The Ecological and Functional Roles of Commercially Important Rabbitfish Species in a Marine Environment

The ecological and functional roles of commercially important rabbitfish species in a marine environment

Ameer Ebrahim BSc (Hons)

A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2019 School of Biological Sciences

Abstract

Herbivorous fishes have consistently been found to have a significant positive impact on the resilience of coral reefs. Their feeding and foraging actions shape the capacity of reefs to both resist, and recover from, disturbance. Relatively little is known, however, about which ecological features in particular influence their distributions, movements, and foraging behaviours. Furthermore, the manner in which their biomass is altered by human actions remains poorly understood. Understanding and appreciating these key factors may help improve global management strategies of fisheries and coral reef habitats. Overall, this thesis provides a framework to establish the ecological functions of commercially important herbivorous fishes. This will not only help authorities establish effective management strategies to curb fishing pressure, but will also provide an indication of the capacity of reefs to resist or recover from disturbances that lead to algal dominance.

At present, there is a noticeable gap in the literature pertaining to the ecological role that individual herbivorous fish species play; contemporary research instead focusses on their overall contribution. To bridge this gap and to render this research relevant to management efforts in the Seychelles, this thesis explored the distribution, ecological functions, and spatial ecology of key rabbitfish species in this archipelago of the Indian Ocean. Rabbitfishes are a crucial food source to the majority of the Western Indian Ocean region, comprising the bulk of their artisanal fisheries. In the Seychelles, in particular, they constitute over 60% of the total artisanal catch and, notably, there are no measures aimed at management of stocks currently being implemented or enforced.

This thesis demonstrates that the overall distribution of herbivorous fishes is shaped by the habitat characteristics of individual reefs, and not by the presence of marine reserves. Species level analyses on rabbitfishes reveal that they were also influenced by the features of reef habitats, but not the conservation status of individual reefs. Furthermore, this research reveals that fishes from different functional groups occupy distinct reef habitats, suggesting that this may lead to spatial separation in the

ii distribution of different forms of herbivory, and possibly reef resilience, in the Seychelles.

This thesis investigated the ecological functions of rabbitfishes, focussing on their foraging rates and substrate preference, among diverse sites with varying benthic habitats. It reveals that rabbitfishes display both browsing and grazing traits, indicating possible functional complementarity between species. This is an important distinction as it implies that the conservation of different species may result in distinct shifts in the competitive dominance of coral and algae.

Finally, this thesis examines the spatial ecology of the most commercially important fish species of the artisanal fishery in the Seychelles: the shoemaker spinefoot (Siganus sutor); it investigated whether they act as a mobile link between networked habitats, and whether their movements differed between day and night. Detection patterns reveals them to be diurnal herbivores, with only rare nocturnal movements. Furthermore, their movements are influenced by seagrass and coral abundance. Identification of such links between networked habitats are relevant for the purposes of implementing effective fisheries and habitat management strategies, and have a crucial bearing on sites around the archipelago which are to be declared marine reserves.

iii

Declaration by author

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, financial support and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my higher degree by research candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis and have sought permission from co-authors for any jointly authored works included in the thesis.

Publications during candidature

Peer-reviewed papers

Ebrahim, A., Martin, T. S. H., Mumby, P. J., Olds, A. D., Tibbetts I. R. (2020) Differences in diet and foraging behaviour of commercially important rabbitfish species on coral reefs in the Indian Ocean. Coral Reefs. https://doi.org/10.1007/s00338-020-01918-6

Ebrahim, A., Bijoux, J. P., Mumby, P. J., Tibbetts I. R. (2020) The commercially important Shoemaker spinefoot, Siganus sutor connects coral reefs to neighbouring seagrass meadows. Journal of Fish Biology. https://doi.org/10.1111/jfb.14297

Conference abstracts

Ebrahim, A., Bijoux, J. P., Mumby, P. J., Olds, A. D., Tibbetts I. R. Reef features, not marine reserves, shape the distribution of herbivorous fishes on coral reefs in the Seychelles (Oral presentation). Workshop on status of Seychelles coral reefs, reef resilience and coral reef restoration, Seychelles.

Ebrahim, A., Bijoux, J. P., Mumby, P. J., Olds, A. D., Tibbetts I. R. Reef features, not marine reserves, shape the distribution of herbivorous fishes on coral reefs in the Seychelles (Oral presentation). The Eleventh Western Indian Ocean Marine Science Association (WIOMSA) Scientific Symposium, Mauritius.

Manuscripts in prep for peer-review

Ebrahim, A., Bijoux, J. P., Mumby, P. J., Olds, A. D., Tibbetts I. R. Reef features, not marine reserves, shape the distribution of herbivorous fishes on coral reefs in the Seychelles

Manuscripts in prep for peer-review included in this thesis

Ebrahim, A., Bijoux, J.P., Mumby, P.J., Olds, A.D., Tibbetts I.R. (submitted to the Journal of Fish Biology) ‘Reef features, not marine reserves, shape the distribution of herbivorous fishes on coral reefs in the Seychelles

This work was incorporated as Chapter 2

Contributor Statement of contribution

Ameer Ebrahim Designed study (70%) (Candidate) Conducted fieldwork (80%)

Conducted analysis (80%)

Wrote and edited paper (70%)

Jude P. Bijoux Designed study (20%)

Wrote and edited paper (5%)

Peter J. Mumby Designed study (5%)

Conducted analysis (20%)

Wrote and edited paper (5%)

Andrew D. Olds Wrote and edited paper (10%)

Rodney Melanie Conducted fieldwork (20%)

Ian R. Tibbetts Designed study (5%) Wrote and edited paper (10%)

Publications included in this thesis

This work was incorporated as Chapter 3

Contributor Statement of contribution

Ameer Ebrahim Designed study (70%) (Candidate) Conducted fieldwork (80%)

Conducted analysis (60%)

Wrote and edited paper (70%)

Tyson SH. Martin Conducted analysis (30%)

Wrote and edited paper (5%)

Peter J. Mumby Designed study (20%)

Conducted analysis (10%)

Wrote and edited paper (5%)

Andrew D. Olds Wrote and edited paper (10%)

Rodney Melanie Conducted fieldwork (20%)

Ian R. Tibbetts Designed study (10%)

Wrote and edited paper (10%)

vii

This work was incorporated as Chapter 4

Contributor Statement of contribution

Ameer Ebrahim Designed study (70%) (Candidate) Conducted fieldwork (60%)

Conducted analysis (70%)

Wrote and edited paper (75%)

Jude P. Bijoux Designed study (20%)

Conducted fieldwork (20%)

Conducted analysis (20%)

Peter J. Mumby Designed study (10%)

Conducted analysis (10%)

Wrote and edited paper (5%)

Rodney Melanie Conducted fieldwork (10%)

Ian R. Tibbetts Wrote and edited paper (20%)

viii

Contributions by others to the thesis

Tibbetts IR, Mumby PJ, Bijoux JP, and Olds AD contributed to the conception and design of this research, advised on methods and analyses, and provided critical comments on the thesis and all the associated manuscripts that have been submitted for peer-review.

Mumby PJ, and Bijoux JP provided detailed advice on statistical analyses throughout the thesis. Whilst, Martin TSH provided statistical advice specifically on Chapter 3. All contributions are specifically acknowledged in the relevant chapters.

Melanie R, and Bijoux JP assisted with fieldwork and data collection. Their contributions are specifically acknowledged in the relevant chapters.

Statement of parts of the thesis submitted to qualify for the award of another degree

None.

Research Involving Human or Animal Subjects

Animal subjects used in this study were approved under AEC Approval Number: SBS/518/15/SFA by the Animal Welfare Unit, UQ Research and Innovation, The University of Queensland, Australia.

A copy of the ethics approval letter is included in the thesis Appendix A.

Acknowledgements

To my supervisors, Assoc. Prof. Ian Tibbetts, Prof. Peter Mumby, Dr. Jude Bijoux and Dr. Andrew Olds, your guidance, support and supervision have been greatly appreciated. I look forward to many more years of friendship and collaborations with each of you.

To my wife, Michelle, I am grateful for your unconditional support and strength.

To my son, Adam, who joined us right before I was initially due to submit my thesis, you have put everything into perspective and will always be my greatest pride and joy.

To my in-laws, thank you for your unwavering support and guidance.

To my colleague and friend, Tyson Martin, thank you for all your help and advice over the past decade. I look forward to future collaborations in Seychelles with you.

I am grateful to the Seychelles Government for their support, without which I may not have been able to carry out my research of choice for my PhD.

Financial support

This research was supported by the Seychelles Fishing Authority (SFA), Seychelles.

Keywords

Herbivory; Phase-shifts; Siganidae; Fish distributions; Behavioural foraging observations; Ecological functions; Movement; Connectivity; Management

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 060205 Marine and Estuarine Ecology (incl. Marine Ichthyology), 60% ANZSRC code: 060104, 060201 Behavioural Ecology, 30% ANZSRC code: 050205 Environmental Management, 10%

Fields of Research (FoR) Classification

FoR code: 0602, Ecology, 80% FoR code: 0501 Ecological Applications, 20%

Table of contents

Chapter 1 – General introduction ...... 1

1.1 Herbivory on coral reefs ...... 1

1.2 Rabbitfishes and Herbivory ...... 2

1.3 Rabbitfishes in the Seychelles ...... 6

1.4 Current status of coral reefs in the Seychelles ...... 6

1.5 Specific aims and thesis synopsis ...... 7

Chapter 2 – Reef features, not marine reserves, shape the distribution of herbivorous fishes on coral reefs in the Seychelles ...... 9 2.1 Introduction ...... 10

2.2 Materials and Methods ...... 13

2.2.1 Study region and sampling design ...... 13

2.2.2 Assessing benthic habitat characteristics ...... 15

2.2.3 Surveying herbivorous fishes ...... 16

2.2.4 Data analysis ...... 20

2.3 Results ...... 21

2.3.1 Effects on herbivore functional groups ...... 21

2.3.2 Effects on rabbitfishes (Siganidae) ...... 27

2.4 Discussion ...... 31

2.4.1 Effects on herbivore functional groups ...... 32

2.4.2 Effects on rabbitfishes (Siganidae) ...... 35

Chapter 3 – Differences in diet and foraging behaviour of commercially important rabbitfish species on coral reefs in the Indian Ocean ...... 36 3.1 Introduction ...... 37

3.2 Materials and Methods ...... 40

3.2.1 Study region and sampling design ...... 40

3.2.2 Surveying the composition of coral and algal assemblages ...... 41

xii

3.2.3 Quantifying the foraging behaviours of herbivorous rabbitfishes ...... 42

3.2.4 Data analysis ...... 43

3.3 Results ...... 45

3.3.1 Foraging rates and behaviours ...... 45

3.3.2 Foraging selectivity ...... 50

3.4 Discussion ...... 51

3.4.1 Foraging rates and behaviours ...... 51

3.4.2 Foraging selectivity ...... 54

Chapter 4 – The commercially important Shoemaker spinefoot, Siganus sutor connects coral reefs to neighbouring seagrass meadows ...... 56 4.1 Introduction ...... 57

4.2 Materials and Methods ...... 59

4.2.1 Study region ...... 59

4.2.2 Acoustic monitoring system ...... 61

4.2.3 Range test ...... 61

4.2.4 Capture and tagging of fish ...... 62

4.2.5 Data analysis ...... 63

4.3 Results ...... 66

4.3.1 Home Range ...... 66

4.3.2 Daily movement patterns ...... 68

4.3.3 Activity in each habitat ...... 68

4.4 Discussion ...... 74

Chapter 5 – General discussion: next steps in the conservation of rabbitfishes ...... 78 5.1 Overview of key findings ...... 78

5.2 Limitations and avenues for future research ...... 80

5.3 Management of the fishery moving forward ...... 82

xiii

References ...... 84

Appendices ...... 102

Appendix A: Ethics Approval letter ...... 102

Appendix B: Chapter 2, Supplementary Material ...... 104

Appendix C: Chapter 3, Supplementary Material ...... 110

xiv

List of figures

Figure 2.1 Map showing the 24 study sites across six different inner islands of the Seychelles. Highlighted on the map are the MPAs and reef type (carbonate or granitic) ...... 14 Figure 2.2 Stacked bar chart showing the percentage cover of each benthic material at each reef surveyed ...... 22 Figure 2.3 Bar chart showing the average rugosity ratio (±SE) at each surveyed site

...... 23 Figure 2.4 Stacked Bar charts showing the total mean biomass of each functional group at; a. each island (Mahé vs. St. Anne vs. La Digue vs. Praslin vs. Curieuse vs. Cousin), b. by management status (unprotected vs. protected), and c. by reef type (granitic vs. carbonate) ...... 24 Figure 2.5 Two-dimensional Principal Components Analysis (PCA) illustrating the similarity of herbivorous fish biomass (g.m-2) (segmented for each functional group), and benthic habitat characteristics (presented as vectors). Numerical values in the text box demonstrate biomass range (g.m-2) of each functional group...... 27 Figure 2.6 Two-dimensional Principal Components Analysis (PCA) illustrating the similarity of rabbitfishes biomass (g.m-2) (segmented for each species), and benthic habitat characteristics (presented as vectors). Numerical values in the text box demonstrate biomass range (g.m-2) of each species ...... 31 Figure 3.1 Map showing the 16 study sites across seven different inner islands of the Seychelles. Also highlighted on the map are the marine protected areas...... 41 Figure 3.2 Average proportion of bites (± standard errors) on the different resources by the four species of rabbitfish. Resources included; Fleshy macroalgae fronds (FMAF), fleshy macroalgae thallus (FMAT), seagrass epiphytes (SGE), seagrass blades (SGB) and turf algae (TA). Also displayed are the total bites (± standard errors) for each species ...... 47

Figure 3.3 Two-dimensional Principal coordinates analysis (PCO) illustrating the bite rates (bites.min-1) of the four rabbitfish species on each type of resource foraged. Distances between samples on the ordination attempt to match corresponding dissimilarities in community structure ...... 48

Figure 3.4 Bite rates (bites.min-1) for each rabbitfish species at different time periods...... 49 Figure 3.5 Bar charts showing rabbitfish resource electivity (Vanderploeg and Scavia’s relativized index, Ei) averaged across sites for each the five foraged items: fleshy macroalgal fronds (FMAF), fleshy macroalgal thallus (FMAT), seagrass epiphytes (SGE), seagrass blades (SGB) and turf algae (TA). Average electivity is represented on the vertical axis: values >0 represent active selection disproportionate to abundance, values <0 represent resource avoidance ...... 50 Figure 4.1 Map of Denis Island, highlighting the benthic substrate composition, as well as the location of the 22 active receivers (S1-22) deployed. Each station is displayed with the maximum 150 m range of each receiver. Also highlighted on the map are the tagging locations (T1-T3)...... 60 Figure 4.2 Proportion number of detections of S. sutor at each station during the whole study period (separated into diurnal and nocturnal detections) ...... 68 Figure 4.3 Bar chart of the number of detections (averaged over the whole study period, and separated into diurnal and nocturnal detections) (±SE)) within each habitat type…………………………………………………………………………………………..69 Figure 4.4 Bar chart demonstrating the mean diel variation in the number of acoustic detections recorded between 17 November 2016 to 10 May 2017. Time bin 0 corresponds to 00:00 to 00:59 h...... 70 Figure 4.5 Average number of detections recorded in regions with varying covers of coral and seagrass (m2) during diurnal and nocturnal time periods ...... 74 Figure 4.6 Predicted average number of detections in regions with varying covers of coral and seagrass (m2) ...... 75

xvi

List of tables

Table 1.1 Summary of all currently available research on the functional ecology of rabbitfishes on tropical coral reefs ...... 4

Table 2.1 Marine Protected areas around the Inner Seychelles Islands ...... 15

Table 2.2 List of the herbivore species recorded in this study showing their functional behaviour, and total biomass averaged across the 24 study sites ...... 18

Table 2.3 Summary of the permutational multivariate analysis of variance (PERMANOVA) showing the influence of benthic habitat characteristics, and factors on the biomass of herbivore functional groups ...... 25

Table 2.4 Summary of the permutational multivariate analysis of variance (PERMANOVA) showing the influence of benthic habitat characteristics, and factors on the biomass of rabbitfishes species...... 29

Table 3.1 Results of the PERMANOVA performed on the foraging rates (bites.min-1). Species, site and time period were fixed factors. Significant interactions are highlighted in bold ...... 45

Table 4.1 Summary of Movement data of 15 S. sutor tagged with V9-1L transmitters ...... …..67

Table 4.2 Summary of visited stations and principal station location of each tagged fish ...... 71

Table 4.3 Summary of the total coral and seagrass areas (m2) estimated at each station, using satellite imagery. Also displayed are the total number of detections recorded at each station during the experiment ...... 73

xvii

List of Abbreviations

2D Two-dimensional 3D Three-dimensional π pi BCS Bray Curtis Similarity Bites.min-1 Bites per minute cm centimetre CR Coral Rubble DC Dead Coral df Degrees of Freedom EAM Epilithic Algal Matrix FL fork length FMA Fleshy Macroalgae FMAF Fleshy Macroalgal Fronds FMAT Fleshy Macroalgal Thallus F-statistic test statistic that has an F-distribution under the null hypothesis g.L-1 gram per litre g.mˉ² grams per meter-squared (Biomass unit measurement) GBR Great Barrier Reef GLM Generalized Linear Model h hour ha hectares hrs hours km2 kilometre-squared L Litre LC Live Coral m metre m2 metre squared (area) min minute

xviii

min-1 per minute MDT Median Distance Travelled MLD Minimum Linear Distance mm millimetre MPA Marine Protected Area NGO Non-Governmental Organisation nMDS Non-metric multidimensional scaling NS Nature Seychelles PCA Principal Components Analyses PCO Principal Coordinate Analysis PERMANOVA Permutational Multivariate Analysis of Variance PERMDISP Permutational Analysis of Multivariate Dispersions PIT Point Intercept Transect Pperm Permutation calculated probability pseudoF Validity Index of F-statistic P-value Calculated probability RI Residency Index S Sand SCUBA Self-Contained Underwater Breathing Apparatus SE Standard Error SFA Seychelles Fishing Authority SG Seagrass SGB Seagrass Blades SGE Seagrass Epiphytes SNPA Seychelles National Parks Authority STD Dev Standard Deviation TA Turf Algae t statistic ratio of estimated value of a parameter from its hypothesized value to its standard error. UNDP United Nations Development Programme WIO Western Indian Ocean

xix

Chapter 1 – General introduction

1.1 Herbivory on coral reefs

Coral reefs are amongst the most threatened marine systems globally (Walther et al., 2002). They cover less than 0.1% of the world’s oceans (Spalding et al., 2002), but remain one of the most taxonomically diverse places, supporting nearly one third of our marine fish species (Raeka-Kudla, 1996). Coral reefs also provide many key services, including coastal protection, fisheries productivity, and economic revenue linked to tourism (Edwards et al., 2014). However, they are at risk of being destroyed by direct and/or indirect anthropogenic effects, such as overfishing (Dulvy et al., 2004; MacNeil et al., 2015). Natural disturbances, such as the effects of El Niño events, can also cause distress to coral reefs. However, in the absence of anthropogenic threats, reefs are usually able to recover within in a few years to a few decades, depending on the extent of the disturbance (Roff et al., 2015). Unfortunately, some reefs have seemingly gone beyond the point of recovery (Wilkinson, 2004), prompting an urgent need to halt or reverse these declines for the future survival of these ecosystems. The promotion of coral reef resilience through effective fisheries management may be one approach to combat the damage posed by anthropogenic threats (Green and Bellwood, 2009).

Coral reef resilience is the capability of reefs to absorb regular disturbances, such as the effects of recurring El Niño events, and restructure coral-dominated systems, rather than shifting to a system dominated primarily by algae (Hughes et al., 2007; Graham et al., 2015). Herbivorous fish species are known to play a vital role in the promotion of coral reef resilience by rapidly grazing the reef substratum and removing large amounts of algae (Polunin and Klumpp, 1992). Reefs that have healthy populations of these fishes are able to limit the accumulation of algal material/ growth, minimise coral-algal competition, and help to maintain reef health by reducing the potential for phase shifts from coral to algal dominance (Hughes et al., 2007).

Large-scale exclusion experiments that stimulated the effects of overfishing, by excluding herbivorous fishes from large (5m × 5m) experiment plots found a rapid increase of macroalgal cover (e.g. Bellwood et al., 2006; Hughes et al., 2007). They demonstrated that when these fishes were allowed to flourish by removing the effect of fishing effort (i.e. open plots), coral cover increased significantly due to extensive coral recruitment and growth. Contrastingly, the elimination of feeding by herbivorous fishes (i.e. within the exclusion plots) caused a dramatic phase shift from a system dominated by epilithic algae and corals, to one overgrown by fleshy macroalgae (Bellwood et al., 2006; Hughes et al., 2007). When the excluded fishes were allowed into to the phase-shifted area (i.e. when exclusion plots were removed), the macroalgae was rapidly removed (Bellwood et al., 2006; Hughes et al., 2007).

Similarly, larger spatial scale applications investigating the effects of marine reserves have shown negative correlations between herbivorous fish biomass and the cover of macroalgae. These studies (e.g. Friedlander et al., 2007; Mumby and Harborne, 2010; Rasher et al. 2013., Bozec et al., 2016; Mellin et al., 2016) revealed that herbivores can act as crucial top-down regulators of macroalgal growth. Most of these studies (i.e. Friedlander et al., 2007; Mumby and Harborne, 2010; Rasher et al. 2013., Mellin et al., 2016) further demonstrated that coral reefs within marine reserves had a larger population of herbivorous fishes and display higher rates of recovery when compared to reefs open to fishing.

1.2 Rabbitfishes and Herbivory

There are three main families of herbivorous fishes on Indo-Pacific tropical reefs: parrotfishes (Labridae), surgeonfishes (Acanthuridae), and rabbitfishes (Siganidae) (Cheal et al., 2012; Hoey et al., 2013; Puk et al., 2016). Traditionally, all three families were positioned in the same functional group and labelled ‘algal removers’ (Choat, 1991). However, it has now become clear that these fishes consume different types of algae, often with distinct effects on coral-algal dynamics, and they are now typically grouped according to their functional traits (Heenan and Williams, 2013). The broadest functional

distinction is between fishes that graze predominantly on epilithic algal matrix (EAM) (i.e. grazers) and those that browse typically on fleshy macroalgae (i.e. browsers) (Heenan and Williams, 2013). Grazers can be further separated into grazers/ detritivores, scrapers and excavators by the amount, and type, of EAM and underlying substrate they remove during feeding (Green and Bellwood, 2009). Detritivores consume copious amounts of turf algae while brushing the EAM for detritus (Marshell and Mumby, 2012), whereas scrapers clean EAM off the reef matrix, and excavators remove EAM by taking bites off the reef matrix (Steneck, 1988; Bellwood and Choat, 1990).

Herbivorous fish studies on tropical coral reefs have been primarily focussed on the ecological roles of parrotfishes and surgeonfishes (e.g. Marshell and Mumby, 2015). By contrast, limited research is available on the functional roles of rabbitfishes on tropical reefs (for exceptions, see Bryan, 1975; Paul et al., 1990; Fox et al., 2009; Fox and Bellwood, 2013; Hoey et al., 2013; Brandl and Bellwood, 2014 (Table 1.1)). It appears that rabbitfishes perform a complementary role to that of parrotfishes and surgeonfishes (Table 1.1). However, the majority of research on rabbitfishes has been restricted to the Great Barrier Reef in eastern Australia (Table 1.1), and little is known of their behaviour and ecology from reefs elsewhere.

Table 1.1: Table summarizing all current research on the functional ecology of rabbitfishes on tropical coral reefs. Study description Location Main findings Reference A study on the feeding ecology of Guam, Western Gut content analysis revealed a diverse range of algal Bryan, 1975 S.spinus Pacific resources.

Conclusion: S. spinus shows high dietary plasticity towards different types of filamentous algae.

Comparing the feeding preferences of Guam, Western Adult S. argenteus were deterred by certain chemically Paul et al., adult and juvenile S. argenteus Pacific defensed algae, while juveniles were not. 1990

Conclusion: functional diversification found between adults and juveniles.

Comparing the feeding behaviour of two Lizard Island, GBR, Distinct feeding behaviours displayed by both species. Fox et al., similar rabbitfish species (S. doliatus and Australia 2009 S. lineatus) S.doliatus displayed a typical herbivore diurnal feeding pattern, with a diet dominated by red thallate algae, and red and green filamentous algae.

S. lineatus stomach contents were dominated by amorphous organic matter (detritus), and revealed traits of a nocturnal feeder.

Conclusion: both species have distinct functional roles.

Comparing the feeding ecology of three Lizard Island, GBR, Revealed a unique functional role for rabbitfishes. The Fox and species of rabbitfish (Siganus corallinus, Australia species studied specifically fed on crevice-dwelling algal or Bellwood, S. puellus and S. vulpinus) to that of benthic organisms. 2013 parrotfishes and surgeonfishes Conclusion: rabbitfish display extensive functional complementarity compared to parrotfish and surgeonfish.

