Global Vision International,

Seychelles - Mah Report Series No. 164

ISSN 1751-2255 (Print)

GVI Seychelles Mah

Marine Conservation Expedition

January - December 2016

 GVI Seychelles Mah
Marine Conservation Expedition Report January December 2016

Submitted in whole to
Global Vision International
Seychelles National Parks Authority (SNPA)

Produced by
Fanny Vessaz Science Coordinator
Benjamin James Taylor Science Officer
Fanchon Wright Science Officer
Josie Woodgate Science Officer
Rosabella Mangroo Research Assistant

And
Mariliana Leotta Base Manager
Corey Cole Dive Officer
David Edward Rowe Dive Officer
Christophe Mason-Parker Country Director

Thank you also to our hardworking volunteers for the collection of all data.

GVI Seychelles Mah/Marine Conservation Expedition

Address: GVI c/o SNPA, PO Box 1240, Victoria, Mah, Seychelles
Email: seychelles@gviworld.com Web page: http://www.gvi.co.uk and http://www.gviusa.com

2
Summary

This report summarises the findings of data collected from January to December 2016, covering
all monitoring activities carried out by GVI Seychelles on the north-west coast of Mah. This
included surveys to determine benthic cover, hard coral genera diversity and lifeforms, coral
recruitment, reef and commercially important fish density, diversity and biomass estimates of
commercially important fish as well as abundance of invertebrates that are of commercial
importance or ecosystem indicators.

Percentage hard coral cover across all surveyed sites was found to have decreased to 31.58 (
1.71) %, a decrease of 31% compared to the previous survey period (2015: 45.65 1.87 %).
Mean hard coral cover has fallen under 2010 levels (34.66 1.47 %), but it is still 164% higher
than 2005 mean hard coral cover (2005: 11.95 0.79 %). All granitic sites combined remain to
have a higher percentage coral cover than all carbonate sites combined (36.43 (2.66) % and
27.75 (2.13) % respectively). However, highest coral cover was found at the granitic reef
Therese South Reef with a mean of 60.13 (8.88) %. Lowest mean coral cover was found at the
carbonate site Whale Rock (14.37 (4.35) %). On all sites, the genus Acropora was the most
abundant coral genus found (comprising 36 % of mean hard coral cover). Branching lifeforms
are dominant on carbonate reefs, on granitic reefs encrusting lifeforms are mainly present.

The overall density of coral recruits was found at 6.61 (0.11) recruits per m2, a lower density
compared to the previous survey period. Highest coral recruit density was found at the granitic
site Therese South (7.36 (1.37) recruits per m2), lowest being recorded at Willies Bay Reef
(2.81 (0.42) recruits per m2). Highest recruitment was found for the genus Favites with a mean
of 2.03 (0.03) recruits per m2.

Mean fish density recorded in 2016 was 0.43 (0.05) individuals per m2 which equates to a 41%
increase since 2005 and 16% increase since 2015 (2005: 0.25 (0.008) fish m-2; 2015: 0.36
(0.004) fish m-2). Densities post 2008 and 2016 bleaching events are almost identical
(0.423;0.429 fish m-2 respectively) with highest densities recorded in the second half of the year
(post bleaching) as per 2015 but in contrast to all other survey periods. Reef fish densities
continues to increase at a slow steady rate whilst commercial fish densities have accelerated
resulting in a higher density of the latter for the first time since sampling began. This is largely
attributed to the increase in parrotfish (Scaridae) which made up 43% of fish recorded in 2016.
Granitic reefs support a higher density of fish than carbonate reefs, though the difference has
narrowed to 3%. Protected sites support higher densities than non-protected sites, with the

3
highest recorded in Baie Ternay. The gap between management strategies as well as between
marine parks (Port Launay vs Baie Ternay) is widening. The herbivorous parrotfish (Scaridae)
and rabbitfish (Siganidae) as well as the piscivorous groupers (Serranidae) are the most
abundant commercially important fish and have recorded increasing populations over the last
few years. The majority of fish are recorded within the 11-20 and 21-30 cm size classes and
density differences have been identified for Lethrinidae, Haemulidae and Lutjanidae with a
marked preference for granitic sites in some size classes. The majority of sites recorded a high
diversity (above 0.8), however richness varied widely between sites (13-39).

Overall the 10 m surveys all taxa decreased in mean density in comparison to 2015 surveys
except for Platyhelminthes (flat worms) and black spined sea urchins (Diadema sp. and
Echinothrix spp.). Arthropoda were the most commonly encountered invertebrate taxa with 2.24
(0.2) individuals m-2 overtaking Echinodermata which experience a 41% decline in mean
density from 2015 to 2016.

Consistent with previous years, long-spined urchins (Diadema sp.), short-spined urchins
(Echinothrix spp.) as well as Drupella spp. were the most abundant invertebrates recorded on
the 50 m belt surveys during both survey periods of January-June and July-December 2016. In
comparison to 2015, Drupella spp. overall mean density decreased 57% in 2016 (0.21 (0.04)
individuals m-2). The decline in obligate corallivorous Drupella spp. is thought to be as a result of
the large bleaching experienced in early 2016. Sea cucumbers increased from the population
low experienced in 2015 to 10.7 (1.82) individuals m-2 in 2016, however still remain lower than
the pre-2014 levels. Pearsonothuria graeffei and Stichopus spp. were the most commonly
observed taxa with 0.009 (0.002) individuals m-2 and 0.006 (0.003) individuals m-2
respectively. Densities of Pearsonothuria graeffei increased to the highest level observed since
surveys began in 2009.

The results show the impact of the severe coral bleaching event, caused by the El Nio
Southern Oscillation, on benthic communities in early 2016. Fish and invertebrate densities
display fluctuations, however 2017 is expected to show a different trend with a lag effect of
bleaching impacts. Outside survey activities relatively high numbers of Acanthaster planci are
still observed and removal is on-going. Considering the recent bleaching event, it is crucial to
keep Acanthaster planci populations under control through continuous and shared efforts, to
prevent further damage to weakened coral communities. The dataset collected by GVI will be
valuable when looking to understand the impact of future stress events, such as changing
climatic conditions, and will assist with the evaluation of reef resilience in the coming years.

4
 Table of contents
1. Introduction ........................................................................................................................11

1.1 Coral reef monitoring ......................................................................................................... 11

1.2 Aims .................................................................................................................................. 12

2. Methodology ......................................................................................................................13

2.1 Survey sites ....................................................................................................................... 13

2.2 Expedition practice and general methodology .................................................................. 16

2.3 Survey methodologies ....................................................................................................... 17

2.3 Data analysis ..................................................................................................................... 24

3. Results ...............................................................................................................................25

3.1 Benthic cover..................................................................................................................... 25

3.2 Hard coral genera diversity ............................................................................................... 36

3.3 Coral Recruitment ............................................................................................................. 39

3.4 Fish.................................................................................................................................... 44

3.5 Invertebrates ..................................................................................................................... 63

4. Discussion..........................................................................................................................82

4.1 Benthic cover and coral recruitment .................................................................................. 82

4.2 Fish density and community composition ......................................................................... 86

4.3 Invertebrates ..................................................................................................................... 92

5. Additional ecosystem monitoring .......................................................................................95

5.1 Turtles ............................................................................................................................... 95

5.2 Crown-of-thorns sea stars (COTs) .................................................................................... 97

5.3 Sightings of other fauna .................................................................................................... 99

References...........................................................................................................................101

Appendix A...........................................................................................................................107

Appendix B...........................................................................................................................108

Appendix C ..........................................................................................................................113

5
List of Figures
Figure 1: Location and substrate type of survey sites (modified after Samantha Howlett (not published)) ......................................... 14

Figure 2: Survey site schematic: Layout of benthos line intercept and invertebrate belt transects (10 m each) and coral diversity as
well as invertebrate belt transects (50 m each) .................................................................................................................................... 18

Figure 3: Layout of coral recruitment quadrats at each survey site. .................................................................................................... 19

Figure 4: Layout of fish species point counts and 50m fish visual census belt transects. ................................................................... 22

Figure 5: Mean percentage live hard coral cover for each survey period from 2005 2015, including survey results (means only) of
Engelhard (2004) prior to GVIs survey activities. Error bars indicate the standard error of the mean. ............................................... 25

Figure 6. Mean percentage live hard coral cover at carbonate and granitic survey sites for each survey period from 2005 2016,
including survey results (means only) of Engelhardt (2004) prior to GVIs survey activities. Error bars indicate the standard error of
the mean. ............................................................................................................................................................................................. 26

Figure 7. Mean percentage live hard coral cover in Baie Ternay Central prior and after the bleaching event early 2016. Error bars
indicate the standard error of the mean. .............................................................................................................................................. 27

Figure 8. Mean percentage live hard coral cover for each survey period from 2005 2016, including survey results (means only) of
Engelhard (2004) prior to GVIs survey activities. 2016 results are divided into pre- and post- bleaching periods according to survey
conduct dates (May 1 as threshold). Error bars indicate the standard error of the mean. ................................................................. 28
st

Figure 9: Mean percentage coral cover found at each site surveyed between January and July 2016. Error bars indicate the
standard error of the mean. Sites are displayed in a west-east gradient according to location. Dark grey bars indicate granitic reefs,
light grey bars indicate carbonate reefs; framed bars indicate the location within marine protected areas. ........................................ 29

Figure 10: Percentage cover of the coral genera Acropora, Pocillopora, Porites and Favites against total life coral cover (%) of
surveys conducted between 2005 and 2016. Error bars indicate the standard error of the mean. Note that pre-2009 only Acropora
and Pocillopora where surveyed to genus level. .................................................................................................................................. 30

Figure 11: Percentage of coral growth forms on total coral cover on carbonate reefs between 2005 and 2016. The category Other
includes digitate, foliose, mushroom and unrecorded/missing growth forms. ...................................................................................... 31

Figure 12: Percentage of coral growth forms on total coral cover on granitic reefs between 2005 and 2016. The category Other
includes digitate, foliose, mushroom and unrecorded/missing growth forms. ...................................................................................... 31

Figure 13. Mean percentage cover of coral growth forms at each site surveyed between January and July 2016. Error bars indicate
the standard error of the mean. Letters in brackets indicate the substrate type of the reef (granitic or carbonate). Sites written in
capital letters are located within marine protected areas. .................................................................................................................... 32

Figure 14: Percentage cover of different algae categories from 2005 2016. Error bars indicate the standard error of the mean.
Note the categories Turf algae and Algae assemblage where added to the LIT categories in 2009................................................ 33

Figure 15. Mean percentage cover of algae categories at each site surveyed between January and July 2016. Error bars indicate
the standard error of the mean. Letters in brackets indicate the substrate type of the reef (granitic or carbonate). Sites written in
capital letters are located within marine protected areas. .................................................................................................................... 34

6
Figure 16. Mean percentage cover of other benthic organisms from 2005 to 2015. Error bars indicate the standard error of the
mean. ................................................................................................................................................................................................... 35

Figure 17: Mean percentage cover of other benthic organisms on each site surveyed between January and July 2016. Error bars
indicate the standard error of the mean. Letters in brackets indicate the substrate type of the reef (granitic or carbonate). Sites
written in capital letters are located within marine protected areas. ..................................................................................................... 36

Figure 18: Mean number of coral genera found at all survey sites for each year from 2005 to 2016. Error bars indicate the standard
error of the mean.................................................................................................................................................................................. 37

Figure 19: Number of coral genera found at each site surveyed between January and July 2016. Sites are displayed in a west-east
gradient according to location. Dark grey bars indicate granitic reefs, light grey bars indicate carbonate reefs; framed bars indicate
the location within marine protected areas........................................................................................................................................... 38

Figure 20. Simpsons Diversity Index at each site surveyed between January and July 2016 (Simpsons Diversity Index D = 1 [ (
2
n / N) ] ). Sites are displayed in a west-east gradient according to location. Dark grey bars indicate granitic reefs, light grey bars
indicate carbonate reefs; framed bars indicate the location within marine protected areas. ................................................................ 39

2
Figure 21. Mean coral recruit density per m ( SE) recorded at each surveyed site between July December 2016. Sites are
displayed in a west-east gradient according to location. Dark grey bars indicate granitic reefs, light grey bars indicate carbonate
reefs; framed bars indicate the location within marine protected areas. .............................................................................................. 40

2
Figure 22. Mean coral recruit density ( SE) per m across the surveyed sites from 2005 to 2016. Bars represent overall mean coral
2 2
recruit density per m (primary y-axis), lines represent mean coral recruit density of the genera Acropora and Porites per m
(secondary y-axis). ............................................................................................................................................................................... 41

2
Figure 23. Mean coral recruit density ( SE) per m for the most abundant coral genera (number of recruits) of all surveyed sites
from 2005 to 2016. ............................................................................................................................................................................... 42

2
Figure 24. Mean number of coral recruits ( SE) recorded per m on deep and shallow areas of the survey sites for each survey
year. ..................................................................................................................................................................................................... 42

Figure 25. Mean number of coral recruits ( SE) recorded per m in the size classes <2cm and 2-5cm for each survey year. .......... 43
2

Figure 26. Mean total fish density per year. Error bars showing standard error. ................................................................................. 44

Figure 27. Mean total fish density per survey period. Error bars showing standard error.................................................................... 45

Figure 28. Mean total fish density per site (approximately West to East) during 2016. Dark grey indicate granitic sites; light grey
indicate carbonate sties; Outlined bars show sites located within MPAs; trend line shows average total fish density for 2005-2016.46

Figure 29. Mean reef and commercial fish density per survey period. Error bars showing standard error. ......................................... 47

Figure 30. Mean reef and commercial fish density per survey period. Error bars showing standard error. ......................................... 47

Figure 31. Mean reef fish density per survey period. Error bars showing standard error. ................................................................... 48

Figure 32. Mean total fish density per year for granitic and carbonate sites. Error bars showing standard error. ............................... 49

Figure 33. Mean total fish density per year for protected and unprotected sites. Error bars showing standard error.......................... 49

Figure 34. Mean density of commercial and reef fish in protected and unprotected sites. Error bars showing standard error. .......... 50

7
Figure 35. Density of commercial fish with Scaridae family Included and excluded from calculations for protected and unprotected
sites per year. Error bars showing standard error. ............................................................................................................................... 51

Figure 36. Mean fish densities for Port Launay (PL) and Baie Ternay (BT) marine protected areas. Error bars showing standard
error. ..................................................................................................................................................................................................... 52

Figure 37. Mean reef fish densities for Port Launay (PL) and Baie Ternay (BT) marine protected areas. Error bars showing standard
error. ..................................................................................................................................................................................................... 53

Figure 38. Mean commercial fish densities for Port Launay (PL) and Baie Ternay (BT) marine parks including and omitting Scaridae
family from calculations. Error bars showing standard error. ............................................................................................................... 54

Figure 39. Mean density of commercial families per year. Error bars showing standard error. ........................................................... 55

Figure 41. Mean percentage contributions of fish families to communities per year. Tetraodontidae (pufferfish) and Diodontidae
(porcupinefish) are omitted as negligible and not visible to community composition. .......................................................................... 58

Figure 42. Mean density of commercial fish species size classes in 2016. Error bars showing standard errors. Note that the category
100+ cm has been omitted as no individuals were recorded in that category. .................................................................................... 60

Figure 45. Fish diversity, evenness and richness of survey sites in 2016. .......................................................................................... 63

2
Figure 46. Mean density (individual per m ) of Annelida, Arthropoda and Platyhelminthes for every survey period from 2005 to 2016
across all survey sites. Error bars show standard error of mean. Note Platyhelminthes are represented by the secondary y-axis. ... 64

2
Figure 47. Mean density (individual per m ) of Echinodermata, Mollusca and black spine sea urchins for every survey period from
2005 to 2016 across all survey sites. Note that black spine sea urchins are not included in the phylum Echinodermata on the graph
due to their high abundance................................................................................................................................................................. 65

2
Figure 48: Mean density (individual per m ) of Annelida, Arthropoda and Platyhelminthes for every survey period from 2005 to 2016
across all carbonate survey sites. Error bars show standard error of mean. Note Platyhelminthes are represented by the secondary
y-axis. ................................................................................................................................................................................................... 66

2
Figure 49. Mean density (individual per m ) of Echinodermata, Mollusca and Black spine urchins for every survey period from 2005
to 2016 across all carbonitic survey sites. Error bars show the standard error of the mean................................................................ 67

Figure 50. Mean density (individual per m2) of Echinodermata, Mollusca and Black spine urchins for every survey period from 2005
to 2016 across all granitic survey sites. Error bars show the standard error of the mean. .................................................................. 68

2
Figure 51. Mean density (individual per m ) of Annelida, Arthropoda and Platyhelminthes for every survey period from 2005 to 2016
across all granitic survey sites. Error bars show standard error of mean. Note Platyhelminthes are represented by the secondary y-
axis. ...................................................................................................................................................................................................... 69

2
Figure 53. Mean density individuals per m of surveyed invertebrates across all carbonate and granitic sites for the January-June
2016 survey period. Error bars show the standard error of the mean.................................................................................................. 71

2
Figure 55: Mean density individuals per m of surveyed invertebrates across all carbonate and granitic sites for the July-December
2016 survey period. Error bars show the standard error of the mean.................................................................................................. 72

2
Figure 56: Mean density of individuals per m of short and long spine urchins across all carbonate and granitic sites surveyed from
2009 to 2016. Error bars show the standard error of the mean. Note July-December 2015 only 2 surveys were completed, both on
carbonitic sites. .................................................................................................................................................................................... 73

8
 2
Figure 57: Mean density of individuals per m (SE) of corallivorous invertebrates surveyed; cushion sea star (Culcita sp.) combined
with other sea stars, crown-of-thorns (Acanthaster planci) and Drupella spp from all survey periods across all sites. The densities of
cushion sea star (Culcita sp.) combined with other sea stars and crown-of-thorns (Acanthaster planci) is indicated on the secondary
y-axis. Error bars show the standard error of the mean. Note: 2015 data includes reduced surveys.................................................. 74

2
Figure 58: Mean density of individuals per m of Mathaes urchins, pencil urchins, cake urchins, flower urchins, giant clams and
oysters across all sites from 2009 until 2016. Error bars show the standard error of the mean. Note 2015 only contains January-June
2015 dataset as the July-December 2015 had only to surveys and skewed results. Note that oysters were inaccurately recorded in
2015-2016 and values are null. ............................................................................................................................................................ 75

Figure 59: Mean number of sea cucumbers recorded across all surveyed site per year from 2006 until 2016. Error bars show the
standard error of the mean. Note, in July-December 2015 only two carbonitic sites were surveyed................................................... 76

Figure 60: Mean density of sea cucumber taxa across all surveyed sites from January-June 2016 and July-December 2016 survey
periods. Error bars show the standard error of the mean. ................................................................................................................... 77

Figure 61: Mean density of sea cucumbers recorded across carbonate and granitic reefs during January-June 2016. Error bars
show the standard error of the mean. .................................................................................................................................................. 78

Figure 62: Mean density of sea cucumbers recorded across carbonate and granitic reefs during July-December 2016. Error bars
show the standard error of the mean. .................................................................................................................................................. 79

Figure 63: Mean density of individual sea cucumber taxa across all sites from 2008-2016. Error bars show the standard error of the
mean. Holothuria noblis, Holothuria sp. (pentard), Thelenota ananas, Holothuria edulis and Thelenota anax were all excluded from
the graph due to no data or minimal data. Note, in July-December 2015 only two carbonitic sites were surveyed. ........................... 80

Figure 64: Mean density of Stichopus sp. and Pearsonothuria graeffei across all sites from 2008-2016. Error bars show the standard
error of the mean. Note, in July-December 2015 only two carbonitic sites were surveyed.................................................................. 81

Figure 65. Number of COTs and the CPUE from January to December 2016. ................................................................................... 99

9
List of Tables
Table 1: Survey sites information ......................................................................................................................................................... 15

Table 2: Total number of turtle sightings recorded for each dive logged in 2016. ............................................................................... 96

Table 3: Total number of hawksbill and green turtle sightings recorded for each dive logged in 2016................................................ 96

Table 4. Total number of Crown-of-Thorns sightings recorded for each dive logged in 2016. ............................................................ 97

Table 5. Number of sharks, rays, lobster and octopuses recorded for each dive logged in in 2016. Note that number of dives has
been omitted, as data is only exploratory and number of dives has been fairly constant over the year. ........................................... 100

10
1. Introduction
Global Vision International (GVI) is a globally operating volunteering organization, which has
two expedition bases within the inner granitic islands of Seychelles. One expedition base is
situated on Curieuse Island within the Curieuse National Marine Park to the north of Praslin.
The other expedition base is located within the Baie Ternay Marine National Park at Cap Ternay
in the north west of Mahe Island. All of GVIs scientific activities in Seychelles are carried out on
behalf and under the methodological directory of the Seychelles National Parks Authority
(SNPA), which manages all of Seychelles national parks. GVI provides experienced staff, trains
and utilizes volunteers and supplies equipment to support the research section of the SNPA in
their monitoring activities by the collection of long term data sets.

