Spatial Dynamics of Red Sea Coral Reef Fish Assemblages: a Taxonomic And

Spatial dynamics of Red Sea coral reef fish assemblages: a taxonomic and

ecological trait approach

Thesis by

Gloria Lisbet Gil Ramos

In Partial Fulfillment of the Requirements

For the Degree of

Master of Science

King Abdullah University of Science and Technology

Thuwal, Kingdom of Saudi Arabia

April, 2021 2

EXAMINATION COMMITTEE PAGE

The thesis of Gloria Lisbet Gil Ramos is approved by the examination committee.

Committee Chairperson: Dr. Michael Lee Berumen Committee Members: Dr. Susana Carvalho, Dr. Raquel Peixoto

ABSTRACT

Spatial dynamics of Red Sea coral reef fish assemblages: a taxonomic and

ecological trait approach

Gloria Lisbet Gil Ramos

Despite the increases in the intensity and frequency of disturbances on coral reefs in the Red Sea over the past decade, patterns of variability in fish communities are still poorly understood. This study aims to contribute to a better understanding of how fish communities vary along multiple spatial scales (10-100’ of kilometers) and to provide a baseline for future comparisons, fundamental to assess responses to climate change and other disturbances. Coral reefs along the Saudi Arabian Red

Sea coast were surveyed from 2017 to 2019. The reefs ranged from 28° N to 18 °N and were categorized according their geographical location and grouped within three regions, namely north (24-28.5°N; 12 reefs), central (20.4-22.3°N; 11 reefs), and south (18.5-21.2°N; 30 reefs). The quantification of spatial patterns was conducted based on both taxonomic- and trait-based approaches. Considering the dependence of fish communities on the benthic habitat the relationship between different attributes of the fish assemblages and coral cover was also investigated. A consistent pattern of separation between assemblages of the northern and central region from the ones in the south was observed in nearshore reefs but was not evident for offshore reefs. The southern region supported higher densities, biomass, 5 and species richness than the other two regions. The analysis showed that transect and reef scales contributed to the greatest variation in fish communities, suggesting higher levels of variability within small spatial scales. Several parameters of the fish community (total species, total density, total biomass, total functional entities, functional richness, functional redundancy) were positively correlated to coral cover, particularly in the northern region. Responses were not consistent across the Red

Sea basin, suggesting that management plans should be regionally based. This study can be helpful to design management strategies as it provides a current baseline from both taxonomic and trait perspectives for Red Sea reefs that can be used to evaluate future changes due to natural and human-based disturbances.

AKNOWLEDGMENTS

My deepest thanks go to Dr. Michael Berumen for believing in me, for his help, support, and for enforcing an environment of solidarity and cooperativeness in the

Lab and facilitating many expeditions in the Red Sea. Thanks go next to Dr. Susana

Carvalho, who has been so supportive every single step of the way. Thank you,

Susana, for your patience, always encouraging us, and sharing your time and knowledge, and basically for everything. I could never find enough words to tell you how grateful I am to have been under your guidance.

I want to say thanks to João for all his help with the analysis and explanations and to all the lab members who helped in the data collection, particularly to Dr. Darren

Coker, who always was willing to help and giving helpful feedback: thank you,

Darren, for all your input. Also, thank you very much, Lucía, for your time and help with the functional analysis.

Of course, I also want to thank my family, mainly my parents Livier and Martín, and my sister Marisol who have always been so supportive and cheerful and always tell me to “echarle ganas”. This is a tribute to all of you.

I could not finish this section without saying thanks to all my friends who have been with me in every step of this journey. Thank you, Karla, Laura, Anuj, Julian, Bogdan,

Safwan and Abdullah. You are the ones that constantly motivate me to do and to be better. I love you all so much.

-Thank you all.

TABLE OF CONTENTS

EXAMINATION COMMITTEE PAGE ...... 2 COPYRIGHT PAGE…………………………………………………………………..…..3 ABSTRACT ...... 4 AKNOWLEDGMENTS ...... 6 TABLE OF CONTENTS ...... 7 LIST OF ABBREVIATIONS ...... 8 LIST OF FIGURES ...... 9 LIST OF TABLES ...... 12 INTRODUCTION ...... 14 1.1 Fish and habitat relationships...... 15 1.2 Scales of variability in the fish communities ...... 16 1.3 Fish and the functional roles they play ...... 17 1.4 Reef fish communities of the Red Sea ...... 19 1.5 Objectives ...... 21 1.6 Expected results ...... 22 MATERIALS AND METHODS ...... 23 2.1 Study sites ...... 23 2.2 Data collection...... 27 2.3 Data analysis...... 30 RESULTS ...... 35 3.1 General descriptors of the fish community ...... 35 3.2 Multivariate Community Patterns ...... 37 3.3 Components of variability in the community in the fish community ...... 59 3.4 Relationships between the fish variables and Hard Coral cover ...... 60 DISCUSSION ...... 69 REFERENCES ...... 82 APPENDICES ...... 91

LIST OF ABBREVIATIONS

CRS-Central Red Sea

GBR-Great Barrier Reef

MC-Monte Carlo Test

NRS-Northern Red Sea

PCO-Principal Coordinate Analysis

PERMANOVA-Permutational Multivariate Analysis of Variance

SIMPER-Similarity Percentages

SRS-Southern Red Sea

LIST OF FIGURES

Figure 2.1. Survey sites along the Saudi Arabian Red Sea Coast……………..……25 Figure 3.1. Fish community indices for the 53 reefs surveyed…………………...….36 Figure 3.2. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index……………………………………...37 Figure 3.3. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in nearshore reefs of three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index……………….....40 Figure 3.4. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in midshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index…………………...…...41 Figure 3.5. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in offshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index…………………...…...42 Figure 3.6. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index……………..………………….…...43 Figure 3.7. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in the nearshore reefs of three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index…..……45 Figure 3.8. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in the midshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index………………….46 Figure 3.9. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in the offshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index…………….……47 Figure 3.10. Relationship between the number of species and their respective ecological traits……………..…………………………………………………………….48 Figure 3.11. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in assemblages in three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index…………50 Figure 3.12. Principal Components Analysis (PCO) showing the dissimilarity ecological trait diversity in assemblages in nearshore reefs of three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index…………………………………………………………………………....51 10

Figure 3.13. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in assemblages in midshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index……………………………………………………………………………………….52 Figure 3.14. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in assemblages in offshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index……………………………………………………………………………………….53 Figure 3.15. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index………………………………………………………………………………….……54 Figure 3.16. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in nearshore reefs of three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index………………………………..…………………………………..….…...55 Figure 3.17. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in midshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index……..……………………………………………………………….…….56 Figure 3.18. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in offshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index…………………………………………………….……………………...57 Figure 3.19 Number of species as a function of the rank of Functional Entities….58 Figure 3.20. Components of variation in the fish community……………..………....59 Figure 3.21. Total species richness as a function of percentage cover of hard coral in three regions of the Red Sea……………………………………………………….…61 Figure 3.22. Total biomass as a function of percentage cover of hard coral in three regions of the Red Sea……………………………………………………………..……62 Figure 3.23. Total density as a function of percentage cover of hard coral in three regions of the Red Sea…………………………………………………………………..63 Figure 3.24. Total functional entities as a function of percentage cover of hard coral in three regions of the Red Sea…………………………………………………………64 Figure 3.25. Functional richness as a function of percentage cover of hard coral in three regions of the Red Sea……………………………………………………………65 Figure 3.26. Functional divergence as a function of percentage cover of hard coral in three regions of the Red Sea………………………………………….……………. 66 11

Figure 3.27. Functional vulnerability as a function of percentage cover of hard coral in three regions of the Red Sea……………………………………………….……… 67 Figure 3.28. Functional redundancy as a function of percentage cover of hard coral in three regions of the Red Sea…………………………………………………..……78

LIST OF TABLES

Table 2.1. Information of the 53 Saudi Arabian Red Sea coral reefs surveyed….....26 Table 2.2. Fish ecological traits…………………………………………………………30 Table 3.1. Results of a three-factor nested permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish species along three regions of the Red Sea……….…38 Table 3.2. Results of the pairwise test of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish species in three regions of the Red Sea………….…...38 Table 3.3. Similarity results of reef fish density in assemblages in three regions of the Red Sea using SIMPER…………………………………………..……….………. 39 Table 3.4. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish species in nearshore reefs of three regions of the Red Sea………41 Table 3.5. Results of a three-factor nested permutational multivariate analysis of variance (PERMANOVA) on square root transformed biomass of fish species along three regions of the Red Sea………………………………………………………....…43 Table 3.6. Results of the pairwise test of permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish species in three regions of the Red Sea………..……...44 Table 3.7. Similarity results of reef fish assemblages in three regions of the Red Sea using SIMPER………………………………………………………………….…………44 Table 3.8. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish species in nearshore reefs of three regions of the Red Sea………46 Table 3.9. Results of a three-factor nested permutational multivariate analysis of variance (PERMANOVA) on square root transformed density of trait diversity of fish along three regions of the Red Sea………………………………………………….….50 Table 3.10. Results of the pairwise test of permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish traits in three regions of the Red Sea…………..…….51 Table 3.11. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish traits in nearshore reefs of three regions of the Red Sea…………52 Table 3.12. Results of a three-factor nested permutational multivariate analysis of variance (PERMANOVA) on square root transformed biomass of trait diversity of fish along three regions of the Red Sea………………………………………..……………54 13

Table 3.13. Results of the pairwise test of permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish traits in three regions of the Red Sea…………………..55 Table 3.14. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish traits in nearshore reefs of three regions of the Red Sea………….56 Table 3.15. Responses of the fish community to hard coral cover (%) in three regions of the Red Sea……………………………………………………………………………60

INTRODUCTION

Coral reefs are a unique environment that, while occupying only 0.1% of the surface of the ocean, are one of the most valuable ecosystems on Earth, displaying biodiversity levels compared with the tropical rainforests (Connell, 1978; Reaka-

Kudla, 1997). These high levels of biodiversity are mainly supported by scleractinian corals (hard corals), which are foundational species that not only provide the grounds for building a diverse, complex, and heterogenous physical structure, but at the same time, provide food and shelter for a wide array of taxa (Coker et al., 2014; Cole et al., 2008). Moreover, coral reefs are known to support the livelihoods of more than

500 million people globally (Spalding et al., 2017).

It is well documented that approximately one quarter of all marine fish rely on coral reefs for food, shelter, and settlement (Spalding et al., 2001; Allen, 2008). In fact, fish communities reach their peak of diversity within these ecosystems, although they vary within and between reefs within the same geographic area and geographic regions (Harmelin-Vivien, 2002). For instance, different regions support different number of species; for example, the central Indo-Pacific supports more than 3,600 species, whereas the Central Indo-Pacific nearly 3,000 species. Values reported for the Western Indian Ocean are 2,241, whereas the Tropical Eastern Pacific possesses 570 species, the Western Atlantic approximately 900 species, and the

Eastern Atlantic 403 species (Mouillot et al., 2014).

1.1 Fish and habitat relationships

Since most reef fishes spend their juvenile and adult stages near the bottom of the reef (benthic or benthopelagic), benthic habitat features have been investigated to understand better the spatial distribution and abundance of fish (García-Charton &

Pérez-Ruzafa, 2001). More complex habitats provide more ecological niches, refugia, and feeding sources, which may attenuate the role of biological processes such as competition and predation (Cole et al., 2008). Other studies have specifically focused on the influence of coral attributes (e.g., coral cover, coral diversity) in the fish community. Fish species richness has been reported to be positively correlated with both coral species richness (Komyakova et al., 2013) and coral cover (Chabanet et al., 1997; Bozec et al., 2005). Indeed, there is evidence of a decline in fish species richness and functional groups along gradients of decreasing structural benthic complexity (Chong-seng et al., 2012). Also, changes in the structure of fish assemblage have been documented in response to a decline of coral cover (Wilson et al., 2006).

Habitat alterations are likely to influence differentially fish species present in complex systems like coral reefs. Wilson et al. (2006) found that the loss of coral cover was most severe on corallivores, which decreased in abundance. In some cases, it seems that the strongest correlations are detected for obligate coral feeders in response to low coral cover (Bell et al., 1985). Moreover, within fish species that rely on corals, i.e., corallivores and coral dwellers (e.g., gobies, damselfishes), species- specific responses can be found depending on the coral genera and their specific 16 attributes. Habitat-fish association is species dependent and the strength of such relation is influenced by the way that the species utilizes their coral host (Coker et al., 2014).

1.2 Scales of variability in the fish communities

Coral-reef fish communities reach their highest levels of diversity in these ecosystems. However, they vary considerably as a result of a myriad of potential factors (Harmelin-Vivien, 2002). Reef-associated fish assemblages respond to changes in environmental conditions as a result of various processes that operate at different spatial and temporal scales. On a local scale, fish diversity can be driven by ecological processes such as competition for food or shelter, predation, recruitment, and perturbation (Harmelin-Vivien, 1989). However, on the regional scale, the composition of reef fish assemblages can be explained by the processes of larval dispersal and hydrodynamic patterns and tolerance to environmental variables, whereas on a global scale, fish diversity can be explained by the evolutionary history of the region (Floeter et al., 2001; Harmelin-Vivien, 2002).

In order to understand patterns of distribution and abundance of fish, it is crucial to investigate the spatial scale of observation (Anderson and Millar, 2004; Underwood et al., 2000). For instance, when studying large predators (e.g., sharks), the study areas typically occupy whole reefs (few kilometers), to bays (hundreds of km) or even whole basins (thousands of kilometers) (Holmes et al., 2017; Lea et al., 2015; 17

Munroe et al., 2015). In contrast, for the small site attached fish, survey methods can be quadrats as small as 1 square meter (Troyer et al., 2018). The extent of variation in every spatial scale can provide insight into the relevant processes for reef fishes

(Jones & Syms, 1998). Despite these complications, some large-scale patterns emerge even when comparing data from various methods (e.g., biogeography-scale studies).

1.3 Fish and the functional roles they play

Fish is the most diverse group of vertebrates with more than 30,000 valid species and exhibits a large diversity of biological and ecological characteristics. They represent an essential part of the ecosystems as they contribute to critical processes such as food chain dynamics and nutrient cycling. Likewise, they are also key in the connectivity between ecosystems (e.g., seagrass and coral reefs) and are fundamental in the maintenance of genetic, species, and ecosystem biodiversity

(Holmlund & Hammer, 1999). In coral reef communities, fish is one of the most conspicuous groups. They play crucial roles in several biogeochemical fluxes and are fundamental in maintaining the resilience of this ecosystem and a link between humans and marine ecosystems (Allgeier et al., 2014).

