Marine Biodiversity Survey of Coral Reefs in Cabo Delgado in March 2015

Marine biodiversity survey of coral reefs in Cabo Delgado in March 2015

Melita Samoilys, David Obura and Kennedy Osuka CORDIO East Africa

Report prepared for Our Sea Our Life Project (ZSL, AMA, Bioclimate, CORDIO EA, University Lurio, University of Lisbon) and Flora and Fauna International

December 2015

1 Table of Contents

Introduction ...... 3 Methods ...... 4 Survey sites ...... 4 Coral species richness, reef structure and resilience ...... 6 Fish diversity ...... 6 Fish abundance ...... 6 Results and Discussion...... 7 Benthic substrate...... 7 Coral community structure ...... 8 Reef resilience indicators ...... 11 Fish species diversity ...... 12 Fish abundance ...... 15 Summary ...... 23 Coral reef structure and benthos ...... 23 Fish populations ...... 25 References ...... 29 Appendix 1. Fish abundance survey method...... 31 Appendix 2. Reef fish species inventory...... 34 Appendix 3. Mean fish abundance and biomass per site...... 39 Appendix 4. Coral genera and species lists ...... 44

Acknowledgements

We appreciate the hard work and support of Jamen Mussa for data collection and Rebecca Short for diving support on this survey.

We are indebted to Matundo Island Lodge for providing us with accomodation, great food, a perfect dive boat, SCUBA tanks and compressor for the bulk of the dives.

We are grateful to Dr Andrew Halford, Curtin University, for analysing the trends in fish populatons over years.

The field surveys were supported by the financial assistance of the European Union (DCI- ENV/2013/323-897) and the Darwin Initiative (20-023) to the Zoological Society, with additional support from FFI. The contents of this document are the sole responsibility of the authors and can under no circumstances be regarded as reflecting the position of the European Union, Darwin Initiative or Flora and Fauna International.

2 Introduction

This report on the coral reefs of northern Cabo Delgado contributes to the EU funded project Securing marine biodiversity through sustainably-financed and community-managed marine areas in coastal Mozambique, led by the Zoological Society of London (ZSL). The purpose of the survey was to assess the biodiversity status of the coral reefs at the project sites, and to assess the health of these reefs, in particular their fish populations in the context of the local artisanal fisheries. The objective in conducting comprehensive and quantitative surveys of these reefs is to provide a baseline prior to conservation action. Specifically, this report provides input to the following project outputs: i) Establish a community-run model for management of marine areas in Mozambique. v) Evaluate and communicate impacts of this action; and the following project objective: 1. Establishment of a community-run model for management of marine areas in Mozambique to secure marine biodiversity and increase coverage of marine protected areas (MPAs).

Measures of coral and fish diversity; and reef health in terms of fish populations and a range of resilience indices; are described here to provide the ecological basis for designing the community-run management areas, and to provide indices for measuring project impact.

The coral reefs of Cabo Delgado province in Northern Mozambique, particularly in the Quirimbas Archipelago, are increasingly recognized as among the most well developed and diverse in the Western Indian Ocean (Obura 2012, McClanahan et al. 2014). The convoluted coastline of low islands and banks cut by deep canyons and the very narrow continental shelf provide for a complex matrix of sheltered, channel, bank and deep reefs, which are bathed by variable eddies and mesoscale currents from the Mozambique Channel (Ternon et al. 2014).

Due to their remoteness the reefs of this coastline have not been comprehensively studied, though have had a number of monitoring and research projects focus on different parts of the reef system, such as around Pemba town, the southern Quirimbass islands, and the northern islands, in particular Vamizi and Metundu (Schleyer et al. 1999, Davidson and Hill 2006, Samoilys et al. 2011, Obura 2012, Obura, unpubl. 2003). Coral diversity at the provincial scale is at the highest levels for the Western Indian Ocean (WIO), with up to 300 species expected at a site and >350 to 400 species regionally.

Comprehensive surveys and analyses of coral reef fish diversity are few in the Western Indian Ocean (WIO). Consequently our understanding of patterns in reef fish diversity and species richness across the WIO region is still poor. However, a region centred around southern Tanzania/northern Mozambique and Northern Madagascar has been proposed as a centre of coral diversity in the WIO, driven by the Indian Ocean current patterns (Obura 2008, 2012). It is likely that coral reef fishes follow similar gradient patterns (Chabanet and Durville 2005) and therefore the reefs in northern Cabo Delgado would be expected to support high species diversity of reef fishes. Similarly this region is considered more resilient to climate induced coral bleaching (Ateweberhan et al. 2011) and less impacted by human populations and therefore likely to support higher densities and biomass of a suite of reef fish species both targeted and not targeted by local fisheries.

3 The underwater coral reef surveys conducted in March 2015 under the ZSL-led Our Sea Our Life (OSOL) Project (2013-2018) is contributing to understanding patterns in the biodiversity and health of reefs in the WIO. It also sets a baseline for the status of corals and reef fish populations on these reefs in northern Cabo Delgado prior to management and conservation interventions set to start late in 2015. In particular the surveys set baseline values for key indicator variables selected for monitoring the impacts of conservation interventions in thse sites. These reefs contribute important fishing grounds to neighbouring villages and are therefore integral to the livelihoods of coastal people in this region.

Methods

Survey sites The coral reefs surveyed in northern Cabo Delgado under the Our Sea Our Life Project were selected to correspond to the project village sites and are shown in Figure 1 and Table 1. It is important to note that villages share fishing grounds and therefore these reefs cannot be seen as exclusive fishing grounds to only one village. Of further note, Vamizi reef has been afforded some protection through a local Reserve supported by the Lodge on Vamizi Island. This site also represents a fishing ground for Olumbi village. Details of the surveys sites are provided in Table 1. Twenty two dives were done to survey corals and reef structure and resilience while 20 dive stations were done survey fish diversity and population abundance. The latter are aggregated to give 10 sites in total for diversity and 11 for abundance. An additional site from 2014 (Quifuki) was added in the presentation of the results because this site was surveyed for corals in 2015.

Figure 1. Map of all sites surveyed during the OSOL survey (see Table 1 for further details).

Table 1. Survey sites in 2015 listed with dive station names, coordinates and the villages whose fishing grounds most closely link to these reefs. * = coral data only. # fish data collected in 2014, though replication level low.

Country Area Location Site Station(dive) Lat Long Survey Date Village-Fishing ground Mozambique Palma Quirindi Quirindi Quirind2 -10.62036501 40.61030997 14/03/2015 Quirindi Mozambique Palma Quirindi Quirindi Quirind1 -10.64928903 40.58802098 14/03/2015 Quirindi Mozambique Palma Quiwia Quiwia Farol2 -10.67093203 40.64220998 14/03/2015 Quiwia Mozambique Palma Quiwia Quiwia Farol1 -10.69706667 40.65378333 13/03/2015 Quiwia Mozambique Palma Quiwia Quiwia Quiwia -10.72325 40.63063333 13/03/2015 Quiwia Mozambique Palma TekamajiN TekamajiN TekamajN1 -10.75541801 40.64315999 15/03/2015 Ngoji Mozambique Palma TekamajiN TekamajiN TekamajN2 -10.75634203 40.64404001 15/03/2015 Ngoji Mozambique Palma Palma Isles Palma Islands TekamajiSW -10.80747499 40.649174 12/03/2015 Palma Mozambique Palma Palma Isles Palma Islands Rongwe -10.82188399 40.65679098 12/03/2015 Palma Mozambique Vamizi VamiziNR VamiziN VamiziNR-3 -10.99069501 40.68838497 07/03/2015 Olumbi Mozambique Vamizi VamiziNE VamiziN VamiziNE-5 -10.994217 40.71341702 07/03/2015 Olumbi Mozambique Metundo Baixo Pinguim Baixo Pinguim Baixo Pinguim* -11.08083299 40.66075 08/03/2015 Lalane Mozambique Mocimboa Lalane Kiwe Kiwe1 -11.11513501 40.62741903 06/03/2015 Lalane Mozambique Mocimboa Lalane Kiwe Kiwe2 -11.12433298 40.6102 06/03/2015 Lalane Mozambique Mocimboa Metundo NW Metundo NW Shauli2 -11.118571 40.67165804 05/03/2015 Quifuki Mozambique Mocimboa Metundo NW Metundo NW Shauli1 -11.12168304 40.67656699 05/03/2015 Quifuki Mozambique Metundo MetundoNE MetundoNE MetundoNE1 -11.13314998 40.70244997 10/03/2015 Quifuki Mozambique Metundo MetundoE MetundoNE MetundoE -11.15118097 40.70303704 10/03/2015 Quifuki Mozambique Mocimboa Lalane Lalane Lalane1 -11.14911902 40.54054404 06/03/2015 Lalane Mozambique Metundo QuifiukiS-1 QuifiukiS-1 QuifukiS-1# -11.20548304 40.61574999 08/03/2015 Quifuki Mozambique Mocimboa Malinde-Kiwe Malinde-Kiwe KisangaMungo -11.304951 40.50233304 11/03/2015 Malinde Mozambique Mocimboa Malinde-Kiwe Malinde-Kiwe Mal-Kiwe -11.30974604 40.52603798 11/03/2015 Malinde

5 Coral species richness, reef structure and resilience

Coral genera and species were identified in the field, and a full species list was developed based on field IDs using digital photography as a primary reference and references that include underwater photographs (see Obura 2012, Sheppard and Obura 2004). To derive an index of relative abundance at genus level, a 5-point scale was used to estimate relative abundance for each genus at the end of a dive, with 1- rare; 2- uncommon; 3- common; 4- abundant and 5- dominant. Using these scores three variables can be calculated for each genus: number of sites recorded, average abundance at each site (ignoring absences) and maximum abundance across all sites. Averaging these together provides an index of relative abundance of genera, ranging from 1 (rarest) to 5 (dominant). Using species occurrence records from successive dives an accumulation curve for the survey trip is established that asymptotes towards a total species richness for the study area (see Obura 2012 for details). Michaelis-Menten enzyme kinetic equations provide a stable estimate of total species richness at the asymptote (Smax, Keating 1998), using the multivariate analysis software PRIMER v 6.0 (Clarke and Gorley 2006).

Methods for assessing reef resilience were developed by the IUCN working group on Climate Change and Coral Reefs, as a rapid assessment of the resilience of coral reefs to climate change and its most immediate consequence, high seawater temperature (Obura and Grimsditch 2009). The purpose of the method is to provide an overview of multiple factors affecting reef health at a site, giving quick recommendations for prioritizing management actions across sites and to cope with different threats/factors at different sites. Only a subset of indicators was collected here, as indicated in Table 6. Indicators were estimated either in the natural quantity (e.g. % cover, for the dominant cover types such as those analyzed in fig. 11), or on a semi-quantitative scale from 1 to 5. Fish diversity

To measure the diversity of coral reef fishes we compiled a complete species inventory of 19 families (Table 2) at each location. These families were selected based on the following criteria: largest (of all shore fishes); most diverse; known indicators of biogeographical patterns; inclusion of endemics, rare and vulnerable species (special conservation concern); amenable to UVC (diurnal, not cryptic); of fishery relevance/reef health status (Table 2). This group of 19 families includes potentially around 460 species in total from the WIO (Allen 2005, Davidson et al. 2006, Obura 2004, Samoilys 1988) and is considered broad enough and diverse enough to capture patterns in diversity of fishes across Mozambique’s northern reefs as well as broadly within the WIO region, and has been adopted as a standard by CORDIO.

Fish abundance

Surveys of fish abundance of selected taxa were done to assess the health of these coral reefs, where health can be explained in terms of the reef’s ecological resilience – its ability to resist threats and to recover to a healthy state when an impact does occur. A broad range of taxa were selected for surveys that were then assigned to specific trophic groups relevant in assessing resilience (see Appendix 1). Taxa were categorised into seven functional trophic groups: piscivores, omnivores, corallivores, invertivores, planktivores, detritivores, and herbivores (Green and Bellwood 2009, Obura and Grimsditch 2009, Lieske and Myers 1996, Samoilys and Carlos 2000). The herbivores were further broken down into six functional

6 groups (Green and Bellwood 2009): large excavators, small excavators, scrapers, grazers, browsers and grazers/detritivores; each of these groups are thought to play a different ecological role in coral reef resilience to climate change.

