You are on page 1of 20

Palaeogeography, Palaeoclimatology, Palaeoecology (Global and Planetary Change Section), 89 (1991) 379-398 379

Elsevier Science Publishers B.V., Amsterdam

Global coastal hazards from future sea level rise

Vivien G o r n i t z
NASA GSFC Institute [or Space Studies and Columbia University, New York, N Y 10025, USA

(Received 6 December 1989; revised and accepted 7 June, 1990)

ABSTRACT

Gornitz, V., 1991. Global coastal hazards from future sea level rise. Palaeogeogr., Palaeoclimatol., Palaeoecol. (Global
Planet. Change Sect.), 89: 379-398.

A rise of sea level between 0.3 and 0.9 m by the end of the next century, caused by predicted greenhouse climate
warming, would endanger h u m a n populations, cities, ports, and wetlands in low-lying coastal areas, through inundation,
erosion and salinization. T h e consequences of a global sea level rise would be spatially non-uniform because of local or
regional vertical crustal movements, differential resistance to erosion, varying wave climates, and changeable longshore
currents.
Although m a n y factors can influence sea level, leading to a noisy record, various studies utilizing tide-gauge data find
an average global rate of sea level rise of 1-2 m m / y r , over the last 100 years. This trend is part of a general rise over the
last 300 years, since the low point of the Little Ice Age. Sea level rise m a y accelerate 3-8 times over present rates, within
the next century.
The permanently inundated coastal zone would extend to a depth equivalent to the vertical rise in sea level. Major
river deltas, coastal wetlands and coral islands would be most affected. Episodic flooding by storm waves and surges
would penetrate even farther inland. Beach and cliff erosion will be accentuated. Saltwater penetration into coastal
aquifers and estuaries could contaminate urban water supplies and affect agricultural production.
Research on relative risks and impacts of sea level rise on specific localities is still at an early stage. Development of a
global coastal hazards data base, intended to provide an overview of the relative vulnerabilities of t h e world's coastlines,
is described in this paper. To date, information on seven variables, associated with inundation and erosion hazards, has
been compiled for the U.S., and parts of Canada and Mexico. A coastal vulnerability index (CVI) has been designed to
flag high risk coastal segments. Preliminary results are presented for the eastern United States, as a test case.

Introduction with increased risk to human life and property


along low-lying coastal areas.
The greenhouse climate warming, caused by The effects of the worldwide sea level rise
the atomospheric buildup of CO2 and trace gases (SLR) will be spatially nonuniform, for several
(Hansen et al., 1988; Pearman, 1988) could raise reasons. First of all, the global or eustatic change
sea level by 0.3 m by 2050 and 0.6 m by 2100 is superimposed on local vertical crustal move-
(Meier, 1989; Oerlemans, 1989). Although sig- ments. Areas where land is rising, due to glacial
nificantly reduced from previous estimates rebound (e.g. Fennoscandia, the Canadian
(NRC, 1987), these figures still represent an in- shield), or tectonism (much of the Pacific coast
crease of 3-8 times over present rates of sea of the Americas), will be at less risk than subsid-
level rise. Locally, increases could be still greater, ing regions (such as the Mississippi and Nile
depending on land subsidence factors. Accelerat- deltas; the Low Countries). Secondly, the char-
ing sea level trends could seriously exacerbate acteristics of any given coastline results from
erosion, inundation and salinization hazards, the interaction between lithology, landform,

0921-8181/91/$03.50 1991 - Elsevier Science Publishers B.V.


380 V. GO14NITZ

wave climate, longshore currents, and s t o r m fre- the International (ieosphere-Biosphere Pro-
quencies. Relative m a g n i t u d e s of these variables gram.
change from place to place, which induce a non-
uniform response of the coastline to SLR. Thus, Sea level trends--recent past and near-fu-
the coastal vulnerability will v a r y spatially. ture
Coastal vulnerability is the liability of the shore
to respond adversely to a hazard. A coastal T h e detection of sea level change is com-
hazard is a natural p h e n o m e n o n t h a t exposes plicated by the large n u m b e r of factors which
the littoral zone to risk of d a m a g e or other influence sea level. S h o r t - t e r m changes (100-150
adverse effects. An impact, however, is the nega- years) in sea level are obtained from tide-gauge
tive consequence{s) arising from the assumed records, which are geographically biased toward
SLR. Although sea level rise is emphasized in the northern hemisphere, and often consist of
this paper, other factors such as winds, waves, time series which are too s h o r t or too broken to
and s t o r m surges represent additional t y p e s of be useful. F u r t h e r m o r e , the records exhibit vari-
coastal hazard. Sea level rise is a global-scale, ability produced by tides, o c e a n - a t m o s p h e r e ef-
long-term hazard, which may, in the long run, fects such as the E1 N i h o / S o u t h e r n Oscillation
inflict even greater d a m a g e t h a n t h a t of a hur- (ENSO), and vertical land m o v e m e n t s (including
ricane. T h e intensity of " r a p i d - o n s e t " hazards glacial rebound, neotectonism, s e d i m e n t a t i o n
m a y also v a r y in response to global climate and compaction, and subsurface fluid
warming. 1 Because of the complexity of model- withdrawal). Thus, sea level changes m a y reflect
ing the response of these hazards to climate real changes in the fluid-volume of the oceans
change, how these will in t u r n alter coastal (the eustatic change), as well as changes in land
vulnerability lies outside the scope of this paper. elevation on local to regional scales, or shifts in
However, rising sea levels will only exacerbate ocean currents. For these reasons, some scien-
these effects. tists have questioned w h e t h e r an average of
Recent historical and projected sea level tide-gauge m e a s u r e m e n t s can represent t h e eu-
trends (_+ 100 years) are briefly reviewed in the static trend (Pirazzoli, 1986, 1989).
next section. Processes changing the coastal zone Nevertheless, m a n y studies of global m e a n
during a period of rapid sea level rise are dis- sea level changes (_< 100 years) yield rates be-
cussed in the succeeding section. E x a m p l e s of tween 0.5 and 3 m m / y r , with m o s t reported
recent investigations into potential adverse ef- values ranging between 1 2 r a m / y r . (For a com-
fects of S L R on coastal cities and ecosystems are prehensive review, see Pirazzoli, 1989; and a
presented. Finally, preliminary results from a status report, Warrick and Oerlemans, 1990, also
coastal hazards d a t a base are summarized for Fig. 1). T h e m o s t recent studies have a t t e m p t e d
the eastern U.S. Ultimately, this inventory of to filter out long-wavelength crustal m o v e m e n t s
erosion and inundation hazards will classify the by using '4C-dated paleosealevel indicators
relative vulnerability of the world's coastlines to (Gornitz and Lebedeff, 1987) or geophysical
SLR, in order to select high risk areas for more modeling (Peltier a n d T u s h i n g h a m , 1989).
detailed, higher resolution analysis, and to in- Douglas and H e r b r e c h t s m e i e r (1989) h a v e found
tegrate with other d a t a sets monitoring global a trend of 1.6 + 0.6 m m / y r , after screening out
environmental change, such as those planned for stations for wind forcing effects, as well as cor-
recting for glacial rebound.
In spite of noisy data, spatially and tempor-
ally coherent rises in sea level can be recognized,
1 H u r r i c a n e i n t e n s i t y m a y increase in a double-CO 2 world especially for b e t t e r - d o c u m e n t e d regions with
(E manuel, 1987). Areas subject to the strongest hurrica ne s large station populations. M o r e significantly,
(e.g. Bangladesh and t h e U.S. Gulf Coast) will also be
especially vulnerable to S L R for reasons discussed in t he
rates of sea level rise over the last 100 years, in
text. general, exceed those of the late Holocene (the
GLOBAL COASTAL HAZARDS FROM FUTURE SEA LEVEL RISE 381

last 5000 years), and especially those of the last previously estimated. T h e Antarctic, in contrast
2000-3000 years. T h e r e is also some suggestion to Greenland and most m o u n t a i n glaciers, is
of a modest increase in S L R (0.4 m m / y r ) in presently so cold and dry, t h a t increasing tem-
Northwest Europe, during the last 100 years, as peratures in the future will enhance snow accu-
compared with the previous 100 years (Wood- mulation over ablation, leading to a negative SL
worth, 1990). I t is very unlikely t h a t these dif- contribution (Oerlemans, 1989). These results are
ferences represent a recent change in the rate of based upon calculated changes in s a t u r a t e d
land subsidence or sediment compaction, except vapor pressure over ice, as well as correlations
in cases of localized fluid withdrawal (Gornitz, between observed t e m p e r a t u r e s and snow accu-
1990a). mulation (Meier, 1989). Finally, re-examination
Eustatic sea level rise over the next century of ice melting processes suggest the disintegra-
will be the sum of individual contributions from tion of WAIS will take several centuries, a t least
thermal expansion of sea water and ice melting (Budd, 1987).
from alpine glaciers, polar ice sheets (Greenland Summarizing these various processes, Meier
and Antarctica), and possible instability of the (1989) estimates t h a t by 2050, thermal expan-
West Antarctic Ice Sheet (WAIS). sion of the oceans will contribute 0.2 _+ 0.1 m,
T h e latest calculations suggest t h a t polar ice small glaciers 0.16 + 0.14 m, Greenland 0.08 +
sheets will contribute less to future S L R t h a n 0.12 m, the Antarctic - 0 . 3 _+ 0.2 m, and o t h e r

8o[ !

