Bicarbonate dependence of glutamate
Analytical protocol for identification of BMAA and DAB in biological samples. receptor activation by beta-N-methylamino-L-alanine: channel recording and
Analyst 135, 127–132. study with related compounds. Neuron 3, 321–326.
Trojsi, F., Monsurro, M.R., Tedeschi, G., 2013. Exposure to environmental toxicants WHO, 2011. Safety evaluation of certain contaminants in food. In: Seventy-second
and pathogenesis of amyotrophic lateral sclerosis: state of the art and research Meeting of the Joint FAO/WHO Expert Committee on Food Additives, WHO
perspectives. Int. J. Mol. Sci. 14, 15286–15311. Press, Geneva, pp. 605–673.
USEPA, 2002. Method 1631, Revision E: Mercury in Water by Oxidation, Purge and Xie, X., Basile, M., Mash, D.C., 2013. Cerebral uptake and protein incorporation of
Trap, and Cold Vapor Atomic Fluorescence Spectrometry. U.E.P.A. EPA (Ed.). cyanobacterial toxin beta-N-methylamino-L-alanine. NeuroReport 24, 779–784.
USP, 2013. Elemental Impurities – Limits (232) and Elemental Impurities –
Procedures (233). United States Pharmacopeial Concention.
toxins
Article
Cyanobacterial Neurotoxin BMAA and Mercury
in Sharks
Neil Hammerschlag 1,2 , David A. Davis 3 , Kiyo Mondo 3 , Matthew S. Seely 3 , Susan J. Murch 4 ,
William Broc Glover 4 , Timothy Divoll 5 , David C. Evers 5 and Deborah C. Mash 3,6, *
1 Rosensteil School of Marine and Atmospheric Science, University of Miami, Miami, FL 33149, USA;
nhammerschlag@rsmas.miami.edu
2 Leonard and Jayne Abess Center for Ecosystem Science and Policy, University of Miami, Coral Gables,
FL 33124, USA
3 Department of Neurology, Miller School of Medicine, University of Miami, Miami, FL 33136, USA;
d.davis12@med.miami.edu (D.A.D.); kmondo@hiroshima-u.ac.jp (K.M.); mxs1904@miami.edu (M.S.S.)
4 Department of Chemistry, 3247 University Way, University of British Columbia, Kelowna, BC V1V 1V7,
Canada; susan.murch@ubc.ca (S.J.M.); williambrocglover@gmail.com (W.B.G.)
5 Biodiversity Research Institute, 276 Canco Road, Portland, ME 04103, USA; tim.divoll@briloon.org (T.D.);
david.evers@briloon.org (D.C.E.)
6 Department of Molecular and Cellular Pharmacology, Miller School of Medicine, University of Miami,
Miami, FL 33136, USA
* Correspondence: dmash@med.miami.edu; Tel.: +1-305-243-5888; Fax: +1-305-243-3649
Academic Editor: Luis Botana

Received: 28 April 2016; Accepted: 26 July 2016; Published: 16 August 2016
Abstract: Sharks have greater risk for bioaccumulation of marine toxins and mercury (Hg), because
they are long-lived predators. Shark fins and cartilage also contain β-N-methylamino-L-alanine
(BMAA), a ubiquitous cyanobacterial toxin linked to neurodegenerative diseases. Today, a significant
number of shark species have found their way onto the International Union for Conservation of
Nature (IUCN) Red List of Threatened Species. Many species of large sharks are threatened with
extinction due in part to the growing high demand for shark fin soup and, to a lesser extent, for shark
meat and cartilage products. Recent studies suggest that the consumption of shark parts may be a
route to human exposure of marine toxins. Here, we investigated BMAA and Hg concentrations
in fins and muscles sampled in ten species of sharks from the South Atlantic and Pacific Oceans.
BMAA was detected in all shark species with only seven of the 55 samples analyzed testing below
the limit of detection of the assay. Hg concentrations measured in fins and muscle samples from the
10 species ranged from 0.05 to 13.23 ng/mg. These analytical test results suggest restricting human
consumption of shark meat and fins due to the high frequency and co-occurrence of two synergistic
environmental neurotoxic compounds.
Keywords: β-N-methylamino-L-alanine; conservation; cyanobacteria; total mercury; methylmercury;

neurodegenerative disease; neurotoxin; sharks
1. Introduction
Sharks are exploited in both target fisheries [1–3] and as bycatch (both discarded and incidental
catch) that is also sold [4–6]. The estimates suggest total annual mortality of 100 million sharks killed
in 2000 and about 97 million sharks in 2010, with a total range between 63 and 273 million per year [7].
At least 126 countries worldwide catch sharks, and the global annual value of trade in shark parts is
approximately $1 billion US. [8]. Though sharks are harvested for meat consumption and/or for their
cartilage used in alternative medicine products, the largest driver of shark mortality is directed fishing
to obtain their fins for human consumption in shark fin soup [9–12]. Given their relatively low natural
Toxins 2016, 8, 238; doi:10.3390/toxins8080238 www.mdpi.com/journal/toxins

Toxins 2016, 8, 238 2 of 14
population growth rates, many sharks are undergoing population declines [5,7] rendering about 16%
of the ocean’s shark species threatened with extinction [13].
Shark fin soup primarily consumed in China is also a delicacy in other Asian countries and
their diaspora communities worldwide [11,12]. Records from the Chinese Song Dynasty (960–1279)
describe the use of shark fin soup as a traditional banquet staple [14]. Today, shark fin soup is in
increasingly high demand, popular at weddings and other celebrations across Asia [15]. Dietary
supplements containing shark cartilage, the health benefits of which are purportedly bolstered by
traditional Chinese medicine claims, have gained popularity in western nations. However, the U.S.
Food and Drug Administration (FDA) has been unable to confirm any proclaimed benefits [16] and
available reports of health benefits are questionable [17].
There is growing concern as to the potential negative health consequences associated with
consuming shark parts, including fins, meat and cartilage. The neurotoxic compound methyl
Hg (MeHg) has been known to bioaccumulate in sharks over their lifespans [18–21]. As such,
Hg levels in shark muscle often exceed advisory guidelines for safe human consumption [21–24].
For example, the Florida Department of Health (FDOH) advises that people should not eat sharks
greater than ~109 cm and further recommends that children and pregnant woman not eat any shark
meat [23]. Moreover, recent studies have reported that commercial shark cartilage supplements
contain pro-inflammatory compounds that could pose health risks for consumers, especially those
with inflammatory diseases [17].
Recently, the cyanobacterial neurotoxin β-N-methylamino-L-alanine (BMAA), has been detected
in shark fins [25] and shark cartilage supplements [16]. BMAA has been linked to amyotrophic lateral
sclerosis/Parkinsonism dementia complex (ALS/PDC) of Guam and has been detected in the brains
of North American Alzheimer’s disease and ALS patients [26,27], suggesting that BMAA plays a role
as an environmental toxin in neurodegenerative disease. Recent evidence suggests that merely living
near a body of water with cyanobacterial blooms, which contaminate the water, fish, and even the air,
may increase the risk of developing ALS [28]. In vitro exposures have demonstrated BMAA’s acute
neurotoxicity and animal studies show that BMAA exposure leads to motor impairments in rats [29,30].
Thus, the consumption of shark fins and dietary cartilage supplements may pose a risk for human
exposure to environmental neurotoxins BMAA and Hg [16,25].
A causal role for BMAA toxicity in humans is still uncertain due to a lack of epidemiological data
with human intake estimated from dietary exposures. Thus, it remains unclear whether detection of
BMAA in shark fin or cartilage supplements on its own poses a threat to human health. Likewise,
it has been noted that the concentrations of MeHg found in fish and marine products are unlikely
to cause significant adverse CNS health effects [31]. However, a synergistic toxicity between these
two neurotoxic compounds has been suggested, since BMAA concentrations in a range of 10–100 µM
were potentiated by MeHg (3 µM) when these were combined. BMAA and MeHg have been shown to
decrease the main cellular antioxidant glutationine, which would increase vulnerability of the brain
to oxidative stress [31]. Recent studies by Cox and coworkers demonstrate that vervet monkeys fed
BMAA for 140 days develop neurofibrillary tangles and β-amyloid deposits in the brain similar to
what is seen in patients with neurodegenerative diseases, including ALS and Alzheimer’s disease [32].
Given the potential synergistic toxicity of Hg and BMAA and their likely prevalence in top marine
predators, we conducted an expanded analysis to test fin and muscle from an opportunistic sample
of 10 different shark species collected from different ocean basins. Shark samples were analyzed
for BMAA using high performance liquid chromatography with fluorescence detection (HPLC-FD)
and Hg concentrations were quantified by cold vapor atomic fluorescence spectrometry (CVAFS).
Since the total Hg (THg) that is measured in shark muscle and fin is mostly in the form of MeHg+,
measures of total Hg are generally equivalent to MeHg+ [33–35] Here, THg concentrations in fin
and muscle samples were measured and compared to MeHg+ for confirmation of levels in select
samples. Our results demonstrated that all 10 shark species tested positive for both BMAA and Hg.
Independent laboratory confirmation of BMAA and its isomers 2, 4-diaminobutyric acid (DAB) and
Toxins 2016, 8, 238 3 of 14
N-(2-aminoethyl) glycine (AEG) was determined by ultra-performance liquid chromatography/mass

spectrometry/mass spectrometry (UPLC-MS/MS).
2. Results and Discussion

A total of 55 sharks were analyzed for contaminations of BMAA and Hg in selected fin and/or
muscle. Our cohort contained 10 different shark species sampled from the Atlantic and the Pacific
Ocean (Table 1). These shark species sampled range in threat status from Least Concern (bonnethead
shark) to Endangered (great hammerhead) by the International Union for Conservation of Nature
(IUCN). Several species (tiger, great hammerhead, and bull) are known to be common in the shark fin
trade [12], and the fins and meat of all species sampled are subject to exploitation (Table 1) [36].
Table 1. A summary of shark species, sampling times and locations sites.
International Union for

Scientific Name Conservation of Nature Common Name Location Month
Red List Category
Carharhinus acronotus Near Threatened Blacknose b 25.09417o N 81.04234o W March
- - Blacknose b 25.00858o N 81.00089o W April
- - Blacknose a 25.62099o N 80.15602o W October
- - Blacknose a Biscayne Bay June
- - Blacknose b 25.09417o N 81.04234o W December
- - Blacknose b 25.01089o N 81.00419o W April
Carcharhinus limbatus Near Threatened Blacktip b 25.00644o N 80.99969o W March
- - Blacktip b 25.00644o N 80.99969o W September
- - Blacktip a 25.59968o N 80.15205o W July
- - Blacktip b 25.00644o N 80.99969o W March
- - Blacktip a 25.62592o N 80.15442o W October
- - Blacktip a 25.61905o N 80.1714o W October
- - Blacktip a 25.64757o N 80.1881o W April
- - Blacktip a 25.67199o N 80.18144o W September
- - Blacktip b 25.01715o N 81.01056o W October
- - Blacktip b 25.01715o N 81.01056o W February
- - Blacktip b 25.01089o N 81.00419o W April
- - Blacktip b 25.00858o N 81.00089o W December
- - Blacktip b 25.00623o N 80.99723o W March
Sphyrna tiburo Least concerned Bonnethead a 25.36711o N 80.14806o W March
- - Bonnethead a 25.36711o N 80.14806o W March
- - Bonnethead a 25.40807o N 80.21806o W October
- - Bonnethead b 25.36711o N 80.14806o W March
Carcharhinus leucas Near threatened Bull b 25.01715o N 81.01056o W September
- - Bull b 25.01309o N 80.00129o W September
- - Bull b 25.00623o N 80.99723o W March
Sphyrna mokarran Endangered Great Hammerhead a 25.62138o N 80.15656o W July
- - Great Hammerhead b 25.01715o N 81.01056o W September
- - Great Hammerhead a 25.740092o N 79.967258o W May
- - Great Hammerhead b 26.61587o N 79.96725o W February
- - Great Hammerhead b 26.457892o N 80.053938 o W April
Negaprion brevirostris Near threatened Lemon b 25.00644o N 80.99969o W June
- - Lemon b 25.00644o N 80.99969o W March
Ginglymostoma cirraum Data deficient Nurse a 25.61942o N 80.1835 o W September
- - Nurse b 24.88335o N 80.84475 o W April
- - Nurse b 25.00644o N 80.99969 o W March
- - Nurse a 25.62311o N 80.15626o W August
- - Nurse a 25.60062o N 80.15214 o W August
- - Nurse b 25.00858o N 80.00089 o W September
- - Nurse b Florida Bay January
- - Nurse b 25.00983o N 80.99305o W March
Toxins 2016, 8, 238 4 of 14
Toxins 2016, 8,‐ 238 ‐ Nurse a 25.60062oN 80.15214 oW 4 of 14

August
‐ ‐ Nurse a 25.60569oN 80.1534 oW August
‐ ‐ Nurse a 25.62311 N 80.15626 W
o o August
‐ ‐ Table 1. Cont.
Nurse b 25.00858oN 80.00089 oW September
‐ ‐ Nurse b Florida Bay January
International Union for oN 80.99305oW
‐ Name
Scientific ‐ of Nature
Conservation Nurse
Common Name
b 25.00983
Location March
Month
Rhizoprionodon Red List Category
Least Concerned Atlantic Sharpnose b 25.00858oN 81.00089oW April
terraenovae
Rhizoprionodon
Least Concerned Atlantic Sharpnose b 25.00858o N 81.00089o W April
terraenovae
‐ ‐ Atlantic Sharpnose b 25.10566oN 81.04757oW April
- - Atlantic Sharpnose b 25.10566o N 81.04757o W April
‐ - ‐ - Atlantic Sharpnose
Atlantic Sharpnose b
b Florida Bay
Florida Bay AprilApril
Sphyrna zygaena
Sphyrna zygaena Vulnerable
Vulnerable Smooth Hammerhead a a
Smooth Hammerhead o o
26.117727NN
26.117727 o
W W February
80.09734
80.09734 o February
Galeocerdo cuvier
Galeocerdo cuvier Near threatened
Near threatened Tiger cc
Tiger 21.12055o N
−−21.12055 149.22416
oN
o
149.22416E oE January
January
- - Tiger c o
−32.78278 N 152.41171 E o
January
‐ - ‐ - Tiger −−32.78278
24.81665o NN 152.41171E E September
January
c o o
Tiger c 152.47257 o
‐ - ‐ - Tiger cc
Tiger 24.81665o N
−−24.81665 152.47257
oN
o
152.47257E oE March
September
‐ ‐ a Biscayne Bay; b Florida Bay;c c Pacific Ocean.
Tiger −24.81665oN 152.47257 oE March
a Biscayne Bay; b Florida Bay; c Pacific Ocean
We used a rapid and sensitive HPLC-FD method for detection of 6-aminoquinolyl-N-

We used a carbamate
hydroxysuccinimidyl rapid (AQC)
and tagged
sensitive
BMAA HPLC‐FD method
(Figure 1) [25]. for BMAA
We detected detection of
in shark
6‐aminoquinolyl‐N‐hydroxysuccinimidyl
fins of all 10 species surveyed in this study carbamate (AQC) tagged
in concentrations ranging BMAA
from 34(Figure 1) [25]. (wet
to 2011 ng/mg We
detected BMAA in shark fins of all 10 species surveyed in this study in concentrations ranging from
weight) (Table 2). The average concentration for this survey of BMAA in sharks was 366 ± 72 ng/mg
34 to weight)
(wet 2011 ng/mg
(Table(wet
2). weight) (Table 2). method
The HPLC-FD The average concentration
has lower sensitivityfor compared
this survey
to of BMAA in
LC-MS/MS
sharks was 366 ± 72 ng/mg (wet weight) (Table 2). The HPLC‐FD method
methods [37,38]. The unambiguous detection and identification of BMAA in complex biologicalhas lower sensitivity
compared to LC‐MS/MS methods [37,38]. The unambiguous detection and identification of BMAA
samples requires mass spectrometry validation and AQC derivatization to distinguish BMAA from its
in complex
positional biological
isomers DABsamples
and AEGrequires
(Table 3)mass spectrometry
[16,25]. UPLC-MS/MS validation
was usedand toAQC
confirmderivatization
the identity to
of
distinguish BMAA from its positional isomers DAB and AEG (Table 3) [16,25]. UPLC‐MS/MS was
BMAA in a representative sample of shark fins (Figure 2).
used to confirm the identity of BMAA in a representative sample of shark fins (Figure 2).

