FROM MEGAFAUNA TO MICROPLASTICS: UNDERSTANDING HABITAT USE REVEALS POTENTIAL THREATS TO INDONESIA’S MANTA RAYS Submitted by Elitza Germanov MSc, BSc (Hon.), Microbiology & Immunology, Dalhousie University, Canada This thesis is presented for the degree of Doctor of Philosophy, Environmental and Conservation Sciences, Murdoch University, Western Australia 2020 ii iii DECLARATION I declare that this thesis is my own account of my research and contains as its main content work which has not previously been submitted for a degree at any tertiary education institution. Elitza Germanov June 2020 iv ABSTRACT Manta rays Mobula alfredi are large, charismatic, filter-feeding elasmobranchs that are threatened with extinction. Limited information on populations within Indonesia exists, despite the growing popularity of manta ray tourism. As a wide ranging, coastal dwelling species, the main threats to manta rays come from fisheries. Within Indonesia, directed fisheries are now prohibited, but incidental capture still occurs and indirect threats from entanglement with fishing gear persist. While the impacts are more cryptic, if unregulated, tourism can also pose threats to manta rays through boat strikes and excessive SCUBA diving/snorkeling disrupting natural behaviors and impacting on important habitats. As filter-feeders, manta rays are susceptible to plastic ingestion in areas heavily contaminated by plastic pollution, like Indonesia. The potential threats posed by microplastics to filter-feeding megafauna are reviewed in Chapter 2. In this thesis, I have investigated the habitat use, demographics and movement patterns of the manta ray inhabiting the Nusa Penida Marine Protected Area (MPA) (Chapter 3) and the Komodo National Park (NP) (Chapter 4), Indonesia. Maximum likelihood techniques and a Markov movement analysis were used to analyze data logged online (www.mantamatcher.org) by researchers and citizen scientists. A larger number of manta rays were recorded within Komodo NP than in the Nusa Penida MPA (~1,100 vs. ~ 600). Within the Nusa Penida MPA, manta rays displayed diverse use of habitats with sites being identified as social areas, foraging grounds and potential nurseries. Sex-based differences in site affinity were also identified with females and immature males having higher site affinity than mature males. Within the Komodo NP there was less differentiation in habitat use between the sexes and between juvenile and older individuals, apart from one site, which had high re-sightings of immature individuals. Tourism, measured by the number of boats present at dive sites, increased by 60% in the Nusa Penida MPA (from 2012 to 2017) and by 34% in the Komodo NP (from 2014 to 2017). Further, persistent threats from artisanal fisheries remain, with 14% of manta rays v in the Nusa Penida MPA and 5% of those in the Komodo NP observed with line entanglements or permanent injuries from fishing gear. To assess the abundance of plastic marine debris, including small sized plastics – microplastics, I completed a series of quantitative visual and trawl surveys (n ≥ 6 per season) of feeding grounds for manta rays (Nusa Penida MPA and Komodo NP) and for whale sharks Rhincodon typus (Pantai Bentar, East Java) (Chapter 5). Manta ray feeding grounds were assessed during the two prevailing seasons, the north-west (wet) and south-east (dry) monsoons from 2016 – 2018. Manta ray egested material was collected opportunistically within the Nusa Penida MPA. Floating plastic marine debris was counted visually in transects, while a plankton net trawled for near surface plastics. Plastics were identified, measured and categorized visually from trawl and egested material samples. Linear models found significant seasonal and location differences in plastic abundance with the highest estimates being for the Nusa Penida MPA and coinciding with the wet season. I provided the first theoretical plastic ingestion estimates for both manta rays (up to ~ 63 pieces h-1) and whale sharks (~ 137 pieces h-1) within the Coral Triangle region and confirmed that manta rays both ingest and egest plastic. The results of the population studies (Chapters 3 and 4) were considered when making management recommendations for manta ray conservation in the region, including mandatory codes of conduct for dive operators, carrying capacity restrictions and management zoning. Knowledge gaps were identified and the value and shortcomings of citizen science data collection were outlined. Further, the plastic marine debris studies (Chapter 5) highlighted the potential threat to manta ray and whale shark populations and identified key times for prioritizing plastic marine debris prevention actions. I underline the need for further investigation into the health impacts of plastic ingestion and exposure to plastic associated pollutants, and provide future research directions for establishing the population level effects to these threatened species (Chapters 2, 5, 6 and Appendix I). Focusing research on these umbrella and flagship species could serve to increase public engagement, awareness and foster greater stewardship for overarching marine conservation issues. vi STATEMENT OF CONTRIBUTIONS SUPERVISION Professor Neil Loneragan, Environmental and Conservation Sciences, Centre for Sustainable Aquatic Ecosystems, Harry Butler Institute, Murdoch University. Professor Lars Bejder, Hawaiʻi Institute of Marine Biology, University of Hawaii, U.S.A. Professor Mike van Keulen, Environmental and Conservation Sciences, Centre for Sustainable Aquatic Ecosystems, Harry Butler Institute, Murdoch University. External supervisor: Dr. Andrea Marshall, Manta Ray Department, Marine Megafauna Foundation, USA. KEY COLLABORATORS Dr. I Gede Hendrawan, Faculty of Marine Sciences and Fisheries, Udayana University, Indonesia Dr. Simon Pierce, Whale Shark Department, Marine Megafauna Foundation, USA. Dr. Chris Rohner, Whale Shark Department, Marine Megafauna Foundation, USA. Dr. Delphine Chabanne, Aquatic Megafauna Research Unit, Centre for Sustainable Aquatic Ecosystems, Harry Butler Institute, Murdoch University. Dr. Ryan Admiraal, School of Mathematics and Statistics, Victoria University of Wellington, Wellington, New Zealand. vii GENERAL CONTRIBUTION I conceived the study and developed the overall approach and methodology with direction from my supervisors