4.o .................................................... ~-i.... i


%

~g A ~

-8.0 '" ~: ' . . . .

, ; ~. .

-i2.o + .................................
; ........ .7...............................................
: ~ - 4.0

':,'~ .,! ' ~ ~'i,~ :" , ' ,.'

0 ~: ...... -; ,: ,, -8.0

[ -; ...... : . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

i : -12.0
1880 1900 1920 1940 1960 1980 1990
DATE
Fig. 1. Global mean sea level rise over the last century. T h e baseline is obtained by setting the average for the period 1951-1970
equal to zero. The dashed line represents the annual mean and the solid line represents the 5-year r u n n i n g mean. T h e top curve
( A ) is modified from Gornitz and Lebedeff (1987), and the b o t t o m curve ( B ) is from B a r n e t t (1988).
382 v ~;Ol:tNITZ

effects 0.2 _ 0.3 m or a total S L R of 0.34 0.42 the eastern coast of S u m a t r a , and large sections
m. Although Oerlemans (1989) projects a similar of coastal Borneo ( K a l i m a n t a n ) will face flood-
increase of 0.33 + 0.32 m by 2050, reaching 0.66 ing. Other highly p o p u l a t e d deltas and coastal
0.57 m b y 2100, in his model, the contribu- plains are at risk; however, m a n y of these areas
tions from glaciers and t h e r m a l expansion are (e.g. Amazon, Orinoco, Niger Deltas, etc.) are
comparable, whereas those from Antarctica and still relatively underpopulated.
Greenland largely cancel. Coastal wetlands will be a m o n g the m o s t
Given the great uncertainties due to the in- severely affected ecosystems, since these form
completeness of our knowledge of processes and largely in the intertidal zone. T h e response of a
observations, these projections can only be salt m a r s h to rising sea level depends on the
viewed as t e n t a t i v e scenarios of sea level rise, relative rates of submergence vs. vertical accre-
against which coastal hazards m u s t be assessed. tion or sedimentation. A m a r s h m a y m a i n t a i n
its areal extent or even grow in the face of SLR,
R e s p o n s e s of t h e s h o r e to a c c e l e r a t e d s e a if sedimentation rates at least m a t c h submer-
level rise gence rates. Even so, t h e m a r s h (or m a n g r o v e in
the tropics) will extend landward, as the upland
T h e following are the m a j o r processes t h a t edge is colonized by inland migrating m a r s h
will affect the coastal zone, during a period of vegetation. However, this landward translation
accelerated sea level rise: (1) inundation, b o t h of ecological zones might be hindered by an
p e r m a n e n t and episodic, (2) increased erosion, increase in slope upland, as well as by economic
and (3) salt w a t e r intrusion of estuaries and development of the interior.
aquifers. These are now briefly discussed in turn. In the U.S., present rates of m a r s h accretion
are generally keeping pace with present SLR,
Inundation except for p a r t s of Louisiana and portions of the
Chesapeake and I)elaware B a y s (NRC, 1987).
P e r m a n e n t inundation, .by an a m o u n t corre- Around 140 km 2 of Louisiana wetlands are inun-
sponding to the vertical increase in relative m e a n d a t e d annually (EPA, 1987). T h i s loss is aggra-
sea level, will affect an area which depends on vated by a high submergence r a t e ( - 10 m m / y r ) ,
the local gradient. Low gradient coastal land- which, in part, is due to the c u r t a i l m e n t of
forms m o s t susceptible to inundation include sediment supply from the Mississippi River, in
beach ridge and chenier plains, deltas, mudflats, recent decades, because of u p s t r e a m dams, chan-
estuaries, lagoons, and bays. In the U.S., this nels and levees. T h e construction of dikes along
includes m a r s h areas along the Atlantic Coastal the Mississippi River and canals in the wet-
Plains, m u c h of the Florida Everglades, and the lands, have accelerated s a l t w a t e r intrusion into
Gulf Coast. O t h e r high risk areas, elsewhere, a b a n d o n e d delta lobes, thereby destroying the
include the estuaries of eastern U.K., the vegetation cover and exposing the area to in-
E u r o p e a n Low Countries, the Southern Baltic, creased erosion (Day and T e m p l e t , 1989). T h e
and m a j o r river deltas, such as those of the Nile, high subsidence rate may be also exacerbated by
the m o u t h s of the Indus, G a n g e s - B r a h m a p u t r a oil and gas pumping.
and I r r a w a d d y Rivers, and the Chao P h r a y a Under extremely high rates of S L R (2.2 m by
River, in southern Asia. Within the next 100 the year 2100), 739~ of all U.S. wetlands existing
years, 26% of the land in Bangladesh and up to in 1975 would be inundated, based on a sample
21.5% of habitable land in the Nile Delta could of 57 coastal sites, occupying 485,000 hectares
be lost, under S L R scenarios t h a t include local ( A r m e n t a n o et al., 1988). T h i s loss could be
subsidence and d a m m i n g (Milliman et al., 1989). reduced to 56~, by the f o r m a t i o n of new wet-
T h e t h r e a t to B a n g k o k is acute; local subsidence lands further inland. Under a lower S L R scenario
due to groundwater p u m p i n g has reached up to (1.4 m), a 40% inundation could be reduced to
13 c m / y r in recent decades (ibid). In Indonesia, 22%, by inland migration of wetlands.
G L O B A L COASTAL H A Z A R D S F R O M F U T U R E SEA L E V E L R I S E 383
A.
In the tropics, clearing of coastal mangrove
swamps and forests, such as the 6000 km 2
Sundarban mangrove forest in southwestern

.,,,..;i?
E
Bangladesh, could accelerate erosion trends and
expose the interior to damaging storm surges. ,
o
This forest may be inundated by even a 1 m o
SLR. Another highly vulnerable ecosystem is
freshwater peat bogs, bordered by mangroves, Qa
which occur in many parts of Indonesia.
Many coral islands have an average elevation
of only 1.5-2 m above present sea level, and are
103 I0 z I0 I I0 "I I0 -~ I 0 ~ - ~ -7
therefore at risk to inundation. The ability of
Probability of exceedance per year
corals to keep up with SLR depends on the
relative rates of coral reef growth. If the rate of
S LR approaches 10-12 m m / y r , which is close to B.
the maximum calcification rates (Hopley and
Kinsey, 1988; Buddemeier and Smith, 1988), , I0-4 I

even fast-growing coral species may begin to lag


behind and become progressively submerged. Do-3i
Furthermore, coral growth may be retarded at
sea surface temperatures above 30C, which may
become prevalent in tropical oceans, in a high
I
CO 2 world. Already, within the past few years, "6
~ I 0-1
widespread severe coral "bleaching" episodes
5~
have occurred, which have been linked to the
general global warming trend of the 1980's 13._ I I I
0.5 1.0 1.5 2.0
(Bunkley-Williams and Williams, Jr., 1990). If
Mean sea level rise, m
sea-surface temperatures continue to rise in the
Fig. 2A. S t o r m surge w a t e r levels for Hoek v a n Holland, t h e
future, such bleaching events may become even
N e t h e r l a n d s (inset, t o p left). T h e s t o r m s u r g e for t h e 1953
more persistant, inhibiting coral growth, and s t o r m is s h o w n for comparison. A s t o r m s u r g e of t h i s magni-
tipping the balance in favor of submergence. t u d e ( - 4 m) occurs w i t h a frequency of a r o u n d once every
T he survival of coral islands will depend on 250 years, a t p r e s e n t sea levels. (after G o e m a n s , 1986). B.
C u r v e of m e a n sea level rise vs. probability of exceedance of
whether or not living coral growth rates can
s t o r m surge levels, for p r e s e n t dikes, central N e t h e r l a n d s
match SLR and sufficient coral rubble can accu- (after Goemans, 1986).
mulate to maintain protective storm ridges.
Potentially vulnerable islands include Indian
Ocean islands, such as the Maldives, and many heights. A great deal depends on fetch and aspect
Pacific and Caribbean islands; also Australia's (facing direction).
Great Barrier Reef. Coastal engineers and planners calculate
Episodic inundation results from storm standard recurrence intervals of storm tides
surges, which are anomalously high tides pro- above present mean sea level. A rise in mean sea
duced by a combination of low atmospheric level will result in a greater frequency of occur-
pressure and wind-driven waves, especially if rence of a storm surge at a given height. For
superimposed on astronomical high tides. Sea example, a 4 m surge is calculated to occur at
level varies inversely with atmospheric pressure Hoek van Holland, on average once in around
by 1 c m / m b . Variation in wind speed also alters 250 years (Fig. 2a). If sea level were to rise by 1
sea level, particularly in shallow water, due to m, a surge of only 3 m would be needed to reach
the effects of changes in wind stress on wave the 4 m water level. T he 3 m surge has a
384 v GORNI'FZ