Figure
Figure 1.1. High
High performance
performance liquid
liquid chromatography
chromatography with with fluorescence
fluorescence detection
detection (HPLC‐FD)
(HPLC-FD)
identification of β‐N‐methylamino‐ L‐alanine (BMAA) in shark fins. (A) Separation of
identification of β-N-methylamino-L-alanine (BMAA) in shark fins. (A) Separation of
6‐aminoquinolyl‐N‐hydroxysuccinimidyl
6-aminoquinolyl-N-hydroxysuccinimidyl carbamate carbamate (AQC)(AQC) derivatized
derivatized aminoamino acid standards
acid standards tyrosine
tyrosine
(Try), (Try),
valine valine
(Val), (Val), (Met),
methionine methionine
and BMAA (Met), standard;
and BMAA standard; (B)
(B) representative representative
chromatogram of
chromatogram of Australian Tiger shark fin (black) and BMAA standard (green).
Australian Tiger shark fin (black) and BMAA standard (green). Chromatogram shows BMAA has a Chromatogram
shows BMAA has a distinct peak with a retention time of 31.8 mins.
distinct peak with a retention time of 31.8 mins.
Toxins 2016, 8, 238 8, 238
Toxins 2016, 5 of 14 6 of 14

Figure 2. Figure 2. Ultra-performance
Ultra‐performance liquid chromatography/mass
liquid chromatography/mass spectrometry/mass spectrometryspectrometry
spectrometry/mass
(UPLC-MS/MS) detection and conformation of BMAA in shark fins. (A) Chromatograms depicting
(UPLC‐MS/MS) detection and conformation of BMAA in shark fins. (A) Chromatograms depicting
detection of ACQ derivatized standards of BMAA, and structural isomers N-(2-aminoethyl) glycine
detection of ACQ derivatized standards of BMAA, and structural isomers N‐(2‐aminoethyl) glycine
(AEG) and 4-diaminobutyric acid (DAB); (B) UPLC-MS/MS chromatograms of BMAA detection in
fins from Australian sharks. The diagnostic selected reaction monitoring (SRM) transitions of the
(AEG) and 4‐diaminobutyric acid (DAB); (B) UPLC‐MS/MS chromatograms of BMAA detection in
parent ion m/z 459 to daughter ions 289, 171 and 119 are common to all three isomers. The BMAA
fins from Australian sharks. The diagnostic selected reaction monitoring (SRM) transitions of the
(blue) peak is selectively identified at 6.55 min by the transition 459 > 258. AEG (green) is selectively
parent ion m/z 459 to daughter ions 289, 171 and 119 are common to all three isomers. The BMAA
identified at 6.66 min by the transition 459 > 214. DAB (grey) is selectively identified at 6.81 min by the
transition 459 > 188.
(blue) peak is selectively identified at 6.55 min by the transition 459 > 258. AEG (green) is selectively
identified at 6.66 min by the transition 459 > 214. DAB (grey) is selectively identified at 6.81 min by
the transition 459 > 188.
BMAA was below the level of quantitation in only 12% of shark fins tested (Table S1). The
highest BMAA concentrations were measured in bonnethead sharks (Figure 3, Table 2; 707 ± 395
ng/mg wet weight and 925 ng/100 cm fin length). This result is in keeping with the elevated levels of
Toxins 2016, 8, 238 6 of 14
Table 2. β-N-methylamino-L-alanine (BMAA) concentrations detected by high performance liquid

chromatography with fluorescence detection (HPLC-FD) in shark fins
Species Range (ng/mg) Detected Mean ± SE (ng/mg) BMAA/Length (ng/100 cm)

Blacknose (n = 6) a ND–1663 573 ± 322 * 473
Blacktip (n = 17) a ND–811 282 ± 72 * 203
Bonnethead (n = 4) a 40–1836 707 ± 395 925
Bull (n = 3) a 43–264 180 ± 69 103
Great Hammerhead (n = 5) a 42–1528 576 ± 272 273
Lemon (n = 2) a 556–628 592 ± 36 322
Nurse (n = 10) a ND–2011 442 ± 315 * 216
Sharpnose (n = 3) a 40–115 68 ± 24 47
Smooth Hammerhead (n = 1) a - 43 21
Tiger (n = 4) b 34–44 39 ± 2 11
ND, Below limit of detection; SE: Standard Error; *: Only detected samples averaged; a : Atlantic Ocean; b :
Pacific Ocean.
Table 3. Comparison of BMAA concentrations detected by HPLC-FD and ultra-performance liquid

chromatography/mass spectrometry/mass spectrometry (UPLC-MS/MS).
Species HPLC-FD * (ng/mg) UPLC-MS/MS * (ng/mg)

Galeocerdo cuvier - -
Tiger a 35.60 ± 1.90 19.20 ± 7.10
Tiger a 31.50 ± 2.60 20.68 ± 3.50
Tiger a 39.60 ± 4.70 33.15 ± 5.60
Tiger a 38.90 ± 5.10 20.17 ± 2.40
* Data presented as Mean ± Standard Error; a : Pacific Ocean; Four replicate biological samples were analyzed
in triplicate to determine method reproducibility and ruggedness.
BMAA was below the level of quantitation in only 12% of shark fins tested (Table S1). The highest
BMAA concentrations were measured in bonnethead sharks (Figure 3, Table 2; 707 ± 395 ng/mg wet
weight and 925 ng/100 cm fin length). This result is in keeping with the elevated levels of BMAA
in benthic organisms [39]. The preferred prey of bonnetheads is found in coastal inshore areas that
feed primarily on blue cabs (Callinectes sapidus) and other crustaceans [40]. In an examination of
cyanobacteria in South Florida and BMAA concentrations in resident fish and invertebrates [41],
we found that blue crab and shrimp had among the highest concentrations of BMAA reported in
animals (6976 µg/g). Macroalgal abundance per square meter in such habitats can be typically
around 20 times higher on the sediment than in the water column [39]. The elevated levels of BMAA
may be due to the high occurrence of benthic cyanobacteria associated with the microalgae and detritus
Toxins 2016, 8, 238
that the blue crab and shrimp feed on [41]. 7 of 14

Figure 3. BMAA concentration per unit length of shark fin. Bar graphs depict the mean and standard
Figure 3. BMAA concentration per unit length of shark fin. Bar graphs depict the mean and standard
error of BMAA concentration per 100 cm in ten shark species from the Atlantic and Pacific oceans.
error of BMAA concentration per 100 cm in ten shark species from the Atlantic and Pacific oceans.
Cold vapor atomic fluorescence spectroscopy (CVAFS) and thermal decomposition methods for
total Hg in sharks gave positive results in all 10 species surveyed. The THg concentrations ranged
from 0.048 to 13.23 ng/mg with a mean concentration of 2.3 ± 0.4 ng/mg (Table 4; Table S2). These
values are higher than those reported safe for human consumption, which range from 0.3 to 1.0 μg/g
wet weight based on different criteria and benchmark dose estimates reported by health
organization or government agencies [42,43]. The highest THg concentrations were found in the bull
Toxins 2016, 8, 238 7 of 14
Cold vapor atomic fluorescence spectroscopy (CVAFS) and thermal decomposition methods
for total Hg in sharks gave positive results in all 10 species surveyed. The THg concentrations
ranged from 0.048 to 13.23 ng/mg with a mean concentration of 2.3 ± 0.4 ng/mg (Table 4; Table S2).
These values are higher than those reported safe for human consumption, which range from 0.3
to 1.0 µg/g wet weight based on different criteria and benchmark dose estimates reported by health
organization or government agencies [42,43]. The highest THg concentrations were found in the bull
sharks, averaging 7.26 ± 3.04 ng/mg. Bull sharks are large coastal apex predators with high Hg levels
reported previously in agreement with our results [21,44,45]. Shark muscle samples contained nine
times the amount of THg as compared to fins (Student’s t-test p < 0.0001; n = 26/20) (Table S1). THg in
muscles ranged from 0.27 to 13.23 ng/mg with a mean concentration of 3.8 ± 0.6 ng/mg. In addition,
MeHg was measured in a subset of sharks with concentrations ranging from 0.05 to 1.95 ng/mg and a
mean concentration of 0.42 ± 0.11 ng/mg (Table 4). THg and MeHg concentration in shark samples
tested were highly correlated (Spearman correlations r = 0.94, p < 0.0001; n = 18) in our shark cohort as
expected [33–35].
Table 4. Mercury concentrations detected in shark fin and muscle.
Species Range Hg (ng/mg) THg (ng/mg) * MeHg (ng/mg) * BMAA:THg

Blacknose a 0.05–5.65 1.93 ± 2.27 (n = 3) 0.71 ± 0.02 (n = 2) 429:1
Blacktip a 0.22–7.73 3.70 ± 0.69 (n = 16) 1.40 ± 0.75 (n = 7) 368:1
Bonnethead a 0.41–1.77 0.96 ± 0.32 (n = 4) 0.56 ± 0.44 (n = 4) 668:1
Bull a 3.24–13.23 7.26 ± 3.04 (n = 3) 2.32 (n = 1) 27:1
Great Hammerhead a - 3.29 (n = 1) N/A 465:1
Lemon a 0.27–1.34 0.81 ± 0.54 (n = 2) 0.26 ± 0.08 (n = 2) 1390:1
Nurse a 0.06–0.48 0.24 ± 0.04 (n = 10) N/A 1509:1
Sharpnose a 0.44–2.41 1.42 ± 0.98 (n = 2) 0.25 (n = 1) 70:1
Smooth Hammerhead a - 2.85 (n = 1) N/A 15:1
Tiger b 0.12–1.61 0.74 ± 0.36 (n = 4) N/A 23:1
* Data presented as mean ± standard error; a : Atlantic Ocean; b : Pacific Ocean. N/A, Samples not available
for measurement.
Previous studies of MeHg in shark fins and soup reported variable levels, with higher
concentrations measured in higher trophic levels species consistent with biomagnification [35].
However, a recent report suggests that THg levels are low in shark fin soup posing only a minor risk
for human exposure [35]. We reported that THg and BMAA are detected in shark cartilage dietary
supplements [16]. Despite the low levels of THg in shark fin soup, the co-occurrence of BMAA and
elemental and Hg in shark fin and muscle should be considered a potential human health concern due
to their possible synergistic toxicity to neural tissues [31].
In all shark species surveyed, BMAA levels averaged 15 to 1500 times higher than the
concentration of Hg (Table 4). We correlated BMAA and Hg levels by comparing the linear mass
density of each contaminant. The concentrations of Hg were positively correlated with BMAA
among individuals when both values were normalized for length (concentration/100 cm; Spearman
correlations r = 0.37, p = 0.02; n = 39). The positive correlation demonstrates that sharks with higher
BMAA concentrations show increased Hg levels. Shark fins are often dried or cooked prior to
human consumption. While these preparation methods are known to remove other marine toxins,
neither BMAA nor Hg would likely be significantly affected because both are associated with stable
incorporation into proteins [46]. BMAA is misincorporated into neural proteins [47–49] and Hg-binding
proteins are a likely source of harmful accumulation of Hg in the marine food web [50].
The sharks surveyed show both inter- and intra-specific variation in BMAA and Hg concentrations.
There are various biological and physical properties of the environment that may affect exposures of
sharks to spatial and temporal differences in the accumulation of Hg and BMAA through the food web.
In the case of Hg, exposure can be through both point source pollution and atmospheric emissions
Toxins 2016, 8, 238 8 of 14
from fossil fuel combustion [51]. The bioaccumulation will depend in part on the presence of anaerobic
bacteria that can convert inorganic Hg to the organic form for trophic transfer up the food web to
sharks [52] Similarly, variation in the bioaccumulation and exposure of sharks to BMAA will depend
on environmental levels of cyanobacteria that increase with nutrient pollution derived from land-based
sources [25,41]. While evaluating patterns of spatial, temporal and even individual variation in BMAA
and Hg toxicity are important for mitigation efforts to reduce exposure, our data suggest that the risk
of BMAA exposure may be greatest between spring and summer seasons (Table S2). Thus, further
studies are warranted based on the limited reports of BMAA in the marine food web across diverse
geographical locales.
The prevalence of dementia and Alzheimer’s disease is significantly higher in certain Asian
countries [53]. In China alone, the number of people with dementia has increased significantly
from 3.7 million in 1990 to 9.2 million in 2010 [54]. Moreover, a report in Lancet on global disease
burden found that the number of deaths in China due to Alzheimer’s disease and other dementias
doubled between 1990 and 2010, while mortality rates, especially among women, fell steeply during
the same period [54]. With the continuing growth of China’s aging population, these findings suggest
that the nation is heading for a bigger dementia burden than anticipated [55]. The present study
suggests that ingestion of shark fin and shark dietary supplements is a route for human exposure to
the environmental toxins BMAA and Hg. Although there are no estimates to help benchmark exposure
risk of BMAA to humans from dietary exposures, in China, male infertility has been linked to Hg
exposure through consumption of seafood, including shark fins [56]. Because sharks were sampled
in South Florida waters, there is a concern of BMAA exposures also to USA residents. In Florida,
there is an estimated 0.5 million people over the age of 65 with Alzheimer’s disease. These numbers
are anticipated to increase to 0.7 million by the 2020 [57].
Given the decline in many shark populations from overfishing, more research is needed to fully
understand the potential toxicity of BMAA and Hg to the health and fitness of shark species [1].
Systemic exposure to BMAA and Hg is likely to worsen the problem and limit recovery efforts if not
considered in conservation management efforts. Since sharks often occupy high trophic levels in the
marine food web, they are vulnerable to bioaccumulation and biomagnification of neurotoxins and
other toxic compounds. Given that humans and sharks are both top predators, the results reported
here support the view that sharks serve as bioindicators of ecosystem health from human stressors
and marine contaminants [58,59].
3. Materials and Methods
3.1. Sample Collection

Fin clips and muscle biopsies were collected from shark species (n = 10) sampled from areas
with or without documented cyanobacterial blooms in the Atlantic and Pacific Oceans as described
previously [25,26]. Small clips were sampled from archived frozen dorsal fins for analysis of BMAA
(n = 55), total Hg (n = 46). Shark specimens where available were assayed for Hg concentrations in
muscle (n = 26) and fins (n = 20). Tissue specimens from blacknose (Carcharhinus acronotus), blacktip
(Carcharhinus limbatus), bonnethead (Sphyrna tiburo), bull (Carcharhinus leucas), great hammerhead
(Sphyrna mokarran), lemon (Negaprion brevirostris), nurse (Ginglymostoma cirratum), Atlantic sharpnose
(Rhizoprionodon terraenovae), smooth hammerhead (Sphyrna zygaena) and tiger (Galeocredo cuvier) sharks
were included in this survey (Table 1).
3.2. HPLC Sample Preparation

BMAA in shark fin clips was detected and quantified using high performance liquid
chromatography (HPLC) methods as reported previously [25]. Briefly, fin clips (50 mg) were
hydrolyzed at 110 ◦ C for 18 h in 6 N HCl (1:8 w/v) followed by filtration using centrifugation
at 15,800× g for 3 min. Sample extracts were concentrated and dried in a speed-vac (Thermo-Savant
Toxins 2016, 8, 238 9 of 14
SC250DDA Speed Vac Plus with a Savant refrigerator trap RVT 4104, ThermoFischer; Waltham, MA,
USA). Extracts were re-suspended in 0.1 M trichloroacetic acid and washed with chloroform to remove
any residual lipids. The dried extract was resuspended to 1000 µL in 20 mM HCl. A 100 µL aliquot of
the sample extract was derivatized with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC)
using the AccQ-Fluor reagent (Waters Crop; Milford, MA, USA). The derivatized samples (20 µL
resuspended HCl extract, 60 µL of borate buffer (AccQ-Fluor Reagents A and B; Waters), and 20 µL
AccQ-Tag) were run in parallel with buffer and AQC blanks and BMAA, AEG, DAB and reference
amino acid standards. The sample matrix was spiked with known amounts of BMAA to determine
recovery of the extraction procedure and confirm peak identity. Each sample was prepared in triplicate
for quantitative studies and orthogonal detection method comparisons. For the orthogonal method
comparisons, the shark samples were prepared by different analysts in different labs as a measure of
method ruggedness.
3.3. Fluorescence HPLC Methods for Analysis of BMAA

BMAA was separated from amino-acids by reverse-phase high pressure chromatography
(Waters Nova-Pak C18 column, 3.9 mm × 300 mm; Waters Crop; Milford, MA, USA) eluted in
a gradient of 140 mM sodium acetate, 5.6 mM triethylamine, pH 5.2 (mobile phase A), and 52%
(v/v) acetonitrile in water (mobile phase B) at 37 ◦ C using a flow rate of 1.0 mL/min, and 10 µL
sample injection volume. The samples were eluted using a 60 min gradient: 0.0 min = 100%
A; 2 min = 90% A curve 11; 5 min = 86% A curve 11; 10 min = 86% A curve 6; 18 min = 73% A curve 6;
30 min = 57% A curve 10; 35 min = 40% A curve 6; 37.5 min = 100% B curve 6; 47.5 min = 100% B curve 6;
50 min = 100% A curve 6; 60 min = 100% A curve 6. Detection of the AQC fluorescent tag was achieved
using a Waters 2475 Multi λ-Fluorescence Detector (Milford, MA, USA) with excitation at 250 nm
and emission at 395 nm. Experimental samples were compared with standard spiked shark fin
matrix negative for endogenous BMAA and commercial BMAA reference standard (Sigma B-107; >95%
purity, St. Louis, MO, USA). The limits of detection (LOD) and limits of quantification (LOQ)
were 2.7 and 7.0 ng, respectively. The percentage of recovery of BMAA was 88%.
3.4. UPLC/MS/MS of BMAA

BMAA and the isomers N-(2-aminoethyl) glycine (AEG) and 2,4-diaminobutyric acid (DAB)
were separated, detected and quantified by ultra-performance liquid chromatography/mass
spectrometry/mass spectrometry (UPLC/MS/MS) using a fully validated method as previously
described [60]. Briefly, 50 mg samples of frozen shark fin clips were accurately weighed and suspended
in 1.0 mL of 6 N HCl sealed with N2 gas blown into the tubes for 30 s to displace oxygen. Samples
were hydrolyzed for 18 h at 110 ◦ C. A subsample of 400 µL was filtered (0.22 µm PVDF Ultrafree
MC centrifuge filters; EMD Millipore; Billerica, MA, USA) and a 100 µL aliquot was dried overnight
(Labconco Centrivap; Kansas City, MO, USA). The sample was reconstituted in 1.0 mL 20 mM HCl
and a 20 µL aliquot was derivatized with 20 µL 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate
(AQC) in 60 µL borate buffer (AccQ-Fluor Reagents A and B; Waters, Milford, MA, USA). BMAA, AEG
and DAB were separated by reverse phase C18 chromatography (BEH column 150 × 2.1 mm 1.7 µm;
Waters) and eluted with a gradient of 20 mM ammonium formate with 0.2% formic acid (A) and 0.1%
formic acid in acetonitrile; (B). Gradient was delivered by a Waters Acquity I-Class UPLC (Milford,
MA, USA) (0 min, 95% A; 1.0 min, 95% A; 7 min, 85% A; 7.5 min, 78% A; 8 min, 15% A; 8.5 min, 15% A;
8.6 min, 95% A; 10 min, 95% A) with a flow rate of 0.7 mL/min at 52 ◦ C. Compounds were clearly
separated with BMAA elution at 6.56 min (%RSD = 0.23), AEG at 6.67 min (%RSD = 0.22) and DAB at
6.82 min (%RSD = 0.26) (see Figure 2). Triplicate measures were performed on each shark fin sample
(Table 3).
Ions were detected on a triple quadrupole tandem mass spectrometer (Waters Xevo TQS,
Milford, MA, USA) with the following parameters: cone voltage was 16 V. Capillary voltage was set
to 2500 V with a source offset of 50 V. Desolvation temperature was 550 ◦ C, with a corresponding
Toxins 2016, 8, 238 10 of 14
gas flow of 800 L/h. and a cone gas flow of 150 L/h. Collision-induced-dissociation was performed
with 99.999% pure argon pressurized to 7.0 bar with a dwell time of 0.05 s. The characteristic transitions
were detected as: BMAA 459 > 258 at collision voltage 18 V, DAB 459 > 188 at collision voltage 20 V,
AEG 459 > 214 at collision voltage 20 V.
3.5. Determination of Hg in Fins (CVAFS Method)