and collaborators. I was responsible for developing the individual research components, collecting data and/or directing data collection, processing and analyzing data, drafting and revising thesis chapters and resulting manuscripts. During the publication process, I submitted the initial manuscripts, organized input from collaborators, developed responses to reviewers and handled all editorial input. I developed and wrote the first and last Chapters (1 and 6) of the thesis with comments on structure and editorial input from Andrea Marshall (AM) and Neil Loneragan (NL). Requests for additional information and discussion that resulted from the examination process were included in these chapters. Thus, the thesis was much improved by the examiners Andrew Chin, Heidi Dewar, and Julia Reisser. Appendix I, which I wrote with help from Daniel Fernando and editorial input from Josh Stewart, was published as a section of a larger treatment on mobulid research. CONTRIBUTION TO DATA CHAPTERS Chapter 3 – Andrea Marshall (AM) and I conceived the study. Data collection was initiated by Aquatic Alliance under the direction of Helen Mitchell and Peter Bassett from 2012-2016. Starting 2016, data collection was continued by Marine Megafauna Foundation (MMF) under the direction of Lillian Ellevog (LE, 2016 – 2018), Lisa Auditore (LA, 2017), Rosie Cooper (RC, 2018) and I. I analyzed the data with Lars Bejder (LB), Delphine Chabanne (DC), and Simon Pierce (SP) advising on the analysis. I wrote the manuscript with contribution to drafting, critical review, and editorial input from LB, DC, Dharmadi, I Gede Hendrawan (IGH) AM, SP, Mike van Keulen and Neil Loneragan (NL). All authors critically reviewed the manuscript with final proofs. Gail Schofield, Queen Mary University of London, United Kingdom, Paul Brickle, South Atlantic Environmental Research Institute, Falkland Islands, and Joshua Stewart, University of California, San Diego, United States critically reviewed the manuscript for publication into Frontiers in Marine Science: Megafauna. viii Chapter 4 – AM and I conceived the study. I initiated data collection in 2012, which was continued by MMF under the direction of LE (2015 – 2018), LA (2017), RC (2018) and I. I analyzed the data with SP and LB advising on the analysis. I wrote the manuscript with contribution to drafting, critical review, and editorial input from SP, LB, AM and NL. Chapter 5 – AM and I conceived the study. Data collection was completed under the direction of LE (2016, 2018), LA (2017, 2018) and I with assistance from Wulandari Raka (WR), Janis Argeswara (JA), and Mahardika R. Himawan. I (2016-2017), WR (2018), and JA (2018) processed the samples. I analyzed the data with Christoph Rohner (CR) and Ryan Admiraal (RA), advising on the analysis. I wrote the manuscript with contribution to drafting, critical review, and editorial input from AM, IGH, CR, RA and NL. Julia Reisser, the University of Western Australia edited, and Monica F. Costa, Federal University of Pemambuco, Brazil and Rodrigo Riera, Catholic University of the Most Holy Conception, Chile critically reviewed the manuscript for Frontiers in Marine Science: Marine Pollution. ix PROJECT FUNDING This research was made possible through funding from Ocean Park Conservation Foundation (grant number: FH04_1516), Foundation FortUna, the PADI Foundation (grant number: 14668), Idea Wild, Mantahari Ocean Care, Fish-Are-Friends, R. Horner, private donors, as well as additional financial support from Murdoch University. The project also greatly benefitted from in-kind support from the Lembongan Marine Association on Nusa Lembongan, especially Big Fish Diving and Blue Corner Diver; the Dive Operators Community of Komodo, and especially Arenui Boutique Liveaboard, Current Junkies, Happy Days Yacht and Scuba Junkie Komodo, and Wunderpus Liveaboard. SCHOLARSHIP AND STIPEND I am a recipient of an International Posgraduate Award & Murdoch International Top Up For International Postgraduate Research Studentship Award. ANIMAL ETHICS STATEMENT This study was carried out in accordance with the approval of the Animal Ethics Committee, Murdoch University (R2781/15). Photographic and sighting data was collected opportunistically by the public and contributed to a public online repository (www.mantamatcher.org) developed explicitly to facilitate citizen science contributions to manta ray research. RESEARCH PERMIT STATEMENT This study was conducted under permits issued by the Indonesian Ministry of Research and Technology (Permit #458/SIP/FRP/E5/Dit. KI/XII/2015; Permit Extension#11/TKPIPA/E5/Dit. KI/XI/2016 and #86/EXT/SIP/FRP/E5/Dit.KI/XI/2017) and the Komodo National Park (#SI.1432/BTNK-1/2016). x THESIS PUBLICATIONS The following publications are associated with this thesis: Chapter 2 was published as: Germanov, E. S., Marshall, A. D., Bejder, L., Fossi, M. C., & Loneragan, N. R. (2018). Microplastics: no small problem for filter-feeding megafauna. Trends in Ecology & Evolution, 33(4), 227-232. Chapter 3 was published as: Germanov, E.S., Bejder, L., Chabanne, D.B.H., Dharmadi, D., Hendrawan, I.G., Marshall, A.D., Pierce, S.J., van Keulen, M. & Loneragan, N.R., (2019). Contrasting habitat use and population dynamics of reef manta rays within the Nusa Penida marine protected area, Indonesia. Frontiers in Marine Science, 6, 215. Chapter 4 has been formatted for journal submission (Peer J): Germanov, E.S., Pierce, S.J., Marshall A.D., Hendrawan, I.G., Bejder, B., & Loneragan, N.R. No place like home? Photo-identification reveals fine-scale habitat use of a large reef manta ray population in Komodo National Park, Indonesia. Chapter 5 was published as: Germanov, E.S., Marshall A.D., Hendrawan, I.G., Admiraal, R., Rohner, C.A., Argeswara, J., Raka, W., Himawan, M.R., & Loneragan, N.R. (2019). Microplastics on the menu: Plastics Pollute Indonesian Manta Ray and Whale Shark Feeding Grounds. Frontiers in Marine Science, 6, 679 xi Appendix I was published as: Stewart, J.D., Jaine, F.R., Armstrong, A.J., Armstrong, A.O., Bennett, M.B., Burgess, K.B., Couturier, L.I.E., Croll, D.A., Cronin, M.R., Deakos, M.H., Dudgeon, C.L., Fernando, D., Froman, N., Germanov, E.S., Hall., M.A, Hinojosa-Alvarez, S., Hosegood, J.E., Kashiwagi, T., Laglbauer. B.J.L., Lezama-Ochoa., N., Marshall, A.D., McGregor, F., Notarbartolo di