Considering that coral-reefs are currently one of the most vulnerable ecosystems as a result of, among others, overexploitation, pollution, and climate change, there is a growing concern regarding the response of their communities to disturbances

(Pandolfi et al., 2003; Bellwood et al., 2004) These responses are essential to take into consideration to determine their implications on the functioning of the 18 ecosystems and, ultimately, what management actions should be implemented

(Villéger et al., 2017).

As ecosystem function is affected by the biological and ecological attributes of organisms, rather than by their taxonomic identity (Hooper et al., 2006), a trait-based approach can offer comparative advantages to help predict ecosystem functioning

(Gagic et al., 2015). Although this approach has its limitations, it does offer a way to evaluate broad-scale trends that may be functionally relevant. Even if there are detectable shifts in taxonomic community composition, it may be that the ecological functions or processes are stable through time or space. It is therefore critical to consider community function metrics in assessing ecosystem health. This approach has been increasingly applied in response to the current face of the rapid degradation of coral reefs (Bellwood et al., 2019). Traits (i.e., species attributes) can be obtained from online resources such as FishBase for many species and the literature (Bellwood et al., 2019) or even from expert judgment. Within the most commonly used traits are size (maximum adult size), trophic group, mobility, grouping behavior, activity patterns, and position in the water column (Mouillot et al.,

2014; Stuart-Smith et al., 2013). These traits have been reported to be informative regarding fish ecology (Mouillot et al., 2014).

1.4 Reef fish communities of the Red Sea

The Red Sea is regarded as a unique environment due to its relatively extreme conditions of high salinity and temperature (Voolstra & Berumen, 2019). Yet, this narrow basin harbors one of the largest reef systems on the planet and is home to more than 1100 fish species and more than 300 species of scleractinian corals

(DiBattista et al., 2016; Bogorodsky & Randall, 2018).

There have been a few studies focused on regional trends and patterns on the assemblage structure of fish on the Red Sea (Roberts et al., 1992; Roberts et al.,

2016). However, they have been focused on large spatial scales without considering the scale of variability in the fish community (from meters to hundreds of kilometers) or trait fish diversity (but see Pombo-Ayora et al., 2020 in Central Red Sea).

Fish community composition on the coast of the Red Sea has been shown to be different when comparing the west (Sudan) and the eastern coast (Saudi Arabia).

Furthermore, biomass estimates have also been proved to change significantly, with the reefs in Sudan yielding 1.04 tones ha-1 more than those in the Saudi Red Sea

(Kattan et al., 2017), probably due to the removal of predators as a result of higher fishing pressure in Saudi Arabia.

Studies conducted on specific fish taxa (butterflyfishes and angelfishes) in nearshore reefs along the latitudinal axis of the Red Sea have found regional changes in the species abundance and assemblage composition (Roberts et al., 1992). The authors reported that the abundance was greater in the central Red sea, decreasing towards 20 the extremes of the basin, and the changes in the assemblages were more conspicuous from the central to the north and the southern Red Sea to the Gulf of

Aden. However, other studies restricted to offshore reefs have not found significant differences in the fish assemblages in terms of abundance, species richness, and diversity indices along the latitudinal gradient of the Red Sea (Roberts et al., 2016).

These authors also reported that only a few common species (e.g., Pseudianthias squamipinnis, Chromis dimidiata, Chromis flavaxilla, Pseudochromis fridmani) tended to drive the similarities among regions. Changes in fish community composition have also been found across a cross-shelf gradient in the central Saudi

Arabian Red Sea, with nearshore reefs being markedly different from midshore and offshore reefs (Khalil et al., 2017).

Since Saudi Arabia is a country that depends to a large extent on the Red Sea as a fishing resource and more recently as a destination for marine tourism (Vision 2030)

(https://www.vision2030.gov.sa/), it is important to assess the changes in the fish community and their response to the benthic conditions along with different spatial scales. Furthermore, it is also important to describe the biological traits of the fish, which can be a proxy for understanding the functioning of the reefs in the Red Sea, as trait-based analyzes have been shown to provide a better resolution resolving spatial and temporal patterns compared to taxonomic approaches (DeVictor, 2010).

Furthermore, traditional taxonomic metrics are not always related to the ecological functions of the reef fish (Cadotte et al., 2011)). This information can be considered for future managing plans, which are fundamental in periods of frequent disturbances on coral reefs (Bellwood et al., 2004). 21

1.5 Objectives

This thesis aims to identify differences in Red Sea fish assemblages in three regions of the Red Sea. Particularly how attributes of the fish community structure (based on taxonomic and trait approaches) change across multiple scales (from transect, i.e.,

<20m, to the scale of the Red Sea basin, i.e., >1000km). Taking this into consideration, this study aims to identify the most relevant scales of variability in the

Red Sea fish community and investigate whether taxonomic and trait approaches show consistent patterns. Furthermore, considering the pivotal role of hard corals and the ongoing degradation, namely due to more frequent and long-lasting/intense bleaching events (Furby et al., 2013; Monroe et al., 2018), the relationships between hard coral cover and specific attributes of the fish assemblages will be investigated.

This study may benefit the Kingdom as it aims to provide useful insight regarding the fish assemblages and their relationships with coral reefs (e.g., hard corals) and can be helpful to inform the design of monitoring plans, identify sensitive functional groups, and sensitive areas for management. Providing a current baseline from both taxonomic and trait perspectives for Red Sea reefs that can be used to evaluate future changes due to natural and human-based disturbances

1.6 Expected results

Based on previous findings, I expect that there will be changes in the fish community composition among the regions analyzed, which may be clearer in nearshore reefs

(sensu Roberts et al., 1992). Furthermore, there will be a significant variation in the density and biomass of the fish assemblages in the scales analyzed (Tuya et al.,

2011). I further expect that coral cover will be positively correlated to fish variables such as biomass, density, and trait diversity. Finally, I predict that reef communities will exhibit changes in diversity at the regional scale; specifically, in areas impacted by bleaching I would expect to have lower diversity than the regions where coral cover is higher.

MATERIALS AND METHODS

2.1 Study sites

A total of 53 reefs along the Saudi Arabian Red Sea coast were surveyed from 2017 to 2019. The reefs surveyed ranged from 28° N to 18 °N and were categorized according to their geographical location and grouped within three regions, namely

North (NRS), Central (CRS), and South (SRS), that fairly coincide with the designations of Sheppard and Sheppard (1991), Raitsos et al., 2013 and Pearman et al., 2019. Survey sites ranged from ~300 m to ~70 km from shore and were classified into three categories according to their position in the continental shelf: i) nearshore reefs (nearest to shore and with adjacent waters approximately 20 m deep); ii) midshore reefs (located between nearshore and offshore reefs, characterized by neighboring waters ranging from 50 to 200 m deep) and iii) offshore reefs (farther apart from shore and surrounded by waters deeper than 200 meters)

(following Khalil et al., 2017). The surveys were carried out in either the exposed or sheltered site of the reef, with 31 reefs surveyed on the exposed side and 22 surveyed on the sheltered margin (figure 2.1, table 2.1). 24

30°0'0"N

### Survey sites # # 2017-2019 NRS

### 25°0'0"N ## ______# Egypt Saudi Arabia

a e S d e R " CRS "" " ""

______" !!! 20°0'0"N !! Sudan !!!!!!! !!!!! SRS !!!!! ! ! !!

Location

# Duba # Um Lujj Eritrea " Jeddah 15°0'0"N Yemen " Thuwal ! Al Lith ! Farasan Banks Ethiopia

± Djibouti 0 125 250 500 Kilometers 35°0'0"E 40°0'0"E 45°0'0"E

Figure 2.1 Survey sites along the Saudi Arabian Red Sea Coast. Every reef surveyed is represented with a dot. Triangles represent the Northern Red Sea (NRS), squares the Central Red Sea (CRS), and circles the Southern Red Sea (SRS). Locations of the reef surveys are represented in dark blue (Duba), light blue (Um Lujj), pink (Jeddah), light purple (Thuwal), light green (Al Lith), and dark green (Farasan Banks). 25

Table 2.1- Information of the 53 Saudi Arabian Red Sea coral reefs surveyed. Surveys were conducted in three regions of the Red Sea and six locations. The locations were comprised of four (Thuwal), six (Duba, Um Lujj), seven (Jeddah, Al Lith) and 23 (Farasan Banks) reefs. Latitude and Longitude represent the geographic location of every reef. The shelf column enlists the position in the shelf of every reef. The exact survey date is found in the last column; 11 reefs were surveyed a second time that was considered as a replicate. Asterisk represents benthic surveys.

Region Location Reef Latitude Longitude Shelf Survey Date North Duba DR17 28.0841 35.00462 Nearshore 20/03/2017* Duba DR16 27.99347 34.95137 Nearshore 21/03/2017* Duba DR14 27.94567 35.16775 Nearshore 24/03/2017* Duba DR11 27.78145 35.4078 Nearshore 24/03/2017* Duba DR12 27.68122 35.44215 Nearshore 03/11/2018 Duba DR7 27.27333 35.64228 Nearshore 03/11/2018* Um Lujj AWBR1 25.46301 36.74962 Nearshore 10/02/2017* Um Lujj DR4 25.3542 36.8914 Nearshore 26/03/2017* Um Lujj DR13 25.35088 37.0707 Nearshore 08/02/2017 Um Lujj DR3 24.98674 36.94556 Nearshore 10/02/2017* Um Lujj DR2 24.96507 37.04622 Nearshore 15/02/2017* Um Lujj DR1 24.76122 37.08896 Nearshore 07/02/2017* Central Thuwal ASH 22.2983 39.04632 Nearshore 21/08/2017* 12/05/2019* Thuwal AFH 22.2294 38.96397 Midshore 20/08/2017* 11/06/2019* Thuwal AST 22.13872 38.96758 Midshore 20/08/2017* Thuwal ABM 22.09045 38.7788 Offshore 21/08/2017* 28/11/2019* Jeddah JPR2 21.47778 39.12056 Nearshore 18/06/2019* Jeddah JDH1 21.45422 39.11102 Nearshore 01/07/2017* 16/06/2019* Jeddah JR14 21.41361 39.11333 Nearshore 18/06/2019* Jeddah JR18 21.39833 39.10611 Nearshore 24/06/2019* Jeddah JDH2 21.22513 39.12062 Nearshore 01/07/2017* 26/06/2019* Jeddah JDH3 21.08202 39.20105 Nearshore 01/07/2017* 30/06/2019 Jeddah AL5 20.49567 39.63635 Nearshore 05/08/2017* 17/04/2019* South Al Lith AL1 20.17192 40.16915 Nearshore 08/12/2019* Al Lith AL2 20.14762 40.23765 Nearshore 09/08/2017 08/12/2019* Al Lith AL6 20.13535 40.09597 Midshore 10/12/2019* Al Lith AL7 20.12233 40.21752 Nearshore 04/08/2017* 19/04/2019* Al Lith AL8 20.11077 40.21985 Nearshore 07/08/2017* 08/12/2019*

Survey Location Reef Latitude Longitude Shelf Region Date

South Al Lith AL4 20.08981 40.30896 Nearshore 08/12/2019*

Al Lith AL3 19.90703 40.52297 Nearshore 18/04/2019*

Farasan FB14 19.79339 40.39909 Midshore 16/02/2019 Banks

Farasan FB15 19.78995 39.95045 Offshore 26/04/2019* Banks

Farasan FB1 19.78952 40.14564 Midshore 06/02/2019 Banks

Farasan FB16 19.74922 39.9039 Offshore 26/04/2019* Banks

Farasan FB13 19.67271 40.4433 Midshore 16/02/2019 Banks

Farasan FB12 19.62758 40.59488 Nearshore 15/02/2019 Banks

Farasan FB31 19.61235 40.39762 Nearshore 04/05/2019* Banks

Farasan FB30 19.6066 40.66793 Nearshore 04/05/2019* Banks

Farasan FB17 19.56185 39.9974 Offshore Banks 27/04/2019*

Farasan FB24 19.42237 40.40557 Midshore 30/04/2019 Banks

Farasan FB18 19.41827 40.16267 Offshore 27/04/2019 Banks

Farasan FB9 19.39212 40.86092 Nearshore 12/02/2019 Banks

Farasan FB25 19.38982 40.54957 Nearshore 30/04/2019 Banks

Farasan FB22 19.2729 40.4419 Midshore 29/04/2019 Banks

Farasan FB23 19.25923 40.67475 Midshore 30/04/2019 Banks

Farasan FB19 19.24097 40.15217 Offshore 28/04/2019 Banks

Farasan FB20 19.18812 40.26937 Midshore 28/04/2019 Banks 27

Farasan FB26 19.1775 40.80743 Nearshore 01/05/2019* Banks

Farasan FB21 19.12263 40.58707 Offshore 29/04/2019 Banks

Farasan FB8 18.97358 40.97324 Midshore 11/02/2019 Banks

Farasan FB4 18.81844 40.64141 Offshore 09/02/2019 Banks

Farasan FB6 18.80802 40.89627 Midshore 10/02/2019 Banks

Farasan FB5 18.74263 40.93583 Midshore 10/02/2019 Banks

2.2 Data collection

2.2.1 Fish surveys

At each reef, three replicate belt transects were laid at 10 meters depth in order to standardize across sites. As data were collected under the framework of different projects, two belt lengths were used: i) 20 m length or ii) 50 m length. The width of each belt transect varied according to the size and activity of the fishes: smaller sedentary fish had a narrower width (2m and 4m), whereas larger mobile fish had a transect width of 5m and 8m (following Sandin et al., 2008). Most of the surveys

(62.5%) were conducted using belt parameters ii). A diver swimming along the replicate transects identified the fish to the lowest taxonomic level possible and registered their abundance. Size estimates were also recorded per fish and later used to calculate biomass following the allometric length-weight conversion W= aLb, where W is weight in kilograms, L is total length in centimeters, and a and b are species-specific constants that can be obtained from the online resource FishBase

(Friedlander and DeMartini, 2002; Froese and Pauli, 2016). 28

To account for the variability in the belt transect area, count data are reported as density (individuals m-2) and biomass as kg m-2. All fish species recorded are listed in Appendix 1.