Total Hg (THg) includes inorganic and organic forms of Hg. THg and MeHg analyses were
performed on shark fin clips following the Standard Operating Procedure modified from the U.S.
Environmental Protection Agency (EPA) Test Method 1631 [61]. THg in a sample was isolated and
oxidized to mercuric ion using acid digestion, and then reduced to elemental Hg by stannous chloride,
purged from the liquid by a carrier gas (Argon). MeHg was extracted from the sample matrix with
sodium hydroxide in methanol on a hot block. The Hg species on the traps were desorbed, pyrolyzed
and detected by Cold Vapor Atomic Fluorescence Spectrometry (CVAFS) (Millennium Merlin 10.035, PS
Analytical, Deerfield Beach, FL, USA). Briefly, the samples (about 0.2 g) were weighed into 10-mL glass
ampoules to which 1 mL of deionized water and 2 mL of concentrated HNO3 were added. The ampules
were then sealed and the samples were autoclaved for 1 h at 105 ◦ C for sample digestion. The samples
were diluted with 1% HCl and introduced into CVAFS, reduced with 2% (v/v) SnCl2 (in 2.5% HCl).
Daily analytical runs began with an initial calibration containing 5 non-zero points and a system blank.
The mean calibration factor (CFm), calculated from the calibration factor (CFx) for Hg in each of the
five standards using the system blank-subtracted peak height, was used for the calculation of sample
concentration. Each analytical batch included at least one method blank, a Continuing Calibration
Check Samples (CCS), and a Quality Control Sample (QCS). All method blanks during analysis were
below the method detection limits (MDLs). The readings of CCS were always within acceptable range
(85%–115% for THg of initial calibration). Certified reference material (CRM), DORM-2, was used as
a QCS sample throughout the analysis and the recoveries for the QCS samples (84%–128% for THg)
were always within acceptable range specified in standard operating procedure (SOPs) (70%–130% for
THg). The method limit of detection for the instrument was 0.002 mg/kg.
3.6. Determination of Total Mercury in Muscles (Thermal Decomposition Method)

Muscle samples were placed into nickel sample boats, weighed, and analyzed for THg using
thermal decomposition technique with an automated direct Hg analyzer (DMA 80, Milestone
Incorporated, Shelton, CT, USA) using the US EPA Method 7473 [62]. Assays were run with one
sample each of two standard reference materials (DORM-3 and DOLT-4), two method blanks, and one
sample blank.
Supplementary Materials: The following is available online at www.mdpi.com/2072-6651/8/8/238/s1. Table S1

BMAA and Total Hg concentrations determined by HPLC in sharks. Table S2: BMAA and Mercury concentration
detected by seasons in sharks.
Acknowledgments: The Herbert W. Hoover Foundation (Akron, OH, USA) provided the funding for this
study. Guangling Liu PhD and Yong Cai PhD, Department of Chemistry and Bioinorganic and Environmental
Analytics, Florida International University, Miami, FL provided contract services for Hg analyses. We acknowledge
Bonnie J. Holmes PhD, University of Queensland Brisbane, for providing Australia tiger shark fins for this survey.
Author Contributions: N.H. analyzed data and wrote the manuscript; D.A.D. analyzed data and wrote the
manuscript; K.M. performed experiments, analyzed data, and assisted with graphics; M.S.S. conducted the
literature review and wrote the manuscript; S.J.M. performed experiments, analyzed data and wrote the
manuscript; W.B.G. performed experiments and analyzed data; T.D. performed experiments and analyzed
data; D.C.E. performed experiments and analyzed data; and D.C.M. conceived the project, designed experiments,
analyzed data; and wrote the manuscript.
Toxins 2016, 8, 238 11 of 14
Abbreviations
ALS/PDC Amyotrophic lateral sclerosis/parkinsonism dementia complex

AEG N-(2-aminoethyl) glycine
AQC 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate
BMAA β-N-methylamino-L-alanine
CVAFS Cold vapor atomic fluorescence spectrometry
CCS Calibration Check Samples
CFm Calibration factor
CFx Calibration factor
DAB 2,4-diaminobutyric acid
FDOH Florida Department of Health
Hg Mercury
HPLC-FD High performance liquid chromatography with fluorescence detection
LOD Limits of detection
LOQ Limits of quantification
MDLs Method detection limits
MeHg Methyl mercury
QCS Quality Control Sample
THg Total mercury
UPLC-MS/MS Ultra-performance liquid chromatography/mass spectrometry/mass spectrometry
References
1. Baum, J.K.; Myers, R.A.; Kehler, D.G.; Worm, B.; Harley, S.J.; Doherty, P.A. Collapse and conservation of
shark populations in the Northwest Atlantic. Science 2003, 299, 389–392. [CrossRef] [PubMed]
2. Dulvy, N.K.; Baum, J.; Clarke, S.; Compagno, L.J.; Cortes, E.; Domingo, A.; Fordham, S.; Fowler, S.;
Francis, M.P.; Gibson, C.; et al. You can swim but you can’t hide: The global status and conservation
of oceanic pelagic sharks and rays. Aquat. Conserv. 2008, 18, 459–482. [CrossRef]
3. Camhi, M.D.; Valenti, S.V.; Fordham, S.V.; Fowler, S.L.; Gibson, C. The Conservation Status of Pelagic Sharks
and Rays: Report of the IUCN Shark Specialist Group Pelagic Shark Red List Workshop; IUCN Species Survival
Commission Shark Specialist Group: Newbury, UK, 2009.
4. Molina, J.M.; Cooke, S.J. Trends in shark bycatch research: Current status and research needs. Rev. Fish
Biol. Fish. 2012, 22, 719–737. [CrossRef]
5. Gallagher, A.J.; Kyne, P.M.; Hammerschlag, N. Ecological risk assessment and its application to elasmobranch
conservation and management. J. Fish Biol. 2012, 80, 1727–1748. [CrossRef] [PubMed]
6. Oliver, S.; Braccini, M.; Newman, S.J.; Harvey, E.S. Global patterns in the bycatch of sharks and rays.
Mar. Policy 2015, 54, 86–97. [CrossRef]
7. Worm, B.; Davis, B.; Kettemer, L.; Ward-Paige, C.A.; Chapman, D.; Heithaus, M.R.; Kessel, S.T.; Gruber, S.H.
Global catches, exploitation rates, and rebuilding options for sharks. Mar. Pol. 2013, 40, 194–204. [CrossRef]
8. The state of world fisheries and aquaculture. Rome: Food and agriculture organization of the United States.
Available online: http://www.fao.org/3/a-i3720e.pdf. (accessed on 15 January 2016).
9. Shiffman, D.; Hammerschlag, N. Shark conservation and management policy: A review and primer for
non-specialists. Anim. Consv. 2016. [CrossRef]
10. Cunningham-Day, R. Sharks in Danger: Global Shark Conservation Status with Reference to Management Plans
and Legislation; Universal Publishers: Boca Raton, FL, USA, 2001.
11. Spiegel, J. Even Jaws Deserves to Keep His Fins: Outlawing Shark Finning Throughout Global Waters.
Boston Coll. Law Rev. 2000, 24, 409–437.
12. Clarke, S.C.; Milner-Gulland, E.J.; Bjorndal, T. Social, economic, and regulatory drivers of the shark fin trade.
Mar. Resour. Econ. 2007, 22, 305–327. [CrossRef]
13. Dulvy, N.K.; Fowler, S.L.; Musick, J.A.; Cavanagh, R.D.; Kyne, P.M.; Harrison, L.R.; Carlson, J.K.;
Davidson, L.N.; Fordham, S.V.; Francis, M.P.; et al. Extinction risk and conservation of the world’s sharks
and rays. Elife 2014, 3, e00590. [CrossRef] [PubMed]
14. Templer, R. Food for Thought: Chicken Soup for the Ostentatious Soul—Tonic or Tragedy? East, West
Divided Over Shark’s Fin. Asian Wall St J. 1999. The Wall Street Journel. Availble online: http://www.wsj.
com/articles/SB934473109149930399 (accessed on 31 January 2011).
Toxins 2016, 8, 238 12 of 14
15. Mahr, K. Shark-Fin Soup and the Conservation Challenge. Available online: http://content.time.com/time/
magazine/article/0,9171,2021071,00.html (accessed on 25 June 2016).
16. Mondo, K.; Broc Glover, W.; Murch, S.J.; Liu, G.; Cai, Y.; Davis, D.A.; Mash, D.C. Environmental
neurotoxins beta-N-methylamino-L-alanine (BMAA) and mercury in shark cartilage dietary supplements.
Food Chem. Toxicol. 2014, 70, 26–32. [CrossRef] [PubMed]
17. Merly, L.; Smith, S.L. Pro-inflammatory properties of shark cartilage supplements. Immunopharmacol.
Immunotoxicol. 2015, 37, 140–147. [CrossRef] [PubMed]
18. Delshad, S.T.; Mousavi, S.A.; Islami, H.R.; Pazira, A. Mercury concentration of the whitecheek shark,
Carcharhinus dussumieri (Elasmobranchii, Chondrichthyes), and its relation with length and sex. Pan-Am. J.
Aquat. Sci. 2012, 7, 135–142.
19. Escobar-Sanchez, O.; Galvan-Magana, F.; Rosiles-Martinez, R. Mercury and selenium bioaccumulation in
the smooth hammerhead shark, Sphyrna zygaena Linnaeus, from the Mexican Pacific Ocean. Bull. Environ.
Contam Toxicol. 2010, 84, 488–491. [CrossRef] [PubMed]
20. Pethybridge, H.; Cossa, D.; Butler, E.C. Mercury in 16 demersal sharks from southeast Australia: Biotic and
abiotic sources of variation and consumer health implications. Mar. Environ. Res. 2010, 69, 18–26. [CrossRef]
[PubMed]
21. Rumbold, D.; Wasno, R.; Hammerschlag, N.; Volety, A. Mercury accumulation in sharks from the coastal
waters of southwest Florida. Arch. Environ. Contam. Toxicol. 2014, 67, 402–412. [CrossRef] [PubMed]
22. Adams, D.H., Jr. Mercury levels in four species of sharks from the Atlantic coast of Florida. Fish. Bull. 1999,
97, 372–329.
23. Health FDo. Seafood Consumption, Get Fresh with Florida Fish. Available online: http://www.doh.state.fl.
us/floridafishadvice/ (accessed on 25 June 2016).
24. Scientific Opinion on the risk for public health related to the presence of mercury and methylmercury in
food. Avaible online: http://onlinelibrary.wiley.com/doi/10.2903/j.efsa.2012.2985/abstract (accessed on 28
June 2016).
25. Mondo, K.; Hammerschlag, N.; Basile, M.; Pablo, J.; Banack, S.A.; Mash, D.C. Cyanobacterial neurotoxin
beta-N-methylamino-L-alanine (BMAA) in shark fins. Mar. Drugs 2012, 10, 509–520. [CrossRef] [PubMed]
26. Murch, S.J.; Cox, P.A.; Banack, S.A.; Steele, J.C.; Sacks, O.W. Occurrence of beta-methylamino-l-alanine
(BMAA) in ALS/PDC patients from Guam. Acta Neurol. Scand. 2004, 110, 267–269. [CrossRef] [PubMed]
27. Pablo, J.; Banack, S.A.; Cox, P.A.; Johnson, T.E.; Papapetropoulos, S.; Bradley, W.G.; Buck, A.; Mash, D.C.
Cyanobacterial neurotoxin BMAA in ALS and Alzheimer’s disease. Acta Neurol. Scand. 2009, 120, 216–225.
[CrossRef] [PubMed]
28. Field, N.C.; Metcalf, J.S.; Caller, T.A.; Banack, S.A.; Cox, P.A.; Stommel, E.W. Linking
beta-methylamino-L-alanine exposure to sporadic amyotrophic lateral sclerosis in Annapolis, MD. Toxicon
2013, 70, 179–183. [CrossRef] [PubMed]
29. Shen, W.B.; McDowell, K.A.; Siebert, A.A.; Clark, S.M.; Dugger, N.V.; Valentino, K.M.; Jinnah, H.A.;
Sztalryd, C.; Fishman, P.S.; Shaw, C.A.; et al. Environmental neurotoxin-induced progressive model of
parkinsonism in rats. Ann. Neurol. 2010, 68, 70–80. [CrossRef] [PubMed]
30. Spencer, P.S.; Nunn, P.B.; Hugon, J.; Ludolph, A.C.; Ross, S.M.; Roy, D.N.; Robertson, R.C. Guam amyotrophic
lateral sclerosis-parkinsonism-dementia linked to a plant excitant neurotoxin. Science 1987, 237, 517–522.
[CrossRef] [PubMed]
31. Rush, T.; Liu, X.; Lobner, D. Synergistic toxicity of the environmental neurotoxins methylmercury and
beta-N-methylamino-L-alanine. Neuroreport 2012, 23, 216–219. [CrossRef] [PubMed]
32. Cox, P.A.; Davis, D.A.; Mash, D.C.; Metcalf, J.S.; Banack, S.A. Dietary exposure to an environmental toxin
triggers neurofibrillary tangles and amyloid deposits in the brain. Proc. R. Soc. B Biol. Sci. 2016, 283.
[CrossRef] [PubMed]
33. Storelli, M.; Ceci, E.; Storelli, A.; Marcotrigiano, G. Polychlorinated biphenyl, heavy metal and methylmercury
residues in hammerhead sharks: Contaminant status and assessment. Mar. Pollut. Bull. 2003, 46, 1035–1039.
[CrossRef]
34. Nam, D.; Adams, D.; Reyier, E.; Basu, N. Mercury and selenium levels in lemon sharks (Negaprion brevirostris)
in relation to a harmful red tide event. Environ. Monit. Assess. 2010, 176, 549–559. [CrossRef] [PubMed]
Toxins 2016, 8, 238 13 of 14
35. Nalluri, D.; Baumann, Z.; Abercrombie, D.L.; Chapman, D.D.; Hammerschmidt, C.R.; Fisher, N.S.
Methylmercury in dried shark fins and shark fin soup from American restaurants. Sci. Total Environ.
2014, 496, 644–648. [CrossRef] [PubMed]
36. Nature IUfCo. THE IUCN Red List of Threatened Species. Available online: http://www.iucnredlist.org
(accessed on 28 February 2016).
37. Al-sammak, M.A.; Hoagland, K.D.; Snow, D.D.; Cassada, D. Toxicon Methods for simultaneous detection
of the cyanotoxins BMAA, DABA, and anatoxin-a in environmental samples. Toxicon 2013, 76, 316–325.
[CrossRef] [PubMed]
38. Faassen, E.J.; Gillissen, F.; Lürling, M. A comparative study on three analytical methods for the determination
of the neurotoxin BMAA in cyanobacteria. PLoS ONE 2012, 7, e36667. [CrossRef] [PubMed]
39. Brand, L.; Suzuki, M. Distribution of Benthic Chlorophyll in Florida Bay Sediments. In Proceedings of the
1999 Florida Bay and Adjacent Marine Systems Science Conference, Key Largo, FL, USA, 1–5 November
1999; p. 129.
40. Bethea, D.M.; Hale, L.; Carlson, J.K.; Cortés, E.; Manire, C.A.; Gelsleichter, J. Geographic and ontogenetic
variation in the diet and daily ration of the bonnethead shark, Sphyrna tiburo, from the eastern Gulf of Mexico.
Mar. Biol. 2007, 152, 1009–1020. [CrossRef]
41. Brand, L.E.; Pablo, J.; Compton, A.; Hammerschlag, N.; Mash, D.C. Cyanobacterial Blooms and the
Occurrence of the neurotoxin beta-N-methylamino-L-alanine (BMAA) in South Florida Aquatic Food Webs.
Harmful Algae 2010, 9, 620–635. [CrossRef] [PubMed]
42. Adminstration USFaD. 2016. Available online: http://www.fda.gov/Food/GuidanceRegulation/
GuidanceDocumentsRegulatoryInformation/ChemicalContaminantsMetalsNaturalToxinsPesticides/
ucm077969.htm-merc (accessed on 15 June 2016).
43. Agency USEP. Available online: http://www.epa.gov/mercury (accessed on 20 April 2015).
44. Hueter, R.; Fong, W.; Henderson, G.; French, M.; Manire, C. Methylmercury concentration in shark muscle
by species, size and distribution of sharks in Florida coastal waters. Water Air Soil Pollut. 1995, 80, 893–899.
[CrossRef]
45. Adams, D.H., Jr.; Henderson, G.E. Mercury Levels in Marine and Estuarine Fishes of Florida 1989–2001; Florida
Marine Research Institute: Port Charlotte, FL, USA, 2003.
46. Holtcamp, W. The emerging science of BMAA: Do cyanobacteria contribute to neurodegenerative disease?
Environ. Health Perspect. 2012, 120, 110–116. [CrossRef] [PubMed]
47. Xie, X.; Basile, M.; Mash, D.C. Cerebral uptake and protein incorporation of cyanobacterial toxin
beta-N-methylamino-L-alanine. Neuroreport 2013, 24, 779–784. [CrossRef] [PubMed]
48. Dunlop, R.A.; Cox, P.A.; Banack, S.A.; Rodgers, K.J. The non-protein amino acid BMAA is misincorporated
into human proteins in place of L-serine causing protein misfolding and aggregation. PLoS ONE 2013, 8,
e75376. [CrossRef] [PubMed]
49. Glover, W.B.; Mash, D.C.; Murch, S.J. The natural non-protein amino acid N-beta-methylamino-L-alanine
(BMAA) is incorporated into protein during synthesis. Amino Acids 2014, 46, 2553–2559. [CrossRef] [PubMed]
50. Weiyue, F.; Meng, W.; Ming, G.; Yuan, H.; Junwen, S.; Bing, W.; Motao, Z.; Hong, O.; Yuliang, Z.; Zhifang, C.
Mercury speciation and mercury-binding protein study by HPLC-ICP-MS on the estimation of mercury
toxicity between maternal and infant rats. J. Anal. At. Spectrom. 2011, 26, 156–164. [CrossRef]
51. Pacyna, E.G.; Pacyna, J.M.; Sundseth, K.; Munthe, J.; Kindbom, K.; Wilson, S.; Maxson, S.P. Global emission
of mercury to the atmosphere from anthropogenic sources in 2005 and projections to 2020. Atmos. Environ.
2010, 44, 2487–2499. [CrossRef]
52. Compeau, G.C.; Bartha, R. Principal Methylators. Microbiology 1985, 50, 498–502.
53. Kalaria, R.N.; Maestre, G.E.; Arizaga, R.; Friedland, R.P.; Galasko, D.; Hall, K.; Luchsinger, J.A.; Ogunniyi, A.;
Perry, E.K.; Potocnik, F.; et al. Alzheimer’s disease and vascular dementia in developing countries:
Prevalence, management, and risk factors. Lancet Neurol. 2008, 7, 812–826. [CrossRef]
54. Chan, K.Y.; Wang, W.; Wu, J.J.; Liu, L.; Theodoratou, E.; Car, J.; Middleton, L.; Russ, T.C.; Deary, I.J.;
Campbell, H.; et al. Epidemiology of Alzheimer’s disease and other forms of dementia in China, 1990–2010:
A systematic review and analysis. Lancet 2013, 381, 2016–2023. [CrossRef]
55. Loef, M.; Walach, H. Midlife obesity and dementia: Meta-analysis and adjusted forecast of dementia
prevalence in the United States and China. Obesity 2013, 21, 51–55. [CrossRef] [PubMed]
Toxins 2016, 8, 238 14 of 14
56. Dickman, M.D.; Leung, C.K.; Leong, M.K. Hong Kong male subfertility links to mercury in human hair and
fish. Sci. Total Environ. 1998, 214, 165–174. [CrossRef]
57. 2016 Alzhemer's Disease Facts And Figures. Availble online: https://www.alz.org/documents_custom/
2016-facts-and-figures.pdf (accessed on 25 July 2016).
58. Schindler, D.W. Detecting ecosystem responses to anthropogenic stress. J. Fish. Aquat. Sci. 1987, 44, S6–S25.
[CrossRef]
59. Torres, P.; da Cunha, R.T.; Maia, R.; Dos Santos Rodrigues, A. Trophic ecology and bioindicator potential of
the North Atlantic tope shark. Sci. Total Environ. 2014, 481, 574–581. [CrossRef] [PubMed]
60. Glover, W.B.; Baker, T.C.; Murch, S.J.; Brown, P.N. Determination of beta-N-methylamino-L-alanine,
N-(2-aminoethyl)glycine, and 2,4-diaminobutyric acid in Food Products Containing Cyanobacteria by
Ultra-Performance Liquid Chromatography and Tandem Mass Spectrometry: Single-Laboratory Validation.
J. AOAC Int. 2015, 98, 1559–1565. [CrossRef] [PubMed]
61. Agency USEP. Method 1631, Revision E: Mercury in Water by Oxidation, Purge and Trap, and Cold Vapor
Atomic Fluorescence Spectrometry. Available online: http://www.epa.gov/sites/production/files/2015-
08/documents/method_1631e_2002.pdf (accessed on 17 September 2002).
62. EPA Method 7473 (SW-846): Mercury in Solids and Solutions by Thermal Decomposition, Amalgamation,
and Atomic Absorption Spectrophotometry. Available online: https://www.epa.gov/homeland-security-
research/epa-method-7473-sw-846-mercury-solids-and-solutions-thermal-decomposition (accessed on
15 September 2015).
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
RESEARCH ARTICLE
Cyanobacterial neurotoxin BMAA and brain