Sciara, G.N., Palacios, M.D., Peel, L.R., Richardson, A.J, Rubin, R.D., Townsend, K.A., Venables, S.K., & Stevens, G.M. (2018). Research priorities to support effective manta and devil ray conservation. Frontiers in Marine Science, 5, 314 Data (Chapters 3 and 4) collected and used in this thesis contributed to the following publication: Venables, S.K., Marshall, A.D., Germanov, E.S., Perryman, R.J., Tapilatu, R.F., Hendrawan, I.G., Flam, A.L., Van Keulen, M., Tomkins, J.L., & Kennington, W.J. (2019). It's not all black and white: investigating colour polymorphism in manta rays across Indo-Pacific populations. Proceedings of the Royal Society B, 286(1912), 20191879. xii CONFERENCE PRESENTATIONS Germanov, E.S., and Marshall A.D. (2015) Regional Movement Patterns of Manta alfredi through a Complex of Parks and Fisheries. Oceania Chondrichthyan Society, Auckland, New Zealand. Germanov, E.S., Marshall A.D., Hendrawan, I.G., and Loneragan, N.R. (2016) Plastics on the Menu: Microplastics Present in Indonesian Manta Ray Feeding Habitats. American Elasmobranch Society, New Orleans, USA. Germanov, E.S., Marshall A.D., Hendrawan, I.G., and Loneragan, N.R. (2016) Plastics on the Menu: Microplastics Present in Indonesian Manta Ray Feeding Habitats. International Marine Conservation Congress, St. John’s, Canada. Germanov, E.S., Marshall A.D., Hendrawan, I.G., and Loneragan, N.R. (2016) Plastics on the Menu: Microplastics Present in Indonesian Manta Ray Feeding Habitats. Presented by I Gede Henrawan to Conference Marine Biodiversity, Bali, Indonesia. Germanov, E.S., Marshall A.D., Hendrawan, I.G., and Loneragan, N.R. (2018) Plastik untuk Makanan: Keberadaan Mikroplastik Pada Habitat Makan Pari Manta di Indonesia. Indonesian Shark and Ray Simposium, Jakarta, Indonesia. Germanov, E.S., Marshall A.D., Hendrawan, I.G., and Loneragan, N.R. (2018) Plastics on the Menu: Microplastics Present in Indonesian Manta Ray and Whale Shark Feeding Habitats. International Marine Conservation Congress, Kuching, Malaysia. Germanov, E.S., Marshall A.D., Hendrawan, I.G., Admiraal, R., Rohner, C.A., Argeswara, J., Raka, W., Himawan, M.R., and Loneragan, N.R. (2019) Plastics Pollute Indonesian Manta Ray and Whale Shark Feeding Grounds. International Whale Shark Conference, Ningaloo, Australia. xiii AKNOWLEDGEMENTS Every thesis has a story and mine begins with Helen Mitchell and Peter Bassett, founders of the Aquatic Alliance. Thank you for showing me that I can become a conservation biologist if I put my heart to it. Thanks to Helen and Peter, I met Andrea Marshall, Chris Rohner, Simon Pierce and the rest of the Marine Megafauna Foundation (MMF) crew, who took me under their fin. Andrea, thank you for your confidence in me and shaping the conservationist I am today by helping me get a scholarship and gain acceptance into a PhD program, sharing personal experiences, advice and more. Enter stage left, my primary supervisor, Neil Loneragan (highly recommended by Vanessa Jaiteh and Marianne Nyegaard). Thank you Neil for being exactly the supervisor I needed and gently guiding me back if I got “too feral”. The last few years my secret PhD mantra has been “make Neil proud” and that has helped me through many tough battles at the lab bench and writing desk. The next key player to enter this story was Bpk I Gede Hendrawan from Udayana University, Indonesia without whom this project would have never taken shape in Indonesia. Thank you Bpk Gede, for all the assistance with the paper work to allow my research in your country and to get samples transported for analyses, also for your gracious patience as we learn to work together through the cultural divide. Thank you for teaming me up with the best of your students that have rekindled the teacher in me and kept me motivated. There are many other players that have come in along the thesis journey and there is a good chance I might leave someone out, in case I do, please know that you shared in this journey with me and this thesis is your win as well. Early on, Lydie Couturier critically reviewed my first marine science manuscript, giving me a taste of what is to come and introduced me to many colleagues. Marianne Nyegaard helped me with grants, forms and lent me sampling gear during the pilot phase of my research. Clare Prebble helped with grants and ethics approvals. Thanks to you both, I managed to secure three grants and get the project off the ground. Chris Rohner was never more than an email away and ready to give advice on methodology, writing xiv and be a field and travel buddy. Kady Lyons and Sarah Hajbane helped with establishing methodology. Christine Parfitt helped with methodology, a microscope, and moral support (+) as a fellow non-for-profit manager/PhD Student in Indonesia. Katherine Burgess provided solid advice and dragged my sampling equipment from New York to Australia. Marten Welly, Ranny Ramadhani, Iqbal Putra and Susi Kamil were instrumental in helping me navigate the research process in Indonesia with connections, introductions and helping me with my Indonesian writing and translations. Betty Laglbauer was always ready to share permit stories and offer advice. Simon Pierce has truly been a massive thesis MVP and did his absolute best to be available to sit and nut things out with me in his easy going and thoughtful way. Lars and Mike helpfully sent along relevant papers, read drafts and provided thought provoking discussions. Thank you to the thesis examiners, manuscript reviewers and editors for their time and input. All have gone a long way towards much improving this thesis. Many people and businesses lent me their space, facilities and time. This study would not have been possible without the generous support of Lembongan Marine Association and the Dive Operators Community of Komodo, the local dive community and many citizen scientists to whom I am indebted. Particularly, Tim Birkett (Big Fish and Secret Garden), Cody and Andrew (Blue Corner), Ed Statham (Wunderpus), Current Junkies Liveaboard and Scuba Junkie dive shops have been instrumental. I thank the Aquatic Alliance for spearheading the manta ray ecology project and Marine Megafauna Foundation staff and volunteers for carrying it forward. Notably, Lillian Ellevog, Lisa Auditore, Rosie Cooper, Stuart Ecob, Doug Keim, Mickey Cobussen and Elias Sinderson were forefront on the data processing battle (and