Table 2. Families selected for coral reef fish diversity surveys for biogeographic analyses. Family Notes Labridae (wrasse)* Largest and most diverse families of all Serranidae (groupers) shore fishes Pomacentridae (damselfishes)* Chaetodontidae (butterflyfishes)* Known indicator families of Scaridae (parrotfishes)* biogeographical patterns and/or coral Acanthuridae (surgeonfishes)* reef health Lutjanidae (snappers) Pomacanthidae (angelfishes)* Lethrinidae (emperors) Additional families included for fishery Haemulidae (grunts) importance Mullidae (goatfishes) Siganidae (rabbitfish) Nemipteridae (bream) Carangidae (trevally) Caesionidae (fusiliers) Additional families included to broaden Balistidae (triggerfish) taxonomic range Monacanthidae (filefish) Ostraciidae (boxfish) Tetraodontidae (pufferfish) *= CFDI taxa (Allen and Werner 2002, see text)

Results and Discussion

Benthic substrate Coral cover varied greatly across the sites, from a maximum of 70% at Metundu NW to minima of 5% at Lalane and Rongwe (fig. 2). The pattern was broadly similar to that of coral genera, in that outer reef sites had higher coral cover, while inner sites tended to have lower cover. The cover of rubble increased as coral cover decreased, matching observations of coral damage and breakage where reef condition was poorest, which may have been due to physical breakage, as the reefs are relatively exposed to wave energy, and/or breakage by fishermen. In addition the sum of fleshy and turf algae was inversely related to coral and coralline algae cover, indicating the reefs are exhibiting a classic feature of reefs exposed to disturbance and threats, where lost coral cover is reflected in higher fleshy and turf algae cover. No reefs were dominated by macroalgae, with the highest levels being found at Rongwe and Lalane of under 20%.

Figure 2. Benthic cover of survey sites, ordered from highest to lowest coral cover.

Coral community structure Fifty-eight coral genera were recorded across all the sites (Appendix 4), with site-level richness varying from highs of 49 at Tekamaji N2 and 47 at Vamizi NR3 (fig. 3) . Genus composition was uniformly high and consistent at 39 and more genera for over half the sites, decreasing to lows below 30 genera mostly for inner reef sites of 20-22 per site at Tekamaji SW and Rongui. The low diversity at Tekamaji SW and Rongui is notable, partly due to their being shallow almost-intertidal sites unlike any of the other sites surveyed, but also they are heavily exposed to fishing pressure by fishers from Palma peninsula, as well as being on the route of the main gas pipelines proposed for the area. The other sites with generic diversity < 30 were all inner sites and exposed to high impact from fishermen, as well as higher sediment influence.

Figure 3. Diversity of coral genera by site.

Figure 4. Coral genus abundance patterns shown by Relative Abundance (RA) an aggregate score combining average abundance, occurrence across all sites and maximum abundance at any one site.

The reefs were strongly dominated by Acropora and then Porites (fig. 4), both being dominant at one or more sites, and Acropora having the highest average abundance across all sites. After these two, about seven genera were very common or abundant (Favia, montipora, Pocillopora, Fungia, Platygyra, Galaxea and Millepora), after which the relative abundance line shows a long progressive decline. At the rarest end of the scale, Caulastrea, Horastera and Stylarea were recorded once each.

Coral species diversity was high and comparable to other sites surveyed in the Northern Mozambique Channel (fig. 5). The accumulation of species was slightly less than for surveys farther north and south along the same coast, and this might be due to a relatively constrained set of habitats sampled on inner and the shallower platforms on outer reefs, compared to Mnazi Bay and Pemba town, where deeper reef slopes were more accessible and sampled to a greater extent.

Figure 5. Number of coral species recorded in this study (Metundu- Palma), compared to other surveys conducted nearby in N. Mozambique, S. Tanzania and NE Madagascar.

Coral colony size class structure illustrates the maturity of the coral community at each site. Farol had the most well developed coral community and the highest biomass (area of corals) of all sites, dominated by 81-160 and 161-320 cm diameter corals. Interestingly, the site was not dominated by Acropora, illustrated by it being demoted to 9th place for biomass of Acropora only (Fig. 6). Farol and Metundu NW were very mixed and vibrant fore reef communities. By contrast, Kisanga nungu and Quiwia were inner reefs dominated by extensive staghorn Acropora beds, shown by the dominance of >3.2 m colony sizes (Fig. 6). Vamizi NR(3) and the sites following it were also dominated by Acropora, shown by the size class compositions for Acropora and all corals being quite similar.

Figure 6. Size class composition of all corals at the top sites (left) and of Acropora colonies (right).

Figure 7. Number (left) and area (right) of colonies in the top seven genera sampled for size classes.

After Acropora, Echinopora was the next dominant genus in terms of colony area (among the 23 genera selected for size class measurement). Figure 7 (left) shows the classic decline in number of colonies with increasing sizein all the main genera, while on the right shows what this means in terms of total biomass. 41-80 cm Acropora colonies, mostly corymbose and tabular heads, contributed the most biomass across all sites, followed by the large staghorn thickets found at Kisanga nungu and Quiwia. For Echinopora, the greatest biomass was in 81-320 cm corals, large expanses of encrusting/plating colonies on the outer fore reefs at Farol and Metundu NW.

10 Reef resilience indicators Factors indicative of reef health/resilience are summarized in fig. 8. The sites show a consistent decline from high to low, though with natural breaks at an aggregate score of 25 and 19, separating groups that can be described as having high, medium and low resilience.

Figure 8. Resilience indicators, showing the status of the current coral populations, competition/negative interactions, substrate consolidation, sediment influence, cooling, shading and extremes/ponding of water. The table is colour coded to illustrate high, medium and low levels of individual factors and the aggregate factor (obtained by summing the individual factors). High (green) levels indicate conditions that are good for corals.

The figure emphasizes how the overall coral population state is relatively consistent across all the sites, indicating that where disease and bleaching, size class distributions and evidence of recent mortality are included in consideration, there coral communities are judged as being relatively similar and healthy. Similarly, substrate condition and shading factors don’t vary greatly across the sites. By contrast, the factors that vary more between high and low resilience sites are competitive interactions, sediment exposure and cooling/ponding factors. The first two are indicative of stressful or sub-optimal conditions for corals where these score poorly, while the last is indicative of potentially high exposure to thermal stress during bleaching events. Overall, the impression is that resilience among the sites varies due to interactions with other benthic species, sediment exposure and cooling factors, but the actual status of the coral community is stable across the sites.

Combining all the benthic and corl variables into a multivariate analysis, an MDS plot (fig. 9) shows that the primary axis (horizontal) separates sites characterized by high fleshy and turf algal cover, rubble and other indicators of degraded reefs to the left, from sites with high hard coral cover and high levels of most of the resilience indicators and coral size class variables to the right. The vertical axis is aligned with indicators of Acropora dominance – total Acropora area, area in the 41-80 cm size class for Acropora, and total area/biomass of coral. The two bubble plots are shown to illustrate these main patterns, showing Acropora biomass/area (left), which is highest at Kisanga nungu and Malindi-Kiwe (due to extensive staghorn thickets), and fleshy algal cover, which was highest in the inner sites Lalane and Rongwe. Of the classification variables (reef type, topographic complexity, sediment

11 influence and latitude) the strongest pattern on the MDS was for reef type, distinguishing the inner patch reefs with low coral/high algal cover to the left, and fringing reefs strung vertically across the middle (thus with consistency in most indicators of reef community structure but variable levels of Acropora development). To the right, reefs with good coral condition include all three fore reef slope sites, the one bank (Baixo Pinguin) and two of the patch reefs (Shauli 1 and Vamizi NR(3).

Figure 9. Multi-dimensional Scaling plots of benthic and coral variables. Sites are coded by their reef type, and the bottom two plots show bubbles scaled by Acropora biomass/total area (left) and macroalgal cover (right).

Fish species diversity

A total of 263 species were recorded across 11 sites (Table 3) from the pre-determined list of 19 families of reef-associated fishes (Appendix 2). The Coral Reef Fish Diversity Index (CFDI), developed by Allen and Werner (2002) for estimating reef fish diversity based on the six most abundant, speciose and characteristic families of coral reef fishes (marked * in Table 4) was calculated to assess the overall diversity of coral reef fishes in northern Cabo Delgado and to compare with other locations in Mozambique and neighbouring countries. A value of 167 from 11 sites is one of the highest recorded in the WIO (Samoilys and Alvarez-Filip unpubls), and increases the index value of only 137 from a previous survey in 2010 around Metundo Island just further south in Mocimboa District.

12 Table 3. Coral reef fish diversity index (CFDI) values from other WIO countries and an earlier survey in Cabo Delgado at sites around Metundo Island. CFDI = total no. species in 6 families: Chaetodontidae, Pomacanthidae, Pomacentridae, Labridae, Scaridae and Acanthuridae. Location CFDI Source/date of survey Mozambique – Mocimboa 167 This survey/ 2015 and Palma Mozambique – Mocimboa 137 Samoilys et al. 2011/ 2010 Northeastern Madagascar 172 Samoilys & Randriamanantsoa 2011 / 2010 Northwestern Madagascar 176 Allen 2005 / 2002 Comoros - Grande Comore 147 Samoilys unpubl., 2010 and Moheli

The total number of species per site (in >75 minutes of observations) is shown in Table 4 and show that the highest diversity of species (144) was recorded at Quiwia, followed by Kiwe (133) and then Quirindi (131) (Fig. 10). Data from a comprehensive survey (3 weeks, over 16 reefs) conducted in NE Madagascar in 2010 gave diversity per site values that ranged from 60 to 127, with a combined value of 271 for all sites (Table 4). These values are provided for comparison and illustrate that reef fish diversity in Mocimboa and Palma Districts is high.

300

250

200

150

# of species of # 100

Sites

Figure 10. Fish species diversity by site from North to South.

13 Table 4. Total number of species per family per site, sites listed from north to south. * marks families that comprise the CFDI index. Madagascar 2010 values (Samoilys and Randriamanantsoa 2011) provided for comparison. Family Madag- All sites this Quiwia Quirindi TekamajiN Palma Vamizi Kiwe Metundo MetundoE Malindi- QuifukiS Lalane1 ascar 10 survey Islands N NW Kiwe Pomacentridae* 38 40 19 16 15 22 13 29 15 15 16 15 20 Pomacanthidae* 6 7 6 3 4 2 5 1 3 4 1 1 1 Labridae* 57 50 29 31 28 18 28 31 22 25 18 19 14 Chaetodontidae* 22 21 18 16 14 7 14 6 14 14 13 9 5 Scaridae* 20 26 12 10 16 8 13 12 10 11 8 7 1 Acanthuridae* 28 23 15 11 12 9 16 11 9 18 8 9 3 Serranidae 14 18 8 8 5 3 4 3 3 7 0 2 1 Lethrinidae 10 9 2 5 3 4 4 5 2 2 1 2 1 Lutjanidae 11 10 6 6 4 1 6 5 3 3 3 3 1 Caesionidae 7 8 4 4 1 2 0 6 4 2 4 3 0 Haemulidae 6 8 3 3 2 1 3 3 1 1 3 2 2 Nemipteridae 2 3 1 1 1 0 0 2 0 0 1 0 2 Mullidae 8 6 4 4 3 1 2 3 1 4 1 1 1 Siganidae 6 6 1 2 3 2 2 2 1 1 3 0 1 Balistidae 12 9 7 5 4 1 8 3 2 6 1 2 1 Monacanthidae 7 2 0 1 0 1 1 2 1 0 1 0 1 Ostraciidae 2 5 3 2 4 1 3 4 2 3 1 1 0 Tetraodontidae 8 8 4 2 1 2 2 4 2 1 2 2 2 Carangidae 7 4 2 1 0 0 0 1 0 1 1 0 0 Grand Total 271 263 144 131 120 85 124 133 95 118 86 78 57

14 Fish abundance We first examined differences in fish abundance between reefs based on the following attributes: geography (latitude), reef type, topographical complexity and a resilience index that combined sediment layer and texture. Sediment layer measures the depth of silt on the reef and texture measures how calcareous or muddy the substrate is. The scaled values (1-5) of these two factors were averaged to get a combined resilience index that represents the negative influence of terrigeneous sediments on coral growth - high sediment causes stress in corals. Thus a site with very high sediment influence (5) was assigned a very low resilience index (1). Put simply, silty muddy reefs are not very resilient to disturbance, though in conditions of thermal stress, turbidity associated with fine terrigenous sediment can protect reefs from light stress.