pathology in stranded dolphins
David A. Davis ID1*, Kiyo Mondo1, Erica Stern1☯, Ama K. Annor1☯, Susan J. Murch2,
Thomas M. Coyne3, Larry E. Brand4, Misty E. Niemeyer5, Sarah Sharp5, Walter G. Bradley1,
Paul Alan Cox ID6, Deborah C. Mash1,7*
1 Department of Neurology, Miller School of Medicine, University of Miami, Miami, Florida, United States of
America, 2 Department of Chemistry, University of British Columbia, Kelowna, British Columbia, Canada,
3 Office of the District 21 Medical Examiner, Fort Myers, Florida, United States of America, 4 Divisions of
Marine Biology and Fisheries and NSF/NIEHS Oceans and Human Health Center, Rosenstiel School of
a1111111111 Marine and Atmospheric Science, University of Miami, Miami, Florida, United States of America, 5 Marine
a1111111111 Mammal Rescue and Research, International Fund for Animal Welfare (IFAW), Yarmouth Port,
a1111111111 Massachusetts, United States of America, 6 Brain Chemistry Labs, Institute for Ethnomedicine, Jackson
a1111111111 Hole, Wyoming, United States of America, 7 Department of Molecular and Cellular Pharmacology, Miller
School of Medicine, University of Miami, Miami, Florida, United States of America
a1111111111
☯ These authors contributed equally to this work.
* dmash@med.miami.edu (DM); d.davis12@med.miami.edu (DD)
OPEN ACCESS
Abstract
Citation: Davis DA, Mondo K, Stern E, Annor AK,
Murch SJ, Coyne TM, et al. (2019) Cyanobacterial Dolphin stranding events occur frequently in Florida and Massachusetts. Dolphins are an
neurotoxin BMAA and brain pathology in stranded
excellent sentinel species for toxin exposures in the marine environment. In this report we
dolphins. PLoS ONE 14(3): e0213346. https://doi.
org/10.1371/journal.pone.0213346 examine whether cyanobacterial neurotoxin, β-methylamino-L-alanine (BMAA), is present
in stranded dolphins. BMAA has been shown to bioaccumulate in the marine food web,
Editor: Susana Caballero, Universidad de los
Andes, COLOMBIA including in the muscles and fins of sharks. Dietary exposure to BMAA is associated with the
occurrence of neurofibrillary tangles and β-amyloid plaques in nonhuman primates. The
Received: June 5, 2018
findings of protein-bound BMAA in brain tissues from patients with Alzheimer’s disease has
Accepted: February 20, 2019
advanced the hypothesis that BMAA may be linked to dementia. Since dolphins are apex
Published: March 20, 2019 predators and consume prey containing high amounts of BMAA, we examined necropsy
Copyright: © 2019 Davis et al. This is an open specimens to determine if dietary and environmental exposures may result in the accumula-
access article distributed under the terms of the tion of BMAA in the brains of dolphins. To test this hypothesis, we measured BMAA in a
Creative Commons Attribution License, which
series of brains collected from dolphins stranded in Florida and Massachusetts using two
permits unrestricted use, distribution, and
reproduction in any medium, provided the original orthogonal analytical methods: 1) high performance liquid chromatography, and 2) ultra-per-
author and source are credited. formance liquid chromatography with tandem mass spectrometry. We detected high levels
Data Availability Statement: All relevant data are of BMAA (20–748 μg/g) in the brains of 13 of 14 dolphins. To correlate neuropathological
within the paper and its Supporting Information changes with toxin exposure, gross and microscopic examinations were performed on corti-
files. cal brain regions responsible for acoustico-motor navigation. We observed increased num-
Funding: This work was supported by the Herbert bers of β-amyloid+ plaques and dystrophic neurites in the auditory cortex compared to the
W. Hoover Foundation Award 2012-2017: visual cortex and brainstem. The presence of BMAA and neuropathological changes in the
Cyanobacterial toxin BMAA in the Marine Food
Web, Cyanobacterial toxin BMAA in Stranded
stranded dolphin brain may help to further our understanding of cyanotoxin exposure and its
Dolphins, and Cyanobacterial Toxin BMAA and the potential impact on human health.
Marine Environment: Findings in Stranded
Dolphins, BMAA in Marine Environments,
Cyanobacterial Toxin BMAA in Dolphins, http://
www.hwhfoundation.org/; to D.M. and D.D. D.M.
PLOS ONE | https://doi.org/10.1371/journal.pone.0213346 March 20, 2019 1 / 18

BMAA, neuropathology and stranded dolphins
and L.B. were supported by a National Science Introduction

Foundation Award 2008-2009, Award Abstract
#0742285, "BMAA in the food web of South Florida Harmful algal blooms (HABs) are becoming increasingly frequent in fresh water lakes, estuar-
coastal waters," https://www.nsf.gov/awardsearch/ ies, and the sea [1–3] likely due to climate warming and increased phosphorous and nitrogen
showAward?AWD_ID=0742285. effluents [2]. HABs are sometimes associated with widespread dying-off of sea grass and fish
Competing interests: The authors have declared ascribed to the large biomass of algae producing hypoxic conditions in the water [4]. HABs
that no competing interests exist. composed of cyanobacteria contain multiple cyanotoxins including microcystins, nodularin,
cylindrospermopsin, and anatoxins, while those composed of dinoflagellates can contain bre-
vitoxins [5–7]. Acute and chronic exposures to such toxins can be harmful to both humans
and animals resulting in respiratory illnesses, severe dermatitis, mucosal damage, cancer,
organ failure and death [7].
The cyanotoxin, β-N-methylamino-L-alanine (BMAA), has been linked to several neurode-
generative diseases [8–11]. BMAA has been shown to pass the blood-brain barrier where it is
incorporated into brain proteins [3,12–14] inducing misfolding and aggregation [13]. Chronic
dietary exposure to BMAA has been shown to trigger neurodegenerative changes in experi-
mental models including non-human primates [15–17]. BMAA has also been detected in post-
mortem brain tissues of patients with Amyotrophic Lateral Sclerosis (ALS), Parkinsonism
Dementia Complex of Guam (ALS/PDC) and Alzheimer’s disease (AD) [18–20]. Epidemio-
logical studies support a link between BMAA and neurodegenerative disease, since people liv-
ing in close proximity of lakes with frequent cyanobacteria blooms have increased an
incidence of ALS [21–25]. BMAA has been shown to accumulate in long-lived apex predators
like sharks [26], as well as throughout the South Florida marine food web [27,28].
Dolphins that inhabit Florida coastal waters are often exposed to recurring HABs [29–31].
Coastal subpopulations of bottlenose dolphins in the southeastern United States have an
increased risked of cyanobacteria bloom exposure, compared to common dolphins, a pelagic
species, which spend relatively little time in coastal or near-shore environments [27,32]. Bottle-
nose dolphins have a median lifespan of 30 to 32 years with an maximum of 50 years in captiv-
ity [33]. Irrespective of age or sex, bottlenose dolphins have been observed stranded all along
the coastal United States [34–36]. Stranding events have been attributed to trauma [37], infec-
tions [35], neuropathology [38], navy sonar [39,40] or toxin exposures [29]. Regardless of
cause of stranding, dolphins are valuable bio-indicators of the health of the marine environ-
ment and provide clues of potential environmental risks to humans. We hypothesize that dol-
phins are likely to accumulate BMAA both by exposure to HABs and by the ingestion of prey
previously exposed to the cyanotoxin [27,28].
In this study, we examined an opportunistic sample of brain tissues from fourteen stranded
dolphins obtained from the Rosenstiel School of Marine and Atmospheric Science and the
International Fund for Animal Welfare (IFAW). BMAA was measured in select brain regions
of stranded dolphins. In addition, we examined brain tissues of seven dolphins stranded in
Massachusetts to seek a correlation between neuropathological changes and the presence of
BMAA.
Material and methods

Dolphins
Adult (n = 9) and sub-adult (n = 5) stranded dolphins were obtained from ten different sites in
Florida and Massachusetts between 2005 and 2012 (S1 Table). The seven Florida dolphins
were found stranded in sites known to have recurring harmful algal blooms: Atlantic Ocean
[41], Banana River [42], Indian River Lagoon [42,43] and Gulf of Mexico [30]. The seven Mas-
sachusetts dolphins were found stranded in various sites in Cape Cod Bay [44]. Cyanobacteria

species capable of producing BMAA were detected in all these regions as determined by sur-
veillance data from the Phytoplankton Monitoring Network (S2 Table) [45]. Dolphins were
classified as adult or sub-adult based on body length from the Geraci & Lounsbury 1993
stranding guide [46]. Dolphins varied in manner of stranding, species and sex (S1 Table). Six
of fourteen dolphins (43%) were observed alive while stranded. Physical assessments were per-
formed on the dolphins when first sighted [46]; those assessed to be in poor health and unable
to be released were euthanized on site, followed by a full necropsy within 24 to 48 hours by
IFAW. Dolphin specimens in Massachusetts were obtained under a Stranding Agreement
between National Ocean and Atmospheric Administration (NOAA) National Marine Fisher-
ies Services Greater Atlantic Region Fisheries Office and IFAW. That agreement enables access
to and management of stranded pinnipeds and cetaceans, and where applicable, landowners
were consulted when animals were located on private property. Clinical assessments and
euthanasia were conducted according to IFAW’s protocols approved by the Woods Hole
Oceanographic Institution Institutional Animal Care and Use Committee. Euthanasia was
done by intravenous barbiturate, following guidelines from Barco et al 2016 [47]. Euthanasia
decisions were made on the basis of prognosis. The Florida specimens were obtained from
deceased dolphins found off the Florida coast, either beached or floating. Necropsies on these
dolphins were performed at the Hubbs Seaworld Research Institute. These specimens were
obtained with permission from the National Marine Fisheries Services (S1 Table).
At necropsy, external and internal examinations were performed and tissue samples were
collected for histopathology and infectious disease testing (Morbillivirus, Brucella spp., etc.
[48]) where available (S1 Table). Brains from the Massachusetts’ cohort (n = 7 dolphins) were
harvested, and a mid-sagittal cut was made to separate each brain into right and left hemi-
spheres. One hemisphere was frozen and stored in an ultra-low freezer at -80˚C for toxicologi-
cal analysis and the other hemisphere was fixed in 10% neutral buffered formalin for one
month for histopathological assessments. Cryopreserved brain tissues were sampled for analyt-
ical measures of BMAA (see methods below). Formalin-fixed brain hemispheres were used to
assess neuropathological changes (see methods below). Toxicological and histological assess-
ments of dolphin brain specimens were performed with the approval from the NOAA South-
east Region Stranding Program and National Marine Fisheries Service.
Analytical reagents
BMAA, a non-protein amino acid, was measured in dolphin brain tissue using fluorescence
high performance liquid chromatography and ultra-performance liquid chromatography/ tan-
dem mass spectrometry (UPLC-MS/MS) as described in Mondo et al. 2012 & 2014 [26,49]:
AQC Waters AccQTag reagent (PN WAT052880) Reagent Kit was purchased from the Waters
Corporation (Milford, MA, USA). Acetonitrile (ACN), hydrochloric acid (HCl) and HPLC
grade water, were acquired from VWR (Suwanee, GA, USA). β-N-methylamino-L-alanine
hydrochloride (BMAA) DL-2,4-diaminobutyric acid dihydrochloride (DAB)were purchased
from Sigma Aldrich (B-107 and 32830, respectively, St. Louis, MO, USA). N-2(amino)ethylgly-
cine (AEG) was obtained from TCI America (A1153, Portland, OR).
Brain extraction and analytical methods