1.1 Coral reef monitoring

The 1997/98 El Nio Southern Oscillation and the subsequent coral bleaching event caused
severe coral mortality worldwide (Spencer et al. 2000; Engelhardt 2002). Scleractinian coral
mortality in the inner granitic islands of the Seychelles exceeded 90% due to the combined
effects of bleaching and an Acanthaster planci outbreak (Engelhardt 2002), with dominantly
branching genera Acropora and Pocillopora often suffering high rates of mortality (Spencer et al.
2000). Monitoring of recovery of the reefs surrounding the North West coast of Mah was
initiated in 1998 by the Schoals of Capricorn, a three-year programme funded by the Royal
Geographic Society in conjunction with the Royal Society. Reef states and development were
further assessed between 2001 and 2004 as part of the Seychelles Marine Ecosystem
Management Project (SEYMEMP), which was the most comprehensive assessment of the coral
reefs within the inner islands of the Seychelles to date. Eighty-one carbonate and granitic reef
sites throughout the inner islands were monitored using fine scale monitoring techniques.
Monitoring efforts were continued by Reefcare International, a non-governmental organisation
based in Australia. The protocols established by Reefcare International provided a foundation
for those adopted by GVI Seychelles which continued reef monitoring along the North West
coast of Mah at sites selected by SNPA.

This continuing long term data set, covering more than 10 years of surveying, shows a unique
trajectory of reef development, allowing the assessment of ecosystem recovery after the 1998
bleaching event and an assessment of the impacts of the recent 2015 El Nio Southern
Oscillation that caused severe damage to coral reefs worldwide. Evaluation of ecosystem

11
recovery after two major bleaching events will provide crucial data and implications for future
coral reef and fisheries management.

1.2 Aims

The aim of the continuous survey activities is to monitor hard coral cover, recruitment and
diversity, fish density and diversity as well as the density of invertebrates. Specifically, the aims
of GVIs survey activities along the northwest coast of Mah between January 2016 and
December 2016 were to:

Assess diversity and density of reef and commercially important fish species
Assess sizes of commercially important fish species
Assess benthic assemblage, including evaluation of hard coral, soft coral, sessile
organisms coverage and substrate composition
Assess diversity of hard coral genera
Evaluate coral juvenile recruitment rates
Assess density of invertebrate hard coral predators and sea urchins
Assess abundance and diversity of commercially targeted invertebrate species including
sea cucumbers, lobster and octopus
Monitor and manage the abundance of crown-of-thorns seastars (Acanthaster planci)
Monitor sea turtle populations and develop a photo identification database.

12
2. Methodology

2.1 Survey sites

Surveys are conducted at 13 granitic and 11 carbonate reefs around the northwest coast of
Mah (Figure 1). Each survey site is divided into shallow and deep zones, with the shallow
zone being defined at 1.5 5.0 m depth and the deep zone being between 5.1 15.0 m depth.
Each site has a central point, marked by a distinctive landmark on the coastline, and is further
divided into left, centre and right sections. These areas are loosely defined as such by their
position with respect to the centre marker of the site (left and right are reached by a 25 m swim
from the centre point). All depths are standardised with respect to tidal chart datum so as to
eliminate tidal influence. See Table 1 for further site details.

13
Figure 1: Location and substrate type of survey sites (modified after Samantha Howlett (not published))

14
Table 1: Survey sites information

Site
Site Name GPS Reef type
No
1 Conception North Point S 0439.583, E 05521.654 Granitic
2 Conception Central East Face S 0439.891, E 055 22.258 Carbonate
4 Port Launay West Rocks S 0439.416, E 05523.382 Granitic
5 Port Launay South Reef S 0439.158, E 05523.695 Carbonate
7 Baie Ternay Lighthouse S 0438.373, E 05521.993 Granitic
8 Baie Ternay Reef North East S 0438.013, E 05522.405 Granitic
9 Baie Ternay Reef Centre S 0438.321, E 05522.504 Carbonate
10 Baie Ternay Reef North West S 0438.382, E 05522.133 Carbonate
11 Rays Point S 0437.347, E 05523.145 Granitic
12 A Willies Bay Reef S 0437.650, E 05522.889 Carbonate
12 B Willies Bay Point S 0437.589, E 05522.776 Granitic
13 A Anse Major Reef S 0437.546, E 05523.121 Carbonate
13 B Anse Major Point S 0437.509, E 05523.010 Granitic
14 Whale Rock S 0437.184, E 05523.424 Granitic
15 Auberge Reef S 0437.024, E 05524.243 Carbonate
16 Corsaire Reef S 0437.016, E 05524.447 Carbonate
17 White Villa Reef S 0436.935, E 05524.749 Carbonate
18 Lilot North Face S 0438.652, E 05525.932 Granitic
19 Site Y S 0437.771, E 05522.660 Granitic
21 Therese North End S 0440.101, E 05523.737 Granitic
22 Therese North East S 0440.099, E 05523.891 Carbonate
23 Therese South S 0440.764, E 05524.310 Granitic
24 Site X S 0437.059, E 05523.783 Granitic
25 Secret Beach Reef N/A Carbonate
* Sites listed in bold are located within marine protected areas

15
2.2 Expedition practice and general methodology

Expedition and survey periods: The GVI expedition comprises of volunteering programs that
are four, eight or twelve weeks long, running continuously throughout the year from January -
December. Within one year, each site is aimed to be surveyed for fish and invertebrates twice,
with the first set of surveys being conducted from January June and the second set conducted
from July December. Line Intercept Transects and coral diversity transects are undertaken
from January June to evaluate coral coverage and diversity. Coral recruitment quadrats are
used from July December to survey newly recruited colonies.

Health and Safety: The safety of all volunteers is paramount. All volunteers are given a health
and safety induction on base upon arrival and conservative diving guidelines are adhered to for
the duration of the expedition. In addition, volunteers complete the PADI Emergency First
Response course, and are taught how to administer oxygen in the event of a diving related
incident.

Dive Training: All volunteers must be at least PADI Open Water qualified to join the expedition.
Volunteers then receive the PADI Advanced Open Water course covering Boat, Peak
Performance Buoyancy, Navigation, Underwater Naturalist, and Deep Dive. Volunteers also
complete the PADI Coral Reef Research Diver (CRRD) course, which is specifically developed
for GVI. All volunteers are trained in the use of surface marker buoys, delayed surface marker
buoys and tape reels, plus any other survey equipment specific to the surveys they will be
conducting. Volunteers gain sufficient dive experience during the training period prior to
conducting surveys. Particular attention is given to the training of good buoyancy skills as
surveys are conducted in water as shallow as two metres and over delicate reef ecosystems.

Species Identification and survey methodology training: Volunteers are required to learn
identification of fish, coral or invertebrates along with all additionally recorded megafauna such
as turtles, rays, cetaceans and sharks. Training is provided in the form of presentations,
workshops and informal discussion with the expedition staff in the classroom. Self-study
materials are also available in the form of electronic and hard copy flashcards, as well as Indian
Ocean identification publications. Volunteers are taken on identification dives with staff
members for in-water testing; their responses are recorded and the dives continue until the
volunteer has demonstrated accurate identification of all necessary species/genera. Volunteers
need to pass a final classroom exam with at least 95% before they can proceed with the training
in survey methodology. To learn GVIs survey methodology for the respective surveys they aim

16
to conduct, volunteers receive initial on land training and subsequent in water training during
which volunteers conduct practice surveys together with a staff member. This training continues
until volunteers are deemed confident and reliable to conduct actual surveys.

2.3 Survey methodologies

2.3.1 Coral and benthic cover surveys

Genera surveyed
During all benthic surveys, hard corals are surveyed to genus level, including 47 genera from 14
families. Genera were first introduced for the LIT surveys in 2009, prior to which only Acropora
and Pocillopora were surveyed to this level, with all other genera categorised as other coral,
and broken down into growth forms. See Appendix A for the full list of coral genera surveyed as
well as the benthic categories used during the LIT surveys.

Line Intercept Transects (LIT)

Benthic cover and substrate composition around northwest Mah was assessed between
January and June 2016 with six 10 metre LITs at each site. Each transect was placed parallel to
the shore, with three transects placed within the shallow depth range (1.5 m 5.0 m) and three
transects placed within the deep depth range (5.1 m 15.0 m). All survey depths were
standardised to the respective chart datum at the time of the survey. Transects were
haphazardly spread amongst the left, centre and right of the site with at least 15.0 m distance
between them to avoid overlap (Figure 2). The benthic assemblage encountered directly under
the tape as well as the respective substratum was identified and recorded at each transition
point to the nearest centimetre. Coral was identified to genus level and growth form of the
majority of the colony was recorded.

Coral Diversity Belt Transects

Two belt transects were conducted at each site to assess the diversity of coral genera between
January and June 2016. The transect tapes were laid out from the shallow centre towards the
deep left (Belt A) and the deep right (Belt B) of the site at a 45 angle from shore where possible
(Figure 2). Due to the topography of some sites, various transects had to follow the reef instead
of a 45 angle. Each diver in a buddy pair surveyed 2.5 m in a tight S-shape pattern to the left or
the right of the transect tape, recording any coral genus encountered once.

17
Figure 2: Survey site schematic: Layout of benthos line intercept and invertebrate belt transects (10 m
each) and coral diversity as well as invertebrate belt transects (50 m each)

Coral recruitment quadrats

Reef regeneration around north-west Mah was investigated using haphazardly placed 1m2
benthic quadrats. Placement of quadrats was done across a specified depth range (1.5 m 5 m
for shallow, and 5.1 m 15 m for deep surveys). See also the schematic presentation of the
orientation of quadrats across a site on figure 3. Quadrats were placed over reef substratum,
not over large patches of sand or silt. Quadrats were held to a height of 1m above the area to
be sampled, carefully dropped then allowed to settle before examining the area contained. To
ensure safe diving practices, surveys were conducted in a buddy pair with each diver working
on any one quadrat and quadrats were placed 2m apart to maintain buddy contact. The
percentages of substrate cover (rock, rubble and sand) was described for each quadrat together
with percentage algal cover and the depth. Individual coral recruits located within the quadrats
were assigned to one of two size classes (1-2 or 2-5 cm size class), identified to genus level
and counted. All recruits with distinct grazing marks or any other damage (e.g. bleaching signs)
were recorded separately. Ideally 36 quadrats should be completed at each site; 18 for each
depth range, although a minimum of 30 quadrats per site was aimed for. This methodology is
based on the works done by Engelhardt (2002) for the coral recruit section.

18
Figure 3: Layout of coral recruitment quadrats at each survey site.

2.3.2 Fish surveys

Species list
The fish species chosen for survey represent a range of species that are commercially
important and those that play an important ecological role within the reef community as chosen
by SNPA. This data can be used to assess the status of coral reef fish assemblages as well as
giving an insight into coral reef dynamics and the state of local fisheries and the community
responses to current fishing pressure compared with historical data spanning over a decade.

Fish are generally surveyed to the highest resolution possible with the majority, over 80, being
surveyed to species level. Resolution is dependent on the commercial or ecological importance
of each species; and surveyed to the highest appropriate level as required by our partners. For
example; the majority of parrotfish species fill the same ecological niche and are therefore
surveyed to family level (Scaridae); whereas genera that encompass species of more than one

19
feeding guild are generally identified to species level. For a full list of species surveyed and the
taxonomic levels used please see Appendix B.

Stationary Point Counts (SPC)

Stationary point counts (SPC) are a commonly used underwater visual census technique for
assessing reef fish populations (Kulbicki 1998; Engelhardt 2004) and have been employed, in
different variations, by numerous studies internationally (Hill and Wilkinson 2004) as well as
locally by several studies within Seychelles (Jennings et al. 1996; Spalding and Jarvis 2002;
Engelhardt 2004; Graham et al. 2006). For coral reef assemblages point counts with a radius of
7-7.5 metres are thought to be the most appropriate for the size categories that reef fish
typically fall into (Samoilys and Gribble 1997). The post bleaching surveys undertaken as part of
the SEYMEMP project by Reefcare international utilised point counts with a radius of 7 metres
(Engelhardt 2001, 2004), when GVI took responsibility for the monitoring program in 2005 a
similar point count methodology was adopted.

At each site eight SPCs were conducted, spread evenly between the deep and shallow zones
(Figure 4). One SPC was conducted at the left and right sides of the site with two further point
counts conducted at the centre of the site in both the deep and shallow areas. Surveys were
always conducted by two divers, each responsible for counting a different selection of fish
species thus reducing the number of species each person had to count in order to increase
accuracy (Samoilys and Gribble 1997). A tape measure was used to delineate the 7 metre
radius of the SPC and also served as visual reference for the survey area. The tape was laid
perpendicularly towards the shore and the depth of the centre of the point count was recorded
as well as the start time of the survey period. Before starting the survey, divers waited for at
least one minute at the centre of the point count for fish to resume normal behaviour after the
disturbance of laying the tape. Each survey lasted a total of seven minutes with the two
surveying divers hovering above the reef at the centre point whilst rotating slowly for the first six
minutes so minimising behavioural disturbance. A brief search of the survey area was
conducted for the final minute in order to give a more accurate count of cryptic species.

Belt Transects
Belt transects were used in conjunction with stationary point counts as they allow surveyors to
cover a greater area for a similar level of effort (Colvocoresses and Acosta 2007). However
behavioural avoidance of fish species towards divers has been frequently noted and may lead

20
to lower densities of fish than those recorded from SPCs; therefore, steps were incorporated
into the methodology in order to minimise this (Samoilys and Gribble 1997; Hill and Wilkinson
2004).

At each site 4 transect belts were conducted running parallel to the shore, two in the deep zone
and two in the shallow, completed in conjunction with the left and right SPCs (Figure 4). On
sites where it was not possible to follow a straight bearing, belts were set following a contour
line parallel to the shore. Divers were instructed to count fish above the transect line if it did not
touch the substrate. Transect belts were 50m long and 5m wide; a standard survey area used
by a number of previous studies (Samoilys and Gribble 1997; Hill and Wilkinson 2004).
Transects were conducted by a pair of divers with one diver leading and counting one group of
fish, while the second diver laid the tape behind. The diver counts the commercially important
species on this pass, which can include the more errant species that show a greater level of
avoidance behaviours. This method of simultaneously surveying and laying the tape has been
recommended by Samoilys and Gribble (1997) as it avoids disturbing fish prior to the start of the
survey. After the initial survey divers waited outside of the survey area for three minutes before
the second diver returned down the belt counting the second group of fish (non-commercial)
while the tape was reeled up behind them. Each diver completed their surveys in a time of
between 8 and 12 minutes allowing a more accurate count of fish abundances as well as
decreasing the impact of diver disturbance.

21
Figure 4: Layout of fish species point counts and 50m fish visual census belt transects.

Commercial fish size estimation

As well as assessing the abundance, diversity and densities of commercial species from point
counts and belt transects, size estimation was used as a surrogate for the biomass of
commercial reef fish species and to assess community responses to fishing pressure (Samoilys
and Gribble 1997; Jennings and Polunin 1997). Surveyed species that are considered
commercially important include the emperors (Lethrinidae), groupers (Serranidae), rabbitfish
(Siganidae), snappers (Lutjanidae) and sweetlips (Haemuiidae). The diver surveying the
commercially important fish recorded sizes in 10 centimetre bands during both the SPC and belt
transect surveys. Observer bias was minimised by training volunteers on sizing during their
species identification dives and ensuring that surveys matched that of the instructors.

2.3.3 Invertebrates surveys

Species surveyed
Invertebrate species, which influence and can indicate the health and conditions of coral reefs
are surveyed along with commercially viable species which are under fishing pressure. The full
list of surveyed invertebrate species is included in Appendix C.

22
Belt transects (10m)
Invertebrate surveys are conducted in a buddy pair with the coral LIT diver, who lays out the
10m transect tape (see 2.3.1 Coral Line Intercept Transects (LIT)). At each site, six 10 metre
invertebrate surveys are carried out between January and June. After the coral LIT diver lays
out the transect, the invertebrate surveyor typically starts, along the same transect belt, 5
minutes later to minimise underwater clashes. Each transect is placed parallel to the shore, with
three transects placed within the shallow depth range (1.5 m 5.0 m) and three transects
placed within the deep depth range (5.1 m 15.0 m). All survey depths are standardised to the
respective chart datum at the time of the survey. Transects are haphazardly spread amongst
the left, centre and right of the site with at least 15.0 m distance between them to avoid overlap
(Figure 2). Using a systematic S bend swimming pattern, targeted cryptic invertebrate species
(see Appendix C) are recorded within 1 meter either side of the 10 m transect, covering a total
20m2 area.

Belt transects (50m)

The 50 m belt transects aim to quantify the abundance of key macro-invertebrate groups in a
given dive site. Two 50 m transect tapes are laid out at each site, from the shallow centre point
towards the deep. Belt A runs 45 to the left and belt B 45 to the right at each site where
possible (Figure 2); due to the topography of some sites, transects have to follow the reef
instead of a 45 angle. All survey depths are standardised to the respective chart datum at the
time of the survey. Each diver in the buddy pair surveys the target invertebrate taxa (see
Appendix C) within 2.5 m on the left and right side of the transect, using the systematic
Sshaped swimming pattern; surveying a 250 m2 area.
The extent of hard coral predation is measured by the density of the gastropods in the genus
Drupella and of two types of sea stars; the cushion stars (Culcita sp.) and the crown-of thorns
sea stars (Acanthaster planci). Algal grazing pressure is measured through recording the
density of sea urchins. Sea cucumbers and other species important to fisheries are also
recorded.

Environmental parameters

During each survey dive, the boat captain records the following environmental parameters:

Turbidity, as measured with a Secchi disk

23
 Cloud cover, as estimated in eights
Wind speed, as evaluated via the Beaufort wind force scale
Surface and bottom sea temperatures based on divers personal dive computers.

2.3 Data analysis

Data collected was analyzed through MicrosoftExcel spreadhseets without calculation of

statistical significance as it is outside the scope of this report. Simpsons Diversity Index (D = 1
[ ( n / N)2 ] ) was used to calculate diversity of different sites (for hard coral and fish
diversity) taking into account evenness of taxa distribution.

24
3. Results

3.1 Benthic cover

Live hard coral cover

Percentage hard coral cover was determined from line intercept transects completed across 20 survey
sites between the survey period January July 2016, equating to 119 LIT transects and 1190 m
surveyed. Mean live hard coral cover was 31.58 ( 1.71) % across all sites (Figure 5); a decrease of 31%
compared to the previous survey period (2015: 45.65 1.87 %). Mean hard coral cover has fallen under
2010 levels (34.66 1.47 %), but it is still 164% higher than 2005 mean hard coral cover (2005: 11.95
0.79 %). On carbonate reefs, mean hard coral cover was 27.75 2.13 % which represents a 34%
decrease compared to 2015 (41.75 ( 2.89) %). Granitic reefs showed a higher mean coral cover than
carbonate reefs with 36.43 2.66 %, a decrease of 27% compared to 2015 (50.12 2.08 %) (Figure 6).

Figure 5: Mean percentage live hard coral cover for each survey period from 2005 2015, including
survey results (means only) of Engelhard (2004) prior to GVIs survey activities. Error bars indicate the
standard error of the mean.

25
Figure 6. Mean percentage live hard coral cover at carbonate and granitic survey sites for each survey
period from 2005 2016, including survey results (means only) of Engelhardt (2004) prior to GVIs survey
activities. Error bars indicate the standard error of the mean.

The site Baie Ternay Central was surveyed twice, before and after the bleaching event. Mean
hard coral cover before bleaching was 54.68 7.3 %, in shallow and deep areas it was 55.57
3.32 % and 53.8 9.33 % respectively. After bleaching, mean hard coral cover experienced a
decrease of 62% overall with a mean hard coral cover of 20.77 4.7 % and a decrease of 71%
and 52% in shallow and deep areas respectively (mean hard coral cover: 15.9 5.34 % and
25.63 7.67 % respectively) (Figure 7).

26
Figure 7. Mean percentage live hard coral cover in Baie Ternay Central prior and after the bleaching
event early 2016. Error bars indicate the standard error of the mean.