The effect of these factors on fish populations were assessed separately a posteriorly using a Bray-Curtis cluster analysis of similarity. ANOSIM results showed that reef type and the resilience index of sediment layer/texture were the most significant factors that explained the similarity in fish populations at the different reefs (Fig. 11, 12, Table 5). Fish abundance on patch reefs was broadly different to fore reef slopes and fringing reefs. However this distinction was less apparent for fish biomass (Fig. 11). Reef type, for example, well developed, deeper or exposed fore reefs versus smaller generally shallow patch reefs is likely to correlate with population abundance – with higher densities on fore and fringing reef slopes. The pattern differs with biomass with some patch reefs: those at Metundo NW and the Palma Islands site clustering with the fore reef and fringing reef sites elsewhere which is surprising since the patch reef biomass values are the lowest at these two sites (Table 6). Reef sediment was the most significant factor in driving the dissimilarity patterns between fish populations on the reefs (Fig. 3b; Table 5). Metundo E was most dissimilar from all other sites with the highest sediment score, whereas Malinde-Kiwe had the lowest sediment score and was most similar to Kiwe and different from all other sites (Fig. 12). Fish populations on the northern reefs in Palma District clustered with the most offshore (eastern) sites on Vamizi and Metundo Islands further south, with high levels of sediment. Fish populations on reefs closer to shore and in the south were associated with less sediment. This might be related to sediment effluent from the large Ruvuma River just north of Palma District on the border with Tanzania. The analyses show the shallower nearshore reefs at Malinde-Kiwe and Kiwe are more similar to Palma Islands, while the deeper further offshore reefs of Metundo NE and Vamizi are more similar to Quiwia, Quirindi and Tekamaji N. Patterns differ slightly in the fish biomass, possibly due to differences in fishing pressure.

Table 5: ANOSIM results of four factors: Latitude (North and south), Reef type (fringing reef, patchy reef and fore reef slope), resilience index on topography (low and moderate) and resilience index on sediment layer/texture (very low, low, moderate, high and very high).

Factor Latitude Reef type Topography Sediment layer and texture a) Species abundance Global R 0.133 0.413 -0.01 0.504 p-value 0.123 0.08 0.459 0.001 b) Species biomass Global R 0.100 0.335 -0.004 0.574 p-value 0.169 0.018 0.426 0.015 c) Functional group abundance Global R 0.191 0.277 0.096 0.644 p-value 0.039 0.031 0.234 0.05 d) Functional group biomass Global R 0.12 0.254 0.215 0.539 p-value 0.186 0.056 0.095 0.05

15 Table 6. Total fish biomass per site. Site Total biomass (kg/ha) VamiziNE 1186.00 Quifuki1 983.43 TekamajiN 950.05 Quirindi 925.86 MetundoE 922.53 Quiwia 765.15 Malinde-Kiwe 516.92 Kiwe 514.57 VamiziNR 495.51 MetundoNE 393.17 MetundoNW 360.32 Palma Islands 315.70

Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 40 20 Reef type Reef type Patch reef Patch reef Fore reef slope Fore reef slope Fringing reef Fringing reef 40 60

)

(

t i

r 60

a l

l i

S 80 80

100 100

e e s 1 a

E E E

N i d i R

e e s 1 a

E E E

W i

i R d i N

d j

W i

o k

d j w w N N

k o i N

n i i w i

w w N N n N a i

N i d u i

i i i w r

n N i a o

u d i f i z

r z

o K K a o i i

f z i n u i

z l m

K K a o i i d

n i u - u

l m

d d u

- u u

s a m

d u n m

u t e Q

s m a I

n m n

e t Q k Q

I a

n Q u a

Q k e d

Q u a u t

d e e

u t t V

n V

t V

M i e

n V

i M e e

T m

e l

a M

a M M

M P

P Sites Sites Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 40 20 Reef type Reef type Patch reef Patch reef Fore reef slope Fore reef slope Fringing reef Fringing reef 40 60

)

(

t i

r r 60

a l

l i

S 80 80

100 100

e e s 1 a

E E E

N i d i R

e e s 1 a

E E E

W i

i R d i N

d j

W i

o k

d j w w N N

k o i N

n i i w i

w w N N n N a i

N i d u i

i i i w r

n N i a o

u d i f i z

r z

o K K a o i i

f z i n u i

z l m

K K a o i i d

n i u - u

l m

d d u

- u u

s a m

d u n m

u t e Q

s m a I

n m n

e t Q k Q

I a

n Q u a

Q k e d

Q u a u t

d e e

u t t V

n V

t V

M i e

n V

i M e e

T m

e l

a M

a M M

M P

P Sites Sites Figure 11. Cluster analysis on fish species abundance (top) and biomass (bottom) using reef type as a factor.

16 Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 40 20 Resilience index (sediment) Resilience index (sediment) 1 (Very low) 1 (Very low) 2 (Low) 2 (Low) 3 (Moderate) 3 (Moderate) 40 4 (High) 4 (High) 60 5 (Very high) 5 (Very high)

)

(

t i

r 60

a l

l i

m i

S 80 80

100 100

e e s 1 a

E E E

N i d i R

e e s 1 a

E E E W i

i R d i N

d j

W i o k

d j w w N N

k o i N

n i i w i

w w N N n N a i

N i d u i

i i i w r

n N i a o

u d i f i z

r z

o K K a o i i

f z i n u i

z l m

K K a o i i d

n i u - u

l m

d d u

- u u

s a m

d u

n m

t e Q

s m a

n m

e t Q

k Q

n Q u a

Q k e d

Q u a u t

d e e

u t t V

e n V

t V M i e

n V T

i M e e

T m

e l

a M

a M M

M P

P Sites Sites Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 40 20 Resilience index (sediment) Resilience index (sediment) 1 (Very low) 1 (Very low) 2 (Low) 2 (Low) 3 (Moderate) 3 (Moderate) 40 4 (High) 4 (High) 60 5 (Very high) 5 (Very high)

)

(

r r 60

S 80 80

100 100

e e s

R d i

k o

w w

N i

i i

u d

K a o i

d u

n m

u a

P Sites Sites Figure 12. Cluster analysis on fish species abundance and biomass using resilience index on sediment layer and texture as a factor.

To examine which species are driving these patterns we ran a SIMPER analysis and the first 10 species with the highest scores of similarity between reef types are presented in Table 7. The results show the abundance and biomass of the small (<35cmTL) common excavating parrotfish, Chlororus sordidus, is the most important species contributing to within group similarities of the fish populations on fringing and patch reef types. The small detritivore surgeon Ctenochaetus striatus contributed highly to the similarity in fore reef slope sites. The grazer surgeonfish, Acanthurus nigrofuscus, was among the 10 most abundant species driving the within group similarity of all reef types. The abundance of the fusiliers Caesio spp. was important in driving the dissimilarity between all reef types and their biomass was higher in patch reefs compared to fringing reefs. The dissimilarity in biomass between fringing reefs versus patch and fore reef slope was due to a higher biomass of the grunt Plectorhynchus gaterinus in fringing reefs. These results show the top 10 species responsible for site differences and therefore these might be likely species to change in response to conservation interventions.

17 Table 7. SIMPER results of 10 species driving the patterns in abundance and biomass a) within reef type groups b) between reef type groups. a) Within groups abundance Patch reef Fore reef slope Fringing reef Average similarity: 50.44 Average similarity: 55.32 Average similarity: 53.46 Chlor. sordidus S Ct. striatus Chlor. sordidus S C. trifasciatus Chlor. sordidus S Ct. striatus Caesio spp. A. nigrofuscus A. nigrofuscus Ct. striatus Centropyge spp. Centropyge spp. A. nigrofuscus Ct. truncatus C. trifasciatus Scarus spp. Z. scopas Z. scopas Z. scopas C. trifasciatus Ct. truncatus Centropyge spp. C. guttatissimus Monotaxis grandoculis P. gaterinus C. trifascialis L. fulviflamma C. auriga B. undulatus B. undulatus b) Between groups abundance Patch reef & Fore reef slope Patch reef & Fringing reef Fore reef slope & Fringing reef Average dissimilarity = 59.52 Average dissimilarity = 57.53 Average dissimilarity = 58.01 Caesio spp. Caesio spp. Caesio spp. Ct. truncatus Scarus spp. O. niger O. niger Ct. truncatus P. gaterinus Hemitaurich. zoster S. russelli Hemitaurich. zoster C. guttatissimus Ct. striatus Ct. truncatus Scarus spp. P. gaterinus N. elegans Monotaxis grandoculis S. globiceps Monotaxis grandoculis B. undulatus Hemitaurich. zoster L. fulviflamma Ct. striatus A. nigrofuscus N. hexacanthus B N. elegans B. undulatus A. leucosternon a) Within groups biomass Patch reef Fore reef slope Fringing reef Average similarity: 40.95 Average similarity: 39.14 Average similarity: 40.29 Chlor. sordidus S Chlor. sordidus S Chlor. sordidus S P. gaterinus Ct. striatus Ct. striatus Caesio spp. A. nigrofuscus Monotaxis grandoculis S. ghobban N. elegans A. nigrofuscus Scarus spp. Ct. truncatus P. gaterinus Ct. striatus Z. scopas B. undulatus L. fulviflamma C. argus L. fulviflamma A. nigrofuscus B. undulatus Caesio spp. S. frenatus S. tricolor Z. scopas Z. scopas Monotaxis grandoculis Aprion viriscens b) Between groups biomass Patch reef & Fore reef slope Patch reef & Fringing reef Fore reef slope & Fringing reef Average dissimilarity = 72.49 Average dissimilarity = 62.24 Average dissimilarity = 62.56 Caesio spp. P. gaterinus P. gaterinus N. elegans Caesio spp. N. elegans Monotaxis grandoculis Monotaxis grandoculis Monotaxis grandoculis S. ghobban Ct. striatus Caesio spp.

18 Ct. truncatus S. ghobban O. niger N. hexacanthus B Chlor. sordidus S N. hexacanthus B O. niger Chlor. sordidus B Chlor. sordidus B P. gaterinus S. russelli N. unicornis L. fulviflamma L. fulviflamma L. fulviflamma N. unicornis A. xanthopterus Chlor. sordidus S

Similar patterns across the reef survey sites are seen when the Bray Curtis cluster analyses are done with fish functional groups (Fig.13, 14) illustrating that the functional group categories are representative of the fish assemblages. The trophic groups differed significantly in abundance and biomass (K-W test abundance: H=264.2; p<0.001; biomass H=208.4; p<0.001). The sites also showed significant differences in density and biomass of functional groups (K-W test abundance: H=76.12; p<0.001; biomass H=92.93; p<0.001).

Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 70 60 Reef type Reef type Patch reef Patch reef 75 Fore reef slope Fore reef slope Fringing reef Fringing reef 70 80

)

(

r 85 r 80

S 90 90 95

100 100

P Sites Sites Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 70 60 Reef type Reef type Patch reef Patch reef 75 Fore reef slope Fore reef slope Fringing reef Fringing reef 70 80

) )

% %

( (

y y

t t

i i r 85 r 80

a a

l l

i i

m m

i i

S S 90 90 95

100 100

i i

e e s 1 a a e 1 s e

E E E E E E

i d i R N d i N i R

W i i W

d j j d

o k k o

n n

w w N N N w N w

i N i N

i i w i i w i i

n N i a a n i N

d u i i u d

r r

o o

f i z i z f

z z

K K a o i i i K i a o K

n u i u i n

l m m l

d d

- u u -

d u u d

u u

s m a m a s

m n m n

e t Q Q e t

I I

n n

Q k Q k

a a

Q a u Q a u

d e d e

u t t u

e e

a a

t V V t

n V V n

i M e e i M

T T

l l

e e

m m

l l

a M M a

M M

a a

M M

P P Sites Sites Fig. 13. Cluster analysis on fish functional groups abundance and biomass using reef type as a factor.

19 Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 70 60 Resilience index (sediment) Resilience index (sediment) 1 (Very low) 1 (Very low) 75 2 (Low) 2 (Low) 3 (Moderate) 3 (Moderate) 70 4 (High) 4 (High) 80 5 (Very high) 5 (Very high)

)

(

r 85 r 80

S 90 90 95

100 100

P Sites Sites Abundance Biomass Transform: Log(X+1) Transform: Log(X+1) Resemblance: S17 Bray Curtis similarity Resemblance: S17 Bray Curtis similarity 70 60 Resilience index (sediment) Resilience index (sediment) 1 (Very low) 1 (Very low) 75 2 (Low) 2 (Low) 3 (Moderate) 3 (Moderate) 70 4 (High) 4 (High) 80 5 (Very high) 5 (Very high)

)

(

r 85 r 80

S 90 90 95

100 100

P Sites Sites Fig. 14. Cluster analysis on fish functional groups abundance and biomass using resilience index on sediment layer and texture as a factor.

The density and biomass of the different trophic groups varied between the 12 sites, and for some groups was very low (Fig. 15). In terms of the herbivorous functional groups (see Appendix 1), known to play an important role in conferring resilience to coral reefs under climate change, the following patterns are notable. The large excavators (parrotfish) were either missing completely (Palma Islands, Malinde-Kiwe, MetundoNW, MetundoNE, MetundoE and Vamizi NR) or their abundance was low (Fig. 15). This important functional group was most abundant at Quifuki1 (8.0 fish per 1000m2) and least abundant at Tekamaji N (0.8 fish per 1000m2). Further, the parrotfish groups – the small excavators and scrapers, which should be highly abundant on healthy reefs, were also low, particularly at Quirindi, Quiwia, and both Vamizi sites. Densities were higher at Palma Island and Tekamaji N. The herbivores were in fact dominated by grazers, primarily the surgeonfish Acanthurus nigrofuscus, and these densities were high. Grazers were most abundant at Tekamaji N and at Vamizi both inside and east of the community Reserve (Fig. 15). These grazers are also important in maintaining low levels of algae, which aids coral recovery from bleaching. The low densities of parrotfish may be caused by fishing while A. nigrofuscus, a small brown surgeon fish is not a target fishery species. The very low biomass values for the grazers (Fig. 15) illustrates this point – abundant but small.