Frozen punch biospecimens (200 mg) from the cerebral cortex of each dolphin were made
using a 6.0mm Biopunch (Ted Pella, Inc., Redding CA, USA). Punch biospecimens were
obtained from the auditory cortex (ACtx) and the visual cortex (VCtx). Following sampling, tis-
sue biospecimens were subjected to total acid hydrolysis and assayed for total concentrations of
BMAA and BMAA structural isomers N-(2-aminoethyl) glycine (AEG) and 2,4-diaminobutyric

acid (DAB) as described by Mondo et al 2012 [26]. No attempts were made to differentiate free
from protein bound BMAA. Briefly, a 1:6 W/V of 6 N HCl (1000 μL) was added to each cortical
biopsy, followed by sonication. Samples were then hydrolyzed on a heat block for 18 hours at
110˚C, filtered at 13,800 x g for 3 minutes at 4˚C (0.22 lm Ultrafree-MC, Millipore, Bedford,
MA, USA), and concentrated in a speed-vacuum (Thermo-Savant SC250DDA Speed Vac
Plus with a Savant refrigerator trap RVT 4104). The dried extract was then resuspended to
1000 μL of 20 mM HCl. An aliquot of the sample extract was derivatized with AQC using the
AccQ-Fluor reagent (Waters Crop, Milford, MA). The derivatized samples (20 μL resuspended
HCl extract, 60 μL of borate buffer, and 20 μL AccQ-Tag) were run in parallel with buffer and
AQC blanks, and L-BMAA, AEG, DAB and reference amino acid standards. Since control dol-
phin specimens could not be obtained, fresh frozen human brain tissues were used in our assay.
For a negative control, frozen punch biopsies were obtained from the cortex of a 68 year-old
human female with no known neurological changes and previously found to be free of BMAA
[18]. For a positive control, punch biopsies were obtained from a 77 year-old human male with
ALS whose tissues were previously shown to contain a high concentration of BMAA [18]. In
addition, control sample matrixes were also spiked with known amounts of BMAA to confirm
the efficiency of extraction.
HPLC method
BMAA and its structural isomers (AEG & DAB) were detected and quantified using a vali-
dated HPLC method [50,51] with minor modifications [19,26]. Briefly, the non-protein amino
acid BMAA was separated from its structural isomers and other protein amino acids using
reverse-phase elution (Waters Nova-Pak C18 column, 3.9 mm x 300 mm) on a Waters 1525
Binary HPLC pump and a Waters 717 autosampler. The mobile phase consisted of 140 mM
sodium acetate, 5.6 mM triethylamine, pH 5.7 (Eluent A), and 52% (v/v) aqueous acetonitrile
(Eluent B) using a flow rate of 1.0 mL/minute and a 10 μl sample injection volume. The sam-
ples were eluted using a 60 min gradient: 0.0 minute 100% A; 2.0 minutes 90% A; 5.0 minutes
86% A; 10.0 minutes 86% A; 18.0 minutes 73% A; 30.0 minutes 57% A; 35.0 minutes 40% A;
37.5 minutes 100% B; 47.5 minutes 100% B; 50.0 minutes 100% A; 60.0 minutes 100% A. Sam-
ples were derivatized with the AQC fluorescent tag using 20 μL aliquot of sample plus 20 μL
6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) in 60 μL of borate buffer. Com-
pounds were clearly separated with AEG at 29.6 min, BMAA elution at 31.1 min, and DAB at
33 minutes.
Detection was achieved using a Waters 2475 Multi k-Fluorescence Detector with excitation
at 250 nm and emission at 395 nm. Measurements of BMAA in dolphin cortical tissues were
compared with those of the human positive and negative controls (described above), and those
spiked with known amounts of a reference BMAA standard (Sigma B-107; >95% purity,
St. Louis, MO, USA). The limits of detection (LOD) and the limit of quantification (LOQ)
were based on the standard deviation (SD) of response and slope (S), calculated from linearity
of the response of BMAA. The LOD (2.7 ng) and LOQ (7.0 ng) were obtained by using formula
(3.3 x SD)/S and (10 x SD)/S, respectively. The efficiency of the analyte recovery was estimated
adding known amounts of a BMAA standard spiked into a reference sample that was below
the LOD.
UPLC-MS/MS method
BMAA and its structural isomers AEG and DAB were separated, detected and quantified by
ultra-performance liquid chromatography/ tandem mass spectrometry (UPLC-MS/MS) using
a fully validated method as previously described [52,53]. In brief, punch biospecimens of

cerebral cortex (200 mg) were weighed and suspended in 1.0 mL of 6 N HCl sealed with N2 gas
blown into the tubes for 30 s to displace oxygen. Brain samples were hydrolyzed for 18 hours
at 110 ˚C. A subsample of 400 μL was filtered (0.22 μm PVDF Ultrafree MC centrifuge filters;
EMD Millipore; Billerica, MA, USA) and a 100 μL aliquot was dried overnight (Labconco Cen-
trivap; Kansas City, MO, USA). The sample was reconstituted in 1.0 mL 20 mM HCl and
derivatized as described above. BMAA, AEG and DAB were separated by reverse phase C18
chromatography (BEH column 150 × 2.1 mm 1.7 μm; Waters) and eluted with a gradient of 20
mM ammonium formate with 0.2% formic acid (A) and 0.1% formic acid in acetonitrile; (B).
Gradient was delivered by a Waters Acquity I-Class UPLC (Milford, MA, USA) (0 min, 95%
A; 1.0 min, 95% A; 7 min, 85% A; 7.5 min, 78% A; 8 min, 15% A; 8.5 min, 15% A; 8.6 min, 95%
A; 10 min, 95% A) with a flow rate of 0.7 mL/min at 52 ˚C. Compounds were clearly separated
with BMAA elution at 6.56 min (%RSD = 0.23), AEG at 6.67 minute (%RSD = 0.22) and DAB
at 6.82 minutes (%RSD = 0.26). Triplicate measures were performed on each dolphin sample.
Ions were detected on a triple quadrupole tandem mass spectrometer (Waters Xevo TQS, Mil-
ford, MA, USA) with the following parameters: cone voltage was 16 V. Capillary voltage was
set to 2500 V with a source offset of 50 V. Desolvation temperature was 550 ˚C, with a corre-
sponding gas flow of 800 L/hour and a cone gas flow of 150 L/hour. Collision-induced-dissoci-
ation was performed with 99.99% pure argon pressurized to 7.0 bar with a dwell time of 0.05
sec. The characteristic transitions were detected as: BMAA 459 > 258 at collision voltage 18 V,
DAB 459 > 188 at collision voltage 20 V, AEG 459 > 214 at collision voltage 20 V. The method
performance characteristics were determined by definitions of the Environmental Protection
Agency (EPA). The Method Detection Limit (MDL) was < 0.01 pg on column. The lower
Limit of Quantification was > 0.02 pg on column. The average % recovery of BMAA in tripli-
cate samples spiked at 80%, 100% and 120% of the expected value was 98.3%. The precision
was <6% relative standard deviation (%RSD) for all replicated analyses in the sample matrix.
The HorRat value was calculated for triplicate analyses and falls within the AOAC recom-
mended range of 0.5–2.0 [52].
Immunohistochemistry
To investigate neuropathological changes, formalin-fixed brain tissue blocks (2 x 2 x 0.2 cm3)
were collected from the ACtx, VCtx and medulla (Md) from seven dolphins. Tissue blocks
were processed through xylenes and alcohols and embedded in paraffin wax. Microtome sec-
tions were cut at 5 μm and mounted on 1 x 3 inch glass slides. Tissue sections were deparaffi-
nized in three changes of xylene for 10 minutes each, followed by two changes of absolute
ethanol for 5 minutes each, then 95% ethanol for 5 minutes. Tissues sections were stained with
hematoxylin and eosin (H&E) (for cytoarchitecture) and thioflavine-S (for plaques, tangles,
and dystrophic neurites). Modified Bielschowsky’s (MB) silver staining was performed at
AML laboratories using an American MasterTech stain kit (Lodi, CA, USA) [54].
To probe for the specific protein antigens noted below, after rehydration sections were
incubated in 3% H2O2 in methanol for 10 minutes, followed by rinsing in 3 changes of distilled
water (DiH2O) each for 5 minutes. Slides were then incubated in 98% formic acid for 45 sec-
onds, followed by a wash in 3 changes of DiH2O on a Thermolyne Roto Mix shaker, and incu-
bation in phosphate-buffered saline pH 7.4 (PBS) for 5 minutes. To block non-specific
antibody bindings, 10% normal donkey serum in PBS pH 7.4 (PBS) was applied to tissue sec-
tions in a humidity chamber and incubated at room temperature for 30 minutes. Sections
were then probed with mouse anti-β-amyloid (1:800, Covance, Princeton, NJ, USA); anti-neu-
ronal nuclei protein (Anti-NeuN Antibody, clone A60; 1:200, Millipore, MA, USA); or anti-
glial fibrillary acidic protein (GFAP; 1:200, DAKO/Agilent, CA, USA). Antibodies were

allowed to incubate at 4˚C overnight, and then the slides were rinsed with PBS for 3 x 10 min-
utes, followed by additional application of 2% normal donkey serum for 10 minutes then PBS
rinse. A donkey anti-mouse-biotin (1:200; Jackson Immunoresearch, West Grove, PA, USA)
conjugated goat anti-mouse secondary antibody was incubated on tissue sections for 2 hours
at room temperature, rinsed with PBS wash for 10 minutes followed by application of ExtrAvi-
din peroxidase (1:5,000, Sigma-Aldrich, St. Louis, MO, USA) in PBS for incubation for 1 hour.
ExtrAvidin peroxidase was detected using DAB solution (100 mL DAB = 98 mL PBS + 2 mL
25mg/mL DAB + 16.6 μL 3% H2O2) for 10 minutes. Slides were then washed in two changes of
PBS, rinsed with distilled water, counterstained with Gill No. 1 Hematoxylin for 20 seconds,
and rinsed under running tap water for 5 minutes. Frontal cortex samples from a patient with
advanced AD were used as a positive control for all neuropathology markers listed above.
Digital and fluorescent microscopic pathology

Histology slides were scanned at 40x resolution using a TissueScope LE (Huron Digital Pathol-
ogy, Waterloo, ON, CAN). Digital scanning allowed for complete micropathological review of
the entire dolphin brain tissue sections and margins at an optimal resolution of (0.2 μm/pixel
at 40x). High quality tif file images (975 x 477 pixels) were exported from TissueScope LE and
imported into NIH ImageJ 64 VER1.44o (NIH, Bethesda, Maryland, USA) for analysis of Aβ+
density and Aβ+ plaque area. To determine number and area size of Aβ+ plaques, a total of 6
images were analyzed per dolphin (ACtx n = 3 scans and VCtx n = 3 scans). Aβ+ plaques were
annotated and quantified using ImageJ cell counter plug-in software and presented as the
mean and standard error of Aβ+ deposits per 5x scan. A threshold was applied to regions of
interest to determine the area of each Aβ+ plaques. Slides stained with MB silver, thioflavine-S,
4’,6-diamidino-2-phenylindole (DAPI), GFAP and Neu-N were visualized at 20x magnifica-
tion using a Zeiss Apotome fluorescent microscope (USA).
Statistics
Statistical analyses were performed using Prism Version 7 software (Graph Pad, La Jolla, CA,
USA). Single comparisons tests were analyzed using Mann-Whitney U and Wilcoxon
matched-pairs test. Multiple comparisons were analyzed using Friedman test with Dunn’s
multiple comparison test. For correlation analyses, the Pearson’s Correlation Coefficient and
Spearman correlation tests were used to determine significance. All data and results were
expressed as the Mean ± Standard Error; significance level of alpha = 0.05. Due to the limited
size of our sample cohort, Cohen’s d was used to determine effect size and D’Agostino & Pear-
son test was used to determine normality of data. A table of descriptive statistics for each com-
parison can be found in supplementary material (S3 Table).
Results
Stranded dolphins
A total of fourteen dolphins were investigated from the following dolphin species: bottlenose
dolphin (Turiops truncates) and common dolphin (Delphinus delphis) (S1 Table). Nine dol-
phins (64%) in our sample cohort were observed dead. Five dolphins (36%) were part of a
mass stranding. Four dolphins (29%) had conditions that have been shown to cause strand-
ings: one with mechanical trauma (Hubbs 0805 Tt) and three cases of Brucella infections
(IFAW 12-223Dd, 12-228Dd and 12-229Dd) [55]. Seven dolphins (50%) were female; seven
dolphins were male (50%). One dolphin (7%) was pregnant at the time of death. Nine dolphins
(64%) in the study were categorized as adult and five as sub-adult (36%). The bottlenose

dolphins had 1.3-fold greater mean length than the common dolphins (P<0.0001, t-Test) (S1
Table). Bottlenose dolphins were also 4-fold greater in weight than the common dolphin
(P<0.0001, t-Test) (S1 Table). All dolphins except for two (Hubbs 0805-Tt and Hubbs
0630-Tt) had no detectable injury resulting from human interaction. Dolphin Hubbs 0805-Tt
had an old scar on her dorsal fin from a boat injury. Dolphin Hubbs 0630-Tt had fishing gear
entangled around the flukes and hooks in the oral cavity. External examinations of several dol-
phins (29%) showed signs of stranding stress (IFAW12-201Dd), cauliflower-like lesions on
flukes, peduncle and dorsal fin (IFAW12-228Dd), a large circular wound (Hubbs 0805 Tt) and
several broken ribs (Hubbs 0541-Tt). One dolphin (7%) (IFAW12-198Dd) had incidental
post-stranding injuries to the eyes as a result of gull pecking. Gross examinations of internal
structures were unremarkable, except for two dolphins (14%) with non-specific lung (Hubbs
0541 Tt) and spleen (IFAW12-198Dd) pathology. One dolphin had signs of hepatitis
(IFAW12-201Dd) [48].
Detection of BMAA
BMAA was detected in 13 of the 14 dolphins (93%) beached in both Florida and Massachusetts
(Table 1, S1 Table, Fig 1). One dolphin injured by boat collision was negative for BMAA
(Hubbs 0805 Tt). BMAA levels in the other 13 dolphin brains ranged from (20–748 μg/g). The
overall mean concentration of BMAA was 287.9 ± 64.3 μg/g and the median concentration
was 170 μg/g. The bottlenose dolphins that stranded in Florida had 3-fold (P = 0.01; t-Test)
higher concentrations of BMAA than the common dolphins from Massachusetts (Table 1, S1
Table). Sex had no effect on the mean concentration of BMAA. The average BMAA/length of
dolphin was 1.2 ± 0.2 μg/g/cm. BMAA concentrations were positively correlated with the
length of dolphin (n = 13; Pearson r = 0.6002; P = 0.01) (S3 Table). The mean concentration of
BMAA in the stranded dolphin brains was 1.4-fold higher than in reference brains of patients
with AD and ALS (S1 Table) [18]. In some dolphins, BMAA levels detected were up to 3.6-fold
greater than found in those of AD and ALS reference brains (S1 Table). The levels of BMAA
and the BMAA isomers (AEG and DAB) were not region-specific and the mean concentration
of BMAA, AEG, and DAB did not differ in auditory or visual cortex (Table 2). DAB was
detected in 7 of 7 (100%) of dolphins from Massachusetts. The mean concentration of DAB
detected is 2.4-fold higher than BMAA averaged across all dolphins. AEG was detected in 6 of
7 (86%) of dolphins stranded in Massachusetts. The mean concentration of AEG was 4-fold
lower than BMAA and 9.4-fold lower than DAB (P<0.0001; ANOVA). The detection of
BMAA and BMAA structural isomers in the dolphin brain was confirmed using LC-MS/MS
in representative animals sampled from the cohort (S4 Table).
Neuropathology in stranded dolphins

Gross examinations of external brain structures were unremarkable for all stranded dolphins
examined. Brain structures showed normal anatomic landmarks with normal gyral and sulcal
patterns (Fig 2A). No atrophy, white matter lesions, infarcts or hemorrhages were observed on
gross inspection of coronal sections of the brain. Internal gray and white matter structures
were also unremarkable, with clear delineation. The lateral, third and fourth ventricles were
normal in size (Fig 2B). Microscopic examination of H&E sections from the ACtx, VCtx and
Md tissue sections revealed evidence of cellular injury and age-related changes (see below).
Microscopic examination showed hypoxic, shrunken, and eosinophilic neurons in upper and
lower cortical layers and within brainstem nuclei (Fig 2C). Gliosis, mild neuronal satellitosis
and neuronal atrophy were present in cortical layers and brainstem (Fig 2D). Age-related his-
tological changes included accumulation of lipofuscin, chromatolysis, and corpora amylacea

Table 1. Comparison of BMAA detected in brains of stranded dolphins.

Agency ID Necropsy Findings Sex Age Class BMAA (μg/g)
Bottlenose Dolphin (FL)
Hubbs 0805 Tt Boat Injury Female Adult ND
Hubbs 0720 Tt Unknown Female Adult 114
Hubbs 0717 Tt Unknown Male Adult 295
Hubbs 0630 Tt Unknown Female Adult 335
Hubbs 0541 Tt Unknown Male Adult 541
Hubbs 0636 Tt Unknown Male Sub-Adult 675
PCNMF S08-01 Tt Unknown Female Adult 748
Mean ± SE 451 ± 99 ��
Min—Max 114–748
Common Dolphin (MA)
IFAW 12–228 Dd Brucellosis Male Adult 20
IFAW 12–223 Dd Brucellosis Male Sub-Adult 111
IFAW 12–200 Dd Unknown Male Sub-Adult 127
IFAW 12–198 Dd Unknown Female Adult 129
IFAW 12–229 Dd Brucellosis Male Sub-Adult 157
IFAW 12–205 Dd Unknown Female Sub-Adult 170
IFAW 12–201 Dd Unknown Female Adult 320
Mean ± SE 147 ± 34
Min—Max 14–320
FL, Florida; MA, Massachusetts;

��
, P = 0.01 (t-Test); ND, Not Detected; ± SE: Standard Error
https://doi.org/10.1371/journal.pone.0213346.t001
(Fig 2E–2G). In addition, moderate numbers of eosinophilic plaques were seen (Fig 2H) and
edema was observed throughout parts of the cerebral cortex and brain stem of all dolphins
(Fig 2C–2H). The pathological changes described above were present in all dolphins in varying
degrees regardless of age, sex or BMAA concentrations.
Aβ+ Pathology
Anti-Aβ staining showed Aβ+ plaques, ranging in occurrence from rare to very frequent, and
intracellular localization of Aβ in the cerebral cortex of all stranded dolphins (Fig 3A–3C). Aβ+
plaques were observed throughout the ACtx and VCtx. Rare to sparse Aβ+ plaques were local-
ized adjacent to neurons in the Md region of the brainstem (Fig 3D). The Aβ+ plaques varied
from focal clusters of well-defined plaques to diffused and poorly organized plaque-like struc-
tures. Aβ+ plaques were observed on both routine H&E and MB silver stained sections (Fig
3E–3P). Most Aβ+ plaques were senile (diffused) plaques, but rare argophyllic compact neu-
ritic plaques and neuropil threads were observed in the cerebral cortex (Fig 4A–4C). Intracel-
lular silver staining neurofibrillary tangles were observed in cortical neurons with dystrophic
neurites, indicating the presence of intracellular inclusions of insoluble proteins (Fig 4D–4F).
Thioflavin-S staining was used to confirm the Aβ+ staining of plaques and neuronal inclusions
(Fig 5A–5H). GFAP+ fluorescent immunohistochemistry (IHC) provided evidence of reactive
astrocytes around thioflavin-S+ plaques (Fig 5A–5D). Semi-quantitative analysis showed the
densities of Aβ+ plaques were positively correlated in the ACtx and VCtx (n = 7; r = 0.89;
P<0.05) but, was 1.3-fold denser in the ACtx than in the VCtx (P = 0.01; t-Test) (Table 3, S3
Table). Dolphins with Brucella spp. infections had fewer Aβ+ plaques when compared to