Transects conducted in 2016 were split in two different phases according to survey dates (pre-
and post-bleaching) marking the beginning of dying coral colonies (end of April-early May (pers.
obs.), May 1st taken as reference to sort results). Pre-bleaching (before May 1st), 60 surveys
were conducted and 64 surveys post-bleaching (after May 1st). In protected areas, 27 surveys
were conducted pre-bleaching and 2 post-bleaching (unprotected areas: 33 and 62 surveys pre-
and post-bleaching respectively). The post-bleaching dataset of Baie Ternay Central was
included as an additional site in this section only. Mean hard coral cover post-bleaching across
all sites is 55% lower than pre-bleaching levels (Pre-: 44.65 2.48 %; post-: 20.26 1.20 %).
On carbonate sites, mean coral cover pre- and post-bleaching were respectively 42.94 4.08 %
and 19.74 1.40 % which accounts for a decrease of 54%. On granitic sites, mean coral cover
was 45.88 3.04 % pre-bleaching and 21.33 2.37 % post bleaching, which is a 53% decrease
from pre- to post-bleaching hard coral cover. Compared to the 2015 mean hard coral cover, pre-
bleaching mean hard coral cover is 3% higher on carbonate sites and 8% lower on granitic sites
(Figure 8).

27
Figure 8. Mean percentage live hard coral cover for each survey period from 2005 2016, including
survey results (means only) of Engelhard (2004) prior to GVIs survey activities. 2016 results are divided
st
into pre- and post- bleaching periods according to survey conduct dates (May 1 as threshold). Error bars
indicate the standard error of the mean.

Lowest mean coral cover was found at the granitic site Whale Rock with 14.37 (4.35) % and
the carbonate site Secret Beach with 16.33 (2.69) %. Highest mean coral cover was found at
the granitic site Therese South Reef with 60.13 (8.88) % and the carbonate site Baie Ternay
Central with 54.68 (7.28) % (Figure 9). The six reefs that lay within marine protected areas
were found to have a combined mean coral cover of 37.20 (3.22) % which is higher than the
combined mean of sites outside the protected areas (mean coral cover: 29.21 (1.97) %).

28
Figure 9: Mean percentage coral cover found at each site surveyed between January and July 2016.
Error bars indicate the standard error of the mean. Sites are displayed in a west-east gradient according
to location. Dark grey bars indicate granitic reefs, light grey bars indicate carbonate reefs; framed bars
indicate the location within marine protected areas.

Coral genera dominance

Mean cover of Acropora on granitic and carbonate reefs combined decreased from 22.84
(1.77) % in 2015 to 11.56 (1.44) % in 2016 (Figure 10). With the modification of the survey
methodology in 2009, identification of the most dominant corals became possible. In 2009,
Porites were dominant on carbonate sites (~38% of mean live coral cover); Acropora and
Favites dominated on granitic sites (~21% and 20% of mean live coral cover, respectively). In
2010, Acropora dominance was observed for the first time; with 2015 levels showing highest
dominance with Acropora representing 52% of mean live hard coral cover. In 2016 Acropora
corals still dominate across all sites (comprising 36 % of mean live hard coral cover), being the
most recorded genera on both granitic and carbonate reefs (Figure 10). Porites corals however
have increased in relative abundance across all sites, with 21% of mean live hard coral cover
compared with 2015 where they represented 16% of mean live hard coral cover.

29
Figure 10: Percentage cover of the coral genera Acropora, Pocillopora, Porites and Favites against total
life coral cover (%) of surveys conducted between 2005 and 2016. Error bars indicate the standard error
of the mean. Note that pre-2009 only Acropora and Pocillopora where surveyed to genus level.

Coral growth forms as proxy for structural complexity

Since 2010 the branching growth form is the most dominant on carbonate reefs, a trend that can
be observed throughout the years (Figure 11). However in 2016, branching corals decreased
from 56 % in 2015 to 39%; in comparison massive corals presence increased from 18% in 2015
to 28% in 2016. On granitic reefs, tabulate growth forms are more present since 2011 with the
highest cover in 2015 with 8%, and 6% in 2016 (Figure 12). On granitic reefs encrusting corals
have been most prevalent since surveying began, until 2014 (Figure 12). In 2014 and 2015,
branching corals dominated at about 40% cover. In 2016, encrusting corals became more
prevalent again with 41 % cover (2015: 27 %) in comparison to 32 % of branching corals. The
three sites Conception North Point, Therese North End and Therese South reef show a
prevalence of encrusting corals (Figure 13).

30
Figure 11: Percentage of coral growth forms on total coral cover on carbonate reefs between 2005 and
2016. The category Other includes digitate, foliose, mushroom and unrecorded/missing growth forms.

Figure 12: Percentage of coral growth forms on total coral cover on granitic reefs between 2005 and
2016. The category Other includes digitate, foliose, mushroom and unrecorded/missing growth forms.

31
Figure 13. Mean percentage cover of coral growth forms at each site surveyed between January and July
2016. Error bars indicate the standard error of the mean. Letters in brackets indicate the substrate type of
the reef (granitic or carbonate). Sites written in capital letters are located within marine protected areas.

Benthic assemblage

Turf algae represented the highest cover of non-scleractinian organisms in 2016 with a mean of
46.33 (1.66) % across all sites surveyed (Figure 14). Since the inclusion of turf algae as a LIT
category in 2009, it is the highest mean turf algae cover. Macro algae cover stayed low
throughout the years; the cover of coralline algae stayed relatively similar across the years
increasing slightly since 2014, with a mean cover of 7.44 (0.50) % recorded in 2016 (Figure
14).

32
Figure 14: Percentage cover of different algae categories from 2005 2016. Error bars indicate the
standard error of the mean. Note the categories Turf algae and Algae assemblage where added to the
LIT categories in 2009.

Between carbonate and granitic sites, coralline algae cover is relatively similar (mean cover for
carbonate sites: 7.60 0.66 %; mean cover for granitic sites: 7.25 0.75 %). Turf algae cover is
slightly higher on carbonate sites (mean cover for carbonate sites: 48.54 2.08 %; mean cover
for granitic sites: 43.53 2.65 %).

In 2016, the highest cover of turf algae for granitic sites was recorded at Whale Rock with a
mean of 66.52 (5.36) % and for carbonate sites, Willies Bay Reef with a mean of 61.31 (7.98)
%. Eight out of the twenty sites surveyed have a mean turf algae cover higher than 50%.
Highest cover of coralline algae was recorded at the granitic site Baie Ternay North West with
10.33 (2.60) % (Figure 15).

33
Figure 15. Mean percentage cover of algae categories at each site surveyed between January and July
2016. Error bars indicate the standard error of the mean. Letters in brackets indicate the substrate type of
the reef (granitic or carbonate). Sites written in capital letters are located within marine protected areas.

The mean cover of other benthic organism is shown in Figure 16. Across all sites in 2016, mean
cover of corallimorphs and zoanthids was 2.65 (0.34) %, whereas for soft corals and sponges
a mean cover of 1.70 (0.37) % and 1.30 (0.17) % respectively was recorded. Zoanthids and
corallimorphs are more prevalent than soft corals and sponges, a trend observed from 2005 to
2008 and last in 2014, in contrast to soft corals being dominant.

The percentage cover of soft corals, corallimorphs and zoanthids was shown to be higher on
carbonate reefs than granitic reefs (soft corals, carbonate: 2.73 0.63 % / granitic: 0.39 0.14
%; corallimorphs and zoanthids, carbonate: 3.60 0.52 % / granitic: 1.44 0.34 %), a trend
observed throughout the majority of the years since GVI surveys started.

34
Figure 16. Mean percentage cover of other benthic organisms from 2005 to 2015. Error bars indicate the
standard error of the mean.

The carbonate reef White Villa had the highest mean soft coral cover, recorded at 10.47
(4.72) %, similar to 2015 levels where White Villa showed the highest mean soft coral cover
with 10.33 (2.46) %. The carbonate reef Baie Ternay Central, had the highest mean cover of
corallimorphs, recorded at 8.55 (2.76) % (Figure 17).

35
Figure 17: Mean percentage cover of other benthic organisms on each site surveyed between January
and July 2016. Error bars indicate the standard error of the mean. Letters in brackets indicate the
substrate type of the reef (granitic or carbonate). Sites written in capital letters are located within marine
protected areas.

3.2 Hard coral genera diversity

Mean hard coral genera richness seems to fluctuate around 28 and 33 genera since surveying
began in 2005 (Figure 18). The highest genera richness was recorded on granitic reefs in 2008
with a mean of 32.58 ( 0.64) coral genera found across all granitic sites surveyed. In 2011 a
mean of 32.67 ( 1.12) coral genera was found on all carbonate reefs surveyed that year. Rare
genera are Alveopora, Coeloseris, Diaseris, Pectinia, Seriatopora and Siderastrea. Especially in
the first two years of GVIs survey activities (2005 and 2006) these rare genera were recorded
on more sites. In the following years recordings of rare genera was reduced to a few recordings
of especially Alveopora, with additional recordings of Siderastrea in 2007 and Pectinia in 2008.
With 2011 as an exception, recordings of rare genera were seldom. In 2011 Coeloseris,
Alveopora, Pectinica and Siderastrea where recorded on different survey sites, however
Seriatopora and Diaseris where last recorded in 2005 and 2006, respectively. In 2016,
Siderastrea, Halomitra, Alveopora and Pectinia were not recorded and Coeloseris, Pachyseris,
Podabacia and Symphillia were observed only once during surveys. From personal

36
observations however, Podabacia, Pectinia, Pachyseris and Symphillia are present at several
sites and especially in Baie Ternay.

Figure 18: Mean number of coral genera found at all survey sites for each year from 2005 to 2016. Error
bars indicate the standard error of the mean.

In 2016 a mean of 30.21 ( 0.63) coral genera were recorded across all carbonate and granitic
sites. On the carbonate site Baie Ternay Central 35 coral genera were recorded; 34 coral
genera were recorded on the carbonate site Port Launay South Reef and the granitic site
Conception Central East Face. The lowest coral genera richness was recorded at the granitic
site Therese North End and the carbonate site Anse Major Reef with 26 coral genera found on
both of these sites (Figure 19).

37
Figure 19: Number of coral genera found at each site surveyed between January and July 2016. Sites are
displayed in a west-east gradient according to location. Dark grey bars indicate granitic reefs, light grey
bars indicate carbonate reefs; framed bars indicate the location within marine protected areas.

The Simpsons Diversity Index takes into account the species richness and abundance,
therefore is a measure of each sites biodiversity (from 0 to 1, 1 = most equally diverse). It was
highest at the carbonate site Port Launay South Reef with 0.88 and lowest at the carbonate site
Baie Ternay North West with 0.46. Baie Ternay Central has a diversity index of 0.5, third lowest
across all sites (Figure 20). Baie Ternay Central was surveyed twice to investigate the effect of
the bleaching event taking place during the January-June 2016 survey period, the Simpsons
Diversity Index for the second survey (post-bleaching) for Baie Ternay Central is 0.81.

38
Figure 20. Simpsons Diversity Index at each site surveyed between January and July 2016 (Simpsons
2
Diversity Index D = 1 [ ( n / N) ] ). Sites are displayed in a west-east gradient according to location.
Dark grey bars indicate granitic reefs, light grey bars indicate carbonate reefs; framed bars indicate the
location within marine protected areas.

3.3 Coral Recruitment

The overall mean density of coral recruits in 2016 across all survey sites has been recorded at
6.61 (0.11) coral recruits per m2, with 36 coral genera recorded from 14 different families. The
highest recorded density was at Therese South with a mean of 7.36 (1.37) coral recruits per
m2, the lowest recorded density was at Willies Bay Reef with a mean of 2.81 (0.42) coral
recruits per m2 (Figure 21).

The genus Favites recorded the highest density with 2.03 (0.03) coral recruits per m2, the
lowest being 0.0016 ( 0.01) coral recruits per m2 recorded from the genera Podabacia and
Physogyra (Error! Reference source not found.).

39
Figure 21. Mean coral recruit density per m2 ( SE) recorded at each surveyed site between
July December 2016. Sites are displayed in a west-east gradient according to location. Dark
grey bars indicate granitic reefs, light grey bars indicate carbonate reefs; framed bars indicate
the location within marine protected areas.

There has been a decrease in the mean coral recruits per m2 from the initial surveys in 2005
(9.74 0.13 recruits per m2) to 2016 (6.60 0.1 recruits per m2). The mean density of coral
recruits per m2 for all genera decreased in 2016, falling from 13.04 ( 0.14) in 2015 to 6.60 (
0.1) in 2016, decreasing past 2005 recorded levels of 9.74 ( 0.13). Analysis of percentage
composition for the most dominant coral genera have shown the genus Acropora has increased
from 5.47% in 2005 to 12.9% in 2016, whereas the genus Porites has decreased from 22.7% in
2005 to 16.83% in 2016. However, both Porites and Acropora have experienced a decline from
2015 (Acropora; 15.65% and Porites 19.77% in 2015) (Figure 22).

40
 2
Figure 22. Mean coral recruit density ( SE) per m across the surveyed sites from 2005 to 2016. Bars
2
represent overall mean coral recruit density per m (primary y-axis), lines represent mean coral recruit
2
density of the genera Acropora and Porites per m (secondary y-axis).

There has been a decline in the coral recruit density per m2 from 2015 to 2016 for all of the most
abundant coral genera (Error! Reference source not found.). Pocillopora was recorded at its
lowest value since records began (0.64 in 2005 to 0.05 in 2016). Similarly, Porites decreased
from 2.21 in 2005 to 1.11 in 2016 after peaking in 2014 with 4.78 coral recruits per m2. Acropora
and Favites similarly declined from 2014 to 2016, (2.14 to 0.85 and 2.41 to 1.26 respectively).

41
 2
Figure 23. Mean coral recruit density ( SE) per m for the most abundant coral genera (number of
recruits) of all surveyed sites from 2005 to 2016.

2
Figure 24. Mean number of coral recruits ( SE) recorded per m on deep and shallow areas of the
survey sites for each survey year.

42
Depth specific results showed the lowest values since recording began; 2016 recorded mean
densities of 7.28 ( 0.19) and 5.92 ( 0.18) coral recruits per m2 in the deep and shallows
respectively (Figure 23). There was a decrease at deep and shallow sites from 2015 to 2016
(decreasing trend starting in 2014); deep decreasing from 13.07 ( 0.25) in 2015 to 7.28 (
0.19) coral recruits per m2 in 2016 and shallow falling from 12.98 ( 0.23) in 2015 to 5.92 (
0.18) in 2016 (Figure 24). Other results from 2016 showed greater values from the 2-5 cm size
class compared to the 0-2 cm size class, with mean densities of 3.45 ( 0.05) and 1.42 ( 0.06)
coral recruits per m2 respectively. Prior to 2016, both size classes maintained increasing values
throughout the years from the initial surveys in 2005. However, 2016 displays a decrease in
both size classes for the mean recruit densities with 2-5 cm class decreasing from 6.96 ( 0.07)
in 2015 to 3.45 ( 0.05) in 2016 and 0-2cm from 5.98 ( 0.08) in 2015 to 1.42 ( 0.06) in 2016
(Figure 25).

2
Figure 25. Mean number of coral recruits ( SE) recorded per m in the size classes <2cm and 2-5cm for
each survey year.

The average diversity (using Simpsons Index of Diversity (1-D) across the years was similar,
with 2010 having the greatest diversity (0.91) and 2014 the lowest (0.87). 2006 displayed the
greatest species richness (50) yet the lowest evenness (0.28), whereas 2016 had the lowest
richness (36) and greatest evenness (0.36). The species diversity was high for all sites (0 = no

43
diversity, 1 = high diversity), however the evenness was low, indicating that individuals were not
evenly distributed between the species and therefore less diverse. Post-hoc analysis would be
useful in determining the significance of the data, however this is unfeasible due to the time
restraints of this project. Consequently the diversity, evenness and richness graph has been
omitted from this report.

3.4 Fish

Fish Densities

During 2016, a total of 42,147 fish were surveyed with 16,955 in the first survey period (Jan Jun) and
25,192 in the second survey period (Jul Dec). The mean density of fish was 0.43 (0.05) individuals per
2
m . This demonstrates an increase of 41% since 2005 (0.25 (0.008)) when sampling commenced and
1
an increase of 16% since 2015 (0.36 (0.004)) (Figure 26).

Figure 26. Mean total fish density per year. Error bars showing standard error.

Fish densities display some overall seasonal variation with higher densities typically recorded in
the first half of the year. However, this trend was reversed in 2007, 2008, 2015 and 2016 with
the highest densities recorded in the second half of the year (Figure 27). Previous mass
bleaching events were recorded in late 2008-early 2009 with effects being seen in mid-2009. A
second mass bleaching event was recorded early 2016 and with effects shown in the second

1
Increase in densities of parrotfish (Scaridae) is responsible for this increase, see Figure 30.

44
half of the year. Total fish densities recorded in the 2008 and 2016 peaks were almost identical
at 0.423 and 0.429 fish m-2 respectively.

Figure 27. Mean total fish density per survey period. Error bars showing standard error.

Sites supporting the highest densities of fish were all located within Baie Ternay Marine Park
with the highest recorded at Baie Ternay Central (BTC) (0.86 fish m-2), following Baie Ternay
North East (BTNE) with 0.64 fish m-2, Baie Ternay North West (BTNW) with 0.54 fish m-2 and
Secret Beach (SB) with 0.53 fish m-2. Sites supporting the lowest densities of fish were Port
Launay South Reef (PLSR) (0.23 fish m-2), Auberge Reef (0.26 fish m-2) and Anse Major Reef
(AMR) (0.30 fish m-2) (Figure 28).

45
Figure 28. Mean total fish density per site (approximately West to East) during 2016. Dark grey indicate
granitic sites; light grey indicate carbonate sties; Outlined bars show sites located within MPAs; trend line
shows average total fish density for 2005-2016.

Reef fish species occur in higher densities than commercial species for all survey periods up to
Jul Dec 2015 where densities are virtually equal with 0.193 (0.06) for commercial and 0.192
(0.004) for reef species. Commercial species density overtakes reef species for the final
sample period of 2016 for the first time (0.26 (0.006 and 0.20 (0.005) respectively) (Figure
29). Commercial fish density has demonstrated an increase of 98% (0.13 0.26 fish m-2) in the
sampling period Jan 15 Dec 16 compared with an overall increase of 150% (0.10 0.13 fish
m-2) between Oct 05 and Dec 14 (Figure 29, Figure 30). The Scaridae family (parrotfish) made
up 43% of the fish recorded in 2016 and is prevalent in virtually all surveys. The trend in density
of Scarus spp. is reflected in overall commercial fish densities since the Jan Mar 2015
sampling period. However, when the Scaridae family is removed from calculations, population
densities show a slow but steady overall increase of 56% since 2005; nevertheless the rate has
accelerated and a increase of 64% was recorded between January 2015 and December 2016
(Figure 30).

46
Figure 29. Mean reef and commercial fish density per survey period. Error bars showing standard error.

Figure 30. Mean reef and commercial fish density per survey period. Error bars showing standard error.

Reef fish densities are recording a slow but steady increase with 35% recorded overall since
2005 (0.051 0.203 fish m-2) and a 5% increase recorded since 2015 (0.193 0.203 fish m-2)
(Figure 31). The most conspicuous and prevalent family of reef fish is the Acanthuridae
(surgeon fish, bristletooth and unicorn fish), which make up 19% of the reef fish recorded in

47
2016. The Holocentridae (soldier and squirrel fish) and the Chaetodontidae (butterflyfish) both
account for 11% of reef fish, therefore overall reef fish density is less impacted by family
population spikes than the commercial fish (Figure 40).

Figure 31. Mean reef fish density per survey period. Error bars showing standard error.

Fish densities and substrates

Granitic sites continue to support the highest density of fish in 2016 with 3% more fish than
carbonate sites. This has been the overall trend during bleaching recovery periods (2005-2008
and 2010-2016) (Figure 32).

48
Figure 32. Mean total fish density per year for granitic and carbonate sites. Error bars showing standard
error.

Management Strategies

Protected areas continue to support higher densities of fish than non-protected areas with 0.55
(0.016) and 0.19 (0.003) fish m-2 respectively in 2016 (Figure 33). The difference between
densities in protected and unprotected sites is also continuing to increase with a 188%
difference for 2016 compared with 119% in 2015.

Figure 33. Mean total fish density per year for protected and unprotected sites. Error bars showing
standard error.

49
Both reef and commercial fish recorded higher densities in protected than non-protected areas
(Figure 34). The difference in commercial fish densities between protected and unprotected
sites has grown substantially; 3% higher in protected sites in 2013 compared to 33% higher in
2016. The rate of increase also differs between management strategies with 104% increase in
density recorded from 2013 to 2016 for unprotected sites and 165% increase for protected sites.
The difference between reef fish densities in protected and unprotected sites is also increasing
with 12% recorded in 2014 and 39% in 2016. The rate of increase is much slower for reef fish
with a 42% increase recorded since 2014 for protected sites, however a decrease of 9% was
recorded post 2014.

Figure 34. Mean density of commercial and reef fish in protected and unprotected sites. Error bars
showing standard error.