20 Table 8. SIMPER results of 10 functional groups that most highly correlate with the patterns in a) abundance and b) biomass within reef type groups (i), and between reef type groups(ii). i) Within groups abundance Patch reef Fore reef slope Fringing reef Average similarity: 80.67 Average similarity: 85.99 Average similarity: 84.58 Scrapers Grazers Small excavators Small excavators Detritivores Grazers Grazers Invertivores Detritivores Corallivores Small excavators Omnivores Planktivores Grazer-detritivores Corallivores Detritivores Scrapers Grazer-detritivores Omnivores Corallivores Invertivores Invertivores Planktivores Scrapers Grazer-detritivores Omnivores Planktivores Browsers ii) Between groups abundance Patch reef & Fore reef slope Patch reef & Fringing reef Fore reef slope & Fringing reef Average dissimilarity = 21.76 Average dissimilarity = 19.28 Average dissimilarity = 14.42 Detritivores Detritivores Browsers Browsers Planktivores Planktivores Piscivores Grazer-detritivores Omnivores Planktivores Piscivores Piscivores Invertivores Small excavators Large excavators Grazer-detritivores Omnivores Small excavators Omnivores Large excavators Detritivores Grazers Invertivores Invertivores Scrapers Browsers Scrapers Large excavators Scrapers Corallivores i) Within groups biomass Patch reef Fore reef slope Fringing reef Average similarity: 69.66 Average similarity: 72.72 Average similarity: 69.34 Scrapers Grazers Small excavators Small excavators Scrapers Omnivores Omnivores Small excavators Grazers Planktivores Invertivores Scrapers Grazers Browsers Detritivores Detritivores Omnivores Planktivores Browsers Detritivores Invertivores Piscivores Grazer-detritivores Planktivores Piscivores ii) Between groups biomass Patch reef & Fore reef slope Patch reef & Fringing reef Fore reef slope & Fringing reef Average dissimilarity = 36.30 Average dissimilarity = 32.91 Average dissimilarity = 27.79 Browsers Grazer-detritivores Browsers Invertivores Omnivores Omnivores Planktivores Planktivores Large excavators Omnivores Detritivores Planktivores Detritivores Large excavators Grazer-detritivores Large excavators Piscivores Invertivores Piscivores Small excavators Small excavators Grazers Grazers Scrapers Grazer-detritivores Invertivores Piscivores Scrapers Browsers Detritivores

21 Kiwe Quifiki 1 Malinde-Kiwe Metundo-NW

Abundance (#) Biomass (kg) Abundance (#) Biomass (kg) Abundance (#) Biomass (kg) Abundance (#) Biomass (kg)

150 250 200 150

2 200 2

150

0 100 0 0 100

0 150 0

r r 100

100

a 50

a a 50

e e 50

M 50 M

0 0 0 0

r r r r r r r r t r r v r r r v v r r r it iv v v ri o o lli o o o a a e e e tr r r v r r r v v r r r it r r v r r r v v r r r it o o ll o o o a a e e e t iv iv a iv iv iv c c p s z e o o li o o o a a e e e tr o o li o o o a e e e tr iv iv a iv iv iv c c p s z e c n r rt t it x x a w ra d iv iv l iv iv iv c c p s z e iv iv l iv iv iv c a p s z c n r rt t it x x ra w ra d s o e k tr e e r o - c n ra t t it x x a w a c a t t it x c a w a e is o e k tr e e c o - i m v n e l c r G z s r k r e r o r -d s n r r k r x r r -d m v n e l r G z P O C la e S B a i m o e n t l e c r G z i m o e n t e e c ro G z P O C n la e S B a In D g m r P C v a e e S B a P C v a e e l S a I P D g m r P L S G O In l D g m r O In l D g m B r L S G P L S G P L S G Functional groups Functional groups Functional groups Functional groups

MetundoE MetundoNE Palma Islands Quirindi

Abundance (#) Biomass (kg) Abundance (#) Biomass (kg) Abundance (#) Biomass (kg) Abundance (#) Biomass (kg)

300 250 250 250

2 250 2 200 200 2 200

0 200 0

0 150 150 0 150

r 150 r

100 100 100

a 100 a

M 50 50 50 M 50

0 0 0 0

r r v r r r v v r r r it r r v r r r v v r r r it r r v r r r v v r r r it r r r r r r r r t o o li o o o a a e e e tr o o li o o o a a e e e tr o o li o o o a a e e e tr o o liv o o o v v e e e ri iv iv l iv iv iv c c p s z e iv iv l iv iv iv c c p s z e iv iv l iv iv iv c c p s z e v v l v v v a a p s z t c n ra t t it x x a w a c n ra t t it x x a w a c n ra t t it x x a w a i i a ti ti ti c c a a e s o r k r e e r o r -d s o r k r e e r o r -d s o r k r e e r o r -d c n r r k ri x x r w r -d i m e n t l c r G z i m e n t l c r G z i m e n t l c r G z is m o e n t e e c ro G z P C v la e e S B a P C v la e e S B a P C v la e e S B a P C v a e e l S a O In D g m r O In D g m r O In D g m r O In l D g m B r P L S G P L S G P L S G P L S G Functional groups Functional groups Functional groups Functional groups

Quiwia TekamajiN VamiziNE VamiziNR

Abundance (#) Biomass (kg) Abundance (#) Biomass (kg) Abundance (#) Biomass (kg) Abundance (#) Biomass (kg)

150 250 150 150

2 200 2

0 0 100 100 100 0

0 150 0

100

a 50

a 50 a 50

M 50 M

0 0 0 0

r r r r r v r r r it r r r r r v r r r it r r r r r v r r r it r r iv r r r v v r r r it o o liv o o o a v e e e r o o liv o o o a v e e e r o o liv o o o a v e e e r o o ll o o o a a e e e tr iv iv l iv iv iv c a p s z t iv iv l iv iv iv c a p s z t iv iv l iv iv iv c a p s z t iv iv a iv iv iv c c p s z e c a t t it x c a w a e c a t t it x c a w a e c a t t it x c a w a e c n r rt t it x x ra w ra d s n r r k r e x r r -d s n r r k r e x r r -d s n r r k r e x r r -d is o e k tr e e c o - i m o e n t e c ro G z i m o e n t e c ro G z i m o e n t e c ro G z m C v n e l r G z P C v a e e l S B a P C v a e e l S B a P C v a e e l S B a P O n la e S B a O In l D g m r O In l D g m r O In l D g m r I P D g m r P L S G P L S G P L S G L S G Functional groups Functional groups Functional groups Functional groups

Fig. 15. Fish functional groups density and biomass by site. Each site consists of transects from two dive stations (see Methods) except for Quifiki, VamiziNE, VamiziNR and MetundoE

22 A SIMPER analysis was also run on the functional groups and these are presented in Table 8. These show that the herbivores groups of scrapers, excavators and browsers (parrotfish), and the Detritivores, Grazers and Planktivores are most correlated with the fish assemblage patterns found across the reefs.

In terms of the fishery species and higher trophic level groups, numbers varied. Throughout all sites piscivore abundance was very low, even in the protected Vamizi Reserve, and they were absent from the Palma Islands (Fig. 15). This result is in part explained by the reef structure with the Palma Island sites being very shallow inner reef sites which do not usually support high numbers of higher trophic level groupers, trevally and job fish. However, at the other reefs with well formed outer reef slopes and depth, such as Quiwia and Quirindi, Tekamaji N, Vamizi N and Metundo, these piscivore densities are low and suggest moderately heavy fishing pressure. Other important fishery groups are the omnivores (snapper, emperor and grunts), which were abundant at Quirindi and Vamizi NE, but generally not very abundant at other sites (Fig. 15). Further, their biomass was low, except at Quirindi, showing that at most reefs these fish were reasonably abundant but small. This is a typical sign of growth overfishing. It was notable that omnivores in the unprotected Vamizi NE were more abundant than in the Reserve, Vamizi NR. This is possibly explained by inadequate protection in the Reserve, and the easternmost Vamizi NE site is perhaps protected from fishing due to its distance from shore and exposure to strong seas.

Overall the reefs surveyed at Tekamaji N, Quiwia and the two Vamizi sites combined both the highest populations of reef fishes and fish abundance spread across a range of functional groups suggesting a healthy mixed species assemblage of both fishery and non-fishery species. The highest density of planktivores was found at Malindi-Kiwe and Metundo E. These species (planktivorous surgeonfish, triggerfish and fusiliers) are generally associated with well formed reef slopes, depth and good currents. It is therefore not surprising they were low in abundance at the shallow lagoonal type habitats of for example the Palma island sites or sites close to Lalane. The lack of planktivores at Vamizi NR is surprising and may reflect the wide submerged reef plateau of this site, and that our survey did not reach the reef edge. Our understanding of how planktivores may act as an indicator of reef health is still poor (Samoilys and Randriamanantsoa 2011).

Summary We have no prior surveys of the reefs in this northern Cabo Delgado area and to our knowledge they have not been surveyed before, at least to the level of biodiversity and reef and fish population health that we have recorded here. These surveys in 2015 therefore represent the first baseline. Clearly this is very late to have a baseline with the ever increasing levels of human impact. However, analysis of data collected on our surveys further south near Metundo and Vamizi islands during the last five years provide some indication of recent changes in the condition of these reefs. In addition, our surveys from elsewhere in the northern Mozambique Channel, particularly from Madagascar also help put the Mocimboa and Palma reefs in context.

Coral reef structure and benthos Overall, the diversity and composition of species and community structure of hard corals was very consistent with other sites in the Northern Mozambique Channel (Obura 2012). The dominance of Acropora across all size classes and in multiple reef locations and habitats is indicative of a mature state of reefs in the area, suggesting advanced recovery from any prior mass mortality from coral bleaching or other factors that there might have been earlier (e.g. Davidson and Hill 2006).

23 Coral cover was low and algal cover high at most inner sites. The poor condition of these sites is exacerbated by other indicators – high levels of competitor/negative interactions (such as higher cover of sponges, higher abundance of bioeroding sea urchins), unconsolidated substrate, high sediment influence due to proximity with the mainland and rivers, low levels of cooling due to distance from open water and being surrounded by extensive shallow platforms, low levels of shading/screening factors (these were low across all sites) and high tendency for ponding of waters due to low circulation in inner bays and platforms. These inner sites are also strongly impacted by fishing, being closest to fisher communities on the mainland and inner islands, which is reflected in their fish populations (next section).

Poor reef condition is limited to inner reefs, which are all patch reefs geomorphologically, while reefs in good condition are found in the three other reef classes – fringing, fore and bank reefs. While this poor condition of inner reefs may reflect limitations on their resilience and robustness, their higher exposure to fishing and terrestrial impacts is likely the driver of their poor condition, exacerbated by lower levels of resilience to enable full recovery from chronic or acute impacts.

Table 10. Resilience factors aggregated from multiple indicators, see caption to Fig. 6 for details.

Sites Coralspopn. Competition Substrate Sediment Cooling Shade & screen Ponding Aggregate

Metundu E 3.9 4.3 3.4 4.5 4.8 1.6 5.0 27.4

Farol 4.5 3.3 3.2 5.0 4.0 2.0 5.0 27.0 Vamizi NE(5) 4.4 3.3 3.2 3.5 4.3 2.2 5.0 25.8

High Metundu NE1 4.1 4.3 2.6 4.0 3.8 2.0 5.0 25.7 Shauli 1 4.2 4.5 3.0 4.5 3.3 2.2 4.0 25.7 Shauli 2 4.1 4.3 2.6 3.5 3.5 2.0 4.0 23.9 Vamizi NR(3) 4.1 3.8 2.4 4.0 3.0 1.8 4.0 23.0 Baixo Pinguin 4.2 2.8 2.2 3.5 3.0 2.0 5.0 22.7

Tecomaji N1 3.6 2.3 2.6 4.0 3.8 2.0 4.0 22.2 Quirindi 2 3.9 3.5 2.2 2.0 3.0 1.6 5.0 21.2

Quirindi 1 4.4 2.3 3.4 3.5 2.5 2.0 3.0 21.0 Medium Quiwia 3.9 3.3 2.6 3.0 2.3 1.8 4.0 20.8 Tecomaji N2 4.0 3.5 2.0 3.0 2.5 1.4 3.0 19.4 Quifuki South 3.9 3.0 3.0 2.5 2.0 1.8 3.0 19.2 Kiwe 1 3.4 2.3 2.0 2.5 2.3 1.4 3.0 16.8 Kiwe 2 3.4 3.0 2.0 3.5 1.5 1.2 2.0 16.6

Tecomaji 3.3 2.3 2.2 2.0 1.8 1.2 2.0 14.7 Rongui 2.7 2.0 2.0 2.5 2.0 1.2 2.0 14.4

Low Kisanga mungu 3.5 2.3 2.2 1.0 1.8 1.6 2.0 14.3 Malinde-Kiwe 3.4 2.5 2.2 1.0 1.3 1.6 2.0 13.9 Lalane 2.9 1.5 1.6 2.0 1.5 1.6 2.0 13.1

Table 10 summarises the resilience factors illustrated in Fig. 6 in a framework more suitable to management, providing clear indications of those factors score poorly and may be changed by interventions, contrasting with those factors scoring highly and could be capitalized on for conservation.