frequency of occurrence of approximately once Bruun, 1962, 1983). Limitations of its applicabil-
in 50 years. Possible changes in tidal range, ity include fulfillment of the assumption of equi-
storm frequency, and storm tracks m a y require librium conditions, with no t r a n s p o r t of sedi-
revision of present surge recurrence curves. ment into and out of the study area, and diffi-
Presently, dikes in the Netherlands are desig- culty in defining the offshore limit of sediment
ned to withstand a storm surge level with a transport. Also unaccounted for are processes
probability of occurrence of 1/10,000 in the such as thinning of barrier islands (Leatherman,
central part of the country, and 1/4000 in the 1983), or washover and inlet sedimentation,
north and south. However, a S L R of + 1 m which could represent up to 62% of the total
would increase the probability of exceedance erosion (Fisher, 1980). Dean and M a u r m e y e r
from 1/10,000 to 1/250 (Goemans, 1986; Fig. (1983) have extended the Bruun Rule to barrier
2b). Therefore, Dutch law now provides for re- island systems t h a t retreat landward by filling
building the dikes, in response to future in- in on the bay side to maintain constant width,
creases in sea level. ms they erode on the ocean side. This model m a y
not apply to very high rates of SLR, as is
Erosion occurring presently in Louisiana ( - 10 m m / y r ) ,
where barrier islands have decreased in area by
Even at present rates of sea level rise, over 37%, between 1890 and 1979 (Sallenger et al.,
70% of the world's sandy beaches are retreating 1987).
(Bird, 1985). However, the worldwide erosion A three-dimensional sediment budget analy-
can be attributed to numerous factors in ad- sis (Everts, 1985) incorporates landward trans-
dition to S L R (Bird, 1988). Beach erosion is port of eroded sediment, and adjusts for particle
frequently intensified by anthropogenic inter- size. Kriebel and Dean (1985) have developed a
vention, such as by intercepting silt and sand by dynamic equilibrium model, in which the theo-
upstream reservoirs (e.g. the Aswan dam, Smith retical beach profile will evolve toward a new
and Abdel-Kader, 1988), interruption of littoral equilibrium shape, in the face of SLR, or in-
drift by breakwaters, or beach sand mining. creased wave heights. Sand t r a n s p o r t is assumed
T h e rate and extent of coastal erosion is ex- to maintain constant volume. M e h t a and Cush-
pected to intensify as a result of increased SLR. man (1989) further contrast kinematic and dy-
T h e " B r u u n Rule" is widely used to predict the namical modeling approaches.
shoreline response to SLR. It states t h a t a typi- In some cases, projection of historical shore-
cal concave-upward beach profile erodes sand line erosion with respect to local sea level changes
from the beachface and deposits it offshore, so may be the most feasible approach to predicting
as to maintain constant water d e p t h (Fig. 3; future trends. While less q u a n t i t a t i v e t h a n the

" ~:7~'. Initial shore profile

\~'~;"~~ ~" new meon seo level


Profile Odlusted to ~ ~;.~ ! ~=\ ~, " ~ h
seo,eve,, se,, \

'E' - _

( X ) ~ ?: - : ." ~

( D

Fig. 3. S c h e m a t i c i l l u s t r a t i o n of t h e s h o r e l i n e r e s p o n s e t o s e a level rise, a c c o r d i n g t o t h e '" B r u u n R u l e " . h = s e a level rise,


D = t o t a l s h o r e l i n e d i s p l a c e m e n t , I = c o m p o n e n t o f s h o r e l i n e d i s p l a c e m e n t d u e t o d i r e c t i n u n d a t i o n , E = c o m p o n e n t of
shoreline displacement to erosion.
GLOBAL COASTAL HAZARDS FROM FUTURE SEA LEVEL RISE 385

above-cited models, this method accounts for Relative sea level change
FALLING SEA LEVEL I RISING SEA LEVEL
the inherent variability of shoreline response, OR LAND UPLIFT I OR SUBSIDENCE
based on varying coastal geomorphology, beach RAPID SLOW STABLE SLOW RAPID
composition and exposure to waves and tides. = ~
,sos,o',c(_k
or tectonic
', J<-2oGoAo
X ~ ' ~ \ , x'~X.'~ .~ I . '
The technique has been applied to Galveston uplift ~X,.~ \', ~,=",,~ "5)~.1"~_7
Bay, Texas (Leatherman, 1984) and Ocean City,
Maryland (Leatherman, 1985).
Another approach to the assessment of future "6 ~ d~ I \",,7 'q~,x",X I u%
shoreline changes is to use the post-glacial
= ~ Fluvial ~X,~\ ~'~'~.~, u~u,0
marine transgression (-16,000-6000 yr B.P., [deposition ~ . ~ \ ~,~N~BP
Fairbanks, 1989) as an analog, because of simi-
Fig. 4. Sketch illustrating t h e relationship between relative
larities in rates of SLR. Thom and Roy (1988) rates of sea level change (either rise or fall) and sediment
find that coastal evolution, among other factors, supply to t h e beach (erosion or deposition), which will de-
is also closely linked to the gradient of the termine t h e prevalence of marine transgression or regression
equilibrium shoreline profile. They identify two (modified from Orford, 1987).

basic responses to the post-glacial marine trans-


gression in SE Australia. The first, on very low- in place (for example, the present Louisiana
gradient coasts, is a landward migration of sand coast). However, drowning may also occur where
barriers during SLR ("roUover"), followed by sediment influx retards barrier migration, or
dune accretion, as SL stabilizes. The second, where the interaction of various coastal processes
more prevalent on higher gradient coasts, is results in a dynamic equilibrium of the barrier
transport of sand offshore, in accordance with position (Carter, 1988).
the "Bruun Rule".
Sediment supply and breaker-wave height Saltwater intrusion
also modify the adjustment of the shoreline to
the post-glacial marine transgression (Orford, Another inpact of rising sea level is that of
1987; Short, 1988). At high rates of SLR, but increasing saltwater Intrusion into rivers and
low sediment supply, other factors remaining estuaries, and also infiltration into coastal
constant, the shoreface will erode and shift land- aquifers. The upstream penetration of saltwater
ward, as for example, the barriers along the will be similar to that which occurs at present
Delaware coast, during the Holocene (Kraft et during extreme droughts, when river runoff is
al., 1987). At reduced rates of SLR, with suffi- diminished. Higher estuarine salt levels would
cient sediment supply, seaward progradation can also contaminate urban surface water supplies.
occur (Fig. 4; after Orford, 1987). Examples of For example, a S L R of 0.73 m cause the salinity
prograded beach ridges that formed after the level of the Delaware River, at the intake sta-
deceleration of the post-glacial transgression, tion above Philadelphia, to exceed New Jersey's
creating a "stillstand" ( < 6000 yr B.P.), include drinking water standard (50 mg/1 Na + ) during
the German North Sea Coast (Streif, 1989); the 15% of the tidal cycles, while a SLR of 2.5 m
Netherlands (Jelgersma, 1979); and eastern would cause this value to be exceeded over 50%
Australia (Short, 1988; Chappell, 1987). How- of the cycles (Hull and Titus, 1986).
ever, within the last few thousand to hundred In some major tropical deltas with seasonally
years, seaward deposition has given way to ero- fluctuating river discharge (particularly in SE
sion in numerous localities (Bird, 1985; Short, Asia), fresh and saline conditions alternate
1988; Tanner, 1988). In places, the offshore sand within a broad zone. Reduced runoff in the dry
source has now been largely depleted, and may season causes tidal penetration of saline water,
not be available to replenish beaches in the near forcing a fallow period in agriculture. Fresh-
future. Where the rate of SLR greatly exceeds water flooding during the wet season desalinizes
that of sedimentation, barrier islands may drown the zone, which can then be used for a single
386 v GORNITZ

wetland rice crop (Brinkman, 1984). As sea level hence increase seepage, requiring additional
rises the tidal saltwater zone will penetrate fur- flushing of the polders with freshwater. An anal-
ther upstream. This zone therefore becomes un- ogous situation can be anticipated for the Stock-
fit for tidal swamp rice cultivation, over a longer ton-Sacramento agricultural district, California,
period of the annual cycle. where much of the land is also at or below
In humid, equatorial climates, gradual SLR present sea level (Smith and Tirpak, 1988).
would cause the brackish-water zone, occupied
chiefly by mangrove, to migrate inland. Peaty Coastal hazards and impact studies
soils, formed under mangroves, tend to accu-
mulate pyrite, which will be oxidized and acidify
Studies of coastal hazards, or vulnerability,
the soil, once the mangroves are cleared for
delineate risk factors and estimate their relative
cultivation. The encroachment of the brackish
magnitudes, extent, and likelihood of occur-
zone, with its potentially acid soil, will come at
rence. A hazard analysis provides a baseline of
the expense of land currently suitable for eco-
present (and historical) data from which future
nomically important crops such as wetland rice,
trends can be extrapolated. However, future
sugarcane or rubber (Brammer and Brinkman,
trends may differ substantially from those of
1990).
the recent past. From a slightly different per-
Sea level rise will also promote saltwater in-
spective, impact studies examine physical
trusion into coastal aquifers. Along barrier
a n d / o r socio-economic consequences of an as-
coasts, volcanic, and coral islands, a freshwater
sumed SLR scenario, based on temperature in-
lens overlies saltwater. According to the
creases from general circulation models (GCM's;
Ghyben-Herzberg-Dupuit model, the freshwater
e.g. Hansen et al., 1988), and assumptions of
lens is forty times thicker than the elevation of
ocean heat diffusion, and increases in polar ice-
the water table above mean sea level. Thus, each
sheet melting. Some examples of studies using
increment of SLR will reduce the freshwater
both approaches, on regional to international
capacity by 40 times. On many low coral atolls,
scales are-presented in the following pages.
less permeable semi-lithified Holocene sediments
typically overlie a highly permeable Pleistocene
karstic subsurface, through which seawater can Coastal hazard studies
infiltrate. A much broader transition zone of
mixed fresh and saline water underlies the Existing coastal hazards have been mapped
freshwater lens, thereby reducing the potential for the U.S. (Kimball et al., 1985), including
freshwater storage capacity of the aquifer from dangers from storms, waves, erosion, also lands-
that suggested by the above-cited model (Bud- lides and earthquakes. Professor R. Dolan (Uni-
demeier and Oberdorfer, 1989). Any coastal ero- versity of Virginia) is currently preparing a'more
sion accompanying future SLR would further detailed hazards evaluation of both the East
reduce the freshwater storage, with serious con- and West Coasts (pers. comm., 1990).
sequences for the water supplies of small islands The Geological Survey of Canada is imple-
and coastal dune areas. Excessive freshwater menting a Coastal Information System (CIS),
pumping has already provoked an upward which will provide a national coastal inventory
migration of the saltwater-freshwater interface of geology, geomorphotogy, offshore characteris-
in many coastal localities. tics, and wave, tide and surge records (Fricker
A large part of Holland presently lies below and Forbes, 1988).
mean SL, so that saltwater seeps upward into Coastal hazard zones in New Zealand are
the subsoil, via groundwater flow. This problem being mapped in order to mitigate exposure to
already affects agriculture, especially during dry natural hazards by appropriate land use plan-
summers (De Ronde, 1989). Increasing SL will ning (Gibb, 1984). Factors considered include
steepen the saltwater-freshwater gradient, and historical shoreline fluctuations, lithology, beach
GLOBAL COASTAL HAZARDS FROM FUTURE SEA LEVEL RISE 387