2.2.2 Benthic surveys

In order to investigate the responses of the fish community to the benthic composition, 34 reefs were selected from the original set of 53 reefs (please refer to table 1). A picture (1 m2) was taken at every meter of the transects laid for the fish surveys. These pictures were further analyzed, and the benthic groups in every photo were identified and quantified using the software Coral Point Count with Excel extensions (CPCe; Kohler and Gill, 2006) with 48 random points. A total of 90 taxonomic groups were identified, and their percentage cover was calculated. From all the categories, only coral cover was used.

2.2.3 Fish ecological traits

Following Mouillot et al., 2014, every fish species registered in the reef surveys were characterized with six traits that describe the main aspects of fish ecology. The trait information was retrieved from online datasets and available literature and are described in table 2.2.

Table 2.2. Fish ecological traits. These traits describe the main facets of fish ecology.

Trait Classification Description Diet Planktivore Feeds on plankton Invertivore Feeds on invertebrates Pelagic omnivore Generalist, living in water column Benthic omnivore Generalist, living in the benthos

Grazer Feed on macroalgae and epilithic algae Piscivore Feeds on fish Corallivore Feeds on corals Browser Feed on macroalgae General Carnivore Feed on meat (e.g., fish, invertebrates) Excavator Take deeper excavating bites on coral surfaces Scraper Consume epilithic algae and remove sediment Feeds on crustacean ectoparasites and mucus of Ectoparasite other fishes. Common adult S1 Individuals from 0-7 cm length S2 7-15 cm S3 15-30 cm S4 30-50 cm S5 50-80 cm S6 More than 80 cm Mobility Sedentary Species staying in a restricted area (~ 100 m2) Mobile within the reef Travel large distances over a reef Mobile between the Species which frequently change reefs reef Position in the Benthic Species staying on the bottom at all times water column Species living above the bottom but may at time rest Bentho-pelagic on the bottom Species spending most of their active time on the Pelagic water column Activity Diurnal Fish active during daytime hours Diurnal and nocturnal Fish active both in day and night Nocturnal Fish active on night hours Gregariousness Solitary 1 individual Pair 2 individuals Small groups 3 to 20 individuals Medium groups 20 to 50 individuals Large groups More than 50 individuals

With these traits, every fish species was assigned a “functional entity”, which is a code made of a series of characteristics that can be unique to a species or shared with other species. This code was used together with the species’ density and biomass (separately), species with the same code were unified and their biomass or 30 density added together for every transect. This information was used for the multivariate analysis that will be described in section 2.3.6.

2.3 Data analysis

2.3.1 Community indexes

From the data obtained from the surveys, estimates of species richness, evenness, and diversity were calculated. Species richness was determined by the total number of species recorded on each transect. Species diversity was calculated using the

Shannon’s diversity index (H’). To determine the evenness of the diversity, Pielou’s evenness (Jʹ) was used. Values of Jʹ lie between 0–1, where 0 indicates an uneven distribution of abundances among species within a site, and values closer to 1 show more evenly composed communities. Species richness, Pielou’s evenness, and

Shannon diversity indexes were calculated at the level of the transect using

PRIMER-v6 (Clarke and Gorley, 2006).

2.3.2 Functional diversity indexes

Following the methodology proposed by Mouillot et al., (2013), five functional indexes were calculated and plotted with the six ecological traits described in section

2.2.3, using the package FD using the software R.

1.- Functional richness. Functional richness is expressed in terms of both the number of functional entities and the functional volume occupied by the species of the community (Muillot et al., 2005; Mouillot et al., 2014). It is orthogonal to both species richness and species evenness, therefore functional richness may remain 31 unchanged or increase as species richness increase. High functional richness is considered as low (0.5) or high (0.857). Low functional richness indicates that the niches (resources) available to the community are unused (Mason et al., 2005).

2.- Functional divergence. It defines how the abundance is distributed in a functional trait space volume. High values indicate that there is a high niche differentiation between the species which can happen when high abundant species exhibit extreme traits (away from the center) (Mason et al., 2005).

3.-Functional redundancy. the degree to which organisms have evolved to do similar things. It is calculated as the ratio between the total number of species and the total functional entities in an assemblage.

4.- Functional vulnerability. Is the percentage of entities represented by only one species (Mouillot et al., 2014).

5.- Functional over-redundancy. Is the percentage of functional entities that are over represented i.e., that have more species than expected from functional redundancy

(Mouillot et al., 2014).

Because the matrix built with this information contains both categorical and continuous data, several transformations had to be done. Continuous data were standardized using the function Scale in R (R Development Core Team, 2013).

Furthermore, Gower distances were calculated with the function Daisy from Cluster package (Gower, 1971; Maechler et al., 2013). This data transformation gives the same weight to each trait (Legendre and Legendre, 2003). Finally, with the density 32 of individuals and their presence or absence in each zone, the indexes were calculated using the software R. 4.0.3 (R Development Core Team, 2013).

2.3.3 Components of variability in the fish community

With the purpose of investigating the spatial scale with the highest variability, univariate attributes of the reef fish assemblage included species density, biomass and density and biomass of functional entities were used to perform Permutational

Multivariate ANOVA (PERMANOVA), using the factors “region” (fixed), “location”

(fixed, nested within region) and “site” (random, nested within location and region).

The α value was set at 0.05 value. Pair-wise comparisons between each pair of regions were executed as well as the Monte Carlo test. The contribution of each scale to explain the total variation in univariate and multivariate fish responses were estimated by calculating their variance components and expressing the values of total explained variation. All analyses in this section were carried out in the software

PRIMER v6 (Clarke and Gorley, 2006).

2.3.4 Multivariate analysis

In order to understand the patterns of distribution of the fish communities (general, by region and by shelf position), Principal Coordinates Analysis (PCO) were performed to visualize the relative dissimilarity distances based on community compositions among the 53 reefs, using the density, biomass and the density and biomass of fish entities at the level of the reef (site). PCO were used because it 33 reduces the dimensions of a complex data set and can be used to visualize it. The first principal component accounts for as much of the variability in the data as possible, and each succeeding component accounts for as much of the remaining variability.

All the data was square root transformed prior to analysis to reduce the effect of the most abundant species. Every PCO was calculated using the Bray Curtis resemblance matrix (Bray and Curtis, 1957; Clarke and Gorley, 2006). This resemblance matrix is commonly used in studies of biological communities and has shown to be sensitive to differences in community structure when using species abundance data.

Following this analysis, and in order to determine the species that most contributed to the differences between the regions and shelf positions, a pairwise comparison of similarity percentages (SIMPER; Clarke and Warwick, 2001) of abundance data was used to determine which fish species and contributed most to the differences between regions and between shelf positions. This analysis first identifies the species that are most influential in characterizing an area (region/shelf) and then ranks the species that drive average dissimilarity between them.

2.3.5 Linear regressions between hard corals and fish data

In order to investigate the relationship between parameters of the fish community and hard coral cover at the level of the transect, total biomass, total species and total density in addition to the indexes calculated in section 1.3.2 were used to perform 34 linear regressions. All linear regressions were done for every region (north, center, and south) using the data collected in the subset of 34 reefs (see section 2.2.2) using the software R. 35

RESULTS

3.1 General descriptors of the fish community

A total of 53627 fish were registered, representing 41 families, 113 genera, and 220 species (Appendix 1). Families Labridae and Pomacentridae recorded the highest richness with 35 and 26 species, respectively, followed by the families Serranidae

(17 species), Scaridae (15 species), and Chaetodontidae (11 species). These five families accounted for 47% of the total species recorded. The species that registered more individuals per m2 were Chromis dimidiata (141), Pseudanthias squamipinnis

(74), Chromis flavaxilla (39), Amblyglyphidodon flavilatus (18), and Pseudochromis fridmani (18), that altogether contributed to the 51.86% of the total density. In contrast, the grouper Epinephelus fuscoguttatus, the porcupinefish Diodon hystrix, and the wrasses Cheilinus undulatus and Cheilinus lunulatus registered the lowest density, with 0.04 ind. m-2 each. In the case of the biomass, the species that most contributed to the total were the fusiliers Caesio lunaris and Caesio striata, the surgeonfishes Acanthurus sohal, and Ctenochaetus striatus, and the unicornfish

Naso elagans, which accounted for 28.21%.

The NRS reefs had the highest mean values in density and biomass, although they had higher variability than the other two regions. Regarding species richness and

Shannon diversity index (average), CRS was higher than the other two regions (NRS

S=28, CRS S=35, SRS S=32) (Figure 3.1). However, absolute values of species richness for the regions showed that the SRS had 185 species, whereas the NRS and the CRS registered 146 and 144 species, respectively. 36

A NRS

CRS Density

SRS

H’ Biomass

J’ S

Figure 3.1. Fish community indices for the 53 reefs surveyed. The reefs are allocated within 3 regions along the latitudinal axis of the Red Sea, denominated: north (NRS) (n=12), central (CRS) (n=11) and southern (SRS) (n=30). Each reef was comprised of three transects. Mean density (A) and biomass (B) of the recorded individuals per reef and year. C) Total species richness recorded at every reef. D)Mean Shannon diversity index (H’) (log (e)). Mean Pielou’s evenness values. Every boxplot represents the median (transversal line on every box) with the interquartile range. Whiskers show the upper and lower extremes of the values.

3.2 Multivariate Community Patterns

3.2.1 Taxonomy: fish density

The PCO analysis of the fish density illustrates the differences in community structure among the three analyzed regions (Figure 3.2). The fish assemblage in the Central Red Sea reefs shows more similarity (SIMPER average similarity 40.75 %). On the other hand, the southern reefs have more variation(SIMPER average of similarity of 32.58 %). Northern reefs show more variation than the central but less than those in the south (SIMPER average similarity of 29.10 %). CRS is closer to the NRS, although certain reefs in the SRS also show similarity with them. Permutational multivariate analysis of variance (PERMANOVA) confirmed statistical differences among the regions (pseudo-f=4.46, p=0.001) (Table

3.1), with the pairwise analysis showing that all the pairs of regions are different

Transform: Square root (Table 3.2). Resemblance: S17 Bray Curtis similarity 40 PCO Density: all reefs Region NRS CRS SRS

) 20 n

o t

f 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (26.1% of total variation) Figure 3.2. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every symbol on the graph represents a reef. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS) and circles the ones in the Southern Red Sea (SRS). 38

Table 3.1. Results of a three-way nested permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish species along three regions of the Red Sea. Region is a fixed factor; location is nested within region; site is nested within location and region. df=degrees of freedom, MS=Mean Sum of Squares, Pseudo- F= Mean value by permutation P(perm)= p-value of permutations.

Source df MS Pseudo-F P(perm) Unique perms Region 2 24205 4.4593 0.001 997 Location (Region) 3 10862 1.9936 0.01 999 Site(Location(Region)) 47 5129.3 4.8018 0.001 990 Transect 139 1068.2

Total 191

Table 3.2. Results of the pairwise test of permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish species in three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms CRS, SRS 2.2818 0.002 999 CRS, NRS 1.8859 0.003 999 SRS, NRS 2.0554 0.001 998

Ten planktivore species were found to contribute with more than 30% of the dissimilarity between the regions. Of these species, Amblyglyphidodon flavilatus,

Amblyglyphidodon indicus, Chromis dimidata, Chromis flavaxilla, Dascyllus abudafur, Paracheilinus octotaenia, Plectroglyphidodon lacrymatus, Pomacentrus sulfereus, and Pseudanthias squamipinnis were consistently present in the three regions (Table 3.3).

Table 3.3. Results of the SIMPER analysis of density of fish assemblages in three regions of the Red Sea. The table shows the one-way SIMPER comparison between the regions. The top ten species are listed for each region, which contributed approximately a cumulative 30% of the dissimilarity (numbers in bold).

Regions compared Species Average Average Contribution Cumulative Abundance Abundance to % Region 1 Region 2 dissimilarity (%) CRS vs SRS Chromis dimidata 0.69 0.4 7.44 7.44 Average dissimilarity = Chromis flavaxilla 0.39 0.13 4.55 11.99 70.98 Amblyglyphidodon flavilatus 0.16 0.24 3.1 15.09

Pseudochromis fridmani 0.02 0.26 2.88 17.96

Pomacentrus sulfereus 0.36 0.17 2.81 20.77

Plectroglyphidodon lacrymatus 0.26 0.05 2.73 23.51

Pseudanthias squamipinnis 0.11 0.19 2.7 26.21

Paracheilinus octotaenia 0.02 0.23 2.67 28.88

Amblyglyphidodon indicus 0.2 0.12 2.33 31.21

Dascyllus abudafur 0.19 0.01 2.08 33.29 CRS vs NRS Chromis dimidata 0.69 0.73 8.32 8.32 Average dissimilarity = Chromis flavaxilla 0.39 0.2 4.57 12.9 71.03 Pseudanthias squamipinnis 0.11 0.43 4.19 17.09

Pomacentrus sulfereus 0.36 0.16 3.26 20.35

Amblyglyphidodon indicus 0.2 0.26 3.02 23.37

Plectroglyphidodon lacrymatus 0.26 0.02 2.78 26.16

Amblyglyphidodon flavilatus 0.16 0.18 2.7 28.86

Dascyllus abudafur 0.19 0.02 2.06 30.92

Chromis viridis 0.18 0.03 2.04 32.96

Caesio lunaris 0.06 0.16 1.86 34.82 SRS vs NRS Chromis dimidata 0.4 0.73 7.87 7.87 Average dissimilarity = Pseudanthias squamipinnis 0.19 0.43 4.86 12.73 74.83 Pseudochromis fridmani 0.26 0.1 3.09 15.82

Amblyglyphidodon flavilatus 0.24 0.18 3 18.82

Amblyglyphidodon indicus 0.12 0.26 2.88 21.69

Chromis flavaxilla 0.13 0.2 2.85 24.55

Paracheilinus octotaenia 0.23 0.03 2.83 27.38

Pomacentrus trichrourus 0.15 0.12 2.54 29.91

Pomacentrus sulfereus 0.17 0.16 2.34 32.25

Caesio lunaris 0.06 0.16 1.87 34.13

When analyzed by shelf position, nearshore reefs show a clear separation between the regions (Figure 3.3). Reefs in the CRS are more clustered together and similar to the NRS reefs. Lastly, reefs belonging to the SRS are separated from the reefs in the CRS and NRS. PERMANOVA results confirmed statistical differences among the regions (Table 3.4). Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Density: Nearshore reefs Region NRS CRS SRS

)

f 0

(

O C P -20

-40 -40 -20 0 20 40 PCO1 (23.2% of total variation) Figure 3.3. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef surveyed. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS) and circles the ones in the Southern Red Sea (SRS).