The surveys findings on coral cover and reef health are summarised as follows:  coral cover varied from 70 – 10% across sites, suggesting a broad gradient of reef state.  Reefs were overwhelmingly dominated by Acropora, including and corymbose heads at most sites, and staghorn thickets on some fringing and patch reef locations. Acropora colony sizes were large, indicating mature community development and high recovery from any past major disturbances.  After Acropora, Porites and a high diversity of other genera were common to abundant.  Coral reef health was high to moderate in 2/3 of the sites, predominantly outer and fore reef/fringing reef sites. Poor condition of inner reefs appeared to be associated with high terrestrial influence (sediments, turbidity) and fishing impacts.  Factors contributing to reef resilience that varied most across sites, and thus differentiated high from low resilience sites were sediment exposure and cooling/ponding factors, and ecological interactions (competition, bioerosion, etc).  256 species of corals in 58 genera were recorded, the location being at the low end of diversity compared to other sites within the northern Mozambique Channel center of diversity.

Fish populations Fish diversity at the Mocimboa and Palma sites was relatively high with a total species count of 263 from the 19 families. The species diversity CFDI index of 167 is also high when compared with other sites in the WIO, notably with northern Madagascar (values of 172 and 176, Samoilys & Randriamanantsoa 2011, Allen 2005, respectively) suggesting these reefs with varied topography and benthos support a wide range of fish species. Sites with the highest species diversity were in northern Palma at Quiwia, Quirindi and Tekamaji N, and further south at Kiwe.

Overall the reefs surveyed at Tekamaji N, Quiwia and the two Vamizi sites combined both the highest populations of reef fishes and fish abundance spread across a range of functional groups suggesting a healthy mixed species assemblage of both fishery and non-fishery species. A similar species assemblage was seen at the distant offshore site at Vamizi NE. However, across all sites piscivore (groupers, trevally and job fish) abundance and biomass was low, even at the deeper outer reef slopes surveyed at Quiwia and Quirindi, Tekamaji N and the two Vamizi sites. These results suggest moderately heavy fishing pressure for these target fishery species. Other fishery species, the omnivores (snapper, emperor and grunts) were abundant at Quirindi and Vamizi NE, but not at other sites and all were small with low biomass, except at Quirindi. This result suggests that fishing pressure is affecting the biomass across most sites.

The herbivorous parrot fish groups – the small excavators and scrapers, are generally abundant on healthy reefs and play a critical role in confering resilience to reefs under climate change (Bellwood et al. 2003). In Palma District their densities were relatively low, particularly at Palma Islands, MetundoE, and Malinde-Kiwe. Densities were higher at Quifiki1 and Tekamaji N. The herbivores were in fact dominated by grazers, primarily the surgeon fish Acanthurus nigrofuscus. This species plays a functional role in resilience but is not a fishery species. These results also suggest fishing pressure is depleting the parrot fish.

25 Information on changes over time from previous surveys by CORDIO in northern Cabo Delgado is limited and inconclusive, and summarised in Table 11 below. Three sites were tested where Vamizi sites were re-surveyed and the Metundo and Baixo Pinguim sites were compared because they were close by and structurally similar.

Table 11. Changes in fish abundance over time (tested by Primer Bray Curtis and SIMPER).

2010/11 2014/15 Difference Vamizi N Reserve 2011 Vamizi NReserve 2015 * R=0.463; p=0.057 (community reserve) (community reserve) Vamizi NE 2011 (exposed Vamizi NE 2015 (exposed NS R= 0.02; p=0.48 distant site) distant site) Baixo Pinguim NE 2010 Metundo NE 2014 ** R=0.494; p=0.008 (unprotected site) (unprotected site)

Populations changed over time at Vamizi Reserve and in the Baixo Pinguim / Metundo NE area. The difference at Vamizi Reserve showed a mix of some species increasing and other decreasing. For Baixo Pinguim 2010 and Metundo NE 2014, the species driving the difference over years were Scarus, Chlororus and Naso species which increased in abundance, while species of Chaetodon, Ctenochaetus and Acanthurus declined. These results do not provide a clear picture of impacts. They suggest that Vamizi Reserve is protecting some species and therefore their numbers are increasing, but the same is happening near Metundo Island with no Reserve. This may therefore be because coral health has improved.

Comparisons of fish assemblages based on fish species abundance and biomass across a range of survey sites in Cabo Delgado including the Palma District sites showed that reef type and sediment play a significant role in driving species abundance, but this pattern is less clear with fish biomass, possibly due to differences in fishing pressure across the sites. Thus the less developed patch reefs support lower levels of biomass, as would be expected, but this pattern is not seen throughout.

When compared with regional surveys from elsewhere (MacNeil et al. 2015, McClanahan et al. 2011) total biomass values in this survey reveal that several sites are seen to support good levels of biomass (Table 6). These authors recommend that managers should aim to maintain biomass levels at 300-600 kg/ha to keep reefs healthy and supporting ecosystem services, and that unfished biomass should be at 1,200 kg/ha. However, we caution using these recommended management targets without clarity on the families/species that are included in such total biomass figures. In addition, MacNeil and McClanahan’s surveys are shallow, confined to <11 m in depth, sometimes on snorkel and not SCUBA, and often in lagoons. In contrast our surveys span the whole depth of the reef, up to 25-30m. Since many larger fish species are associated with depth this difference is clearly going to affect the biomass values. Further, which species/families are included in such aggregate biomass values needs to be clearly stated. Our surveys were on reefs that are all fished except for Vamizi NR (a local reserve) and show that five of the sites have biomass measures that are close to the “unfished biomass level” reported by McNeil et al. (2015), and all are within or above the recommended targets for sustainable fishing levels recommended by McClanahan et al. (2011) except for the Metundo and Palma Islands sites which are at the minimum biomass level. Based on these values, and the generally low values of Piscivores, we would suggest that the reefs of Mocimboa-Palma are showing some signs of overfishing, with depleted biomass values, but these are not yet critically low.

Measures of coral and fish diversity, and reef health in terms of fish populations and a range of resilience indices, are described here to provide the ecological basis for designing the community-run management areas, and to provide indices for measuring project impact. Previous recommendations on the fish and fishery indicators, in conjunction with the artisanal fisheries monitoring data (Table 12), now need to be evaluated against the results obtained in the coral reef survey reported here. For example, our analyses here reveal 10 fish species responsible for site differences and therefore these might be likely species to change in response to conservation interventions.

It is recommended that discussions are now held with project partners on the values obtained here for setting baseline ecological indicators for both designing MPAs and measuring project impact. The relevance and effectiveness of the values obtained during this coral reef survey need to be debated by the broader range of partners.

27 Table 12. Indicator species for the OSOl project for measuring trends in populations of IUCN Red List species and impacts on fishery biomass. Species not counted in reef surveys removed. * = potential species for length data collection from fishery surveys.

No. Species Local Name Family/Trophic Ecology/habitat IUCN Red List Other factors category 1* Naso brevirostris Puju Acanthuridae/ coral reef Potentially important ecological role in 2 Acanthurus tennenti Angadja Planktivore fish community; easily identified. 3* Siganus sutor Safi Siganidae/ non-coral demersal, key fishery spp. Herbivore seagrass 4* Scarus ghobban Pono Scarinae/ Herbivore coral reef role in reef resilience 5 Leptoscarus vaigiensis Pono Scarinae/ Herbivore seagrass key fishery spp. 6 Bolbometapon muricatum ? Scarinae/ Herbivore coral reef Vulnerable Key ecological /resilience role, easily identified 7* Lethrinus lentjan Ndjana Lethrinidae/ mixed habitats – potentially resilient to CC Omnivore demersal 8 Epinephelus fuscogutttatus Kicheua Serranidae/ coral reef Vulnerable Piscivore 9 Plectropomus laevis Chewa Serranidae/ coral reef Vulnerable Piscivore 10 Epinephelus marginatus Chewa Serranidae/ coral reef Endangered Rare Piscivore 11* Lutjanus fulviflamma Querera Lutjanidae/ coral reef key fishery spp. Omnivore 12 Cheilinus undulatus Gombezi Omnivore coral reef Endangered 13* Caranx tille Xereua/ Carangidae/ Pelagic / coral reef Ndjolue Piscivore 15 Triaenodon obesus Papa hite tip shark / top pelagic Near piscivore Threatened

References Allen, G.R. (2005) Reef fishes of Northwestern Madagascar. In: McKenna S & G.R. Allen (eds.) A Rapid Marine Biodiversity Assessment of the Coral reefs of Northwest Madagascar. pp. 39-48. Allen, G.R. and T.B. Werner. (2002) Coral reef fish assessment in the coral triangle of southeastern Asia. Env. Biol. Fish. 65: 209-214. Ateweberhan M., McClanahan T.R., Graham N.A.J., and Sheppard C.R.C. (2011) Episodic heterogeneous decline and recovery of coral cover in the Indian Ocean. Coral Reefs, 30, 739–752. Chabanet, P. and Durville, P. (2005). Reef fish inventory of Juan De Nova’s Natural Park (Western Indian Ocean). Western Indian Ocean J. Mar. Sci. 4(2): 145-162. Davidson, J. Hill, N. Muaves, L., Mucaves, S. Silva, I., Guissamulo, A., and Shaw, A. (2006) Vamizi Island Mozambique: Marine Ecological Assessment October 2006. ZSL/Maluane Report. 96pp. Green, A.L. and Bellwood, D.R. 2009. Monitoring functional groups of herbivorous reef fishes as indicators of coral reef resilience – A practical guide for coral reef managers in the Asia Pacific region. IUCN working group on Climate Change and Coral Reefs. IUCN, Gland, Switzerland. 70 pp. Lieske E. and Myers R. (1996) Coral Reef Fishes. Princeton University Press. 400pp. MacNeil, M. A., Graham, N. A., Cinner, J. E., Wilson, S. K., Williams, I. D., Maina, J., ... and McClanahan, T. R. (2015). Recovery potential of the world's coral reef fishes. Nature, 520(7547), 341-344. McClanahan, T. R., Graham, N. A., MacNeil, M. A., Muthiga, N. A., Cinner, J. E., Bruggemann, J. H., and Wilson, S. K. (2011). Critical thresholds and tangible targets for ecosystem- based management of coral reef fisheries. Proceedings of the National Academy of Sciences, 108(41), 17230-17233. McClanahan TR, Ateweberhan M, Darling ES, Graham NAJ, Muthiga NA (2014) Biogeography and Change among Regional Coral Communities across the Western Indian Ocean. PLoS ONE 9(4): e93385. doi:10.1371/journal.pone.0093385 Obura, DO (2003, unpublished) Coral species survey, Pemba, Mozambique; October 2003 Obura D.O. (2008). Scleractinian coral fauna of the Western Indian Ocean. In: Obura, D.O., Tamelander, J., and Linden, O. (Eds). Ten years after bleaching – facing the consequences of climate change in the Indian Ocean. CORDIO Status Report 2008. CORDIO/ SIDA-SAREC. Mombasa. pp. 139-147. Obura, D. (2012). The diversity and biogeography of Western Indian Ocean reef-building corals. PloS ONE, 7(9). Obura, D. (2004). Biodiversity Surveys of the Coral Reefs of Mnazi Bay - Ruvuma Estuary Marine Park. IUCN EARO Publication Obura, D.O. and Grimsdith G. (2009). Resilience Assessment of coral reefs – Assessment protocol for coral reefs, focusing on coral bleaching and thermal stress. IUCN working group on Climate Change and Coral Reefs. IUCN, Gland, Switzerland. 70 pp. Samoilys, M.A. (1988). Abundance and species richness of coral reef fish on the Kenyan coast: the effects of protective management and fishing. Proc. 6th Int. Coral Reef Symp. 2: 261-266. Samoilys, M.A. and Carlos, G. (2000). Determining methods of underwater visual census for estimating the abundance of coral reef fishes. Env. Biol. Fish. 57:289-304.