eagerly and enthusiastically followed all new, however small discoveries). Debbie and Jerry, Sonja Geier, Sabrina Inderbitzi, Jo Marlow, Fabi Nompas shared their beautiful photos. The help from MMF volunteers (AKA ‘Manta Minions’) was priceless. Thank you to: Satu Ahola, Soraya Al Rashid, Tatjana Baleta, Giorgi Barford, Rebecca Danieli, Kalina Grabb, Claudia Grenier, Sebastian and Sara Kalucza, Niamh Lynch, Sam Mathews, Janine Meers, Jannifer Ward, Dion Whittle, all the others. Special thanks to I Kadek Riadiyasa and the local dive xv communities. I am grateful for the assistance (and enthusiasm) of students from Udayana University: Janis Argeswara, Dewi Damayanti, Bintang Gustavina, Walundari Raka, Surya Risuana, Rai Ayu Saraswati, and Ayudian Swarry. Giles Winstanley (Stan) provided technical support and the use of ‘MantaUtil’ for streamlined data processing and facilitated ‘batch’ lodging of legacy data to ‘Manta Matcher’. Amelia Armstrong helped me through the pattern assigning teething problems. Thanks to the Manta Mather.org/Wild Me team for providing and supporting the data sharing platform and collaboration tool. Thanks to Hal Whitehead for the guidance he provided on the use of Markov movement models, Ryan Admiraal for linear models using R, Delphine Chabbane for SOCPROG, and Mike Calver for the use of Vasserstats. Thanks to members of Faculty of Marine Sciences and Fisheries, Udayana University and Clara Lafuente for assisting in mapping using QGis. Thanks to Rich Horner and Peter Coulson for providing microscopy tools and/or assistance. Thanks to Bpk Budi Wirayawan for assisting in access to precipitation data. For their logistical help in the field, I thank Captains Diasa and Edy for assisting at Manta Bay; the captains and crew of the Busy Girl and Happy Days yachts, and Komodo National Park rangers Bpk Ande Kefi, Bpk Yusuf, Dewi Indriasari and others from Scuba Junkie Komodo, Manta Rhei and the Arenui for assisting at the Komodo National Park; and Mahardika Himawan and the local boat captains and representatives for assisting at Pantai Bentar. I am also grateful for the study permit from the Indonesian Ministry of Research (Ristek Dikti) that permitted this research. Not all research undertaken over the last four years made it into this thesis. However, the help provided was nevertheless appreciated. Thank you to Shabri Nambiar and Sze How Bong for their assistance using mass spectrometry. Thank you to Ale Ponzo, Josh Rambahiniarison, MJ Lamoste, Gonzo Araujo, Ivan Tulla and the LAMAVE Bohol team for assisting in the mobulid study (and keeping me in bukko). Thank you to Carol Warren, Bpk Kennedy and Wayan Lena for helping me to create and disseminate stakeholder surveys and socialize the study results. I look forward to updating you on the progress on these remaining studies. xvi Some of the unsung heroes of this thesis are MMF’s volunteer accountant, Charis Gambil (thanks for making sure I don’t go over budget), media and public relations person, Sabrina Weiss (thanks for getting the research out into the public realm), and graphic designer, Giulia De Amicis (thanks for the gorgeous info graphs). Friends and family were often roped in to help as well. Thanks to Marek Golasch and my mom for the graphic design help, and to Nick Longfellow (my husband that somehow proposed and married me during this PhD madness) for the countless hours in the sun driving the speed boat or droning for manta rays (all the meals and pep talks too). Special thanks to all who joined my research expeditions and made PhD Life fun. Of course without my family this thesis would have ended in teary, snotty mess a long time ago. Thanks to my parents for always pushing me to use my brain to its fullest potential, my newfound family through marriage (the Longfellows and clan) for the support and motivation, my Jakarta family (Steph and Agus) for the company and logistical support and string pulling, my Australian family (Lin, Babs, Zos and Leu) for the peaceful home, the full belly and the ready ear. My self-proclaimed manta wife, Steph Venables for always checking in, offering to read chapters, plotting using R, offering to do my references (I wish I took you up on that), and always reminding me of why we do this in the first place. Finally, to my husband. Thank you for always reminding me that there is more to life than a PhD and the success (or failure) of this battle ultimately does not determine my self-worth. We survived the PhD, the house built and (hopefully) the pandemic, and that is something for both of us to be proud of. xvii LIST OF FIGURES Box 1 Definitions and graphical depictions of the common site use terminology applied throughout this thesis. ……………………………………..…….. 8 Figure 1.1 The (A) global habitat range of Mobula alfredi as indicated by thatched lines and (B) known aggregation and fisheries landing sites sites within (orange) and adjacent to (green) the Coral Triangle region. Indonesian locations are listed from left to right as: a) Pulau Weh, b) Sangalaki Island c) Nusa Penida MPA, d) Tanjug Luar, e) Komodo National Park, f) Rote Island, g) Lamakera and h) Raja Ampat. …..…………….……... 11 Figure 1.2 Schematic diagram of thesis structure. This schema will be repeated at the beginning of each chapter to orient the reader to the location of the particular chapter in the overall thesis framework. The position on the thesis roadmap is indicated in orange. …………………………..……………….………. 016 Figure 2.1 Classification, origins and trophic transfer of microplastics. Microplastic classification is based on (A) size in the environment, with microplastics defined as < 5 mm in size; and (B) size of manufacture and origin, with primary microplastics being those manufactured to be < 5 mm in size and secondary microplastics resulting from environmental degradation of larger plastic items. (C) Microplastics contain toxic plastic additives and adsorb persistent organic pollutants (POPs), such as such as dichloro-diphenyl-trichloroethanes (DDTs) and polychlorinated biphenyls (PCBs), that flow through marine food webs (depicted are the chemical structures of four common plastic-associated pollutants). Filter-feeding marine megafauna are particularly prone to microplastic ingestion and contamination by plastic-associated pollutants because of the large volumes of water they ingest during feeding. Modified, with permission, from The Ocean Cleanup, Copyright & Associates and designed by G.D. Amicis and M. Golasch. ...