A study on the diet and distribution of Lizard Island and Analysis of stomach contents revealed four distinct groups: Hoey et al., eleven rabbitfish species (S. argenteus, MacGillivray Reef, 1. browsers of leathery brown macroalgae (Siganus 2013 S. canaliculatus, S. corallinus, GBR, Australia canaliculatus, S. javus); S. doliatus, S. javus, S. lineatus, S. 2. croppers of red and green macroalgae (S. puellus, S. punctatissimus, S. punctatus, argenteus, S. corallinus, S. doliatus, S. spinus); S. spinus and S. vulpinus)

3. mixed feeders of diverse algal material, cyanobacteria, detritus and sediment (S. lineatus, S. punctatissimus, S. punctatus, S. vulpinus); and 4. Feeder of sponges (S. puellus).

The distribution of these rabbitfishes displayed clear cross- shelf variation; Biomass was greatest on inner-shelf reefs, decreasing markedly on mid- and outer-shelf reefs.

Conclusion: Diet and distribution vary within the Siganidae family, and emphasis must be placed on the importance of examining function on a species-by-species basis.

A study comparing the functional ecology Lizard Island, GBR, Showed that rabbitfishes feed over a greater range of Brandl and of the three main herbivorous fish groups, Australia microhabitats than either surgeonfishes or parrotfishes, and Bellwood, parrotfish, surgeonfish and rabbitfish that they penetrated the reef matrix to exploit concealed 2014 surfaces of various substratum types.

Conclusion: rabbitfishes display extensive functional complementarity compared to parrotfishes and surgeonfishes.

A study comparing the feeding ecology of Northern Red Sea S. rivulatus and Z. desjardinii, both known to be herbivorous, Bos et al., S. rivulatus (rabbitfish) and Zebrasoma region were observed targeting and feeding on ctenophores and 2017 desjardinii (surgeonfish) scyphozoans. S. rivulatus dominated the feeding.

Conclusion: both species show dietary plasticity and their trophic role should be reevaluated.

1.3 Rabbitfishes in the Seychelles

Within the Western Indian Ocean (WIO) region, rabbitfishes are considered to be a vital protein source that sustains local human populations (Grandcourt and Cesar, 2003; Kaunda-Arara et al., 2003). Often, rabbitfishes constitute a significant proportion (> 40% by weight) of inshore reef fishery catches in these areas (Kamukuru, 2009; Hicks and McClanahan, 2012), dominated mostly by the WIO endemic shoemaker spinefoot, Siganus sutor (Kamukuru, 2009; Samoilys et al., 2011; Hicks and McClanahan, 2012). In the small archipelagic nation of the Seychelles, rabbitfishes are the main target species (approx. 60 % by weight) of the artisanal fishery catch (Grandcourt and Cesar, 2003; Robinson et al., 2011), with S. sutor being the most sought-after species (Grandcourt and Cesar, 2003; Seychelles Fishing Authority, 2016). Other targeted rabbitfishes include the streamlined spinefoot, Siganus argenteus, the blue-spotted spinefoot, S. corallinus, and the brown-spotted spinefoot, S. stellatus. They are mainly caught using heart-shaped bamboo traps chiefly set from small (6 m length) outboard-powered vessels (Grandcourt and Cesar, 2003; Robinson et al., 2011). The local authorities have established minimum mesh size requirements (40 mm) for these traps, but these requirements are poorly enforced (Robinson et al., 2011). Therefore, juveniles are often captured and landed illegally (Grandcourt and Cesar, 2003). These species are widely considered to be important browsers and grazers (Green et al., 2009; Cheal et al., 2012; Fox and Bellwood, 2013), but empirical data on their ecological functions is lacking.

1.4 Current status of coral reefs in the Seychelles

Seychelles experienced a significant El Niño event in 1998, and, as a result, underwent mass coral bleaching, which destroyed > 90% of live coral cover around the inner islands (Graham et al., 2006, 2015). Subsequently, the cover of fleshy macroalgae on many reefs had increased significantly from < 1% in 1994 to > 40% in 2011 (Graham et al., 2015). The dramatic increase of fleshy macroalgae (particularly Sargassum sp.) caused the majority of reefs to phase shift from coral-dominated systems to be dominated by macroalgae (Graham et al., 2015).

Over the past few decades, the Seychelles has witnessed great economic restructuring through coastal reclamation and development projects (Clifton et al., 2012). Many of these activities have occurred within, or in close proximity to, the marine protected areas (MPAs) established around the inner islands of the Seychelles, and sedimentation through these activities has had significant deleterious impacts on the reefs (Ahamada et al., 2002, Clifton et al., 2012). Poaching is also widespread in the local artisanal fishery (Domingue et al., 2000, Clifton et al., 2012), with herbivorous fish species, such as rabbitfishes, parrotfishes and surgeonfishes, being the main targets (Seychelles Fishing Authority, 2016). Given the ecological significance of herbivory in coral reef ecosystems, a deeper understanding of the ecological functions of these commercial fishes would be beneficial for local management efforts.

1.5 Specific aims and thesis synopsis

Fishes that perform crucial ecological roles on coral reefs (e.g., herbivory), and which are also important targets of fisheries, should be a primary focus for research efforts and management actions (Bellwood et al., 2012; Edwards et al., 2014). The principal aim of this thesis was to determine the ecological roles of commercially important rabbitfishes, a family often understudied and overlooked within the global context. Their distributions around the inner islands of the Seychelles, in relation to parrotfishes and surgeonfishes, were investigated. This research determined which attributes of reefs potentially help drive these distributions (Chapter 2).

Secondly, the foraging behaviors of four rabbitfish species that dominate artisanal fisheries in the Seychelles were explored through the use of direct observations in the field, an approach that has been used widely to quantify the functional roles of herbivorous fishes on coral reefs. This thesis tested how their feeding ecologies changed spatially and temporally, with variation in the relative abundance of algal food resources amongst various reefs. The aim of this research was to define the ecological roles of harvested rabbitfishes on coral reefs in the Seychelles, and to determine whether, and how, these

species potentially contribute to complementarity in their functions on coral reefs (Chapter 3).

Finally, in Chapter 4, acoustic telemetry was used to study the spatial ecology of the shoemaker spinefoot, S. sutor. The spatial ecology of this species is poorly understood, yet it remains the main target species of the artisanal fisheries of many countries within the Western Indian Ocean region. The study was designed to investigate connectivity between different habitats by placing acoustic stations within dense coral reef patches and seagrass meadows. The identification of such links may help authorities incorporate networked habitats into management strategies aimed at conserving this heavily fished resource.

Collectively, the information gathered from each chapter was used to identify key ecological roles that are influenced by commercially significant species on tropical coral reefs. This research may not only act as a guide for the management of fishing pressure on vital rabbitfishes, but may also aid to promote the capacity of some reefs to resist, or recover from, disturbances that have led to algal dominance.

Chapter 2 – Reef features, not marine reserves, shape the distribution of herbivorous fishes on coral reefs in the Seychelles

Abstract

Herbivorous fishes perform important functions on coral reefs, and their feeding actions shape the capacity of reefs to both resist, and recover, from disturbance. Relatively little is known, however, about which ecological features of reefs influence the distribution of some of these fishes (e.g. rabbitfishes), or how their biomass is altered by human actions. We examined whether, and how, the benthic habitat characteristics of reefs (e.g. reef type, complexity, coral cover) and marine reserves combine to influence the biomass of herbivorous fishes. We focussed on different functional groups of herbivorous fishes, with a species level analysis of rabbitfishes, as this family comprises the bulk of the local artisanal fisheries on Seychelles reefs. The biomass of herbivorous fishes was shaped by the habitat characteristics of individual reefs, but not by the presence of marine reserves. Herbivore biomass was greatest in areas with high live coral cover, and also positively correlated with the cover of coral rubble and dead coral. Fish functional groups were, however, located in distinct habitats on local reefs: browsers were positively correlated with reef complexity, the cover of live coral and macroalgae; detritivores were positively correlated with dead coral cover; and scrapers were positively correlated with coral rubble. The biomass of locally exploited rabbitfishes was also influenced by the features of reef habitats, but not the conservation status of individual reefs. Siganus sutor and S. argenteus were strongly associated within areas of dense macroalgae, whereas S. corallinus was associated with areas where the cover of live coral was high and epilithic algal matrix cover was low. Our results show that existing marine reserves in the Seychelles do not appear to support increased fish biomass compared to non-protected areas and that other factors, such as benthic conditions of reefs seem to be driving the distribution of fish biomass. This suggests that this might lead to spatial separation in the distribution of different forms of herbivory, and possibly reef resilience, in the Seychelles.

2.1 Introduction

The consumption of algae by herbivorous fishes is a key ecological function on coral reefs (Hoey and Bellwood, 2009). Reefs that have healthy populations of herbivorous fishes, are able to limit the growth of algae, minimise coral-algal competition, and help to maintain reef health by reducing the potential for phase shifts from coral to algal dominance (Hughes et al., 2007; Mumby and Steneck, 2008). Phase shifts can also occur following major natural disturbance, such as mass coral bleaching, which first impacted reefs on the Inner Islands of Seychelles in 1998, and reduced the cover of live coral by > 90% (Graham et al., 2006, 2015). Since that time, the cover of macroalgae on some reefs has also increased significantly, from < 1% in 1994 to > 40% in 2011 (Graham et al., 2015). The effects of natural disturbance, including some mass coral bleaching events, can be reversed but this may take years or decades (Cheal et al., 2010; Gilmour et al., 2013; Roff et al., 2015), and the rate of recovery can be further delayed when other anthropogenic stressors (e.g. overfishing) are not managed (Dayton et al.,1995). These shifts from coral to algal dominance are also often associated with long-term changes to both the abundance and diversity of reef fishes, and the ecological function of herbivory that fishes perform in reef ecosystems (Mumby and Steneck, 2008).

There are three main families of herbivorous fishes on tropical Indo-Pacific coral reefs: parrotfishes (Labridae), surgeonfishes (Acanthuridae), and rabbitfishes (Siganidae) (Cheal et al., 2012; Hoey et al., 2013; Puk et al., 2016). The majority of research on herbivorous fishes within the Indo-Pacific have concentrated on the ecological roles of parrotfishes and surgeonfishes (e.g. Bellwood and Choat, 1990; Goatley and Bellwood, 2010; Bellwood et al., 2012; Marshell and Mumby, 2012; 2015). By contrast, the functional roles of rabbitfishes within the Indo-Pacific remains quite limited (for exceptions, see Bryan, 1975; Paul et al., 1990; Fox et al., 2009; Fox and Bellwood, 2013; Hoey et al., 2013; Brandl and Bellwood, 2014). It appears that rabbitfishes perform a role that is complementary to that of parrotfishes and surgeonfishes (e.g. Fox and Bellwood, 2013; Brandl and Bellwood, 2014). The majority of research on rabbitfishes has been restricted

to the Great Barrier Reef in eastern Australia, and little is known of their behaviour and ecology from reefs elsewhere.

Herbivorous fishes, particularly rabbitfishes, are a vital food source in many countries in, and surrounding the Western Indian Ocean (McClanahan and Mangi, 2004). In the Seychelles, for example, four rabbitfish species, namely; the streamlined spinefoot, Siganus argenteus, the blue-spotted spinefoot, S. corallinus, the brown-spotted spinefoot, S. stellatus and the shoemaker spinefoot, S. sutor make up a significant proportion of the artisanal trap fishery, constituting 60% of the average annual reported catch by weight (Seychelles Fishing Authority, 2016). The Republic of Seychelles is an archipelago of 115 granitic and carbonate islands spread over an exclusive economic zone of 1.37 million km² in the Western part of the Indian Ocean (Robinson et al., 2011), with a population of about 93,400 inhabitants which is mostly centralised on Mahé Island (National Bureau of statistics, 2016) (Fig. 1). The islands’ economy is driven mainly by tourism and fisheries, and over the past few decades, the Seychelles has witnessed great economic restructuring through coastal reclamation and development projects (Clifton et al., 2012). Many of these activities have occurred within, or in close proximity to, the marine protected areas (MPAs) established around the inner islands, and sedimentation through these activities has had significant deleterious impacts on the reefs (Ahamada et al., 2002, Clifton et al., 2012). Poaching is also widespread in the local artisanal fishery (Domingue et al., 2000, Clifton et al., 2012), with herbivorous fish species such as rabbitfishes, parrotfishes and surgeonfishes being the main targets (Seychelles Fishing Authority, 2016).

Fleshy macroalgae (particularly Sargassum sp.) around the inner islands has increased significantly over the past two decades, causing the majority of reefs systems to shift from coral domination to being overtaken by this macroalgae (Graham et al., 2015), and the excessive removal of these particular fishes may be detrimental. It is not clear, however, to what extent the heavy fishing pressure impacts herbivorous fishes in the Seychelles, or the ecological function of herbivory they perform.

Traditionally, parrotfishes, surgeonfishes and rabbitfishes were placed in the same functional group and labelled ‘algal removers’ (Choat, 1991). It has now become clear that herbivorous fishes consume different types of algae, often with distinct effects on coral-algal dynamics, and they are now frequently grouped based on their functional traits (Heenan and Williams, 2013). The broadest functional distinction is between fishes that graze predominantly on the epilithic algal matrix (EAM) (i.e. grazers) and those that browse typically on fleshy macroalgae (i.e. browsers) (Heenan and Williams, 2013). Grazers can be further separated into grazers/ detritivores, scrapers and excavators by the type, and amount, of the EAM and underlying substrate they remove during feeding (Green and Bellwood, 2009). Detritivores devour copious amounts of turf algae while brushing the EAM for detritus (Marshell and Mumby, 2012), whereas scrapers clean EAM off the reef matrix, and excavators remove EAM by taking bites off the reef matrix (Steneck, 1988; Bellwood and Choat, 1990). These actions may help promote coral larvae settlement by exposing the carbonate substrate. By contrast, browsers have the potential to reverse phase shifts once macroalgae become established (Hoey and Bellwood, 2009; Streit et al., 2016). However, the ecological roles of individual browsers remain poorly known (Streit et al., 2016).

It is clear that herbivorous fishes perform important ecological functions on coral reefs, but there is limited knowledge on the ecological attributes of reefs that influence the spatial distribution of these fishes (Tootell and Steele, 2016). Their distributions can be influenced by resource availability (Mumby et al., 2013), reef complexity (Mumby and Wabnitz, 2002), reef type (Graham et al., 2006), and fishing pressure (Strain et al., 2019). In areas where herbivorous fishes are harvested, MPAs have been established with the aim of conserving herbivore populations and the ecological functions these fishes perform (Topor et al., 2019). Positive effects of MPAs have been reported on herbivore biomass and, diversity. MPAs may play a role in restoring the biomass of important functional groups, and increasing fish catch through spill-over effects and egg subsidy (Gell and Roberts, 2003; Strain et al., 2019). Given the ecological significance of herbivory in coral reef ecosystems, it is imperative that we seek to better understand which ecological features of reefs influence the distributions of these important fishes, and the ecological

functions they perform (Cheal et al., 2012). This is particularly true for the Seychelles, where pressures from coral bleaching, and commercial and artisanal harvesting of herbivorous fishes are severe.

In this study, we examined how the benthic habitat features of reefs (e.g. reef type, complexity, coral cover) and MPAs influence the biomass of species of rabbitfishes, parrotfishes and surgeonfishes, sorted into key functional groups. We further tested these habitat characteristics and management status separately on rabbitfishes at species level. We hypothesized that the biomass of: 1) herbivorous fishes would be influenced by MPA status and reef type; 2) browsers would be correlated with the cover of macroalgae; and 3) grazers (i.e. detritivores, excavators and scrapers) would be correlated with the cover of epilithic algal matrix.

2.2 Materials and Methods

2.2.1 Study region and sampling design

We surveyed herbivorous fishes from 24 reefs across six different islands (i.e. Mahé, St. Anne, La Digue, Praslin, Curieuse and Cousin) of the inner Seychelles in May 2016 (Fig. 2.1). Our sampling design included two types of reef (i.e. carbonate and granitic), reefs that were protected in MPAs and reefs that were open to fishing. (Fig. 2.1, Table 2.1).

Figure 2.1 Map showing the 24 study sites across six different inner islands of the Seychelles. Highlighted on the map are the MPAs and reef type (carbonate or granitic).

Table 2.1 Marine Protected areas around the Inner Seychelles Islands.

Sources: (1) UNDP. Strengthening Seychelles’ protected area system through NGO management modalities; 2010. http://www.thegef.org/gef/node/3957 (2) Jennings et al., 1995.

Designation Year Site Ownership Management Marine Designated area (ha) Marine National 1979 Baie Ternaie (A2) Government SNPA 80 Park Marine National 1973 St. Anne (B1-B4) Government SNPA 1073 Park Marine National 1979 Praslin and Curieuse Government SNPA 1176 Park (D5, D6, E1-E4) Special Reserve 1968 Cousin (F1-F4) NGO NS 1200 Key: SNPA- Seychelles National Parks Authority: NS- Nature Seychelles

Specifically, forereef sites of 5-9 m in depth were investigated as this habitat type was available at all survey locations. This depth range is also known to sustain a high abundance and diversity of herbivorous reef fishes (Bellwood and Choat, 1990; Fox and Bellwood, 2007), including the focal species and functional groups in this study (Edwards et al., 2014). Furthermore, this is also the depth range over which the near-shore artisanal trap fishery operates (Seychelles Fishing Authority, unpublished data).

2.2.2 Assessing benthic habitat characteristics

To quantify variation in benthic community structure on each forereef site, benthic surveys using the Point Intercept Transect (PIT) method were carried out (sensu Hodgson et al., 2003). Specifically, a 10 m tape highlighted at 0.25 m intervals was laid out in a haphazard fashion, covering varying reef contours. A diver swam directly above the tape and identified and recorded the benthos at 0.25 m intervals along the transect line. Benthic habitat was identified and classified as epilithic algal matrix (EAM) (i.e. multispecies assemblage of miniscule algae, or predominantly filamentous algae with a canopy height

<1 cm (Steneck, 1988)) (Note: this classification of EAM was specific to alternative benthic material, other than those found growing on coral rubble or dead coral), fleshy macroalgae (i.e. large fleshy algae with canopy heights >1 cm (Steneck, 1988)), in this case dominated by Sargassum sp., live coral, sand, seagrass, coral rubble and dead coral. Dead coral was distinguished from coral rubble by the status of the coral structure. Each transect was laid out 10m apart in random directions along the forereefs. In total, 40 points were surveyed on each transect line. This technique was repeated five times, resulting in a total of 200 points for analysis on each reef, and covering an area of 50 m at each site (10 m transects replicated five times). Care was taken in not overlapping transects. The data were averaged, and presented as percentages.

Habitat complexity is a significant reef feature influencing the abundance patterns of many organisms on coral reefs (McCormick, 1994). We quantified habitat complexity as the rugosity of each reef, and measured this using the chain and tape method (Mumby and Wabnitz, 2002). Briefly, a 2 m chain (segment length 3 cm) was placed over the substrate surface, covering varying reef contours and the linear horizontal distance between the start and end of the chain was measured with a tape measure. Rugosity was calculated as the ratio of chain length (2 m) to horizontal distance. Each transect was laid out 10 m apart in random directions along the forereefs, in direct vicinity to the benthic transects. Twelve replicate measurements were made per studied reef, covering an area of 24 m at each site (2 m transects replicated 12 times). Care was taken in not overlapping transects. The data were later averaged for each site.

2.2.3 Surveying herbivorous fishes

Fish surveys across the twenty-four forereef sites were performed by the same observer (AE) using 7 m radius point counts by SCUBA, following similar methods as Ledlie et al., (2007), and Graham et al., (2015). Prior to each survey the visibility was estimated and ensured to always be more than 7 m. Furthermore, the 7 m radius was measured and re- calibrated at each site. Point counts were deemed most suitable since spearfishing is prohibited in the Seychelles, as is the feeding of fish during recreational dives, which

otherwise might cause fish to either flee from divers or aggregate around them, respectively, and skew survey results (Ledlie et al., 2007). All rabbitfishes, surgeonfishes and parrotfishes were identified to species level, and estimates of their fork lengths made to the closest ±1 cm. Prior to carrying out these surveys, baseline assessments of the observer’s ability to accurately assess fork length was performed. Fork length measurements were chosen, as it is the most commonly used measurement utilised in fisheries science for estimating weight from length. Accordingly, measurements were then converted to biomass using length-weight relationships from Froese and Pauly (2000). Biomass was standardized to unit measurements (g.mˉ²), and averaged for each site. Herbivorous fishes were then separated into groups based on their functional roles (i.e. browsers, detritivores, excavators, and scrapers of reef substratum) (Table 2.2). Point counts were repeated 8 times at each site, and timed to last 8 minutes each. Altogether an area of approximately 1232 m² (π×7²×8) was surveyed on each reef. This practice maximises area coverage and replication, yet allows for comprehensive searching for territorial species, providing a quantitative approximation of the number of fishes of differing behaviours and sizes (Graham et al., 2006).

Table 2.2 List of the herbivore species recorded in this study showing their functional behaviour, and total biomass averaged across the 24 study sites.

Family Functional behaviour Species Reference Avg Total Biomass (g.m-2) STD Dev Siganidae Algal browsers S. sutor Fox and Bellwood, 2013 4174.8 514.5

Grazers/ Detritivores S. corallinus Putra et al., 2018 151.0 141.9 S. argenteus Putra et al., 2018 6385.7 760.5 S. stellatus Green et al., 2009 227.1 21.0

Acanthuridae Algal browsers N. elegans Chong-Seng et al., 2014 452.7 65.2

Grazers/ Detritivores A. lineatus Green and Bellwood, 2009 317.6 62.8 A. nigrofuscus Green and Bellwood, 2009 1603.8 114.0 A. triostegus Green and Bellwood, 2009 58.2 7.4 A. leucosternon Roberts et al., 1987 359.6 39.8 A. leucocheilus Randall, 1956 3.6 0.7 A. tennenti Samoilys et al., 2018 1689.3 130.4 C. striatus Cheal et al., 2012 111.0 17.6 C. strigosus Cheal et al., 2012 36.5 6.4 C. truncatus Randall and Clements, 2001 10.8 1.6 Z. desjardinii Alwany, 2008 182.3 16.1 A. dussumieri Cheal et al., 2012 101.6 15.9 A. nigricauda Cheal et al., 2012 1064.1 108.1

Labridae Algal browsers C. carolinus Cheal et al., 2012 45.3 6.4 C. spinidens Cheal et al., 2012 149.9 13.8

Excavators C. bicolor Choat and Bellwood, 1990 376.4 37.7 C. artilunula Choat and Bellwood, 1990 285.1 24.6 C. gibbus Alwany, 2008 333.9 38.2 C. sordidus Cheal et al., 2012 2766.5 125.1 C. strongylocephalus Plass-Johnson et al., 2013 64.1 13.1

Scrapers H. harid Afeworki et al., 2013 1817.2 311.9 S. caudofasciatus Plass-Johnson et al., 2013 126.3 21.7 S. falcipinnis Plass-Johnson et al., 2013 168.1 30.7 S. frenatus Cheal et al., 2012 583.8 40.2 S. ghobban Cheal et al., 2012 1797.3 122.2 S. niger Cheal et al., 2012 89.9 9.1 S. prasiognathos Stockwell et al., 2009 227.7 32.7 S. psittacus Cheal et al., 2012 21.7 2.4 S. russelii Plass-Johnson et al., 2013 878.7 66.7 S. tricolor Plass-Johnson et al., 2013 251.0 17.8 S. rubroviolaceus Cheal et al., 2012 211.3 19.2 S. scaber Plass-Johnson et al., 2013 1356.9 82.4

2.2.4 Data analysis

We tested for effects of benthic habitat characteristics and MPAs on the biomass of: (i) the overall herbivorous fish community; (ii) each functional group of herbivorous fishes; (iii) the overall rabbitfish community; and (iv) each rabbitfish species. It must be noted that we were not testing MPA performance, but rather investigating whether they had a significantly greater biomass (g.m-2) of herbivorous fishes compared to fished areas.

Multivariate variations in mean biomass of herbivorous fishes was calculated by means of a three-way permutational analyses of variance (PERMANOVA) with the following factors: Island (6 levels; Mahé, St. Anne, La Digue, Praslin, Curieuse and Cousin islands; random), management status (2 levels; protected and unprotected; fixed) and reef type (2 levels; granitic and carbonate; fixed) (similar approach used in Cáceres et al., 2020). Bray-Curtis similarity matrices were used in multivariate community analyses (i.e. all fish together). This method was used because it takes species identity into account. By contrast, Euclidean similarity matrices were used in univariate analyses, which tested for effects on individual fish functional groups (i.e. algal browsers, grazers/detritivores, excavators, and scrapers), the overall rabbitfish community, and individual rabbitfish species (i.e. S. argenteus, S. corallinus, S. stellatus, and S. sutor). This method was used to simply investigate the distance measures between the different sites. All data were fourth-root transformed prior to analyses in order to limit the effects of numerically large values (e.g. highly abundant schooling species) (Clarke, 1993).

To test for correlations between benthic habitat characteristics and the biomass of herbivorous fish functional groups, data on reef rugosity and the cover of benthic habitat features were included in PERMANOVA designs as covariates. Benthic data were fourth- root transformed and normalised prior to analysis. Principal Coordinate Analyses (PCOs) were then carried out to assess and display the influence of environmental variables on different functional groups of herbivorous fishes, as well as on the different rabbitfish species. In order to determine the level of significance of each variable in the study, we

analysed the variance components from the PERMANOVA outputs and express them as a proportion of the total.