Fig 1. HPLC-FD detection of BMAA in the cerebral cortex of stranded dolphins. (A) Separation of 6-aminoquinolyl-N-
hydroxysuccinimidyl carbamate (AQC) derivatized amino acid standards tyrosine (Try), valine (Val), methionine (Met), Lysine
(Lys), Leucine (Leu) and Phenylalanine (Phe). BMAA and BMAA structural isomers N-(2-aminoethyl)-glycine (AEG) &
2,4-diaminobutyric acid (DAB) standards are shown in red. (B) Representative chromatogram from the visual cortex (VCtx) of
stranded dolphin IFAW12-201Dd with low-concentration spikes of BMAA and BMAA isomer standards in the same dolphin
sample shown in blue. Chromatogram shows BMAA has a distinct peak with a retention time of 31.1 minutes. AEG and DAB have
distinct retention times of 29.6 and 33.0 minutes, respectively.
https://doi.org/10.1371/journal.pone.0213346.g001
dolphins without Brucella spp. infections (P = 0.02; t-Test) (S3 Table). The size of the Aβ+ pla-
ques did not differ between the two cortical regions. Neither age group nor sex of dolphins
affected the Aβ+ plaque deposition (S3 Table). Nor was Aβ+ plaque deposition correlated with
brain BMAA concentration (S3 and S5 Tables).
Discussion
As the world’s climate warms, HABs are becoming more frequent, including in eastern China,
which has seen some of the largest cyanobacterial blooms on Earth, and North America. Cya-
nobacteria produce powerful cyanotoxins that impact aquatic and terrestrial life. Exposures to
cyanotoxins are a public health concern as they are linked to organ system damage and disease.
Examining the levels of the cyanobacterial toxin BMAA in apex predators, such as dolphins
and sharks, provides a powerful bio-indicator of the potential for human exposures [26,27,49].
BMAA is produced by cyanobacteria that are sometimes seen as surface blooms, but can
occur in the water column and in benthic mats in lakes, shallow estuaries and bays. In marine

Table 2. Region specific detection of BMAA and BMAA Isomers in the brains of stranded dolphins.
Agency ID ACtx VCtx
BMAA DAB AEG BMAA DAB AEG
(μg/g) (μg/g) (μg/g) (μg/g) (μg/g) (μg/g)
IFAW 12–228 Dd 20.2 ± 0.3 106.4 ± 1.1 NQ ND 97.7 ± 0.0 ND
IFAW 12–223 Dd 52.6 ± 0.6 210.8 ± 0.6 NQ 168.5 ± 0.1 411.1 ± 1.0 50.2 ± 0.7
IFAW 12–229 Dd 120.6 ± 1.0 300.0 ± 4.0 NQ 193.3 ± 1.0 419.5 ±1.1 31.6 ± 0.3
IFAW 12–205 Dd 136.20 ± 1.5 333.7 ± 5.8 NQ 204.4 ± 0.6 461.8 ± 4.3 42.9 ± 3.1
IFAW 12–198 Dd 193.7 ± 5.1 439.9 ± 9.5 NQ 63.9 ± 0.2 244.2 ± 0.4 27.6 ± 1.4
IFAW 12–200 Dd 209.6 ± 2.1 465.0 ± 7.9 42.2 ± 1.0 43.7 ± 0.5 212.9 ± 2.6 17.2 ± 1.7
IFAW 12–201 Dd 328.4 ± 3.5 741.8 ± 7.0 73.5 ± 1.7 312.2 ± 0.9 644.6 ± 3.1 ND
Mean ± SE 151.6 ± 39.2 371.1 ± 77.6�� 57.9 ± 15.7 164.3 ± 40.4 356.0 ± 69.2�� 33.9 ± 5.8
Min—Max 20–328 106–742 42–74 44–312 98–645 17–50
ND, Not Detected; NQ, Not Quantifiable; ± SE: Standard Error;

��
, P<0.0001 (ANOVA)
systems, BMAA enters the food chain via crustaceans and bottom-feeding fish, and then accu-
mulates into long-lived apex predators like sharks [26,27]. We now report that another apex
predator, dolphins, when exposed to cyanobacteria or diets of crustaceans and fish containing
BMAA, bioaccumulate the toxin in their neuroproteins. In our study, BMAA was present in
the brains of stranded dolphins at 1.4-fold greater amounts than in the brains of patients with
AD and ALS [18]. Dolphins beached in Florida were observed to have nearly 3-fold higher
concentrations of BMAA in their brains than dolphins beached in Massachusetts. The differ-
ences in BMAA concentration may be due to differences in diet, occurrences of algal blooms,
varying phytoplankton species or modes of stranding [56].
The recognition that the very high incidence of neurodegenerative disease Guam ALS/PDC
was likely due to the production of BMAA by endophytic cyanobacteria resident in specialized
coralloid roots of cycad trees led to the hypothesis that BMAA is a cause of ALS/PDC and
non-endemic ALS, and provided a link to neurodegenerative disorders [19,20]. In Guam,
cycad seeds contain BMAA and are used to make flour by the indigenous Chamorro. BMAA is
also biomagnified up the food chain to animals like pigs, deer and flying foxes that are part of
the Chamorro diet [20,50,57]. Once ingested, BMAA can cross the blood-brain barrier and
become incorporated into proteins where it is associated with neuropathological changes as
seen in Guam ALS/PDC [12,20]. In this paper, we have now shown that BMAA is detectable
in brains of dolphins that also show neuropathological changes characteristic of human neuro-
degenerative disease. Here, BMAA is also shown to accumulate with increasing length and age
class. Thus, our data suggest dolphins may provide a naturalistic model of BMAA toxin expo-
sures in marine environments. The effects of this exposure however are not yet clear.
Dolphins have a highly evolved cerebral cortex that underlies a sophisticated sonar naviga-
tion system [58]. Previous studies have shown that dolphins have age-related Aβ+ deposits in
their brains [58,59]. Aβ+ plaques, a diagnostic hallmark of AD, were observed to be widespread
in the auditory and visual cortex and to a lesser extent the brainstem of the stranded dolphins
examined in our study. Numerous Aβ+ intracellular inclusions and dystrophic neurites were
also observed. The increased deposition of Aβ+ plaques with accompanying histological
changes (e.g. lipofuscin, corpora amylacea) suggest cellular brain aging. Accumulation of
BMAA in the brains of dolphins may exacerbate these age related changes. This observation
also raises concern about the potential risk to human health of chronic exposure to BMAA
associated with increasingly frequent algal blooms in surrounding lakes and coastal waters.

Fig 2. Gross and microscopic evaluation of postmortem brains from stranded dolphins. (A) External examination was
performed on the cerebral cortex and cerebellum of formalin-fixed hemispheres from stranded dolphins (n = 7). (B) Following
external examinations, brain hemispheres were cut into a series of coronal slices to investigate internal gray and white matter
structures. Tissue blocks were sampled from anatomical regions in the dolphin cerebral cortex and brainstem involved with
acoustico-motor navigation: auditory cortex (ACtx), visual cortex (VCtx), and the medulla oblongata (Md). (C) Digital pathology
scans were obtained from routine histological stain. H&E stain shows hypoxic and eosinophilic changes in neurons of both upper
and lower cortical layers. (D) Gliosis was also observed in the cerebral cortex. (E) Advanced age-related changes were observed
including, corpora amylacea and (F) lipofuscin granules. (G) Karyorrhexis nuclear changes (black arrow) and chromatolysis (white
arrow) were observed. (H) Representative scans of eosinophilic plaques and a rare hemosiderin deposits were observed in the ACtx
of stranded dolphins. Representative scale bar: 5 cm (A, B), 1000 μm (C), 200 μm (D, F, G), 50 μm (E, H).

Fig 3. Aβ deposition in the cerbral cortex of stranded dolphins. (A) Anti-Aβ IHC demonstrates Aβ+ plaques in the cerebral cortex
of stranded dolphins. (B) Intraneuronal Aβ+ accumulation was aslo observed throughout upper and lower cortical layers. (C) High-
resolution digital patholgy scans of pyramidal neurons in the ACtx containing dense intracellalur Aβ+ inclusions. (D) Large and
sparse Aβ+ plaques were observed in the Md of stranded dolphins. (E-P) Aβ+ plaques were different in morphology as observed by
H&E staining (E-H), MB silver staining (I-L) and Aβ+ IHC (M-P). Aβ+ plaques ranged from small focal with compact cores (E, I, M),
primative immature cotton wool-like (F, J, N), to large and diffuse (G, K, O) and ill-defined (H, L, P). Representative scale bar:
500 μm (A, B & D), 100 μm (C), 50 μm (E-P).
Lastly, Marine mammals and fish have long been known to concentrate methylmercury
(MeHg). Many water bodies have both high concentrations of mercury and frequent algal
blooms. High consumption of fish has been linked both to increased blood levels of mercury
and to higher incidence of neurodegenerative disease [60]. The co-occurrence of MeHg and
BMAA has synergistic neurotoxicity compared to single exposures [61] and has been sug-
gested as a link to dementia [62]. We detected BMAA in all but one dolphin beached in Florida
and Massachusetts. Further studies are needed to investigate the synergistic effect of MeHg on

Fig 4. Neurodegenerative changes observed in the brains of stranded dolphins. (A) MB silver staining illustrates
AD-like compact neuritic plaque in the ACtx of a stranded dolphin. (B) A representative digital scan of an ill-defined
agrophyllic plaque containing dystrophic neurites. (C) Dense neuropil threads were observed in the VCtx. (D, E)
agrophyllic neurons containing dystrophic neurites. (F) Coiled body in the ACtx. Representative scale bars: 50 μm.
Fig 5. Thioflavin-S+ pathology in the auditory cortex of stranded dolphins. Thioflavin-S+ staining and IHC with GFAP and Neu-N was used to
determine co-localization of plaques and intracellular inclusions with astroglia and neurons in the stranded dolphin brain. (A, E) Thioflavin-S+ plaques
and intracellular staining was observed in the ACtx (arrows). (B, F) Activated GFAP+ astrocytes are demonstrated surrounding a cluster of dense
thioflavin-S+ plaques. Neu-N+ neurons are shown with intracellular thioflavin-S+ inclusions. (C, G) DAPI staining highlights intact cellular nuclei. (D,
H) The merged panels show the relative co-localization of thioflavin-S+ plaques, blood vessels, and intracellular inclusions within astrocytes and
neurons. Panel inserts (dotted lined boxes) show high magnification detailing thioflavin-S+ structures. Representative scale bar: 150 μm.

Table 3. Region specific quantification of Aβ+ plaques in the stranded dolphin brain.
Agency ID ACtx VCtx ACtx VCtx
Plaques/ Field Plaques/ Field Plaque Area (Px) Plaque Area (Px)
IFAW12-228 Dd 41.3 ± 7.7 25.0 ± 2.2 84.0 ± 9.8 79.6 ± 8.2
IFAW12-229 Dd 42.7 ± 7.9 37.0 ± 6.7 78.7 ± 14.8 90.3 ± 6.8
IFAW12-201 Dd 54.3 ± 2.8 46.0 ± 6.1 55.4 ± 6.6 67.82 ± 9.5
IFAW12-223 Dd 54.7 ± 9.8 30.0 ± 2.4 125.9 ± 34.8 107.16 ± 15.0
IFAW12-198 Dd 77.7 ± 32.3 66.7 ± 9.7 82.0 ± 8.1 102.16 ± 14.7
IFAW12-205 Dd 84.3 ± 7.5 84.3 ± 33.4 89.7 ± 9.8 89.0 ± 8.3
IFAW12-200 Dd 101.0 ± 28.5 70.3 ± 13.0 78.6 ± 6.9 81.0 ± 9.6
Mean ± SE 65.1 ± 8.6�� 51.3 ± 8.5 84.9 ± 8.0 88.1 ± 5.1
Min—Max 41–101 6–84 55–126 68–181
��
, P<0.01 (Paired t-Test); Px, Pixels; ± SE: Standard Error
the dolphin brain. The potential combined neurotoxicity of BMAA and MeHg underscores
the impending impact of climate change on the marine food web.
Conclusion
We have detected BMAA in the cerebral cortex of stranded dolphins. We also report Alzhei-
mer-like neurodegenerative changes in the brains of dolphins containing BMAA. The pres-
ence of BMAA suggests that dolphins provide an excellent sentinel species for toxin exposures
in the marine environment.
Supporting information
S1 Table. Stranded dolphin demographics.
(DOCX)
S2 Table. Surveillance data from the Phytoplankton Monitoring Network obtained in
regions of dolphin stranding.
(DOCX)
S3 Table. Statistical analyses.
(DOCX)
S4 Table. LC-MS/MS detection of BMAA.
(DOCX)
S5 Table. Comparison of BMAA and Aβ+ plaques in dolphin brain.
(DOCX)
Acknowledgments
We gratefully acknowledge the assistance of Megan K. Stolen, MS of Hubbs Seaworld Research
Institute for assistance with Florida dolphin specimen collection and Michael Moore, PhD of
the Woods Hole Oceanographic Institution. We wish to acknowledge Angela M. Amatruda,
HT (ASCP) and the staff at AML Laboratories, Saint Augustine, FL for providing the applica-
tion of silver stains to dolphin brain. We like to thank Sandra A. Banack, PhD of the Institute
for Ethnomedicine and James Hungerford, PhD of the US Food and Drug Administration for
their technical review. We also like to thank the Miami Brain Endowment Bank, a NIH

NeuroBioBank, for providing donated postmortem human brain tissues used in these studies.
We gratefully acknowledge Margaret Basile, MS and John Pablo, PhD formerly of University
of Miami Department of Neurology for their technical assistance.
Author Contributions
Conceptualization: Deborah C. Mash.
Data curation: David A. Davis.
Formal analysis: David A. Davis, Kiyo Mondo, Erica Stern, Ama K. Annor, Susan J. Murch,
Thomas M. Coyne, Larry E. Brand, Paul Alan Cox.
Funding acquisition: David A. Davis, Larry E. Brand, Deborah C. Mash.
Investigation: David A. Davis, Kiyo Mondo, Erica Stern, Ama K. Annor, Susan J. Murch,
Misty E. Niemeyer, Sarah Sharp.
Methodology: David A. Davis, Kiyo Mondo, Susan J. Murch, Larry E. Brand, Misty E. Nie-
meyer, Sarah Sharp, Deborah C. Mash.
Project administration: David A. Davis, Deborah C. Mash.
Resources: Deborah C. Mash.
Software: David A. Davis.
Supervision: David A. Davis, Deborah C. Mash.
Validation: David A. Davis, Susan J. Murch, Paul Alan Cox, Deborah C. Mash.
Visualization: David A. Davis.
Writing – original draft: David A. Davis, Susan J. Murch, Walter G. Bradley, Paul Alan Cox,
Deborah C. Mash.
Writing – review & editing: David A. Davis, Misty E. Niemeyer, Sarah Sharp, Deborah C.
Mash.
References
1. Yan X, Xu X, Wang M, Wang G, Wu S, et al. (2017) Climate warming and cyanobacteria blooms: Looks
at their relationships from a new perspective. Water Res 125: 449–457. https://doi.org/10.1016/j.
watres.2017.09.008 PMID: 28898702
2. Erdner DL, Dyble J, Parsons ML, Stevens RC, Hubbard KA, et al. (2008) Centers for Oceans and
Human Health: a unified approach to the challenge of harmful algal blooms. Environ Health 7 Suppl 2:
S2.
3. Glover WB, Mash DC, Murch SJ (2014) The natural non-protein amino acid N-beta-methylamino-L-ala-
nine (BMAA) is incorporated into protein during synthesis. Amino Acids 46: 2553–2559. https://doi.org/
10.1007/s00726-014-1812-1 PMID: 25096519
4. Lapointe BE, Tomasko DA, and Matzie WR (1994) Eutrophication and trophic state classification of sea-
grass communities in the Florida Keys. Bulletin of Marine Science 54: 696–717.
5. Morabito S, Silvestro S, Faggio C (2017) How the marine biotoxins affect human health. Nat Prod Res:
1–11.
6. van Apeldoorn ME, van Egmond HP, Speijers GJ, Bakker GJ (2007) Toxins of cyanobacteria. Mol Nutr
Food Res 51: 7–60. https://doi.org/10.1002/mnfr.200600185 PMID: 17195276
7. Buratti FM, Manganelli M, Vichi S, Stefanelli M, Scardala S, et al. (2017) Cyanotoxins: producing organ-
isms, occurrence, toxicity, mechanism of action and human health toxicological risk evaluation. Arch
Toxicol 91: 1049–1130. https://doi.org/10.1007/s00204-016-1913-6 PMID: 28110405