sediment supply, landslides, neotectonic activity SLR scenarios of 2.6-3.3 m (including local sub-
and storm frequency and flooding potential. sidence, damming), by the year 2100, between
The CORINE project (Quelennec, 1989) 21.5-26% of the habitable land in the Nile delta
evaluates coastal erosion risks in 11 European would be lost, and 19-24% of the population
Community countries. It contains information displaced, with a comparable loss in the Gross
on the morphosedimentologic characteristics of Domestic Product. In Bangladesh, due to higher
the coastal zone, shoreline displacement trends, rates of local subsidence, SLR could reach 3.3-
and presence on absence of coastal defense works. 4.5 m by 2100, drowning 26-34% of the land
Beach erosion occurs in all countries, especially area, displacing 27-35% of the population, and
in Portugal, France, Greece, Belgium and the causing a loss of 22-31% of the GDP, in the
Netherlands, where as much as 40-50% of the affected area (Milliman et al., 1989).
sandy beaches are retreating. A preliminary impact study for the Nether-
Dr. S. Jelgersma, Geological Survey of the lands (Jelgersma et al., 1987) points out that
Netherlands, is preparing a computerized data 57% of the population today lives below present
base for the UN FAO, which will include infor- mean sea level. Industrial and agricultural activ-
mation on geomorphology, coastal sediment ities are concentrated on land below 5 m eleva-
transport, vertical movements, vegetation, land tion. Even a modest SLR of 0.5 m will upset the
use, suspended river sediment a_~d discharge, intricate water management system in the area
tidal ranges, frequency and severity of cyclones behind the coastal dunes and sea dikes, unless
of hurricanes (S. Jelgersma, pers. comm., 1989). steps are taken to raise and adapt the surround-
The results will be represented on a 1 : 5,000,000 ing dikes. Goemans (1986) has estimated the
scale map. costs of raising the dikes to protect against a 1
m SLR at $4.4 billion, and $8.8 billion for a 2 m
SLR.
Impact studies
The Netherlands in cooperation with U N E P
has prepared a report on the "Impact of Sea
A U.S. EPA study examines regional and Level Rise on Society" (ISOS; U N E P and Delft
national impacts (Smith and Tirpak, 1988). The Hydraulics, 1989). Data for 10 priority areas i
study projects a loss of 13,200-26,700 km 2 of have been compiled on cyclone and hurricane
drylands without shoreline protection, assuming flood damage, soil types, sensitive ecosystems,
1 m SLR by 2100. Under these conditions, 26- such as wetlands, mangroves, and coral reefs,
66% of the wetlands would be inundated. Shore- and population density.
line protection measures for densely developed A joint United Nations U N E P - U N E S C O In-
areas could cost $73-111 billion. Louisiana, the tergovernmental Oceanographic Commission
Gulf Coast, and the mid- to south Atlantic coasts (IOC) program is investigating the implications
will be most severely affected. of climate change in five regions (Caribbean,
The A R I S (Australian Resources Inventory central-east Atlantic, western Pacific,southwest
System) data base contains information on geol- Atlantic and the central Indian Ocean). A n im-
ogy, landforrns, vegetation and land use for each portant component of this undertaking includes
of 3027 10 by 3 k m sections (Galloway et al., comprehensive reviews of coastal changes, sea-
1984). Priority areas (Brisbane, Sydney and the level variations and impacts on coastal ecosys-
N S W North Coast) have been selected for de- tems. (C. Latouche, IOC consultant, Paris; M.
tailed S L R impact studies, based on coastal Hendry, 1989, U. West Indies, pers. comm).
population densities, and risk from inundation,
erosion and storm flooding (Cocks et al., 1988).
Milliman et. al. (1989) have analyzed impacts
1 T h e s e a r e a s are Bangladesh, Egypt, Indonesia, Maldives,
of rising sea level on the Nile and Ganges- Mozambique, Pakistan, Senegal, Surinam, Thailand and
Brahmaputra deltas. They conclude t h a t under Gambia.
388 V. GORN ITZ

The coastal h a z a r d s data base sediment transport, although also important


factors, have not been included at the present
time. However, as the data are incorporated into
A coastal hazards data base is being devel- a Geographic Information System (GIS), these
oped to provide a global overview of the relative layers ca_n be added as information becomes
vulnerabilities of the world's coastlines to inun- available. The GIS approach also allows even-
dation and erosion hazards associated with tual integration with other climatological and
accelerated SLR (Gornitz and Kanciruk, 1989). socioeconomic data sets.
The data base integrates information on seven Here, we briefly discuss the methodology of
variables, including: (1) relief (elevation), (2) constructing the data base, and its application
lithology (rock type), (3) coastal landforms geo- to the East Coast, U.S.A., as a test case. We
morphology, (4) vertical land movements (rela- then evaluate the differential vulnerability of
tive sea level changes), (5) horizontal shoreline the East Coast, in terms of the individual varia-
changes (erosion or accretion), (6) tidal ranges, bles and the combination of these into a Coastal
and (7) wave heights. Although not specifically Vulnerability Index. Procedures are still under
dealt with here, data on storm frequency and development, and the outline presented here
intensity have been collected by others at provides a demonstration of the approach rather
ORNL, in a related study. Storm surges and than a final ~ssessment.

TABLE 1
C o a s t a l risk classes

Variable Rank
Very low Low Moderate High Very high risk
1 2 3 4 5
Relief (m) >_ 30.1 20.1-30.0 10.1-20.0 5.1-10.0 0-5.0

Rock t y p e Phitonic I ~ w - g r a d e met a mor. Mos t s e d i m e n t a r y Coarse a n d / o r F i ne uncon-


(relative Volcanic (lava) S a n d s t o n e and rocks poorly-sorted s ol i da t e d
resistance H i g h - m e d i u m grade conglomerate unconsolidated sediment
to erosion) metamorphics (well-cemented) sediments Volcanic a s h

Lan dfo rm Rocky, cliffed M e d i u m cliffs Low cliffs Beaches (pebbles) B a r r i e r beaches
Coasts I n d e n t e d coasts Glacial dri ft Estuary. B e a c he s (sand)
Fiords Salt marsh Lagoon Mudflats
Fiards Coral Reefs Alluvial pl a i ns Deltas
Mangrove
Vertical
movement < - 1.1 - 1.0-0.99 1.0-2.0 2.1-4.0 _> 4.1
( R S L change)
(mm/yr) L and rising ( w i t h i n ra nge of - - Land sinking )
e u s t a t i c rise

Shoreline >_ 2.1 1.0-2.0 - 1.0- + 1.0 - 1.1 .... 2.0 _< - 2 . 0
displacement
(m/yr) Accretion ( Stable ~ Eros i on