Table 3.4. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish species in nearshore reefs of three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms SRS, CRS 2.3015 0.001 999 SRS, NRS 2.5373 0.001 999 CRS, NRS 1.5005 0.01 999

In the midshore reefs, there appears to be more variation between them. Moreover,

PERMANOVA analysis did not find any differenceTransform: amongSquare root the regions (p=0.239) Resemblance: S17 Bray Curtis similarity 40 Region PCO Density: Midshore reefs CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 60 PCO1 (39.6% of total variation) Figure 3.4. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in midshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

Similarly, offshore reefs do not show a clear trend, and no statistical differences were found (p=0.553) (Figure 3.5). 42 Transform: Square root Resemblance: S17 Bray Curtis similarity 20 PCO Density: offshore reefs Region CRS

SRS

)

t a

i 0

(

2 -20

C P

-40 -40 -20 0 20 40 PCO1 (35.2% of total variation)

Figure 3.5. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in offshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Central Red Sea (CRS) reefs are represented in squares and Southern Red Sea (SRS) reefs in circles.

3.2.2 Taxonomy: fish biomass

The PCO analysis of the fish biomass illustrates the differences in community structure among the three analyzed regions (Figure 3.6). The fish biomass in the

Central Red Sea reefs show less variation between them (average similarity of

33.49%), and they are in proximity to the reefs on the north and nearly half of the reefs of the south. On the other hand, the SRS reefs show more variation than the ones in the center (Average similarity of 29.39%). However, the NRS reefs are the ones that exhibit more dissimilarity out of the three regions (average similarity of

23.42%). The Permutational multivariate analysis of variance (PERMANOVA) also 43 confirmed significant differences among the regions (pseudo-f=3.34, p=0.001)

(Table 3.5), with the pairwise analysis showing that all the pairs of regions are different (Table 3.6). Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Biomass: all reefs Region NRS CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 60 PCO1 (14.4% of total variation) Figure 3.6. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Squares represent the reefs in the Central Red Sea (CRS) and circles the ones in the Southern Red Sea (SRS).

Table 3.5. Results of a three-way nested permutational multivariate analysis of variance (PERMANOVA) on square-root transformed biomass of fish species along three regions of the Red Sea. Region is a fixed factor; location is nested within the region; the site is nested within location and region. df=degrees of freedom, MS=Mean Sum of Squares, Pseudo-F= Mean value by permutation P(perm)= p-value of permutations. P(MC)=Monte Carlo p-value.

Source df MS Pseudo-F P(perm) Unique P(MC) perms Region 2 17337 3.3387 0.001 999 0.001 Location (Region) 3 10783 2.0696 0.001 997 0.001 Site(Location(Region)) 47 4920.7 3.1657 0.001 996 0.001 Transect 139 1554.4

Total 191

Table 3.6. Results of the pairwise test of permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish species in three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms CRS, SRS 1.8929 0.001 997 CRS, NRS 1.5452 0.001 998 SRS, NRS 1.9338 0.001 997

Results of the SIMPER analysis show that ten species contribute with approximately

25% of the dissimilarity between the regions. Nine of these species are herbivores:

Acanthurus sohal, Cheilinus quinquecinctus, Chlorurus gibbus, Chlorurus sordidus,

Ctenochaetus striatus, Hipposcarus harid, Naso elegans, Scarus ferrugineus, and

Scarus niger were present in the three regions (Table 3.7).

Table 3.7. Results of the SIMPER analysis of biomass of reef fish assemblages in three regions of the Red Sea. The table shows the one-way SIMPER comparison between the regions. The top ten species are listed for each region, which contributed with approximately a cumulative 25% of the dissimilarity (numbers in bold in the last column).

Regions compared Species Average Average Contribution Cumulative Abundance Abundance to % Region 1 Region 2 dissimilarity (%) CRS vs SRS Naso elegans 1.5 1.39 3.52 3.52 Average dissimilarity=72.65 Scarus niger 1.43 1.07 2.82 6.34 Scarus ferrugineus 1.26 1.04 2.58 8.92 Ctenochaetus 1.9 1.98 2.54 11.46 striatus Chlorurus sordidus 1.24 1.39 2.42 13.87 Hipposcarus harid 1.13 0.73 2.37 16.25 Acanthurus gahhm 0.41 1.02 2.34 18.59 Chlorurus gibbus 1.03 0.18 2.01 20.59 Cheilinus 0.45 1.07 1.93 22.52 quinquecinctus Acanthurus sohal 0.34 0.74 1.81 24.33 CRS vs NRS Caesio lunaris 0.5 1.8 3.58 3.58 Average dissimilarity= 75.2 Naso elegans 1.5 0.88 3.11 6.69 Scarus niger 1.43 1.15 2.87 9.56 Hipposcarus harid 1.13 0.94 2.63 12.2 Acanthurus sohal 0.34 1.4 2.53 14.73 Scarus ferrugineus 1.26 0.47 2.5 17.23 Ctenochaetus 1.9 1.6 2.23 19.46 striatus Chlorurus sordidus 1.24 0.84 2.09 21.55 Chlorurus gibbus 1.03 0.18 1.97 23.53 Chromis dimidata 0.7 0.74 1.93 25.45 45

SRS vs NRS Caesio lunaris 0.44 1.8 3.31 3.31 Average dissimilarity = 77.90 Naso elegans 1.39 0.88 3.07 6.38

Acanthurus sohal 0.74 1.4 3.02 9.4 Scarus niger 1.07 1.15 2.45 11.85 Chlorurus sordidus 1.39 0.84 2.32 14.17 Scarus ferrugineus 1.04 0.47 2.21 16.38 Hipposcarus harid 0.73 0.94 2.17 18.55 Ctenochaetus 1.98 1.6 2.08 20.63 striatus Acanthurus gahhm 1.02 0.05 2.01 22.63 Cheilinus 1.07 0.48 1.88 24.51 quinquecinctus

In the analysis by shelf position, nearshore reefs show a clear separation between the regions (Figure 3.7). Reefs in the CRS are more clustered together and more similar to the NRS reefs. Lastly, reefs belonging to the SRS (except the reefs FB25 and FB31) are separated to the reefs in the CRS and NRS by a wide band.

Differences between south and north and south and center were confirmed by

Transform: Square root PERMANOVA results (p=0.001) (Table 3.8). Resemblance: S17 Bray Curtis similarity 40 PCO Biomass: nearshore reefs Region NRS CRS SRS

) 20

a i

a t

o 0

1 1

(

2 O

C P -20

-40 -40 -20 0 20 40 60 PCO1 (15.3% of total variation) Figure 3.7. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in the nearshore reefs of three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS) and circles the ones in the Southern Red Sea (SRS). 46

Table 3.8. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish species in nearshore reefs of three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms SRS, CRS 1.8107 0.001 998 SRS, NRS 2.1939 0.001 997 CRS, NRS 1.2239 0.103 998

On the other hand, midshore reefs do not show a clear separation between the regions (Figure 8). Moreover, no statisticalTransform: differences Square root were found (p=0.556). Resemblance: S17 Bray Curtis similarity 40 PCO Biomass: midshore reefs Region CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (19.4% of total variation) Figure 3.8. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in the midshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

Offshore reefs also do not show a clear trend, and PERMANOVA analysis did not find any statistical differences (p=0.294) (Figure 3.9). 47 Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Biomass: offshore reefs Region CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (29.4% of total variation) Figure 3.9. Principal Components Analysis (PCO) showing the dissimilarity of fish assemblages in the offshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

3.2.3 Trait diversity

In figure 3.10 is summarized the information obtained of the species regarding their ecological traits. From the 220 species found, in terms of diet, 24.09 % were invertivores, followed by planktivores (18.18%) and general carnivores (15.5%). With regards to the size, 70% of the species were contained mainly in the categories S4

(61 species), followed by S2 (52 species) and S3 (48 species). Regarding mobility, 48

46.8% of the species were categorized as sedentary, whereas 38.6% were mobile within the reef, and the rest, 14.55%, were mobile between reefs. In terms of vertical position in the water column, nearly two-thirds are benthopelagic fish. In terms of activity, the vast majority (83.18%) of fish species are active during the day. Finally, regarding the gregariousness, 35.9% of the fish species were cataloged as solitary, followed by small groups with 31.36% of the species. The group that had fewer species in this category was medium groups with 4.54% of the species.

Diet Size 80 60

50 60 40

30 40

20 20

10 Numberofspecies Numberofspecies

0 0

Pl Pi

Ex Sc Ec

Pe Be Co Ge S1 S2 S3 S4 S5 S6

Mobility Vertical position 150 100

75 100

50 50

Numberofspecies Numberofspecies 0 0 B BP P Sedentary Mwith Mbet 49

Activity Gregariousness 200 80

150 60

100 40

50 20

Numberofspecies Numberofspecies

0 0 DaN Diu Noc Sol Par SG MG LG

Figure 3.10. Relationship between the number of species and their respective ecological traits.

3.2.4 Ecological traits: density

The PCO analysis of the density illustrates the differences in trait diversity in the assemblages among the three analyzed regions (Figure 3.11). Central Red Sea reefs show more similarity between them and group with the NRS reefs. On the other hand, the northern reefs show more variation than those in the center but less than the southern reefs. SRS reefs separate from NRS and CRS, although certain reefs

(FB15, FB16, FB17, and FB20) group with these. The Permutational multivariate analysis of variance (PERMANOVA) also confirmed statistical differences among the regions (pseudo-f=4.59, p=0.001) (Table 3.9), with the pairwise analysis showing that all the pairs of regions are different (Table 3.10).

50 Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Trait density: all reefs Region NRS CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (25.5% of total variation) Figure 3.11. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in assemblages in three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS), and circles the ones in the Southern Red Sea (SRS).

Table 3.9. Results of a three-way nested permutational multivariate analysis of variance (PERMANOVA) on square-root transformed density of trait diversity of fish along three regions of the Red Sea. Region is a fixed factor; location is nested within the region; the site is nested within location and region. df=degrees of freedom, MS=Mean Sum of Squares, Pseudo-F= Mean value by permutation P(perm)= p-value of permutations. P(MC)=Monte Carlo p-value.

Source df MS Pseudo- P(perm) Unique P(MC) F perms Region 2 19863 4.5988 0.001 997 0.001 Location (Region) 3 9807.6 2.2609 0.005 997 0.001 Site(Location(Region)) 47 4081.5 4.6402 0.001 996 0.001 Transect 139 879.59

Total 191

Table 3.10. Results of the pairwise test of permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish traits in three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms CRS, SRS 2.2616 0.001 999 CRS, NRS 1.8801 0.001 999 SRS, NRS 2.1774 0.001 999

In figure 3.12 is shown the Principal Component Analysis for the density of functional entities in nearshore reefs. In the CRS, reefs are more clustered together and closer to the NRS reefs. Furthermore, SRS reefs are separated from the reefs in the CRS and NRS. Within the SRS, reefs have a similarity of 38.54%, being the NRS the ones that show more variation (average similarity 37.61%). Differences were confirmed

Transform: Square root by PERMANOVA results (Table 3.11). Resemblance: S17 Bray Curtis similarity 40 PCO Trait density: nearshore reefs Region NRS CRS SRS

) 20

a i

l a

o 0

% 2

1 (

2 O

C P -20

-40 -40 -20 0 20 40 PCO1 (22.3% of total variation) Figure 3.12. Principal Components Analysis (PCO) showing the dissimilarity ecological trait diversity in assemblages in nearshore reefs of three regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS) and circles the ones in the Southern Red Sea (SRS). 52

Table 3.11. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed densities of fish traits in nearshore reefs of three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms SRS, CRS 2.2973 0.001 999 SRS, NRS 2.7225 0.001 998 CRS, NRS 1.5136 0.011 998

On the other hand, in the midshore reefs, SRS reefs exhibit more variation than the reefs on the CRS (Figure 3.13). PERMANOVA did not show any significant difference between these regions (p=0.286). Transform: Square root Resemblance: S17 Bray Curtis similarity PCO Trait density: midshore reefs 20 Region CRS

SRS

)

t a

i 0

(

2 -20

C P

-40 -40 -20 0 20 40 PCO1 (40.2% of total variation) Figure 3.13. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in assemblages in midshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

Offshore reefs, on the other hand, do not show a clear trend. However, CRS reefs are less dispersed than the ones in the south (Figure 14). Monte Carlo test did not show any statistical difference among the regions (p=0.519). 53 Transform: Square root Resemblance: S17 Bray Curtis similarity PCO Trait density: offshore reefs 40 Region CRS SRS

)

o i

t a

i 20 r

a t

f o

% 2

(

2 0 O

C P

-20 -40 -20 0 20 PCO1 (36.1% of total variation) Figure 3.14. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in assemblages in offshore reefs of two regions of the Red Sea based on density after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

3.2.5 Ecological traits: biomass

The Principal Component Analysis for the biomass of functional traits shows the dissimilarity of the three regions analyzed (Figure 3.15). Assemblages in the Central

Red Sea reefs show similarity with the reefs of the NRS. On the other hand, the SRS reefs show more variation than those in the CRS and the NRS reefs. The

Permutational multivariate analysis of variance (PERMANOVA) also confirmed statistical differences among the regions (pseudo-f=3.45, p=0.001) (Table 3.12), with the pairwise analysis showing that all the pairs of regions are different (Table 3.13).

54 Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Trait biomass: all reefs Region NRS CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (13.6% of total variation) Figure 3.15. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS), and circles the ones in the Southern Red Sea (SRS).

Table 3.12. Results of a three-way nested permutational multivariate analysis of variance (PERMANOVA) on square-root transformed biomass of trait diversity of fish along three regions of the Red Sea. Region is a fixed factor; location is nested within the region; the site is nested within location and region. df=degrees of freedom, MS=Mean Sum of Squares, Pseudo-F= Mean value by permutation P(perm)= p-value of permutations. P(MC)=Monte Carlo p-value.