PERMDISP tests for homogeneity of multivariate dispersion between groups were used to test for variability of the tested factors (i.e. island, management status and reef type) (Anderson et al., 2008). Non-metric multidimensional scaling (nMDS) was used to display the dispersion of variability for any significant interactions discovered during PERMDISP tests.

All multivariate statistical analyses were performed using Primer-E v7 software (Clarke and Gorley, 2015) with the PERMANOVA+ add-on package (version 7.0.13) (Anderson et al., 2008).

2.3 Results

2.3.1 Effects on herbivore functional groups

This study recorded 36 species of herbivorous fishes from the three different families (Table 2.2). Fifteen were grazers/ detritivores, twelve were scrapers, five were excavators and four were algal browsers (Table 2.2). The spatial distribution of fish functional groups varied considerably among reefs: grazers/ detritivores and scrapers were most common during the surveys, being present on 83% and 100% respectively of reefs; whilst excavators were present on 75% of reefs; and algal browsers were present on only 21% of reefs (Fig. 2.4). The streamlined spinefoot, Siganus argenteus (a grazer/ detritivore), had the highest total mean biomass across the 24 sites (6,386 ± 760 g.m-2), and accounted for ~47% of all grazers/ detritivores in this study, whilst the pale-lipped surgeonfishes, Acanthurus leucocheilus (also a grazer/ detritivore), had the lowest total mean biomass across the sites (3.6 ± 0.7 g.m-2), accounting for <1% of grazers/ detritivores (Table 2.2).

The overall distribution of herbivorous fishes was shaped by benthic habitat characteristics (Figs. 2.2 and 2.3). Live coral cover was found to be the only influential variable tested (pseudoF = 3.20, Pperm = 0.04), accounting for ~23% of the total variation observed (Table 2.3 and Appendix B: Table S2.2). Coral rubble, dead coral, fleshy macroalgae, epilithic algae, sand, seagrass and rugosity had no detectable influence on the overall distribution (Pperms > 0.05) (Table 2.3 and Appendix B: Table S2.2). Also, the tested factors (island, management status and reef type), and the interaction between these factors had no significant influence on the overall biomass of herbivorous fishes (Pperms > 0.05) (Figure 2.4, Table 2.3 and Appendix B: Table S2.2). PERMDISP results of factors confirmed that dispersions between sites were non-significant (Pperms > 0.05) (Appendix B: Table S2.4).

Coral rubble 100 Dead coral Epilithic algal matrix

80 Fleshy Macroalgae Live coral Sand 60 Seagrass

% % Cover 40

F1 F2 F3 F4

E1 E2 E3 E4

B2 A1 A2 A3 A4 B1 B3 B4 C1 C2 D1 D2 D3 D4 D5 D6 Sites

Figure 2.2 Stacked bar chart showing the percentage cover of each benthic material at each reef surveyed.

4 3.5 3 2.5 2 1.5

Rugosity Rugosity Ratio 1 0.5 0 A1 A2 A3 A4 B1 B2 B3 B4 C1 C2 D1 D2 D3 D4 D5 D6 E1 E2 E3 E4 F1 F2 F3 F4 Sites

Figure 2.3 Bar chart showing the average rugosity ratio (±SE) at each surveyed site.

Figure 2.4 Stacked Bar charts showing the total mean biomass of each functional group at; a. each island (Mahé vs. St. Anne vs. La Digue vs. Praslin vs. Curieuse vs. Cousin), b. by management status (unprotected vs. protected), and c. by reef type (granitic vs. carbonate).

Source df Overall Algal browsers Grazers/Detritivores Excavators Scrapers pseudo-F P(perm) pseudo-F P(perm) pseudo-F P(perm) pseudo-F P(perm) pseudo-F P(perm) Coral rubble 1 2.57 0.10 0.01 0.10 0.08 0.81 2.88 0.11 4.97 0.04 Dead coral 1 1.83 0.17 2.55 0.16 1.28 0.36 2.97 0.17 0.39 0.52 EAM 1 0.59 0.77 0.34 0.81 0.51 0.64 0.77 0.52 1.56 0.29 FMA 1 1.23 0.35 6.24 0.04 0.82 0.46 0.19 0.83 0.10 0.78 Live coral 1 3.20 0.04 1.60 0.21 3.56 0.10 4.00 0.10 0.26 0.61 R ratio 1 2.69 0.08 2.43 0.16 0.30 0.60 2.84 0.12 0.02 0.91 Sand 1 0.80 0.52 0.29 0.61 0.01 0.99 0.47 0.52 0.14 0.71 Seagrass 1 2.05 0.15 5.09 0.06 0.16 0.68 0.07 0.80 1.43 0.71 Island 5 1.15 0.41 1.21 0.48 2.57 0.19 2.57 0.19 0.77 0.62 Mngt 1 0.78 0.53 0.72 0.42 3.67 0.10 0.29 0.82 0.92 0.71 R type 1 1.40 0.24 1.28 0.27 0.41 0.43 2.35 0.21 0.47 0.57 Island × 1 0.61 0.73 0.23 0.69 0.15 0.73 1.82 0.26 0.58 0.49 Mngt Island × R type 3 0.50 0.85 0.38 0.15 0.23 0.86 0.43 0.72 1.85 0.29

Key: EAM- Epilithic algal matrix: FMA- Fleshy macroalgae: R ratio- Rugosity ratio: Mngt- Management: R type: Reef type: df- Degrees of Freedom Note: the interaction terms of Mngt × R type, and Island × Mngt × R type were not reported as there was not enough replication.

When herbivores were tested at functional level, we found that different functional groups were associated with different benthic habitats on local reefs. The biomass of algal browsers was associated with fleshy macroalgae (pseudoF = 6.24, Pperm = 0.04) accounting for ~29%, of the total variation observed (Table 2.3 and Appendix B: Table S2.2). The biomass of scrapers was positively correlated with coral rubble (pseudoF = 4.97, Pperm = 0.04), accounting for ~22% of the total variation observed (Appendix B: Table S2.2). By contrast, the biomass of grazers/ detritivores and excavators were not correlated with the variation of any benthic habitat features tested (Pperms > 0.05) (Table 2.3 and Appendix B: Table S2.2). These patterns were confirmed through observation of the PCO plot, which displays the similarity of each herbivorous fishes’ biomass (grouped at functional level) distribution to the benthic habitat variables (displayed as vectors in 2D space) (Fig. 2.5). The tested factors (island, management status and reef type), and the interaction between these factors had no significant influence on any functional group of herbivorous fish (Pperms > 0.05) (Figure 2.4, Table 2.3 and Appendix B: Table S2.2). PERMDISP results of factors confirmed that dispersions between sites were non- significant (Pperms > 0.05) (Appendix B: Table S2.4).

Figure 2.5 Two-dimensional Principal Coordinate Analysis (PCO) illustrating the similarity of herbivorous fish biomass (g.m-2) (segmented for each functional group), and benthic habitat characteristics (presented as vectors). Numerical values in the text box demonstrate biomass range (g.m-2) of each functional group.

2.3.2 Effects on rabbitfishes (Siganidae)

This study recorded four species of rabbitfishes that are harvested from reefs in the Seychelles, including: Siganus argenteus, S. corallinus, S. stellatus and S. sutor. These four species had a total mean biomass of 12,297.6 (± 1437.9) making up ~41% of the total mean biomass estimates of all fish recorded in this study (Table 2.2). S. argenteus (an algal browser) had the highest total mean biomass of 6,386 (± 760g.m-2), accounting for ~53% of the total mean biomass estimates of all rabbitfish species, whilst S. sutor (an algal browser) had a total mean biomass of 4,174 (±514.5 g.m-2), accounting for 35% of the total mean biomass estimates for rabbitfishes, S. corallinus (an algal browser) had a 27

total mean biomass of 1,510.0 (±141.9 g.m-2), accounting for 12% of the total mean biomass, and S. stellatus (a detritivore), had the lowest total mean biomass of 227.1 (±21.0 g.m-2) recorded, accounting for <1% of the total mean biomass of all rabbitfish species (Table 2.2).

The overall distribution of rabbitfishes were influenced by certain benthic habitat characteristics (Figs. 2.2 and 2.3), some more than others, with live coral cover being the most influential (pseudoF = 2.72, Pperm = 0.04), accounting for ~12% of the total variation observed (Table 2.4 and Appendix B: Table S2.3). Other influential variables included epilithic algal matrix (pseudoF = 3.11, Pperm = 0.03), accounting for ~8% of the total variation observed and fleshy macroalgae (pseudoF = 3.39, Pperm = 0.04), accounting for ~7% of the total variation observed (Table 2.4 and Appendix B: Table S2.3). The other benthic habitat characteristics tested (i.e. coral rubble, dead coral, sand, seagrass and rugosity) had no significant influence on the overall rabbitfish biomass distribution (Pperms > 0.05) (Table 2.4 and Appendix B: Table S2.3). Also, the tested factors (island, management status and reef type) had no significant influence on the overall rabbitfish distribution (P perms >0.05) (Table 2.4 and Appendix B: Table S2.3). Although, PERMDISP results showed differences in the dispersions of the overall rabbitfish distribution between reef types (carbonate vs. granitic) (F = 18.10, Pperm = 0.01) (Appendix B: Table S2.5) from the nMDS plot of the variability (Appendix B: Fig. S2.1), it was apparent that much of this variability was occurring on carbonate reefs.

Source df Overall S. argenteus S. corallinus S. stellatus S. sutor pseudo-F P(perm) pseudo-F P(perm) pseudo-F P(perm) pseudo-F P(perm) pseudo-F P(perm) Coral rubble 1 0.40 0.71 0.42 0.53 0.03 0.86 0.20 0.65 0.49 0.50 Dead coral 1 0.45 0.69 0.03 0.92 1.44 0.28 0.07 0.85 0.51 0.51 EAM 1 3.11 0.03 0.23 0.66 4.81 0.05 0.05 0.81 0.11 0.78 FMA 1 3.39 0.04 4.92 0.04 0.16 0.68 0.80 0.36 19.9 0.001 Live coral 1 2.72 0.04 2.20 0.16 4.16 0.05 3.16 0.09 1.74 0.18 R ratio 1 1.51 0.24 1.61 0.25 0.09 0.77 0.10 0.80 0.27 0.64 Sand 1 1.77 0.17 3.28 0.14 0.55 0.50 0.43 0.53 2.28 0.17 Seagrass 1 2.10 0.11 2.50 0.14 0.15 0.70 1.06 0.31 8.42 0.01 Island 5 1.60 0.23 1.40 0.36 2.06 0.25 3.52 0.10 1.12 0.44 Mngt 1 2.27 0.21 4.48 0.06 0.93 0.77 0.52 0.86 2.21 0.15 R type 1 1.75 0.28 1.03 0.37 2.96 0.14 0.53 0.53 0.78 0.49 Island × 1 0.30 0.82 0.28 0.61 0.62 0.47 1.27 0.31 0.07 0.80 Mngt Island × R type 3 0.95 0.51 1.75 0.30 1.26 0.40 2.16 0.26 1.32 0.37 Key: EAM- Epilithic algal matrix: FMA- Fleshy macroalgae: R ratio- Rugosity ratio: Mngt- Management: R type: Reef type: df- Degrees of Freedom Note: the interaction terms of Mngt × R type, and Island × Mngt × R type were not reported as there was not enough replication.

When rabbitfishes were tested at species level, we found that different species were associated with different benthic habitats on local reefs. The biomass of S. argenteus was positively correlated with fleshy macroalgae (pseudoF = 4.92, Pperm = 0.04), accounting for ~20% of the total variation observed (Appendix B: Table S2.3). The biomass of S. corallinus was correlated with live coral cover (pseudoF = 4.16, Pperm = 0.05), accounting for ~24% of the total variation observed, and epilithic algal matrix (pseudoF = 4.81, Pperm = 0.05), accounting for ~14% of the total variation observed (Appendix B: Table S2.3). The biomass of S. sutor was primarily associated with fleshy macroalgae (pseudoF = 19.9, Pperm = 0.001), accounting for ~27% of the total variation observed, and seagrass cover (pseudoF = 8.42, Pperm = 0.01), accounting for ~19% of the total variation observed (Table 2.4 and Appendix B: Table S2.3). By contrast, the biomass of S. stellatus was not correlated with variation in any of the benthic habitat features we measured (Pperms >0.05), (Table 2.4 and Appendix B: Table S2.3). These patterns were confirmed through the PCA plot which displays the similarity of each rabbitfishes’ biomass distribution, to the benthic habitat variables (displayed as vectors in 2D space) (Fig. 2.6). The tested factors (island, management status and reef type) had no significant influence on any species of rabbitfishes (Pperms >0.05) (Table 2.4). However, PERMDISP results showed differences in dispersions of S. argenteus (F = 14.36, Pperm = 0.01) and S. sutor (F = 9.74, Pperm = 0.02) (Appendix B: Table S2.5). The nMDS plot of variability (Appendix B: Fig. S2.1) highlighted that much of the variability was occurring on carbonate reefs.

Figure 2.6 Two-dimensional Principal Coordinate Analysis (PCO) illustrating the similarity of rabbitfishes biomass (g.m-2) (segmented for each species), and benthic habitat characteristics (presented as vectors). Numerical values in the text box demonstrate biomass range (g.m-2) of each species.

2.4 Discussion

Identifying processes that regulate the health of coral reefs is key to understanding how management activities might defend them from anthropogenic threats (Strain et al., 2019). Herbivorous fishes may contribute to coral reef resilience (Hoey and Bellwood, 2009); however, the majority of studies have primarily investigated the top-down effects. We examined potential bottom-up effects by identifying whether, and how, the benthic

habitat characteristics of reefs (e.g. reef type, complexity, coral cover) and marine reserves combine to influence the biomass of key functional groups of herbivorous fishes on Seychelles reefs. We also tested specifically for effects of reef features and reserves on four rabbitfish (Siganidae) species, which are the primary target in local artisanal fisheries.

2.4.1 Effects on herbivore functional groups

Grazers/ detritivores and scrapers accounted for the majority of herbivorous fishes, being observed on most reefs. This suggests that they are both abundant and live in a greater range of habitats around the Seychelles inner islands. A very small proportion of the population were algal browsers, suggesting that this functional group of herbivores are either few or restricted to specific habitats.

The variation of distribution of each species may reflect factors that affect their survival (e.g. dietary preferences, food availability or predation risk). In our case, a large proportion of this variation was associated with environmental variables (i.e. the benthic habitat substrates and reef rugosity, Appendix B: Table S2.2) which accounted for over 53% of total variation observed during this study. In particular, the cover of live coral significantly influenced the overall biomass of these fishes (Appendix B: Table S2.2)., though when analysing each functional group separately, live coral was not a significant influence. This demonstrates that overall, herbivorous fish communities’ benefit from the framework provided by certain coral morphologies that provide a refuge from predation (reviewed in Graham and Nash, 2013; Richardson et al., 2017). However, this may only be essential to certain individual species.

Although live coral appeared to control the overall distribution of herbivorous fishes, other species, particularly grazers (i.e. detritivores, excavators and scrapers), benefit from the increase of algae following the loss of live coral tissue (e.g. Adam et al., 2011; Gilmour et al., 2013). These grazers primarily feed on epilithic algae over dead coral substrata and rubble. Our results show scrapers to be positively influenced by coral rubble, probably

because of the epilithic algae emerging. Their feeding behaviours can promote settlement of new coral larvae by increasing exposure of the carbonate substrate (Steneck, 1988; Bellwood and Choat, 1990). These fishes may play an important role to the recovery of Seychelles reefs.

Our results support our hypothesis that browsers would be correlated with the cover of macroalgae, suggesting that these fishes may be distributed as predicted by foraging models, i.e. the optimal foraging theory (Schoener, 1971; Charnov, 1976) and the ideal free distribution (Fretwell and Lucas, 1970), whereby individual foragers take full advantage of their energy gain and distribute themselves proportionally to resources. This was also confirmed through personal observations in the field, particularly for S. sutor that was recorded feeding freely and frequently on this food source (Ebrahim et al., 2020). . To our surprise we also observed S. argenteus (typically classed as a grazer/ detritivore) also foraging on fleshy macroalgae. This may indicate that S. argenteus may only feed on fleshy macroalgae when it dominates the seascape (i.e. after regime shifts), as observed in the Seychelles (Ebrahim et al., 2020).

The overall biomass of herbivorous fishes was not influenced by the presence of marine reserves, reef type or island location. This result disproves our priori hypothesis that herbivorous fishes would be influenced by MPA status and reef type. MPAs are established to increase fish abundance by reducing or eliminating fishing mortality (Halpern, 2003). Herbivorous fish biomass and herbivory have been found to be higher on reefs that are protected within marine reserves (Russ and Alcala, 1996; Mumby and Steneck, 2008; Strain et al., 2019). However, our results indicate that the management initiatives applied to the areas we studied, most of which have been in place for over 40 years, are either not effective or are being circumvented. Over the past few decades, Seychelles has witnessed great economic restructuring through coastal reclamation and development projects lacking adequate or effective impact assessment procedures (Clifton et al., 2012). Many of these activities have occurred within, or in close proximity to, the MPAs established around the inner islands of Seychelles, and sedimentation through these activities has had significant deleterious impacts on the reefs (Ahamada et

al., 2002; Clifton et al., 2012), resulting in inadequate function. Poaching is another serious strain on MPAs in Seychelles (Domingue et al., 2000; Clifton et al., 2012; personal observations). The methods used (i.e. fish traps) directly target crucial herbivorous fish species, such as rabbitfishes, parrotfishes and surgeonfishes (Seychelles Fishing Authority, 2016).

We should also consider the fact that many reefs within MPAs around the inner islands have phase shifted to fleshy macroalgal (FMA) dominance (Graham et al., 2015), and chemical cues may deter coral and fish larvae settlement on these reefs (Dixson et al., 2014). The effects of the 1998 El Niño caused major phase shifts in the Seychelles, with some reefs showing over a 40% increase in FMA cover (Graham et al., 2015), and our results revealed higher percentage covers in some MPA regions (up to 60%). Also, resident herbivorous fishes can become overwhelmed by sudden increases in macroalgal growth, reducing their ability to crop it down (Williams et al., 2001). Therefore, although herbivorous fish can be effective in preventing phase shifts, natural disturbance such as the 1998 El Niño may have created a situation where accelerated macroalgal growth had reduced the impact of these herbivores present. Once reefs have shifted, reversing these shifts may prove to be an even greater task (e.g. Mumby et al., 2007). The long-term recovery and resilience of reefs may be more intricate than just increasing herbivore biomass, but may also hinge on the extent and length of the phase shift, the distinctiveness of particular species, and their function in space and time (Puk et al., 2016).

Reef type not being a significant component from our findings contradicts previous work done around the Seychelles Inner Islands by Graham et al., (2006). They reported that herbivore abundance was significantly greater on granitic reefs than on carbonate reefs, expressing concern that the recovery of carbonate reefs would thus be impeded (Graham et al., 2006). The reason for the disparity between the two studies, and whether it represents a real shift in biomass or reflects differences in methodologies, is beyond the scope of this study.

2.4.2 Effects on rabbitfishes (Siganidae)

The distribution of rabbitfishes, which make up the bulk of the artisanal fishery in Seychelles, was most influenced by live coral cover. Epilithic algae and fleshy macroalgae were also significant in influencing the overall biomass distribution, but to a lesser extent. However, analysed at species level, of the four species, live coral only positively influenced the distribution of S. corallinus, indicating that they may be using the framework provided by coral for evading predation (Graham and Nash, 2013). Fleshy macroalgal cover influenced the distribution of S. argenteus and S. sutor.

The distribution of rabbitfishes was not influenced by management status, reef type or island location. Although, we did find significant variability in the biomass for S. argenteus and S. sutor on carbonate reefs when testing reef type, indicating that these species were found in different abundances across the various carbonate reefs; perhaps due to varying abundance of macroalgae present. The fact that management initiatives are not promoting the biomass of the most targeted fish species in the artisanal fishery is concerning. If the fishery is to be sustained, certain measures such as effective enforcement will need to be taken up by management.

S. stellatus was not influenced by any of the variables or factors in this study probably due to their very low abundance during our surveys. Rabbitfishes are not a uniform group of herbivorous fishes on coral reefs as previously reported (e.g. Russ, 1984; Green and Bellwood, 2009; Hoey and Bellwood, 2010), rather there are two guilds in the Seychelles; S. sutor and S. argenteus are possibly feeding primarily on fleshy macroalgae on macroalgal-dominated reefs and, S. corallinus possibly feeding on epilithic algae on coral- dominated reefs.

Chapter 3 – Differences in diet and foraging behaviour of commercially important rabbitfish species on coral reefs in the Indian Ocean

Abstract

Herbivorous fishes consume algae on coral reefs and this ecological function is pivotal in helping reefs to resist and recover from disturbance. Although numerous studies have differentiated between those fishes that graze on low-profile algae versus those that browse on larger fleshy macroalgae, little is known about the feeding behaviours of some herbivorous fishes (e.g. rabbitfishes, Siganidae), limiting our understanding of whether, and how, these species contribute to ecological functions on coral reefs. Here, we examine how the feeding ecology of four species of rabbitfishes that dominate the artisanal fishery in the Seychelles changed spatially and temporally. Siganus argenteus and S. sutor were generalist herbivores feeding on a range of substrata (namely turf algae, macroalgae, seagrass and epiphytic algae), whereas S. corallinus and S. stellatus were specialist herbivores feeding primarily on substrata covered in turf algae. Bite rates of S. argenteus and S. sutor were positively correlated with the cover of macroalgae, seagrass and epiphytic algae. By contrast, bite rates of S. corallinus and S. stellatus were not correlated with changes in the cover of turf algae. These findings illustrate possible differences in the ecological contributions among rabbitfish species on coral reefs, and emphasize the need for caution when assigning species to functional groups and assuming within-group functional equivalence. The results also support the classic niche theory that species within a community must use resources differently in order to coexist over evolutionary timescales. These results further provide valuable insights for the management of rabbitfishes in tropical fisheries because it implies that the conservation of different species might result in distinct shifts in the competitive dominance of coral and algae.

3.1 Introduction

Fishes that perform crucial ecological roles on coral reefs (e.g. herbivory), and that are also important targets of fisheries, should be a primary focus for research effort and management action (Bellwood et al., 2012; Mumby, 2014). Coral reefs with relatively intact fish assemblages experience intense herbivory from the feeding activities of fishes that consume algal material (Green and Bellwood, 2009; Fong et al., 2018). The foraging actions of herbivorous fishes help to maintain a low biomass of algae on reefs (Hughes et al., 2007; Rasher and Hay, 2010), and can prevent the competitive dominance of canopy-forming macroalgae (e.g. Hoey and Bellwood, 2009; Green and Bellwood, 2009; Bonaldo and Hay, 2014). If, however, large and/ or dense beds of canopy-forming macroalgae become established they may be difficult to remove as these algae are unpalatable to many species of herbivorous fish (Bellwood et al., 2006; Ledlie et al., 2007; Hoey and Bellwood, 2011).

Herbivorous fishes are commonly placed in functional groups in an attempt to understand and examine ecological functions on coral reefs (Brandl et al., 2019). The broadest distinction in the herbivore functional groups is between fishes that graze predominantly on algal turfs and those that browse on fleshy macroalgae (Heenan and Williams, 2013). ‘Grazers’ may prevent macroalgae from becoming established by feeding on diminutive macroalgae and turf algae (Paddack et al., 2006; Hughes et al., 2007). ‘Browsers’ have the potential to reverse macroalgal phase shifts as they can reduce the overgrowth and shading of coral by selectively feeding on mature macroalgae (Hoey and Bellwood, 2009; Green and Bellwood, 2009). However, recent findings suggest that the relationship between nominally herbivorous fishes and benthic algae is not a simple relationship, but influenced by many differences of consumer and producer ecological traits, as well as environmental factors (Brandl and Bellwood, 2016). Thus, more detailed work on the relationship between herbivorous fishes and the benthic community is required in order to develop a deeper understanding of consumer-producer dynamics on tropical coral reefs (Adam et al., 2015).

The diet and foraging behaviours of many herbivorous fishes on coral reefs are characterized by high variability, both among and between habitat features on reefs, with differences often relating to changes in both the available nutritional content and chemical defenses of algal resources (e.g. Fox et al., 2009; Bruggemann et al., 1994; Hanmer et al., 2017). Coral reefs support a high diversity of algal species, which vary in their nutritional value to herbivores, and it is widely accepted that herbivores regulate their foraging efforts to optimise nutritional benefits (Zemke-White et al., 2002; Simpson et al., 2004; Dromard et al., 2015). To better appreciate the ecological roles that herbivorous fishes perform on reefs, we require empirical data to explain why they select certain food items, and to describe how preferences vary in time and space (Suding et al., 2004; Miller et al., 2011). Identification of such behaviour is a necessary step towards understanding feeding preferences in herbivores, and how different species modify their feeding actions and help reefs to either resist, or recover from, disturbances that would otherwise lead to macroalgae overgrowth (Bellwood et al., 2006; Adam et al., 2015; Loffler et al., 2015; Johnson et al., 2017).