T i d a l range _< 0.99 1.0-1.9 2.0-4.0 4.1-6.0 _> 6.1


m (mean)
Microtidal < Mesotidal ~ Macrotidal

Wave height, 0-2.9 3.0-4.9 5.0-5.9 6.0-6.9 _> 7.0


m (max.)
GLOBAL COASTAL HAZARDS FROM F U T U R E SEA LEVEL R I S E 389

Data base components and risk classes Kanciruk (1989). Because these factors are dif-
ficult to quantify, they are ranked into classes of
For the purposes of this paper, a vulnerable increasing risk (Table 1).
coastline is characterized by low coastal relief, Digitized historical U.S. shoreline changes,
an erodible substrate (e.g. sand, unconsolidated averaged into 3' cells, come from the CEIS data
sediment), present and past evidence of subsi- base (Dolan et al., 1983). Rates within _+1 m lie
dence, extensive shoreline retreat, and high within the measurement error and are consid-
wave/tide energies., These attributes serve as ered at relatively low risk. Shores with rates of
guidelines for the ranking scheme outlined in - 1 m / y r or less (more negative) are eroding,
the following paragraphs. and at relatively higher risk (Table 1). Con-
Among the variables considered here, relief versely, shores with rates > +1 m / y r are
and vertical land movements (particularly subsi- accreting, and at correspondingly low risk.
dence), are primarily indicators of inundation Waves and tidal currents actively transform
risk. As a simple means of determining relief, the the shoreline. Wave heights bare proportional to
average elevation of 5 latitude-longitude land the square root of wave energy, which is a mea-
data points (from ETOP05 Gridded World sure of the capacity for erosion. U.S. wave data
Elevations, National Geophysical Data Center, come from the Wave Information Study (WIS),
Boulder, CO) aggregated into 1/4 coastal cells, U.S. Army Corps of Engineers (CERC), for 166
represents a first-order approximation of the nearshore segments at roughly 10 mi (16 krn)
areal extent of inundation, suitable for a global spacing along the East Coast (Jensen, 1983).
scale. This data set, although not without prob- (The ranks assigned in Table 1 are based on
lems, and at relatively coarse resolution, never- maximum significant wave heights).
theless represents the most complete global The tidal range is linked to both inundation
coverage currently available. While the eleva- and erosion hazards. Although a large tidal range
tion zone within 1 m faces the highest probabil- dissipates wave energy, limiting beach or cliff
ity of permanent inundation, the coastal strip erosion to a brief period of high tide (Bird,
within 5 m of present SL is also at high risk to 1985), it also delineates a broad zone of inter-
above normal tides from severe storm surges. tidal wetlands, which will be most susceptible to
The hazard decreases progressively for higher inundation following SLR. Furthermore, the
average elevations (Table 1). velocity of tidal currents in estuaries depends on
Vertical land movements are obtained from the tidal range, as well as the asymmetry of the
relative sea level trends, from a worldwide net- tidal cycle and channel morphology (Pethick,
work of - 1000 tide-gauge stations (Pugh et al., 1984). Therefore, holding these other factors
1987). The eastern U.S. is covered by 33 stations constant, high tidal range is associated with
(Lyles et al., 1987). The relative sea level (RSL) stronger tidal currents, capable of eroding and
change at each locality includes a eustatic com- transporting sediment. Therefore, macrotidal
ponent (1-2 m m / y r ) , as well as glacioisostatic, coasts (> 4 m) will be more vulnerable than
neotectonic and local subsidence components. those with lesser ranges (Table 1). Tide range
Subsiding areas, or those with RSL in excess of data are listed in the annual Tide Tables (NOS,
the eustatic range (> 2 m m / y r ) , regardless of 1988).
ultimate cause, face greater inundation hazards
(Table 1). Coastal Vulnerability Index
The other variables of the data base are asso-
ciated with erodibility risk. Bedrock lithology, Because the data base comprises qualitative,
shore materials, and coastal landforms vary sub- as well as quantitative information, at different
stantially in their resistance to erosion. A gener- scales and units, each variable for each coastal
alized scale of lithologic and geomorphologic re- segment has been assigned a rank from 1 to 5,
sistance to erosion is discussed in Gornitz and with 5 representing the most vulnerable class
390 V. GORN l"I'Z

T A B L E 2. Relative shorelength of landform a n d rock t y p e s for t h e east coast a n d subregions, USA

Landform Percent Beaches Marshes


East Coast
Rocky, glaciated coast 12.3 0.5 0.2
Estuaries 41.9 0.8 l 5.5
Coastal plain beach 0.8 0.5
Barrier coast 18.2 15.0 3.1
Lagoonal coast 15.3 0.2 9.5
Glacial deposit 6.0 2.1 0.6
Reef 1.4 1.4
Mangrove 3.5
Other 0.6
100.0 18.6 30.3
Rock type Percent
Resistant, crystalline rocks 13.2
S e d i m e n t a r y rocks 10.2
Sand 31.3
O t h e r unconsolidated s e d i m e n t s 45.3
100.0
Landform Percent Beaches
New England
Rocky, glaciated coast
(incl. cliff) 63.4 2.9
Estuaries 10.4
Coastal plain beach 1.2 1.2
Barrier coast 8.6 8.5
Glacial deposits 15.2 8.3
Other 1.2
100.0 20.9%
Rock t y p e Percent
Resistant, crystalline rocks 66.8
S e d i m e n t a r y rocks 9.6
Unconsolidated s e d i m e n t s 23.6
100.0
Landform Percent Beaches Marshes
Mid-Atlantic states
Estuaries 64.0 3.0 11.0
Coastal plain 0.8 \
Barrier coast 13.2 ] 9.3 3.1
Lagoonal coast 12.3 11.0
Glacial deposits 9.2 2.0 1.2
Other 0.6
100.1 14.3% 26.3%
Rock type Percent
Resistant, crystalline rocks 0.7
S e d i m e n t a r y rocks 0.2
Sand 52.1
M u d a n d silt 0.8
Glacial till 0.7
Calcareous s e d i m e n t 2.6
Mixed or undifferentiated 42.9
100.0
G L O B A L C O A S T A L H A Z A R D S F R O M F U T U R E SEA L E V E L R I S E 391

TABLE 2 (continued)

Landform Percent Beaches Marshes


Southeast A tlantic
Estuaries 39.2 0.1 26.4
Coastal plain beach 0.8 0.8 -
Barrier coast 26.1 20.4 4.3
Lagoonal coast 22.3 0.4 11.3
Reef 3.2 - 3.2
Mangrove 7.7 - -
Other 0.8 - -
100.1 21.7% 45.2%
Rock type Percent
Sedimentary rocks 18.1
Sand 22.9
Mud, silt 4.6
Calcareous sediment 11.8
Mixed lithology 6.9
Undifferentiated sediment 35.3
Other 0.4
100.0

(highest risk; Table 1). These individual risk D a t a entry into the Geographic In[ormation
classes can then be combined into a Coastal S y s t e m (GIS)
Vulnerability Index, C V I which can be com-
p u t e d as either the sum or p r o d u c t of the varia- T h e A R C / I N F O G I S ( E S R I , Inc.) software
bles. T h e p r o d u c t has the a d v a n t a g e of expan- at O R N L can relate and m a n i p u l a t e d a t a in
ding the range of values. On the o t h e r hand, it various formats and spatial resolutions, such as
m a y be quite sensitive to small changes in indi- (1) point d a t a (e.g. tide-gauge stations), (2) line
vidual ranking factors. Therefore, it m a y be nec- or arc d a t a (lithology, landforms, waves), (3)
essary to introduce a factor to d a m p e n the ex- polygons or cells (relief, shoreline displacements;
treme range. For the purposes of demonstration Gornitz and Kanciruk, 1989). E a c h variable
in this paper, the C V I is taken as the square forms a feature class (coverage), which can be
root of the geometric mean, or the square root of displayed graphically. Individual feature classes
the p r o d u c t of the ranking factors, divided by can be superposed, and areas with a c o m m o n set
the n u m b e r of variables present. of attributes can be identified.

CVI = n (al x a2 x ... an Results

where a i - - v a r i a b l e and n = total n u m b e r of Estuaries represent the d o m i n a n t landform


variables present. along the East Coast (41.9% by length), followed
T h e total range of CVI was divided into four by barrier coasts (18.2%), and lagoons (15.3%).
equal parts, and t h e upper quarter, or C V I > 33.0 Rocky, glaciated coasts occupy 12.3% of the
was taken as " v e r y high risk coastline." Based shore, while glacial deposits form 6.0% (Table 2).
on the shorelength frequency distribution, this Around three-quarters of the E a s t Coast is un-
corresponds to the 96 percentile (4% of the E a t derlain by unconsolidated sediments, the bal-
Coast, including bays, lagoons and estuaries, has ance divided between crystalline ( i g n e o u s /
a C V I score of 33.0 or greater). metamorphic) rocks (13.2%) and s e d i m e n t a r y
392 v. GORNITZ

rocks (10.2%), predominantly in New England tively. The East Coast can be divided for con-
(Table 2). Elevations range from a high of 100 m venience into three regions, t hat also differ in
to near sea-level along barrier coasts. Around geologic and terrain characteristics, described in
two-thirds of the East Coast is relatively stable more detail below.
(shoreline displacement within _+1 m / y r ) , with
25.2% eroding and 7.7% accreting (based upon New England (Maine through Connecticut)
the length of coast for which data are available,
Table 3). The East Coast is subsiding. Rates of The New England coastline consists of 63.4%
sea level rise exceeding 2 m m / y r affect 89.0% of strongly to weakly embayed rocky, glaciated
the region (Table 3). Values of CVI for the East shores. While beaches constitute 20.9% of the
Coast range between 1.79 and 46.29. The median total New England shoreline, 39.7% of these
value (by shorelength) is 10.12, while the upper occur on unconsolidated glacial deposits, 13.9%
and lower quartiles are 15.12 and 6.87, respec- are pocket beaches, and 46.4% are barrier

'FABL E 3
S u m m a r y of relative proportions coastal risk classes for t h e east coast and subregions, USA

Variable Risk classes (percent shorelength)


0 * 1 2 3 4 5

East cocaut
Relief 6.9 3.5 8.9 6.9 1:~.I 60.6
Ro ck t y p e 0.2 7.4 5.8 10.2 ;36.;] 40. l
Landform 0.4 11.0 t .5 36.6 32.8 17.7
Vertical m o v e m e n t 1.4 0 0.1 9.6 88.5 0.5
Shoreline d i s p l a c e m e n t 49.3 2.7 1.2 34.0 6.4 6.4
T i d a l range 0 45.7 33.4 20.1 0.9 0
Wave height 47.7 1.5 40.6 10.1 () 0