……………………………………………………………………...…. 023 Figure 2.2 Key buoyant microplastic hotspots overlap with habitat ranges of filter-feeding marine megafauna. The habitat ranges for (A) Mobula (Manta) alfredi, (B) Rhincodon typus, and (C) Balaenoptera physalus, as indicated by thatched, lined, or dotted overlay, respectively, overlap with regions containing high levels of buoyant microplastic pollution. Adapted, with permission, from data-calibrated models of buoyant microplastic concentrations (van Sebille et al., 2015) and habitat range data from the IUCN (2017). Designed by V. Brakalova. ………………………………………………………………………...… 026 Figure 3.1 Location of study sites in the Indonesian archipelago along the west coastline of Nusa Penida, Manta Bay (MB), and Manta Point (MP) shown in relation to Bali and the Lesser Sunda Islands, Indonesia. Popular SCUBA diving locations around the Nusa Penida MPA with infrequent manta ray sightings are also marked as described in the legend. Sites marked as “other” are ‘Pontoon,’ ‘Toyapakeh,’ ‘Sekolah Desa,’ ‘Pura Ped,’ ‘Sental,’ and ‘Pura Mas Gading.’ xviii Bathymetry information was available from: GEBCO_2014 Grid, version 20150318; www.gebco.net. ...………..…………..…………………...… 039 Figure 3.2 Dive operator logs of dives around the Nusa Penida MPA and those with manta ray sightings. The percent of total dives logged (top; n = 1,562) and of dives with manta ray sightings per location (bottom; n = 327) based on 2016 and 2017 logs from one dive operator. MB = Manta Bay; MP = Manta Point; BY = Buyuk; CB = Crystal Bay; MG = Mangrove; Other = six locations on the north coast of Nusa Penida. ...……………....…………………..…… 041 Figure 3.3 The combined number of annual survey days at manta ray sites Manta Bay and Manta Point by trained observers (Logs) and the manta ray survey days deposited to ‘Manta Matcher’ between 2012 and 2018 in the Nusa Penida MPA. Note: 2018 = January to April only. …………………………..............… 042 Figure 3.4 The number of individual manta rays observed with permanent injuries, including pregnant manta rays. Photos are copyright of the Marine Megafauna Foundation and the following individuals (from top): 1) Peter Bassett, 2) Jennifer Ward, 3) Stuart Ecob 4) Elias Sinderson, and 5) Lillian Ellevog. ………………………………………………………………..……...…... 045 Figure 3.5 Example photographs and observational attributes indicative of manta ray behavior. Photos are copyright of the Marine Megafauna Foundation and the following individuals (from top): 1) E. Germanov, 2) Lillian Ellevog, 3) E. Germanov, 4) Nicholas Longfellow, 5 -7) E. Germanov. ...…….… 046 Figure 3.6 The average annual (A) and monthly (B) manta ray sightings per unit effort (day) recorded between 2012 and 2017 in the Nusa Penida MPA at Manta Bay (MB) and Manta Point (MP) presented as the daily means (±1 SE). Data from 2012 and 2015 is excluded from (B) due to lower year-round effort. (A) Sightings (MB = 2,995, MP = 2,918), days (MB = 581, MP = 511). (B) Sightings (MB = 2,295, MP = 2,137), days (MB = 420, MP = 336). …... 051 Figure 3.7 The number of survey days (n = 659) logged by trained observers for each month between 2012 and 2017 in the Nusa Penida MPA. *The average per month is pooled and calculated across years (excluding prior 2013 and 2015 due to lower year-round effort). ..…………........… 051 Figure 3.8 Discovery curves for newly identified manta rays against elapsed days. Discovery curves are presented for the Nusa Penida MPA (all) and broken down by sites Manta Bay (MB) and Manta Point (MP) from January 2012 until April 2018. The x-axis gray lines indicate survey years. NS = number of survey days and NI = number of individuals. …………. 053 Figure 3.9 Population structure of manta rays in the Nusa Penida MPA, with (A) the number of individuals and (B) the number of sightings for Nusa Penida (All) and the two sites Manta Bay (MB) and Manta Point (MP). Based on sightings from January 2012 to April 2018. Percentages were calculated from the total number of individuals (A) or sightings (B) for each site. ……………………………………………………………………………. 055 xix Figure 3.10 The proportion of manta rays identified in the Nusa Penida MPA (All), and study sites Manta Bay (MB) and Manta Point (MP) plotted against the number of times individuals were sighted between January 2012 and April 2018. Total number of individuals for All, MB, MP, are shown for each number of sightings. ...…….………..…………….…………………..… 057 Figure 3.11 The top ten most sighted manta rays at Manta Bay (MB) and Manta Point (MP) are all males (A) and females (B), respectively. Colored circles show sighting locations. Solid lines indicate the first sighting where the individual is deemed mature, whereas dashed lines indicate observed pregnancy (female). ….………………………………………………..... 058 Figure 3.12 Lagged Identification Rates (LIR ± SE) of manta rays within the two study areas, Manta Bay (MB) and Manta Point (MP), Nusa Penida MPA, Indonesia. The mean probability of re-identifying an individual manta ray, from the time of first identification until approximately 6 years later within the same or different area in the Nusa Penida MPA (A), MB and MP (B); and broken down by sex within the sites MB (C); and MP (D). The predicted LIRs for models of best fit (Table 1) are shown for each group. Records from January 2012 to April 2018. ….………………………………………………………...… 059 Figure 3.13 Lagged Identification Rate (LIR ± 1 SE) of immature and mature male manta rays within Manta Bay, Nusa Penida MPA, Indonesia. The predicted LIRs for models of best fit (Table 3.4) are shown for each group. Only sighting records of individuals whose maturity status remained the same throughout the duration of the study (January 2012 to April 2018) are used. Immature male sightings = 487, individuals = 59. Mature male sightings = 707, individuals = 127. …………………………………………………………………….…...… 062 Figure 3.14 (A) The percentage of observed manta ray behaviors determined from sighting logs and photographs from Manta Bay (MB, n = 1,211) and Manta Point (MP, n = 2,481) from January 2012 to April 2018. Foraging and cleaning, as well as courtship and cleaning behaviors by an individual manta ray were on occasion both witnessed during the same sampling event and identified as such. (B) The average monthly sightings per unit effort (day) of manta rays engaging in courtship behavior. Data recorded between 2013 and 2017 (excluding 2015 due to lower year-round effort) in MP and presented as the daily means (± 1 SE). ………….