29 Samoilys, M.A. and Randriamanantsoa, B. (2011). Reef fishes of northeast Madagascar. In: Obura, D., Di Carlo, G., Rabearisoa, A. and Oliver, T. (eds.) (2011). A Rapid Marine Biodiversity Assessment of the coral reefs of northeast Madagascar. RAP Bull Biol Assessment, 61. Schleyer, M., Obura, D., Motta, H., Rodrigues, M.-J. 1999. A Preliminary Assessment of Bleaching in Mozambique. In: Coral Reef Degradation in the Indian Ocean. Status Reports and Project Presentations, 1999. Eds Linden, O., Sporrong, N. Sida/SAREC Marine Science Programme, Sweden Sheppard C.R.C. and Obura D.O. (2004). Corals and reefs of Cosmoledo and Aldabra atolls: extent of damage, assemblage shifts and recovery following the severe mortality of 1998. J. Nat. Hist. 39(2): 103–121. Ternon J.F., Bach P., Barlow R., Huggett J., Jaquemet S., Marsac F., Ménard F., Penven P., Potier M., & Roberts M.J. (2014) The Mozambique Channel: From physics to upper trophic levels. Deep Sea Research Part II: Topical Studies in Oceanography, 100, 1–9

30 Appendix 1. Fish abundance survey method. Taxa surveyed for fish abundance with their assigned trophic functional group (after Green and Bellwood 2009, Obura and Grimsditch 2009). Functional Group/family English name or species Notes on feeding habits and selection of species Group Piscivores Carangidae Trevally top level predators, they exert top-down control on lower Serranidae Groupers trophic levels of fish, are very vulnerable to overfishing, and Lutjanidae Aprion viriscens good indicators of the level of anthropogenic disturbance (fishing) on a reef. Elasmobranchii Sharks and rays

Scombridae Tunas Omnivores Haemulidae Sweetlips Second-level predators with highly mixed diets including small Lethrinidae Emperors fish, invertebrates and dead animals, their presence/absence is Lutjanidae Snappers except Aprion also a good indicator of anthoropogenic disturbance (fishing). viriscens Corallivores Chaetodontidae Butterflyfish Obligate and facultative corallivores are a secondary indicator of coral community health. 8 species: C. bennetti, C. lineolatus, C. melannotus, C. meyeri, C. ornatissimus, C. trifascialis, C. trifasciatus, C. zanzibarensis

Invertivores Pomacanthidae Angelfish Feed on coral competitors such as soft corals and sponges, their relative abundance may be a secondary indicator of abundance/stability of these groups and of a phase shift. Except Centropyge spp. which are grazer-detrivores Balistidae Triggerfish Benthic triggerfish (e.g. Sufflamen spp.) Chaetodontidae Butterflyfish Non-corallivore species - all others except the 8 above and H. zoster and H. diphreutes which are planktivorous Labridae Cheilinus undulatus Only this labrid surveyed Planktivores Balistidae Triggerfish Resident on reef surfaces, but feed in the water column. Their presence/absence may be related to habitat for shelter and water column conditions Trigger fish in the water column (eg. Melichthys spp., Odonus

31 Functional Group/family English name or species Notes on feeding habits and selection of species Group niger)

Chaetodontidae Hemitaurichthys zoster

Heniochus diphreutes

Acanthuridae Naso spp. >20cm Large Naso in water column, except unicornis and tuberosus (below) which are always Browsers

A. mata A. nubilus A. thompsoni Paracanthurus Caesionidae Fusiliers Detritivores Acanthuridae Ctenochaetus Feed on organic matter in sediment and on reef surfaces, their relative abundance may be an indicator of eutrophication and conditions unsuitable for corals. Herbivores Exert the primary control on coral-algal dynamics and are implicated in determining phase shifts from coral to algal dominance especially in response to other pressures such as eutrophication, mass coral mortality Large Scaridae Bolbometopon Take few, large, deep bites, remove much substratum and play excavators a key role in bioerosion Chlorurus spp. >35cm Cetoscarus bicolor Small Chlorurus spp. <35cm Remove substrate - play a smaller role in bioerosion excavators

32 Functional Group/family English name or species Notes on feeding habits and selection of species Group Scrapers Scarus spp. Remove algae, sediment and other material by closely cropping or scraping the substrate Hipposcarus spp. Browsers Scaridae Calotomus spp. Feed on large macro-algae Leptoscarus spp. Acanthuridae Naso unicornis Naso tuberosus Naso spp. <21cm Ephippidae Bat fish – Platax spp. Siganidae S. canaliculatus Kyphosidae Rudder fish Grazers Acanthuridae Zebrasoma spp. graze epilithic algal turfs, which can also limit growth of macroalgae A. nigrofuscus Acanthurus spp. Small surgeon species, incl. lineatus Siganidae Siganus spp. Except Siganus canaliculatus Grazer- Acanthuridae A. blochii Ring tails. Feed on algal turf, sediment and some animal detritivores material. Similar role to grazers - remove macroalgae A. dussumieri A. leucocheilus A. nigricauda A. xanthopterus A. tennenti Pomacanthidae Centropyge spp

33 Appendix 2. Reef fish species inventory.

Family Species Acanthuridae A. leucosternon Acanthuridae A. mata Acanthuridae A. nigricaudus Acanthuridae A. nigrofuscus Acanthuridae A. tennenti Acanthuridae A. thompsoni Acanthuridae A. triostegus Acanthuridae A. xanthopterus Acanthuridae C. striatus Acanthuridae C. truncatus Acanthuridae Ctenochaetus binot Acanthuridae N. brachycentron Acanthuridae N. brevirostris Acanthuridae N. elegans Acanthuridae N. hexacanthus Acanthuridae N. thynnoides Acanthuridae N. unicornis Acanthuridae Naso annulatus Acanthuridae Z. scopas Acanthuridae Z. vellifer Balistidae B. viridescens Balistidae Balistapus undulatus Balistidae Balistoid. conspicillum Balistidae Melichthys indicus Balistidae Melichthys niger Balistidae Odonus niger Balistidae S. chrysopterum Balistidae Sufflamen bursa Caesionidae C. lunaris Caesionidae C. xanthonota Caesionidae Caesio caerulaurea Caesionidae Pterocaesio marri Caesionidae Pterocaesio tile Carangidae C. fulvoguttatus Carangidae C. melampygus Carangidae Carang. gymnostethus Chaetodontidae C. bennetti Chaetodontidae C. falcula Chaetodontidae C. guttatissimus Chaetodontidae C. interruptus Chaetodontidae C. kleinii Chaetodontidae C. lunula

34 Family Species Chaetodontidae C. madagaskariensis Chaetodontidae C. melannotus Chaetodontidae C. meyeri Chaetodontidae C. trifascialis Chaetodontidae C. trifasciatus Chaetodontidae C. vagabundus Chaetodontidae C. xanthocephalus Chaetodontidae C. zanzibarensis Chaetodontidae Chaetodon auriga Chaetodontidae F. longirostris Chaetodontidae Forcipiger flavissimis Chaetodontidae Hemitaurichthys zoster Chaetodontidae Heniochus acuminatus Chaetodontidae Heniochus monoceros Haemulidae P. gaterinus Haemulidae P. plagiodesmus Haemulidae P. playfairi Haemulidae Plectorhinchus albovittatus Haemulidae Plectorhinchus flavomac. Haemulidae Plectorhinchus vittatus Labridae A. melanurus Labridae A. meleagrides Labridae A. twistii Labridae Anampses caeruleopunt Labridae B. axillaris Labridae B. diana Labridae Bodianus anthioides Labridae C. batuensis Labridae C. caudimacula Labridae C. fasciatus Labridae C. freiri Labridae C. oxycephalus Labridae C. trilobatus Labridae C. undulatus Labridae Cirrhilabrus exquisitus Labridae Coris aygula Labridae Coris cuvieri Labridae Epibulus insidiator Labridae Gomphosus caeruleus Labridae H. doliatus Labridae H. hortulanus Labridae H. nebulosus Labridae H. scapularis Labridae Halichoeres cosmetus Labridae Hemigymnus fasciatus

35 Family Species Labridae Hemigymnus melapterus Labridae Hologymnosus annulatus Labridae L. dimidiatus Labridae Labrichthys unilineatus Labridae Labroides bicolor Labridae Labropsis xanthonota Labridae Macropharyngodon bipartitus Labridae Novaculichthys taeniourus Labridae Oxycheilinus digrammus Labridae Oxycheilinus mentalis Labridae Pseudocheilinus evanidus Labridae Pseudocheilinus hexataenia Labridae Pseudodax moluccanus Labridae Pteragogus cryptus Labridae S. bandanensis Labridae Stethojulis strigiventer Labridae T. hardwicki Labridae T. herbracium Labridae T. lunare Labridae Thalassoma amblycephal Lethrinidae Gnathoden. aurolinea Lethrinidae Gymnoc. grandoculis Lethrinidae L. erythracanthus Lethrinidae L. lentjan Lethrinidae L. mahsena Lethrinidae L. obsoletus Lethrinidae L. olivaceus Lethrinidae Monotaxis grandoc Lutjanidae Aphareus furca Lutjanidae Aprion virescens Lutjanidae L. bengalensis Lutjanidae L. bohar Lutjanidae L. fulviflamma Lutjanidae L. fulvus Lutjanidae L. gibbus Lutjanidae L. kasmira Lutjanidae L. monostigma Lutjanidae Macolor niger Monacanthidae Amanses scopas Monacanthidae C. fronticinctus Monacanthidae C. pardalis Monacanthidae Oxymonacanth longir Monacanthidae Pervagor janthinosoma Mullidae M. vanicolensis Mullidae Mulloidi. flavolineatus

36 Family Species Mullidae P. cyclostomus Mullidae P. pleurostigma Mullidae P. trifasciatus Mullidae Parupeneus barberinus Nemipteridae S. ghanam Ostraciidae O. meleagris Pomacanthidae Apolemichthys trimacul Pomacanthidae C. bispinosa Pomacanthidae C. multispinis Pomacanthidae P. imperator Pomacanthidae P. semicirculatus Pomacanthidae Pomacanthus chrysurus Pomacanthidae Pygoplites diacanthus Pomacentridae A. allardi Pomacentridae A. clarkii Pomacentridae A. sexfasciatus Pomacentridae A. sparoides Pomacentridae Amblygl. indicus Pomacentridae Amphiprion akallopisos Pomacentridae C. atripectoralis Pomacentridae C. dimidiata Pomacentridae C. lepidolepis Pomacentridae C. nigroanalis Pomacentridae C. nigrura Pomacentridae C. ternatensis Pomacentridae C. viridis Pomacentridae C. weberi Pomacentridae Chromis agilis Pomacentridae Chromis cf. leucura Pomacentridae D. trimaculatus Pomacentridae D.carneus Pomacentridae Dascylus aruanus Pomacentridae Neoglyphid. melas Pomacentridae P. baenschi Pomacentridae P. caeruleus Pomacentridae P. sulfureus Pomacentridae P. trichourus Pomacentridae P. trilineatus Pomacentridae Pl. johnstonianus Pomacentridae Pl. lacrymatus Pomacentridae Plectroglyphid. dickii Pomacentridae Pristotis obtusirostris Pomacentridae S. nigricans Scaridae Bolbo. muricatum Scaridae Calotomus carolinus

37 Family Species Scaridae Cetoscarus ocellatus Scaridae Chlor. capistratoides Scaridae Chlor. sordidus Scaridae Chlor. strongylocephalus Scaridae Chlor. atrilunula Scaridae Hipposcarus harid Scaridae S. falcipinnis Scaridae S. frenatus Scaridae S. ghobban Scaridae S. globiceps Scaridae S. niger Scaridae S. psittacus Scaridae S. rubroviolaceus Scaridae S. russelli Scaridae S. scaber Scaridae S. tricolor Scaridae S. viridifucatus Serranidae Aethaloperca rogaa Serranidae C. leopardus Serranidae C. miniatus Serranidae C.urodeta sb nigripinnis Serranidae Cephalopholis argus Serranidae E. fasciatus Serranidae E. malabaricus Serranidae E. merra Serranidae E. spilotoceps Serranidae Plectropomus laevis Serranidae V. louti Siganidae S. luridus Siganidae S. stellatus Siganidae S. sutor Siganidae Siganus argenteus Tetraodontidae A. mappa Tetraodontidae A. nigropunctatus Tetraodontidae C. janthinoptera Tetraodontidae C. solandri Tetraodontidae C. valentini Tetraodontidae Canthigaster rivulata