N e w England
Relief 18.8 13.3 22.5 5.5 8.6 ;31.3
Rock t y p e 1.1 36.8 30.0 9.6 9.9 12.5
Landform 0 56.5 6.9 7.6 16.'~ 12.7
Vertical m o v e m e n t 7.3 0 0.6 12.5 79.5 0
Shoreline d i s p l a c e m e n t 1.4 0.5 0.6 ,90.7 4.6 2.2
T i d a l range 0 0 29.8 65.9 4.3 0
W a v e height 10.4 0 67.5 22. i i) 0

Mtd- A tlantic states


Relief 3.8 2.7 13.7 15.6 14.3 49.9
Rock t y p e 0 0.1 0.6 0.2 43.6 55.5
Landform 0.1 0 0.4 31,4 53.9 14.2
Vertical m o v e m e n t 0 0 0 5.1 93.5 1.4
Shoreline d i s p l a c e m e n t 53.5 1.7 (1.4 24.9 ~.1.5 8.0
Tidal range 0 59.6 38.0 2A l) (1
W a v e height 63.0 0 29.7 7.0 (i 0

Southeast atlantic
Relief 3.8 0 (~ 0.2 14.5 81.6
Rock t y p e 0.1 0 0 18.1 42.5 39.']
Landform 0.8 0 0 52.8 22.9 23.6
Vertical m o v e m e n t 0 0 0 10.9 89,1 (I
Shoreline d i s p l a c e m e n t 67.8 4.1 1.9 16.4 2,7 7.0
T i d a l range 0 53.8 32.2 14.(1 t~ 0
Wave height 52.4 3.4 36. [ 8.1 (t 0

* No data/unclas.u.
G L O B A L C O A S T A L H A Z A R D S F R O M F U T U R E SEA L E V E L R I S E 393

beaches. Geologically, 66.8% of the New England less, and nearly half is 5 m or less. T he average
shore is underlain largely by Paleozoic and Pre- regional elevation is 8.9 m, ranging from a high
cambrian crystalline rocks (igneous, metamor- of 70 m to near sea level. Because of the preva-
phic), 9.6% is on sedimentary rocks, and 23.6% lence of sandy sediments and relatively mobile
on unconsolidated sediments (Table 2). landforms throughout the region, nearly the en-
In this region, 39.9% of the shore lies at an tire coast is stable or eroding (Table 3). How-
average altitude of 10 m or less (Table 3). The ever, erosion rates are extremely variable, par-
average regional elevation is 13.2 m, ranging ticularly near tidal inlets. Severe erosion ( < - 2
from a high of 100 m, to near sea level. Only m / y r ) occurs on the barrier beaches of Jones
6.8% of the coastline is eroding at rates of - 1 Island, NY, central New Jersey and especially
m / y r or greater, 2.2% more than - 2 m / y r (Ta- the Atlantic shore of Maryland-Virginia, where
ble 3). several islands are retreating at rates exceeding
Tidal ranges are predominantly high micro- - 1 0 m / y r . On the other hand, some of the
tidal to mesotidal (Table 3). There is a progres- inlets and spits are accreting rapidly, adjacent
sive decrease southward from northern Maine ( to severely eroding beaches.
6.1 m), to Connecticut (1.0-1.9 m), increasing Tidal ranges decrease southward from New
again toward the western end of Long Island York toward Chesapeake Bay. Regional maxi-
Sound (2.0-4.0 m). Maximum wave heights in mum wave heights on the open coast are mod-
New England are only moderately high (Table erate, ranging between 3.8 and 5.2 m.
2), with regional maxima on Martha's Vineyard The mid-Atlantic region is marked by above
(5.1 m) and Nantucket (5.2 m), off Cape Cod. average RSL rise (> 3 m m / y r ) , which roughly
T he regional RSL, ranges from 1.8 m m / y r to 2.7 coincides with a zone of maximum peripheral
ram/yr. bulge collapse south of the edge of the former
Wisconsin ice sheet.
Mid-Atlantic Coast (New York to Virginia)
Southeast Atlantic (North Carolina to Florida)
The East Coast, south of New England, lies
on poorly consolidated to unconsolidated Meso- In the southeast Atlantic region, 81.8% of the
zoic to Holocene Coastal Plains sediments. Long shoreline lies on unconsolidated sediments; the
Island, like Cape Cod, is formed largely of gla- remainder is chiefly limestone (Florida). Barriers
cial moraine and outwash deposits. constitute 26.1% of the total regional shore-
Th e region is dominated by two major length, lagoons 22.4% and estuaries another
estuaries: Delaware Bay and Chesapeake Bay, 39.2%, with the remainder mangroves and reefs.
both of which are river valleys submerged by the Marshes occupy 45.2% of the coastline, of which
post-glacial marine transgression. The estuarine 58.4% are located in estuaries and 25.0% along
environment occupies 64% of the shorelength, lagoonal coasts, the remainder on reefs or back-
with barriers and lagoons comprising only 26.3%. barriers. Nearly all beaches in this region occur
Unconsolidated sediments form 99.1% of the on barrier islands.
shore. Beaches occupy 14.3% of the total shore The southeast is the lowest region on the
length. Of the beaches, 65.0% are located on East Coast (av. elevation 2.13 m, ranging from
barriers or coastal plains, 21% are along estuaries 11 m to near sea level. Around 96.1% of the
and 14.0% on glacial debris (e.g., Long Island). shoreline is 10 m or less; 81.6% is 5 m or less
Marshes constitute 26.3% of the mid-Atlantic (Table 3).
coast, 41.8% of which occur along estuaries, 41.8% Around 51.1% of the mapped coastline is sta-
are associated with lagoons and only 11.8% are ble (within + 1 m), while 30.2% is eroding, and
on barriers (Table 2). 18.7% is accreting. Although severe erosion oc-
Elevations, in general, are lower than in New curs on numerous barrier beaches, north of Cape
England. Here, 64.2% of the coastline is 10 m or Canaveral, Florida, to the south, erosion rates
394 V. GORNITZ

0 2 O KM

~O
Fig. 6. Cape Cod area. Distribution of CV1 values greater
than or equal to 33.0 (heavy line).

Fig. 5. The East Coast, U.S.A. Distribution of CVI values


greater than or equal to 33.0 (heavy line). Boxed areas are
shown in greater detail in Figs. 6-9.

are g e n e r a l l y fairly low. As i n o t h e r regions,


s h o r e l i n e d i s p l a c e m e n t t r e n d s are s p a t i a l l y
h i g h l y variable. T i d a l r a n g e s grade f r o m micro-
t i d a l ( < 2 m) c o n d i t i o n s in N o r t h C a r o l i n a to
m e s o t i d a l ( 2 - 4 m) c o n d i t i o n s a l o n g t h e G e o r g i a
coast, a n d b a c k to m i c r o t i d a l i n Florida. Sea- \
w a r d shoals, off t i d a l i n l e t s r e f r a c t i n c o m i n g Fig. 7. Mid-Atlantic region. Distribution of CVI values
waves, c a u s i n g r a p i d erosion o n o n e s e c t i o n of greater than or equal to 33.0 (heavy line).
GLOBAL COASTAL HAZARDS FROM F U T U R E SEA LEVEL R I S E 395

beach, while nearby, wave energy is reduced,


and sand buildup can occur, particularly near
ebb-tidal deltas (Kana, 1989). ~L

Maximum wave heights throughout the re-


gion range between 2.4 and 5.9 m. Regional
highs occur north of Cape Hatteras (5.9 m), and
north of Cape Canaveral (5.1-5.2 m). Along the
southeast Atlantic coast, RSL trends range be- t00 Ku

tween 1.8-3.4 m m / y r .
Figures 5-9 show the distribution of shore-
lines with C V I scores of 33.0 or greater for the
East Coast and four selected areas: Cape COd,
the mid-Atlantic region, Cape Hatteras and
Southern Florida. These composite very high
risk areas occur on the Atlantic side of southern
Maryland-Virginia, the northern Cape Hatteras,
and also parts of the Georgia-South Carolina,

Fig. 9. Southern Florida. CVI values greater than or equal to


33.0.