We know surprisingly little about the feeding behaviours of many herbivorous fish species and this limits our understanding of the ecological roles they perform on reefs (Fox and Bellwood, 2013; Hoey et al., 2013; Yabsley et al., 2016). Rabbitfishes are a family comprising of 28 species, characterized by their morphology and ecology; dull colored, fusiform species that typically occur in schools within seagrass and or macroalgal habitats (e.g. the streamlined spinefoot, Siganus. argenteus; and the shoemaker spinefoot, S. sutor), and brightly colored, deep-bodied reef-associated species that typically occur in in pairs (e.g. the blue-spotted spinefoot, S. corallinus; and the brown-spotted spinefoot, S. stellatus) (Woodland, 1990; Borsa et al., 2007). They are considered to be important browsers and grazers on tropical reefs, and feed on a diverse assortment of algae within these seascapes (Cvitanovic and Bellwood, 2009; Hoey et al., 2013; Brandl and Bellwood, 2015). This family has recently gained attention, particularly on the Great Barrier Reef (GBR), for not only their functional role but also their ability to coexist with closely related species. Fox and Bellwood, (2013) demonstrated that rabbitfishes had very different feeding microhabitats from other herbivorous fish species, such as

parrotfishes and surgeonfishes. Moreover, Hoey et al., (2013) showed that 11 rabbitfish species on the GBR exhibited species-specific variations in diet composition, and Fox et al., (2009) demonstrated that two closely related rabbitfishes had clear differences in diet composition and feeding periods. Thus, it appears that feeding niche partitioning is an important component of coexistence among herbivorous species, even within families.

Rabbitfishes are vital food sources for humans in many tropical coastal regions, and are subjected to heavy fishing pressures in the Indo-West Pacific coastal regions of the world (Woodland, 1990; Kuiter, 1993; Kaunda-Arara and Rose, 2004; McClanahan and Mangi, 2004). The harvesting of rabbitfishes is particularly intense in the Republic of Seychelles, where four species, including S. argenteus, S. corallinus, S. stellatus, and S. sutor constitute over half (approx. 60 %) of the annual artisanal fishery catch (Grandcourt and Cesar, 2003; Robinson et al., 2011). Unfortunately, empirical data on their feeding behaviours is lacking for many of these species (e.g. S. sutor), and for most coral reefs outside of the GBR, (but see Chong-Seng et al., 2014). This information is necessary to better understand how rabbitfishes, which dominate artisanal fisheries in the Seychelles, possibly contribute to the function of herbivory (i.e. either graze and/or browse on algal material) on coral reefs.

The majority of the reefs around the inner islands have undergone macroalgal (predominantly Sargassum sp.) regime shifts following the major bleaching event of 1998 (Graham et al., 2015), resulting in increased herbivore productivity that has sustained the local artisanal reef fishery (Robinson et al., 2019). With recent habitat restructuring, the region provides a unique opportunity to study foraging behaviours of the family Siganidae, as a contrast to the well-studied GBR herbivores where coral cover is generally higher, macroalgal habitat minimal (except on inshore reefs (Wismer et al., 2009)), and species are not crucial fishery targets (Fox and Bellwood, 2008). We tested how the feeding ecology of rabbitfishes changed spatially and temporally, with variation in the relative abundance of algal food resources among reefs. This was done by focusing on the feeding rates and types of substrata targeted on each reef using direct bite rate observations in the field, an approach that has been adopted widely to quantify the

functional roles of herbivorous fishes on coral reefs (e.g. Burkepile and Hay, 2008; Cardoso et al., 2009; Fox and Bellwood, 2013; Hanmer et al., 2017). The overall aim was to describe the ecological roles of harvested rabbitfishes on coral reefs in the Seychelles, and to determine whether, and how, these species display differences in their diet and foraging behaviours. Indeed, classic niche theory suggests that species within a community must use resources differently in order to coexist over evolutionary timescales (Chesson et al., 2001), therefore we hypothesized that the four species would show variations in their diet and foraging behaviours by targeting different microhabitats on reefs in the Seychelles.

3.2 Materials and Methods

3.2.1 Study region and sampling design

We surveyed the composition of benthic assemblages, and quantified the foraging behavior of harvested rabbitfishes, on 16 coral reefs, which provided a strong gradient in the cover and composition of algal communities across seven of the inner islands of the Seychelles (Fig. 3.1, Appendix C: Fig. S3.1). The Republic of Seychelles is an archipelago of 115 granitic and carbonate islands spread over an exclusive economic zone of 1.37 million km² in the Western part of the Indian Ocean (Robinson et al., 2011). The islands’ economy is driven mainly by tourism and fisheries, whereby rabbitfish dominate the local artisanal fishery catch (Grandcourt et al., 2003; Robinson et al., 2011).

Figure 3.1 Map showing the 16 study sites across seven different inner islands of the Seychelles. Also highlighted on the map are the marine protected areas.

All data were collected from shallow (<10 m) reef slopes using SCUBA. This is the primary operating depth for the near-shore artisanal trap fishery that targets rabbitfishes, as it supports the four focal species in high abundance (Seychelles Fishing Authority, 2016).

3.2.2 Surveying the composition of coral and algal assemblages

To quantify variation in the composition of coral and algal assemblages among reefs, benthic surveys were performed using the Point Intercept Transect (PIT) method (sensu 41

Hodgson et al., 2003). At each reef, coral and algal assemblages were surveyed at 40 points along a 10m tape. A diver swam directly above the tape and identified and recorded the benthos at 0.25m intervals along the transect line. The survey was replicated five times at each site, yielding a total of 200 points for analysis from each reef. Transect tapes were laid in a haphazard fashion at each site, approximately 10 m apart without any overlap. Benthic life forms were identified and classified as fleshy macroalgae (i.e. Sargassum sp.) (FMA), turf algae (TA), live coral (LC), dead coral (DC), coral rubble (CR), sand (S), and seagrass (SG).

3.2.3 Quantifying the foraging behaviours of herbivorous rabbitfishes

Foraging behaviours of the four focal rabbitfish species were recorded using timed foraging observations (e.g. Cardoso et al., 2009; Fox and Bellwood, 2013; Hanmer et al., 2017), which were conducted between January 2015 and November 2016. During each observation, an individual fish was followed for a period of 8-10 minutes, at a distance of 2-3 m, and data were recorded to describe the species and size (fork length) of each fish, and the number of bites taken on each type of benthic substrata (i.e. turf algae, fleshy macroalgal fronds, fleshy macroalgal thallus, seagrass, or epiphytic algae). Feeding on epiphytes, as distinct from feeding on either the macroalgal fronds or seagrass blades that supported epiphytes, was distinguished by the rapid bite rates of fishes (Fox and Bellwood, 2008; Hoey and Bellwood, 2009), as well as the absence or very little macroalgal and seagrass material removed. To test whether foraging behaviours varied throughout the day, divers followed individuals of each species in six distinct time periods: early morning (06:00 to 08:00 h); mid-morning (08:00 to 10:00 h); late morning (10:00 to 12:00 h); early afternoon (12:00 to 14:00 h); mid-afternoon (14:00 to 16:00 h); late afternoon (16:00 to 18:00 h). If fish behavior appeared to be affected by the presence of divers, observations were stopped and data for that fish were excluded from analyses. In total, 480 foraging observations were completed to describe the feeding behaviours of S. sutor (n = 142), S. argenteus (n = 140), S. corallinus (n = 110) and S. stellatus (n = 88). Data from behavioral observations were summarized as the foraging rate (bites.min–1) of each rabbitfish species on different food items and times of day at each reef.

3.2.4 Data analysis

Foraging rates and behaviours

Bite rate data (bites.min-¹) for each individual fish, on each algal resource (i.e. fleshy macroalgae, turf algae, and seagrass) were transformed using log (x + 1) to normalise the data (Anderson et al., 2008). Bray Curtis similarity (BCS) distances (BCS = 1 − Bray Curtis dissimilarity distance) were then calculated for transformed data (Clarke and Gorley, 2006). A Principal Coordinate Analysis (PCO) plot was then used to visualize the BCS values among the different rabbitfish species’ bite data on different feeding substrata. We then averaged the PCOs per species, per site using the first two axes of the eigenvalues, as they accounted for over 90% of the total variation observed, and plotted the centroids in order to visualize the variation among species, among sites. Standard error bars were also displayed to show the dispersion from the mean.

To understand whether foraging behaviors differed among species, sites and at various times of the day, we used a three-way permutation-based multivariate analysis of variance (PERMANOVA; Anderson et al., 2008), in which fish species, site and time period were the fixed factors, and pseudo-F was calculated using 999 restricted permutations of data. Post-hoc pairwise comparisons were used to test for significant differences in bite rates between species, time periods and the reef sites. All multivariate statistical analyses were performed using Primer-E v7 software (Clarke and Gorley, 2015) with the PERMANOVA+ add-on package (version 7.0.13) (Anderson et al., 2008). Bar charts with standard error bars were used to display the foraging behaviours of each species, at each time period, at each site.

To display the diurnal patterns of feeding of each species, we used simple scatter plots of the average foraging rates across all sites. The foraging rates (bites.min-¹) of each species on each substratum, was averaged across the sites and displayed using a stacked bar chart with the corresponding error bars. In order to determine how the rates

differed between sites, simple regressions were plotted for each species at each site, with the abundance of the target resource as the independent variable, and the foraging rate of each rabbitfish as the dependent variable.

Foraging selectivity

We examined foraging selectivity using Vanderploeg and Scavia’s Relativized Electivity Index (Vanderploeg and Scavia, 1979; Lechowicz, 1982). This index is calculated by first finding the selectivity coefficient for foraged item i, Wi:

where ri is the proportion of bites taken in each category i and pi is the proportional cover of each category i. The index Wi ranges from 0 (total avoidance) to 1 (total preference). The relativized index is then

where n represents the number of foraged categories available at each site. The values of Ei range from −1 (total avoidance) to 1 (total preference).

Only the algal resources (i.e. fleshy macroalgae fronds (FMAF) and thallus (FMAT); seagrass epiphytes (SGE) and blades (SGB); and turf algae (TA) were used in the calculations of the electivity indices.

The calculated Electivity indices of each algal resource were averaged across sites for each species, and are presented as bar graphs.

3.3 Results

3.3.1 Foraging rates and behaviours

The overall foraging behaviour of each rabbitfish species was significantly different among species, sites, and the time of day (PERMANOVA Pseudo-F = 3.7, Pperm = 0.001; Table 3.1; Appendix C: Figs. S3.2, S3.3, S3.4 and S3.5). When analyzing the pairwise tests between sites, for each species at the different time intervals, S. argenteus and S. sutor showed significant values for the majority of the interactions (Appendix C: Figs. S3.2, S3.3, S3.4 and S3.5; Tables S3.1 and S3.4), suggesting that most of their foraging rates varied amongst reefs and perhaps over time. Contrastingly, S. corallinus and S. stellatus demonstrated very few significant interactions between sites at the different time intervals (Appendix C: Figs. S3.3 and S3.4; Tables S3.2 and S3.3), suggesting that these two species had similar foraging rates at different reefs, and possibly over time.

Table 3.1 Results of the PERMANOVA performed on the foraging rates (bites.min-1). Species, site and time period were fixed factors. Significant interactions are highlighted in bold.

Source of variation Pseudo-F P (perm) Species 7.6 0.004 Site 92.6 0.001 Time 7.3 0.001 Species × Site 3.6 0.001 Species × Time 5.5 0.001 Site × Time 3.8 0.001 Species × Site × Time 3.7 0.001 Residuals - Total -

Individual species comparisons confirmed that S. corallinus displayed quite similar foraging habits among sites (~77% similarity) (Appendix C: Table S3.5). Likewise, S. stellatus showed high foraging similarity between sites (~73%), and when compared to each other, showed ~ 70% similarity (Appendix C: Table S3.5). Conversely, individual comparisons of S. argenteus, and S. sutor showed only ~43-44% similarity between sites (Appendix C: Table S3.5), and when compared to each other, they were only ~43% similar (Appendix C: Table S3.5). Correspondingly, when S. sutor and S. argenteus were compared to S. corallinus and S. stellatus, similarity ranged from ~44-47% (Appendix C: Table S3.5). These patterns were confirmed through the average PCO plot, which displays the average foraging rate (bites.min-1) for each species (± SE), at the different sites (Appendix C: Table Fig. S3.6).

All four species were observed predominantly biting on turf algae (TA) (Fig. 3.2). On average, S. corallinus took the most bites on TA (9.0 bites.min-1), followed by S. stellatus (7.4 bites.min-1), S. sutor (5.3 bites.min-1) and least by S. argenteus (4.3 bites.min-1) (Fig. 3.2). In addition to bites taken on TA, S. sutor took the most bites on fleshy macroalgae fronds (FMAF) (3.1 bites.min-1), followed by S. argenteus (2.1 bites.min-1), S. corallinus (0.9 bites.min-1) and least by S. stellatus (0.2 bites.min-1) (Fig. 3.2). Bites were also taken on the thallus of the macroalgae (FMAT), mostly by S. sutor (1.8 bites.min-1), followed by S. argenteus (0.8 bites.min-1) and least by S. corallinus (0.2 bites.min-1) (Fig. 3.2). No bites were taken by S. stellatus on FMAT (Fig. 3.2). Bites on seagrass blades (SGB) were mostly taken by S. sutor (1.8 bites.min-1), followed by S. argenteus (1.7 bites.min-1) and least by S. stellatus (0.4 bites.min-1) (Fig. 3.2). Bites on seagrass epiphytes (SGE), was done generally by S. argenteus (2.6 bites.min-1), followed by S. sutor (1.8 bites.min-1) and least by S. stellatus (0.7 bites.min-1) (Fig. 3.2). S. corallinus was not observed foraging on seagrass (Fig. 3.2). These observations were confirmed by the PCO plot, which displays the similarity of each rabbitfish surveyed, distinguished by species, and their distribution in 2D space in relation to the preferred foraged resource (Fig. 3.3).

Figure 3.2 Average proportion of bites (± standard errors) on the different resources by the four species of rabbitfish. Resources included; Fleshy macroalgae fronds (FMAF), fleshy macroalgae thallus (FMAT), seagrass epiphytes (SGE), seagrass blades (SGB) and turf algae (TA). Also displayed are the total bites (± standard errors) for each species.

For the majority of feeding substrata, foraging rates were higher when the feeding substrata abundance was high (Appendix C: Figs. S3.7a, b, c and d). Specifically, foraging rates by S. argenteus, on the fronds of macroalgae (R2 = 0.84), seagrass epiphytes (R2 = 0.77) and seagrass blades (R2 = 0.88) increased linearly as the abundance of these resources increased (Appendix C: Fig S3.7a). S. sutor demonstrated similar characteristics when foraging on the thalli of macroalgae (R2 = 0.64), seagrass epiphytes (R2 = 0.93) and seagrass blades (R2 = 0.98) (Appendix C: Fig S3.7d). 48

Conversely, S. corallinus, and S. stellatus showed no strong linear increase in foraging rates as the resource abundances increased (Appendix C: Figs. S3.7b and c).

All rabbitfishes exhibited a pattern of foraging rate typical of diurnal herbivores, with bite rates increasing through the morning periods, peaking in the mid afternoon and decreasing again in the late afternoon or early evening (Fig. 3.4; Appendix C: Figs. S3.2, S3.3, S3.4 and S3.5). At T1 (06:00 to 08:00 h), overall foraging rates averaged between 5.8 and 8.7 bites.min-1 across the 16 sites, increasing to a peak of between 13.1 and 21.0 bites.min-1 at T4 (12:00 to 14:00 h) (Fig. 3.4). At T6 (16:00 to 18:00 h), average foraging rates had declined to between 3.6 and 7.1 bites.min–1 (Fig. 3.4).

Figure 3.4 Bite rates (bites.min-1) for each rabbitfish species at different time periods.

3.3.2 Foraging selectivity

Resource selection, measured by the Vanderploeg and Scavia Relativized Index, showed that the average electivity of S. argenteus, and S. sutor were similar. Both showed neutral or positive electivity (Ei ≥ 0; Fig. 3.5) for all types of algae on offer. The electivity for S. corallinus, and S. stellatus were similar (Fig. 3.5). On average, both showed positive electivity (Ei ≥ 0) for TA, and a negative electivity for FMAF and FMAT (Ei < 0; Fig. 3.5).

Figure 3.5 Bar charts showing rabbitfish resource electivity (Vanderploeg and Scavia’s relativized index, Ei) averaged across sites for each the five foraged items: fleshy macroalgal fronds (FMAF), fleshy macroalgal thallus (FMAT), seagrass epiphytes (SGE), seagrass blades (SGB) and turf algae (TA). Average electivity is represented on the vertical axis: values >0 represent active selection disproportionate to abundance, values <0 represent resource avoidance.

3.4 Discussion

3.4.1 Foraging rates and behaviours

Rabbitfishes in the Seychelles exhibited differences and similarities in foraging behaviours and rates. S. corallinus and S. stellatus displayed rather specific foraging preference, whereby both species foraged primarily on turf algae. This finding, specifically for S. corallinus, is consistent with that found on the Great Barrier Reef (GBR), Australia (Fox and Bellwood, 2013; Hoey et al., 2013). Contrastingly, our results also show S. argenteus and S. sutor to have a much more generalist foraging behaviour compared to the other two species. They were observed foraging on turf algae, brown fleshy macroalgal (i.e. Sargassum sp.) epiphytes and thalli, seagrass epiphytes and blades, indicating dietary plasticity. Dietary plasticity is not uncommon in rabbitfishes (e.g. Wu, 1984; Fox and Bellwood, 2011; Hoey et al., 2013), although our results for S. argenteus oppose those from other global regions. Hoey et al., (2013) on the GBR demonstrated S. argenteus had similar foraging habits to S. corallinus, made up primarily of turf algae. Furthermore, gut content analyses in their study confirmed no presence of Sargassum sp., nor seagrass (Hoey et al., 2013). It must be noted however, that Hoey et al., (2013) only sampled S. argenteus from mid-shelf reefs where Sargassum sp. is rare/ absent (Wismer et al., 2009). Although other studies in the Pacific Ocean have found little evidence of S. argenteus feeding on Sargassum sp. when available (Fiji: Rasher et al., 2013; Guam: Paul et al., 1990). This signifies that S. argenteus may only feed on Sargassum sp. when it dominates the seascape (i.e. after regime shifts), as observed in the Seychelles. Our findings for S. sutor seem to be consistent with others within the Western Indian Ocean region. Almeida et al., (1999), and Lugendo et al., (2006) found seagrass in their gut contents in eastern Africa, while Chong-Seng et al., (2014), and Humphries et al., (2015) documented them foraging and assimilating macroalgae within this same region.

Turf algae was the primary feeding substrata of all four rabbitfish species in the present study. This is common with many other herbivorous fishes globally, whereby turf alga is

selected over other algal sources due to their morphology and nature (Kelly et al., 2016; Tootell and Steele, 2016). Turf algae are primarily composed of filamentous, palatable, and fast-growing species that can be readily digested as compared to more structurally or chemically-defended macroalgae species (Kelly et al., 2016), such as Sargassum sp. However, we also discovered foraging on Sargassum sp., dominated mostly by S. argenteus, and S. sutor. We distinguished between thallus and frond foraging of macroalgae, as it has been shown that bites on the fronds by fishes typically target epiphytes, rather than actual macroalgal material (Fox and Bellwood, 2008; Hoey and Bellwood, 2009). Our results also indicate that S. stellatus may not browse on Sargassum sp. at all, but rather graze on the epiphytes. The targeting of the epiphytes has little impact on the reduction of macroalgal biomass (Hoey and Bellwood, 2009), and may actually enhance macroalgal growth and longevity (Fox and Bellwood, 2008) by increasing photosynthetic capacity through cleaning of the fronds. However, incidental ingestion or dislodgement of the fronds may occur, therefore the role of epiphyte grazing should not be underestimated (Streit et al., 2016; Puk et al., 2016). Interestingly, in a recent review by Puk et al., (2016), it was declared that rabbitfishes are only able to keep macroalgae growth in check by targeting the fronds, but are unlikely to remove whole thalli. We found that S. argenteus, S. sutor, and to a smaller extent, S. corallinus, were indeed foraging on the thallus tissue of Sargassum sp., suggesting that their roles as macroalgal browsers should not be ignored. From an ecosystem perspective, foraging on different parts of macroalgae by different species can have a significant effect on its overall removal (Streit et al., 2016). This may lead to overall positive effects of browser diversity on macroalgal removal (Topor et al., 2019).

Seagrass habitats occupy a small proportion of the world’s oceans, but provide a disproportionately large range of ecological services, including nutrient recycling, sediment stabilisation and carbon sequestration (Waycott et al., 2009; Fourqurean et al., 2012). They are important habitats and foraging areas for many key fish species (Orth et al., 2006; Unsworth and Cullen, 2010). During our surveys, we distinguished between foraging on seagrass blades vs. foraging on the epibiota because herbivores can be effective at reducing algal loads, but some species consume seagrass directly and may

have a more detrimental effect (Heck and Valentine, 2006). Most of the blade foraging was done by S. argenteus and S. sutor, and to a small extent by S. stellatus. Although, when compared to grazing by macro-herbivores, such as dugongs and turtles (e.g. Fourqurean et al., 2012), their impact is considered light, and can actually stimulate seagrass growth and productivity (e.g. Valentine et al., 1997, Christianen et al., 2012). Foraging on the epiphytes of seagrass have a positive effect by increasing light availability to seagrass, which in turn improves photosynthetic capacity (e.g. Whalen et al., 2013). We found that, again, S. argenteus, and S. sutor foraged the most on seagrass epiphytes, and to a smaller extent, S. stellatus. We also discovered that S. corallinus did not forage on seagrass blades or seagrass epiphytes, demonstrating a functional distinction between the species.

Foraging rates varied significantly between sites, with a general trend of increased foraging on resources as they became more abundant. Specifically, foraging rates by S. argenteus on the fronds of macroalgae, seagrass epiphytes and seagrass blades increased as the abundance of these resources increased. Similarly, S. sutor foraged most on the thalli of macroalgae, seagrass epiphytes and seagrass blades, when these resources were in high abundance. This contradicts many other foraging studies (e.g. Hoey and Bellwood, 2010; 2011, Bennett and Bellwood, 2011; Chong-Seng et al., 2014), which all found the opposite trend towards resource increase. This may be explained by the findings of Boyer et al., (2004), who found that nutrient enrichment on coral reefs and seagrass habitats cause increased quantity and quality of plant and algal material, thereby increasing the foraging rates by herbivorous fishes in these areas. All of the contradicted studies (i.e. Hoey and Bellwood, 2010; 2011; Bennett and Bellwood, 2011; Chong-Seng et al., 2014) used transplanted bioassays of macroalgae, which may have had reduced nutritional quality from transportation and handling (due to algal fragility) (Lefèvre and Bellwood, 2010), thus rendering conflicting results to our study. Therefore, our results indicate that S. argenteus, and S. sutor may be foraging at a greater rate in areas where the quality of seagrass and macroalgae is higher. This finding, particularly for areas that have a high macroalgal coverage, insinuates that macroalgal overgrowth may have been favored in locations with high nitrogen loads, and suggests that nutrient

enrichment enhanced macroalgal regime shifts (Graham et al., 2015). As a result, this may have contributed to the higher quality of algal tissue for S. argenteus and S. sutor. This could be an interesting area for future research; by understanding the links between function, foraging and food quality.

All species of rabbitfishes exhibited a pattern of foraging rate typical for diurnal herbivores, with bite rates increasing through the morning periods, peaking in the mid afternoon and decreasing again in the late afternoon or early evening (Fox et al., 2009, Zemke-White et al., 2002, Polunin et al., 1995). Zemke-White et al., (2002) showed that the nutritional value of algal sources increases until mid-day and remains high throughout the afternoon, and this correlates with the diel-pattern of feeding by herbivorous fishes, which may be seeking nutrient rich sources of algal material (Zemke-White et al., 2002). Furthermore, the patterns exhibited may be attributed to predator avoidance at certain times of the day. Several studies (e.g. Catano et al., 2017; Madin et al., 2019) demonstrate highest predator presence at dusk and lowest in the mid-afternoon; corresponding well to our observations.

3.4.2 Foraging selectivity

Understanding the selectivity of a particular resource by a certain species may allow us to determine their precise impact on the ecosystem, especially following large increases in that particular resource. Following the 1998 bleaching event, most of the reefs surrounding the Inner Seychelles Islands underwent macroalgal regime shifts (Graham et al., 2015), and the ecological roles of many herbivorous fishes post-bleaching is poorly understood. However, increases in their biomass and heavy exploitation suggests that the current status of reefs around these islands are sustaining these fisheries (Robinson et al., 2019), further insinuating that herbivore populations have responded positively to habitat change, but species-specific responses are not well known.

We showed that the average electivity of S. argenteus and S. sutor across sites were similar by having neutral or positive electivity for all types of algae on offer, highlighting

them as generalist foragers. On the other hand, S. corallinus and S. stellatus both showed a positive electivity for only turf algae on average, indicating that they may be specialists in turf algae foraging. Furthermore, they both selected against Sargassum sp. fronds and thalli, signifying that they may not be efficient browsers (generally speaking). We also demonstrate that, on average, S. stellatus selected against seagrass blades and epiphytes. These patterns may be due to chemical defences released by Sargassum sp. (e.g. Soliman et al., 2008; Rasher et al., 2013) and seagrass (e.g. Vergés et al., 2007; 2011), acting as deterrents towards them.

Many reefs globally have already phase shifted from coral dominance to alternative states dominated predominantly by algae, and future coral reefs are likely to vary in appearance and functionality to those of the past and present. Our results emphasize the need for caution when assigning species to functional groups and assuming within-group functional equivalence (cf. Streit et al., 2016). The utilisation of foraging substrata needs to be considered when characterising species with regards to their functional impact (Adam et al., 2015).