The rapid increase in density of commercial fish can be largely explained by the Scaridae
family. On average (across all years) the difference between including and excluding Scaridae
in calculations for protected sites in 2016 was 0.11 (285%) and for unprotected was 0.08

50
(213%). In 2016 the difference between protected sites that included and omitted Scaridae in
calculations was 0.22 (407%) and for unprotected sites 0.17 (443%) (Figure 35).

Density calculations that include the Scaridae are always higher for protected rather than
unprotected sites for all years, whereas the higher densities fluctuate when Scaridae spp. are
omitted from calculations. Densities for both management strategies where calculations include
Scaridae are increasing at a fast rate since 2013, whereas when they are omitted densities
remain relatively stable (Figure 35).

Figure 35. Density of commercial fish with Scaridae family Included and excluded from calculations for
protected and unprotected sites per year. Error bars showing standard error.

Marine Protected Areas

Overall Baie Ternay supports a higher population density of fish than Port Launay except for
2011, when Port Launay showed higher densities (Figure 36). Prior to 2010 the difference in
density between the two marine parks was more pronounced, with Baie Ternay supporting
higher fish densities. Between 2010 and 2013, densities were relatively similar in both marine
parks. After 2012 higher densities were recorded in Baie Ternay and the difference between
marine parks has widened from the mean density being 7% higher in Baie Ternay in 2013
compared with 88% higher in 2016. The change in density differences between the parks is
exaggerated by a slight decrease in density in Port Launay in 2015 but overall remaining
consistent with 0.34 fish per m2 recorded in 2014 and 2016 ( 0.013; 0.006 respectively).

51
Figure 36. Mean fish densities for Port Launay (PL) and Baie Ternay (BT) marine protected areas. Error
bars showing standard error.

Reef fish density follow a very similar pattern to the overall fish densities with higher densities
recorded in Baie Ternay prior to 2010; this trend is inverted from 2010-2012; higher densities
were recorded in Baie Ternay, with an increasing difference between the parks (Figure 37). This
similarity between trends can be explained by looking at total abundances recorded for reef fish;
always higher than commercial fish, with an average of 16,533 and 11,109 fish recorded
respectively, with the exception of 2016. In 2016 the density of reef fish recorded in Baie Ternay
was 0.31 (0.02) and 0.16 (0.0007) in Port Launay, equating to a difference of 84%. Since
2014 reef fish density has increased in BT by 28% whilst it has decreased in PL by 12%.

52
Figure 37. Mean reef fish densities for Port Launay (PL) and Baie Ternay (BT) marine protected areas.
Error bars showing standard error.

Overall Baie Ternay supports higher densities of commercially important fish with the exception
of 2007 where Port Launay densities are higher. 2013-14 are also exceptions, with densities
that are virtually identical (Figure 38). From 2011-2016 fish density in Port Launay increased by
75% (0.1-0.17 fish m-2), compared with a 176% (0.12-0.33) increase in Baie Ternay. The overall
difference in density between marine parks has increased annually from 2014-2016 with a 4%,
55% and 91% recorded annually in 2014 to 2016. When Scaridae are removed from
calculations, the density of commercially important fish has remained relatively stable in both
marine parks. Average densities for 2005-2016 is 0.039 fish m-2 for Port Launay and 0.040 for
Baie Ternay, with fluctuations between Baie Ternay and Port Launay holding the highest
density. Average trends from 2011 to 2016 show density has decreased by 26% in Port Launay
compared to a 68% increase in Baie Ternay. When comparing fish density between the two
MNPs, Baie Ternay supports a higher number 26%; 6%; and 117% in 2014, 2015 and 2016
(Figure 38).

53
Figure 38. Mean commercial fish densities for Port Launay (PL) and Baie Ternay (BT) marine parks
including and omitting Scaridae family from calculations. Error bars showing standard error.

Commercial Species Analysis

Commercially important species surveyed include the rabbitfish (Siganidae), snappers

(Lutjanidae), emperors (Lethrinidae), groupers (Serranidae), sweetlips (Haemulidae) and the
parrotfish (Scaridae) families. Results show that Scaridae are the most abundant commercial
species with a density of 0.18 fish m-2 (0.002), followed by the Siganidae 0.14 (0.0004);
Serranidae 0.01 (0.0003); Lutjanidae 0.01 (0.0005); Lethrinidae 0.004 (0.00009) and
Haemulidae 0.0007 (0.00004) (Figure 39). Density of the Scaridae generally increased prior to
2009 then decreased and levelled off in 2013 at approximately the same density as when
surveys started in 2005, with 0.068 fish m-2. Density increased 168% from 2013-2016; and 49%
from 2015-2016. Population density of the Siganidae has shown small fluctuations over the
years; peaking in 2010 with 0.010 (0.0002) and 2016 with 0.014 (0.0004). Density fluctuated
from 0.005 (0.0001) in 2007 to 0.010 (0.0003) in 2010 to 0.005 (0.0001) in 2014 (0.0004)
to 0.014 in 2016. This equates to an increase of 84% for the 2010 peak (from 2007 to 2010) and
157% for the 2016 (from 2014 to 2016) peak. The Serranidae show a slow but gradual increase
in average mean density since the start of sampling (0.008-0.121 fish m-2) with some
fluctuations in density. Peak densities were observed in 2008 and 2014 (0.011 (0.0006) and
0.013 (0.0001) respectively) and the lowest density was recorded in 2010 at 0.005 (0.0002)
fish m-2. From the last peak in 2014 there has been a gradual decrease in density to 0.012

54
(0.0003); a decrease of 8%. Between 2006 and 2008, the Lutjanidae were one of the most
abundant commercial families with a substantial population decline in 2007. It is probable that
this family is responsible for the 2007 decrease observed in other trends (Figure 35, Figure 38).
The population recovered from 2009 to 2011 and has since remained relatively stable with small
fluctuations. In 2015-16, the population has increased from 0.011 (0.0001) to 0.012 (0.0003)
fish m-2 or 4%. The Lethrinidae saw a 46% (0.008 0.12) population density increase from 2005
to 2006, however recorded a slow but steady population decrease of (0.12 0.004) or 69% from
2006 to 2016. The population density of Haemulidae has been low throughout our survey
period, fluctuating between 0.0005 0.0015 (0.00002 0.00006). The current trend displays a
slow decrease in population density from 0.001 in 2014 to 0.0007 in 2016, a 39% decrease
(Figure 39).

Figure 39. Mean density of commercial families per year. Error bars showing standard error.

The herbivorous fish species Siganidae (rabbitfish) and Scaridae (parrotfish) have both
recorded an overall increase in population density. Scaridae populations are universally higher
across all years with a more rapid density increase in protected sites. From 2013, the rate of
population growth diverges between management strategies with a 5% difference in 2013 to a
32% difference in 2016 (Figure 40b). The Siganidae show a similar pattern from 2014 with
higher densities recorded in protected sites and a divergence in the rate of increasing
population density. Protected sites supported a 42% higher density rising to 85% higher in 2016

55
(Figure 40a). There is an overall trend of slow but decreasing density for the Lethrinidae
(emperors) in both protected and unprotected sites, with fluctuations between the highest
density occurring between the two differing management strategies. Densities were highest at
protected sites in 2006, 2012, and 2014, and for unprotected sites in 2009, 2013 and 2016. The
difference between peaks in density for protected sites is reducing in comparison to unprotected
sites over time. In 2016 unprotected sites supported 13% more fish than protected sites (Figure
40c). Trends for the Haemulidae are hard to discern due to the low numbers recorded and high
error margins. Highest densities between management strategies fluctuate due to the larger
flux between peaks and troughs for protected sites. At peak densities, protected sites are only
marginally higher, however they fall to lower levels compared with unprotected sites during
troughs. Due to the low numbers recorded, trends presented are with a lower reliability (Figure
40d). For the piscivirous fish the Serranidae show a general trend of increasing density, with
higher densities recorded in protected sites, except in 2005 and 2008. The rate of increase is
higher in protected sites with a 157% increase recorded from 2005-2016, compared with 20%
for unprotected sites. Since 2014 density fluctuated for protected sites and increased by 0.0002
or 1% whilst protected sites decreased by 0.001 or 15% over the same period (Figure 40f).
Density of the Lutjanidae (snappers) has remained relatively stable over time with the exception
of 2007 where a high peak was recorded with the highest densities recorded in protected sites
for that year and fluctuate between management strategies for other years. Since 2009
densities have been very similar for both strategies, however higher densities have been
recorded outside of protected areas since 2014 (Figure 40e).

56
 0.025 0.25
Protected Protected
Density m-2(SE)

Density m-2(SE)
0.02 Unprotected 0.2 Unprotected
0.015 0.15
0.01 0.1
0.005 0.05
0
0
2005
2006
2007
2008
2009

2011

2013

2015
2016
2010

2012

2014

2005
2006
2007
2008
2009

2011

2013

2015
2016
2010

2012

2014
a.
Year b. Year

0.02 0.002
Protected Protected

Density m-2(SE)
Density m-2(SE)

Unprotected Unprotected
0.015 0.0015

0.01 0.001

0.005 0.0005

0 0

2005
2006
2007
2008
2009

2011

2013

2015
2016
2005
2006
2007
2008
2009

2011

2013

2015
2016
2010

2012

2014

2010

2012

2014
c. d.
Year Year

0.07 0.02
Protected Protected
0.06
Density m-2(SE)

Density m-2(SE)

Unprotected Unprotected
0.05 0.015
0.04
0.01
0.03
0.02 0.005
0.01
0 0
2005
2006
2007
2008
2009

2011

2013

2015
2016
2010

2012

2014

e. Year f. Year

Figure 40. Mean density of commercial fish species per year. a) Siganidae, b) Scaridae, c) Lethrinidae, d)
Haemulidae, e) Lutjanidae, f) Serranidae. Error bars showing standard error.

57
Species assemblages

Figure 41. Mean percentage contributions of fish families to communities per year. Tetraodontidae
(pufferfish) and Diodontidae (porcupinefish) are omitted as negligible and not visible to community
composition.

During the recovery period from the 2008/9 bleaching event, the proportion of Chaetodontidae
(Butterflyfish) increased in conjunction with live coral cover (Figure 41; Figure 5). The highest
assemblage was recorded in 2013 (19.4%) and remained relatively stable until 2015. The
assemblage decreased from 17.2% to 11.4% from 2015 to 2016; a 34% decrease. The
Acanthuridae (surgeons, bristletooth and unicornfish) make up the largest proportion of the reef
associated fish. Surgeonfish and bristeltooths are herbivorous fish and account for the majority
of this family group. The highest proportion of Acanthuridae was recorded in 2011 (30%) and
decreased gradually to 19.8% in 2015. From 2015 to 2016 the proportion increased by 30.2% to
25.7%. The Labridae made up 16-20% of the population between 2006 and 2008. The
population crashed in 2009 and represented only 1.8% of fish. The population has been
gradually increasing since then and in the 2015-2016 survey period increased from 4.5% to
9.2%; equivalent to an increase of 104%. The Pomacanthidae (angelfish) have also recorded an
increase in proportional representation from 2015 to 2016 (1.1% - 4.1%). Populations have

58
been relatively stable, making up approximately 1% of the fish assemblage since 2012. The
Scaridae (parrotfish) have made up the highest proportion of commercially important fish since
surveys started. The proportion of Scaridae has been gradually increasing from 2011 to 2015 at
25.4 41.3%. Although the total numbers of parrotfish has increased from 2015-16 (Figure 35;
Figure 38; Figure 40b) the proportion of fish has decreased by 37% in one year, to 26%.

Commercial fish size analysis

Since 2009 GVI has also estimated the sizes of all rabbitfish, snappers, emperors, groupers and
sweetlips surveyed in order to give an estimation of the biomass of commercial families. A full
analysis of this data has yet to be undertaken, however the dataset for 2016 is presented below.
Large individuals of commercial fish are rare with only one individual recorded in the 51-60cm,
61-100cm and no individuals in the 100+ cm category in 2016. Most represented size classes
are 11-20 cm and 21-30 cm (Figure 42).

Siganidae (rabbitfish), the most abundant commercial fish species after Scaridae (parrotfish),
show the highest density in the size class 21-30 cm with 0.011 (0.0019) fish m-2. Serranidae
(groupers) have comparable densities in size classes 11-20 cm and 21-30 cm with 0.0064
(0.00040) and 0.0067(0.00038) fish m-2 respectively. Similarly, Lutjanidae (snappers) have
comparable high densities in smaller size classes; 11-20 cm and 21-30 cm with 0.0054
(0.0010) and 0.0043 (0.00073) fish m-2 respectively. Lethrinidae (emperors) have been
observed more frequently in the size class 21-30 cm (0.0023 0.00024 fish m-2); Haemulidae
(sweetlips) are present in low densities, the highest density occurring in the size class 21-30 cm
with 0.00028 (0.000085) fish m-2 (Figure 42).

59
Figure 42. Mean density of commercial fish species size classes in 2016. Error bars showing standard
errors. Note that the category 100+ cm has been omitted as no individuals were recorded in that
category.

For the Lethrinidae, Haemulidae and Lutjanidae, density differences can be observed between
carbonate and granitic sites (Figure 43). For Lethrinidae, this difference is more marked in the
size class 11-20 cm, with more fish recorded on carbonate sites (0.0017 (0.00054) fish m-2)
compared to 0.00063 (0.00016) fish m-2 on granitic sites (Figure 43b). Granitic sites appear to
be preferred by Haemulidae for all classes, except the 0-10 cm size class. This is most marked
for the size class 21-30 cm with 0.00009 (0.00005) fish m-2 on carbonate sites compared to
0.00049 (0.00017) fish m-2 on granitic sites (Figure 43c). Similarly, Lutjanidae prefer granitic
habitats; for the size class 21-30 cm this difference is emphasized by 0.00674 (0.0014) fish m-2
compared to 0.00195 (0.00057) fish m-2 on carbonate sites (Figure 43d).

Protected areas show slightly higher densities of commercial reef fish species overall, however
due to insufficient data, trends have to be investigated cautiously. Siganidae show higher
densities in protected areas, especially for the size class 21-30 cm, with 0.0173 ( 0.005) fish m-
2
in protected areas compared to 0.00843 ( 0.002) fish m-2 in unprotected areas (Figure 44a).
The size class 31-40 cm for Serranidae has the highest densities of fish in protected areas, with
0.0037 (0.0005) fish m-2 in protected areas and 0.0017 (0.0002) fish m-2 in unprotected areas
(Figure 44e). In contrast, Lethrinidae and Haemulidae show higher densities in unprotected
areas, both generally for the size class 31-40 cm. In this size class, Lethrinidae can be found at
0.0016 (0.0002) fish m-2 in unprotected areas compared to 0.00078 (0.0003) fish m-2 in
protected areas (Figure 44b); Haemulidae at 0.00027 (0.00007) fish m-2 and 0.000029
(0.00006) fish m-2 (Figure 44c).

60
Figure 43. Mean density of commercial fish families in 2016 per size class (0-10, 11-20, 21-30, 31-40, 41-50, 51-60, 61-
100 cm) on carbonate and granitic sites. A) Siganidae B) Lethrinidae C) Haemulidae D) Lutjanidae E) Serranidae F) All
above families. Error bars showing standard error. Note that the size class 100+ cm has been omitted since no individuals
were recorded.

61
 0.025 0.003
Protected Protected
0.02 0.0025
Density m-2 SE

Density m-2 SE
Unprotected Unprotected
0.015 0.002
0.0015
0.01
0.001
0.005 0.0005
0 0
A 0 - 10 11-20 21 - 30 31 - 40 41 - 50 51 - 60 61 - 100 B 0 - 10 11-20 21 - 30 31 - 40 41 - 50 51 - 60 61 - 100
Size classes (cm) Size classes (cm)
0.0005 0.008
Protected 0.007 Protected
0.0004
Density m-2 SE

0.006
Density m-2 SE

Unprotected
0.0003 0.005
0.004
0.0002 0.003
0.002
0.0001
0.001
0 0
0 - 10 11-20 21 - 30 31 - 40 41 - 50 51 - 60 61 - 100 0 - 10 11-20 21 - 30 31 - 40 41 - 50 51 - 60 61 - 100
C D
Size classes (cm) Size classes (cm)
0.01 0.009
0.008
0.008 Protected 0.007 Protected
Density m-2 SE

Density m-2 SE

Unprotected 0.006 Unprotected

0.006
0.005
0.004 0.004
0.003
0.002 0.002
0.001
0 0
E 0 - 10 11-20 21 - 30 31 - 40 41 - 50 51 - 60 61 - 100 F 0 - 10 11-20 21 - 30 31 - 40 41 - 50 51 - 60 61 - 100
Size classes (cm) Size classes (cm)
Figure 44. Mean density of commercial fish families in 2016 per size class (0-10, 11-20, 21-30, 31-40, 41-50, 51-60, 61-100
cm) in protected and unprotected areas. A) Siganidae B) Lethrinidae C) Haemulidae D) Lutjanidae E) Serranidae F) All above
families. Error bars showing standard error. Note that the size class 100+ cm has been omitted since no individuals were
recorded.

62
Diversity indices

In summary the average diversity index (Simpsons Diversity index (1-D)) was similar across the
sites, with Willies Bay Point having the greatest (0.965) and Port Launay South Reef the lowest
(0.81) (Figure 45). Post hoc analysis would be useful in determining the significance of the data
however this is unfeasible due to the time restraints of this project. Willies Bay Point had the
greatest diversity, evenness and richness (0.965; 0.822; 39 respectively). The species richness
fluctuated more between sites, and Conception Central East Face had the lowest value (13),
whereas the lowest species evenness was at Conception North Point (0.368) (Figure 45).

Figure 45. Fish diversity, evenness and richness of survey sites in 2016.

3.5 Invertebrates

3.5.1 10m transects

Specific surveyed invertebrate species have increased densities across all sites when
compared with 2005 surveys, with the exception of long spine sea urchins and short spine sea
urchins, referred to collectively as black spine sea urchins. Since the 2015 January-June
surveys, all surveyed invertebrate species have decreased with the exception of
Platyhelminthes and black spine sea urchins. This pattern is the opposite of the survey results

63
for 2015, where invertebrate phyla densities Echinodermata, Arthropoda, Annelida and Mollusca
mean densities across sites increased from the 2014 surveys and Platyhelminthes and black
spine urchins decreased (Figure 46 and Figure 47). The highest densities are observed in the
Arthropoda, Echinodermata and Mollusca phyla. The Arthropoda phylum has increased from the
low densities found in 2005 of 0.01 individuals per m2 (0.1) up to the most abundant
invertebrate found in 2016, with 2.24 individuals per m2 (0.2) (Figure 46).

2
Figure 46. Mean density (individual per m ) of Annelida, Arthropoda and Platyhelminthes for every survey
period from 2005 to 2016 across all survey sites. Error bars show standard error of mean. Note
Platyhelminthes are represented by the secondary y-axis.

The Echinodermata phylum has seen a 41% decrease since 2015 in overall mean density
across all sites, with 2.39 individuals per m2 (0.2) in 2015 compared to 1.32 individuals per m2
(0.12) in 2016 (Figure 47). Overall the mean density of the Echinodermata phyla has increased
by 44% since 2005 surveys, with 0.73 individuals per m2 (0.11) in 2016. The Mollusca phyla
however, decreased from 2015-2016 from 1.93 individuals per m2 (0.16) to 1.28 individuals per
m2 (0.11) respectively. However overall, the Mollusca mean density has increased since 2005

64
by 0.12 individuals per m2. The Annelida phylum appears to be relatively stable, showing a
slight decrease from 2015 to 2016, 0.47 individuals per m2 (0.06) and 0.36 individuals per m2
(0.04) respectively, yet this is an increase of 85% from 0.07 individuals per m2 (0.01)
observed in 2005 (Figure 47). Black sea spine urchins increased 16% from 2015 to 2016 with
0.48 individuals per m2 (0.06) to 0.57 individuals per m2 (0.05) respectively, but overall
appear to remain at a relatively stable density compared with previous years (Figure 47).

2
Figure 47. Mean density (individual per m ) of Echinodermata, Mollusca and black spine sea urchins for
every survey period from 2005 to 2016 across all survey sites. Note that black spine sea urchins are not
included in the phylum Echinodermata on the graph due to their high abundance.

Carbonate sites
Several Invertebrate phyla show differences in mean densities when sorted into the substrate
type Carbonate and Granitic reefs. Annelida, Echinodermata and Mollusca all decreased in
mean density from 2015 to 2016, whilst Arthropoda overtook Echinodermata as the most
abundant phyla across the carbonitic sites. Arthropoda densities increased by 19% from 1.82
(0.15) individuals per m2 in 2015 to 2.18 (0.23) individuals per m2 in 2016. Echinoderms

65
decreased by 35%, from 2.37 (0.23) individuals per m2 in 2015 to 1.53 (0.2) individuals per
m2 in 2016. Mollusca densities also decreased, by 20% from 2015 to 2016, from 1.61 (0.16)
individuals per m2 to 1.27(0.12) individuals per m2 respectively (Figure 48 and Figure 49).