New Jersey coasts, Cape Canaveral, Florida, and


parts of Cape Cod (Gornitz, 1990b).
Summary and conclusions
Global climate warming in the next century
will produce an estimated sea level rise of 0.3 to
0.9 m. Such an increase would cause inundation,
both permanent and episodic, of low-lying areas,
increased erosion, saltwater intrusion into
coastal aquifers and estuaries and damage to
man-made facilities and structures. T h e vulnera-
bility of the coast to SL R will be spatially
non-uniform because of variations in coastal
topography, rock hardness, vertical crustal
movements, wave climate and tidal regime. Low
:/ gradient coasts underlain by unconsolidated
/ sediments, such as deltas, mudflats, barrier is-
lands and estuaries, will be most susceptible to
permanent inundation and erosion. High subsi-
dence rates, particularly in deltaic areas, also
Fig. 8. Cape Hatteras-Myrtle Beach. Distribution of CVI frequently exacerbated by fluid-withdrawal at
values greater than or equal to 33.0 (heavy line). coastal cities, increase the inundation risk. Tidal
396 v. G,ORNtTZ

marshes, mangroves and coral reefs are ecosys- the relative risk factor. These risk factors are
tems most threatened by SLR. Their survival then combined into an overall coastal vulnera-
will depend on the ability of sedimentation or bility index, CVI, here taken as the square root
growth rate of plants and corals to compensate of the geometric mean of the risk classes. B y the
for submergence. T h e 1 m elevation zone will be criteria of coastal vulnerability as defined in this
most directly affected by inundation, but a zone study, the sections of coasts with the highest
of up to 5-10 m elevation could still experience CVI ratings include the Atlantic coast of Mary-
some flooding due to storm surges from extreme land-Virginia, northern Cape Hatteras, parts of
events (hurricanes, typhoons). Defensive struc- New Jersey, Georgia and S o u t h Carolina. South-
tures, such as dikes and sea walls, will have to be ern Florida in general does not rank as high.
erected in many ports and coastal cities. T h e Although it is very vulnerable to inundation due
D u t c h experience can serve as a model (Goe- to low elevation, and to erosion due to high risk
mans, 1986). rock types and landforms, nevertheless historic
T h e rate and extent of coastal erosion is ex- rates of erosion, wave energies and tide ranges
pected to intensify as a result of increased SLR. are considerably lower t h a n in the abovemen-
However, erosion trends are not easily predicted, tioned areas. On the other hand, incorporation
because of the interplay between numerous fac- of storm frequencies, intensities and surges, as
tors, including the sediment budget and oceano- well as population densities, as additional risk
graphic-climatic variables. Analogs for future factors, could very well place south Florida in
S L R include the Holocene marine transgression the highest risk category. Extension of this d a t a
(Jelgersma, 1979; K r a f t et al., 1987; Short, 1988; base to other regions is in progress.
T h o m and Roy, 1988) and presently rapidly
subsiding areas, such as Louisiana (Sallenger et Acknowledgements
al., 1987). Other approaches include modeling
the response of beach profiles to SLR, based on This research was sponsored by the U.S. De-
historic trends (Leatherman, 1984; 1985), sedi- p a r t m e n t of Energy, Office of Energy Research,
ment budget analysis (Everts, 1985), and dy- u n d e r c o n t r a c t DE-AC05-840R21400 w i t h
namic equilibrium models (Kriebel and Dean, Martin Marietta Energy Systems, Inc. and sub-
1985). contract M R E T T A 19X-91348V with Columbia
Studies such as those sponsored by the U N E P University. Programming assistance was provide
Regional Seas Program have begun to investi- by Mr. Jack Jia and Mr. Z.Y. Zhang. Apprecia-
gate potential adverse impacts of S L R on coastal tion is expressed to Mr. Robert M. Cushman
cities and ecosystems. An international con- and Dr. Paul Kanciruk, Oak Ridge National
ference convened in Venice, Dec. 1989, to discuss Laboratory, Oak Ridge, TN, and to Prof. R.W.
the impact of sea level rise on coastal cities. Fairbridge, GISS, and Dr. S. Jelgersma, Nether-
Other impact studies cited here include the Nile lands Geological Survey, for their helpful
and G a n g e s - B r a h m a p u t r a deltas (Milliman et suggestions and encouragement, and to Ms.
al., 1989), U.S. wetlands (Armentano et al., 1988), T a m m y W. White of O R N L for GIS d a t a entry,
and the Netherlands (Jelgersma et al., 1987). analysis and preparation of coastal risk maps.
Erosion risks h a v e been i n v e n t o r i e d for
European C o m m u n i t y countries (Quelennec, References
1989).
Armentano, T.V., Park, t~.A., and Cloonan, C.L., 1988. Im-
Preliminary statisitical summaries and d a t a p a c t s on coastal wetlands t h r o u g h o u t t h e U.S., in: J.G.
analyses have been presented for a coastal Titus, eel., Greenhouse Effect, Sea Level Rise a n d Coastal
hazards d a t a base, covering the East Coast, Wetlands, U.S. EPA, 230-50-86-0F~, 87--149.
Barnett, T.P., 1988. Global sea level change, in: NCPO,
U.S.A. Each of seven variables, relating to
Climate variations over t h e pant c e n t u r y and t h e
coastal inundation or erosion hazards, has been Greenhouse effect. A report based on t h e First C l i m a t e
assigned a rank ranging from 1 to 5, based on T r e n d s Workshop, 7 - 9 September, 1988, W a s h i n g t o n ,
GLOBAL COASTALHAZARDSFROM FUTURE SEA LEVEL RISE 397

D.C., National Climate Program Office/NOAA, Rock- level rise, EOS, Am. Geophys. Union Trans., 70: 1050.
ville Maryland. (Abstr.).
Bird, E.C.F., 1985. Coastline Changes, A Global Review, J. Emanuel, K.A., 1987. The dependence of hurricane intensity
Wiley & Sons, Chichester, 219 pp. on climate, Nature, 326: 483-485.
Bird, E.C.F., 1988. Physiographic indications of a sea-level Environmental Protection Agency, 1987. Saving Louisiana's
rise, in: G.I. Pearman, ed. Greenhouse-Planning for Coastal Wetlands: the need for a long-term plan of
Climate Change, CSIRO, Australia, 60-73. action: Report of the Louisiana Wetland Protection
Brammer, H. and Brinkman, R., 1990. Changes in soil re- Panel, U.S. EPA 230-02-87-026, 109 pp.
sources in response to a gradually rising sea-level, in: Int. Everts, C.H. 1985. Effect of sea level rise and net sand
Workshop, Effects of expected climate change on soil volume change on shoreline position at Ocean City,
processes in the tropics and subtropics, Nairobi, 12-14 Maryland, in: Potential Impacts of Sea Level Rise on the
February, 1990, ISSS, UNEP, CTA, 12p. Beach at Ocean City, MD, U.S EPA 230-10-85-013, 67-97.
Brinkman, R., 1984. Hydrology of tidal swamps and ad- Fairbanks, R.G., 1989. A 17,000-year glacio-eustatic sea level
jacent parts of coastal plains, in: Workshop on Research record: influence of glacial melting rates on the Younger
Priorities in Tidal Swamp Rice, International Rice Re- Dryas event and deep ocean circulation, Nature, 342:
search Institute, Manila, Philippines, pp. 127-141. 637-642.
Brnun, P., 1962. Sea level rise as a cause of shore erosion, J. Fisher, J.J., 1980. Shoreline erosion, Rhode Island and North
Water Ways, Harb. Div., ASCE, 71: 1729-1754. Carolina coasts-test of Bruun rule, in: Proc. Per Bruun
Bruun, P., 1983. Review of conditions for uses of the Bruun Symp., IGU Comm. on Coastal Environment, J.J. Fisher
rule of erosion, Coastal Eng., 7: 77-89. and M.L. Schwartz, eds., Western Washington Univer-
Budd, W.F., 1987. The expected sea-level rise from climatic sity, Bellingham, WA, 32-54.
warming in the Antarctic, in: G. I. Peannan, ed., Fricker, A. and Forbes, D.L., 1988. A system for coastal
Greenhouse-Planning for Climate Change, CSIRO, description and classification, Coastal Management, 16:
Australia, 74-82. 111-137.
Buddemeier, R.W. and Oberdorfer, J.A., 1989. Climate Galloway, R.W., Story, R., Cooper, R. and Yapp, G.A., 1984.
change and island groundwater resources, Report sub- Coastal Lands of Australia, CSIRO Div. Water and Land
mitted to South Pacific Regional Environment Pro- Resources, Nat'l Resources Ser. No. 1, 53 p.
gramme and the United Nations Environment Pro- Gibb, J.G., 1984, Coastal Erosion, in: Scientific, Economic
gramme, Lawrence Livermore National Laboratory and Social Reviews of Natural Hazards in New Zealand,
UCLR-100627, 20p. Task Force on Natural Hazards, New Zealand National
Buddemeier, R.W. and Smith, S.V., 1988. Coral reef growth Commission for UNESCO, Wellington, N.Z., 134-158.
in an era of rapidly rising sea level: Predictions and Goemans, T., 1986. The sea also rises: The ongoing dialogue
suggestion for long-term research, Coral Reefs, 7: 51-56. of the Dutch with the sea, in: Effects of Changes in
Bunkley-Williams, L. and Williams, Jr., E.H., 1990. Global Stratospheric Ozone and Global Climate, Vol. 4, Sea
Assault on coral reefs, Natural History, April: 47-54. Level Rise, UNEP and U.S. EPA, 47-56.
Carter, R.W.G., 1988. Coastal Environments, Academic Press Gornitz, V., 1990a. Mean sea-level changes in the recent
Ltd., London, Chap. 6, Sea-Level Changes, 245-279. past, in: R.A. Warrick and T.M.L. Wigley, eds., Climate
Chappell, J., 1987. Late Quaternary sea-level changes in the and Sea Level Change: Observations, Projections and
Australian region, in: M.J. Tooley and I. Sherman, eds., Implications (in press).
Sea-Level Changes, Basil Blackwell, Ltd., Oxford, 296- Gornitz, V., 1990b. Vulnerability of the East Coast, U.S.A.
331. to future sea level rise. J. Coast. Res., 9: 201-237.
Cocks, K.D., Gilmour, A.J. and Wood, N.H., 1988. Regional Gornitz, V., and Kanciruk, P., 1989. Assessment of global
impacts of rising sea levels in Coastal Australia, in: G.I. coastal hazards from sea level rise, Coastal Zone 89, Proc.
Pearman, ed., Greenhouse - Planning for Climate Change, 6th Syrup. Coastal and Ocean Management/ASCE,
CSIRO, Australia, 105-120. 1345-1359.
Day, J.W. Jr., and Templet, P.H., 1989. Consequences of sea Gornitz, V., and Lebedeff, S., 1987. Global sea-level changes
level rise: Implications from the Mississippi Delta, during the past century, in: D. Nummedal, O.H. Pilkey,
Coastal Management, 17: 241-257. and J.D. Howard, eds., Sea-Level Fluctuation and Coastal
Dean, R.G. and Maurmeyer, E.M., 1983. Models for beach Evolution, SEPM Special Publication No. 41: 3-16.
profile response, in Handbook of Coastal Processes and Hansen, J., Fung, I., Lacis, A., Rind, D., Lebedeff, S., Ruedy,
Erosion, Ch. 7: 151-166, CRC Press. R., and Russell, G., 1988. Global climate changes as
De Ronde, J.G., 1989. Past and future sea level rise in the forecast by Goddard Institute for Space Studies three-di-
Netherlands, in: A.J. Mehta and R.M. Cushman, eds., mensional model, J. Geophys. Res., 93: 9341-9364.
Workshop on sea level rise and coastal processes, Hopley, D. and Kinsey, D.W., 1988. The effects of a rapid
DOE/NBB-0086, U.S. Dept. Energy, Washington, D.C., short-term sea level rise on the Great Barrier Reef, in:
253-280. G.I. Pearman, ed., Greenhouse-Planning for Climate
Dolan, R., Hayden, B. and May, S., 1983. Erosion of the U.S. Change, CSIRO, Australia, 189-201.
shorelines, in: P.D. Komar, ed., CRC Handbook of Hull, C.H.J. and Titus, J.G., eds., 1986. Greenhouse Effect,
Coastal Processes and Erosion, Chap. 14, 285-299, CRC Sea Level Rise and Salinity in the Delaware Estuary,
Press, Inc., Boca Raton, FL. EPA 230-50-86-010, 86 pp.
Douglas, B. and Herbrechtsmeier, E., 1989. Determining sea Jelgersma, S., 1979. Sea-level changes in the North Sea
398 v GORNITZ