Source df MS Pseudo-F P(perm) Unique P(MC) perms Region 2 15963 3.4504 0.001 998 0.001 Location (Region) 3 10362 2.2312 0.001 996 0.001 Site(Location(Region)) 47 4383.7 3.0867 0.001 992 0.001 Transect 139 1420.2

Total 191

Groups t P(perm) Unique perms

CRS, SRS 1.9026 0.001 999 CRS, NRS 1.5316 0.002 999

SRS, NRS 2.01 0.001 997

When analyzed by shelf position, nearshore reefs show a separation between CRS and NRS from SRS. Differences were confirmed by the pairwise test for southern and central assemblages (p=0.002) as well as the southern and northern (p=0.001) but not for central and northern (p=0.113)Transform: (table Square 3.14) root. Resemblance: S17 Bray Curtis similarity 40 PCO Trait biomass: nearshore reefs Region NRS CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (15.6% of total variation) Figure 3.16. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in nearshore reefs of three regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Triangles represent the reefs in the Northern Red Sea (NRS), squares the reefs of the Central Red Sea (CRS) and circles the ones in the Southern Red Sea (SRS). 56

Table 3.14. Results of the permutational multivariate analysis of variance (PERMANOVA) based on Bray-Curtis similarity index after square root transformed biomass of fish traits in nearshore reefs of three regions of the Red Sea. T= t-value, P(perm)= p-value of permutations.

Groups t P(perm) Unique perms SRS, CRS 1.7493 0.002 999 SRS, NRS 2.2923 0.001 998 CRS, NRS 1.2222 0.113 998

On the other hand, midshore reefs exhibit more variation than the nearshore reefs, and statistical analysis did not find any difference between the regions analyzed

(p=0.536) (Figure 3.17). Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Trait biomass: midshore reefs Region CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (17.9% of total variation) Figure 3.17. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in midshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

Similarly, offshore reefs do not show a clear trend. Moreover, both regions analyzed have high variability within the reefs with no statistical differences (Monte Carlo p=0.418) (Figure 3.18). Transform: Square root Resemblance: S17 Bray Curtis similarity 40 PCO Trait biomass: offshore reefs Region CRS SRS

) 20

o 0

(

O C

P -20

-40 -40 -20 0 20 40 PCO1 (27.1% of total variation) Figure 3.18. Principal Components Analysis (PCO) showing the dissimilarity of ecological trait diversity in fish assemblages in offshore reefs of two regions of the Red Sea based on biomass after square root transformation using Bray Curtis similarity index. Every element on the graph represents a reef. Reefs of the Central Red Sea (CRS) are represented in squares, and Southern Red Sea (SRS) reefs in circles.

3.2.6 Functional entities

A total of 142 functional entities were determined for the 214 fish species found in the subset of 34 reefs. A redundancy of 1.5 was found, and additionally, twenty percent of the functional entities were represented for more than one species

(functional over redundancy), being an entity in particular common for ten species.

On the other hand, 70% of the entities were represented by only one species

(functional vulnerability) (Figure 3.19).

Figure 3.19. Number of species as a function of the rank of Functional Entities. 59

3.3 Components of variability in the community in the fish community

Total variation was partitioned among the scales (Figure 3.20). In the fish density and density of fish traits, the factor location had the highest percentage (33.8% and

30%, respectively), followed by the transect level, which had 32.7% and 29.7%. On the contrary, for biomass and functional entities biomass, transect level had the highest variability with 39.4% of the total variation for the former and 37.7% for the latter. The factor location had the lowest variability (approximately 14%) in the four fish variables.

variation of components of Estimates

Figure 3.20. Components of variation in the fish community. A) Density, B) Biomass, C) Density of functional entities D) Biomass of functional entities. Colors represent the factors: white= region, light gray=location, dark gray=site and black=transect.

3.4 Relationships between the fish variables and Hard Coral cover

The results of the linear regressions showed that many variables of the fish community were correlated to differences in percentage cover of hard corals. R squared values were generally low, but some showed a significant trend (Table

3.15). There were more positive trends in the northern region than in the center and the south. However, the three regions shared a positive relationship between the hard coral cover and the total density and functional redundancy and a negative relationship between hard coral cover and functional divergence and vulnerability.

Table 3.15. Responses of the fish community to hard coral cover (%) in three regions of the Red Sea. T= trend (positive/negative), R2= regression coefficient, p= p-value. *= statistical difference

North Center South Fish community variables T R2 p T R2 p T R2 p

Total species + 0.16 0.027 * - 0.04 0.17 - 0 0.983

Total Biomass + 0.15 0.033* - 0.01 0.461 - 0.01 0.589

Total Density + 0.14 0.043* + 0.01 0.562 + 0 0.766 Total Functional Entitites + 0.12 0.06 - 0.05 0.13 - 0 0.863 Functional Richness + 0.53 0.000* - 0.02 0.353 + 0 0.68 Functional Divergence - 0.07 0.17 - 0.09 0.034* - 0.34 0.000*

Functional Redundancy + 0.22 0.009* + 0.01 0.498 + 0.02 0.345 Functional vulnerability - 0.15 0.036* - 0.06 0.098 - 0.06 0.097

There is a positive correlation with the percentage cover of hard corals and total species and total biomass of fish in the northern region (R2=0.163 and 0.152, respectively), being both statistically significant. However, this relationship is negative in the central and southern regions (Figures 3.21 and 3.22). 61

Total species Total

% Hard coral cover

Figure 3.21. Total species richness as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

ass Total Total biom

% Hard coral cover

Figure 3.22. Total biomass as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

In the case of total density, the three regions exhibit a positive relationship with the hard coral cover, being only the northern region with statistical significance (Figure

3.23).

density Total Total

% Hard coral cover

Figure 3.23. Total density as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

As for the functional entities, there is a positive relationship with the hard-coral cover in the northern region. However, in the central and southern regions, the trend is negative (Figure 3.24).

Total Total funcional entities

% Hard coral cover

Figure 3.24. Total functional entities as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

Positive relationships between hard coral cover were also found in the functional richness in the northern and southern region with statistical significance for the former (p=0.000) (Figure 3.25).

Functional richness Functional (density)

% Hard coral cover

Figure 3.25. Functional richness as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

On the other hand, hard coral cover was negatively correlated with the functional divergence and functional vulnerability in the three regions (Figures 3.26 and 3.27).

e(density) Functionaldivergenc

% Hard coral cover

Figure 3.26. Functional divergence as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: North in blue, Central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

Functionalvulnerability

% Hard coral cover

Figure 3.27. Functional vulnerability as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

Functional redundancy, however, was positively correlated with hard coral cover in the three regions (Figure 3.28).

Functionalredundancy

% Hard coral cover

Figure 3.28. Functional redundancy as a function of percentage cover of hard coral in three regions of the Red Sea. Every element in the graph represents a reef. Colors represent every region: north in blue, central in pink, and south in green. Fitted lines are linear regressions ± 95% confidence intervals.

DISCUSSION

The present study aimed to analyze scales of spatial variability in fish assemblages in three regions of the Red Sea assessed from a taxonomic and trait perspective.

Considering the pivotal role of hard corals on the dynamics of the reef fish, it was investigated the relationships between hard coral cover and attributes of the fish community.

4.1 Patterns in the Fish Community

In this study, fish assemblages displayed consistent patterns among the regions from both taxonomic and ecological trait approaches. The most consistent pattern observed between the two approaches was the separation in species composition between the north and central assemblages from those in the south. However, this regional distinction was only evident for nearshore reefs.

These findings agree with what was previously reported by Roberts et al. (1992), who found a latitudinal change in the composition of butterflyfishes (Chaetodontidae) and angelfishes (Pomacanthidae) in nearshore reefs of the Red Sea. Also, a clear differentiation when analyzing the fish assemblages of offshore reefs was unnoticed, with authors reporting a transition in the community from north to south (Roberts et al., 2016). Major latitudinal changes in either species richness or community composition of endemic species were also unclear in the Red Sea basin (DiBattista et al., 2016). Therefore, it seems that the driving forces or gradients that define the species distribution in the Red Sea communities are stronger in the nearshore communities, becoming attenuated towards the offshore. 70

The latitudinal differences found in the nearshore reefs can result from the width of the eastern Red Sea shelf, narrow in the north, about 40 km, and broadening in the south reaching up to 130 km (Sheppard et al., 1992; Tesfamichael and Pauly, 2016).

Furthermore, northern reefs are mostly narrow fringing reefs, whereas the central

Red Sea is dominated by patchy nearshore reefs (Sheppard, 1992). On the other hand, the southern region comprises a network of coral reefs that occupy the whole width of the shelf that extends from approximately 20°N to 18°N constituting one of the largest systems within the basin (Ormond et al., 1984; Sheppard, 1992).

A possible explanation for the higher number of species, density, and biomass in the south is an extensive reef system with a wide range of habitats such as shoals, reef platforms, isolated reefs, pinnacles, and atoll-like formations and islands that support vegetation such as mangroves that serve as nursery areas for many reef-associated fish (Price et al., 1987; Bruckner et al., 2012). Furthermore, more extensive reef areas can support higher diversity and can be key variables explaining most variation in reef fish diversity patterns (Malcolm et al., 2010). Moreover, increased reef area can be translated into more abundant and diverse substrate types, which have been found to support fish assemblages of different compositions, richness, and density, i.e., higher biodiversity (Anderson & Millar, 2004; Chittaro, 2004). However, it is important to mention here that the sites and depths were standardized in order to make them comparable.

In the Red Sea, Roberts et al. (1992) found that the reef development was one of the most important factors determining the patterns of distribution of richness and abundance of butterflyfishes. Other factors may also play a role in the regional 71 differences in fish communities, particularly in nearshore reefs. In the south, the water is more turbid and richer in nutrients due to the intrusion of the Gulf of Aden

Intermediate water from the Indian Ocean (Bruckner et al., 2012; Raitsos et al.,

2013). Therefore, differences in density and biomass between south and other regions can also result from natural processes occurring in the Red Sea, such as the differences in productivity. The northern of the basin is also known for having higher salinity and more extreme winters, which in addition to the low productivity, can result in limited survival of fish larvae compared to the south (Osman et al., 2018).

In addition, a higher number of species may also be the result of weak top-down regulations (Sandin et al., 2008), and differences in species richness, biomass, and density of fish can be evidence of human-related pressure. Rowlands et al. (2012) reported that fishing effort is not evenly distributed along the Saudi coast. In this regard, inshore reefs in Thuwal, one of the locations surveyed in the CRS, registered the lowest numbers in terms of species richness, biomass, and densities of fish, with numbers even lower than those reported by Khalil et al. (2017), presumably due to the high fishing intensity. Similarly, Kattan et al. (2017) reported that reef fish in the southern Saudi Arabian Red Sea had on average 62% less abundance and 20% less biomass than their counterparts in the Sudanese Red Sea.

Another point that is important to take into consideration is that only a few species drove nearly 30% of the differences between the regions analyzed. These species accounting for a third of the differences in densities were planktivores, which are a major trophic group on coral reefs (Hobson, 1991). Even though planktivores dominated the whole community, their relative contribution was higher in the south, 72 where nutrient availability is higher (Raitsos et al., 2013; Pearman et al., 2016). A study by Kattan et al. (2017) also found that the fish community composition in two regions of the central-south Red Sea (Sudan and Saudi Arabia) was mainly driven by planktivore species, with eight species constituting 73% of the differences.

Therefore, the results obtained in this study may be driven by the availability of nutrients in addition to an extensive reef area that provides suitable conditions for concentrations of zooplankton and phytoplankton (Malcolm et al., 2010).

4.2.1 Relationships between hard coral cover and fish community parameters

Since several factors related to habitat features can intervene in the fish community

(e.g., fish density and biomass), the following section tries to explain the relationship between specific attributes of the fish community and hard coral cover.

Fish species and their surrounding environment share complex relationships.

Certain studies have addressed the relationship between live coral cover and fish diversity and abundance (e.g., Bouchon-Navaro & Bouchon, 1989; Chong-seng et al., 2012; Komyakova et al., 2013). Coral cover can be a proxy for habitat complexity and a good predictor for abundance of species that feed or inhabit corals or rely on coral for recruitment (Munday, 2004; Gratwicke & Speight, 2005).

In this study, significant correlations were found between hard coral cover and several fish community attributes, particularly in the northern region. There were also positive correlations between total density and the functional redundancy in the central and southern regions. The higher coral cover has implications for fish from 73 recruitment to prey-predator interactions (Jones & Syms, 2004; Coker et al., 2012).

These findings are supported by other studies in the Red Sea that have found a correlation between the density of chaetodontid fishes and the cover and diversity of corals in the Gulf of Aqaba (northern Red Sea) (Bouchon Navarro & Bouchon, 1989).

In the central Red Sea, Pombo-Ayora et al. (2020) also found that herbivorous fish biomass and density were higher in the coral-dominated habitats than the algae- dominated habitats. Interestingly, Khalil et al. (2017) did not find a correlation of coral cover with fish species richness or fish biomass in the reefs located in the central

Red Sea.

Direct relationships between coral cover and fish assemblages in the central and southern regions may be difficult to detect, presumably because some of the reefs within both regions have experienced two bleaching events, one in 2010 and the second in 2015/2016. The 2010 event was particularly severe for reefs in the central region where nearshore reefs lost up 85-95% of live coral cover in depths shallower than 10 m (Furby et al., 2013), while the 2015-2016 bleaching event was more harmful to reefs in the south (DeCarlo et al., 2020). In the southern Red Sea, reefs decreased their live coral cover considerably after the bleaching events but have been found to remain structurally intact (Shellem, 2020). Declines in coral cover are often followed by increases in the algal cover, which can support high herbivorous biomass (Williams et al., 2001). In this regard, when analyzing the assemblages by ecological trait, it was observed that the density planktivores were dominant, but herbivores contribute the most to the total biomass, particularly in the central and southern regions. These results agree with what was reported by Khalil et al. (2017), 74 who found that herbivores dominated the biomass of the reefs surveyed across the shelf in Thuwal (CRS).

A study by Wilson et al. (2009) in the GBR found that fish species richness remained fairly similar between disturbed reefs and those that were used as control over a period of 11 years. They found that even though there was a decline between 46-

96% in coral cover, structural complexity was retained in half of the reefs analyzed.

The declines in coral cover resulted in declines of specialized species, a loss that was compensated by the increase of species that feed on the epilithic algal matrix.

It is generally assumed that most fishes associate with live coral habitat for the physical structure provided by the corals. Certain studies have found that the complexity provides habitat and shelter for many species and mediates interactions between predators and prey (Kerry & Bellwood, 2012; Harborne et al., 2012).

Therefore, complexity can affect different aspects of the fish community, such as species richness, species diversity, and biomass (Luckhurst and Luckhurst, 1978;

Darling et al., 2017). Even though there was no relationship found between hard coral cover and most of the variables in the fish community in the south, the results obtained support the maintenance of the physical structure established by dead corals (Shellem, 2019), which is still important to maintain the levels of diversity, biomass and density of fish in this region.