2
Figure 48: Mean density (individual per m ) of Annelida, Arthropoda and Platyhelminthes for every survey
period from 2005 to 2016 across all carbonate survey sites. Error bars show standard error of mean. Note
Platyhelminthes are represented by the secondary y-axis.

66
 2
Figure 49. Mean density (individual per m ) of Echinodermata, Mollusca and Black spine urchins for every
survey period from 2005 to 2016 across all carbonitic survey sites. Error bars show the standard error of
the mean.

67
Granitic sites

Granitic reef sites have shown a decrease in densities of surveyed invertebrates in 2016, except
for black spine sea urchins (Figure 50, Figure 51). Black spine sea urchins increased in density by
7% from 0.52 individuals per m2 (0.11) in 2015 to 0.56 individuals per m2 (0.08) in 2016.
Despite a decrease of 24% in density, Arthropoda is now the most dominant phyla across all
granitic sites, with 3.07 individuals per m2 (0.4) in 2015 and 2.31 individuals per m2 (0.28) in
2016 (Figure 51). Echinodermata was the most dominant in 2015 with a density of 2.42
individuals per m2 (0.31) and in 2016 showed a decrease of 56% to 1.05 individuals per m2
(0.1). However, the 2016 Echinodermata survey results represent a 47% increase since the
surveys began in 2005 (0.71 individuals per m2 (0.14)). Mollusca phyla had a 44% decrease in
mean density across granitic sites from 2015-2016; 2.31 individuals per m2 (0.28) to 1.29
individuals per m2 (0.18) in 2016.

Figure 50. Mean density (individual per m2) of Echinodermata, Mollusca and Black spine urchins for
every survey period from 2005 to 2016 across all granitic survey sites. Error bars show the standard error
of the mean.

68
 2
Figure 51. Mean density (individual per m ) of Annelida, Arthropoda and Platyhelminthes for every survey
period from 2005 to 2016 across all granitic survey sites. Error bars show standard error of mean. Note
Platyhelminthes are represented by the secondary y-axis.

3.5.2 50m transects

In total 77 surveys were completed during 2016 comprising of 31 surveys during the January-
June period and 46 during the July-December period, covering a total area of 19,250m2. As in
previous years, Drupella spp., long spine urchins (Diadema spp.) and short spine urchins
(Echinothrix spp.) had the highest densities of all the invertebrates surveyed in the periods
January-June and July-December during 2016 (Figure 52). Of the most dominant invertebrates,
Short spine urchins and Drupella spp had higher densities in January-June period than the
subsequent July-December period, where it was noted that Drupella spp. decreased by 50%
from 0.28 individuals per m2 (0.05) to 0.14 individuals per m2 (0.04) respectively (Figure 52).
Short spine urchins decreased by 15%, from 0.39 individuals per m2 (0.05) in January-June
and 0.33 individuals per m2 (0.03) in July-December (Figure 52). Conversely, long spine urchin

69
 density increased by 27% from January-June to July-December with 0.11 individuals per m2
(0.03) and 0.14 individuals per m2 (0.04) respectively (Figure 53).

2
Figure 52. Mean density of individuals per m of surveyed invertebrates across all sites for the January-
June 2016 and July-December 2016 survey periods. Error bars show the standard error of the mean.

When dividing the survey periods into substrate type some interesting patterns can be
observed; in January-June 2016 Drupella spp. (carbonate 0.30.07 individuals per m2 and
granitic 0.2440.09 individuals per m2), Mathaes urchins (Echinometra mathaei) (carbonate
0.030.01 individuals per m2 and granitic 0.00080.01 individuals per m2), pencil urchins
(Heterocentrotus spp.) (carbonate 0.030.01 individuals per m2 and granitic 0.0010.01
individuals per m2) and long spine urchins (carbonate 0.160.07 individuals per m2 and granitic
0.040.02 individuals per m2) all had a higher density on carbonate reef sites. Short spine
urchins, however, (carbonate 0.260.05 individuals per m2 and granitic 0.540.08 individuals per
m2) had higher densities across granitic sites.

70
 2
Figure 53. Mean density individuals per m of surveyed invertebrates across all carbonate and granitic
sites for the January-June 2016 survey period. Error bars show the standard error of the mean.

The results for the July-December 2016 period also share the same pattern as January-June
2016. Drupella spp. (carbonate 0.15 0.08 individuals per m2 and granitic 0.13 0.03 individuals
per m2), Mathaes urchins (carbonate 0.010.004 individuals per m2 and granitic 0.005 0.01
individuals per m2), pencil urchins (carbonate 0.03 0.01 individuals per m2 and granitic 0.001
0.01 individuals per m2) and long spine urchins (carbonate 0.18 0.05 individuals per m2 and
granitic 0.1 0.05 individuals per m2) all had a higher density on carbonate reef sites.
Conversely, short spine urchins (carbonate 0.26 0.03 individuals per m2 and granitic 0.39 0.04
individuals per m2) and other sea stars (carbonate 0.01 0.003 individuals per m2 and granitic
0.012 0.03 individuals per m2) had higher densities across granitic sites (Figure 54).

71
 2
Figure 54: Mean density individuals per m of surveyed invertebrates across all carbonate and granitic
sites for the July-December 2016 survey period. Error bars show the standard error of the mean.

Yearly trends (50m transects)

Importantly, when reviewing the overall yearly trends, the result accuracy is compromised for
the 2015 period, as during the July-December 2015 period only two surveys were completed,
which were both conducted on carbonate sites, meaning no granitic sites were surveyed.

Long spine and short spine urchins

Short spine urchins (Echinothrix spp.) on carbonitic sites have seen an 11% increase in density
from initial observations in 2009 to 2016 with 0.230.03 individuals per m2 and 0.260.04
individuals per m2 respectively. Overall the short spine urchin densities across all the carbonitic
sites appear to relatively stable; between approximately 0.19-0.26 individuals per m2 in
comparison with those found on the granitic sites. On granitic sites short spine urchin mean
densities have increased 14% since the 2014 decline, from 0.40.07 individuals per m2 in 2014
to 0.470.06 individuals per m2 in 2016. Overall the short spine densities are 26% higher in
2016 than when surveys began in 2009, which had a density of 0.370.04 individuals per m2
(Figure 55). Long spine sea urchins (Diadema spp.) on carbonate reefs have declined 35%
since the 2015 surveys, decreasing from a mean density of 0.230.1 individuals per m2 in 2015
to 0.170.06 individuals per m2 in 2016. However, the 2016 density appears relatively similar in

72
comparison to the density found when surveys began in 2009 (0.170.07 individuals per m2)
(Figure 55). Conversely, on granitic reefs long spine urchins density began to increase in the
2016 survey period when compared with the reductions in density observed from 2011 until
2015. The long spine urchin 2016 density (0.070.03 individuals per m2) is 46% lower than 2009
densities of 0.130.06 individuals per m2 and 36% lower than the average yearly density of
2009-2015 of 0.11 individuals per m2 (Figure 55).

2
Figure 55: Mean density of individuals per m of short and long spine urchins across all carbonate and
granitic sites surveyed from 2009 to 2016. Error bars show the standard error of the mean. Note July-
December 2015 only 2 surveys were completed, both on carbonitic sites.

Corallivorous invertebrates
Drupella spp are the most abundant corallivorous invertebrate in 2016, although densities have
decreased 57% from 2015 to 2016, with 0.49 individuals per m2 (0.05) in 2015 and 0.21
individuals per m2 (0.04) in 2016 (Figure 56). The Drupella spp density in 2016 is 67% higher
than the average of all years from 2009 through to 2014, and is 33% higher than the average
from 2009 to 2015; suggesting a higher abundance of Drupella spp since the data records
began. Cushion stars (Culcita sp) and other sea stars referred to collectively as other sea stars,
have increased by 63% from 2009 to 2016, however they still remain at relatively low levels,

73
with 0.0031 individuals per m2 (0.001) in 2009 and 0.0084 individuals per m2 (0.002) in 2016
(Figure 56). Crown of Thorns (Acanthaster planci) have remained at stable low levels since
records began in 2009 (Figure 56).

2
Figure 56: Mean density of individuals per m (SE) of corallivorous invertebrates surveyed; cushion sea
star (Culcita sp.) combined with other sea stars, crown-of-thorns (Acanthaster planci) and Drupella spp
from all survey periods across all sites. The densities of cushion sea star (Culcita sp.) combined with
other sea stars and crown-of-thorns (Acanthaster planci) is indicated on the secondary y-axis. Error bars
show the standard error of the mean. Note: 2015 data includes reduced surveys.

Other Invertebrates
July-December 2015 data has been excluded from the yearly density calcualtion, as only two
surveys were conducted and presented anomalous results. Oysters have been accidentally
excluded from the survey list in surveys from 2015 and 2016 and hence no data from these
years have been collected for oysters.

Mathaes urchin shows a strong increase in mean density across the years from 2013 to 2014,
this trend then decreases from 2015 to 2016. The mean densities across all sites for 2016 is
0.013 individuals per m2 (0.004), which is 84% higher when compared to 2009 (0.002
individuals per m2 (0.001)) and 23% higher than the average mean density from 2009-2015
(0.01 individuals per m2) (Figure 57). Pencil urchins show regular fluctuations but appear
relatively stable around the 2009-2016 average with 0.015 individuals per m2. There has been a

74
slight decrease, of 16%, from 2009 when surveys began to 2016, 0.018 individuals per m2
(0.007) to 0.015 individuals per m2 (0.003) respectively (Figure 57).

2
Figure 57: Mean density of individuals per m of Mathaes urchins, pencil urchins, cake urchins, flower
urchins, giant clams and oysters across all sites from 2009 until 2016. Error bars show the standard error
of the mean. Note 2015 only contains January-June 2015 dataset as the July-December 2015 had only to
surveys and skewed results. Note that oysters were inaccurately recorded in 2015-2016 and values are
null.

75
Sea cucumbers (50m transects)
A total of 139 sea cucumbers were found across all surveyed sites in January-June 2016 and
305 individuals in July-December 2016. Figure 58 shows the mean number of cucumbers
encountered across all sites per year. The graph shows a gradual increase from 2006 until 2014
in the mean number of sea cucumbers observed at each site per year. The mean abundance
then decreases from 2014 to 2015 (9.2 mean individuals per site 1.17) and increases again in
the most recent period 2016 (10.7 mean individuals per survey 1.82) (Figure 58).

Figure 58: Mean number of sea cucumbers recorded across all surveyed site per year from 2006 until
2016. Error bars show the standard error of the mean. Note, in July-December 2015 only two carbonitic
sites were surveyed.

In 2016 the most common sea cucumber taxa by mean density is the Bohadschia spp.,
Actinopyga mauritiana, Stichopus spp. and Pearsonothuria graeffei (Figure 59). The
Pearsonothuria graeffei is the most dominant sea cucumber taxa across both periods with
January-June 2016 having 0.009 individuals per m2 (0.002) and July-December 2016 having
0.009 individuals per m2 (0.001). Specifically, during July-December 2016, Stichopus spp. had
the highest mean density across all surveyed sites with 0.009 individuals per m2 (0.006). In

76
January-June 2016 Stichopus spp. only had a mean density of 0.003 individuals per m2
(0.0007) (Figure 59). Holothuria fuscopunctata, Holothuria fuscogilva, Holothuria sp. (pentard)
and Thelenota anax were not observed throughout the 2016 survey period. Furthermore,
Actinopyga mauritiana was only observed in the July-December 2016 period, at a mean density
of 0.004 individuals per m2 (0.001).

Figure 59: Mean density of sea cucumber taxa across all surveyed sites from January-June 2016 and
July-December 2016 survey periods. Error bars show the standard error of the mean.

77
When comparing the sea cucumbers by substrate across both survey periods, it is apparent that
some species have a preferred substrate. In the January June 2016 period Stichopus sp.,
Pearsonothurian graeffi, Thelenota ananas and Actinopyga sp were at higher densities on
granitic sites than carbonate reefs (Figure 60). Pearsonothuria graeffei was at a 66% higher
mean density on granitic reefs than on carbonate reefs with 0.015 individuals per m2 ( 0.005)
and 0.005 individuals per m2 ( 0.0013) respectively. Conversely, Bohadschia spp. and
Holothuria edulis were more commonly found on carbonate reefs, with 0.004 individuals per m2
( 0.002) for carbonate sites and 0.002 individuals per m2 ( 0.0009) on granitic sites for
Bohadschia spp., and with 0.0009 individuals per m2 ( 0.0004) for carbonate sites and 0.00001
individuals per m2 ( 0.004) on granitic sites for Holothuria edulis.

Figure 60: Mean density of sea cucumbers recorded across carbonate and granitic reefs during January-
June 2016. Error bars show the standard error of the mean.

78
In the July December 2016 period Actinopyga mauritiana, Stichopus sp., Pearsonothuria
graeffei, Thelenota ananas and Actinopyga sp had greater densities on granitic sites than
carbonate reefs (Figure 61). Stichopus spp had 80% higher mean density on granitic reefs than
on carbonate reefs, with 0.01 individuals per m2 ( 0.01) and 0.002 individuals per m2 ( 0.001)
respectively. Actinopyga mauritiana was only found at granitic sites at a mean density of 0.007
individuals per m2 ( 0.002). Conversely, Bohadschia spp., Holothuria atra, Actinopyga sp and
Holothuria edulis were more commonly found on carbonate reefs.

Figure 61: Mean density of sea cucumbers recorded across carbonate and granitic reefs during July-
December 2016. Error bars show the standard error of the mean.

Combining both survey periods of 2016 Stichopus spp. and Pearsonothuria graeffei occurred at
much higher densities on granitic sites. Whereas, by combining both survey periods,
Bohadschia sp., Holothuria atria and Holothuria edulis were found at higher densities on
carbonitic sites, albeit at much smaller density than the granitic preferring sea cucumbers.

There was a large reduction in the majority of the mean densities recorded from individual sea
cucumber taxa in 2015, as reported in the previous annual report, however it appears that the
majority of sea cucumber taxa started to increase in the 2016 period, excluding Stichopus spp.
(Figure 62). Stichopus spp. declined to its lowest mean density since surveys began in 2008

79
(from 0.0068 (0.0003) individuals per m2) to current 0.006 (0.003) individuals per m2 (Figure
63). Stichopus spp. mean density across all sites in relation to 2015 declined 14% in 2016,
(2015 at 0.007 individuals per m2 (0.002)), whilst being 40% lower than the averaged mean
density of years 2009-2015 combined. Pearsonothuria graeffei, which experienced declines in
mean density during 2015 (from 0.005 (0.0007) individuals per m2), has increased 44% to
0.009 (0.002) individuals per m2 in 2016, which is the highest mean density for the taxa
observed since surveys began (Figure 63). Actinopyga mauritiana mean density in 2016 was
the highest observed in survey records, excluding 2015 results (where the survey period was
insufficient with no surveys being conducted on granitic reefs in July-December where
A.mauritiana is common), with 0.0018 (0.0005) individuals per m2, larger than the 0.0013
(0.0006) individuals per m2 observed in 2014. This could indicate a potential increase in
density across the sites. Holothuria atria mean density increased from 2015 to 2016 (<0.0001
(<0.0001) individuals per m2 and 0.0007 (0.0003) individuals per m2) but did not reach the
average mean density of years 2009-2014 combined, which was 0.001 individuals per m2.
Holothuria noblis, Holothuria sp (pentard), Thelenota ananas, Holothuria Edulis and Thelenota
anax have all been omitted from the graphs due these sea cucumber taxa not frequently being
observed.

Figure 62: Mean density of individual sea cucumber taxa across all sites from 2008-2016. Error bars
show the standard error of the mean. Holothuria noblis, Holothuria sp. (pentard), Thelenota ananas,
Holothuria edulis and Thelenota anax were all excluded from the graph due to no data or minimal data.
Note, in July-December 2015 only two carbonitic sites were surveyed.

80
Figure 63: Mean density of Stichopus sp. and Pearsonothuria graeffei across all sites from 2008-2016.
Error bars show the standard error of the mean. Note, in July-December 2015 only two carbonitic sites
were surveyed.

81
4. Discussion
With the completion of the 2016 surveys GVIs data set now represents 11 years of coral reef
monitoring activities. Data sets of this size and detail are vital in furthering our understanding of
coral reef population dynamics as well as providing an insight into the mechanisms and
trajectory of coral reef recovery following a large scale stochastic event.

4.1 Benthic cover and coral recruitment

4.1.1 Coral cover

The 1998 mass coral bleaching event caused 90% mortality in reef building or scleractinian hard
corals. Prior to 2016, surveys suggested moderate increases in coral cover and reef physical
complexity of lengthy process towards a recovery (Engelhardt 2004; Graham et al. 2006). This
might be explained by the physical isolation of Seychelles, impeding coral recruitment from
potential neighbouring healthy reefs (Graham et al. 2006). However, from the data collected in
2016 it appears that the El Nio Southern Oscillation event that occurred from 2015 to 2016
resulted in another mass bleaching event, reaching the Seychelles in February 2016 with high
water temperatures that remained until May 2016. This caused significant coral mortality with a
31% decrease in hard coral cover from 2015. Considering that surveys were conducted before,
during and after the bleaching event, coral cover in 2017 can be expected to be even lower than
2016, as healthy reefs surveyed before the bleaching hadnt experienced thermal stress yet.
According to preliminary results from a study conducted around Seychelles inner granitic islands
during May 2016 by Karen Cheung-Son, Nick Graham and Shaun Wilson (pers. comm.), the
2016 bleaching event may be equivalent or worse than the 1998 event. Nevertheless, with
water temperatures dropping from August 2016 onwards (24C pers. obs.) and the arrival of
southeast trade winds in December, there is hope some colonies can recover.

Similar to previous years, the mean percentage coral cover was found to be higher on granitic
reefs than carbonate reefs. Due to higher water flow and more exposed conditions, granitic sites
usually have a better water clarity (Engelhardt 2004, previous GVI reports); whereas carbonate
sites located in more sheltered areas are inclined to receive more sediment runoff and nutrients
which could negatively influence coral growth and cover (Engelhardt 2004). Following this, the
highest mean coral cover was found on the granitic site Therese South, followed by the
carbonate site Baie Ternay Central (BTC), which showed the highest mean coral cover in 2015.
Note that Therese South was not surveyed in 2015, and surveyed prior the bleaching event in

82
2016; BTC being surveyed twice (before and after), showing a 70% decrease in mean hard
coral cover from pre- to post-bleaching.

Granitic and carbonate reefs however, appear to have been equally impacted by the 2016 on-
going bleaching event, considering mean coral cover on granitic sites had started decreasing
already pre-bleaching, whereas carbonate reefs mean coral cover showed a slight increase
prior to bleaching. It might indicate that bleaching impacts started earlier on granitic reefs, which
may be more sensitive to a change in water temperature due to water clarity. Water turbidity,
higher on carbonate reefs located in more sheltered environments, may provide protection from
high irradiance levels, that in combination with thermal stress can cause coral bleaching.
Carbonate reefs located in more sheltered environments might have experienced bleaching
impacts later, as some studies have shown that reefs located in areas with more sediment
runoff have been found to be more resilient to changing temperatures (Browne et al. 2015;
Guest et al. 2016).

Reefs located in protected areas present a higher mean coral cover than reefs in unprotected
areas, as in 2015. However, results need to be compared with next years monitoring data in
order to draw conclusions (as the data was collected during the bleaching event). 90% of
surveys in protected areas were conducted prior to the bleaching impacts, which may bias the
trend observed.

Corals from the genus Acropora appear to be more impacted with a significant decrease in
percentage cover. Fast growing branching corals, such as Acropora and Pocillopora, are highly
susceptible to bleaching, a pattern consistent over a wide geographic range (Loya et al. 2001;
Baird and Marshall 2002; McClanahan et al. 2004). In contrast, massive and encrusting colonies
are found to be more resistant due to their morphological features (e.g. tissue thickness and
shape-dependent energy transfer efficiency) (Loya et al. 2001; Baird and Marshall 2002).
Massive corals are typically found in marginal environments with conditions outside preferred
range of corals (Hennige et al. 2010). Following this pattern, results show an increase in relative
proportion of massive, submassive and encrusting growth forms. Coral cover of Porites spp.
and Favites spp.; developing massive and encrusting colonies, display similar levels to 2015,
which may indicate a higher resistance of these genera due to their growth form. This pattern
can be observed in the recruitment results as well where the genus Favites dominates the
recruit composition.