………………………….…………...… 063 Figure 3.15 The annual (A) and monthly (B) daily mean (± 1 SE) number of boats present at manta ray sites in the Nusa Penida MPA overall, and separately in Manta Bay (MB) and Manta Point (MP), between 2012 and 2017. ….……....… 064 Figure 3.16 Movement patterns of male and female manta rays in the Nusa Penida MPA. ............................…………………………………….………….……...… 072 Figure 3.17 A manta ray forages at Manta Bay while people fish with pole and line from Nusa Penida cliffs. ………………………………….………..………...… 075 xx Figure 4.1 Location of study sites within the north (Cauldron – CL), central (Karang Makassar – KM, Mawan – MA), and south (Manta Alley – MA) regions of the Komodo National Park (NP) zone, in the Lesser Sunda region of the Indonesian archipelago. Other popular SCUBA diving locations with infrequent manta ray sightings are marked as “other” sites. These locations include (from north to south): ‘Castle’ and ‘Crystal Rock’, ‘Lighthouse’, ‘Tatawa Besar’, ‘Tatawa Kecil’, ‘Batu Bolong’, ‘Siaba Besar’, and ‘German Flag’. Bathymetry information was obtained from: GEBCO_2014 Grid, version 20150318; www.gebco.net. .…………...… 086 Figure 4.2 Mean daily (A, B) and monthly (C, D) sightings of identified manta rays for the Komodo National Park at Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) from 2013 – 2017. Data are presented as annual (A, C) and monthly (B, D) mean sightings (± 1 SE) per unit effort (day – A, B; month – C, D). KM = 1,730 sightings, 415 days; MW = 911 sightings, 233 days; MA = 1,011 sightings, 102 days. .………………………………… 092 Figure 4.3 Individual manta rays were sighted in both Komodo NP and Nusa Penida MPA. The sighting location (site) and date are plotted for each individual according to their identification (ID) codes. The sex of the individual is indicated within the brackets following the ID code, where (F) indicates female and (M) male. The ID codes, excluding the bracketed sex information, can be used to call up full sighting records for the individuals from ‘Manta Matcher’. ………………….………….…………………………..……………....… 098 Figure 4.4 The combined number of annual survey days at manta ray sites in Komodo National Park (Karang Makassar, Mawan, Manta Alley and Cauldron) by trained observers (Logs), and additional manta ray survey days submitted to ‘Manta Matcher’. ………….…………………………..….………...… 098 Figure 4.5 The mean individual sightings per dive (ndives = 617) logged by trained observers per (A) year and (B) month in Karang Makassar (KM), Mawan (MW), and Manta Alley (MA), Komodo National Park. ..…....…..…...… 099 Figure 4.6 Discovery curves for newly-identified manta rays over time (in days) from January 2013 until April 2018. The x-axis gray lines indicate survey years. Discovery curves are presented for the Komodo NP (Comb.) and broken down by sites Karang Makassar (KM), Mawan (MW), and Manta Alley (MA). NS = number of survey days and NI = number of individuals. ………………………………………………………………………...…. 0101 Figure 4.7 Population structure of manta rays in the Komodo National Park from 2013 – 2018. (A) Population structure and (B) sightings for the combined (Comb.) core Komodo National Park sites and broken down by sites (from north to south): Cauldron (CL), Karang Makassar (KM), Mawan (MW) and Manta Alley (MA). ….............…………………………………….………………...… 0104 Figure 4.8 Manta rays sighted more than once at the Cauldron (CL) and their sighting locations. Colored circles show sighting locations. Solid vertical line indicates the first sighting where the individual is deemed mature. xxi *indicates individuals that were immature throughout the duration of the study. ……………………………….………………………………..… 0105 Figure 4.9 The proportion of individual manta rays in Komodo National Park plotted against the number of times individuals were sighted between January 2012 and April 2018. (A) Combined sightings from the three core sites (Comb.), (B) Karang Makassar (KM), (C) Mawan (MW) and (D) Manta Alley (MA). Total number of individuals are shown for each number of sightings. …………………………………………………………………...………. 0106 Figure 4.10 The top most sighted manta rays at (A) Karang Makassar (KM), (B) Mawan (MW) and (C) Manta Alley (MA). Colored circles show sighting locations. ............................…………………………………….………………...… 0107 Figure 4.11 Lagged Identification Rates (± 1 SE) of manta rays within and between the three core study areas in Komodo National Park (2013 – 2018): Karang Makassar (KM), Mawan (MW), and Manta Alley (MA). (A) The probability of an individual manta ray being identified in the same or a different site within the Park over time (days). (B) The probability of an individual manta ray being identified over time (days) within each of the three core sites. Best-fit LIR models (Table 4.2; dotted lines) are shown for each site. ……………………………………………………………..……..……... 0110 Figure 4.12 The mean (± 1 SE) probability of re-identifying an individual manta ray within the same or different central core site (Karang Makassar and Mawan) within Komodo National Park, Indonesia. Records from January 2013 to April 2018. ...............................………………………………….………………...… 0110 Figure 4.13 The mean probability (± 1 SE) of re-identifying an individual manta ray, broken down by sex, from the time of first identification until an average of approximately 3.5 years later within the three study areas, (A) Karang Makassar, (B) Mawan, and (C) Manta Alley, Komodo National Park, Indonesia. The predicted LIRs for models of best fit (Table 4.3) are shown for each group. Records from January 2013 to April 2018. …………..………………………………………………………….…...... 111 Figure 4.14 (A) Manta ray behaviors observed between January 2013 and April 2018 from four sites within Komodo NP. Sightings at Cauldron (CL, 49), Karang Makassar (KM, 1,067), Mawan (MW, 741) and Manta Alley (MA, 498). The number of foraging individuals (B) and feeding events (C) broken down by site and season (November – April = north-west monsoon, NW; May – October = south-east monsoon, SE). (D) The average monthly sightings per unit effort (day) of identified manta rays engaging in courtship behavior. Data recorded between 2013 and 2017 across sites KM, MW and MA and presented as the daily means (± 1 SE). The textured and solid backgrounds indicate NW and SE seasons, respectively. .......................................................................