From the results obtained in this study, we can infer that declines in coral cover may not result in a short-term impact on the diversity of coral reef fishes. However, the assemblages may undergo profound and radical changes that compromise the 75 ability of the reefs to thrive in the face of the more common and intense disturbance events occurring in the Red Sea. A study by Pombo-Ayora et al. (2020) in reefs of the Central Red Sea revealed that in coral-dominated habitats, assemblages of herbivorous fish displayed higher morphological and ecological richness, whereas, in areas with low coral cover, they have limited trait diversity. However, the authors suggested that algae-dominated areas may serve as an important nursery habitat for some species, with the potential to support nearby habitats. Indeed, other habitats are also important to fish such as mangroves and sandy bottoms; the formers act as nurseries, and the latter can act as resting zones for predators such as sharks (Price et al., 1987; Vroom et al., 2006).

4.2.2 Importance of analyzing the fish communities from an ecological trait perspective

Coral reefs are home to an extensive array of fishes that carry out key processes.

Such processes are affected by the biological and ecological characteristics of the organisms rather than by their taxonomic identity (Hooper et al., 2005).

Understanding the functional attributes of individual species is the first step towards linking their presence to the properties of a certain ecosystem. As stated by Bellwood et al. (2019), it is the processes that fish carry out that are key to the survival of the reefs, and therefore they are a focal point for management. The use of traits has provided useful insights from the distribution of trait combinations to changes in community composition after disturbances (Brandl et al., 2016; Mouillot et al., 2014) which can be masked with the use of a taxonomical approach alone, as ecosystems 76 with more species not necessarily show higher functional diversity (Cardoso et al.,

2011).

Another example can be fish biomass. It has been reported that overfishing of predators may result in species turnover with little change in total fish biomass. Yet, these changes may alter the energy flow within the ecosystem and, therefore, in the services provided (Fogarty & Murawski, 1998). Therefore, using a more detailed trait-based approach could be useful to complement the results obtained in this study.

On the other hand, an important result obtained from the trait base approach was that 70% of the functional entities were represented by one species. This result shows the vulnerability of the Red Sea to the losses of diversity in the face of more severe and constant coral bleaching events. Therefore, any changes in biodiversity may affect the functioning of the reef. Consequently, the loss of an ecosystem has far more reaching consequences. The key here would be to prioritize the protection of the functions of the ecosystems rather than protecting species in specific.

The levels of redundancy obtained in this study show that only 20 percent of the functional entities were overrepresented (functional over-redundancy). However, they are necessary to ensure ecosystem resilience to perturbation (Brandl et al.,

2016). Also, it is important to remember that the context of the assemblage plays a crucial role, as it is not the same a species of a pool of 20 species than of a pool of

100 species. Another aspect to consider is that redundant species in small scales may be functionally differentiated when the environment changes over larger spatial scales. 77

Overall, the results obtained in this study provide a bigger picture of the ecological trait diversity of the Red Sea. However, further studies may benefit from the use of other indices such as functional specialization, functional evenness, functional dispersion, and functional originality, which on the whole provide better baselines for understanding the functioning of the reefs (Brandl et al., 2016; Mouillot et al., 2014).

4.2.3 Importance of taking into consideration different spatial scales

The scale on which organisms are studied has major implications for our understanding of processes affecting them. This study showed that the smallest scales (i.e., transect and reef) accounted for the greatest variability in the fish density and biomass (both for taxonomy and trait approaches). These results are not surprising as fish move over short periods and distances, resulting in high small spatial-scale variability, as described elsewhere (Jones & Syms, 1998; García-

Charton et al., 2004; Tuya et al., 2011). Furthermore, it can also be inferred that habitat characteristics can result from the influence of the benthic habitat on the fish assemblages, which in the Red Sea can be quite variable. Therefore, future investigations of patterns of distribution of species may benefit from increased replication within and between reefs.

Even though this study found little variability at the broader scales, it remains important to continue studies that quantify the variability at the level of the region, as this may provide insight into the processes that affect other aspects of the fish biology, such as environmental tolerances or larval dispersal (Taylor & Hellberg.,

2003; Kinlan et al., 2005). 78

4.3 Importance of this research for future coastal management in the Saudi

Arabian Red Sea

Worldwide, coral reef ecosystems have been observed to undergo profound changes. These habitats are under constant threat, and protection efforts should be placed in order to attenuate the ongoing degradation (Guidetti, 2000; Tunesi et al.,

2006). The Saudi Arabian Red Sea has also been subject to changes in coral cover, particularly after the bleaching events of 2010 and 2015/2016 (Furby et al., 2013;

DeCarlo et al., 2020). A better understanding of the roles underpinning reef fish ecosystem functioning, how they respond to disturbance events, and how changes in fish communities and associated functions can affect the delivery of goods and services is critical for effective coastal zone management.

The results of this study can be used as a baseline for the implementation of a network of marine reserves that can effectively address three main principles of conservation and management, which are: biodiversity protection, enhancement of fisheries, and adaptation to climate change. In order to achieve these goals, an array of ecological guidelines for designing marine reserve networks have been proposed, including habitat representation and replication, protection of unique and critical areas, incorporation of connectivity, allow time for recovery, consider local threats, and adaptation to climate change (Green et al., 2014; Munguia-Vera et al., 2018).

As observed in this study, there was more variability at the levels of the transect and reef than at the levels of location and region. Furthermore, the regions and shelves analyzed showed to be different. Therefore, if a network of marine reserves were to 79 be implemented in the Red Sea, they should include reefs of the different regions and the three shelf positions to represent the different fish assemblages present in the basin.

Likewise, and as observed in the correlation analysis, many aspects of the fish community were positively correlated with coral cover. Particularly, fish redundancy was positively correlated with the hard-coral cover; therefore, such reefs where redundancy is higher may serve as a point of dispersion of larvae, juveniles, or adults towards other reefs. Furthermore, such reserves must include areas that fish use in critical phases of their life, such as mangroves which are used as nurseries (Abu El-

Regal & Ibrahim, 2017). Thus, by adding reefs and other habitats along the latitudinal and across the shelf gradient, it allows the continuation of a biological corridor, providing in this way connectivity between different ecosystems within the Red Sea, and in this way, ensure the movement of biological diversity and the continuation of their ecological and evolutionary processes.

On the other hand, this study also highlights the need to implement or reinforce management actions, particularly in areas that can exhibit lower fish diversity, which can result from disturbance events (e.g., coral bleaching) or the pressure due to small-scale artisanal fisheries. Most of the artisanal fishing effort is directed on or around coral reefs, with fishers primarily targeting predators such as groupers and snappers or herbivorous fish (Jin et al., 2012; Shellem et al., 2021). Carnivores are critical for regulating food webs, whereas herbivorous fish are key in regulating algal growth, which competes with corals for space. An interesting way to understand the 80 effect of fishing pressure is through exclusion experiments, where cages with holes of different sizes are deployed to mimic the intensity of herbivory. Cages with higher algal biomass mirror those reefs or sections of the reef in which herbivory is reduced

(Bellwood et al., 2006). With this regard, Giga Projects Red Sea Project and NEOM will implement fishing bans within their economic exclusive zones. This will offer an unprecedented opportunity to evaluate the recovery of reef fish communities and coral reef systems overall in the Red Sea. In particular, it will also be possible to evaluate whether restoration of natural food webs with dominant herbivorous fish and top predators could give resilience to other pressures like those resulting from global warming (e.g., coral bleaching).

In view of the preceding, other strategies are also needed to determine the regulation of the fisheries, taking into consideration functions, species, the minimum and maximum size of the catch, and biology of fish in terms of growth rates and reproduction. In this sense, seasonal closures, particularly in spawning season, applied in conjunction with other management strategies, such as catch quotas and gear restrictions, can lead to the successful preservation of marine resources

(Jennings, 2001; Sadovy et al., 2005). Furthermore, aquaculture can also be promoted. With careful planning in terms of location, design, production of local species, management of food, and disposition of residual nutrients, this practice can be a viable alternative for fishing in terms of revenue and protein source (Gentry et al., 2017).

On a final note, it can be said that future research must focus on a better understanding of the drivers and the pressures that affect fish communities. 81

Likewise, future studies can also focus on key functional groups such as herbivorous fish, which are important to mediate the structure of the reef by feeding on algae, which translates into available space for coral growth. Furthermore, attention should be given to top predators who regulate the food webs and structure the populations of the fishes down in the web. Particularly in Giga-projects that are happening on the

Saudi Arabian Red Sea coast will benefit from implementing the no-take areas, as reef-based tourism can be more profitable than reef-based fisheries (Spalding et al.,

2017). As suggested by Cziesielski et al. (2021), investing in “blue natural capital”

(i.e., coral reefs, mangroves, and seagrass beds in terms of the ecosystem services that result from their functional integrity) may lead to the conservation and enhancement of natural resources but on parallel be a profitable investment opportunity that can diversify the local economies (Schmidt-Roach et al., 2020).

REFERENCES

Abu El-Regal, M. A., & Ibrahim, N. K. (2014). Role of mangroves as a nursery ground for juvenile reef fishes in the southern Egyptian Red Sea. Egyptian Journal of Aquatic Research, 40(1), 71–78. https://doi.org/10.1016/j.ejar.2014.01.001 Allen, G.R. (2008). Conservation hotspots of biodiversity and endemism for Indo- Pacific coral reef fishes. Aquat Conserv. Mar Freshwat Ecosyst 18(5):541–556. Allgeier, J. E., Layman, C. A., Mumby, P. J., & Rosemond, A. D. (2014). Consistent nutrient storage and supply mediated by diverse fish communities in coral reef ecosystems. Global Change Biology, 20(8), 2459–2472. https://doi.org/10.1111/gcb.12566 Anderson, M. J., & Millar, R. B. (2004). Spatial variation and effects of habitat on temperate reef fish assemblages in northeastern New Zealand. 305, 191–221. https://doi.org/10.1016/j.jembe.2003.12.011 Bell, J. D., Craik, G. J. S., Pollard, D. A., & Russell, B. C. (1985). Estimating length frequency distributions of large reef fish underwater. Coral Reefs, 4(1), 41–44. https://doi.org/10.1007/BF00302203 Bellwood, D. R., Hughes, T. P., & Hoey, A. S. (2006). Sleeping Functional Group Drives Coral-Reef Recovery. Current Biology, 16(24), 2434–2439. https://doi.org/10.1016/j.cub.2006.10.030 Bellwood, D. R., Hughes, T. P., Folke, C., & Nyström, M. (2004). Confronting the coral reef crisis. Nature, 429(6994), 827–833. https://doi.org/10.1038/nature02691 Bellwood, D. R., Streit, R. P., Brandl, S. J., & Tebbett, S. B. (2019). The meaning of the term ‘function’ in ecology : A coral reef perspective. November 2018, 948– 961. https://doi.org/10.1111/1365-2435.13265 Bouchon-Navaro, Y., & Bouchon, C. (1989). Correlations between chaetodontid fishes and coral communities of the Gulf of Aqaba (Red Sea). Environmental Biology of Fish, 25(1), 47–60. https://doi.org/10.1007/978-94-009-2325-6_3 Bozec, Y. M., Dolédec, S., & Kulbicki, M. (2005). An analysis of fish-habitat associations on disturbed coral reefs: Chaetodontid fishes in New Caledonia. Journal of Fish Biology, 66(4), 966–982. https://doi.org/10.1111/j.0022- 1112.2005.00652.x Brandl, S. J., Emslie, M. J., & Ceccarelli, D. M. (2016). Habitat degradation increases functional originality in highly diverse coral reef fish assemblages. Ecosphere, 7(11). https://doi.org/10.1002/ecs2.1557 Bruckner, A., & Williams, A. (2012). Assessment of the Community Structure, Status, Health and Resilience of Coral Reefs off St. Kitts and Nevis. June, 64. 83

Bruckner, A., G. Rowlands, B. Riegl, S. Purkis, A. Williams, and P. Renaud (2012). Khaled bin Sultan Living Oceans Foundation Atlas of Saudi Arabian Red Sea Marine Habitats. Panoramic Press, pp. 262. Cardoso, P., Pekár, S., Jocqué, R. & Coddington J.A. (2011). Global patterns of guild composition and functional diversity of spiders. PLoS ONE 6(6):e21710. Chabanet, P., Ralambondrainy, H., Amanieu, M., Faure, G., & Galzin, R. (1997). Relationships between coral reef substrata and fish. May 2014. https://doi.org/10.1007/s003380050063 Chittaro, P. M. (2004). Fish-habitat associations across multiple spatial scales. Coral Reefs, 23(2), 235–244. https://doi.org/10.1007/s00338-004-0376-z Chong-seng, K. M., Mannering, T. D., Pratchett, M. S., Bellwood, D. R., & Graham, N. A. J. (2012). The Influence of Coral Reef Benthic Condition on Associated Fish Assemblages. 7(8), 1–10. https://doi.org/10.1371/journal.pone.0042167 Clarke, K.R. & Gorley, R.N. (2006). PRIMER v6: User Manual/Tutorial. PRIMER-E. Coker, D. J., Graham, N. A. J., & Pratchett, M. S. (2012). Interactive effects of live coral and structural complexity on the recruitment of reef fishes. 919–927. https://doi.org/10.1007/s00338-012-0920-1 Coker, D. J., Wilson, S. K., & Pratchett, M. S. (2014). Importance of live coral habitat for reef fishes. Reviews in Fish Biology and Fisheries, 24(1), 89–126. https://doi.org/10.1007/s11160-013-9319-5 Cole, A. J., Pratchett, M. S., & Jones, G. P. (2008). Diversity and functional importance of coral-feeding fishes on tropical coral reefs. Fish and Fisheries, 9(3), 286–307. https://doi.org/10.1111/j.1467-2979.2008.00290.x Connell, J. H. (1978). Diversity in tropical rain forests and coral reefs. Science, 199(4335), 1302–1310. https://doi.org/10.1126/science.199.4335.1302 Cziesielski, M. J., Duarte, C. M., Aalismail, N., Al-Hafedh, Y., Anton, A., Baalkhuyur, F., Baker, A. C., Balke, T., Baums, I. B., Berumen, M., Chalastani, V. I., Cornwell, B., Daffonchio, D., Diele, K., Farooq, E., Gattuso, J. P., He, S., Lovelock, C. E., Mcleod, E., … Aranda, M. (2021). Investing in Blue Natural Capital to Secure a Future for the Red Sea Ecosystems. Frontiers in Marine Science, 7(January). https://doi.org/10.3389/fmars.2020.603722 DeCarlo, T. M., Gajdzik, L., Ellis, J., Coker, D. J., Roberts, M. B., Hammerman, N. M., et al. (2020). Nutrient-supplying ocean currents modulate coral bleaching susceptibility. Science Advances, 6, eabc5493. Development Core Team.(2013), R: A language and environment for statistical computing. R Found Stat Comput, R Foundation for Statistical Computing. Vienna, Austria. 84