Chapter 4 – The commercially important Shoemaker spinefoot, Siganus sutor connects coral reefs to neighbouring seagrass meadows

Abstract

Spatial management of fish populations can potentially be optimised by determining the area of influence of a particular species. We performed an acoustic tagging study implemented on Denis Island in the Seychelles to assess the area of influence of the heavily targeted shoemaker spinefoot, Siganus sutor. We investigated whether this species acts as a mobile link between coral patches and seagrass meadows, and whether their movements differed between day and night. The study incorporated an array of twenty-two acoustic stations deployed within dense coral patches, seagrass meadows and mixed habitats of both seagrass and coral. Fifteen S. sutor carrying internal acoustic tags, were monitored from November 2016 until May 2017. Detection patterns revealed them to be diurnal herbivores, with only rare nocturnal movements. Home-range estimates showed that individuals differed in their spatial range extents and habitats used, covering ~15% of the total shallow subtidal coastline of the island. However, they displayed very small daily movements (< 200 m), concentrated mainly around sites within mixed coral and seagrass habitats. Optimal number of detections were recorded when the coral to seagrass area ratio was approximately 1.6: 1. This ratio was confirmed through statistical prediction modelling. Identification of such links of commercially important species between networked habitats may help authorities consider incorporating seagrass meadows of the Seychelles into management discussions, that are currently lacking.

4.1 Introduction

Understanding the spatial ecology of fishes is an imperative component in determining the functioning of marine ecosystems, such as coral reefs (Stocks et al., 2015). Today, coral reefs are faced with increased threats from, overfishing, sedimentation, and climate change; and an insight into the spatial ecology of functionally important reef fishes is proving necessary (Welsh and Bellwood, 2012b). Several studies have shown that key functional groups of reef fishes, which, through their interactions with their environment, contribute significantly to coral reef resilience (e.g. Bellwood et al., 2004; Graham et al., 2006; Ebrahim et al., 2020). Species displaying site fidelity and habitat utilisation patterns can help researchers identify potentially significant spawning, feeding and shelter areas within the home range of fishes (Meyer and Holland, 2005; Jadot et al., 2006), and consequently lead to their effective management (Welsh and Bellwood, 2012b). A growing number of studies have used acoustic telemetry to identify species’ home ranges and their movement patterns (Welsh and Bellwood, 2012b). Studies have revealed that some species of parrotfishes; for example, the Redtail Parrotfish, Sparisoma chrysopterum, in the Caribbean (Munõz and Motta, 2000), and the Surf Parrotfish, Scarus rivulatus, in the Great Barrier Reef (GBR) (Welsh and Bellwood, 2012a), do not venture far, exerting a small area of influence. On the other hand, certain species of sea chub; for example, the Brassy Chub, Kyphosus vaigiensis, in the GBR (Welsh and Bellwood, 2014), and the Brown Chub, K. bigibbus, in North-western Australia (Pillans et al., 2017), reveal large-scale movements, and these are considered the most vital mobile links on coral reefs (Welsh and Bellwood, 2014; Allgeier et al., 2015).

Seldom do coral reefs function in isolation; they tend to form a component of a larger habitat network (Lugendo et al., 2006; Olds et al., 2012). Consequently, when designing spatial management strategies (e.g. marine protected areas, MPAs), authorities should be aware of the possible significance of habitat linkages between coral reefs and surrounding coastal ecosystems, given that certain commercially important species may depend upon these networked habitats for the purposes of refuges, nurseries, and feeding (Lugendo et al., 2006; Honda et al., 2013). In fact, MPAs with high connectivity

between reefs and other habitats (e.g. mangroves), have shown to be more effective than MPAs with low connectivity (Olds et al., 2012; 2013). Consequently, improved fishery resource management efforts are presumed to be caused by the effective conservation of these networked habitats, since they are known to promote fish biomass and species richness on neighbouring coral reefs (e.g. Nagelkerken et al., 2002; Dorenbosch et al., 2005; Olds et al., 2012). The spatial extent of fishes can primarily be determined by observing environmental cues such as the diel cycle (Honda et al., 2016; Davis et al., 2017). Accordingly, monitoring the use of different habitats during different times of the day can further explain important biological and ecological processes of fish species; such as foraging, predator avoidance and resting, and thus help support mobile link evidence.

Within the Western Indian Ocean (WIO) region, small herbivorous fish species of the family Siganidae, known as rabbitfish, are considered a vital protein source that sustains local human populations (Grandcourt and Cesar, 2003; Kaunda-Arara et al., 2003). Frequently, rabbitfishes constitute a sizeable proportion (> 40% by weight) of inshore reef fishery catches in these areas (Kamukuru, 2009; Hicks and McClanahan, 2012), dominated mostly by the WIO endemic, shoemaker spinefoot, Siganus sutor (Kamukuru, 2009; Samoilys et al., 2011; Hicks and McClanahan, 2012). In the small island nation of the Seychelles, rabbitfishes are the main target species (approx. 60 % by weight) of the artisanal fishery catch (Grandcourt and Cesar, 2003; Robinson et al., 2011), with S. sutor being the most sought-after species (Grandcourt and Cesar, 2003; Seychelles Fishing Authority, 2016). The local authorities have established minimum mesh size requirements for fish traps used (i.e. 40 mm), but these restrictions are poorly regulated (Robinson et al., 2011), so juvenile fish are often captured and landed illegally (Grandcourt and Cesar, 2003). This species frequents coral reefs and feeds on seagrass (Ebrahim et al., 2020), and so there is an assumption that they may be acting as mobile links between these habitats. Identification of such links may help authorities incorporate networked habitats into management strategies aimed at conserving this heavily fished resource.

The spatial ecology of S. sutor remains little studied, mainly focussed on spawning aggregations, and their response to the lunar cycle (e.g. Bijoux et al., 2013; Samoilys et

al., 2013). Consequently, in this study we aimed to use acoustic telemetry to understand the spatial ecology of this commercially important species. Our goals were to (a) determine their home range size (b) assess whether their range acts as a mobile link to neighbouring seagrass habitats, (c) compare the duration of their occupancy of different habitats, and (d) determine whether this varied with the diel cycle (diurnal vs. nocturnal).

4.2 Materials and Methods

4.2.1 Study region

This study was carried out on Denis Island (3∘ 48′ S, 55∘ 40′ E), a 1.4 km2 coralline cay located on the northern border of the Seychelles Bank (Feare et al., 2016). The island is privately owned and has ~100 inhabitants that maintain and manage a small luxury hotel. The island is surrounded by a lush seagrass habitat (primarily composed of Thalassodendron spp.) along the Eastern side, and dense coral patches mainly scattered around the North-western and Southern tip of the island (Fig. 4.1). The traditional Seychelles’ method of baited bamboo traps that primarily target herbivorous fish species, such as rabbitfishes, is banned on Denis Island; however, a few local resort staff are permitted to use a small drifting gillnet (measuring approximately 6 m by 3 m) to specifically herd small numbers of rabbitfishes on a low tide in the shallows for personal consumption.

Figure 4.1 Map of Denis Island, highlighting the benthic substrate composition, as well as the location of the 22 active receivers (S1-22) deployed. Each station is displayed with the maximum 150 m range of each receiver. Also highlighted on the map are the tagging locations (T1-T3).

4.2.2 Acoustic monitoring system

An array of twenty-two 69 kHz VR2W acoustic receivers (Amirix Systems) were placed around Denis Island, in dense coral patches, within lush seagrass meadows, and on the edge of either coral or seagrass patches that allowed signals from both coral and seagrass to be detected (Fig. 4.1). Receivers were placed at an approximate water depth of 1m near the shoreline to about 10m water depth at the deepest sites. It was ensured that receivers were never exposed out of the water, with at least 1m of water above on a low tide for the shallowest deployments. The array encompassed three specific habitat types, including seagrass meadows, patch reefs dominated by Porites spp. or widespread shallow coral reefs dominated by Acropora spp., and a mixed habitat of both seagrass and coral. Receivers were strategically spaced approximately 200 - 800 m apart at areas which represented the best coverage of each habitat type (i.e. seagrass meadows, coral reefs and mixed habitats), and where the detection ranges generally did not overlap (Fig. 4.1). The array was deployed from the 27th to the 29th of October 2016, and retrieved on the 6th of May 2017, allowing for just over six months of data collection.

4.2.3 Range test

Range testing to determine the range of acoustic reception was carried out using a sample acoustic receiver (VR2W) and tag (Vemco, V9-2H, 120 second nominal delay) (Amirix Systems) near a patch reef during daylight hours following similar methods to Bijoux et al., (2013). Briefly, the acoustic receiver was attached to a rope with the sensor side up, 3 m above the reef surface, and was anchored using lead weights. A small subsurface floating buoy was tied to the other end in order to keep the receiver line taught and upright in the water column. A line to deploy the tag was similarly prepared, and was deployed at different intervals (0, 25, 50, 100, 200, and 250 m distances) from the sample receiver. At each distance interval, the tag line was underwater for 30 mins. The amount of detections recorded at 0 m was used as a reference to compare the number of detections attained at the other intervals. The detection level was 100% at 25 m, 85% at

50 m, 50% at 100 m, and 15% at 150 m. No detections were recorded at 200 m and beyond.

4.2.4 Capture and tagging of fish

Procedures were conducted under the Animal Welfare Unit Ethics Permit (AEC Approval Number: SBS/518/15/SFA). Although baited traps are banned on the island, we were given permission to use them for the purposes of this study. We were also assisted by one of the resort staff using his gillnet to herd and capture rabbitfishes at low tides. Based on the advice and experience of this local fisherman, three specific sites were chosen for the capture (Fig. 4.1). After capture, fishes were placed in a large 100 L holding tank that was filled and aerated with seawater. Using a small aquarium net, individuals were transferred one at a time into a 30L tub of seawater containing an anaesthetic (540 g.L-1 Isoeugenol), until a total loss of equilibrium occurred. The fish was then removed from the solution and placed onto a wet, plastic measuring mat, whereby the fork length (FL) to the nearest mm was recorded. The fish was then transferred to a makeshift sling that was submerged partially into a small aerated tank allowing for their gills to be submerged in seawater, while its body exposed for easy handling during the tagging process. Once in position, the fish was tagged using a uniquely labelled T-Bar anchor tag (Floy tag and Mfg), that was recorded in a database. Secondly, an incision was made slightly off the mid-line between the pelvic fins and anus, and a coded acoustic transmitter (Vemco, V9- 2H, 120 second nominal delay) that was coated in antiseptic was inserted. Again, the uniquely labelled tag number was recorded. Inspection of the gonads was carried out to determine signs of potential spawning which may impact movement patterns observed. The opening was then closed using three non-overlapping sutures, and antiseptic cream placed over the incision. The fish was then placed into a second 100L holding tank with well aerated, running seawater until it had regained full equilibrium. This process was repeated for all individuals captured. The average time from capture to completion of surgery was 6–7 min, whereas recovery times ranged from 5 to 14 min. Once fully recovered, fish were released at their site of capture. In total, 25 individuals, ranging in FL from 190 mm to 291 mm were tagged and released for the purpose of this study

between the 30th of October and 16th of November 2016. The majority of all tagged fishes were considered to be adults, as the size at first maturity for S. sutor is ~22 cm (Kamukuru, 2009).

The care and use of experimental animals complied with The Seychelles Fishing Authority and the University of Queensland animal welfare laws, guidelines and policies as approved by the Animal Welfare Unit, UQ Research and Innovation, The University of Queensland under AEC Approval Number: SBS/518/15/SFA.

4.2.5 Data analysis

Before the data were analysed statistically, detection plots from each individual fish was examined using the VUE software package (Vemco) to verify for signs of mortality. Mortality was recognised by an apparent change in an individual’s movement patterns, to long periods of inactive behaviour (Welsh and Bellwood, 2014). In this case, all subsequent detections were excluded from the analyses (10 S. sutor were excluded on this basis). Furthermore, the first 24 h of detection data from each individual following release was omitted to account for any unusual behaviour which may have arisen from tagging (Welsh and Bellwood, 2014). Over the study period, we consistently received active pings from fifteen individuals, and so only their data were used in analyses.

Home Range

To determine the fidelity to the array region around Denis Island, we calculated the Residency Index (RI). RI was defined as, “the number of days when a fish was detected on any of the receivers within the array (the 22 receivers established for this study) divided by the time interval between the first day of the experiment and the last day each fish was detected” (Villegas-Ríos et al., 2013). A value of 0 specifies no residency and a value of 1 indicates permanent residency. We also determined whether each individual showed preference for a certain region of Denis Island (within the array), by calculating the proportion number of detections at each station.

The detection data of each individual fish was divided into diurnal and nocturnal sampling periods based on the average sunrise and sunset times of the Seychelles during the length of the experiment. Diurnal periods were set from 0600 to 1830 hrs to capture crepuscular movements, and nocturnal periods were set from 1831 to 0559 hrs. The package adehabitatLT in R (Calenge, 2006) was used to analyse the variability and frequency of any large-scale movements. The home range length of each fish was determined for diurnal and nocturnal periods using the minimum linear dispersal (MLD) and the median distance travelled (MDT). The MLD is defined as, “the shortest possible distance between the two most distant receivers where an individual had been detected” (Murchie et al., 2010). This metric was used to signify the minimum home range length (see Welsh and Bellwood, 2014). “The MDT provides a metric of the median dispersal of an individual from its principal area of residence” (Welsh and Bellwood, 2014). The MDT of each fish was determined by firstly identifying the most popular receiver visited by that fish, and secondly, calculating the median distance between that receiver and all other receivers where the fish was detected (Murchie et al., 2010).

To identify diurnal vs. nocturnal movement, the proportion of detections at an individual’s main detection site was calculated for each fish in both time periods. The data collected throughout the entire study period was used to calculate all the above metrics. Whilst the data for five randomly selected days was used to quantify the average size of an individual’s daily movements. We then compared the diurnal and nocturnal sampling periods using paired T-tests for the overall MLD and the daily MLD.

In order to get a home range estimate for S. sutor on Denis Island, we initially determined the approximate width of a constant habitat type (either coral, seagrass or mixed) at each site using visual analysis, as well as satellite imagery from the Ministry of Environment, Energy and Climate Change (MEECC). Accordingly, we established that a 50 m width was a reasonable estimate, and used this to calculate the possible diurnal and nocturnal home range areas. The diurnal home range area was 44935.0 ± 8290.0 m2 and the nocturnal home range area was 500.0 ± 484.0 m2 (Table 4.1).

Activity in each habitat

Statistical modelling can contain complex relationships among variables (e.g. how detection rates vary with the abundance of different substrates (in our case, seagrass and coral cover)). A standard way to view and evaluate the relative importance of different variables is to use the model coefficients to plot the behaviour in an orthogonal design by holding one variable constant but varying the second (e.g. Roff et al., 2015). This technique reveals the rate at which the second variable influences the response.

To discriminate among habitat types (i.e. coral, seagrass or mixed), that attracted the greatest activity from S. sutor, we used a Generalised Linear Model (GLM) in R. The number of detections at each site were set as our response variable and the area of coral and seagrass within each detection zone of each receiver as our predictor variables. To calculate the area coverage of coral and seagrass, we used 150 m as the radius index of each receiver, as no detections were recorded beyond 150 m during the range testing exercise. We then estimated coral and seagrass cover within this range from satellite imagery obtained from the MEECC, Seychelles. Given that the stations were not equally independent from each other, we verified for spatial autocorrelation (Moran, 1950) with Moran’s I tests in R (Paradis et al., 2004), by calculating the minimum path between two sites. We discovered that no spatial correlation structure was required by analysing the model fit statistics of each spatial correlation structure (i.e. exponential, linear, gaussian, quadratic, and spherical), and so we ran the model without spatial autocorrelation. We then used a quasi-Poisson error distribution model (Crawley, 2007) to account for over- dispersed data and analysed the coefficients for the significance of each variable. The GLM predict function (glm. predict) (Schlegel, 2018), in R was then applied to explore how changes in the cover of coral and seagrass affected the number of detections. The model produced was displayed using a bar chart of the average number of detections at different levels of coral and seagrass.

4.3 Results

4.3.1 Home Range

In total, fifteen individual S. sutor (average length 25.3 ± 0.7 cm SE; range 21.1–29.1 cm), (Table 4.1) were monitored for just over six months. During this period, individuals travelled an overall average minimum linear distance (MLD) of 898.7 ± 165.8 m during the day and 10 ± 10 m at night (and overall ranging from 300 to 2400 m during the day, and 0 to 150 m at night; Table 4.1). Their diurnal values were extensive and, on average, covered ~15% of the available coastline of Denis Island (Fig. 4.1; Table 4.1). A significant difference was detected between the diurnal and nocturnal MLDs over the whole study for S. sutor (t = 5.1, p = 0.002), with most detections occurring during the day compared to at night (Figs. 4.2 and 4.3). Particularly, from visual inspection of Fig. 4.4, it was clear that S. sutor were most active between 6-8 am.

The patterns for the MLD’s were reflected by the median distance travelled (MDT), with practically every tagged fish travelling great distances away from their primary detection zone (Table 4.2). They had an average MDT of 323 ± 99 m during diurnal periods and 0 ± 0 m during nocturnal periods (Table 4.1). The values stated for the MLD and MDT are likely to be conservative, with actual movement range predicted to be larger, as the residency index was on average 70 ± 5%, signifying that individuals were beyond the detection zones of the array (sites) ~30% of the time during the length of the experiment (Table 4.1).

Table 4.1 Summary of Movement data of 15 S. sutor tagged with V9-1L transmitters.

Fish Length Diurnal Nocturnal Diurnal Nocturnal Daily Daily Diurnal daily Nocturnal daily Diurnal Nocturnal Residency (FL; MLD* MLD* home range home range diurnal nocturnal proportion of proportion of MDT‡ MDT‡ index cm) estimate (m2) estimate (m2) † MLD* MLD* overall MLD* overall MLD* †

40939 29.1 450 0 22500 0 100 0 0.2 0 125 0 1.0

40941 27.7 450 150 22500 7500 90 0 0.2 0 185 0 0.3

40943 26.1 1400 0 70000 0 266 0 0.19 0 285 0 0.5

40944 27.7 690 0 34500 0 138 0 0.2 0 250 0 0.6

40945 21.8 750 0 37500 0 138 0 0.18 0 0 0 0.6

40947 21.1 1400 0 70000 0 244 0 0.17 0 370 0 0.7

40950 22 690 0 34500 0 138 0 0.2 0 125 0 0.5

40954 25.6 1800 0 90000 0 400 0 0.2 0 1200 0 0.5

40960 24.2 750 0 37500 0 138 0 0.18 0 100 0 0.5

40961 21.8 2400 0 120000 0 480 0 0.2 0 600 0 0.6

40962 26.5 900 0 45000 0 200 0 0.2 0 400 0 1.0

40963 27.4 900 0 45000 0 200 0 0.2 0 1100 0 0.6

40964 26 300 0 15000 0 100 0 0.3 0 0 0 0.9

40965 26.2 300 0 15000 0 100 0 0.3 0 100 0 0.6

40967 26.9 300 0 15000 0 60 0 0.2 0 0 0 1.0

Mean 25.3 ± 898.7 ± 10 ± 10 44935.0 ± 500 ± 484.1 186.1 ± 0 ± 0 0.2 ± 0.0 0 ± 0 322.7 ± 0 ± 0 0.7 ± 0.05 (±SE) 0.7 165.8 8290.0 32.8 98.5 *Minimum linear distance (m) † Home ranges calculated assuming a constant width of 50 m ‡ Median distance traveled (m)

Table 4.2 Summary of visited stations and principle station location of each tagged fish.

Fish ID Visited Principal station Furthest location stations 40939 S9, 11 S11 S11 40941 S9, 11 S11 S9 40943 S7, 9, 10, 11, 12 S11 S7 40944 S7, 8, 9 S8 S9 40945 S8, 9 S9 S9 40947 S7, 9, 10, 11 S11 S11 40950 S8, 9 S9 S9 40954 S2, 3, 7, 20 S20 S9 40960 S8, 9 S8 S9 40961 S3, 4, 5, 7, 8, 9, 10, 11, 12, 20 S9 S20 40962 S3, 4, 5, 6, 7, 8, 9, 19 S7 S20 40963 S2, 10, 20 S20 S20 40964 S4, 22 S22 S22 40965 S4, 22 S22 S22 40967 S4, 22 S22 S22

4.3.2 Daily movement patterns

Analysing a single 24-h period (with five randomly selected days used for replication), demonstrated that the MLD values of individual fishes were relatively small, with an average distance of 186 ± 33 m (Table 4.1), occurring solely during the day time, and on average, 20% of an individual’s total diurnal MLD and 0% of an individual’s total nocturnal MLD (Table 4.1). The average daily MLD values of each individual were significantly higher during the day, compared to at night (t = 5.5, p = 0.00008).

4.3.3 Activity in each habitat

Individual fish were detected over the majority of the array, but overall, they showed high site fidelity to Station 9, where ~60% of all detections occurred (Figs. 4.1 and 4.2). We evaluated individual detections in diurnal versus nocturnal subclasses, and it was confirmed that S. sutor were generally more mobile in diurnal periods (Figs. 4.2, 4.3 and 4.4). Diurnal samples were characterised by pings received at most stations, but

predominantly higher at Station 9 (~35%), whilst total nocturnal detections were recorded ~25% at this same station (Fig. 4.2).

Looking at each habitat type (i.e. coral, seagrass and mixed), there was a significant difference in the number of detections coming from each habitat (F= 6.82, p = 0.006). From the detection data, it was clear that most detections originated from mixed habitats (Fig. 4.3). Statistically, we showed that seagrass cover significantly influenced the number of detections (F = 2.14, p = 0.05), whilst coral cover was a marginally significant influence (F = 1.98, p = 0.06).

Figure 4.2 % number of detections of S. sutor at each station during the whole study period (separated into diurnal and nocturnal detections).

Figure 4.3 Bar chart of the number of detections (averaged over the whole study period and separated into diurnal and nocturnal detections ± S.E.) within each habitat type.

Figure 4.4 Bar chart demonstrating the mean diel variation in the number of acoustic detections recorded between 17 November 2016 to 10 May 2017. Time bin 0 corresponds to 00:00 to 00:59 h.

A total of 6585 detections were recorded over the whole study period (Table 4.3). We determined that the optimal mixed habitat that attracted the most detections (both during the day and at night), had a coral area between 50000 and 60000 m2, and seagrass area was between 30000 and 40000 m2 (~1.6:1 ratio) (Fig. 4.5; Table 4.3). This was reflected in the GLM model, which predicted that most detections occurred at the same coral to seagrass ratio (Fig. 4.6).

Station Habitat Coral area (m2) Seagrass area (m2) Number of detections type 1 Mixed 17600 17600 0 2 Mixed 13000 39700 17 3 Seagrass 0 70600 9 4 Seagrass 0 70600 45 5 Seagrass 0 53000 24 6 Seagrass 0 70600 12 7 Seagrass 0 53000 145 8 Mixed 17600 53000 354 9 Mixed 35300 53000 3713 10 Coral 70600 0 21 11 Mixed 17600 53000 1296 12 Mixed 35300 35000 8 13 Coral 53000 0 0 14 Coral 70600 0 0 15 Coral 70600 0 0 16 Coral 70600 0 0 17 Coral 70600 0 0 18 Seagrass 0 70600 0 19 Mixed 17600 53000 2 20 Mixed 17600 53000 30 21 Coral 70600 0 0 22 Mixed 26500 26500 909

Total: 6585

Figure 4.5 Number of detections (averaged over the whole study period, and separated into diurnal and nocturnal detections) recorded in regions with varying covers of coral and seagrass (m2).

Figure 4.6 Predicted number of detections (averaged over the whole study period) in regions with varying covers of coral and seagrass (m2).

4.4 Discussion

Previous studies have recognised that some herbivorous fishes exert their vital functional processes over multiple habitats, by acting as mobile links, within marine ecosystems (Nyström and Folke, 2001; Lundberg and Mober, 2003). Our study is one of the first to provide approximations of movement and home range size of the commercially important S. sutor, and subsequently strengthens our knowledge of their role as mobile herbivorous trophic links on coral reefs. Previous observations on the behaviour of this species confirm that they travel in big schools (Randall et al., 1997; Robinson et al., 2011; 2017), and so the home ranges displayed by the tagged individuals expectedly represent a significant proportion of the population.

Several reviews comparing the home range movements of small-bodied vs. large- bodied reef fishes, show that the latter often display greater movements (Kramer et al., 1999; Sale et al., 2005). S. sutor is considered to be a small-bodied fish (Mbaru et al., 2018), and, on average occupied a section of reef ~900 m long. This estimate is about 2 - 8 times greater than other similar sized schooling herbivorous fishes such as

the parrotfish, S. rivulatus, in the GBR (Welsh and Bellwood, 2012a) and S. chrysopterum, from the Caribbean (Munõz and Motta, 2000). Both these species displayed limited home range sizes between sites (Welsh and Bellwood, 2012a; b), and are therefore not considered to be important mobile links on coral reefs. Conversely, other studies have demonstrated very large home ranges in similar sized, schooling herbivorous fishes. For example, Welsh and Bellwood, (2014) showed that the sea chub, K. vaigiensis in the GBR, and Pillans et al., (2017) demonstrated that another chub species, K. bigibbus, in North-western Australia, to have the greatest movements recorded to date (> 2 km), and are considered crucial links over extensive areas. Therefore, although S. sutor did not show such extreme movements, their travels still encompassed large enough areas that covered both coral reef habitats, as well as neighbouring seagrass meadows.