…..………...… 113 Figure 4.15 The daily average (± 1 SE) number of boats present at core manta ray sites in the central region of the Komodo National Park overall, and separately for xxii Karang Makassar (KM) and Mawan (MW), across year (A) and month (B) for 2014 and 2017. ………………….…………………………………….…... 115 Figure 4.16 Movement patterns of male and female manta rays in the Komodo NP. Pink arrows denote governing movements of females, blue of males, and gray arrows denote no observed differences between the sexes. .…...………………... 119 Figure 5.1 Regional oceanographic details, river emission points and population density adjacent to study areas. Bathymetry information was available from GEBCO_2014 Grid, version 20150318; www.gebco.net. River emission points available from Lebreton & Reisser (2018). Population information in15 min × 15 min grids (~ 760 km2) using 2025 population projections available from Yetman et al., (2004). ……………………………………….……... 148 Figure 5.2 Map of study locations at (A) North and South Manta Bay (N-MB and S-MB, respectively), Nusa Penida Marine Protected Area; (B) Karang Makassar (KM), Komodo National Park; and (C) Pantai Bentar (PB), East Java in Indonesia. Different color lines in (A – C) represent individual survey GPS tracks. ……………………………………………………….…….………..…….. 149 Figure 5.3 Total monthly rainfall for study regions (A) Bali, (B) Komodo Island, and (C) East Java during study years. 0 = no precipitation recorded; nd = no data available. Precipitation data was available from the Indonesian Meteorology, Climatology and Geophysics Agency Data online, Pusat Database BMKG; http://dataonline.bmkg.go.id/. …………………………..….………...…... 150 Figure 5.4 Samples collected by trawling the near surface water adjacent to feeding animals. ………………………………………………...……….……..…. 153 Figure 5.5 (A) Sampling from the surface-water interface using nets with round or rectangular openings. (B) Example calculations for converting plastic per volume to plastic per surface area and unit conversion. …….…......…...... 155 Figure 5.6 Boxplot depicting plastic abundance estimated through (A) swimmer towed trawl surveys (plastic min-1), (B) boat trawl surveys (plastic m-3), and (C) visually counted anthropogenic debris (plastic km-2) from north (N-MB) and south (S-MB) Manta Bay, Nusa Penida; Karang Makassar (KM), Komodo National Park; and Pantai Bentar (PB), East Java. …….…..…... 164 Figure 5.7 The size (A) and type (B) of anthropogenic debris recorded through trawl and visual (C) surveys from South (MB-S) and North (MB-N) Big Manta Bay, Nusa Penida, Karang Makassar (KM), Komodo National Park and Pantai Bentar (PB), East Java pooled across years and presented as a fraction of the total amount. ……………….………………………………….…….…..…....... 166 Figure 5.8 Plankton abundance pooled by season for south (S-MB) and north (N-MB) Manta Bay, Nusa Penida; Karang Makassar (KM), Komodo National Park; and Pantai Bentar (PB), East Java. ……………….………….……….…... 169 xxiii Figure 5.9 Low (A) and high magnification photographs of smaller (B) and larger (C) plastic debris found in manta ray faecal material (n = 1) from Nusa Penida, Indonesia. …………………………………………….……..………..…... 170 Figure 5.10 Various sizes and types of plastic debris found in manta ray vomit material (n = 1) from Nusa Penida, Indonesia. Larger debris are photographed under high (A, B) and low (C) magnification, and smaller debris (D) under high magnification. …………………...…………………………..………..…... 171 Supplementary Figure 1 The available estimated size classes of manta rays using Manta Bay (A, B) and Manta Point (C, D). Size classes are grouped by estimated disk width size (m) in 0.5 m bins and presented by sex (A, C) and for males according to maturity status (B, D). J = juvenile, SA = sub-adult, Mat = mature. …………..…... 233 xxiv LIST OF TABLES Table 2.1 The world’s large marine vertebrate filter feeders, their IUCN conservation status, and their presence in key microplastic pollution hotspots. ……..…. 026 Table 3.1 Individual manta rays, sightings and survey days, recorded between June 2004 and April 2018, within Nusa Penida overall, Manta Bay (MB) and Manta Point (MP). Daily duplicates are removed (n = 163). .……....……………..…… 042 Table 3.2 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of (A) all manta rays, (B) males and (C) females at Manta Bay (MB) and Manta Point (MP) in the Nusa Penida MPA, Indonesia. …...…………………………………………………………………..….… 049 Table 3.3 The estimated probability (± 1 SE) of an individual manta ray originally identified from one area being re-sighted within the same or different area. Movement probabilities are presented for (A) all manta rays (n = 624), (B) males (n = 342) and (C) females (n = 244). …...…..…..…..…….…… 049 Table 3.4 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of (A) immature and (B) mature male manta rays at Manta Bay (MB) in the Nusa Penida MPA, Indonesia. …...………...………..………. 061 Table 3.5 A comparison of manta ray M. alfredi population studies and demographics from the available literature. ..…...…...………………..…..………...……. 066 Table 4.1 Individual manta rays, sightings and survey days recorded between July 2004 and April 2018. Data are reported for Komodo National Park (KNP) overall, Cauldron (CL), Karang Makassar (KM), Mawan (MW), Manta Alley (MA), other sites within KNP and sites within the Nusa Penida Marine Protected Area (NP MPA). Daily duplicates are removed (n = 755). .………....…… 088 Table 4.2 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of manta rays at Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) in the Komodo NP, Indonesia. ..….....……………..… 094 Table 4.3 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of (A) male (n = 507) and (B) female (n = 498) manta rays at Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) in the Komodo NP, Indonesia...........................................................................................… 095 Table 4.4 The estimated probability (± 1 SE) of an individual manta ray originally identified from one area being re-sighted within the same or different area. Movement probabilities are presented for manta rays (n = 1,061 individuals) at (A) the three core sites Karang Makassar (KM), Mawan (MW) and Manta Alley (MA). (B) Sites grouped into central (KM + MW) and southern (MA) regions.0……………...........................................................................