Dibattista, J. D., Roberts, M. B., Bouwmeester, J., Bowen, B. W., Coker, D. J., Lozano-cort, D. F., Choat, J. H., Gaither, M. R., Hobbs, J. A., Maha, T., Kochzius, M., Myers, R. F., Paulay, G., & Vanessa, S. N. (2016). A review of contemporary patterns of endemism for shallow water reef fauna in the Red Sea. 423–439. https://doi.org/10.1111/jbi.12649

Floeter, S. R., Guimarães, R. Z. P., Rocha, L. A., Ferreira, C. E. L., Rangel, C. A., & Gasparini, J. L. (2001). Geographic variation in reef-fish assemblages along the Brazilian coast. Global Ecology & Biogeography, 10, 423–431. Fogarty, M. J., & Murawski, S. A. (1998). Large-scale disturbance and the structure of marine systems: Fishery impacts on Georges Bank. Ecological Applications, 8(1 SUPPL.), 6–22. https://doi.org/10.2307/2641359 Friedlander, A.M. & Demartini, E.E. (2002). Contrasts in density, size, and biomass of reef fishes between the northwestern and the main Hawaiian Islands: the effects of fishing down apex predators. Mar Ecol Prog Ser 230:253–264 Froese, R. & Pauly, D. (2016). FishBase. http://fishbase.org. Consulted on October, 2020 Furby, K. A., Bouwmeester, J., & Berumen, M. L. (2013). Susceptibility of central Red Sea corals during a major bleaching event. Coral Reefs, 32(2), 505–513. https://doi.org/10.1007/s00338-012-0998-5 Gagic, V., Bartomeus, I., Jonsson, T., Taylor, A., Winqvist, C., Fischer, C., Slade, E. M., Steffan-Dewenter, I., Emmerson, M., Potts, S. G., Tscharntke, T., Weisser, W., & Bommarco, R. (2015). Functional identity and diversity of animals predict ecosystem functioning better than species-based indices. Proceedings of the Royal Society B: Biological Sciences, 282(1801). https://doi.org/10.1098/rspb.2014.2620 García-Charton, J. A., & Pérez-Ruzafa, A. (2001). Spatial pattern and the habitat structure of a Mediterranean rocky reef fish local assemblage. Marine Biology, 138(5), 917–934. https://doi.org/10.1007/s002270000524 García-Charton, J. A., Pérez-Ruzafa, Á., Sánchez-Jerez, P., Bayle-Sempere, J. T., Reñones, O., & Moreno, D. (2004). Multi-scale spatial heterogeneity, habitat structure, and the effect of marine reserves on Western Mediterranean rocky reef fish assemblages. Marine Biology, 144(1), 161–182. https://doi.org/10.1007/s00227-003-1170-0 Genevier, L. G. C., Jamil, T., Raitsos, D. E., Krokos, G., & Hoteit, I. (2019). Marine heatwaves reveal coral reef zones susceptible to bleaching in the Red Sea. Global Change Biology, 25(7), 2338–2351. https://doi.org/10.1111/gcb.14652 Gentry, R. R., Froehlich, H. E., Grimm, D., Kareiva, P., Parke, M., Rust, M., . . . Halpern, B. S. (2017). Mapping the global potential for marine aquaculture. 85

Nature ecology & evolution, 1(9), 1317. Gower, J.C. (1971). A General Coefficient of Similarity and Some of Its Properties. Biometrics, Vol. 27 No. 4, p. 857. Green, A. L., Fernandes, L., Almany, G., Abesamis, R., McLeod, E., Aliño, P. M., White, A. T., Salm, R., Tanzer, J., & Pressey, R. L. (2014). Designing Marine Reserves for Fisheries Management, Biodiversity Conservation, and Climate Change Adaptation. Coastal Management, 42(2), 143–159. https://doi.org/10.1080/08920753.2014.877763 Gratwicke, B., & Speight, M. R. (2005). The relationship between fish species richness, abundance and habitat complexity in a range of shallow tropical marine habitats. Journal of Fish Biology, 66(3), 650–667. https://doi.org/10.1111/j.0022-1112.2005.00629.x Guidetti, P. (2000). Differences among fish assemblages associated with nearshore Posidonia oceanica seagrass beds, rocky-algal reefs and unvegetated sand habitats in the adriatic sea. Estuarine, Coastal and Shelf Science, 50(4), 515– 529. https://doi.org/10.1006/ecss.1999.0584 Harmelin-Vivien, M. L. (1989). Implications of feeding specialization on the recruitment processes and community structure of butterflyfishes. Environmental Biology of Fishes, 25(1–3), 101–110. https://doi.org/10.1007/BF00002204 Harmelin-Vivien, M. L. (2002). Energetics and Fish Diversity on Coral Reefs. In Coral Reef Fishes. Elsevier Science (USA). https://doi.org/10.1016/b978-012615185- 5/50016-5 Hobson, E. S. (1991). Trophic relationships of fishes specialised to feed on zooplankters above coral reefs. In P. F. Sale (Ed.), The ecology of fishes on Coral Reefs (pp. 69–95). Cambridge, MA: Academic Press Inc. Holmes, B. J., Williams, S. M., Otway, M., Nielsen, E. E., Maher, S. L., Bennett, M. B.& Ovenden, J. R., (2017). Population structure and connectivity of tiger sharks ( Galeocerdo cuvier ) across the Indo-Pacific Ocean basin Subject Category : Author for correspondence : Holmlund, C. M., & Hammer, M. (1999). Ecosystem services generated by fish populations. 29, 253–268. Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S. et al. (2005) Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs, 75, 3– 35. Jennings, S., Pinnegar, J. K., Polunin, N. V. C., & Boon, T. W. (2001). Weak cross- species relationships between body size and trophic level belie powerful size- based trophic structuring in fish communities. Journal of Animal Ecology, 70(6), 934–944. https://doi.org/10.1046/j.0021-8790.2001.00552.x 86

Jin, D., Kite-Powell, H., Hoagl, P., & Solow, A. (2012). A bioeconomic analysis of traditional fisheries in the Red Sea. Marine Resource Economics, 27(2), 137– 148. https://doi.org/10.5950/0738-1360-27.2.137 Jones, G. P., & Syms, C. (2004). Introduction. Coral Reefs, 111(1), 101–110. https://doi.org/10.1007/s00338-017-1647-9 Jones, G. P., & Syms, G. (1998). Disturbance, habitat structure and the ecology of fishes on coral reefs. Austral Ecology, 23(3), 287–297. https://doi.org/10.1111/j.1442-9993.1998.tb00733.x Kattan, A., Coker, D. J., & Berumen, M. L. (2017). Reef fish communities in the central Red Sea show evidence of asymmetrical fishing pressure. 1227–1238. https://doi.org/10.1007/s12526-017-0665-8 Kerry J. & Bellwood D. (2012) The effect of coral morphology on shelter selection by coral reef fishes. Coral Reefs 1–10. doi:10.1007/s00338-011-0859-7 Khalil, M. T., Bouwmeester, J., & Berumen, M. L. (2017). Spatial variation in coral reef fish and benthic communities in the central Saudi Arabian Red Sea. 1–23. https://doi.org/10.7717/peerj.3410 Kinlan, B. P., Gaines, S. D. & Lester, S. E. (2005). Propagule dispersal and the scales of marine community process. Diversity and Distributions 11, 139–148. Kohler, K.E., & Gill, S. (2006). Coral Point Count with Excel extensions (CPCe): A Visual Basic program for the determination of coral and substrate coverage using random point count methodology. Computers & Geosciences 32(9):1259- 1269 Komyakova, V., Munday, P. L., & Jones, G. P. (2013). Relative importance of coral cover, habitat complexity and diversity in determining the structure of reef fish communities. PLoS ONE, 8(12), 1–12. https://doi.org/10.1371/journal.pone.0083178 Lea, J. S. E., Wetherbee, B. M., Queiroz, N., Burnie, N., Aming, C., Sousa, L. L., Mucientes, G. R., & Humphries, N. E. (2015). Repeated , long-distance migrations by a philopatric predator targeting highly contrasting ecosystems. Nature Publishing Group, March, 1–11. https://doi.org/10.1038/srep11202 Legendre, P. and Legendre, L. (2003). Numerical Ecology, Developments in Environmental Modeling, 20, Elsevier. Maechler, M., Rousseeuw, P., Struyf, A., Hubert, M. and Hornik, K. (2013). Finding groups in data: Cluster Analysis. R Package. Version 1.14.4. Malcolm, H. A., Jordan, A., & Smith, S. D. A. (2010). Biogeographical and cross- shelf patterns of reef fish assemblages in a transition zone. Marine Biodiversity, 40(3), 181–193. https://doi.org/10.1007/s12526-010-0042-3 87

Mason, N.W.H., Mouillot, D., Lee, W.G. and Wilson, J.B. (2005), Functional richness, functional evenness and functional divergence: The primary components of functional diversity, Oikos, Vol. 111 No. 1, 112–118 Monroe, A. A., Ziegler, M., Roik, A., Ro, T., Emms, M. A., Jensen, T., Voolstra, C. R., & Berumen, M. L. (2018). In situ observations of coral bleaching in the central Saudi Arabian Red Sea during the 2015 / 2016 global coral bleaching event. 1– 13. Mouillot, D., Graham, N.A.J., Villéger, S., Mason, N.W.H. and Bellwood, D.R. (2013), A functional approach reveals community responses to disturbances. Trends in Ecology and Evolution, Vol. 28 No. 3, 167–177. Mouillot, D., Villéger, S., Parravicini, V., Kulbicki, M., & Arias-gonzález, J. E. (2014). Functional over-redundancy and high functional vulnerability in global fish faunas on tropical reefs. PNAS 111(38), 13757–13762. https://doi.org/10.1073/pnas.1317625111 Munday, P.L., 2004. Competitive coexistence of coral-dwelling fishes; the lottery hypothesis revisited. Ecology (85), 623–628. Munguia-Vega, A., Green, A. L., Suarez-Castillo, A. N., Espinosa-Romero, M. J., Aburto-Oropeza, O., Cisneros-Montemayor, A. M., Cruz-Piñón, G., Danemann, G., Giron-Nava, A., Gonzalez-Cuellar, O., Lasch, C., del Mar Mancha-Cisneros, M., Marinone, S. G., Moreno-Báez, M., Morzaria-Luna, H. N., Reyes-Bonilla, H., Torre, J., Turk-Boyer, P., Walther, M., & Weaver, A. H. (2018). Ecological guidelines for designing networks of marine reserves in the unique biophysical environment of the Gulf of California. Reviews in Fish Biology and Fisheries, 28(4), 749–776. https://doi.org/10.1007/s11160-018-9529-y Munroe, S. E. M., Simpfendorfer, C. A., Moloney, J., & Heupel, M. R. (2015). Nearshore movement ecology of a medium-bodied shark, the creek whaler Carcharhinus fitzroyensis. Animal Biotelemetry, 3(1), 1–12. https://doi.org/10.1186/s40317-015-0026-y Ormond RFG, Dawson Shepherd AR, Price ARG, Pitts JR (1984) Management of Red Sea coastal resources: recommendations for marine protected areas. IUCN/MEPA, Kingdom of Saudi Arabia Osman, E. O., Smith, D. J., Ziegler, M., Benjamin, K., Conrad, C., Suggett, D. J., El- haddad, K. M., & Voolstra, C. R. (2018). Thermal refugia against coral bleaching throughout the. June 2017, 474–484. https://doi.org/10.1111/gcb.13895 Pandolfi, J. M., Bradbury, R. H., Sala, E., Hughes, T. P., Bjorndal, K. A., Cooke, R. G., McArdle, D., McClenachan, L., Newman, M. J. H., Paredes, G., Warner, R. R., & Jackson, J. B. C. (2003). Global trajectories of the long-term decline of coral reef ecosystems. Science, 301(5635), 955–958. https://doi.org/10.1126/science.1085706 88

Pearman, J. K., Kürten, S., Sarma, Y. V. B., Jones, B. H., & Carvalho, S. (2016). Biodiversity patterns of plankton assemblages at the extremes of the Red Sea. FEMS Microbiology Ecology, 92(3), 1–13. https://doi.org/10.1093/femsec/fiw002 Pearman, J. K., Aylagas, E., Voolstra, C. R., Anlauf, H., Villalobos, R., & Carvalho, S. (2019). Disentangling the complex microbial community of coral reefs using standardized Autonomous Reef Monitoring Structures (ARMS). Molecular Ecology, 28(15), 3496–3507. https://doi.org/10.1111/mec.15167 Pombo-Ayora, L., Coker, D. J., Carvalho, S., Short, G., & Berumen, M. L. (2020). Morphological and ecological trait diversity reveal sensitivity of herbivorous fish assemblages to coral reef benthic conditions. Marine Environmental Research, 162(March), 105102. https://doi.org/10.1016/j.marenvres.2020.105102 Price, A.R.G., P.A.H. Medley, R.J. McDowall, A.R. Dawson Shepherd, P.J. Hogarth and R.F.G. Ormond. 1987. Aspects of mangal ecology along the Red Sea coast of Saudi Arabia. Journal of Natural History 21: 449–64. Raitsos, D. E., Pradhan, Y., Brewin, R. J. W., Stenchikov, G., & Hoteit, I. (2013). Remote Sensing the Phytoplankton Seasonal Succession of the Red Sea. PLoS ONE, 8(6). https://doi.org/10.1371/journal.pone.0064909 Reaka-Kudla, M. L. (1997). The Global Biodiversity of Coral Reefs: A comparison with Rain Forests. Biodiversity II: Understanding and Protecting Our Biological Resources, October, 83–108. http://www.nap.edu/openbook/0309052270/html/83.html Roberts M.B., Jones, G.P, McCormick, M.I., Munday P.L., Neale, S., Thorrold S.,. Robitzch V.S.N &, Berumen, M.L. (2016). Homogeneity ofcoral reefcommunities across 8 degrees of latitude in the Saudi Arabian Red Sea. Marine Pollution Bulletin, 105, 558-565 Roberts, C. M., Dawson A. Shepherd, R. and Ormond R. F. G. (1992). Large-Scale Variation in Assemblage Structure of Red Sea Butterflyfishes and Angelfishes. Journal of Biogeography, Vol. 19, No. 3, pp. 239-250 Rowlands, G., Purkis, S., Riegl, B., Metsamaa, L., Bruckner, A., & Renaud, P. (2012). Satellite imaging coral reef resilience at regional scale. A case-study from Saudi Arabia. Marine Pollution Bulletin, 64(6), 1222–1237. https://doi.org/10.1016/j.marpolbul.2012.03.003 Sadovy, Y., & Domeier, M. (2005). Are aggregation-fisheries sustainable? Reef fish fisheries as a case study. Coral Reefs, 24(2), 254–262. https://doi.org/10.1007/s00338-005-0474-6 Sandin, S. A., Smith, J. E., Demartini, E. E., Dinsdale, E. A., Donner, S. D., Friedlander, A. M., Konotchick, T., Malay, M., Maragos, J. E., Obura, D., Paulay, G., Richie, M., Rohwer, F., Schroeder, R. E., Walsh, S., Jackson, J. B. C., 89