83
Considering that branching growth forms provide valuable habitat for reef organisms (Gratwicke
and Speight 2005; Graham et al. 2006), a decrease in Acropora cover and overall branching
growth forms can have negative impacts on other reef organisms dependent on branching
corals for shelter and food, therefore impacting the wider ecosystem.

4.1.2 Coral diversity

BTC has a very low diversity index, despite showing high coral cover pre-bleaching. This can be
explained by Acropora spp. dominating this site, which is also responsible for the high decrease
in coral cover from pre- to post-bleaching surveys. As mentioned previously, branching growth
forms such as Acropora spp. are more sensitive to bleaching (Loya et al. 2001). Interestingly,
BTC diversity index is higher for the post-bleaching survey, as the balance between different
genera is restored with a decrease in cover of Acropora spp.

Coral genera richness belt transects are useful to cover a wide area of the survey sites and
increase the chances to find rare coral genera. Several genera have not been recorded or only
once, such as Halomitra, Podabacia, Pectinia, Pachyseris and Symphillia. Nevertheless, these
are present at several sites and especially in Baie Ternay from personal observations.
Komyakova et al. (2013) also showed the association of coral species richness and fish species
richness and total abundance. They suggest using coral species richness as a good indicator of
a healthy and diverse reef ecosystem, equally important as coral cover, therefore emphasizing
the significance of coral diversity belts as a monitoring method.

4.1.3 Non-scleractinian benthic composition

The benthic community composition is relatively similar to previous years. Fast-growing and
short-lived turf algae cover (Birkeland 1977) are the dominant non-scleractinian organisms,
which increased to match pre-2015 levels. Macroalgal cover is still very low, which is a potential
indication of sufficient herbivorous pressure preventing algae succession (Pratchett et al. 2011).

Sponges, soft corals, corallimorphs and zoanthids are present in low numbers. As for
corallimorphs and zoanthids, they are found in higher proportions than soft corals and sponges,
a tendency also observed in 2014 and 2009. In 2016 this higher proportion mostly consists of
corallimorphs. These organisms are secondary colonizers that have a competitive advantage
over corals due to their anatomic and life history traits (e.g. clonal reproduction) (Kuguru et al.
2004; Chadwick and Morrow 2011). Zoanthids also may replicate clonally at rapid rates and
monopolize reef substratum (Chadwick and Morrow 2011). Interestingly, Cruz et al. ( 2016)
showed that increased coral mortality in Brazil usually followed an increase in zoanthids

84
because of competition in disturbed environments (e.g. high nutrient levels). Therefore, an
increase in corallimorphs and zoanthids might be a negative sign for reef health and prevent
successful recovery after the recent bleaching event. Future monitoring data will provide more
information about potential trends in phase shifts or reef recovery.

4.1.4 Coral recruitment

Coral recruitment, dependent on healthy coral colonies, has been highly impacted with a 50%
decrease in coral recruits in 2016 compared with 2015. Considering that the increase in
percentage of live hard coral cover occurs largely as a result of the growth of recently settled
hard coral recruits rather than that due to on going growth of remnant corals (Engelhardt 2004),
a decrease in recruit densities indicates a negative impact for the reef recovery.

Mean coral recruit density results from 2016 displayed a pattern dissimilar from previous survey
years. In contrast to 2014, when results reached a record high, 2016 results reached a record
low since the inception of surveys in 2005, however previous years displayed a trend of mean
coral recruit density increasing. This is a negative sign for the growth in the coral communities
on the reefs within the survey area, indicating a degradation event and decreased resilience to
such events. This characteristic of the newly developing coral communities is a likely reflection
of different rates of bleaching-induced mortality experienced by the major ecological
components of local coral communities (Engelhardt 2004).

The fast-growing genera, Acropora, is a key structural component of healthy and diverse coral
reef ecosystems in the Indo-Pacific (Engelhardt 2004). Despite the decrease in 2016, Acropora
is still among the dominant genera for recruits on the monitored reefs, suggesting its potential to
regain the ecologically dominant status on the reefs in the future. The high morphological
diversity and structural complexity found in this genus promotes ecological relationships with
other reef organisms including many invertebrates and fish species (Engelhardt 2002). Since
surveys began in 2005, the genus Porites has always been dominant in terms of overall
abundance, but in 2016 Favites emerged as the most prolific coral recruit genera per m2. The
2016 survey results suggest that the effects of recent, renewed coral bleaching events on the
mean coral recruit density have been relatively major.

Depth specific density results seem to conform to previous norm with the deeps showing higher
densities compared to the shallows. Variation in coral recruit rates at this spatial scale may be
due to changes in light intensity with both depth and orientation, differences in algal biomass,

85
sediment and grazing intensity (Babcock and Mundy 1996). Recruit size class analysis showed
that both 0-2cm and 2-5cm categories have decreased in the overall mean density of coral
recruits in 2016. This drop could indicate a reduction in the spawning of adult colonies of
increased rates of mortality for newly settles recruits (Engelhardt 2002).

This information on patterns of coral recruitment over the years is crucial. In addition to
understanding population dynamics, knowledge of recruitment patterns is a prerequisite for the
effective management of marine ecosystems, enabling informed responses to disturbances,
such as crown-of-thorns starfish outbreaks, storms and bleaching events (Babcock et al. 2003).

4.1.5 Reef resilience

Reef resilience capacity will be crucial in recovering from the 2016 bleaching event which,
according to our findings for benthic cover and coral recruitment, already displays worrying
impacts on Mahes northwest reefs. Riegl et al. (2015) suggest that reefs in the northwest and
central Indian Ocean are at highest risk of reef degradation from heat stress, coupled with
changes in the nutrient composition that further impact on reef resilience. Regeneration from
such events might be prevented, with an increase in magnitude and frequency of stress events,
leaving no time for the slow reef recovery (Nystrm et al. 2000). To cope with changing
conditions, corals will need to rapidly increase their tolerance; assessments of the adaptive
capacity of reefs will be crucial to evaluate resilience and develop management strategies
taking into account climate change.

Nevertheless, corals have been shown to demonstrate more resilience after a bleaching event
through a change in their symbiotic community, adapting the partnership to one that is more
resistant to higher thermal stress (Silverstein et al. 2012, 2015). Future monitoring might
therefore show a different recovery trajectory than for the 1998 bleaching event, unraveling new
patterns in coral and reef resilience.

4.2 Fish density and community composition

4.2.1 Fish density and benthic cover and composition

There is little data on the fish populations of the reefs around North West Mahe prior to or
immediately after the 1998 bleaching (Engelhardt 2001). Our results show that fish densities
have increased overall since GVI monitoring began and correlate to mean live coral cover, with

86
the exception of 2016 where mean live coral cover experienced a decrease. The relationship
between coral cover and fish abundances and diversity is however highly complex and the
connections hard to discern (Chabanet et al. 1997).

The relationship between corals and overall fish abundance is much less clear with some
studies finding a significant correlation between live coral and abundance (Bell and Galzin 1984;
Wibowo et al. 2016) and others finding no relationship (Roberts and Ormond 1987). Physical
complexity of coral reefs has also been shown to have positive effects of fish abundances in
some studies (Palacios and Zapata 2014). Fish populations are affected by many different
factors, such as predation (including by man), juvenile recruitment, inter and intraspecific
competition for resources and prey availability (Chabanet et al. 1997). The importance of such
factors in determining the observed abundances of fish will vary between species, populations
and temporally thus making the drivers behind fish population dynamics hard to discern. The
effect of coral cover is likely to be an important driver behind the abundances of fish most
dependent on corals for either food or shelter (Chabanet et al. 1995), however results are
inconclusive for our study sites (unquantified). For example the densities of corallivorous
butterflyfish has increased enormously since GVIs monitoring began in 2005 but plummeted in
2016; however densities of parrotfish has been increasing sharply since 2013 and at an
accelerated rate since 2015. The association between percentage coral cover and fish density
is hard to establish prior to further analysis, for example; Baie Ternay Central has the second
highest coral cover, coral diversity and highest fish density, however Therese south has the
highest coral cover and a mid-range fish density. It is also hard to determine the effect of coral
diversity, algal cover and substrate type on fish assemblages until statistical and spatial analysis
are complete.

There may be a stronger correlation between reef fish density and live coral cover since they
are thought to be more sedentary with a range between 10 and 100m2. This is supported by our
results which show total reef fish density increasing in the first half of 2016 (pre-bleaching) but
decreased in the second half (post bleaching). The levels of commercial species (omitting
Scaridae) have seen a slow but gradual increase in density since 2005 with a peak at the end of
2007. The second half of 2015 and 2016 saw an increase in commercial species which could
indicate a seasonal reduction in fishing pressure or the effects of large scale juvenile
recruitment to the population (Engelhardt 2004). Further investigations will be required to
ascertain the cause of this trend.

87
The relationship between coral growth forms and fish is likely to be another key factor in fish
density and diversity since a more complex structure is better able to support overlapping
niches (Gratwicke and Speight 2005; Pratchett et al. 2011). Following this we would expect to
observe highest densities on carbonate sites, which have a higher proportion of branching
corals and tabulate corals compared with granitic sites that support high populations of
encrusting corals. Although granitic sites support a higher fish density overall, the increase in
tabulate corals, massive corals and Porites spp. may explain the increase in parrotfish density,
which in turn accounts for a substantial proportion of the commercial and overall fish densities.
The density of groupers (Serranidae) also rose sharply from 2012 2014. This represents a
substantial recovery in grouper populations which were found to be heavily depleted and, in
many cases, overexploited following the 1998 bleaching event (Engelhardt 2002). Serranids are
known to generally prefer carbonate reefs and often show high habitat affinity (Samoilys 1988;
Engelhardt 2004; Popple and Hunte 2005). Carbonate reefs, particularly those within the Baie
Ternay marine park have shown high levels of recovery from 1998 - 2015, particularly in
branching genera such as Acropora and Pocillopora. Certain grouper species are known to
require the complex reef habitats associated with these branching corals for ambush hunting
and refuge (Engelhardt 2004). It is important to note that even after bleaching and a shift to an
algae dominated environment, the reef still provides the structural complexity required for
ambush hunters.

4.2.2 Trophic spectrum of fish assemblages

Analysis of fish family assemblages over time can be used as a good indicator of the state of
the reef since our species list is selected to include indicator species from different trophic levels
which in turn is a major factor in the structure of biological communities (Bozec et al. 2005;
Gascuel et al. 2005). Our results indicate that the commercial fish are dominated by the
herbivorous parrotfish and rabbitfish; the invertebrate feeding emperors are decreasing, while
piscivorous snappers and groupers are increasing. Detailed temporal analysis of trophic levels
as fractions of the overall community will provide a more defined analysis of overall reef health.

The herbivorous fish assemblage is dominated by counts of small parrotfish individuals.

Interestingly, the presence of herbivorous fish communities operating over a broad range of
spatial scales, therefore a broad range of size classes, is an indicator of reef recovery after a
thermal disturbance (Nash et al. 2016). This is indicative of high response diversity, better than
sole description of species biomass across functional groups. Small individuals should be more
impacted by a bleaching event, however loss of this function is compensated for by an increase

88
in biomass of larger individuals. Considering that counts generally consisted of small individuals
mainly parrotfish individuals, the increase in herbivorous abundance and increasing distance
between herbivorous species and other trophic groups might indicate a frame shift from live
coral to algal dominated communities. An estimation of size would have been useful.

Fish family assemblage and diversity is therefore a key indicator in ecosystem equilibrium and
health. Herbivorous fish abundance and diversity have been shown to be key in supporting reef
resilience by grazing algae and preventing a phase shift to an algal-dominated state (Hughes et
al. 2007; Cheal et al. 2010).

4.2.3 Influence of protected areas

Historically, protected areas were found to have very little influence on fish populations
(Engelhardt 2002). However, since GVI took over monitoring in 2005, higher densities of both
reef and commercial fish have been recorded within the marine parks at Baie Ternay and Port
Launay. This shift in the effectiveness of protection is likely due to the gradual recovery of reefs
within marine parks until the 2016 bleaching event as well as the influence of protection from
fishing.

While protected areas are having a positive impact on the overall fish population, the marine
parks within our study area do have limitations in the recovery of some fish species. The
majority of the sites within the Baie Ternay and Port Launay marine parks are predominantly
sheltered carbonate reef flats surrounded by steep granitic boundaries on the shoreline. As
mentioned above many grouper species have a preference for carbonate reef environments rich
in branching corals (Engelhardt 2004), as such groupers benefit the most from protection of all
the coral families discussed in this report. The majority of the snapper family (Lutjanidae)
however, prefer exposed granitic sites with high levels of water movement and as such gain
little benefit from the MPAs in our study area (Engelhardt 2002), being found in higher densities
outside marine parks. As a result, these species may be more susceptible to the effects of an
increase in fishing pressure. This may be particularly true at Bay Ternay Lighthouse, which
borders the marine park. The site consistently holds aggregations of snappers as it is exposed
to strong currents and wave action, however is often fished by small vessels (personal
observation, unquantified) thereby negating some of the spillover benefits of the park.

The data also reveals that the two marine parks offer differing levels of protection. Baie Ternay
typically holds higher densities of fish than Port Launay. Port Launay is the larger of the two
marine parks so would normally be expected to hold more fish than smaller MPAs. However,

89
the coral cover in Port Launay is lower than that of Baie Ternay, and the water turbidity is often
increased within Port Launay Marine National Park. The morphology of the bay results in high
water flow through a narrow neck, as such there is a high level of mixing and sediment
movement occuring naturally. Nevertheless, a key difference between the two parks is the
presence in Port Launay of a large resort and resultant increase in boat traffic. The frequency
of boat traffic appears to contribute to sustained turbidity (pers. obsv.) that in turn may
negatively impact live coral cover and result in the observed differences in fish densities
between the two MPAs.

The size of the Baie Ternay and Port Launay marine parks may also not be adequate for full
protection of exploited species found within them. Both MPAs are small, 0.8 and 1.58 km2
respectively. The majority of reef fish are thought to be relatively sedentary and may often have
territories or small home ranges between 10 and 100m2. However, many fish associated with
reefs are known to have much larger ranges. For example, grouper species from genera
Cephalopholis (Cephalopholis cruentata) and Plectropomus (Plectropomus leopardus) have
been shown to have average home range sizes of 4 km2 (Popple and Hunte 2005) and 2 km2
(Samoilys 1997) respectively. Larger individuals are thought to move greater distances and
have larger home ranges in order to find food and for mating behaviours (Jones 1991). This
would suggest that larger species of grouper and other fish species are not protected over their
full home ranges or throughout their entire life cycles, which would provide a possible
explanation for why large individuals are rarely recorded on our surveys. In comparison larger
grouper species observed within the Curieuse MPA which has an area of 14.7 km2, e.g.
Epinephelus fasciatus only recorded a maximum size in 2016 of 35 cm (one individual)
compared with regular sightings of <35cm in Curieuse (pers. obs.). Gell and Roberts (2003)
noted the effect of reserve size on the effectiveness of protected areas, they recommend that
the key to success is in matching reserve size to the range of the species that are targeted for
protection; the effective reserve size was estimated to be 16-24 km2 for larger and more mobile
species.

The positive spillover effects of fish populations surrounding MPAs has been documented
(McClanahan and Mangi 2000; Gell and Roberts 2003). This effect is hard to determine from
our results until further statistical analysis has been completed. Sites adjacent to Baie Ternay
Baie Ternay Lighthouse and Site Y do often support relatively high densities of fish, however it
is not clear if this is statistically significant to sites further afield. In addition the environmental
conditions within Baie Ternay and potential spillover sites can be substantially different,

90
particularly Lighthouse, which is an exposed granitic site subject to strong stirring currents.
Other sites with similar conditions such as Conception North Point, Whale Rock and Therese
North End all support similar fish densities. Without further study it is impossible to make any
firm conclusions on whether the change in environment or the proximity to MPAs is the driving
factor in governing fish densities at these sites.

4.2.4 Commercial fish size analysis

Size data for commercial reef fish species needs further analysis to be discussed properly and
assess significant trends over the years. Nevertheless, individuals recorded for the families
Haemulidae, Lutjanidae and Serranidae are mostly juveniles. Mean length at maturity for those
families is 40cm, 33.8 and 35.4cm respectively (Froese and Pauly 2016), whereas the size
classes with highest density are 21-30 cm for Haemulidae,11-30cm for Lutjanidae and
Serranidae. This trend might suggest the area is a preferred nursery site, which requires further
analysis per site for confirmation. Overfishing of sweetlips (Haemulidae), groupers (Serranidae)
and snappers (Lutjanidae) might also be an explanation, where fishing selects larger individuals
therefore impacting fish populations by increasing density of smaller fish and also reducing
genetic pool (Law 2007).

4.2.4 Fish data in a nutshell

In summary, the density of total fish populations continued to increase to an all time high in
2016. The study areas support a higher density of reef fish than commercial fish, however the
commercial fish species are currently increasing at a faster rate than reef species. The reasons
for these trends is likely to be multifaciated and linked to a decrease in live coral cover and
increase in algal coverage in the second half of 2016, fishing pressure, and other environmental
factors. The proportional representation of family groups and feeding guilds has shifted and
could be an indicator of a change in habitat composition however the trend is difficult to
determine: the proportion of coralivorous butterfly fish has decreased; although total numbers of
herbivorous parrotfish and rabbitfish have increased, the proportion of parrotfish have
decreased; yet the proportion of invertivorous wrasses have increased. Further study and
statistical analysis is required to produce more conclusive evidence on the evolution of the fish
assemblage after the bleaching event and impact on the wider ecosystem. Protected areas
support higher densities of fish than non-protected areas, however the results differ between
species and it is probable that due to the size and usage of the MPAs, densities are dictated
more by live coral cover, structural complexity and water quality than management efforts.

91
Protection afforded by the two marine parks differs as higher fish densities are recorded in Baie
Ternay, which is likely linked to the presence of a large resort in Port Launay and associated
effects of the construction and increased boat traffic. The effects of large resorts should be
considered for any further developments in or around MPAs. The importance of protection
offered from MPAs to spillover sites is yet to be determined, pending further study, however it is
probable that Baie Ternay is an important nursery area for fish due to its structural complexity.

4.3 Invertebrates

Invertebrates have been studied as biological indicators within terrestrial and aquatic
ecosystems extensively, including coral reef habitats. Their importance lies in their interactions
with the reef habitat, and density may reflect changes in reef composition and structure.
Densities of surveyed invertebrates from the 10m belt transects are higher than the densities
observed in the beginning of the monitoring in 2005, except black spine urchins which were
observed at densities very similar to that of 2005. The 2016 results show an overall decrease in
all surveyed invertebrates from 2015 with the exception of black spine urchins and
Platyhelminthes. Importantly, Platyhelminthes show low densities due to their lifestyle; these
species are generally nocturnal and found mostly under the rock and rubble of the reef
(Coleman, 2000), but are also hard to spot as most species are small, cryptic and camouflaged.
The decrease in all other invertebrate taxa may be as a result of the strong bleaching event that
occurred in late 2015 to early 2016 (pers. obs.), which lead to reduction of coral cover (31%
decrease in 2016). This large decline in coral would naturally lead to a reduction of certain
invertebrates, which have a high affinity with the coral through a loss of food, productivity,
rugosity and other coral services (Heck and Wetstone 1977). Conversely, the increase of sea
urchins could also be as a result of the bleaching event increasing the abundance of algae or
other food sources.

When comparing the granitic and carbonate sites the results for each invertebrate phyla follow
very similar trends, with the exception of the Arthropoda phyla, which decreased in density on
granitic and increased on carbonitic sites. Anecdotally, the most common arthopods observed
are the coral shrimp and coral crabs such as Periclimenes spp. and Trapezia spp., which live in
the interstitial spaces of branching corals such as Acropora and Pocillopora spp. This
discrepancy could be explained by the coral cover results above, which show that granitic sites

92
lost their structural complexity much sooner after the bleaching event than carbonitic sites.
Interestingly, overall Arthropoda densities across all sites did decline from 2015-2016; yet
remained at higher levels in comparison to the other phyla being the most commonly observed
invertebrates during the surveys. This high density of arthropods may reflect the fact that coral
can often stay structurally intact over long periods of time after a bleaching event, providing
habitat for the organisms inhabiting the coral (Halford and Caley 2009). This may provide a
buffer in the Arthropoda density leading to a subsequent lag effect decline in density (Halford
and Caley 2009). The lag effect will likely be a common issue in all coral dependent
invertebrates if the coral reef does not return to pre bleaching conditions.