Interestingly, other studies have ruled out the importance of seagrass meadows to adult S. sutor (e.g. Gell and Whittington, 2002; Kimirei et al., 2011). Instead, they found seagrass meadows to be critical nursery grounds for the juveniles (Gell and Whittington, 2002, Kimirei et al., 2011). This is an important finding from a management perspective in the Seychelles, as seagrass meadows appear to be crucial habitats for S. sutor at different life stages.

Most detections occurred within mixed habitats, where both coral and seagrass were present, insinuating that this habitat may be most beneficial to this species. We found that there was an optimal number of detections when the coral to seagrass area ratio was approximately 1.6:1. Through statistical modelling we confirmed this coral to seagrass ratio to be most optimal for S. sutor. The coral structure may be providing shelter from predation (reviewed in Graham and Nash, 2013; Richardson et al., 2017), whilst the seagrass may be a vital food source (demonstrated by Ebrahim et al., 2020). In order to limit physical expenditure, these fishes may be distributed as predicted by foraging models, i.e. the optimal foraging theory (Charnov, 1976) and the ideal free distribution (Fretwell and Lucas, 1970), whereby individual foragers maximise their energy gain by distributing themselves proportionally to available resources. This was established when observing the small daily distances (< 200 m on average) covered by S. sutor. Hence, although overall S. sutor have large home ranges, it appears that these distances may be covered over a long time period. They may be utilising a small 75

area for particular resources, and when the resources run scarce, they migrate to another. Indeed, it has been revealed that grazing on seagrass by small herbivorous fish species such as rabbitfishes can be beneficial, and can actually stimulate seagrass growth and productivity (e.g. Valentine et al., 1997; Christiansen et al., 2012). Ebrahim et al., (2020) also recorded S. sutor directly targeting the epiphytes that grow on the seagrass blades. Foraging on the epiphytes of seagrass have a positive effect by increasing light availability to the blades, which in turn improves photosynthetic capacity (e.g. Whalen et al., 2013).

S. sutor were most active during the day, particularly in the early morning, highlighting them as diurnal herbivores that are feeding primarily during the day. This was confirmed through foraging behaviour observations by Ebrahim et al., (2020) who showed that S. sutor would target nutrient rich sources of algal material particularly during the day. Some nocturnal detections were present mainly occurring at one particular station (Station 9), found within a mixed habitat. This may be a possible resting site for the species on Denis Island. Movements of S. sutor due to potential spawning aggregations were ruled out in our study. During the tagging process, we did not observe any mature or maturing gonads. Furthermore, the movement data observed over the six-month period, was fairly consistent, and there was no display of any sudden / sporadic behaviours to indicate other potential activities to have developed over the study period.

Our findings provide supporting evidence that networked habitats to coral reefs, such as seagrass meadows, should be considered when designing Marine Protected Areas. The connectivity of habitats and the strength of the linkages to species within management components must be investigated in order for establishing efficient protection. We must acknowledge however, that the patterns exhibited by S. sutor in our study were concentrated around a small isolated island, where resources (i.e. seagrass) were in fairly close proximity to the reefs. Indeed, movement patterns may vary with location, influenced by other various site factors. Ecological theory suggests that different species may consider dangers (i.e. increased vulnerability to predation while moving across habitats and energetic spending on swimming) and the benefits of having a large home range (i.e. access to higher-quality diet sources) (Lindstedt et

al., 1986; Owen-Smith et al., 2010), and so, it would be interesting to determine if S. sutor have similar movement patterns in other areas.

Furthermore, this study only investigated connectivity between coral reefs and seagrass meadows. Other studies have demonstrated that mangrove habitats are also crucial foraging (e.g. Lugendo et al., 2006), and nursery (Laroche et al., 1997; Lugendo et al., 2005) grounds to S. sutor. Therefore, it is recommended that they also be investigated in order to push for furthering networked habitats being considered for protection. Stable isotope analysis on tissue and/ or otoliths could be investigated to help confirm that S. sutor are using seagrass for foraging purposes, and also ascertain the link among marine ecosystems.

Although it was not intended, we acknowledge that our sample size of fifteen individuals was fairly small when compared to similar studies (e.g. Welsh and Bellwood, 2012a; Pillans et al., 2017); and so, it is suggested that future studies incorporate a larger sample size for a more concrete conclusion.

General discussion: next steps in the conservation of rabbitfishes

5.1 Overview of key findings

Herbivorous fishes may contribute in promoting coral reef resilience (Hoey and Bellwood, 2009). The identification of processes that regulate the health of coral reefs is key to understanding how management efforts might defend them against anthropogenic threats (Strain et al., 2019). Most studies on tropical coral reefs have concentrated on the ecological roles of parrotfishes and surgeonfishes, with few studies investigating the functional roles of rabbitfishes. The majority of these studies have been restricted to the Great Barrier Reef in eastern Australia. Consequently, little is known of their behaviour and ecology from reefs elsewhere.

Rabbitfishes are a crucial food source to the majority of the Western Indian Ocean region, comprising the bulk of their artisanal fisheries. In the Seychelles, in particular, they constitute over 60% of the total artisanal catch (Seychelles Fishing Authority, 2016) and, notably, there are no enforced measures aimed at management of their stocks currently in place. Fishes that perform crucial ecological roles on coral reefs (e.g. herbivory), and that are also important targets of fisheries, should be a primary focus for research efforts and management actions (Bellwood et al., 2012; Mumby, 2014).

This thesis aimed to bridge the gap in the available research by exploring the distribution, ecological functions, and spatial ecology of commercially important rabbitfish species in the Seychelles archipelago of the Indian Ocean. Chapter 2 thereof examined whether, and how, the benthic habitat characteristics of reefs (e.g. reef type, rugosity, and coral cover) and marine reserves combine to influence the biomass of key functional groups of herbivorous fishes on Seychelles reefs. Specific analyses on four rabbitfish species (i.e. the shoemaker spinefoot [Siganus sutor], the streamlined spinefoot [S. argenteus], the blue-spotted spinefoot [S. corallinus], and the brown- spotted spinefoot [S. stellatus]), which are the primary targets in local artisanal fisheries, were carried out separately to determine the effects of various habitat combinations and reserves.

These analyses revealed that the distribution of herbivorous fishes on reefs in the Seychelles was shaped by the biological characteristics of individual reefs. Fish functional groups were found in distinct locations on local reefs, and their specific distributions reflected variation in the cover of fleshy macroalgae, live coral, and coral rubble. For the target rabbitfish species in particular, S. argenteus, and S. sutor, were strongly associated within areas of dense macroalgae, whereas S. corallinus, was associated with areas where the cover of live coral was high and epilithic algal matrix cover was low. S. stellatus was not influenced by any of the factors tested in our study. These results suggest that there might be some spatial separation in the distribution of different components of fish herbivory on reefs in the Seychelles. As herbivory is a central process that underpins the resilience of coral reefs, this spatial discordance in the distribution of herbivorous fish functional groups might affect the capacity of individual reefs to resist, or recover from, disturbance.

Pinpointing distribution patterns is one piece of the puzzle, but understanding what ecological function herbivorous fishes perform in those specific areas is an essential component. Chapter 3 features the first direct evaluation of the ecosystem function of the four commercially important species named above, in a single tropical community in the Seychelles. S. argenteus and S. sutor were similar in their electivity of algal materials, but varying foraging rates were observed between sites and at different time periods. On their part, S. corallinus and S. stellatus were similar in their electivity of algal materials, and showed similarities in their foraging rates between sites and at different time periods. These findings illustrate possible differences in the ecological contributions among rabbitfish species on coral reefs, and emphasize the need for caution when assigning species to functional groups and assuming within-group functional equivalence. The results also support the classic niche theory that species within a community must use resources differently in order to coexist over evolutionary timescales. These results further provide valuable insights for the management of rabbitfishes in tropical fisheries because it implies that the conservation of different species might result in distinct shifts in the competitive dominance of coral and algae.

Herbivorous fishes have been found to be a critical component of coral reef resilience, yet the spatial extent to which individual fish species can, and do, perform their important functional roles is poorly understood (Welsh and Bellwood, 2012b). This insight may be vital to the effective management of certain fish populations. Chapter 4 investigated the area of influence of S. sutor, the most heavily targeted member of the Seychelles Siganids. Home-range estimates showed that individuals differed in their spatial range extents and habitats used, covering a large area of a shallow subtidal coastline. However, they displayed very small daily movements, which were concentrated mainly around sites within mixed coral and seagrass habitats. The movements of S. sutor were influenced by seagrass abundance, with statistical prediction modelling confirming that they preferred an approximate ratio of 1.6m2 of coral to 1m2 of seagrass. Identification of such links for commercially important species between networks of habitats may guide and assist local authorities in the creation and subsequent implementation of an effective management plan for the fishery.

5.2 Limitations and avenues for future research

Chapter 2 revealed that reef features, such as coral and algal cover, can influence the distribution of herbivorous fishes. The results also displayed a large proportion of residual variation, which suggests that other factors that were not considered may be important drivers in the distribution of these fishes. Quantifying certain factors, such as fishing effort (both within MPAs and fished areas) and predation risk at each site could account for this variation.

Furthermore, Chapter 2 only considered top-down effects that may be influencing coral reef resilience. However, other studies (e.g. Lapointe, 1997; Gilby et al., 2015) have shown that nutrient enrichment (i.e. bottom-up approach) can increase algal growth rates, which may adversely impact coral reef health. These factors may be worth considering in future studies of this nature. Moreover, the point count methods used to quantify fish communities have been shown to underestimate the density of large mobile fishes (such as excavators) (Dickens et al., 2011). Perhaps using other methods such as timed swims (e.g. Hill and Wilkinson, 2004), belt transects (e.g. English et al. 1997), or a combination of both (e.g. Choat and Pears, 2003) might prove to be more accurate. 80

Chapter 3 revealed that certain species avoided feeding on algal sources that were in extreme abundances, indicating that phase-shifted environments in the Seychelles may have the capacity to recover following anthropogenic effects and natural disasters. However, if the abundance of the phase-shifted substrate dominates, it may be difficult to reverse. An investigation of this exact threshold is an avenue for future research, and may serve as an indicator of phase-shifts. It is further recommended that future studies on this topic incorporate gut content or stable isotope analyses to complement the current research. The impact of herbivores on reefs are not only based on the type of algae they remove, but can also be determined by the ecological fate of that bite.

Spatial management of fish populations can potentially be optimised by understanding the area of influence of a particular species. By using acoustic telemetry, Chapter 4 identified that the commercially important S. sutor utilised both seagrass and coral habitats. However, we did not show what they use these habitats for. Perhaps this could be an avenue for future research. Furthermore, our study only investigated adult S. sutor movements in shallow areas (<10 m). Yet, research has shown them to frequent deep reefs (>10 m) (e.g. Bijoux et al., 2013). Therefore, we must acknowledge their movements to potentially be much more than what is described in this Chapter; especially during spawning aggregations (Bijoux et al., 2013). Additionally, this research could be used as a basis to investigate whether other networked habitats, such as mangroves, fall within their spatial extent, thereby serving to strengthen possible management strategies. This study was done in an area where networked habitats were within close proximity to one another. Replicating the study in an area where different habitats are farther apart could perhaps identify larger spatial extents of S. sutor. Additionally, other methods such as tissue and otolith stable isotopes can be used as a future research direction to ascertain the link among marine ecosystems (i.e. coral reef, seagrass and mangrove habitats)

5.3 Management of the fishery moving forward

The fisheries sector is one of the main pillars of the Seychelles’ economy (Campling and Rosalie, 2006). Fish remains a vital protein source, contributing on average approximately 35 – 40% to the diet of the locals (Lallemand, 2014). The locally consumed fishes are obtained mainly from the artisanal fishery, which is dominated by rabbitfishes (Seychelles Fishing Authority, 2016). Unfortunately, recent statistics are showing a steady decline in total catches, as well as the catch per unit effort, from 2006 to 2016 (Seychelles Fishing Authority, 2016). Therefore, there is an urgent need to address fisheries management strategies in the country to protect the resource and to promote more sustainable fishing practices.

These declining trends have not gone unnoticed by the local communities. Informal and voluntary protected areas are being increasingly established in the Seychelles by the local inhabitants. Groups of stakeholders and members of the public are recognising certain areas as vulnerable, and are actively generating social awareness and harnessing public support for their conservation efforts at a district level. For instance, the Praslin Fishers Association is presently striving to establish an informal voluntary seasonal fishery reserve in Baie Ste Anne on Praslin Island. However, a scientific basis for such initiatives is often lacking, and without endorsement from the Government in terms of the implementation of a legal framework, enforcement is an issue (Beger et al., 2004).

MPAs have been recognized as important tools for the conservation of herbivorous fishes and habitats (Russ and Alcala, 1996; Mumby and Steneck, 2008; Strain et al., 2019), with reported effects including increases in density and biomass, particularly for commercial species (Russ and Alcala, 1996; Friedlander et al., 2003). However, if management strategies are not effective, MPAs are unlikely to meet their conservation goals (Fox et al., 2014). This thesis did not aim to investigate MPA performance in the Seychelles. Yet, the research carried out in the archipelago revealed that the management initiatives applicable to the areas studied, most of which have been in place for over 40 years, are either ineffective or are being circumvented. This observation is based on the finding in Chapter 2 that increased herbivorous fish biomass was not significantly higher when “protected” areas were compared with 82

fished areas. The most probable cause for this result may be the lack of precise and accurate monitoring data. A crucial aspect of effective management is to adopt sound monitoring programmes that provide information on changes that are occurring within MPAs, and also monitoring information regarding whether the management objectives are being met (Pelletier et al., 2005). It is, therefore, recommended that these strategies be implemented in the Seychelles for the success of future management efforts.

This thesis demonstrated that commercially important rabbitfishes may help to promote the capacity of reefs to resist, or recover from, disturbances that lead to algal dominance in the Seychelles. It further revealed that networked habitats, such as seagrass meadows, are important areas for the targeted S. sutor. Unfortunately, many protected areas globally, and particularly in the Seychelles, fail to include them within the boundaries of MPAs. By ensuring that these networked habitats are reflected in management plans, local authorities may be able to make this fishery sustainable for generations to come.

References

Adam, T.C., Kelley, M., Ruttenberg, B.I., Burkepile, D.E. (2015). Resource partitioning along multiple niche axes drives functional diversity in parrotfishes on Caribbean coral reefs. Oecologia 179, 1173-1185. Adam, T.C., Schmitt, R.J., Holbrook, S.J., Brooks, A.J., Edmunds, P.J., Carpenter, R.C., Bernardi, G. (2011). Herbivory, connectivity, and ecosystem resilience: response of a coral reef to a large-scale perturbation. PloS one 6. Ahamada, S., Bigot, L., Bijoux, J., Maharavo, J., Meunier, S., Moyne-Picard, M., Paupiah, N. (2002). Status of coral reefs in the south west Indian Ocean island node: Comoros, Madagascar, Mauritius, Reunion and Seychelles. Status of the coral reefs of the world. Australian Institute of Marine Science, Townsville, pp79-100. Allgeier, J.E., Adam, T.C., Burkepile, D.E. (2017). The importance of individual and species-level traits for trophic niches among herbivorous coral reef fishes. Proceedings of the Royal Society, 284, 20170307. Allgeier, J.E., Layman, C.A., Mumby, P.J., Rosemond, A.D. (2015). Biogeochemical implications of biodiversity and community structure across multiple coastal ecosystems. Ecology Monographs 85, 117–32 Allgeier, J.E., Valdivia, A., Cox, C., Layman, C.A. (2016). Fishing down nutrients on coral reefs. Nature Communications 7, 12461 Almeida, A.J., Marques, A., Saldanha, L. (1999). Some aspects of the biology of three fish species from the seagrass beds at Inhaca Island, Mozambique. Cybium 23, 369-376. Alwany, M.A. (2008). Behavioural responses towards different light intensities of two Red Sea surgeonfishes. Egyptian Journal of Experimental Biology (Zoology) 4, 199-204. Anderson, M., Gorley, R., Clarke, K. (2008) PERMANOVA+ for PRIMER. Plymouth, UK: Primer-E. Beger, M., Harborne, A.R., Dacles, T.P., Solandt, J.L., Ledesma, G.L. (2004). A framework of lessons learned from community-based marine reserves and its effectiveness in guiding a new coastal management initiative in the Philippines. Environmental Management 34, 786-801.

Bellwood, D.R., Choat, J.H. (1990). A functional analysis of grazing in Parrotfishes (Family Scaridae)—the ecological implications. Environmental Biology of Fishes 28, 189–214. Bellwood, D.R., Hughes, T.P., Folke, C., Nyström, M. (2004). Confronting the coral reef crisis. Nature 429, 827–833. Bellwood, D.R., Hughes, T.P., Hoey, A.S. (2006). Sleeping functional group drives coral-reef recovery. Current Biology 16, 2434-2439. Bellwood, D.R., Renema, W., Rosen, B.R. (2012). Biodiversity hotspots, evolution and coral reef biogeography. Biotic evolution and environmental change in Southeast Asia, 216. Bennett, S., Bellwood, D.R. (2011). Latitudinal variation in macroalgal consumption by fishes on the Great Barrier Reef. Marine Ecology Progress Series 426, 241–252. Bijoux, J.P., Dagorn, L., Berke, G., Cowley, P.D., Soria, M., Gaertner, J.C., Robinson, J. (2013). Temporal dynamics, residency and site fidelity of spawning aggregations of a herbivorous tropical reef fish Siganus sutor. Marine Ecology Progress Series 475, 233-247. Bonaldo, R.M., Hay, M.E. (2014). Seaweed-coral interactions: variance in seaweed allelopathy, coral susceptibility, and potential effects on coral resilience. PLOS One 9, 85786. Borsa, P., Lemer, S., Aurelle, D. (2007). Patterns of lineage diversification in rabbitfishes. Molecular Phylogenetics and Evolution 44, 427-435. Bos, A.R., Cruz-Rivera, E., Sanad, A.M. (2017). Herbivorous fishes Siganus rivulatus (Siganidae) and Zebrasoma desjardinii (Acanthuridae) feed on Ctenophora and Scyphozoa in the Red Sea. Marine Biodiversity 47, 243-246. Boyer, K.E., Fong, P., Armitage, A.R., Cohen, R.A. (2004). Elevated nutrient content of tropical macroalgae increases rates of herbivory in coral, seagrass, and mangrove habitats. Coral Reefs 23, 530-538. Bozec, Y.M., O'Farrell, S., Bruggemann, J.H., Luckhurst, B.E., Mumby, P.J. (2016). Tradeoffs between fisheries harvest and the resilience of coral reefs. Proceedings of the National Academy of Sciences, USA 113, 4536- 4541. Brandl, S.J., Bellwood, D.R. (2015). Coordinated vigilance provides evidence for direct reciprocity in coral reef fishes. Scientific Reports 5, 14556. 85

Brandl, S.J., Bellwood, D.R. (2014). Individual‐based analyses reveal limited functional overlap in a coral reef fish community. Journal of Animal Ecology 83, 661-670. Bruggemann, J.H., Van Oppen, M.J., Breeman, A.M. (1994). Foraging by the stoplight parrotfish Sparisoma viride. I. Food selection in different, socially determined habitats. Marine Ecology Progress Series 106, 41-41. Bryan, P.G. (1975). Food habits, functional digestive morphology, and assimilation efficiency of the rabbitfish Siganus spinus (Pisces, Siganidae) on Guam. Pacific Science 29, 269-77. Burkepile, D.E., Hay, M.E. (2008). Herbivore species richness and feeding complementarity affect community structure and function on a coral reef. Proc Proceedings of the National Academy of Sciences, USA 42, 16201-16206. Cáceres, I., Ibarra-García, E.C., Ortiz, M., Ayón-Parente, M., Rodríguez-Zaragoza, F.A. (2020). Effect of fisheries and benthic habitat on the ecological and functional diversity of fish at the Cayos Cochinos coral reefs (Honduras). Marine Biodiversity 50, 9. Calenge, C. (2006). The package “adehabitat” for the R software: a tool for the analysis of space and habitat use by animals. Ecological Modelling 197, 516- 519. Campling, L., Rosalie, M. (2006). Sustaining social development in a small island developing state? The case of Seychelles. Sustainable Development, 14, 115-125. Cardoso, S.C., Soares, M.C., Oxenford, H.A., Côté, I.M. (2009). Interspecific differences in foraging behaviour and functional role of Caribbean parrotfish. Marine Biodiversity Records 2, 148. Catano, L.B., Barton, M.B., Boswell, K.M., Burkepile, D.E. (2017). Predator identity and time of day interact to shape the risk–reward trade-off for herbivorous coral reef fishes. Oecologia 183, 763-773. Charnov, E.L. (1976). Optimal foraging, the marginal value theorem. Theoretical Population Biology 9, 129–136. Cheal, A., Emslie, M., Miller, I., Sweatman, H. (2012). The distribution of herbivorous fishes on the Great Barrier Reef. Marine Biology 159, 1143-1154. Cheal, A.J., MacNeil, M.A., Cripps, E., Emslie, M.J., Jonker, M., Schaffelke, B.,

Sweatman, H. (2010). Coral–macroalgal phase shifts or reef resilience: links with diversity and functional roles of herbivorous fishes on the Great Barrier Reef. Coral reefs 29, 1005-1015. Chesson, P., Pacala, S., Neuhauser, C. (2001). Environmental niches and ecosystem functioning. Functional Conservation and Biodiversity pp213-245. Choat, J.H. (1991). The biology of herbivorous fishes on coral reefs. The ecology of fishes on coral reefs. Academic Press, San Diego. pp 120–155. Choat, J.H., Clements, K.D., Robbins, W.D. (2002). The trophic status of herbivorous fishes on coral reefs. 1: dietary analyses. Marine Biology 140, 613-623. Choat, J.H., Pears, R. (2003). A rapid, quantitative survey method for large, vulnerable reef fishes. In: Wilkinson, C., Green, A., Almany, J., and Dionne, S. Monitoring Coral Reef Marine Protected Areas. A Practical Guide on How Monitoring Can support Effective Management MPAs. Australian Institute of Marine Science and the IUCN Marine Program Publication. 68pp. Chong-Seng, K.M., Nash, K.L., Bellwood, D.R., Graham, N.A.J. (2014) Macroalgal herbivory on recovering versus degrading coral reefs. Coral reefs 33, 409-419. Christianen, M.J.A., Govers, L.L., Bouma, T.J., Kiswara, W., Roelofs, J.G.M., Lamers, L.P.M., van Katwijk, M.M. (2012). Marine megaherbivore grazing may increase seagrass tolerance to high nutrient loads. Journal of Ecology 100, 546-560. Clarke, K.R. (1993). Non‐parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18, 117-143. Clarke, K.R., Gorley, R.N. (2006). PRIMER v6: User manual/tutorial. PRIMER-E, Plymouth, UK. Clarke, K.R., Gorley, R.N. (2015). PRIMER v6. Plymouth, UK: Plymouth Marine Laboratory Clifton, J., Etienne, M., Barnes, D.K., Barnes, R.S., Suggett, D.J., Smith, D.J. (2012). Marine conservation policy in Seychelles: Current constraints and prospects for improvement. Marine Policy 36, 823-831. Crawley, M. (2007). R Book. John Wiley and Sons Ltd, The Atrium, Southern Gate, Chichester, England. Cvitanovic, C., Bellwood, D.R. (2009). Local variation in herbivore feeding activity on an inshore reef of the Great Barrier Reef. Coral Reefs 28, 127-133. 87

Dayton, P.K., Thrush, S.F., Agardy, M.T., Hofman, R.J. (1995). Environmental effects of marine fishing. Aquatic conservation: marine and freshwater ecosystems 5, 205-232. Davis, K., Carlson, P.M., Lowe, C.G., Warner, R.R., Caselle, J.E. (2017). Parrotfish movement patterns vary with spatiotemporal scale. Marine Ecology Progress Series 577, 149-164. Dickens, L.C., Goatley, C.H., Tanner, J.K., Bellwood, D.R. (2011). Quantifying relative diver effects in underwater visual censuses. PLOS One 6, e18965. Dixson, D.L., Abrego, D., Hay, M.E. (2014). Chemically mediated behavior of recruiting corals and fishes: a tipping point that may limit reef recovery. Science 345, 892-897. Domingue, G., Payet, R., Shah, N.J. (2000). Marine Protected Areas in the republic of Seychelles. EAF document, Nairobi: UNEP. Dromard, C.R., Bouchon-Navaro, Y., Harmelin-Vivien, M., Bouchon, C. (2015). Diversity of trophic niches among herbivorous fishes on a Caribbean reef (Guadeloupe, Lesser Antilles), evidenced by stable isotope and gut content analyses. Journal of Sea Research 95, 124-131. Dulvy, N.K., Ellis, J.R., Goodwin, N.B., Grant, A., Reynolds, J.D., Jennings, S. (2004). Methods of assessing extinction risk in marine fishes. Fish and Fisheries 5, 255-276. Edwards, C.B., Friedlander, A.M., Green, A.G., Hardt, M.J., Sala, E., Sweatman, H.P., Williams, I.D., Zgliczynski, B., Sandin, S.A., Smith, J.E. (2014). Global assessment of the status of coral reef herbivorous fishes: evidence for fishing effects. Proceedings of the Royal Society B: Biological Sciences 281, 20131835. English, S.E., Wilkinson, C., Baker, V. (1997). Survey manual for tropical marine resources. Australian Institute of Marine Science, Townsville, Australia. Emslie, M.J., Cheal, A.J., Johns, K.A. (2014). Retention of habitat complexity minimizes disassembly of reef fish communities following disturbance: a large-scale natural experiment. PLOS One 9, e105384. Feare, C.J., van der Woude, J., Greenwell, P., Edwards, H.A., Taylor, J.A., Larose, C.S., Raines, K. (2017). Eradication of common mynas Acridotheres tristis from Denis Island, Seychelles. Pest Management Science 73, 295-304. Fong, C.R., Frias, M., Goody, N., Bittick, S.J., Clausing, R.J., Fong, P. (2018) 88