8. Nunn PB (2017) 50 years of research on alpha-amino-beta-methylaminopropionic acid (beta-methyla-

minoalanine). Phytochemistry 144: 271–281. https://doi.org/10.1016/j.phytochem.2017.10.002 PMID:
29102875
9. Spencer PS, Nunn PB, Hugon J, Ludolph AC, Ross SM, et al. (1987) Guam amyotrophic lateral sclero-
sis-parkinsonism-dementia linked to a plant excitant neurotoxin. Science 237: 517–522. PMID:
3603037
10. Cox PA, Sacks OW (2002) Cycad neurotoxins, consumption of flying foxes, and ALS-PDC disease in
Guam. Neurology 58: 956–959. PMID: 11914415
11. Bradley WG, Mash DC (2009) Beyond Guam: The cyanobacteria/BMAA hypothesis of the cause of
ALS and other neurodegenerative diseases. Amyotrophic Lateral Sclerosis 10 Suppl 2: 7–20.
12. Xie X, Basile M, Mash DC (2013) Cerebral uptake and protein incorporation of cyanobacterial toxin
beta-N-methylamino-L-alanine. Neuroreport 24: 779–784. https://doi.org/10.1097/WNR.
0b013e328363fd89 PMID: 23979257
13. Dunlop RA, Cox PA, Banack SA, Rodgers KJ (2013) The non-protein amino acid BMAA is misincorpo-
rated into human proteins in place of l-serine causing protein misfolding and aggregation. PLoS One 8:
e75376. https://doi.org/10.1371/journal.pone.0075376 PMID: 24086518
14. Main BJ, Italiano CJ, Rodgers KJ (2017) Investigation of the interaction of beta-methylamino-L-alanine
with eukaryotic and prokaryotic proteins. Amino Acids.
15. Cox PA, Davis DA, Mash DC, Metcalf JS, Banack SA (2016) Dietary exposure to an environmental
toxin triggers neurofibrillary tangles and amyloid deposits in the brain. Proc Biol Sci 283.
16. Cox PA, Davis DA, Mash DC, Metcalf JS, Banack SA (2016) Do vervets and macaques respond differ-
ently to L-BMAA? Neurotoxicology In press
17. Lobner D, Piana PM, Salous AK, Peoples RW (2007) Beta-N-methylamino-L-alanine enhances neuro-
toxicity through multiple mechanisms. Neurobiol Dis 25: 360–366. https://doi.org/10.1016/j.nbd.2006.
10.002 PMID: 17098435
18. Pablo J, Banack SA, Cox PA, Johnson TE, Papapetropoulos S, et al. (2009) Cyanobacterial neurotoxin
BMAA in ALS and Alzheimer’s disease. Acta Neurologica Scandinavica 120: 216–225. https://doi.org/
10.1111/j.1600-0404.2008.01150.x PMID: 19254284
19. Murch SJ, Cox PA, Banack SA, Steele JC, Sacks OW (2004) Occurrence of beta-methylamino-l-ala-
nine (BMAA) in ALS/PDC patients from Guam. Acta Neurol Scand 110: 267–269. https://doi.org/10.
1111/j.1600-0404.2004.00320.x PMID: 15355492
20. Murch SJ, Cox PA, Banack SA (2004) A mechanism for slow release of biomagnified cyanobacterial
neurotoxins and neurodegenerative disease in Guam. Proc Natl Acad Sci U S A 101: 12228–12231.
https://doi.org/10.1073/pnas.0404926101 PMID: 15295100
21. Torbick N, Hession S, Stommel E, Caller T (2014) Mapping amyotrophic lateral sclerosis lake risk fac-
tors across northern New England. Int J Health Geogr 13: 1. https://doi.org/10.1186/1476-072X-13-1
PMID: 24383521
22. Masseret E, Banack S, Boumediene F, Abadie E, Brient L, et al. (2013) Dietary BMAA exposure in an
amyotrophic lateral sclerosis cluster from southern France. PLoS One 8: e83406. https://doi.org/10.
1371/journal.pone.0083406 PMID: 24349504
23. Caller TA, Doolin JW, Haney JF, Murby AJ, West KG, et al. (2009) A cluster of amyotrophic lateral scle-
rosis in New Hampshire: a possible role for toxic cyanobacteria blooms. Amyotroph Lateral Scler 10
Suppl 2: 101–108.
24. Caller TA, Field NC, Chipman JW, Shi X, Harris BT, et al. (2012) Spatial clustering of amyotrophic lat-
eral sclerosis and the potential role of BMAA. Amyotroph Lateral Scler 13: 25–32. https://doi.org/10.
3109/17482968.2011.621436 PMID: 22214351
25. Torbick N, Ziniti B, Stommel E, Linder E, Andrew A, et al. (2018) Assessing Cyanobacterial Harmful
Algal Blooms as Risk Factors for Amyotrophic Lateral Sclerosis. Neurotox Res 33: 199–212. https://
doi.org/10.1007/s12640-017-9740-y PMID: 28470570
26. Mondo K, Hammerschlag N, Basile M, Pablo J, Banack SA, et al. (2012) Cyanobacterial neurotoxin
beta-N-methylamino-L-alanine (BMAA) in shark fins. Mar Drugs 10: 509–520. https://doi.org/10.3390/
md10020509 PMID: 22412816
27. Brand LE, Pablo J, Compton A, Hammerschlag N, Mash DC (2010) Cyanobacterial blooms and the
occurrence of the neurotoxin beta-N-methylamino-L-alanine (BMAA) in South Florida aquatic food
webs. Harmful Algae 9: 620–635. https://doi.org/10.1016/j.hal.2010.05.002 PMID: 21057660
28. Jonasson S, Eriksson J, Berntzon L, Spacil Z, Ilag LL, et al. (2010) Transfer of a cyanobacterial neuro-
toxin within a temperate aquatic ecosystem suggests pathways for human exposure. Proc Natl Acad
Sci U S A 107: 9252–9257. https://doi.org/10.1073/pnas.0914417107 PMID: 20439734

29. Brown A, Foss A, Miller MA, Gibson Q (2018) Detection of cyanotoxins (microcystins/nodularins) in liv-
ers from estuarine and coastal bottlenose dolphins (Tursiops truncatus) from Northeast Florida. Harmful
Algae 76: 22–34. https://doi.org/10.1016/j.hal.2018.04.011 PMID: 29887202
30. Twiner MJ, Fire S, Schwacke L, Davidson L, Wang Z, et al. (2011) Concurrent exposure of bottlenose
dolphins (Tursiops truncatus) to multiple algal toxins in Sarasota Bay, Florida, USA. PLoS One 6:
e17394. https://doi.org/10.1371/journal.pone.0017394 PMID: 21423740
31. Schwacke LH, Twiner MJ, De Guise S, Balmer BC, Wells RS, et al. (2010) Eosinophilia and biotoxin
exposure in bottlenose dolphins (Tursiops truncatus) from a coastal area impacted by repeated mortal-
ity events. Environ Res 110: 548–555. https://doi.org/10.1016/j.envres.2010.05.003 PMID: 20537621
32. Chapra SC, Boehlert B, Fant C, Bierman VJ Jr., Henderson J, et al. (2017) Climate Change Impacts on
Harmful Algal Blooms in U.S. Freshwaters: A Screening-Level Assessment. Environ Sci Technol 51:
8933–8943. https://doi.org/10.1021/acs.est.7b01498 PMID: 28650153
33. Venn-Watson SK, Jensen ED, Smith CR, Xitco M, Ridgway SH (2015) Evaluation of annual survival
and mortality rates and longevity of bottlenose dolphins (Tursiops truncatus) at the United States Navy
Marine Mammal Program from 2004 through 2013. J Am Vet Med Assoc 246: 893–898. https://doi.org/
10.2460/javma.246.8.893 PMID: 25835174
34. Stolen MK, Durden WN, Odell DK (2007) Historical synthesis of bottlenose dolphin (tursiops truncatus)
stranding data in the Indian River Lagoon system, Florida, from 1977–2005. Florida Scientist 70: 45–
54.
35. NOAA (2015) 2013–2015 Bottlenose Dolphin Unusual Mortality Event in the Mid-Atlantic. Health &
Stranding: National Oceanic and Atmospheric Administration
36. Clark LS, Turner JP, Cowan DF (2005) Involution of lymphoid organs in bottlenose dolphins (Tursiops
truncatus) from the western Gulf of Mexico: implications for life in an aquatic environment. Anat Rec A
Discov Mol Cell Evol Biol 282: 67–73. https://doi.org/10.1002/ar.a.20147 PMID: 15622514
37. Bonsembiante F, Centelleghe C, Rossi G, Giglio S, Madeo E, et al. (2017) Clinico-pathological findings
in a striped dolphin (Stenella coeruleoalba) affected by rhabdomyolysis and myoglobinuric nephrosis
(capture myopathy). J Vet Med Sci 79: 1013–1018. https://doi.org/10.1292/jvms.17-0023 PMID:
28442646
38. Pintore MD, Mignone W, Di Guardo G, Mazzariol S, Ballardini M, et al. (2018) Neuropathologic Findings
in Cetaceans Stranded in Italy (2002–14). J Wildl Dis 54: 295–303. https://doi.org/10.7589/2017-02-
035 PMID: 29369721
39. Jepson PD, Deaville R, Acevedo-Whitehouse K, Barnett J, Brownlow A, et al. (2013) What caused the
UK’s largest common dolphin (Delphinus delphis) mass stranding event? PLoS One 8: e60953. https://
doi.org/10.1371/journal.pone.0060953 PMID: 23646103
40. Parsons EC, Dolman SJ, Wright AJ, Rose NA, Burns WC (2008) Navy sonar and cetaceans: just how
much does the gun need to smoke before we act? Mar Pollut Bull 56: 1248–1257. https://doi.org/10.
1016/j.marpolbul.2008.04.025 PMID: 18534632
41. Berry DL, Goleski JA, Koch F, Wall CC, Peterson BJ, et al. (2015) Shifts in Cyanobacterial Strain Domi-
nance during the Onset of Harmful Algal Blooms in Florida Bay, USA. Microb Ecol 70: 361–371. https://
doi.org/10.1007/s00248-014-0564-5 PMID: 25661475
42. Badylak S, Phlips EJ (2004) Spatial and temporal patterns of phytoplankton composition in a subtropical
coastal lagoon, the Indian River Lagoon, Florida, USA. Plankton Research 26: 1229–1247.
43. Sneed JM, Meickle T, Engene N, Reed S, Gunasekera S, et al. (2017) Bloom dynamics and chemical
defenses of benthic cyanobacteria in the Indian River Lagoon, Florida. Harmful Algae 69: 75–82.
https://doi.org/10.1016/j.hal.2017.10.002 PMID: 29122244
44. Sehein T, Richlen ML, Nagai S, Yasuike M, Nakamura Y, et al. (2016) Characterization of 17 New
Microsatellite Markers for the Dinoflagellate Alexandrium Fundyense (Dinophyceae), a Harmful Algal
Bloom Species. J Appl Phycol 28: 1677–1681. https://doi.org/10.1007/s10811-015-0681-7 PMID:
27274617
45. Morton S (2014) US DOC/NOAA/NOS > National Centers for Coastal Ocean Science (2014). Biologi-
cal, chemical, and physical data from the Phytoplankton Monitoring Network from 13 Sep 2001 to 7 Mar
2013 (NODC Accession 0117942). Version 1.1 2014 ed: National Oceanographic Data Center, NOAA.
46. Geraci JR, Lounsbury VJ (1993) Marine Mammals Ashore, A Field Guide for Strandings: Texas A&M
University Sea Grant Publication. 344 p.
47. Barco SG, Walton WJ, Harms CA, George RH, D’Eri LR, et al. (2016) Collaborative development of rec-
ommendations for euthanasia of stranded cetaceans. NOAA Technical Memorandum NMFS-OPR-56:
http://bit.ly/2ydE2CK

48. Montalvo Villalba MC, Cruz Martinez D, Ahmad I, Rodriguez Lay LA, Bello Corredor M, et al. (2017)
Hepatitis E virus in bottlenose dolphins Tursiops truncatus. Dis Aquat Organ 123: 13–18. https://doi.
org/10.3354/dao03085 PMID: 28177289
49. Mondo K, Broc Glover W, Murch SJ, Liu G, Cai Y, et al. (2014) Environmental neurotoxins beta-N-
methylamino-l-alanine (BMAA) and mercury in shark cartilage dietary supplements. Food Chem Toxicol
70: 26–32. https://doi.org/10.1016/j.fct.2014.04.015 PMID: 24755394
50. Banack SA, Cox PA (2003) Biomagnification of cycad neurotoxins in flying foxes: Implications for ALS-
PDC in Guam. Neurology 61: 387–389. PMID: 12913204
51. Banack SA, Johnson HE, Cheng R, Cox PA (2007) Production of the neurotoxin BMAA by a marine cya-
nobacterium. Mar Drugs 5: 180–196. PMID: 18463731
52. Glover WB, Baker TC, Murch SJ, Brown PN (2015) Determination of beta-N-methylamino-L-alanine, N-
(2-aminoethyl)glycine, and 2,4-diaminobutyric acid in Food Products Containing Cyanobacteria by
Ultra-Performance Liquid Chromatography and Tandem Mass Spectrometry: Single-Laboratory Valida-
tion. J AOAC Int 98: 1559–1565. https://doi.org/10.5740/jaoacint.15-084 PMID: 26651568
53. Baker TC, Tymm FJM, Murch SJ (2018) Assessing Environmental Exposure to beta-N-Methylamino-L-
Alanine (BMAA) in Complex Sample Matrices: a Comparison of the Three Most Popular LC-MS/MS
Methods. Neurotox Res 33: 43–54. https://doi.org/10.1007/s12640-017-9764-3 PMID: 28643233
54. Mirra SS, Hart MN, Terry RD (1993) Making the diagnosis of Alzheimer’s disease. A primer for practic-
ing pathologists. Arch Pathol Lab Med 117: 132–144. PMID: 8427562
55. Guzman-Verri C, Gonzalez-Barrientos R, Hernandez-Mora G, Morales JA, Baquero-Calvo E, et al.
(2012) Brucella ceti and brucellosis in cetaceans. Front Cell Infect Microbiol 2: 3. https://doi.org/10.
3389/fcimb.2012.00003 PMID: 22919595
56. Bogomolni AL, Pugliares KR, Sharp SM, Patchett K, Harry CT, et al. (2010) Mortality trends of stranded
marine mammals on Cape Cod and southeastern Massachusetts, USA, 2000 to 2006. Dis Aquat Organ
88: 143–155. https://doi.org/10.3354/dao02146 PMID: 20225675
57. Cox PA, Banack SA, Murch SJ (2003) Biomagnification of cyanobacterial neurotoxins and neurodegen-
erative disease among the Chamorro people of Guam. Proc Natl Acad Sci U S A 100: 13380–13383.
https://doi.org/10.1073/pnas.2235808100 PMID: 14612559
58. Sarasa M, Pesini P (2009) Natural non-trasgenic animal models for research in Alzheimer’s disease.
Curr Alzheimer Res 6: 171–178. https://doi.org/10.2174/156720509787602834 PMID: 19355852
59. Gunn-Moore D, Kaidanovich-Beilin O, Gallego Iradi MC, Gunn-Moore F, Lovestone S (2017) Alzhei-
mer’s disease in humans and other animals: A consequence of postreproductive life span and longevity
rather than aging. Alzheimers Dement.
60. Andrew AS, Chen CY, Caller TA, Tandan R, Henegan PL, et al. (2018) Toenail mercury Levels are
associated with amyotrophic lateral sclerosis risk. Muscle Nerve.
61. Rush T, Liu X, Lobner D (2012) Synergistic toxicity of the environmental neurotoxins methylmercury
and beta-N-methylamino-L-alanine. Neuroreport 23: 216–219. PMID: 22314682
62. Eiser AR (2017) Why does Finland have the highest dementia mortality rate? Environmental factors
may be generalizable. Brain Res 1671: 14–17. https://doi.org/10.1016/j.brainres.2017.06.032 PMID:
28687259

APPENDIX X:
AUDUBON LETTER
4500 Biscayne Blvd., Suite 350
Miami, FL 33137
305-371-6399
fl.audubon.org
April 5, 2019
Colonel Andrew Kelly

District Commander
United States Army Corps of Engineers
701 San Marco Boulevard
Jacksonville, Florida 32207
Mr. Drew Bartlett

Executive Director
South Florida Water Management District
3301 Gun Club Road
West Palm Beach, Florida 33406
VIA E-MAIL:
Andrew.D.Kelly@USACE.Army.Mil
DBartlett@sfwmd.gov
Subject: Lake Okeechobee Discharges
Dear Colonel Kelly and Mr. Bartlett:

This letter is in response to a letter you received on March 15, 2019 signed by 17 entities expressing
concern about recent and current Lake Okeechobee releases. A copy of the letter is attached for your
convenience. The point of the letter is made clear in the second sentence which states “we are stakeholders
interested in . . . water supply.” The authors express concerns about Okeechobee releases and assert that
the Corps and the South Florida Water Management District are taking a “politically-concocted approach,
lacking scientific support.” To support their position, the letter makes arguments we consider inaccurate
and/or misleading. We take this opportunity to set the record straight.
Paraphrasing, the principal arguments in the letter are: 1) that the discharges depart from balancing all
authorized C&SF purposes and that lowering the lake in the dry season in the past has resulted in severe
economic and environmental consequences, especially from water shortages, 2) the discharges do not
follow prior modeling, and 3) the agencies should store more water in the lake, not less.
1) Discharges depart from balancing all authorized C&SF purposes + Lowering the lake in
the dry season in the past has resulted in severe economic and environmental consequences,
especially from water shortages.
Water supply is one C&SF authorized purpose, but the letter omits that Lake Okeechobee, the
Caloosahatchee and St. Lucie Estuaries, the Everglades and their fish and wildlife are authorized purposes
that have suffered extreme harm from the lake reaching 16 or more feet in depth in 6 of the last 7 years.
The submerged aquatic vegetation (SAV) zone is arguably the most important plant zone in the lake and
the recent chronic deep water has reduced its coverage from 44,000 acres in 2012 to only about 5,000
acres in 2018. The high water degradation to the lake’s SAV has been accompanied by repeated
catastrophic estuary releases that reduced their seagrass, oysters and other biota. In many years, the
releases also contained toxin-forming cyanobacteria. The estuary communities have experienced repeated
severe economic, ecological, and probable human health impacts.
In contrast to the letter’s assertion of high risk of low water harm, since 1965, Lake Okeechobee has been
kept harmfully deep in 25 years as compared to only 8 water supply cutback years.1 Yet, the letter implies
management focus should be on drought years over a fear of water cutbacks, which are uncommon
compared with high water years. At this point, the lake and estuary communities urgently need
hydrological conditions conducive to the recovery of their ecosystems and economies and the management
focus should be on them.
In order for lake managers to resuscitate the lake’s SAV community, sunlight must reach the seeds lying
on the bottom in the SAV zone (primarily the 9-11 foot contour). This would require the lake dropping
to about 11 feet for roughly the next 2 – 3 months. This year we appear to have a 1-in-10 year opportunity
to drop lake levels to the 11-foot range to allow SAV seed banks to receive the sunlight they need to
germinate and grow. The estuaries in turn, need relief from harmful releases to allow their grasses and
biota a chance to regrow, and low lake levels this year will make that recovery condition more likely. The
El Nino pattern’s wetter-than-normal dry season may preclude getting the lake down to 11 feet, but even
so, releases now can help the estuaries by lowering the lake before the upcoming wet season. Given the
past seven years of high water events and the loss of so much lake and estuary habitat, the Corps’ approach
is prudent and timely. The efficacy of this approach is based on recovery experiences of the late 1980s,
late 1990s, and after the 2004-05 storms. We support the efforts to manage these ecosystems for recovery.
2) The discharges do not follow prior modeling.
The letter goes on to assert that the discharges “exceed the volumes analyzed in the LORS08 Supplemental
Environmental Impact Statement or the LORS08 2018 (sic) Biological Opinion.” The LORS08 decision
tree makes release recommendations phrased as releases “up to” a certain amount. When modeling, the
agencies tend to use the maximum “up to” values in simulations. In reality, the agencies end up using less
than the recommended “up to” amounts, thus keeping Lake Okeechobee deeper than what the model
predicted. Figure 1 shows that observed lake stages have been deeper than predicted in LORS08. For
example, the LORS08 model predicted Lake Okeechobee would be below 12 feet about 25% of the time,
1
Harmfully deep is defined as the lake staying above 12.5 feet for an entire year which prevents restorative drawdowns. The
8 drought years cited are 1971, 1981, 1991, 2001, 2007, 2008, 2009, and 2011.
but from 2008-2017 the lake was lower than that only about 16% of the time, and from 2013-2017 it was
below 12 feet only about 5% of the time. We consider the releases the Corps is making now to be fully
within this “make-up” zone.
Figure 1. Projected lake levels from LORS08 compared with actual lake levels in recent years. Source: SFWMD. 2018.
Lake Okeechobee stage effects on lake ecology. Presentation to Committee on Independent Scientific Review of Everglades
Restoration Progress. February 22, 2018.
3) The agencies should “store more water in the lake, not less”.
The authors of the letter state “the prudent approach, supported by sound independent science, is to . . .
store more water in the lake, not less.” This is out of touch with the reality of a fragile Herbert Hoover
Dike and the families, businesses and wildlife that had to endure years of deep water and harmful, and at
times toxic, releases. Alarmingly, reaching or exceeding 16 feet in 6 out of the last 7 years has made the
Herbert Hoover Dike a source of significant risk. The occurrence of an extreme weather event while at
or above 16 feet could have put the lake at dangerous levels. Storms routinely raise lake levels 3-5 feet,
as evidenced by Tropical Systems Fay and Irma that elevated lake levels by 4 and 3.5 feet in one month,
respectively.
The ongoing dike rehabilitation project is only partially finished and the dike remains far from safe.
Culvert replacements are ongoing and only about half of the cutoff wall designed to reduce piping (leaks
through the dike that are the greatest threat of dike failure) has been finished. The Corps deemed the reach
from Belle Glade to Clewiston an “Intolerable Risk Area” (Fig. 2) and cutoff wall installation for this long
stretch has barely begun. Recommending storing more water in the lake now is irresponsible. Suggesting
that sound independent science supports doing so is misleading at best. Advocating for storing more water
in the lake once repairs are fully finished, is ignoring the history of lake management. History has shown
time and again that a deeper Lake Okeechobee is a dirtier, more dangerous, unhealthy lake and a source
of harm to the estuaries. Conversely, a deeper lake only benefits water users to the detriment of Florida’s
wildlife, families, and our multi-billion tourism-based economy. A deeper lake represents no shared
adversity.
Figure 2. The blue line below is yet to have a cutoff wall installed and is rated by the USACE as an Intolerable Risk Area
(USACE 20162). Raising lake levels at this time would be dangerous to this 20 plus-mile stretch of dike. (Source: USACE
2016. Herbert Hoover Dike rehabilitation project. Public meeting presentation in Clewiston, FL.)
In summary, the fact is that LORS08 keeps the lake too deep causing harm to the lake and the estuaries,
and the Herbert Hoover Dike remains unsafe posing a real and present danger to human health and safety.
Lowering Lake Okeechobee at this juncture meets the purpose of flood control, has a low chance of
contributing to water shortages, and can start the healing process for beleaguered ecosystems and the fish
and wildlife resources therein.
Conversely, the letter almost singularly focused on water supply and downplayed the other purposes of
the C&SF project, to the point of prioritizing their water supply interests over others under the guise of
“balance.” We urge you to continue to weigh experience-based knowledge into the current process and
continue managing the lake to balance the genuine interests of all stakeholders, including the lake,
estuaries, and the Everglades, for the benefit of all Floridians.
2
USACE 2016. Herbert Hoover Dike rehabilitation project. Public meeting presentation in Clewiston, FL.
On a final note, we applaud and thank the agencies for trying to recover these damaged ecosystems and
economies by facilitating hydrological conditions suitable for recovery. We look forward to working with
your agencies and supporting your efforts to improve lake management.
Sincerely,
Celeste De Palma
Director of Everglades Policy
Audubon Florida
cc: Secretary Noah Valenstein, Florida Department of Environmental Protection