Knowlton, N., & Sala, E. (2008). Baselines and Degradation of Coral Reefs in the Northern Line Islands. 3(2). https://doi.org/10.1371/journal.pone.0001548 Schmidt-Roach, S., Duarte, C. M., Hauser, C. A. E., & Aranda, M. (2020). Beyond Reef Restoration: Next-Generation Techniques for Coral Gardening, Landscaping, and Outreach. Frontiers in Marine Science, 7(August), 1–8. https://doi.org/10.3389/fmars.2020.00672 Shannon, Claude E, and Warren Weaver (1949) Themathematical theory of communication. Urbana: University of Illinois Press:5. Shellem, C. T. (2020). Assessing the impact of a mass coral bleaching event on Red Sea fisheries. Masters thesis. King Abdullah University of Science and Technology. Sheppard C, Price A, Roberts C (1992) Marine ecology of the Arabian region. Academic Press, London, p 359. Spalding M.D., Ravilious C. and Green E.P. (2001). World Atlas of Coral Reefs. Prepared at the UNEP World Conservation Monitoring Centre. University of California Press, Berkeley, USA. Spalding, M., Burke, L., Wood, S. A., Ashpole, J., Hutchison, J., & zu Ermgassen, P. (2017). Mapping the global value and distribution of coral reef tourism. Marine Policy, 82(May), 104–113. https://doi.org/10.1016/j.marpol.2017.05.014 Stuart-Smith, R. D., Bates, A. E., Lefcheck, J. S., Duffy, J. E., Baker, S. C., Thomson, R. J., Stuart-Smith, J. F., Hill, N. A., Kininmonth, S. J., Airoldi, L., Becerro, M. A., Campbell, S. J., Dawson, T. P., Navarrete, S. A., Soler, G. A., Strain, E. M. A., Willis, T. J., & Edgar, G. J. (2013). Integrating abundance and functional traits reveals new global hotspots of fish diversity. Nature, 501(7468), 539–542. https://doi.org/10.1038/nature12529 Taylor, M. S. & Hellberg, M. E. (2003). Genetic evidence for local recruitment of pelagic larvae in a Caribbean reef fish. Science 299, 635–645 Tesfamichael D, Pauly D (2016) The Red Sea ecosystem and fisheries. Springer, New York Troyer, E. M., Coker, D. J., & Berumen, M. L. (2018). Comparison of cryptobenthic reef fish communities among microhabitats in the Red Sea. PeerJ, 2018(6). https://doi.org/10.7717/peerj.5014 Tunesi, L., Molinari, A., & Salvati, E. (2006). Fish assemblage of the marine protected area of Cinque Terre (NW Mediterranean Sea): First characterization and assessment by visual census. Chemistry and Ecology, 22(SUPPL. 1). https://doi.org/10.1080/02757540600720193 Tuya, F., Wernberg, T., & Thomsen, M. S. (2011). The relative influence of local to regional drivers. 217–234. https://doi.org/10.1111/j.1095-8649.2011.03015.x 90

Underwood, A. J., Chapman, M. G., & Connell, S. D. (2000). To make progress on processes without understanding the patterns. Journal of Experimental Marine Biology and Ecology, 250(2000), 97–115. Villéger, S., Brosse, S., Mouchet, M., Mouillot, D., & Vanni, M. J. (2017). Functional ecology of fish: current approaches and future challenges. Aquatic Sciences, 79(4), 783–801. https://doi.org/10.1007/s00027-017-0546-z Voolstra, C. R., & Berumen, M. L. (2019). Coral Reefs of the Red Sea. Springer. https://doi.org/10.1007/978-3-030-05802-9 Vroom, P.S., Page, K. N., Kenyon, J.C. & Brainard, Russell E.(2006) Algae dominated reefs. American Scientist, 94. Williams, I. D., & Polunin, N. V. C. (2001). Large-scale associations between macroalgal cover and grazer biomass on mid-depth reefs in the Caribbean. Coral Reefs, 19(4), 358–366. https://doi.org/10.1007/s003380000121 Wilson, S. K., Nicholas Graham, Morgan S. Pratchet, Geoffrey P. Jones, & Polunin, N. V. C. (2006). Multiple disturbances and the global degradation of coral reefs : are reef fishes at risk or resilient ? Global Change Biology, 12, 2220–2234. https://doi.org/10.1111/j.1365-2486.2006.01252.x

APPENDICES

Appendix 1. Fish taxa found in the surveys. NS=Nearshore, MS=Midshore, OS=Offshore

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Acanthuridae Acanthurus gahhm X X X X X X X X X X X Acanthurus nigrofuscus X X X X X X X X X X X Acanthurus sohal X X X X X Ctenochaetus striatus X X X X X X X X X X X Naso brevirostris X X X X Naso elegans X X X X X X X X X X X Naso hexacanthus X X X X Naso unicornis X X X X X X Zebrasoma desjardinii X X X X X X X X X X X Zebrasoma xanthurum X X X X X X X X X Apogonidae Apogon sp. X X Cheilodipterus macrodon X Taeniamia fucata X X Balistidae Balistapus undulatus X X X X X X X X X X X Balistoides viridescens X X X X X Odonus niger X X X Pseudobalistes flavimarginatus X X X 92

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Pseudobalistes fuscus X X Rhinecanthus assai X X X X X X X X Sufflamen albicaudatus X X X X X X X X Blenniidae Cirripectes castaneus X Ecsenius dentex X Ecsenius frontalis X Meiacanthus nigrolineatus X X Caesionidae Caesio lunaris X X X X X X X X Caesio sp. X Caesio striata X X X X X Caesio suevica X X Carangidae Carangidae sp. X Carangoides bajad X X X X X X X X Caranx ignobilis X Caranx melampygus X X X X X Caranx sexfasciatus X Caranx sp. X Chaetodontidae Chaetodon auriga X X X X X X X X X Chaetodon austriacus X X X X X X X X X X X Chaetodon fasciatus X X X X X X X X X X X 93

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Chaetodon larvatus X X X X X X X Chaetodon lineolatus X X X X Chaetodon melannotus X X X X X X Chaetodon mesoleucos X X X X X X Chaetodon paucifasciatus X X X X Chaetodon semilarvatus X X X X X X X X Chaetodon trifascialis X X X X X X Heniochus intermedius X X X X X X X X X X X Cirrhitidae Cirrhitichthys oxycephalus X Paracirrhites forsteri X X X X X X X X Dasyatidae Taeniura lymma X X X Diodontidae X Diodon hystrix X Ephippidae X X X Platax boersi X X Platax sp. X Platax teira X Ginglymostomatidae Nebrius ferrugineus X Gobiidae Bryaninops natans X Gobiodon citrinus X 94

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Koumansetta hoesei X X X X X Haemulidae Plectorhinchus gaterinus X X X X X X X Holocentridae Myripristis murdjan X X X X X X X X X Myripristis xanthacra X Neoniphon sammara X X X X X X X X Sargocentron caudimaculatum X X X X X X X X X X Sargocentron spiniferum X X X X X X X Kyphosidae Kyphosus cinerascens X X Kyphosus sp. X Kyphosus vaigiensis X Labridae Anampses caeruleopunctatus X Anampses lineatus X Anampses meleagrides X X X X Anampses twistii X X X X X X X X Bodianus anthiodes X X X X X X X Bodianus axillaris X X X X X Bodianus diana X X X Cheilinus abudjubbe X X X X X X X X Cheilinus lunulatus X Cheilinus quinquecinctus X X X X X X X X X X 95

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Cheilinus undulatus X Coris aygula X X X X Coris variegata X X Epibulus insidiator X X X X X X X X X X Gomphosus caeruleus X X X X X X X X X X X Halichoeres hortulanus X X X X X X X X X X X Halichoeres marginatus X Halichoeres scapularis X X X Hemigymnus melapterus X X Hemigymnus sexfasciatus X X X X X X Hologymnosus annulatus X X Labridae sp. X X Labroides dimidiatus X X X X X X X X X X X Larabicus quadrilineatus X X X X X X X X X X X Novaculichthys taeniourus X Oxycheilinus diagramma X X X X X X X X Oxycheilinus mentalis X X X X X X X X X X Paracheilinus octotaenia X X X X X X X X X Pseudocheilinus evanidus X X X X X X X X X X Pseudocheilinus hexataenia X X X X X X X X X X X Pseudodax moluccanus X X X X X Stethojulis albovittata X X X X X X 96

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Thalassoma lunare X X X X X X X X X Thalassoma rueppelli X X X X X X X X Wetmorella nigropinnata X Lethrinidae Lethrinus borbonicus X X Lethrinus harak X X Lethrinus mahsena X X X X Lethrinus microdon X X X Lethrinus obsoletus X X Lethrinus sp. X X X X Lethrinus xanthochilus X X Lutjanus argentimaculatus X Monotaxis grandoculis X X X X X X X X Lutjanidae Lutjanus argentimaculatus X Lutjanus bohar X X X X X X X X Lutjanus ehrenbergii X X X X X X X X X X Lutjanus fulviflamma X X X X X Lutjanus fulvus X X Lutjanus gibbus X X X X Lutjanus kasmira X X X X X X X X X Lutjanus monostigma X X X Macolor niger X X X X X X Monacanthidae 97

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Aluterus scriptus X X Amanses scopas X X X X X Cantherhines pardalis X Monodactylidae Monodactylus argenteus X X Mullidae Mulloidichthys sp. X X Mulloidichthys vanicolensis X X X Parupeneus cyclostomus X X X X X X X Parupeneus forsskali X X X X X X X X X X X Parupeneus macronema X Muraenidae Gymnothorax javanicus X X X X X X Nemipteridae Scolopsis ghanam X X X Scolopsis taeniata X Ostraciidae Ostracion cyanurus X X X X X X X Phempheridae Pempheris vanicolensis X Pervagor randalli X Pinguipedidae Parapercis hexophthalma X X X X X X X X X Pomacanthidae 98

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Centropyge multispinis X X X X X X X X X X X Pomacanthus asfur X X X X X X X Pomacanthus imperator X X X X Pomacanthus maculosus X X X X Pygoplites diacanthus X X X X X X X X X X Pomacentridae Abudefduf sexfasciatus X X X X X Abudefduf vaigensis X X X X Amblyglyphidodon flavilatus X X X X X X X X Amblyglyphidodon indicus X X X X X X X X X X X Amphiprion bicinctus X X X X X X Chromis dimidata X X X X X X X X X X X Chromis flavaxilla X X X X X X X X X X Chromis pembae X X Chromis trialpha X X Chromis viridis X X X X X X Chromis weberi X X X X X X Dascyllus abudafur X X X X X X X X Dascyllus marginatus X X X X Dascyllus trimaculatus X X X X X X X X Neoglyphidodon melas X X X X X X X X Neopomacentrus cyanomos X Neopomacentrus miryae X 99

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Neopomacentrus xanthurus X X X X X Plectroglyphidodon lacrymatus X X X X X X X X X X X Pomacentrus albicaudatus X X X X Pomacentrus leptus X X X X Pomacentrus sulfereus X X X X X X X X X X X Pomacentrus trichrourus X X X X X X X X X Pomacentrus trilineatus X X X X Pristotis cyanostigma X Stegastes nigricans X X Priacanthidae Priacanthus hamrur X X Pseudochromidae Pseudochromis dixurus X X Pseudochromis flavivertex X X X X X X X X Pseudochromis fridmani X X X X X X X X Pseudochromis olivaceus X X X X X Pseudochromis springeri X Ptereleotridae Ptereleotris evides X X X Scaridae Bolbometopon muricatum X X 100

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Calotomus viridescens X X X X X X X X X Cetoscarus bicolor X X X X X X X X Chlorurus genozonatus X X X X X Chlorurus gibbus X X X X X X X Chlorurus sordidus X X X X X X X X X X X Hipposcarus harid X X X X X X X X X X Scarus collana X X Scarus ferrugineus X X X X X X X X X X X Scarus frenatus X X X X X X X Scarus fuscopurpureus X X Scarus ghobban X X X X Scarus niger X X X X X X X X X X X Scarus psittacus X Scarus sp. X X X X X X X X Scombridae Gymnosarda unicolor X Scorpaenidae Pterois cincta X X X X X X X Pterois miles X X Serranidae Aethaloperca rogaa X X X X X X X X Anyperodon leucogrammicus X X X X X Cephalopholis argus X X X X X X X X X X Cephalopholis hemistiktos X X X X X X X X X X X 101

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Cephalopholis miniata X X X X X X X X X Cephalopholis sexmaculata X X Diploprion drachi X X X X Epinephelus fasciatus X Epinephelus fuscoguttatus X Epinephelus malabaricus X Epinephelus summana X X X X X X X Epinephelus tauvina X X X X X Grammistes sexlineatus X Plectropomus areolatus X X X X Plectropomus pessuliferus marisrubri X X X X X Pseudanthias squamipinnis X X X X X X X X X Variola louti X X X X X X X X Siganidae Siganus argenteus X Siganus luridus X X X X X X Siganus rivulatus X X X Siganus stellatus X X X X X X X X X Sparidae Acanthopagrus bifasciatus X X X X

102

Duba Um Lujj Thuwal Jeddah Al Lith Farasan Banks TAXA NS NS NS MS OS NS NS MS NS MS OS Sphyraenidae Sphyraena barracuda X Sphyraena flavicauda X Synodontidae Synodontidae sp. X Synodus dermatogenys X Synodus variegatus X X Tetraodontidae Arothron diadematus X X X X X X X Arothron hispidus X Canthigaster coronata X Canthigaster margaritata X X