The survey list for invertebrates on the 50m belts focuses on commercially important
invertebrates and key species, which indicate ecosystem change. Drupella spp., short spine
urchins and long spine urchins were the most commonly observed invertebrate taxa, excluding
the sea cucumbers. Looking at the densities between the two survey periods January-June and
July-December across all sites, the majority of invertebrate taxa densities are consistent across
the two survey periods, except for Drupella spp., long spine urchins and short spine urchins.
The surveys conducted in the January-June period were during or directly after the bleaching
event, whereas the July-December period was months after the bleaching event and the reef
had become heavily degraded with a high proportion of dead coral (see Results section 2.1).
The Drupella spp., an obligate corallivore, experienced large declines in the July-December
period when compared to the January-June period which may reflect the sharp declines in live
coral cover experienced after the bleaching event (Wilmes 2014). Conversely, after a bleaching
event Diadema spp. typically increase in density, as a result of increased algal growth (Dudgeon
et al. 2010) which may explain the increase in the July-December period.

When viewing the invertebrates by substrate it appears that short spine urchins and other sea
stars are more predominant on granitic sites. Whereas Drupella spp., long spine urchins, pencil
urchins and Mathaes urchins were more typically found on carbonitic sites.

Focusing on the black spine sea urchins across the years, short spine Echinothrix spp. have
relatively increased from 2015 levels, but overall seem relatively stable in comparison to
previous years. Long spine urchins, Diadema spp. seemed to have declined on carbonitic sites
from 2015 densities, which is abnormal considering the bleaching event, where it would be
expected to have increased in population density. This unexpected trend may be as a result of

93
the Diadema spp. typically being quite cryptic on rugged carbonate sites, however this needs
further investigation. Long spine sea urchins Echinothrix spp. increased on granitic sites from
2015, however they are still below that of observed densities pre 2012.

Drupella spp. experienced a large decline from 2015 levels. As mentioned before, this may be
as a result of their strong linkage to the corals, which experienced declines after the bleaching
event (Wilmes 2014). Interestingly, other sea stars continued to increase from the 2015 levels
which were already above the typical densities observed before, however the other sea star
taxa includes numerous sea stars, all with various different feeding styles, behaviours and
responses to bleaching events, making it inappropriate to form any simple conclusions. It would
be suggested that this increase in other sea star densities should be further investigated by
future studies. Mathaes urchins, Echinometra mathaei, continued to decline from 2014 levels,
however, the drivers behind this trend are unclear and need to be further addressed.

Sea cucumbers appear to have recovered from the dramatic decline of 2015, however this
decline may have be a function of the limited surveys conducted in 2015 not being as
representative as the surveys conducted in 2016. However, the overall abundance of sea
cucumbers is still lower than pre 2014 densities. As found in previous years, sea cucumbers
were typically encountered at higher densities on granitic sites than carbonate reefs. The two
most commonly observed taxa were the Stichopus spp. and Pearsonothuria graeffei.
Stichopus sp. continued to decline from 2015 densities to levels lower than when surveys
began, however when comparing January-June to July-December, a much lower density was
observed on granitic sites, which has, in previous reports, been the substrate on which
Stichopus spp. is most commonly found. Conversely, Pearsonothuria graeffei was found at
much higher densities on granitic sites in January-June than in July-December, which may
reflect the loss of habitat on carbonitic sites after the bleaching event. Overall, the
Pearsonothuria graeffei densities increased from 2015, to the highest densities observed since
the surveys began in 2009.

94
5. Additional ecosystem monitoring

5.1 Turtles

There are seven species of turtles worldwide, five of which breed in the Western Indian Ocean:
the leatherback - Dermochelys coriacea, loggerhead - Caretta caretta, olive ridley -
Lepidochelys olivacea, hawksbill - Eretmochelys imbricata, and green - Chelonia mydas (Hewell
2011). The leatherback, loggerhead and olive ridley, although common to parts of the Western
Indian Ocean, are not thought to currently nest in the Seychelles and are rarely seen. In
contrast, the hawksbill and green are residents in coastal waters of the Seychelles, nest on the
beaches, and are commonly observed at dive sites (Diamond 1976). All five species found in
the Seychelles face the combined threats of poaching, pollution and loss of nesting sites
(Diamond 1976), and are listed by IUCN as endangered or critically endangered (Mortimer and
Bresson 1999). Seychelles is considered one of the most important sites for the critically
endangered hawksbill turtle (Spotila 2004) and is one of the few localities in the world where
they can be observed nesting during daylight hours (Mortimer and Bresson 1999).

GVI staff and volunteers are trained in turtle biology and the identification of the two species
commonly seen around north-west Mah, the green (Chelonia mydas) and the hawksbill turtle
(Eretmochelys imbricata). Currently, recordings on incidental sightings from megafauna data are
the only records kept on turtles.

Incidental Turtle Sightings

For every dive undertaken by GVI, a record of megafauna observations is kept of which turtle
encounters are included. The parameters for each of GVIs dives are logged, regardless of
whether or not a turtle was seen, enabling the calculation of turtle frequency per dive and thus
effort-related abundance. The species, carapace length, sex, distinguishing features and
behaviour of all turtles sighted is recorded wherever possible.

Photo Identification of Turtles

Turtle photo identification has been carried out by GVI staff and volunteers until 2013 and was
implemented again in January 2016. All divers with a camera are encouraged to provide any
turtle photos they have taken throughout their time in Cap Ternay. The post-ocular area of
scales on the left and right cheeks of both hawksbill and green turtles are unique to each
individual, allowing for comparisons to be made between identification shots taken on different

95
dives. Individuals can be recognised through analysis of the photographs, based on a code
defined from the localisation and the number of scale of each side of the head profile. This is
aided through the use of the software I3S that, when enough data has been input, can analyse
the photos using its own database.

Results and discussion

Incidental sightings of sea turtles are divided into three month periods to more accurately view
the fluctuations that occur in and outside of nesting season (Table 2). The greatest number of
turtle sightings for 2016 were recorded between January and March, with 110 turtles recorded
(89 Hawksbill turtles and 21 Green turtles), in contrast with 2015 where most turtle sightings
were recorded between July and September (107 turtles). April to June and July to September
had the least sightings; with 23 and 19 turtles recorded for each three month period
respectively. As data able to be analysed is available from April 2015 only, this data still
represents a baseline for any future turtle monitoring activities.

Table 2: Total number of turtle sightings recorded for each dive logged in 2016.

Jan-March Apr-Jun Jul-Sep Oct-Dec

Number of dives 151 171 147 166
Number of turtles 110 23 19 58
Encounter rate per dive (%) 72.85 13.45 12.93 34.94

Table 3: Total number of hawksbill and green turtle sightings recorded for each dive logged in 2016.

Species Jan-March Apr-Jun Jul-Sep Oct-Dec

Hawksbill 89 19 17 45
Green 21 2 1 9
Unidentified 0 2 1 4
Total 110 23 19 58

I3S database

The turtle identification database was started in July 2016 and photos from 2016 back to 2004
have been integrated, accounting for over 450 different sightings of hawksbill and green turtles
and displaying several matches among the years and sites. Unfortunately, an error occurred

96
with the software and the created database is currently inaccessible; this project is therefore on
hold at the moment.

5.2 Crown-of-thorns sea stars (COTs)

Outbreaks of the coral predator, the crown-of-thorns sea star (Acanthaster planci) were first
reported in 1996 and were active until 1998, when the reefs suffered from the bleaching-induced
coral mortality (Engelhardt 2004). Normal density levels are less than one individual per hectare
(Pratchett 2007) and in these numbers A. planci can benefit coral diversity by feeding on the
faster growing corals such as Acropora and Pocillopora; its preferred prey genera (Pratchett
2007) and early colonisers of degraded reefs that out-compete the slower growing corals (Veron
2000). In high numbers however, the level of competition for food drives the sea star to eat any
species of corals which can severely degrade reefs and reduce coral cover to as little as 1%
(CRC Reef Research Centre 2003). The causes of outbreaks are still not completely
understood; it may be connected to overfishing of A. planci predators, such as the giant triton
shell, which is popular with shell collectors, or to natural fluctuations (CRC Reef Research
Centre 2003). The most widely accepted theory is that increased nutrient levels in the ocean
could be an influential factor, originating from agricultural, domestic or industrial sources. A.
planci are surveyed as part of the invertebrate abundance and diversity belts and incidental
sightings are also documented after every dive.

Incidental sightings

In 2016 most COT sightings have been recorded from July to September with 180 individuals.
Lowest number of COTs was recorded from in the January-March period (40 individuals) (Table
4). This data reflects the number of COTs injected during the eradication program (Figure 64);
however this data cant be analyzed further as it is qualitative and not standardized (data
recorded is dependent on various factors e.g. divers, site, dive purpose).

Table 4. Total number of Crown-of-Thorns sightings recorded for each dive logged in 2016.

Jan-Mar Apr-Jun Jul-Sep Oct-Dec

Crown-of-thorns 40 50 180 163

Eradication program

GVI has a COTS eradication program, undertaken when COTs are been observed at surveyed
dive sites. The control dives have been completed by both staff and trained volunteers. Bile

97
salts and sodium bisulphate were previously used for COTs removal, with the former requiring a
single injection to the COT and the latter requiring multiple injections, dependent on the size of
the COT. In 2016, however, vinegar was used as a replacement, which was just as effective,
cheaper and more readily available (Moutardier et al. 2015). The injections were given via a
modified drench gun, which connects to the injection solution (vinegar). As of now, this is the
best-tested method for removing COTs in situ and minimising damage to coral and the
surrounding environment (Rivera Posada and Prachett 2012).

The dedicated COT removal dives were undertaken in buddy pairs spaced along an arranged
target area of a dive site in order to maximize the coverage area. Data was recorded on the total
divers, dive time, depth ranges, total COTs and the size class of the COT (<15 cm, 15-25 cm,
25-40 cm and >40 cm). The Catch per Unit Effort (CPUE) is calculated by:

Total number of COTs injected / Dive time (minutes)

According to the Great Barrier Reef Authority (GBRMPA 2014), a CPUE of 0.05 is a COT
outbreak when the coral cover for the area is below 40%. For areas exceeding coral coverage
of 40% an outbreak is recognised when the CPUE exceeds 0.1. All sites at which COT
eradication dives have been carried out never had a coral cover exceeding 40%. The
eradication dives resulted in a total of 130 injected COTS during 2016, however from April to
September no dives were undertaken. November had the most COTS injections with 36 COTs
removed (Figure 64). As eradication sites were chosen based on previous sightings of COTs
populations and proximity to survey sites, it would therefore be bias to determine if any one site
had a higher COTs population as they were not surveyed at random. The CPUE for 2016 was
calculated as 30.26, well above the GBRMPA outbreak standard of 0.1 for sites above 40%
cover, indicating that there is a COTs outbreak within the Cap Ternay Marine National Park.
An area of concern when calculating the CPUE in accordance to GBRMPA is apparent with the
new, faster methodologies becoming more prevalent, such as the case with the drench gun
vinegar shots. The time needed to give the dosage with the vinegar is greatly reduced in
comparison to the Sodium Bisulphate shots, which is one shot compared to a maximum of 18
injections to one COT; meaning that more time can be used to find and inject COT. Therefore, a
new methodology may be needed to align the different methods in order to create a more robust
CPUE.

98
Figure 64. Number of COTs and the CPUE from January to December 2016.

5.3 Sightings of other fauna

Since 2008, various fauna sighted during every dive undertaken by GVI has been recorded.
This data is purely qualitative and should be taken cautiously as recordings are not systematic
nor standardized, thus vary according to many factors e.g. visibility, the nature of the dive
(training or survey), observer. 2016 data is presented for sharks, rays, lobster and octopuses
sightings divided into three month periods to investigate seasonal fluctuations (Table 5).

Interestingly, the greatest number of shark sightings for 2016 were recorded between January
and March with 65 individuals encountered, followed by October-December with 54 individuals
and lowest rate of encounters from April-June and July-September (respectively 34 and 38
individuals). It might be a seasonal fluctuation, however this pattern needs to be confirmed with
previous and subsequent years. The white tip reef shark (Triaenodon obesus) is the most
observed species, accounting for over 95% of all sightings (Table 5).

Spotted eagle rays (Aetobatus narinari) are the most recorded with 23, 18 and 33 sightings in
January-March. April-June and October-December respectively; in July-September devil rays
(Mobula japonica) were the most sighted with 26 and recordings. Overall, rays are mostly

99
observed from July-September to August-December with 60 and 78 sightings respectively
(Table 5).

October-December was the period with most common reef octopus (Octopus cyanea) sightings
in 2016 (37 recordings). Other periods recorded very few individuals (3 to 5 per 3-months
period) (Table 5). This might be explained by the lack of standardization in data collection.
Painted lobster (Panulirus versicolor) are most recorded in July-September and January-April
with 24 and 18 sightings recorded respectively (Table 5). This data still has to be compared with
previous and subsequent years to assess any potential trend.

Table 5. Number of sharks, rays, lobster and octopuses recorded for each dive logged in in 2016. Note
that number of dives has been omitted, as data is only exploratory and number of dives has been fairly
constant over the year.

Jan-Mar Apr-Jun Jul-Sep Oct-Dec

Whitetip reef shark 62 32 36 51

Other sharks 3 2 2 3
Total sharks 65 34 38 54

Devil ray 4 5 26 29
Spotted eagle ray 23 18 14 33
Feathertail ray 4 3 6 2
Other rays 5 1 14 14
Total rays 36 27 60 78

Common reef octopus 3 5 3 37

Painted lobster 18 8 24 3

100
References
Babcock RC, Baird AH, Piromvaragorn S, Thomson DP, Willis BL (2003) Identification of
scleractinian coral recruits from Indo-Pacific reefs. Zoological Studies 42: 211226.

Babcock R, Mundy C (1996) Coral recruitment: Consequences of settlement choice for early
growth and survivorship in two scleractinians. Journal of Experimental Marine Biology and
Ecology 206: 179201. doi:10.1016/s0022-0981(96)02622-6.

Baird AH, Marshall PA (2002) Mortality, growth and reproduction in scleractinian corals following
bleaching on the Great Barrier Reef. Marine Ecology Progress Series 237: 133141.
doi:10.3354/meps237133.

Bell JD, Galzin R (1984) Influence of live coral cover on coral-reef fish communities. Marine
Ecology Progress Series 15: 265274.

Birkeland C (1977) The importance of rate of biomass accumulation in early successional

stages of benthic communities to the survival of coral recruits. Proc 3rd Int Coral Reef
Symposium 1: 1521.

Bozec Y-M, Kulbicki M, Chassot E, Gascuel D (2005) Trophic signature of coral reef fish
assemblages: towards a potential indicator of ecosystem disturbance. Aquatic Living
Resources 18(2): 103109. doi:10.1051/alr:2005013.

Browne NK, Tay J, Todd PA (2015) Recreating pulsed turbidity events to determine coral
sediment thresholds for active management. Journal of Experimental Marine Biology and
Ecology 466: 98109. doi:10.1016/j.jembe.2015.02.010.

Chabanet P, Dufour V, Galzin R (1995) Disturbance impact on reef fish communities in Reunion
Island (Indian Ocean). Journal of Experimental Marine Biology and Ecology 188(1): 2948.
doi:10.1016/0022-0981(94)00184-F.

Chabanet P, Ralambondrainy H, Amanieu M, Galzin R (1997) Relationships between coral reef

substrata and fish. Coral Reefs 16(2): 93102. doi:http://dx.doi.org/10.1007/s003380050063.

Chadwick NE, Morrow KM (2011) Competition Among Sessile Organisms on Coral Reefs. In: Z
Dubinsky, N Stambler eds. Coral Reefs: An Ecosystem in Transition. Springer Netherlands,
pp.347371. http://link.springer.com/chapter/10.1007/978-94-007-0114-4_20.

101
Cheal AJ, MacNeil MA, Cripps E, Emslie MJ, Jonker M, Schaffelke B, Sweatman H (2010)
Coralmacroalgal phase shifts or reef resilience: links with diversity and functional roles of
herbivorous fishes on the Great Barrier Reef. Coral Reefs 29(4): 10051015.
doi:10.1007/s00338-010-0661-y.

Coleman N (2000) Marine Life of the Maldives. Atoll Editions edition. James Bennett Pty
Ltd:Apollo Bay, Australia, 328pp.

Colvocoresses J, Acosta A (2007) A large-scale field comparison of strip transect and stationary
point count methods for conducting length-based underwater visual surveys of reef fish
populations. Fisheries Research 85(12): 130141. doi:10.1016/j.fishres.2007.01.012.

CRC Reef Research Centre (2003) Crown-of-thorns starfish on the Great Barrier Reef: current
state of knowledge. CRC Reef Research Centre, Townsville, Australia.

Cruz ICS, Meira VH, Kikuchi RKP de, Creed JC (2016) The role of competition in the phase shift
to dominance of the zoanthid Palythoa cf. variabilis on coral reefs. ResearchGate 115:
doi:10.1016/j.marenvres.2016.01.008.

Diamond AW (1976) Breeding biology and conservation of Hawksbill turtles, Eretmochelys

imbricata L., on Cousin Island, Seychelles. Biological Conservation 9(3): 199215.
doi:10.1016/0006-3207(76)90010-0.

Dudgeon SR, Aronson RB, Bruno JF, Precht WF (2010) Phase shifts and stable states on coral
reefs. ResearchGate 413: 201216. doi:10.3354/meps08751.

Engelhardt U (2001) Report on scientific field studies and training activities conducted in June /
July 2001. Seychelles Marine Ecosystem Management Project. Global Environment
Facility/Government of Seychelles/World Wildlife Fund, Interim Report 1, Victoria.

Engelhardt U (2002) Ecological characteristics of scleractinian hard coral communities 4 years

after the 1998 mass coral bleaching event. Seychelles Marine Ecosystem Management Project.
Global Environment Facility/Government of Seychelles/World Wildlife Fund, Interim Report 3,
Victoria.

Engelhardt U (2004) The status of scleractinian coral and reef-associated fish communities 6
years after the 1998 mass coral bleaching event. Seychelles Marine Ecosystem Management
Project. Global Environment Facility/Government of Seychelles/World Wildlife Fund, Victoria.

Froese R, Pauly D (2016) FishBase. World Wide Web electronic publication. www.fishbase.org.

102
Gascuel D, Bozec Y-M, Chassot E, Colomb A, Laurans M (2005) The trophic spectrum: theory
and application as an ecosystem indicator. ICES Journal of Marine Science 62(3): 443452.
doi:10.1016/j.icesjms.2004.12.013.

GBRMPA (2014) Crown-of-thorns starfish control guidelines. Great Barrier Reef Marine Park
Authority, Report, Townsville, Australia.

Gell FR, Roberts CM (2003) Benefits beyond boundaries: the fishery effects of marine reserves.
Trends in Ecology & Evolution 18(9): 448455. doi:10.1016/S0169-5347(03)00189-7.

Graham NAJ, Wilson SK, Jennings S, Polunin NVC, Bijoux JP, Robinson J (2006) Dynamic
fragility of oceanic coral reef ecosystems. Proceedings of the National Academy of Sciences
103(22): 84258429. doi:10.1073/pnas.0600693103.

Gratwicke B, Speight MR (2005) The relationship between fish species richness, abundance
and habitat complexity in a range of shallow tropical marine habitats. Journal of Fish Biology
66(3): 650667. doi:10.1111/j.0022-1112.2005.00629.x.

Guest JR, Low J, Tun K, Wilson B, Ng C, Raingeard D, Ulstrup KE, Tanzil JTI, Todd PA, Toh
TC, McDougald D, Chou LM, Steinberg PD (2016) Coral community response to bleaching on a
highly disturbed reef. Scientific Reports 6: 20717. doi:10.1038/srep20717.

Halford AR, Caley MJ (2009) Towards an understanding of resilience in isolated coral reefs.
Global Change Biology 15(12): 30313045. doi:10.1111/j.1365-2486.2009.01972.x.

Heck KL, Wetstone GS (1977) Habitat Complexity and Invertebrate Species Richness and
Abundance in Tropical Seagrass Meadows. Journal of Biogeography 4(2): 135142.
doi:10.2307/3038158.

Hennige SJ, Smith DJ, Walsh S-J, McGinley MP, Warner ME, Suggett DJ (2010) Acclimation
and adaptation of scleractinian coral communities along environmental gradients within an
Indonesian reef system. Journal of Experimental Marine Biology and Ecology 391(12): 143
152. doi:10.1016/j.jembe.2010.06.019.

Hewell K (2011) Class Reptilia - Reptiles. In: A Field Guide to the Seashores of Eastern Africa
and the Western Indian Ocean Islands. SIDA; 3rd edition, p.464.

Hill J, Wilkinson C (2004) Methods for Ecological Monitoring of Coral Reefs. Australian Institute
of Marine Science:Townsville, Australia.

103
Hughes TP, Rodrigues MJ, Bellwood DR, Ceccarelli D, Hoegh-Guldberg O, McCook L,
Moltschaniwskyj N, Pratchett MS, Steneck RS, Willis B (2007) Phase Shifts, Herbivory, and the
Resilience of Coral Reefs to Climate Change. Current Biology 17(4): 360365.
doi:10.1016/j.cub.2006.12.049.