38 Appendix 3. Mean fish abundance and biomass per site. Table 1: Abundance (number/1000m2) and biomass (kg/1000m2) of fish families at the study sites. Figures are means±SE. Site Fish family Abundance Biomass Site Fish family Abundance Biomass Mean SE Mean SE Mean SE Mean SE Kiwe Acanthuridae 28.80 6.71 4.84 1.99 Quirindi Acanthuridae 130.40 5.48 19.22 9.55 Balistidae 2.40 0.65 0.41 0.13 Balistidae 16.80 1.20 2.54 0.82 Caesionidae 47.60 25.43 6.19 3.47 Caesionidae 32.80 5.85 4.58 3.19 Chaetodontidae 17.60 4.05 0.31 0.09 Chaetodontidae 79.20 4.33 1.49 0.49 Haemulidae 8.40 4.82 3.14 1.46 Haemulidae 42.40 6.42 32.41 19.97 Lethrinidae 0.40 0.40 0.02 0.02 Lethrinidae 6.40 1.12 7.06 6.14 Lutjanidae 11.60 6.32 5.24 2.79 Lutjanidae 18.40 3.31 8.45 5.19 Pomacanthidae 15.20 5.19 0.24 0.08 Pomacanthidae 38.40 2.11 1.68 0.39 Scaridae 118.80 17.50 29.83 6.93 Scaridae 52.00 3.96 13.44 2.71 Serranidae 0.40 0.40 0.07 0.07 Serranidae 6.40 0.51 1.57 0.73 Siganidae 10.80 5.93 1.15 0.53 Siganidae 1.60 0.24 0.15 0.11 Total 262.00 51.46 Total 424.80 92.59 Quifiki Acanthuridae 124.00 15.14 19.06 8.26 Quiwia Acanthuridae 191.20 10.92 33.78 14.87 Balistidae 1.33 1.33 0.53 0.53 Balistidae 15.20 0.80 2.28 0.42 Caesionidae 40.00 30.55 5.42 3.26 Caesionidae 0.00 0.00 0.00 0.00 Chaetodontidae 48.00 8.33 1.58 0.91 Chaetodontidae 87.20 6.85 2.13 0.50 Haemulidae 13.33 13.33 8.87 8.87 Haemulidae 3.20 0.80 1.15 1.15 Lethrinidae 5.33 1.33 4.14 2.47 Lethrinidae 8.00 1.30 2.72 1.14 Lutjanidae 6.67 4.81 6.17 4.03 Lutjanidae 2.40 0.40 0.31 0.24 Pomacanthidae 10.67 5.81 4.13 4.04 Pomacanthidae 24.00 0.84 0.67 0.19 Scaridae 176.00 72.00 48.44 11.65 Scaridae 76.00 4.04 32.35 15.01 Serranidae 0.00 0.00 0.00 0.00 Serranidae 4.80 0.58 0.92 0.70 Siganidae 0.00 0.00 0.00 0.00 Siganidae 0.80 0.20 0.20 0.20 Total 425.33 98.34 Total 412.80 76.52 Malinde-Kiwe Acanthuridae 23.20 3.95 3.29 2.68 TekamajiN Acanthuridae 372.80 12.58 34.55 2.75 Balistidae 0.00 0.00 0.00 0.00 Balistidae 13.60 0.75 1.65 0.35 Caesionidae 179.20 25.41 27.18 16.14 Caesionidae 0.00 0.00 0.00 0.00 Chaetodontidae 46.40 1.75 0.95 0.30 Chaetodontidae 58.40 5.90 1.16 0.37 Haemulidae 13.60 2.68 4.04 3.17 Haemulidae 12.80 2.52 6.38 5.37 Lethrinidae 0.00 0.00 0.00 0.00 Lethrinidae 2.40 0.40 0.72 0.53 Lutjanidae 1.60 0.40 0.43 0.43 Lutjanidae 22.40 4.63 20.88 19.08 Pomacanthidae 1.60 0.24 0.03 0.02 Pomacanthidae 27.20 2.03 1.63 0.66 Scaridae 44.80 6.23 15.78 10.67 Scaridae 148.80 3.87 25.60 2.31 Serranidae 0.00 0.00 0.00 0.00 Serranidae 6.40 0.68 2.09 1.21 Siganidae 0.00 0.00 0.00 0.00 Siganidae 1.60 0.24 0.33 0.21 Total 310.40 51.69 Total 666.40 95.01

39 Site Fish family Abundance Biomass Site Fish family Abundance Biomass Mean SE Mean SE Mean SE Mean SE Metundo NW Acanthuridae 72.40 9.05 7.42 2.96 VamiziNE Acanthuridae 258.67 22.93 32.76 14.27 Balistidae 6.80 2.46 0.76 0.22 Balistidae 49.33 3.67 24.66 7.73 Caesionidae 62.40 42.96 7.03 4.84 Caesionidae 0.00 0.00 0.00 0.00 Chaetodontidae 44.80 6.82 0.78 0.13 Chaetodontidae 52.00 3.21 1.32 0.56 Haemulidae 2.00 1.61 1.79 1.64 Haemulidae 20.00 4.00 8.78 6.39 Lethrinidae 1.20 0.61 0.23 0.12 Lethrinidae 24.00 4.04 9.75 7.22 Lutjanidae 1.20 0.85 1.96 1.32 Lutjanidae 36.00 4.58 15.18 8.42 Pomacanthidae 9.60 2.61 0.80 0.28 Pomacanthidae 20.00 1.53 0.38 0.13 Scaridae 75.20 14.48 12.99 3.88 Scaridae 48.00 3.51 21.56 12.62 Serranidae 2.80 1.04 2.05 0.92 Serranidae 9.33 0.33 4.22 1.01 Siganidae 1.60 0.88 0.21 0.13 Siganidae 0.00 0.00 0.00 0.00 Total 280.00 36.03 Total 517.33 118.60 MetundoE Acanthuridae 126.67 2.85 32.37 10.44 VamiziNR Acanthuridae 157.33 3.18 11.00 1.03 Balistidae 160.00 18.01 37.57 23.46 Balistidae 17.33 0.33 3.01 0.31 Caesionidae 0.00 0.00 0.00 0.00 Caesionidae 0.00 0.00 0.00 0.00 Chaetodontidae 46.67 4.98 1.57 0.91 Chaetodontidae 69.33 3.38 1.03 0.22 Haemulidae 0.00 0.00 0.00 0.00 Haemulidae 0.00 0.00 0.00 0.00 Lethrinidae 0.00 0.00 0.00 0.00 Lethrinidae 28.00 6.51 16.84 15.48 Lutjanidae 1.33 0.33 0.46 0.46 Lutjanidae 8.00 1.53 2.74 1.96 Pomacanthidae 21.33 0.33 0.98 0.37 Pomacanthidae 42.67 0.88 2.62 0.99 Scaridae 38.67 3.18 15.78 8.16 Scaridae 62.67 5.24 8.82 4.36 Serranidae 13.33 0.88 3.52 1.09 Serranidae 4.00 0.58 1.49 0.86 Siganidae 0.00 0.00 0.00 0.00 Siganidae 18.67 4.67 2.00 2.00 Total 408.00 92.25 Total 408.00 49.55 MetundoNE Acanthuridae 137.50 14.50 11.33 1.69 Palma Islands Acanthuridae 17.20 19.61 4.93 1.57 Balistidae 12.50 3.66 2.19 0.61 Balistidae 0.20 0.80 0.32 0.32 Caesionidae 0.00 0.00 0.00 0.00 Caesionidae 0.40 1.60 0.18 0.18 Chaetodontidae 33.00 5.49 0.59 0.14 Chaetodontidae 7.00 13.56 0.46 0.23 Haemulidae 0.00 0.00 0.00 0.00 Haemulidae 0.60 2.40 2.55 2.55 Lethrinidae 16.00 12.92 3.83 3.15 Lethrinidae 1.00 3.10 3.59 2.99 Lutjanidae 3.50 2.97 1.78 1.73 Lutjanidae 1.20 4.80 4.16 4.16 Pomacanthidae 19.50 3.16 0.31 0.05 Pomacanthidae 2.00 2.19 0.39 0.29 Scaridae 134.50 27.29 18.49 3.10 Scaridae 26.80 27.35 14.96 5.99 Serranidae 3.00 1.46 0.80 0.49 Serranidae 0.00 0.00 0.00 0.00 Siganidae 0.00 0.00 0.00 0.00 Siganidae 0.20 0.80 0.03 0.03 Total 359.50 39.32 Total 56.60 31.57

40 Table 2: Abundance (number/1000m2) and biomass (kg/1000m2) of fish functional groups at the study sites. Figures are means±SE. Site Fish functional group Abundance Biomass Site Fish functional group Abundance Biomass Mean SE Mean SE Mean SE Mean SE Kiwe Piscivores 0.40 0.40 0.07 0.07 Quirindi Piscivores 7.20 0.58 3.31 1.64 Omnivores 20.40 7.07 8.41 3.10 Omnivores 66.40 9.56 46.18 28.49 Corallivores 9.60 2.87 0.14 0.04 Corallivores 20.00 1.61 0.30 0.10 Invertivores 10.40 2.68 0.59 0.15 Invertivores 64.00 2.59 4.02 0.86 Planktivores 48.80 25.19 6.63 3.39 Planktivores 53.60 5.42 6.81 2.77 Detritivores 8.40 2.76 0.67 0.21 Detritivores 73.60 4.59 4.40 1.10 Large excavators 1.20 2.54 1.43 1.43 Large excavators 1.60 0.24 2.33 1.46 Small excavators 46.80 1.20 8.96 3.07 Small excavators 36.00 3.10 6.33 0.74 Scrapers 65.60 6.85 18.60 3.24 Scrapers 10.40 0.81 2.83 0.72 Browsers 6.40 11.13 0.99 0.41 Browsers 4.00 0.77 1.95 1.31 Grazers 23.20 7.03 2.00 0.61 Grazers 48.00 2.43 3.10 0.60 Grazer-detritivores 20.80 6.02 2.97 1.49 Grazer-detritivores 40.00 3.29 11.03 10.57 Quifiki Piscivores 0.00 0.00 0.00 0.00 Quiwia Piscivores 4.80 0.58 0.92 0.70 Omnivores 25.33 17.33 19.17 11.21 Omnivores 13.60 1.12 4.18 1.60 Corallivores 36.00 6.93 1.28 0.86 Corallivores 29.60 1.29 0.43 0.07 Invertivores 13.33 2.67 0.84 0.57 Invertivores 42.40 1.78 3.06 0.68 Planktivores 40.00 30.55 5.42 3.26 Planktivores 45.60 10.41 4.82 4.38 Detritivores 62.67 19.37 8.28 3.37 Detritivores 78.40 2.42 4.69 0.58 Large excavators 8.00 6.11 8.30 7.77 Large excavators 6.40 1.12 13.42 10.58 Small excavators 150.67 76.88 27.28 5.47 Small excavators 41.60 2.04 7.66 2.00 Scrapers 17.33 2.67 12.86 6.64 Scrapers 28.00 2.70 11.28 3.17 Browsers 0.00 0.00 0.00 0.00 Browsers 33.60 4.97 20.76 11.84 Grazers 56.00 9.24 8.94 5.81 Grazers 64.80 2.58 4.59 1.03 Grazer-detritivores 16.00 4.00 5.98 3.11 Grazer-detritivores 24.00 1.30 0.72 0.41 Malinde-Kiwe Piscivores 0.00 0.00 0.00 0.00 TekamajiN Piscivores 6.40 0.68 2.09 1.21 Omnivores 15.20 3.07 4.47 3.59 Omnivores 37.60 4.77 27.98 18.12 Corallivores 41.60 2.14 0.59 0.12 Corallivores 28.80 3.28 0.42 0.19 Invertivores 3.20 0.37 0.05 0.02 Invertivores 40.00 2.70 3.26 0.75 Planktivores 181.60 25.68 27.60 16.37 Planktivores 12.00 2.53 0.48 0.31 Detritivores 8.00 1.22 0.48 0.29 Detritivores 232.80 11.93 13.94 2.86 Large excavators 0.00 0.00 0.00 0.00 Large excavators 0.80 0.20 0.93 0.93 Small excavators 12.00 1.41 2.48 1.78 Small excavators 76.80 2.03 13.39 1.92 Scrapers 31.20 4.49 12.85 8.53 Scrapers 71.20 4.37 11.29 1.80 Browsers 7.20 1.56 2.19 2.07 Browsers 44.00 5.07 13.52 6.54 Grazers 8.80 1.96 0.96 0.90 Grazers 95.20 6.61 6.59 1.99 Grazer-detritivores 1.60 0.24 0.03 0.02 Grazer-detritivores 20.80 1.74 1.12 0.89