Empirical data demonstrates risk trade-offs between landscapes for herbivorous fish may promote reef resilience. Marine Environmental Research 133, 1-5. Fourqurean, J.W., Duarte, C.M., Kennedy, H., Marbà, N., Holmer, M., Mateo, M.A., Apostolaki, E.T., Kendrick, G.A., Krause-Jensen, D., McGlathery, K.J., Serrano, O. (2012). Seagrass ecosystems as a globally significant carbon stock. Nature Geoscience 5, 1–5. Fox, R.J., Bellwood, D.R. (2007). Quantifying herbivory across a coral reef depth gradient. Marine Ecology Progress Series 339, 49-59. Fox, R.J., Bellwood, D.R. (2008). Remote video bioassays reveal the potential feeding impact of the rabbitfish Siganus canaliculatus (f: Siganidae) on an inner-shelf reef of the Great Barrier Reef. Coral Reefs 27, 605-615. Fox, R.J, Bellwood, D.R. (2011). Unconstrained by the clock? Plasticity of diel activity rhythm in a tropical reef fish, Siganus lineatus. Functional Ecology 25, 1096-1105. Fox, R.J., Bellwood, D. R. (2013). Niche partitioning creates a unique ecosystem function for rabbitfishes (Perciformes, Siganidae) on coral reefs. Coral Reefs 32, 13-23. Fox, R.J., Sunderland, T.L., Hoey, A.S., Bellwood, D.R. (2009). Estimating functional roles in coral reef ecosystems: behaviour drives contrasting roles in herbivorous fishes (f: Siganidae). Marine Ecology Progress Series 385, 261- 269. Fretwell, S.D. (1969). On territorial behavior and other factors influencing habitat distribution in birds. Acta biotheoretica 19, 45-52. Friedlander, A.M., Brown, E.K., Jokiel, P.L., Smith, W.R., Rodgers, K. S. (2003). Effects of habitat, wave exposure, and marine protected area status on coral reef fish assemblages in the Hawaiian archipelago. Coral reefs 22, 291-305. Friedlander, A.M., Brown, E.K., Monaco, M.E. (2007). Defining reef fish habitat utilization patterns in Hawaii: comparisons between marine protected areas and areas open to fishing. Marine Ecology Progress Series 351, 221-233. Froese, R., Pauly, D. (2000). FishBase 2000: concepts, design and data sources. ICLARM, Los Banos, Laguna, Philippines. 344 p. Gell, F.R., Roberts, C.M. (2003). Benefits beyond boundaries: the fishery effects of marine reserves. Trends in Ecology and Evolution 18, 448-455. 89

Gell, F.R., Whittington, M.W. (2002). Diversity of fishes in seagrass beds in the Quirimba Archipelago, northern Mozambique. Marine and Freshwater Research, 53, 115-121. Gilby, B.L., Maxwell, P.S., Tibbetts, I.R., Stevens, T. (2015). Bottom-up factors for algal productivity outweigh no-fishing marine protected area effects in a marginal coral reef system. Ecosystems 18, 1056-1069. Gilmour, J.P., Smith, L.D., Heyward, A.J., Baird, A.H., Pratchett, M.S. (2013). Recovery of an isolated coral reef system following severe disturbance. Science 340, 69–71. Goatley, C.H., Bellwood, D.R. (2010). Biologically mediated sediment fluxes on coral reefs: sediment removal and off-reef transportation by the surgeonfish Ctenochaetus striatus. Marine Ecology Progress Series 415, 237-245. Goatley, C.H., Bellwood, D.R. (2012). Sediment suppresses herbivory across a coral reef depth gradient. Biology letters 8, 1016-1018. Graham, N.A.J, Jennings, S., MacNeil, M.A., Mouillot, D., Wilson, S.K. (2015). Predicting climate-driven regime shifts versus rebound potential in coral reefs. Nature 518, 94. Graham, N.A.J., Nash, K.L. (2013). The importance of structural complexity in coral reef ecosystems. Coral Reefs 32, 315-326. Graham, N.A.J, Wilson, S.K., Jennings, S., Polunin, N.V., Bijoux, J.P., Robinson, J. (2006). Dynamic fragility of oceanic coral reef ecosystems. Proceedings of the National Academy of Sciences 103, 8425- 8429. Grandcourt, E.M., Cesar, H.S. (2003). The bio-economic impact of mass coral mortality on the coastal reef fisheries of the Seychelles. Fisheries Research 60, 539-550. Green, A.L., Bellwood, D.R. (Eds.). (2009). Monitoring functional groups of herbivorous reef fishes as indicators of coral reef resilience: a practical guide for coral reef managers in the Asia Pacific region. IUCN, Gland, Switzerland. Halpern, B.S. (2003). The impact of marine reserves: Do reserves work and does reserve size matter? Ecological Applications, 13, S117–S137. Hanmer, J., White, J.W., Pawlik, J.R. (2017). Application of diet theory reveals context- dependent foraging preferences in an herbivorous coral reef fish. Oecologia 184, 127-137. 90

Heenan, A., Williams, I.D. (2013). Monitoring herbivorous fishes as indicators of coral reef resilience in American Samoa. PLOS One 8, e79604. Hicks, C.C., McClanahan, T.R. (2012). Assessing gear modifications needed to optimize yields in a heavily exploited, multi-species, seagrass and coral reef fishery. PLOS One 7, e36022. Hill, J., Wilkinson, C. (2004). Methods for ecological monitoring of coral reefs. A resource for managers. Australian Institute of Marine Science, Townsville, Australia. Hodgson, G., Maun, L., Shuman, C. (2003). Reef Check Survey Manual for Coral Reefs of the Indo Pacific, Hawaii, Atlantic/Caribbean, Red Sea and Arabian Gulf. Reef Check, Institute of the Environment, University of California, Los Angeles, California, USA. Hoey, A.S., Bellwood, D.R. (2009). Limited functional redundancy in a high diversity system: single species dominates key ecological process on coral reefs. Ecosystems 12, 1316-1328. Hoey, A.S., Bellwood, D.R. (2010). Cross-shelf variation in browsing intensity on the Great Barrier Reef. Coral Reefs 29, 499–508. Hoey, A.S., Bellwood, D.R. (2011). Suppression of herbivory by macroalgal density: a critical feedback on coral reefs? Ecology Letters 14, 267–273. Hoey, A.S., Brandl, S.J., Bellwood, D.R. (2013). Diet and cross-shelf distribution of rabbitfishes (f. Siganidae) on the northern Great Barrier Reef: implications for ecosystem function. Coral Reefs 32, 973-984. Honda, K., Nakamura, Y., Nakaoka, M., Uy, WH., Fortes, M.D. (2013). Habitat use by fishes in coral reefs, seagrass beds and mangrove habitats in the Philippines. PLOS One 8, e65735. Honda, K., Uy. W.H., Baslot, D.I., Pantallano, A.D.S., Nakamura, Y., Nakaoka, M. (2016). Diel habitat-use patterns of commercially important fishes in a marine protected area in the Philippines. Aquatic Biology 24, 163-174. Hughes, T.P., Rodrigues, M.J., Bellwood, D.R., Ceccarelli, D., Hoegh-Guldberg, O., McCook, L., Moltschaniwskyj, N., Pratchett, M.S., Steneck, R.S. Willis, B. (2007). Phase shifts, herbivory, and the resilience of coral reefs to climate change. Current Biology 17, 360-365. Humphries, A.T., McQuaid, C.D., McClanahan, T.R. (2015). Context-Dependent

Diversity- Effects of Seaweed Consumption on Coral Reefs in Kenya. PLOS One 10, e0144204. doi:10.1371/journal. pone.0144204. Jadot, C., Donnay, A., Acolas, M.L., Cornet. Y., Bégout Anras, M.L. (2006). Activity patterns, home- range size, and habitat utilization of Sarpa salpa (Teleostei: Sparidae) in the Mediterranean Sea. ICES Journal of Marine Science 63, 128- 139. Jennings, S., Grandcourt, E.M., Polunin, N.V.C. (1995). The effects of fishing on the diversity, biomass and trophic structure of Seychelles’ reef fish communities. Coral reefs 14, 225-235. Kamukuru, A.T. (2009). Trap fishery and reproductive biology of the whitespotted rabbitfish Siganus sutor (Siganidae), within the Dar es Salaam Marine Reserves, Tanzania. Western Indian Ocean Journal of Marine Science 8. Kaunda-Arara, B., Rose, G.A. (2004). Effects of marine reef National Parks on "fishery CPUE in coastal Kenya. Biological Conservation 118, 1–13. Kaunda-Arara, B., Rose, G.A., Muchiri, M.S., Kaka, R. (2003). Long-term trends in coral reef fish yields and exploitation rates of commercial species from coastal Kenya. Western Indian Ocean Journal of Marine Science 2, 105-116. Kelly, E.L., Eynaud, Y., Clements, S.M., Gleason, M., Sparks, R.T., Williams, I.D., Smith, J.E. (2016) Investigating functional redundancy versus complementarity in Hawaiian herbivorous coral reef fishes. Oecologia 182, 1151-1163. Kimirei, I.A., Nagelkerken, I., Griffioen, B., Wagner, C., Mgaya, Y.D. (2011). Ontogenetic habitat use by mangrove/seagrass-associated coral reef fishes shows flexibility in time and space. Estuarine, Coastal and Shelf Science 92, 47-58. Kimirei, I.A., Nagelkerken, I., Slooter, N., Gonzalez, E.T., Huijbers, C.M., Mgaya, Y. D., Rypel, A.L. (2015). Demography of fish populations reveals new challenges in appraising juvenile habitat values. Marine Ecology Progress Series 518, 225-237. Kramer, D.L., Chapman, M.R. (1999). Implications of fish home range size and relocation for marine reserve function. Environmental Biology of Fishes 55, 65–79. Kuiter, R. (1993). Coastal fishes of south-eastern Australia. University of Hawaii Press, Honolulu. 92

Lallemand, P. (2014). Economic study on major trends in the tuna industry and its impact on the Seychelles economy over the 5-year period, 2009-2013. SF/215/50. Smart Fish Programme, IOC. Lapointe, B.E. (1997). Nutrient thresholds for bottom‐up control of macroalgal blooms on coral reefs in Jamaica and southeast Florida. Limnology and Oceanography 42, 1119-1131. Laroche, J., Baran, E., Rasoanandrasana, N.B. (1997). Temporal patterns in a fish assemblage of a semiarid mangrove zone in Madagascar. Journal of Fish Biology 51, 3-20. Lechowicz, M.J. (1982). The sampling characteristics of electivity indices. Oecologia 52, 22–30. Ledlie, M.H., Graham, N.A.J., Bythell, J.C., Wilson, S.K., Jennings, S., Polunin, N.V.C., Hardcastle, J. (2007). Phase shifts and the role of herbivory in the resilience of coral reefs. Coral Reefs 26, 641-653. Lefèvre, C.D., Bellwood, D.R. (2010) Seasonality and dynamics in coral reef macroalgae: variation in condition and susceptibility to herbivory. Marine biology 157, 955-965. Lindstedt, S.L., Miller, B.J., Buskirk, S.W. (1986). Home range, time, and body size in mammals. Ecology 67, 413–418. Loffler, Z., Bellwood, D.R., Hoey, A.S. (2015). Among-habitat algal selectivity by browsing herbivores on an inshore coral reef. Coral Reefs 34, 597-605. Lugendo, B.R., Nagelkerken, I., van Der Velde, G., Mgaya, Y.D. (2006). The importance of mangroves, mud and sand flats, and seagrass beds as feeding areas for juvenile fishes in Chwaka Bay, Zanzibar: gut content and stable isotope analyses. Journal of Fish Biology 69, 1639-1661. Lugendo, B.R., Pronker, A., Cornelissen, I., de Groene, A., Nagelkerken, I., Dorenbosch, M., van der Velde, G., Mgaya, Y.D. (2005). Habitat utilization by juveniles of commercially important fish species in a marine embayment in Zanzibar, Tanzania. Aquatic Living Resources 18, 149-158. Lundberg, J., Moberg, F. (2003). Mobile link organisms and ecosystem functioning: implications for ecosystem resilience and management. Ecosystems 6, 0087- 0098. MacNeil, M.A., Graham, N.A.J., Cinner, J.E., Wilson, S.K., Williams, I.D., Maina, J.,

McClanahan, T.R. (2015). Recovery potential of the world's coral reef fishes. Nature 520, 341. Madin, E., Precoda, K., Harborne, A., Atwood, T.B., Roelfsema, C.M., Luiz, O. (2019). Multi-trophic species interactions shape seascape-scale coral reef vegetation patterns. Frontiers in Ecology and Evolution 7, 102 Marshell, A., Mumby, P.J. (2012). Revisiting the functional roles of the surgeonfish Acanthurus nigrofuscus and Ctenochaetus striatus. Coral Reefs 31, 1093- 1101. Marshell, A., Mumby, P.J. (2015). The role of surgeonfish (Acanthuridae) in maintaining algal turf biomass on coral reefs. Journal of Experimental Marine Biology and Ecology 473, 152-160. Mbaru, E.K., Sigana, D., Ruwa, R.K., Mueni, E.M., Ndoro, C.K., Kimani, E.N., Kaunda-Arara, B. (2018). Experimental evaluation of influence of FADs on community structure and fisheries in coastal Kenya. Aquatic Living Resources 31, 6. McClanahan, T.R., Mangi, S.C. (2004). Gear‐based management of a tropical artisanal fishery based on species selectivity and capture size. Fisheries Management and Ecology 11, 51-60. McCormick, M.I. (1994). Comparison of field methods for measuring surface topography and their associations with a tropical reef fish assemblage. Marine Ecology Progress Series. Oldendorf 112, 87-96. Mellin, C., Aaron MacNeil, M., Cheal, A.J., Emslie, M.J., Julian Caley, M. (2016). Marine protected areas increase resilience among coral reef communities. Ecology letters 19, 629-637. Meyer, C.G., Holland, K.N. (2005). Movement patterns, home range size and habitat utilization of the bluespine unicornfish, Naso unicornis (Acanthuridae) in a Hawaiian marine reserve. Environmental Biology of Fishes 7, 201–210. Miller, A.D., Roxburgh, S.H., Shea, K. (2011). How frequency and intensity shape diversity-disturbance relationship. Proceedings of the National Academy of Sciences, USA 108, 5643–5648. Moran, P.A.P. (1950). Notes on continuous stochastic phenomena. Biometrika 37, 17–23. Mumby, P.J., Bejarano, S., Golbuu, Y., Steneck, R.S., Arnold, S.N., Van Woesik,

R., Friedlander, A.M. (2013). Empirical relationships among resilience indicators on Micronesian reefs. Coral Reefs 32, 213-226. Mumby, P.J., Harborne, A.R. (2010). Marine reserves enhance the recovery of corals on Caribbean reefs. PLOS One 5, e8657. Mumby, P.J., Hastings, A., Edwards, H.J. (2007). Thresholds and the resilience of Caribbean coral reefs. Nature 450, 98. Mumby, P.J., Steneck, R.S. (2008). Coral reef management and conservation in light of rapidly evolving ecological paradigms. Trends in Ecology and Evolution 23, 555-563. Mumby, P.J., Wabnitz, C.C. (2002). Spatial patterns of aggression, territory size, and harem size in five sympatric Caribbean parrotfish species. Environmental Biology of Fishes 63, 265-279. Muñoz, R.C., Motta, P.J. (2000). Interspecific aggression between two parrotfishes (Sparisoma, scaridae) in the Florida Keys. Copeia 2000, 674-683. Murchie, K.J., Schwager, E., Cooke, S.J., Danylchuk, A., Danylchuk, S., Goldberg, T., Suski, C., Philipp, D. (2010). Spatial ecology of juvenile lemon sharks (Negaprion brevirostris) in tidal creeks and coastal waters of Eleuthera, The Bahamas. Environmental Biology of Fishes 89, 95–104. Nagelkerken, I., Roberts, C.M., van der Velde, G., Dorenbosch, M., van Riel, M.C., Cocheret de la Morinière, E., Niehuis, P.H. (2002) How important are mangroves and seagrass beds for coral-reef fish? The nursery hypothesis tested on an island scale. Marine Ecology Progress Series 244, 299−305. National Statistics Bureau. (2018) Seychelles in Figures, 2017 Edition. National Statistics Bureau, Victoria, Seychelles pp 1. Nyström, M., Folke, C. (2001). Spatial resilience of coral reefs. Ecosystems 4, 406– 417. Olds, A.D., Connolly, R.M., Pitt, K.A., Maxwell, P.S. (2012). Habitat connectivity improves reserve performance. Conservation Letters 5, 56−63. Olds, A.D., Albert, S., Maxwell, P.S., Pitt, K.A., Connolly, R.M. (2013). Mangrove-reef connectivity promotes the effectiveness of marine reserves across the western Pacific. Global Ecology and Biogeography 22, 1040–1049. Orth, R.J., Carruthers, T.J.B., Dennison, W.C., Duarte, C.M., Fourqurean, J.W., Heck., Jr. K.L., Randall., Hughes, A., Kendrick, G.A., Kenworthy, W., Olyarnik, S., Short, F.T., Waycott, M., Williams, S.L. (2006). A global crisis for seagrass 95

ecosystems. Bioscience 56, 987–996. Owen-Smith, N., Fryxell, J.M., Merrill, E.H. (2010). Foraging theory upscaled: the behavioural ecology of herbivore movement. Philosophical Transactions of the Royal Society B 365, 2267–2278. Paddack, M.J., Cowen, R.K., Sponaugle, S. (2006). Grazing pressure of herbivorous coral reef fishes on low coral-cover reefs. Coral Reefs 25, 461–472 Paul, V.J., Nelson, S.G., Sanger, H.R. (1990). Feeding preferences of adult and juvenile rabbitfish Siganus argenteus in relation to chemical defenses of tropical seaweeds. Marine Ecology Progress Series 60, 23-34. Pelletier, D., García-Charton, J.A., Ferraris, J., David, G., Thébaud, O., Letourneur, Y., Claudet, J., Amand, M., Kulbicki, M., Galzin, R. (2005). Designing indicators for assessing the effects of marine protected areas on coral reef ecosystems: a multidisciplinary standpoint. Aquatic Living Resources 18, 15- 33. Pillans, R.D., Babcock, R.C., Thomson, D.P., Haywood, M.D.E., Downie, R.A., Vanderklift, M.A., Rochester, W.A. (2017). Habitat effects on home range and schooling behaviour in a herbivorous fish (Kyphosus bigibbus) revealed by acoustic tracking. Marine and Freshwater Research 68, 1454-1467. Plass-Johnson, J.G., McQuaid, C.D., Hill, J.M. (2013). Stable isotope analysis indicates a lack of inter-and intra-specific dietary redundancy among ecologically important coral reef fishes. Coral Reefs 32, 429-440. Polunin, N.V.C., Harmelin‐Vivien, M., Galzin, R. (1995). Contrasts in algal food processing among five herbivorous coral‐reef fishes. Journal of Fish Biology 47, 455-465. Polunin, N.V.C., Klumpp, D.W. (1992). Algal food supply and grazer demand in a very productive coral-reef zone. Journal of Experimental Marine Biology and Ecology 164, 1-15. Puk, L.D., Ferse, S.C., Wild, C. (2016). Patterns and trends in coral reef macroalgae browsing: a review of browsing herbivorous fishes of the Indo- Pacific. Reviews in Fish Biology and Fisheries 26, 53-70. Putra, R.D., Suryanti, A., Kurniawan, D., Pratomo, A., Irawan, H., Raja, T.S., Kurniawan, R., Pratama, G., (2018). Responses of Herbivorous Fishes on Coral Reef Cover in Outer Island Indonesia (Study Case: Natuna Island). In E3S Web of Conferences (Vol. 47, p. 04009). EDP Sciences. 96

Randall, J.E. (1956). A revision of the surgeon fish genus Acanthurus. Pacific Science 10, 154-235. Randall, J.E., Allen, G.R., Steene, R.C. (1997). Fishes of the Great Barrier Reef and Coral Sea. University of Hawaii Press, Honolulu, HI, 507 pp. Randall, J.E., Clements, K.D. (2001). Second revision of the surgeonfish genus Ctenochaetus (Perciformes: Acanthuridae), with description of two new species. Indo-Pacific Fishes, Bishop Museum (Honolulu) 32, 1–33. Rasher, D.B., Hay, M.E. (2010). Chemically rich seaweeds poison corals when not controlled by herbivores. Proceedings of the National Academy of Sciences, USA 107, 9683–9688. Rasher, D.B., Hoey, A.S., Hay, M.E. (2013). Consumer diversity interacts with prey defenses to drive ecosystem function. Ecology 94, 1347–1358. Richardson, L.E., Graham, N.A.J, Pratchett, M.S., Hoey, A.S. (2017). Structural complexity mediates functional structure of reef fish assemblages among coral habitats. Environmental Biology of Fishes 100, 193-207. Robinson, J., Graham, N.A.J., Grüss, A., Gerry, C., Bijoux, J. (2017). Fishery benefits from exploiting spawning aggregations not solely dependent on enhanced fish density. African Journal of Marine Science 39, 269-278. Robinson, J., Samoilys, M.A., Grandcourt, E., Julie, D., Cedras, M., Gerry, C. (2011). The importance of targeted spawning aggregation fishing to the management of Seychelles’ trap fishery. Fisheries Research 112, 96-103. Robinson, J.P., Wilson, S.K., Robinson, J., Gerry, C., Lucas, J., Assan, C., Govinden, R., Jennings, S., Graham, N.A. (2019). Productive instability of coral reef fisheries after climate-driven regime shifts. Nature ecology and evolution 3, 183. Roff, G., Zhao, J.X., Mumby, P.J. (2015). Decadal‐scale rates of reef erosion following El Niño‐related mass coral mortality. Global Change Biology 21, 4415-4424. Russ, G.R. (1984). Distribution and abundance of herbivorous grazing fishes in the central Great Barrier Reef. I. Levels of variability across the entire continental shelf. Marine Ecology Progress Series 20, 23-34. Russ, G.R., Alcala, A.C. (1996). Do marine reserves export adult fish biomass? Evidence from Apo Island, central Philippines. Marine Ecology Progress Series 132, 1-9. 97

Sale, P.F., Cowen, R.K., Danilowicz, B.S., Jones, G.P., Kritzer, J.P., Lindeman, K.C., Planes, S., Polunin, N.V., Russ, G.R., Sadovy, Y.J., Steneck, R.S. (2005). Critical science gaps impede use of no-take fishery reserves. Trends in Ecology and Evolution 20, 74-80. Samoilys, M.A., Kanyange, N., Macharia, D., Maina, G.W., Robinson, J. (2013). Dynamics of rabbitfish (Siganus sutor) spawning aggregations in southern Kenya. Reef Fish Spawning Aggregations in the Western Indian Ocean: Research for Management. WIOMSA/SIDA/SFA/CORDIO. WIOMSA Book Series 13, 33-45. Samoilys, M.A., Maina, G.W., Osuka, K.E. (2011) Artisanal fishing gears of the Kenyan coast. Mombasa CORDIO/USAID. 36pp. Samoilys, M.A., Roche, R., Koldewey, H., Turner, J. (2018). Patterns in reef fish assemblages: Insights from the Chagos Archipelago. PLOS One 13, e0191448. Schlegel, B. (2018). Predicted values and discrete changes for GLM. CRAN Schoener, T.W. (1971). Theory of feeding strategies. Annual Review of Ecology and Systematics 2, 369-404. Seychelles Fishing Authority. (2016) Seychelles artisanal fisheries statistics. Seychelles Fishing Authority Technical Report. Government of Seychelles, Seychelles. Simpson, S.J., Sibly, R.M., Lee, K.P., Behmer, S.T., Raubenheimer, D. (2004). Optimal foraging when regulating intake of multiple nutrients. Animal Behaviour 68, 1299-1311. Soliman, V.S., Mendoza, A.B., Yamaoka, K. (2008). Seaweed-associated fishes of Lagonoy Gulf in bicol, the Philippines with emphasis on Siganids (Teleoptei: Siganidae). Kuroshio Science 2, 67–72. Steneck, R.S. (1988) Herbivory on coral reefs: a synthesis. In: Proceedings of the sixth International Coral Reef Symposium Publication, 1:37-49. Steneck, R.S., Arnold, S.N., Mumby, P.J. (2014). Experiment mimics fishing on parrotfish: insights on coral reef recovery and alternative attractors. Marine Ecology Progress Series 506, 115-127. Stocks, J.R., Gray, C.A., Taylor, M.D. (2015). Out in the wash: spatial ecology of a temperate marine shallow rocky-reef species derived using acoustic telemetry. Marine and Freshwater Research 66, 559-571. 98