Jennings S, Marshall S., Polunin NV. (1996) Seychelles marine protected areas: comparative
structure and status of reef fish communities. Biological Conservation 75: 20120.

Jennings S, Polunin NVC (1997) Impacts of predator depletion by fishing on the biomass and
diversity of non-target reef fish communities. Coral Reefs 16(2): 7182.
doi:10.1007/s003380050061.

Jones G (1991) Postrecruitment Processes in the Ecology of Coral Reef Fish Populations: A
Multifactorial Perspective. In: The ecology of fishes on coral reefs. Academic Press, pp.294
328.

Komyakova V, Munday PL, Jones GP (2013) Relative Importance of Coral Cover, Habitat
Complexity and Diversity in Determining the Structure of Reef Fish Communities. PLOS ONE
8(12): e83178. doi:10.1371/journal.pone.0083178.

Kuguru BL, Mgaya YD, hman MC, Wagner GM (2004) The reef environment and competitive
success in the Corallimorpharia. Marine Biology 145(5): 875884. doi:10.1007/s00227-004-
1376-9.

Kulbicki M (1998) How the acquired behaviour of commercial reef fishes may influence the
results obtained from visual censuses. Journal of Experimental Marine Biology and Ecology
222(12): 1130. doi:10.1016/S0022-0981(97)00133-0.

Law R (2007) Fisheries-induced evolution: present status and future directions. Marine Ecology
Progress Series 335: 271277. doi:10.3354/meps335271.

Loya Y, Sakai K, Yamazato K, Nakano Y, Sambali H, van Woesik R (2001) Coral bleaching: the
winners and the losers. Ecology Letters 4(2): 122131. doi:10.1046/j.1461-0248.2001.00203.x.

McClanahan TR, Baird AH, Marshall PA, Toscano MA (2004) Comparing bleaching and
mortality responses of hard corals between southern Kenya and the Great Barrier Reef,
Australia. Marine Pollution Bulletin 48(34): 327335. doi:10.1016/j.marpolbul.2003.08.024.

McClanahan TR, Mangi S (2000) Spillover of Exploitable Fishes from a Marine Park and Its
Effect on the Adjacent Fishery. Ecological Applications 10(6): 17921805. doi:10.1890/1051-
0761(2000)010[1792:SOEFFA]2.0.CO;2.

104
Mortimer JA, Bresson R (1999) Temporal distribution and periodicity in hawksbill turtles
(Eretmochelys imbricata) nesting at Cousin island, Republic of Seychelles, 1971-1997.
Chelonian Conservation and Biology 3: 19711997.

Moutardier G, Gereva S, Mills SC, Adjeroud M, Beldade R, Ham J, Kaku R, Dumas P (2015)
Lime Juice and Vinegar Injections as a Cheap and Natural Alternative to Control COTS
Outbreaks. PLOS ONE 10(9): e0137605. doi:10.1371/journal.pone.0137605.

Nash KL, Graham NAJ, Jennings S, Wilson SK, Bellwood DR (2016) Herbivore cross-scale
redundancy supports response diversity and promotes coral reef resilience. Journal of Applied
Ecology 53(3): 646655. doi:10.1111/1365-2664.12430.

Nystrm M, Folke C, Moberg F (2000) Coral reef disturbance and resilience in a human-
dominated environment. Trends in Ecology & Evolution 15(10): 413417. doi:10.1016/S0169-
5347(00)01948-0.

Palacios M del M, Zapata FA (2014) Fish community structure on coral habitats with contrasting
architecture in the Tropical Eastern Pacific. Revista de Biologa Tropical 62: 343357.

Popple ID, Hunte W (2005) Movement patterns of Cephalopholis cruentata in a marine reserve
in St Lucia, W.I., obtained from ultrasonic telemetry. Journal of Fish Biology 67(4): 981992.
doi:10.1111/j.0022-1112.2005.00797.x.

Pratchett MS (2007) Feeding Preferences of Acanthaster planci (Echinodermata: Asteroidea)

under Controlled Conditions of Food Availability. Pacific Science 61(1): 113120.

Pratchett MS, Hoey AS, Wilson SK, Messmer V, Graham NAJ (2011) Changes in Biodiversity
and Functioning of Reef Fish Assemblages following Coral Bleaching and Coral Loss. Diversity
3(3): 424452. doi:10.3390/d3030424.

Riegl B, Glynn PW, Wieters E, Purkis S, d Angelo C, Wiedenmann J (2015) Water column
productivity and temperature predict coral reef regeneration across the Indo-Pacific. Scientific
Reports 5: 8273. doi:10.1038/srep08273.

Rivera Posada J, Prachett M (2012) A review of existing control efforts for Acanthaster planci:
limitations to successes. NERP, Tropical Environmental Hub, Report, Townsville, QLD,
Australia. http://researchonline.jcu.edu.au/34548/.

Roberts CM, Ormond RFG (1987) Habitat complexity and coral reef diversity and abundance on
Red Sea fringing reefs. Marine Ecology Progress Series 41: 18.

105
Samoilys M (1988) Abundance and species richness of coral reef fish on the Kenyan coast: the
effects of protective management and fishing. Proceedings of the 6th International Coral Reef
Symposium, Townsville 2: 261266.

Samoilys MA (1997) Movement in a large predatory fish: coral trout, Plectropomus leopardus
(Pisces: Serranidae), on Heron Reef, Australia. Coral Reefs 16(3): 151158.
doi:10.1007/s003380050069.

Samoilys M, Gribble N (1997) Manual for assessing fish stocks on Pacific coral reefs. Qld.:
Dept. of Primary Industries, Brisbane.

Silverstein RN, Correa AMS, Baker AC (2012) Specificity is rarely absolute in coral-algal
symbiosis: implications for coral response to climate change. Proceedings Biological Sciences
279(1738): 26092618. doi:10.1098/rspb.2012.0055.

Silverstein RN, Cunning R, Baker AC (2015) Change in algal symbiont communities after
bleaching, not prior heat exposure, increases heat tolerance of reef corals. Global Change
Biology 21(1): 236249. doi:10.1111/gcb.12706.

Spalding MD, Jarvis GE (2002) The impact of the 1998 coral mortality on reef fish communities
in the Seychelles. Marine Pollution Bulletin 44(4): 309321.

Spencer T, Teleki KA, Bradshaw C, Spalding MD (2000) Coral Bleaching in the Southern
Seychelles During the 19971998 Indian Ocean Warm Event. Marine Pollution Bulletin 40(7):
569586. doi:10.1016/S0025-326X(00)00026-6.

Spotila JR (2004) Sea Turtles: A Complete Guide to Their Biology, Behavior, and Conservation.
JHU Press, 254pp.

Veron JEN (2000) Corals of the world. Vol. 1-3. M. Stafford-Smith edition. Australian Institute of
Marine Science:Townsville, 1382pp.

Wibowo K, Abrar M, Siringoringo RM (2016) Trophic Status of Reef Fishes and Correlation
between Herbivorous Fishes and Coral Recruitment in Pari Island Waters, Jakarta Bay.
Oseanologi dan Limnologi di Indonesia 1(2): 7389.

Wilmes J (2014) The Crown-of-Thorns Starfish: Biology, Outbreaks and Control Methods.
Association of Marine Park Tourisms Operators.

106
Appendix A
A.1. Coral genera (LIT, coral diversity belts and recruitment quadrat surveys)

Family Genus Family Genus

Acropora Cycloseris
Acroporidae Astreopora Diaseris
Montipora Fungidae Fungia
Coeloseris Herpolitha
Gardineroseris Podabacia
Agariciidae Leptoseris Hydnophora
Merulinidae
Pachyseris Merulina
Pavona Acanthastrea
Astrocoeniidae Stylocoeniella Blastomussa
Mussidae
Dendrophyllidae Turbinaria Lobophyllia
Euphyllidae Physogyra Symphyllia

Cyphastrea Oculinidae Galaxea

Diploastrea Echinophyllia
Echinopora Pectiniidae Mycedium
Favia Pectinia
Favites Pocillopora
Faviidae Goniastrea Pocilloporidae Seriatopora
Leptastrea Stylophora
Leptoria Alveopora
Montastrea Poritidae Goniopora
Oulophyllia Porites
Platygyra Psammocora
Plesiastrea Siderastreidae Pseudosiderastrea
Siderastrea

Note: The genus Montastrea (Faviidae) has included 3 species since 2009 (start of recording coral genera). One of
those species belongs to the genus Favia (Faviidae), but due to strong similarities has been included in the genus
Montastrea to simplify the identification process. The genus Blastomussa (Mussidae) also includes one species from
the genus Acanthastrea (Mussidae). Considering this has been systematic since 2009, identification procedure has
not been changed so as not to skew the data.

A.2. Benthic organisms and substrate types (LIT surveys)

Algae Macro algae, turf algae, Halimeda sp., coralline algae, algae assemblage

Other lifeforms recorded
Organisms recorded as other
Substrate
Heliopora sp., Millepora sp., Tubipora sp., soft coral, sponge, corallimorphs,
zoanthids
Bryozoans, ascidians, bivalves, anemones, black corals, gorgonians, sea Stars,
holothurians
Sand, rubble, rock, dead coral, silt, water

107
Appendix B
B.1. Fish species list

Family Scientific name Common name Feeding guild Relevance*

Chaetodon vagabundus Vagabond C/I Coral recovery

Chaetodon auriga Threadfin C/I Coral recovery

Chaetodon trifascialis Chevroned C Coral recovery

Chaetodon melannotus Black-backed C/I Coral recovery

Chaetodon mertensii Merten's C/I Coral recovery

Chaetodon triangulum Triangular C Coral recovery

Chaetodon trifasciatus Indian redfin C Coral recovery

Chaetodon interruptus Indian Ocean teardrop C/I Coral recovery

Chaetodon bennetti Bennett's C Coral recovery

Butterflyfish
Chaetodon lunula Raccoon C/I Coral recovery
(Chaetodontidae)
Chaetodon kleinii Klein's C/I Coral recovery

Chaetodon citrinellus Speckled C/I Coral recovery

Chaetodon guttatisimus Spotted C/I Coral recovery

Chaetodon lineolatus Lined C/I Coral recovery

Chaetodon falcula Saddleback C/I Coral recovery

Chaetodon meyersi Meyer's C Coral recovery

Chaetodon xanthocephalus Yellow-headed C/I Coral recovery

Chaetodon zanzibariensis Zanzibar C Coral recovery

Forcipiger sp. Longnose C/I Coral recovery

Apolemichthys trimaculatus Three-spot V Visual appeal

Angelfish Pomacanthus imperator Emperor V Visual appeal

(Pomacanthidae) Pomacanthus semicirculatus Semicircle V Visual appeal

Pygoplites diacanthus Regal V Visual appeal

Acanthurus sp. Surgeonfish H Algae vs. coral

Surgeonfish
Ctenochaetus sp. Bristletooths H Algae vs. coral
(Acanthuridae)
Naso sp. Unicornfish Pl Algae vs. coral
Moorish Idol
Zanclus cornutus Moorish idol V Visual appeal
(Zanclidae)
Siganus puelloides Blackeye H Algae vs. coral
Rabbitfish
Siganus corallinus Coral H Algae vs. coral
(Siganidae)
Siganus stellatus Honeycomb H Algae vs. coral

108
 Siganus argenteus Forktail H Algae vs. coral

Siganus sutor African whitespotted H Algae vs. coral

Lutjanus gibbus Paddletail Pi Fishing pressure

Lutjanus sebae Red emperor Pi Fishing pressure

Lutjanus fulviflamma Longspot Pi Fishing pressure

Lutjanus kasmira Blue-lined Pi Fishing pressure

Lutjanus bengalensis Bengal Pi Fishing pressure

Lutjanus monostigma Onespot Pi Fishing pressure

Snappers
Lutjanus vitta Brownstripe Pi Fishing pressure
(Lutjanidae)
Lutjanus fulvus Flametail Pi Fishing pressure

Lutjanus argentimaculatus Mangrove jack Pi Fishing pressure

Lutjanus bohar Red Pi Fishing pressure

Lutjanus russelli Russell's Pi Fishing pressure

Macolor niger Black Pi Fishing pressure

Aprion virescens Green jobfish Pi Fishing pressure

Balistoides viridescens Titan I Sea urchins & COTs

Triggerfish
Sufflamen chrysopterus Flagtail I Sea urchins & COTs
(Balistidae)
Balistida Other triggerfish I Sea urchins & COTs

Monotaxis sp. Bigeye bream I Sea urchins & COTs

Gymnocranius grandoculis Blue-lined large-eye bream I Sea urchins & COTs

Lethrinus olivaceous Longnosed I Sea urchins & COTs

Lethrinus nebulosus Blue-scaled I Sea urchins & COTs

Lethrinus rubrioperculatus Redear I Sea urchins & COTs

Emperors Lethrinus xanthochilus Yellowlip I Sea urchins & COTs

(Lethrinidae)
Lethrinus harak Thumbprint I Sea urchins & COTs

Lethrinus lentjan Pinkear I Sea urchins & COTs

Lethrinus obsoletus Orange-striped I Sea urchins & COTs

Lethrinus erythracanthus Yellowfin I Sea urchins & COTs

Lethrinus mahsena Mahsena I Sea urchins & COTs

Lethrinus variegatus Variegated I Sea urchins & COTs

Anyperodon leucogrammicus Slender Pi Fishing pressure

Cephalopholi sargus Peacock Pi Fishing pressure

Groupers
Cephalopholis urodeta Flagtail Pi Fishing pressure
(Serranidae)
Cephalopholis miniata Coral hind Pi Fishing pressure

Cephalopholis sonnerati Tomato Pi Fishing pressure

109
 Epinephelus merra Honeycomb Pi Fishing pressure

Epinephelus spilotoceps Foursaddle Pi Fishing pressure

Epinephelus polyphekadion Camouflage Pi Fishing pressure

Epinephelus caeruleopunctatus Whitespotted Pi Fishing pressure

Epinephelus fuscoguttatus Brown-marbled Pi Fishing pressure

Epinephelus tukula Potato Pi Fishing pressure

Epinephelus fasciatus Blacktip Pi Fishing pressure

Aethaloperca rogaa Redmouth Pi Fishing pressure

Variola louti Yellow-edged lyretail Pi Fishing pressure

Plectropomus laevis Saddleback Pi Fishing pressure

Plectropomus punctatus African coral cod Pi Fishing pressure

Plectorhinchus orientalis Oriental I Sea urchins & COTs

Sweetlips
Plectorhinchus picus Spotted I Sea urchins & COTs
(Haemulidae)
Plectorhinchus gibbosus Gibbus I Sea urchins & COTs

Parrotfish Bolbometopon muricatum Bumphead parrotfish C/H Coral damage

(Scaridae) Scaridae Other parrotfish H Algae vs. coral

Cheilinus trilobatus Tripletail I Sea urchins & COTs

Wrasse Cheilinus fasciatus Redbreasted I Sea urchins & COTs

(Labridae) Oxycheilinus digrammus Cheeklined splendour I Sea urchins & COTs

Cheilinus undulatus Humphead I Sea urchins & COTs

Puffers
Tetraodontidae Puffers I Sea urchins & COTs
(Tetradontidae)
Porcupinefish
Diodontidae Porcupinefish I Sea urchins & COTs
(Diodontidae)
Soldierfish & Soldierfish Pl Upwelling areas
squirrelfish
Holocentridae
(Holocentridae) Squirrelfish Pl Upwelling areas

* based on Engelhardt (2004)

110
B.2. Fish species lists divided into commercial and reef fish

Commercial fish families* Reef fish families and species*

Siganidae (rabbitfish) Chaetodontidae (butterflyfish)

Lutjanidae (snappers) Pomacanthidae (angelfish)

Lethrinidae (emperors) Acanthuridae (surgeonfish)

Serranidae (groupers) Balistidae (triggerfish)

Haemulidae (sweetlips) Labridae (wrasse)

Scaridae (parrotfish) Tetradontidae (pufferfish)

Diodontidae (porcupinefish)

Holocentridae (soldierfish & squirrelfish)

Zanclus cornutus (moorish idol)

Bulbometopon muricatum (bumphead parrotfish)

*
according to GVI Seychelles (Mah) methodology

111
 B.3. Fish feeding guilds as referred to in B.1.

Code Feeding guild Description (adapted from Obura and Grimsditch, 2009) Key species

Resident on reef surfaces, but feed in the water column. Their Soldierfish,
Pl Planktivorous abundance is related to quality of reef habitat for refuge, and water squirrelfish,
column conditions. unicornfish

High level predators. Exert top-down control on lower trophic levels.

Pi Piscivorous Important fisheries species but very vulnerable to overfishing thus Groupers, snappers
good indicators of the fishing pressure on a reef.

Butterflyfish
(chevroned,
C Corallivorous Relative abundance is an indicator of coral community health triangular, Bennetts,
Indian redfin, Meyers,
longnose)

Feed on coral competitors such as soft corals and sponges.

Angelfish, moorish
V Varied diet Relative abundances may be an indicator of abundance of these
idol
prey items and of a phase shift.

Sweetlips, emperors,
pufferfish,
porcupinefish, wrasse
Second-level predators with highly mixed diets including small fish, (tripletail, redbreasted,
I Invertivorous* invertebrates and dead animals. Important fisheries species thus cheeklined splendour,
abundances are a good indicator of fishing pressure. humphead),
triggerfish (titan,
flagtail, other
triggerfish)

Exert the primary control on coral-algal dynamics.

Parrotfish,
H Herbivorous surgeonfish,
May indicate phase shift from coral to algal dominance in response bristletooth, rabbitfish
to mass coral mortality or pressures such as eutrophication.

Relative abundance is a secondary indicator of coral community

C/H Corallivorous/Herbivorous Bumphead parrotfish
health

Butterflyfish
(vagabond, threadfin,
blackbacked,
Mertens, Indian
Relative abundance can be a secondary indicator of coral Ocean teardrop,
C/I Corallivorous/Invertivorous
community health racoon, Kleins,
speckled, spotted,
lined, saddleback,
yellow headed,
Zanzibar)

112
Appendix C
C.1. List of invertebrates surveyed on 50 m belts

Mollusca (Gastropoda) Drupella spp Drupella spp

Mollusca (Bivalvia) Tridacnidae Giant clam

Culcita spp. Cushion sea star

Sea stars (Asteroidea) Acanthaster planci Crown-of-thorns sea star
Other sea stars
Diadema spp. Long-spine urchin
Echinometra spp. Mathaes urchin
Echinothrix spp. Short-spine urchin
Sea urchins (Echinoidea)
Pencil urchin
Toxopneustes pileolus Flower urchin
Cake urchin
Holothuria artra Lollyfish
Holothuria fuscopunctata Elephant trunk
Holothuria fuscogilva White teatfish
Holothuria nobilis Black teatfish
Holothuria sp.(undescribed) Pentard
Bohadschia spp. Bohadschia spp.
Sea Cucumbers (Holothuroidea) Actinopyga spp. Actinopyga spp.
Actinopyga mauritiana Yellow surfish
Stichopus spp. Stichopus
Thelenota ananas Prickly redfish
Pearsonothurian graeffei Flowerfish
Thelenota anax Royal
Holothuria edulis Edible sea cucumber
(Cephalopoda) Octopus spp. Common reef octopus
Panulirus spp. Spiny lobster
Lobsters (Palinura)
Parribacus spp./Scyllarides spp. Slipper lobster

113
C.2. List of invertebrates surveyed on 10 m belts

Sabellidae Feather duster worms

Annelida (Polychaeta) Serpulidae Christmas tree worms

Terebellidae Spaghetti worms

(Platyhelminthes) Polycladida Flatworms

Caridea Shrimps
Arthropoda (Crustacea) Stomatopoda Mantis shrimps
- Crabs
Muricidae Murex
Drupella sp. Drupella spp.
Strombidae Conch
Cypraeidae Cowrie
Ranellidae Triton
Mollusca (Gastropoda)
Conidae Cone
Trochidae Top
Cassidae Helmet
- Other shells
Nudibranchia Nudibranchs
Ostreidae Oysters
Mollusca (Bivalvia)
Tridacnidae Giant clam
Sepiidae Cuttlefish
Mollusca (Cephalopoda)
Loliginidae Squid
Culcita sp. Cushion sea star
Sea Stars (Asteroidea) Acanthaster planci Crown-of-thorns sea star
Other sea stars

Brittle stars (Asterozoa) Ophiuroidea Brittle stars

Feather stars (Crinozoa) Crinoidea Feather stars
Diadema sp. Long-spine urchin
Echinometra sp. Mathaes urchin
Echinothrix sp. Short-spine urchin
Sea urchins (Echinoidea)
Pencil urchin
Toxopneustes sp. Flower urchin
Cake urchin
Tripneustes sp.
Other urchins

114