41 Site Fish functional group Abundance Biomass Site Fish functional group Abundance Biomass Mean SE Mean SE Mean SE Mean SE Metundo NW Piscivores 3.60 1.39 3.79 1.89 VamiziNE Piscivores 9.33 0.33 4.22 1.01 Omnivores 3.60 2.02 2.24 1.84 Omnivores 80.00 11.37 33.70 16.11 Corallivores 32.40 5.76 0.59 0.14 Corallivores 24.00 3.21 0.35 0.19 Invertivores 24.40 4.56 1.68 0.39 Invertivores 42.67 1.20 16.16 6.63 Planktivores 62.40 42.96 7.03 4.84 Planktivores 38.67 4.91 9.73 7.05 Detritivores 13.60 2.93 0.90 0.22 Detritivores 114.67 7.75 6.87 1.87 Large excavators 0.00 0.00 0.00 0.00 Large excavators 2.67 0.67 7.20 7.20 Small excavators 51.20 11.04 7.60 2.72 Small excavators 18.67 3.71 2.65 1.91 Scrapers 24.00 7.11 5.40 1.58 Scrapers 26.67 0.33 11.71 6.89 Browsers 4.80 4.37 3.23 3.13 Browsers 26.67 6.17 11.53 8.50 Grazers 55.60 6.76 3.50 0.57 Grazers 86.67 8.11 6.53 3.02 Grazer-detritivores 4.40 1.39 0.07 0.02 Grazer-detritivores 46.67 0.88 7.95 0.78 MetundoE Piscivores 13.33 3.53 3.52 1.09 VamiziNR Piscivores 4.00 0.58 1.49 0.86 Omnivores 1.33 1.33 0.46 0.46 Omnivores 36.00 8.02 19.58 17.41 Corallivores 6.67 4.81 0.09 0.07 Corallivores 42.67 1.45 0.60 0.08 Invertivores 37.33 11.85 6.63 3.38 Invertivores 56.00 3.51 5.58 1.25 Planktivores 188.00 74.22 37.14 17.94 Planktivores 4.00 1.00 0.74 0.74 Detritivores 18.67 5.81 1.12 0.35 Detritivores 66.67 3.53 3.99 0.85 Large excavators 0.00 0.00 0.00 0.00 Large excavators 0.00 0.00 0.00 0.00 Small excavators 20.00 14.05 4.23 2.64 Small excavators 53.33 3.38 7.47 3.13 Scrapers 18.67 6.67 11.55 8.82 Scrapers 9.33 1.86 1.36 1.26 Browsers 14.67 8.11 22.30 11.53 Browsers 1.33 0.33 0.17 0.17 Grazers 68.00 18.33 4.24 1.14 Grazers 102.67 8.74 7.63 3.10 Grazer-detritivores 21.33 1.33 0.97 0.65 Grazer-detritivores 32.00 1.73 0.95 0.45 MetundoNE Piscivores 3.50 1.59 2.21 1.49 Palma IslandsP iscivores 0.00 0.00 0.00 0.00 Omnivores 19.00 15.28 4.20 3.46 Omnivores 11.20 1.96 10.30 7.11 Corallivores 17.50 3.85 0.26 0.06 Corallivores 14.40 2.23 0.22 0.13 Invertivores 23.00 4.77 2.16 0.53 Invertivores 15.20 1.50 0.84 0.46 Planktivores 6.50 4.47 0.78 0.43 Planktivores 2.40 0.40 0.40 0.25 Detritivores 43.00 11.83 2.82 0.80 Detritivores 28.00 1.30 1.72 0.32 Large excavators 0.00 0.00 0.00 0.00 Large excavators 0.00 0.00 0.00 0.00 Small excavators 96.50 24.02 12.29 2.97 Small excavators 36.00 2.85 5.54 2.13 Scrapers 37.00 7.16 6.02 1.16 Scrapers 71.20 6.89 9.42 3.91 Browsers 7.50 4.81 2.67 1.51 Browsers 0.00 0.00 0.00 0.00 Grazers 86.00 6.28 5.57 0.40 Grazers 37.60 3.44 2.55 0.94 Grazer-detritivores 20.00 3.21 0.34 0.06 Grazer-detritivores 10.40 0.68 0.59 0.46

42 Table 3. Density and biomass of key indicator species. Site Indicator species Abundance Biomass Site Indicator species Abundance Biomass Mean SE Mean SE Mean SE Mean SE Kiwe Acanthurus nigrofuscus 10.80 2.92 0.76 0.20 Quirindi Acanthurus nigrofuscus 24.00 10.12 1.50 0.63 Epinephelus fuscoguttatus 0.00 0.00 0.00 0.00 Epinephelus fuscoguttatus 0.00 0.00 0.00 0.00 Cetoscarus ocellatus 0.00 0.00 0.00 0.00 Cetoscarus ocellatus 0.00 0.00 0.00 0.00 Cheilinus undulutus 0.00 0.00 0.00 0.00 Cheilinus undulutus 0.00 0.00 0.00 0.00 Chlororus sordidus B 1.20 1.20 1.43 1.43 Chlororus sordidus B 0.80 0.80 0.95 0.95 Chlororus sordidus S 46.80 6.85 8.96 3.07 Chlororus sordidus S 35.20 12.86 5.70 1.34 Lutjanua fulviflamma 10.80 5.96 4.98 2.65 Lutjanua fulviflamma 7.20 6.25 2.10 1.66 Naso hexacanthus B 0.00 0.00 0.00 0.00 Naso hexacanthus B 0.00 0.00 0.00 0.00 Naso hexacanthus S 0.00 0.00 0.00 0.00 Naso hexacanthus S 0.00 0.00 0.00 0.00 Quifiki Acanthurus nigrofuscus 24.00 10.58 0.86 0.45 Quiwia Acanthurus nigrofuscus 16.00 3.58 1.00 0.22 Epinephelus fuscoguttatus 0.00 0.00 0.00 0.00 Epinephelus fuscoguttatus 0.00 0.00 0.00 0.00 Cetoscarus ocellatus 0.00 0.00 0.00 0.00 Cetoscarus ocellatus 2.40 1.60 5.62 5.58 Cheilinus undulutus 0.00 0.00 0.00 0.00 Cheilinus undulutus 0.00 0.00 0.00 0.00 Chlororus sordidus B 8.00 6.11 8.30 7.77 Chlororus sordidus B 0.00 0.00 0.00 0.00 Chlororus sordidus S 134.67 85.05 24.93 7.82 Chlororus sordidus S 36.80 7.20 5.22 0.51 Lutjanua fulviflamma 4.00 2.31 3.66 1.88 Lutjanua fulviflamma 0.00 0.00 0.00 0.00 Naso hexacanthus B 0.00 0.00 0.00 0.00 Naso hexacanthus B 18.40 13.24 8.19 7.13 Naso hexacanthus S 0.00 0.00 0.00 0.00 Naso hexacanthus S 0.00 0.00 0.00 0.00 Malinde-Kiwe Acanthurus nigrofuscus 1.60 1.60 0.10 0.10 TekamajiN Acanthurus nigrofuscus 18.40 15.57 1.15 0.97 Epinephelus fuscoguttatus 0.00 0.00 0.00 0.00 Epinephelus fuscoguttatus 0.00 0.00 0.00 0.00 Cetoscarus ocellatus 0.00 0.00 0.00 0.00 Cetoscarus ocellatus 0.00 0.00 0.00 0.00 Cheilinus undulutus 0.00 0.00 0.00 0.00 Cheilinus undulutus 0.00 0.00 0.00 0.00 Chlororus sordidus B 0.00 0.00 0.00 0.00 Chlororus sordidus B 0.00 0.00 0.00 0.00 Chlororus sordidus S 12.00 5.66 2.48 1.78 Chlororus sordidus S 70.40 8.45 11.15 1.13 Lutjanua fulviflamma 1.60 1.60 0.43 0.43 Lutjanua fulviflamma 2.40 2.40 1.39 1.39 Naso hexacanthus B 0.00 0.00 0.00 0.00 Naso hexacanthus B 42.40 21.04 12.87 6.82 Naso hexacanthus S 0.00 0.00 0.00 0.00 Naso hexacanthus S 0.00 0.00 0.00 0.00

43 Appendix 4. Coral genera and species lists

Genera 1. Acanthastrea 21. Heliopora 41. Platygyra 2. Acropora 22. Herpolitha 42. Plerogyra 3. Alveopora 23. Horastrea 43. Plesiastrea 4. Astreopora 24. Hydnopohora 44. Pocillopora 5. Blastomussa 25. Isopora 45. Podabacea 6. Caulastrea 26. Leptastrea 46. Polyphyllia 7. Coscinaraea 27. Leptoria 47. Porites 8. Cycloseris 28. Leptoseris 48. Poritipora 9. Cyphastrea 29. Lobophyllia 49. Psammocora 10. Diploastrea 30. Merulina 50. Sandalolitha 11. Echinophyllia 31. Millepora 51. Scolymia 12. Echinopora 32. Montastrea 52. Seriatopora 13. Favia 33. Montipora 53. Stylarea 14. Favites 34. Mycedium 54. Stylocoeniella 15. Fungia 35. Oulophyllia 55. Stylophora 16. Galaxea 36. Oxypora 56. Symphillia 17. Gardineroseris 37. Pachyseris 57. Tubastrea 18. Goniastrea 38. Pavona 58. Turbinaria 19. Goniopora 39. Pectinia 20. Halomitra 40. Physogyra

Species Family Genus Species lutkeni Acroporidae Acropora abrotanoides massawensis acuminata microclados appressa microphthalma arabensis millepora aspera muricata austera nana bifurcata nasuta branchi natalensis cerealis paniculata clathrata pharaonis copiosa polystoma digitifera retusa divaricata robusta gemmifera rosaria grandis roseni granulosa samoensis horrida secale humilis selago hyacinthus squarrosa insignis subulata intermedia tenuis latistella valida listeri variabilis

44 vermiculata Coscinaraeid Coscinarae ae a columna zp crassa Alveopora daedelea exesa spongiosa monile tizardi zpA Astreopora expansa Horastrea indica listeri Dendrophylli

myriophthalma idae Tubastrea micrantha

ocellata zpp

randalli Turbinaria frondens

Isopora palifera irregularis

Montipora aequituberculata mesenterina

calcarea stellulata

caliculata Euphyllidae Physogyra lichtensteini

cryptus Plerogyra sinuosa

efflorescens Faviidae Caulastrea connata

floweri Cyphastrea chalcidicum

foveolata microphthalma

informis serailia millepora Diploastre monasteriata a heliopora nodosa Echinopor a gemmacea spongodes hirsutissima stilosa lamellosa tuberculosa robusta undata Favia danae venosa favus verrucosa Gardineros helianthoides Agariciidae eris planulata lizardensis Leptoseris glabra matthai hawaiiensis pallida incrustans rotumana mycetoseroides speciosa scabra stelligera Pachyseris speciosa truncatus Pavona clavus Favites abdita decussata acuticolis duerdeni complanata explanulata flexuosa frondifera halicora maldivensis pentagona varians russelli venosa spinosa Astrocoeniid Stylocoeni vasta ae ella armata Goniastrea australensis guentheri columella

45 deformis Hydrozoa Heliopora coerulea edwardsi Millepora dichotoma pectinata exesa peresi platyphylla retiformis tenaera Leptastrea aequalis Hydnophor Merulinidae a exesa pruinosa pilosa purpurea rigida transversa Merulina ampliata Leptoria phrygia Acanthastr Montastre Mussidae ea brevis a curta echinata magnistellata hemprichii serageldini

Oulophylli ishigakiensis a crispa regularis levis subechinata Platygyra acuta Blastomus sa merletti crosslandi Cynarina lachrymalis daedalea Lobophylli

lamellina a corymbosa

pini hataii

ryukyuensis hemprichii

sinensis robusta

verweyi Scolymia australis

Plesiastrea devantieri Symphyllia agaricia

versipora erythraea

Fungiidae Cycloseris cyclolytes radians explanulata Oculinidae Galaxea astreata

patelliformis fasicularis wellsi Echinophyl Fungia concina Pectiniidae lia aspera corona echinata danai echinoporoides fungites Mycedium elephantotus granulosa mancaoi paumotensis umbra repanda Oxypora lacera scruposa Pectinia africana scutaria Pocilloporida e Pocillopora damicornis Halomitra pileus elegans Herpolitha limax eydouxi weberi indiania Podabacia crustacea meandrina Polyphillia talpina Sandalolith verrucosa a robusta Seriatopor a caliendrum

46 guttatus hystrix Stylophora madagascarensis pistillata subseriata Poritidae Goniopora albiconus columna djiboutiensis lobata planulata somaliensis stokesi stutchburyi zp. Porites annae cylindrica harrisoni lichen lobata lutea nigrescens profundus rus solida stephensoni superfusa Stylaraea punctata Siderastreida Psammoco e ra contigua niestraazi profundacella