Strain, E.M., Edgar, G.J., Ceccarelli, D., Stuart‐Smith, R. D., Hosack, G.R., Thomson, R.J. (2019). A global assessment of the direct and indirect benefits of marine protected areas for coral reef conservation. Diversity and Distributions 25, 9-20. Streit, R.P., Hoey, A.S., Bellwood, D.R. (2016). Feeding characteristics reveal functional distinctions among browsing herbivorous fishes on coral reefs. Coral reefs 34, 1037-1047. Suding, K.N., Gross, K.L., Houseman, G.R. (2004). Alternative states and positive feedbacks in restoration ecology. Trends in Ecology and Evolution 19, 46–53. Tootell, J.S., Steele, M.A. (2016). Distribution, behavior, and condition of herbivorous fishes on coral reefs track algal resources. Oecologia 181, 13-24. UNDP. (2010) Strengthening Seychelles’ protected area system through NGO management modalities 2010. /http://www.thegef.org/gef/node/3957S. Topor, Z.M., Rasher, D.B., Duffy, J.E., Brandl, S.J. (2019). Marine protected areas enhance coral reef functioning by promoting fish biodiversity. Conservation Letters e12638 Unsworth, R.K.F., Cullen, L.C. (2010). Recognising the necessity for Indo-Pacific seagrass conservation. Conservation Letters 3, 63–73. Valentine, J.F., Heck, Jr. K.L., Busby, J., Webb, D. (1997). Experimental evidence that herbivory can increase shoot density in a subtropical turtlegrass (Thalassia testudinum) meadow. Oecologia 112, 193–200. Vanderploeg, H.A., Scavia, D. (1979). Calculation and use of selectivity coefficients of feeding: zooplankton grazing. Ecological Modelling 7, 135–149. Vergés, A., Alcoverro, T., Romero, J. (2011). Plant defences and the role of epibiosis in mediating within-plant feeding choices of seagrass consumers Oecologia 166, 381-390. Vergés, A., Becerro, M.A., Alcoverro, T., Romero, J. (2007). Experimental evidence of chemical deterrence against multiple herbivores in the seagrass Posidonia oceanica. Marine Ecology Progress Series 343, 107-114. Villegas-Rios, D., Alos, J., March, D., Palmer, M., Mucientes, G., Saborido-Rey, F. (2013). Home range and diel behaviour of the ballan wrasse, Labrus bergylta, determined by acoustic telemetry. Journal of Sea Research 80, 61–71. Walther, G.R., Post, R., Convey, P., Menzel, A., Parmesan., Beebee T.J.C.,

Fromentin J-M., Hoegh-Guldberg, O., Bairlein, F. (2002). Ecological responses to recent climate change. Nature 416, 389−395. Waycott, M., Longstaff, B.J., Mellors, J. (2005). Seagrass population dynamics and water quality in the Great Barrier Reef region: A review and future research directions. Marine Pollution Bulletin 51, 343–350. Welsh, J.Q., Bellwood, D.R. (2012a). How far do schools of roving herbivores rove? A case study using Scarus rivulatus. Coral Reefs 31, 991–1003. Welsh, J.Q., Bellwood, D.R. (2012b). Spatial ecology of the steephead parrotfish (Chlorurus microrhinos): an evaluation using acoustic telemetry. Coral Reefs 31, 55–65. Welsh, J.Q., Bellwood, D.R. (2014). Herbivorous fishes, ecosystem function and mobile links on coral reefs. Coral Reefs 33, 303-311. Whalen, M.A., Duffy, J.E., Grace, J.B. (2013). Temporal shifts in top-down vs. bottom-up control of epiphytic algae in a seagrass ecosystem. Ecology 94, 510–520. Wilkinson, C.R. (2004). Status of Coral Reefs of the World: 2004. Australian Institute of Marine Science, Cape Ferguson, Qld., Volume 1 301 pp. Williams, D.M.B., Hatcher, A.I. (1983). Structure of fish communities on outer slopes of inshore, mid-shelf and outer shelf reefs of the Great Barrier Reef. Marine Ecology Progress Series 10, 239–250. Williams, I.D., Polunin, N.V., Hendrick, V.J. (2001). Limits to grazing by herbivorous fishes and the impact of low coral cover on macroalgal abundance on a coral reef in Belize. Marine Ecology Progress Series 222, 187-196. Wismer, S., Hoey, A.S., Bellwood, D.R. (2009). Cross-shelf benthic community structure on the Great Barrier Reef: relationships between macroalgal cover and herbivore biomass. Marine Ecology Progress Series 376, 45–54. Woodland, D.J. (1990). Revision of the fish family Siganidae with description of two new species and comments on distribution and biology. Indo-Pacific Fishes 19, 1–136. Wu, R.S.S. (1984). The feeding habits of seven demersal fish species in a subtropical estuary. Asian Marine Biology 1, 17–26. Zemke-White, L.W., Choat, J., Clements, K. (2002). A re-evaluation of the diel

100

feeding hypothesis for marine herbivorous fishes. Marine Biology 141, 571- 579.

101

Appendices

Appendix A: Ethics Approval Letter

102

103

Appendix B: Chapter 2, Supplementary material

Table S2.1. Summary displaying the Reef type (Carbonate (C) vs. Granitic (G)), and the current Management status (Protected (P) vs. Unprotected (U). Also displayed are the percentage benthic habitat types per site (± SE). Benthic habitat was identified and classified as fleshy macroalgae (FMA), epilithic algal matrix (EAM), live coral (C), dead coral (DC) coral rubble (CR), sand (S), and seagrass (SG). The Rugosity ratio (± SE) of each site is also shown.

Site Reef type Management FMA EAM LC DC CR S SG Rugosity ratio A1 G U 0 14±0.9 63±1.2 0.5±0.01 17.5±0.5 5±0.1 0 2.5±0.03 A2 C P 0 7±0.07 71.5±1.8 3.5± 14±0.2 4±0.1 0 2.9±0.06 A3 G U 0 1±0.01 15±0.6 0.5±0.01 42±1.5 41.5±1.0 0 2.0±0.05 A4 C U 0 0.5±0.01 30±1.0 3±0.02 58.5±1.6 8±0.2 0 1.4±0.04 B1 C P 0 0.5±0.01 14±0.9 2±0.02 21.5±1.2 62±1.5 0 1.6±0.03 B2 C P 42.5±1.1 5.5±0.2 0 0.5±0.01 14.5±0.5 37±1.0 0 1.2±0.02 B3 C P 0 3±0.05 10.5±0.6 3±0.3 62±1.8 21.5±1.0 0 1.2±0.02 B4 C P 32±1.0 5±0.09 0 2±0.07 6.5±0.3 54.5±1.1 0 1.2±0.03 C1 G U 0 3.5±0.1 17.5±0.9 1±0.01 26.5±1.0 51.5±1.2 0 1.9±0.04 C2 C U 0 0 19±1.0 3±0.02 36.5±1.0 41.5±1.2 0 3.3±0.05 D1 G U 0 7.5±0.8 30.5±1.0 1.5±0.01 4.5±0.04 56±1.8 0 1.6±0.05 D2 C U 0 6±0.2 18±0.8 0 52±0.4 24±1.2 0 1.6±0.03 D3 C U 37.5±1.0 8±0.2 0.5±0.04 0 29±0.8 25±1.2 0 1.2±0.04 D4 G P 0 2±0.03 24±1.2 0.5±0.01 37±0.6 36.5±1.4 0 2.6±0.05 D5 C U 0 40.5±1.2 20±1.0 7.5±0.06 14.5±0.07 17.5±1.0 0 1.6±0.04 D6 C U 0 17.5±0.08 0 0 0 24±1.3 58.5±0.1 2.4±0.05

104

E1 C P 44±1.2 0 0 0 14±0.07 42±1.1 0 1.5±0.07 E2 C P 60±1.3 0 0 0 37.5±0.5 2.5±0.02 0 1.8±0.04 E3 C P 44.5±1.2 0 20±1.0 2±0.01 29±0.9 4.5±0.8 0 2.3±0.06 E4 G P 2±0.03 0 3.5±0.1 0 71±1.0 23.5±1.0 0 1.5±0.03 F1 C P 0 0 1±0.1 0 89.5±1.5 9.5±0.2 0 1.3±0.05 F2 C P 57.5±1.6 0 0 0 38±1.0 4.5±0.2 0 2.9±0.03 F3 C P 5±0.1 0.5±0.01 0.5±0.01 0 82±1.2 12±0.4 0 1.2±0.02 F4 C P 40±1.2 0 0 0 1±0.01 59±1.3 0 1.2±0.05

105

Table S2.2 Summary of the estimates of components of variation demonstrating the percentage contribution of each variable and factor on the mean biomass of herbivorous fish functional groups.

Source Overall Algal browsers Detritivores Excavators Scrapers Est. % Est. % Est. % Est. % Est. % Coral rubble 63.2 9.5 0.3 4.2 0.1 1.7 0.2 3.7 0.3 22.4 Dead coral 61.1 9.1 0.3 4.2 0.1 1.9 0.3 6.4 0 2.8 Epilithic algal matrix 16.4 2.4 0.2 3.8 0.4 8 0 0.5 0 2.6 Fleshy macroalgae 3.3 0.5 1.8 29.3 0.1 3.3 0.1 1.2 0 2.6 Live coral 151 22.6 0.2 3.8 0.4 7 0.5 10 0.1 5.3 Rugosity ratio 35.5 5.5 0.3 4.3 0.1 2.7 0.5 9.7 0.1 5.7 Sand 4.7 0.7 0.3 4.3 0 0.4 0 0.9 0 3.6 Seagrass 23.7 3.5 0.3 4.3 0.4 7 0.3 5.8 0.1 6.5 Island 10.1 1.5 0.1 2.2 0.1 3.3 0.2 4.2 0.1 10.6 Management 12.3 1.8 0.4 5.8 0.3 8.2 0.2 4.5 0.1 3.8 Reef type 28.9 4.3 0.4 6 0.3 8.2 0.4 7.5 0 3.1 Island × 5.6 0.8 0.2 4.7 0.1 3 0.3 5.6 0.1 7 Management Island × Reef type 14.8 2.2 0.2 4.5 0.1 1.6 0.1 2.4 0 0.9 Residuals 238 35.6 0.9 18.6 1.2 43.7 1.8 37.6 0.3 23.1

Table S2.3 Summary of the estimates of components of variation demonstrating the percentage contribution of each variable on the mean biomass of rabbitfish species.

Source Overall S. argenteus S. corallinus S. stellatus S. sutor Est. % Est. % Est. % Est. % Est. % Coral rubble 52.8 3.2 0.14 2.88 0.0 0.4 0.0 2.8 0.1 0.9 Dead coral 83.5 5.0 0.37 7.49 0.1 4.0 0.2 11.9 0.1 1.2 Epilithic algal matrix 125.7 7.5 0.22 4.47 0.3 13.6 0.2 11.8 0.8 10.2 Fleshy macroalgae 117.7 7.1 0.99 20.23 0.1 4.3 0.1 5.2 2.2 26.9 Live coral 207.3 12.4 0.50 10.22 0.6 23.5 0.2 12.0 0.1 1.4 Rugosity ratio 52.8 3.2 0.17 3.50 0.0 0.5 0.1 4.2 0.1 1.5 Sand 61.0 3.7 0.05 0.99 0.0 1.9 0.0 2.7 0.2 1.9 Seagrass 78.0 4.7 0.21 4.37 0.0 0.9 0.1 6.1 1.5 18.8 Island 75.4 4.5 0.07 1.47 0.1 3.1 0.1 8.9 0.1 1.7 Management 18.2 1.1 0.03 0.70 0.1 2.3 0.1 7.5 0.4 5.5

106

Reef type 92.6 5.6 0.07 1.43 0.0 0.5 0.1 4.7 0.2 2.7 Island × Management 154.2 9.2 0.54 11.07 0.1 3.9 0.1 6.8 0.1 0.8 Island × Reef type 69.4 4.2 0.09 1.80 0.0 1.6 0.0 1.8 0.1 1.0 Residuals 480.2 28.8 1.43 29.37 1.0 39.6 0.2 13.6 2.0 25.5

107

Table S2.4 Summary of the homogeneity of multiple dispersions (PERMDISP) of tested factors on the biomass of herbivore functional groups. Significant interactions are highlighted in bold.

Source Overall Algal browsers Detritivores Excavators Scrapers F P(perm) F P(perm) F P(perm) F P(perm) F P(perm) Island 0.67 0.96 1.42 0.45 0.54 0.96 0.32 0.99 0.67 0.82 Management 0.56 0.54 0.98 0.43 1.20 0.34 0.07 0.85 1.44 0.23 Reef type 0.83 0.10 3.79 0.10 2.78 0.13 4.00 0.08 0.07 0.83

Table S2.5 Summary of the homogeneity of multiple dispersions (PERMDISP) of tested factors on the biomass of rabbitfish species. Significant interactions are highlighted in bold.

Source Overall S. argenteus S. corallinus S. stellatus S. sutor F P(perm) F P(perm) F P(perm) F P(perm) F P(perm) Island 2.71 0.18 4.89 0.07 2.16 0.18 3.17 0.49 3.31 0.15 Management 0.01 0.96 1.56 0.47 0.35 0.54 1.40 0.55 1.03 0.54 Reef type 18.10 0.01 14.36 0.01 3.53 0.08 0.17 0.84 9.74 0.02

108

Table S2.6 Summary of Eigenvalues based on the PCA plot of herbivore functional groupings.

PC Eigenvalues %Variation Cum.%Variation 1 2.62 32.8 32.8 2 2.07 25.9 58.7 3 1.35 16.9 75.6 4 0.766 9.6 85.1 5 0.665 8.3 93.5

Table S2.7 Summary of Eigenvalues based on the PCA plot of rabbitfish species groupings.

PC Eigenvalues %Variation Cum.%Variation 1 5.26 49.8 49.8 2 2.22 21.0 70.8 3 1.08 10.2 81.0 4 0.842 8.0 89.0 5 0.656 6.2 95.2

Figure S2.1 nMDS plot demonstrating the variability in biomass distribution of rabbitfish species between the two reef types (carbonate vs. granitic).

109

Appendix C: Chapter 3, Supplementary material

Table S3.1 Pairwise comparisons between sites at each time interval for S. argenteus. Significant interactions are highlighted in bold.

Within Level ‘T1’ of factor Time Groups (Sites) P (perm) A3, A4 0.001 A3, G2 0.001 A4, G2 0.121 Within Level ‘T2’ of factor Time A3, A4 0.001 A3, G2 0.001 A3, B2 0.05 A3, E 0.01 A3, F 0.05 A4, G2 0.67 A4, B2 0.001 A4, E 0.001 G2, B2 0.001 G2, E 0.001 G2, F 0.001 B2, E 0.001 B2, F 0.62 E, F 0.03 Within Level ‘T3’ of factor Time A4, G2 0.001 A4, B2 0.001 A4, F 0.02 A4, A2 0.01 A4, G1 0.05 G2, B2 0.001 G2, F 0.001 G2, A2 0.001 G2, G1 0.001 B2, F 0.53 B2, A2 0.001 B2, G1 0.001 F, A2 0.001 F, G1 0.001 A2, G1 0.001 Within Level ‘T4’ of factor Time G2, B2 0.001 G2, F 0.001 G2, A2 0.001 G2, G1 0.001 G2, D2 0.001 G2, D3 0.001 B2, F 0.36 B2, A2 0.001 B2, G1 0.001 B2, D1 0.18 B2, D3 0.001 F, A2 0.001 F, G1 0.001 F, D2 0.001

110

F, D3 0.001 A2, G1 0.001 A2, D2 0.001 A2, D3 0.68 G1, D2 0.001 G1, D3 0.001 D2, D3 0.001 Within Level ‘T5’ of factor Time A4, G2 0.12 A4, E 0.001 A4, G1 0.79 A4, D2 0.001 A4, D3 0.001 A4, B3 0.001 A4, D4 0.20 G2, E 0.001 G2, G1 0.30 G2, D2 0.001 G2, D3 0.001 G2, B3 0.001 G2, D4 0.68 E, G1 0.001 E, D2 0.02 E, D3 0.001 E, B3 0.001 E, D4 0.001 G1, D2 0.001 G1, D3 0.001 G1, B3 0.001 G1, D4 0.001 D2, D3 0.001 D2, B3 0.001 D2, D4 0.001 D3, B3 0.38 D3, D4 0.001 B3, D4 0.001 Within Level ‘T6’ of factor Time A4, G2 0.16 A4, E 0.001 A4, G1 0.02 A4, B3 0.001 A4, D4 0.05 G2, E 0.001 G2, G1 0.18 G2, B3 0.001 G2, D4 0.001 E, G1 0.001 E, B3 0.001 E, D4 0.001 G1., B3 0.001 G1, D4 0.05 B3, D4 0.001

111

Table S3.2 Pairwise comparisons between sites at each time interval for S. corallinus. Significant interactions are highlighted in bold.

Within Level ‘T1’ of factor Time Groups (Sites) P (perm) B2, A3 0.11 Within Level ‘T2’ of factor Time B1, A3 0.10 B1, D2 0.20 A3, D2 0.10 Within Level ‘T3’ of factor Time B1, A3 0.01 B1, D2 0.14 B1, C 0.25 B1, D3 0.43 A3, D2 0.06 A3, C 0.01 A3, D3 0.01 D2, C 0.10 D2, D3 0.34 C, D3 0.20 Within Level ‘T4’ of factor Time B2, A3 0.10 B2, D2 0.67 B2, C 0.34 B2, D3 0.10 B2, A1 0.31 B2, A2 0.34 B2, D1 0.08 A3, D2 0.27 A3, C 0.23 A3, D3 0.03 A3, A1 0.03 A3, A2 0.10 A3, D1 0.002 D2, C 0.30 D2, D3 0.10 D2, A1 0.36 D2, A2 0.66 D2, D1 0.13 C, D3 0.30 C, A1 0.35 C, A2 0.66 C, D1 0.15 D3, A1 0.51 D3, A2 0.25 D3, D1 0.27 A1, A2 0.30 A1, D1 0.54 A2, D1 0.39 Within Level ‘T5’ of factor Time B2, A3 0.002 B2, D2 0.12 B2, A1 0.03 B2, A2 0.002 B2, D1 0.003 A3, D2 0.12

112

A3, A1 0.04 A3, A2 0.20 A3, D1 0.001 D2, A1 0.31 D2, A2 0.15 D2, D1 0.20 A1, A2 0.71 A1, D1 0.07 A2, D1 0.01 Within Level ‘T6’ of factor Time B2, A3 0.01 B2, D2 0.06 B2, A1 0.47 A3, D2 0.10 A3, A1 0.23 D2, A1 0.26

Table S3.3 Pairwise comparisons between sites at each time interval for S. stellatus. Significant interactions are highlighted in bold.

Within Level ‘T1’ of factor Time Groups (Sites) P (perm) A4, G1 0.01 Within Level ‘T2’ of factor Time A4, G1 0.10 A4, A3 0.10 A4, D2 0.11 A4, D3 0.02 A4, C 0.02 G1, A3 0.10 G1, D2 0.11 G1, D3 0.06 G1, C 0.06 A3, D2 0.33 A3, D3 0.10 A3, C 0.06 D2, D3 0.07 D2, C 0.06 D3, C 0.33 Within Level ‘T3’ of factor Time G1, D3 0.24 G1, C 0.21 G1, G2 0.77 D3, C 0.10 D3, G2 0.33 C, G2 0.25 Within Level ‘T4’ of factor Time G1, B1 0.19 G1, C 0.03 B1, C 0.51 Within Level ‘T5’ of factor Time G1, A3 0.31 G1, D2 0.25 G1, B1 0.10 G1, B2 0.10 G1, C 0.33

113

G1, G2 0.14 A3, D2 0.10 A3, B1 0.31 A3, B2 0.68 A3, C 0.35 A3, G2 0.03 D2, B1 0.24 D2, B2 0.26 D2, C 0.90 D2, G2 0.01 B1, B2 0.10 B1, C 0.30 B1, G2 0.11 B2, C 0.35 B2, G2 0.11 C, G2 0.03 Within Level ‘T6’ of factor Time B1, D1 0.11 B1, G2 0.02 B1, A2 0.10 D1, G2 0.01 D1, A2 0.02 G2, A2 0.01

Table S3.4 Pairwise comparisons between sites at each time interval for S. sutor. Significant interactions are highlighted in bold.

Within Level ‘T1’ of factor Time Groups (Sites) P (perm) D2, E 0.05 D2, G1 0.05 E, G1 0.01 Within Level ‘T2’ of factor Time D2, E 0.26 D2, A3 0.03 D2, B2 0.70 D2, G2 0.23 E, A3 0.20 E, B2 0.24 E, G2 0.01 A3, B2 0.21 A3, G2 0.21 B2, G2 0.25 Within Level ‘T3’ of factor Time E, A3 0.04 E, G2 0.05 E, A2 0.004 E, A4 0.01 E, B1 0.20 A3, G2 0.02 A3, A2 0.001 A3, A4 0.002 A3, B1 0.14 G2, A2 0.03 G2, A4 0.04 G2, B1 0.32 A2, A4 0.002 114

A2, B1 0.14 A4, B1 0.19 Within Level ‘T4’ of factor Time D2, E 0.05 D2, B2 0.05 D2, G2 0.03 D2, A4 0.04 D2, B1 0.37 D2, D3 0.05 D2, F 0.20 E, B2 0.004 E, G2 0.004 E, A4 0.01 E, B1 0.05 E, D3 0.03 E, F 0.03 B2, G2 0.001 B2, A4 0.001 B2, B1 0.05 B2, D3 0.01 B2, F 0.01 G2, A4 0.01 G2, B1 0.03 G2, D3 0.01 G2, F 0.01 A4, B1 0.05 A4, D3 0.02 A4, F 0.02 B1, D3 0.01 B1, F 0.11 D3, F 0.05 Within Level ‘T5’ of factor Time E, A3 0.03 E, B2 0.05 E, G2 0.01 E, B3 0.03 E, D4 0.01 A3, B2 0.03 A3, G2 0.01 A3, B3 0.01 A3, D4 0.02 B2, G2 0.001 B2, B3 0.002 B2, D4 0.01 G2, B3 0.002 G2, D4 0.17 B3, D4 0.01 Within Level ‘T6’ of factor Time G1, A3 0.05 G1, B2 0.04 G1, G2 0.04 G1, A1 0.03 A3, B2 0.87 A3, G2 0.01 A3, A1 0.01 B2, G2 0.003 B2, A1 0.003 G2, A1 0.001

115

Table S3.5 Percent similarity of between and within rabbitfish species.

Average similarity between/within groups

S. corallinus S. stellatus S. sutor S. argenteus S. S. argenteus 43.3%

S. corallinus 44.3% 76.8%

S. stellatus 47.2% 69.7% 72.6%

S. sutor 42.3% 48.2% 46.4% 44.2%

100 CR DC 80 FMA LC 60 S SG

40 TA % % Cover

0 A1 A2 A3 A4 B1 B2 B3 C D1 D2 D3 D4 E F G1 G2 Sites

Fig S3.1 Benthic composition at each site; coral rubble (CR), dead coral (DC), Fleshy macroalgae (FMA), live coral (LC), sand (S), seagrass (SG) and turf algae (TA).

116

Figure S3.2 Average foraging rates (± standard error) for S. argenteus at each site, for each time period.

Figure S3.3 Average foraging rates (± standard error) for S. corallinus at each site, for each time period.

117

Figure S3.4 Average foraging rates (± standard error) for S. stellatus at each site, for each time period.

Figure S3.5 Average foraging rates (± standard error) for S. sutor at each site, for each time period.

118

35 D3D3 B3 C D3B1 D1 A1 25 B1 D2 A2 D2 A3 B2 A3 15 B2

G1 B2 D2

5 F B2 F G2 -35 -15 5 25 D2 45 65 -5

A3 A4 G1 A4 -15 A4 A3 G1

-25 PCO 2 (36.2% of total variation) total of (36.2% 2 PCO E E G2 -35 G2

D4 -45

-55 PCO 1 (54.2% of total variation)

Fig. S3.6 Average Principal coordinates analysis (PCO) of bites.min-1 of rabbitfishes at each site, displayed with the standard errors in both the x and y directions.

119

Fig. S3.7a. Scatter plots of the bites.min-1 of S. argenteus on each substrate at each site. Each dot represents a different site Solid lines represent significant slopes, whilst dotted lines represent non-significant slopes Key: FMA- Fleshy macroalgae, SG- Seagrass

120

Fig. S7b. Scatter plots of the bites.min-1 of S. corallinus on each substrate at each site. Each dot represents a different site Solid lines represent significant slopes, whilst dotted lines represent non-significant slopes Key: FMA- Fleshy macroalgae, SG- Seagrass

121

Fig. S3.7c. Scatter plots of the bites.min-1 of S. stellatus on each substrate at each site. Each dot represents a different site Solid lines represent significant slopes, whilst dotted lines represent non-significant slopes Key: FMA- Fleshy macroalgae, SG- Seagrass

122

Fig. S3.7d. Scatter plots of the bites.min-1 of S. sutor on each substrate at each site. Each dot represents a different site Solid lines represent significant slopes, whilst dotted lines represent non-significant slopes Key: FMA- Fleshy macroalgae, SG- Seagrass

123