noah.valenstein@dep.state.fl.us
Lieutenant Colonel Jennifer A. Reynolds, United States Army Corps of Engineers
jennifer.a.reynolds@usace.army.mil
Chairman Chauncey Goss, South Florida Water Management District
cgoss@sfwmd.gov
Vice Chair Scott Wagner, South Florida Water Management District
swagner@sfwmd.gov
Governing Board Member Jacqui Thurlow-Lippisch, South Florida Water Management District
jthurlowlippisch@sfwmd.gov
Governing Board Member Charlie Martinez, South Florida Water Management District
cmartinez@sfwmd.gov
Governing Board Member Cheryl Meads, South Florida Water Management District
cmeads@sfwmd.gov
Governing Board Member Charlette Roman, South Florida Water Management District
croman@sfwmd.gov
Governing Board Member Jay Steinle, South Florida Water Management District
jsteinle@sfwmd.gov
March 15, 2019
VIA E-MAIL: ANDREW.D.KELLY@USACE.ARMY.MIL AND

DREW.BARTLETT@DEP.STATE.FL.US
Colonel Andrew Kelly

District Commander
United States Army Corps of Engineers
701 San Marco Boulevard
Jacksonville, Florida 32207
Mr. Drew Bartlett

Executive Director
South Florida Water Management District
3301 Gun Club Road
West Palm Beach, Florida 33406
Subject: Lake Okeechobee Discharges
Dear Colonel Kelly and Mr. Bartlett:
The undersigned represent a large cross-section of south Florida’s economy and communities,
both private and public, that rely on Lake Okeechobee (Lake) and predictable functioning of the
Central and Southern Florida Flood Control Project (C&SF Project). We are stakeholders
interested in the decisions that affect water supply in the Lake and have engaged extensively in
water management decisions and the operation of the C&SF Project for decades.
We write to express our grave concerns with the decisions currently being made by the United
States Army Corps of Engineers (Corps), with the South Florida Water Management District’s
(SFWMD) support, that could drive the Lake to extreme low levels. The south Florida region
has lived through prior agency decisions to lower the Lake in the dry season in anticipation of
wet season rain that never came. Severe economic and environmental consequences resulted
from those decisions. Many have experienced the harsh reality of gambling on Mother Nature
and being wrong.
These lessons were unforgettable. Yet, despite this well-known history, the Corps, along with
the SFWMD, appear on track to repeat the mistakes of the past. Your plan to drive the Lake
down, as noted in the Corps’ Memoranda for the Record (MFR) dated October 2018 and
February 2019, is even more aggressive than these past drawdown decisions.
The Lake stage is already low for this time of year, approximately one foot below the bottom of
the Lake’s preferred ecological operating range. The Lake stage recently entered the “Beneficial
Use Band.” In this band, the operating schedule calls for the Corps and SFWMD to conserve
water, but discharges continue. The current release rates are unprecedented, at a combined rate
of 2,300 cfs (up from 1000 cfs in the October 2018 MFR), plus the unquantified volume sent
south, per the MFR dated February 22, 2019. These releases, being made in the absence of any
1
extreme weather conditions, exceed the volumes analyzed in the LORS08 Supplemental
Environmental Impact Statement or the LORS08 2018 Biological Opinion. Already vegetation
is stressed in the stormwater treatment areas, which may affect water quality; water conservation
areas above their regulation schedules; increased discharges to the northern estuaries; and
unwanted discharges to the Lake Worth Lagoon. All of these impacts currently exist, yet the
discharges continue, jeopardizing the environment and water supplies for municipalities
(representing millions of residents), business, and agriculture.
Congress spoke clearly when it stated that the Lake was to serve multiple purposes and the
Lake’s regulation schedule was to balance all authorized C&SF Project purposes. The current
discharges have significantly departed from this mandate. Under the so-called “Additional
Operational Flexibility” (AOF), a tool to be used “infrequently” and for “unique” occurrences,
the Corps has made discharges for months, with no end in sight. The prudent approach,
supported by sound independent science, is to take advantage of the billions of dollars spent to
date to repair the Herbert Hoover Dike, and store more water in the Lake, not less.
With no opportunity for stakeholder engagement, and no meaningful analysis of the adverse
effects to our communities and the environment, we must strenuously request the Corps and the
SFWMD halt these current discharges. We ask the Corps to return to typical operations under
the approved water control plan, while we, as a community, work together to develop the new
Lake Okeechobee System Operating Manual (LOSOM).
We look forward to partnering with the SFWMD and the Corps in the management of the
system, within the approved schedules, not a politically-concocted approach, lacking scientific
support. Our water resources are too precious to gamble with anything less.
Sincerely,
Lake Worth Drainage District

Okeechobee County
City of Okeechobee
City of West Palm Beach
Southeast Florida Utilities Council
Associated Industries of Florida
Okeechobee Economic Council
Florida Land Council
H2O Coalition
Florida Fruit and Vegetable Association
Florida Nursery, Growers and Landscape Association
Florida Farm Bureau Federation
Indian River County Farm Bureau
Richard Budell, Executive Director, Florida Agribusiness Council and Former Director of
Water Policy, Florida Department of Agriculture and Consumer Services
Russ Frydenborg, President, Frydenborg EcoLogic and Former Program Administrator, Division
of Environmental Assessment and Restoration, Florida Department of Environmental
Protection
2
Frank Nearhoof, President, Nearhoof Environmental Consulting and Former Program
Administrator, Division of Environmental Assessment and Restoration, Florida
Department of Environmental Protection
Henry Dean, Former Executive Director of the South Florida Water Management District
cc: Secretary Noah Valenstein, Florida Department of Environmental Protection

(noah.valenstein@dep.state.fl.us)
Lieutenant Colonel Jennifer A. Reynolds, United States Army Corps of Engineers
(jennifer.a.reynolds@usace.army.mil)
Ms. Terrie Bates, Water Resources, South Florida Water Management District
(tbates@sfwmd.gov)
3
APPENDIX Y:
LETTERS REGARDING MINIMUM FLOW LEVELS
FOR THE CALOOSAHATCHEE
APPENDIX Z:
FLORIDA INTERNATIONAL UNIVERSITY STUDY
Jennifer S Rehage
Florida Bays: updates,
fishscience@fiu.edu
ffects & recovery from
Hurricane Irma
5 points on Florida Bay…
2 points: context
Context A: Chronic deficit of water to Florida Bay
Context A: Chronic deficit of water to Florida Bay
Current vs historical inflows into Everglades National Park
Historical
Current
Context A: Highest in Taylor Slough & dry season
Taylor
Slough
Marshall 2009 –
Paleo-based blow
Context A: Highest in Taylor Slough & dry season
Taylor
Slough
Marshall 2009 –
Paleo-based blow
Context A:
Context 1: Highest
Deficit =inhigh
Taylor Slough
salinity & dry season
+ hypersalinity
Hypersalinity
Context A:
Context 1: Highest
5-20 pptinhigher
Taylorsalinities
Slough & dry season
Historical vs. present salinity
Joe
Bay 15 Pre-drainage
24
3 Present
Little
Madeira
29 Bay 29
8
10 17
24
Garfield
Bight 4 Duck
33 Key
33
Terrapin 31
Bay
25 22
Buoy
18
Key
Murray
Key
35 Butternut
Key
27 33
21
Johnson
Key
Bob Allen
34 Key
27
Little Rabbit
36 Paleo-salinities (cores) =
Key
31 Restoration targets
Peterson
Key Marshall et al. 2009, 2014
Context A:
Context 1: Highest
Routine in Taylor Slough & dry season
hypersalinity
29
10
Garfield
Bight
2
Context A:
Context 1: Highest
Routine in Taylor Slough & dry season
hypersalinity
Garfield Bight
29
10 80
Garfield
Bight
2015
Garfield 70
Bight drought
60
2
50
Salinity
ppt
40
30
20
10
0
96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18
Period of Record (1996 - present)

Context 2:
Context 1: Sea
Highest
levelinrise
Taylor Slough &stage,
is increase dry season
salinity
Context A:
Context 1: Highest in rise
Sea level Taylor Slough &stage,
salinity
Context 2:
Context 1: Sea
Highest
levelinrise
Taylor Slough &stage,
salinity
These conditions erode resilience
Resilience: The ability of a system to cope & adapt in
response to disturbance, allowing to sustain form, function,
& identity
These conditions erode resilience
The ability of a system to cope & adapt in response to
disturbance, allowing to sustain form, function, & identity
Could hurricane build resilience?
Hurricane tracks 1900-present

Could hurricane build resilience?
Hurricane
Wilma
• October 25,
2005
• Category 2-3
• 120 mph winds
• 8 ft surge
Context
Point 1: Highestdeposits
1: Hurricane in Taylormud
Slough & dryelevation
to add season
Context
Point 1: Highestdeposits
1: Hurricane in Taylormud
Slough & dryelevation
to add season
Up to 10 cm of Gulf of Mexico mud (rich)

~ 100 years of forest accumulation
Mud
deposit
3x
Phosphorus
Mangrove
soil
Mangroves Surge
attenuation
Attenuate storm surge model
Zhang et al. 2012

Point 2: Hurricane deliver large volumes of water
Stage (cm)
-40
-20
0
20
40
60
80
100
120
01
/2
0
01 01
/2
0
01 02
/2
0
01 03
/2
0
01 04
/2
0
01 05
/2
0
01 06
/2
0
01 07
/2
0
01 08
/2
0
01 09
/2
0
01 10
/2
0
01 11
/2
0
01 12
/2
0
May 2000 - March 2018

01 13
Water Levels for MO215
/2
0
01 14
/2
0
01 15
/2
0
01 16
/2
0
01 17
/2
01
8
MO215
2D Graph 7
44
42
40
38
Average
Salinity (ppt)
Buoy 36 1997-2016
Key 34
32
30
28
2017-18
26
Sept Oct Nov Dec Jan Feb Mar Apr May
Month
Col 1 vs AVERAGE 1997-2016

Col 1 vs 17-18
Point 2: What are effects on fisheries?
Upcoming study will examine how

blooms/dieoff affect fish, their prey
& health
Point 3: Major recruitment event
Recruitment = new baby fish that should have a

positive impact on fishery in years to come
Snook emigration effects

• 2/3 of snook exited
river on day of the
hurricane!
• No effects on bass
• Being observed for multiple species => implication for quality of recreational fisheries
Spotted seatrout Common snook
Reddrum
Bonefish
Garfield Bight
April 2018
Baby bonefish collected in
Garfield Bight (awaiting
genetic testing)
Hurricanes can act as
resilience agents by:
• Stimulating
recruitment
• Building soil
(elevation) &
delivering nutrients
• Replenishing
freshwater

Brian Mast LOSOM Scoping Comment

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Brian Mast LOSOM Scoping Comment

Загружено:

Авторское право:

Доступные форматы

MEMORANDUM:

NEPA Scoping Process Public Comment

I. AUTHORIZED PROJECT PURPOSES ......................................................................... 2

I. AUTHORIZED PROJECT PURPOSES

b. Preservation of Everglades National Park

II. PUBLIC HEALTH

a. Connection Between Discharges and Public Health Threats

Coastal Manag. 2009;52(7):342-347; Koreivienė J, Anne O, Kasperovičienė J, Burškytė V. Cyanotoxin management

III. WATER QUALITY

b. Misinformation Surrounding Septic Tanks

c. Impacts of Freshwater Infusion and Pollution

1. Submerged Aquatic Vegetation (SAV): “Changes in normal salinity ranges can

IV. ADDITIONAL ENVIRONMENTAL CONCERNS

As a result of these trends, the National Academies of Sciences has recommended an

V. RECREATIONAL ACTIVITIES AND ECONOMIC IMPACTS

VI. ENHANCMENT OF FISH AND WILDLIFE

Below are several recommendations to accomplish this goal:

e. LOSOM should incorporate human health and safety protections:

VIII. UNRESOLVED QUESTIONS

Mary Jane Angelo

© University of Florida, March 2015

 400,000 acre-feet of water storage within the Caloosahatchee River watershed,

1. Accelerate funding and completion of existing approved projects

 Obtain federal authorization for CEPP,

2. Provide Water Storage and Treatment North of Lake Okeechobee

4. Deep Well Disposal of Excess Flows

Everglades Technical Review Team Study Objectives and Approach

1. A Complex Inter-Connected System

2. Climatological, Topographic and Geologic Constraints

3. Engineering Constraints due to System Design

4. Need for Sustained, Long-term, Holistic Planning and Execution

Issued Addressed by C&SF authorized purposes

Rights of Existing Legal Users of Water

Water Quality Issues in the EPA

Water Quality Issues in the St. Lucie and Caloosahatchee Estuaries

The Endangered Species Act and Migratory Bird Treaty Act

WY 1998-2014 Annual Average: WY 2014 Annual Total: WY 2014 Wet Season:

Total Inflow Basin Contribution LO Contribution

 A 360,000 acre-ft above-ground reservoir in the Everglades Agricultural (EAA) to provide

b. Northern Everglades and Estuaries Protection Program (NEEPP)

The Caloosahatchee River Watershed Protection Plan (SFWMD, 2009b) identified a

 Provision of approximately 400,000 acre-ft of water storage within the Caloosahatchee

c. River of Grass (ROG) Planning Process

 400,000 acre-feet of water storage within the Caloosahatchee River watershed,

3. Construction Project Status

a. CERP Indian River Lagoon-South (IRL-S) Project

Expected Benefits: The IRL-S project is expected to provide significant water-quality

Condition St. Lucie

d. NEEPP Lake Okeechobee Watershed Construction Project Phase II Technical Plan

Storage ID Sub-watershed Storage Capacity TP load Source Water

ASR ID Sub-watershed Storage TP load Comments

Condition St. Lucie Caloosahatchee

 The LOP2TP elements described in Section III.2.d above.

 Dispersed Water Management: As of 2014, the DWM either in operation or under

f. NEEPP Caloosahatchee River Watershed Protection Plan

g. State of Florida Restoration Strategies

h. CERP Central Everglades Planning Project (CEPP)

A careful analysis of CEPP project construction phasing should be conducted to determine

1. Greater Everglades Watershed Descriptions

Northern Everglades-Lake Okeechobee Watershed

Northern Everglades- Caloosahatchee River and St. Lucie River Watersheds

Southern Everglades- Everglades Agricultural Area and C-139 Basins

2. Greater Everglades Watershed Nutrient Loads

Northern Everglades-Lake Okeechobee Watershed