basin, in: E. Dele, R.T.E. Schuttenhelm and A.J. Wiggers, Orford, J., 1987. Coastal proce.~es: The coastal response to
eds., The Quaternary History of the North Sea, Acta sea-level variation, in: IR.J.N. Devoy, ed., Sea Surface
Univ. Uppsala Syrup. Univ. Ups. Annum Quingestesi- Studies-a Global View, Croom Helm, London, pp. 415-
mum Celebrantis: 2, 233-248. 463.
Jelgersma, S., Tooley, M.J. and Shennan, I., 1987, Volume Pearman, G.l., 1988. Greenhouse gases: evidence for atmo-
H: The impacts of a future rise in sea level on the spheric changes and anthropogenic causes, in: G.I. Pear-
European lowlands, in: European Workshop on Interre- man, ed., Greenhouse-Planning for Climate Change,
lated Bioclimatic and Land Use Changes, Noordwij- CSIRO, Australia, pp. 3 21.
kerhoet, the Netherlands, Oct. 17-21, 1987, 28 p. Peltier, W.R. and Tushingham, A.M., 1989. Global sea level
Jeusen, R.E., 1983, Atlantic Coast Windcast, shallow water, rise and the greenhouse effect: Might they be connected?.
significant wave information. WIS Report 9, USAE, Science, 244: 806-810.
WES, CE, Vicksburg, MS. Pethiek; J,, 1984. An Introduction to Coastal Geomor-
Kana, T.W., 1989. The South Carolina Coast I - Natural phology, Arnold, London, 260p.
Processes and erosion, in: D.K. Stauble and O.T. Mag- Pirazzoli, P.A., 1986. Secular trends of relative sea-level
()on, eds., Barrier Islands: Process and Management, changes (RSL) indicated by tide-gauge records, J. Coast.
ASCE, New York, 265-273. Res., Spec. Issue 1:1 26.
Kimball, S., Andem, F. and Dolan, R., 1985. Coastal Hazards Pirazzoli, P.A., 1989. Present and near-future sea-level
Map, U.S. National Atlas, 1:7,500,000, U.S. Geological changes, Global and Planet Change, 75: 241-258.
Survey, Reston, Va. Pugh, D.T., Spencer, N.E: and Woodworth, P.L., 1987. Data
Kraft, J.C., Chrzastowski, M.J, Belknap, D.F., Toscano, Holdings of the Permanent Service for Mean Sea Level_
M.A. and Fletcher, C.H. III, 1987. The transgressive Bidston Observatory, England.
barrier-lagoon coast of Delaware: morphostratigraphy, Quelennec, R.E., 1.989. The CORINE "Costal Erosion Pro-
sedimentary sequences and responses to relative rise in ject": Identification of coastal erosion problems and data
sea level, in: D. Nummedal, O.H. Pilkey, and I.D. How- base of the littoral environment of eleven European
ard, eels., Sea-Level Fluctuation and Coastal Evolution, countries, Coastal Zone 89, Proc. 6th Symp. Coastal and
SEPM Spec. Pub. No. 41: 129-143. Ocean Management, ASCE: 4594-4601.
Kriebel, D.L. and Dean, R.G., 1985. Estimates of erosion and Sallenger, A.H., Jr., Penland, S. Williams, S.J. and Sutter.
mitigation requirements under various scenarios of sea J.R., 1987. Louisiana barrier island erosion study, Coastal
level rise and storm frequency for Ocean City, Maryland, Sediments 87, WW Div./ASCE, New Orleans, LA, May
in: Potential Impacts of Sea Level Rise on the Beach at 12 14, 1987, 1503-1516.
Ocean City, Maryland, U.S. EPA 230-10-85-013, 99-t76. Short, A.D., 1988. The South Australia coast and Holocene
Leatherman, S.P., 1983: Barrier island evolution in response sea-level transgression, Geog. Rev., 78: 119-136.
to sea level rise: A discussion, J. Sed. Petrol, 53: 1026- Smith, S.E., and AbdeI-Kader, A., 1988. Coastal erosion
1031. along the Egyptian delta, J. Coast. Res., 4: 245-255.
Leatherman, S.P., 1984. Coastal geomorphic responses to sea Smith, J.B., and Tirpak, D.A, 1988. The Potential Effects of
level rise: Galveston Bay, Texas, in: M.C. Barth, and J.G. Global Climate Change on the United States: draft re--
Titus, eds., Greenhouse Effect and Sea Level Rise, Van port to Congress, Executive Summar, 51 pp.
Nostrand Reinhold Co., NY, pp. 151-178. Streif, H., 1989. Barrier islands, tidal flats, and coastal
Leatherman, S.P., 1985. Geomorphic effects of accelerated marshes resulting from a relative rise of sea level in East
sea level rise on Ocean City, Maryland, in: Potential Frisia on the German North Sea coast, Proc. KNGMG
Impacts of Sea Level Rise on the Beach at Ocean City, Syrup. on Coastal Ix}wtands, Geoh)gy and Geotechnol-
Maryland, U.S. EPA 230-10-85-013, 33-65. ogy, pp. 213-223, Kluwer Academic Pub., Dordrecht.
Lyles, S.D., Hickman, L.E., Jr. and Debaugh, H.A., Jr., 1987. Holland.
Sea Level Variations for the United States. 1855-1986. Tanner, W.F., 1988. Beach ridge data and sea level history
NOAA Nat'l Ocean Service, Rockville, MD, 182p. from the Americas, J. Coast Res., 4: 81-91.
Mehta, A.J. and Cushman, R.M., eds., 1989. Workshop on Thorn, B.G. and Roy, P.S.. 1988. Sea-level rise and climate:
level rise and coastal processes, DOE/NBB-0086, lessons from the Holocene, in: Pearman, G.I., ed.,
U.S. Dept. Energy, Washington, D.C., 289 pp. C,reenhouse-Planning for Climate Change, CSIRO,
Meier, M.F., 1989. Reduced rise in sea level, Nature, 343: Australia, pp. 177~ 1~.
115-116. United Natiom~ Enviromnental Programme and Delft Hy-
Milliman, I.D., Broadus, J.M., and F. Gable, 1989. Environ- draulics, 1989. Criteria for assessing vulnerability to sea-
mental and economic impact of ring ~ level and subsid- level ri~; A global inventory to high risk areas, 53 p.
ing deltas: the Nile and Bengal, Ambio, 18: 340-345. Warrick, R.A. and Oerlemans, J., 1990. Sea Level Rise. In:
National Ocean Service, 1988. Tide Tables 1988. NOAA, J.T. Houghton, G.J. Jenkins and J.J. Ephraums (Edivors),
U.S. Govt. Printing Office, Washington, D.C. Climate Change, the IPCC Scientific Assessment. Cam-
National Research Council, 1987. Responding to Changes in bridge Univ. Pres,% Pt). 257 281.
Sea Level, Engineering Implications, National Academy Woodworth, P.L., 1990. A search for accelerations in re~:ords
Pr~,u, Washington, D.C., 148 pp. ()f European mean sea level, [nt'l J. Climat., 10:129 143.
Oerlemarm, J., 1989: A projection of future ~ a level ri~,
Clim. Change, 15:151 ~174.