…… 096 xxv Table 4.5 The estimated probability (± 1 SE) of an individual manta ray originally identified from one area being re-sighted within the same or different area. Movement probabilities are presented for (A) males (n = 507) and (B) females (n = 498) at core sites Karang Makassar (KM), Mawan (MW) and Manta Alley (MA).……………………………………………………………..….. 96 Table 4.6 Model H outputs for time spent within (res time in ± 1 SE) and outside (res time out ± 1 SE) of Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) sites in the Komodo NP, Indonesia. …………………………………………109 Table 5.1 Research locations, sampling seasons and times, and the numbers of (A) feeding, (B) sub-surface, (C) surface, (D) sampling times in Indonesia from January, 2016 to February, 2018 and (E) egested material samples collected from 2016 – 2018. ………………………………………………………………………….….... 152 Table 5.2 Floating plastic debris abundance estimates for each sampling location and season and pooled across years. ……..……………………………...…….... 156 Table 5.3 Estimated coefficients, t-values and significance levels (p), for linear model of log-transformed trawl sampled anthropogenic debris (plastic m-3) on year, season, location and a season-location interaction. …..……………...…....... 162 Table 5.4 Estimated coefficients, t-values and significance levels (p), for linear model of log-transformed visually counted anthropogenic debris (plastic km-2) on year, location, season, and a location-season interaction. …..……….………….... 165 Table 5.5 A comparison of plastic abundance estimates in filter feeding megafauna foraging grounds, theoretical ingestion estimates and plastic associated pollutant levels (when available). …..…...………………..………………………….…….... 168 Supplementary Table 1 The following data tables catalogue the identification (ID) numbers and types of injuries recorded as of April 2018 for individual manta rays that have been sighted with injuries in the A) Nusa Penida Marine Protected Area (MPA) and B) Komodo National Park (NP). ID numbers can be used to search for full sightings information at www.mantamatcher.org. Ceph Inj = Cephalic injury; Pec Inj = Pectoral injury; Line Inj = Line injury. ……………………..…... 234 xxvi TABLE OF CONTENTS DECLARATION iii ABSTRACT iv STATEMENT OF CONTRIBUTION OF OTHERS vi THESIS PUBLICATIONS x AKNOWLEDGEMENTS xiii LIST OF FIGURES xvii LIST OF TABLES xxiv CHAPTER 1. GENERAL INTRODUCTION 1 1.1 THE FLAGSHIP APPROACH TO CONSERVATION …….……………………. 2 1.1.1 Citizen Science …………………………………………………..…….. 4 1.2 CHALLENGES TO MARINE MEGAFAUNA CONSERVATION …………………… 5 1.3 MANTA RAY ECOLOGY …………………………………..…………. 9 1.4 THREATS TO MANTA RAYS ………….…….….…….……….…….…. 13 1.5 AIMS AND OBJECTIVES ……….…..….….….……..….…….…….…. 15 1.5.1 Thesis Structure …………………………..……………………………… 17 CHAPTER 2. MICROPLASTICS: NO SMALL PROBLEM FOR FILTER-FEEDING MEGAFAUNA 20 2.1 ABSTRACT ………………………………………………………. 22 2.2 INTRODUCTION …………….……….……………………………. 22 2.3 SUSCEPTIBILITY TO MICROPLASTIC INGESTION ……………….…………. 24 2.4 EFFECTS OF MICROPLASTIC INGESTION …………………….…….……. 28 2.5 FUTURE DIRECTIONS ….…….…….….….….…….…….……..…… 28 xxvii 2.6 VALUE OF FOCUSING ON MEGAFAUNA …………….…..….…….……… 30 CHAPTER 3. CONTRASTING HABITAT USE AND POPULATION DYNAMICS OF REEF MANTA RAYS WITHIN THE NUSA PENIDA MARINE PROTECTED AREA, INDONESIA 32 3.1 ABSTRACT ……………………………….….….….….…………. 34 3.2 INTRODUCTION …………………………………………….….…. 35 3.3 METHODS …….…………………………………………………. 38 3.3.1 Description of Study Sites ………………………….……..…….……... 38 3.3.2 Data Collection …………………………………….…………………… 40 3.3.3 Analyses of Residency and Movement ……………………….………. 44 3.3.4 Statistical Analyses ……………………….….…………………………. 48 3.4 RESULTS …………………………………………….…….……. 50 3.4.1 Sightings and Survey Effort ………….………………………………… 50 3.4.2 Population Structure ……………….…………………………………… 53 3.4.3 Injury Rates ………………………….………………………………….. 55 3.4.4 Residency and Movement ……………….….…………………………. 56 3.4.5 Behavior …………………………………………………………………. 62 3.4.6 Boats ……………………………………………………………………... 64 3.5 DISCUSSION …………….…….…….…………………………… 65 3.5.1 Manta Bay ………………………………………………………………. 65 3.5.2 Manta Point ……………………………………………………………… 68 3.5.3 Site Fidelity and Productivity …………………………………………… 69 3.5.4 Movement Patterns ……………………………………………………… 71 3.5.5 Implications for Management …………………………………………. 72 xxviii 3.5.6 Manta Ray-Focused Tourism Impacts …………………………………. 73 3.5.7 Artisanal Fisheries ………………………………………………………. 74 3.5.8 Limitations and Future Directions ………………………………………. 76 3.6 CONCLUSION …………….…….…….……….……….…….……. 78 CHAPTER 4. NO PLACE LIKE HOME? PHOTO-IDENTIFICATION REVEALS FINE-SCALE HABITAT USE OF A LARGE REEF MANTA RAY POPULATION IN KOMODO NATIONAL PARK, INDONESIA 79 4.1 ABSTRACT ……………………….…………….…….….……… 81 4.2 INTRODUCTION …………………………………………………. 82 4.3 METHODS ……………………………………….….….…….…. 85 4.3.1 Study Area ………………………………………………………………. 85 4.3.2 Data Collection and Processing ……………………………………….. 89 4.3.3 Statistical Analyses ……………………………………………………… 91 4.3.4 Analyses of residency and movement ……………….…….….………. 93 4.4 RESULTS …………………………….….….……………………. 96 4.4.1 Sightings and Survey Effort …………………………………………….. 96 4.4.2 Population Structure …………………………………………………….. 101 4.4.3 Residency and Movement ……………………………………………… 103 4.4.4 Behavior …………………………………………………………………. 111 4.4.5 Injury Rates ………………………………………………………………. 114 4.4.6 Boating Activity …………………………………………………………. 114 4.5 DISCUSSION ……….……….……….…………………………… 115 4.5.1 Residency and Movement ………………...….….….….….….….…… 116 xxix Population demographics 119 Influence of environment 123 Reproductive behavior 127 4.5.2 Local Threats to Manta Rays …………………………………………… 128 Unregulated tourism 128 Fisheries and natural predators 131 4.5.3 Limitations and Future Directions ………………………………………. 133 4.6 CONCLUSION ……….……….……….……….……….……….…. 136 CHAPTER 5. MICROPLASTICS ON THE MENU: PLASTICS POLLUTE INDONESIAN MANTA RAY AND WHALE SHARK FEEDING GROUNDS 138 5.1 ABSTRACT ……………………………….……………………… 140 5.2 INTRODUCTION …………………………….……………………. 141 5.3 MATERIALS AND METHODS ………………….…….…….…….….…. 145 5.3.1 Study Locations and Oceanographic Conditions ……….…..........… 145 Oceanography in the region 146 5.3.2 Surveys and Sample Analysis ………………………….…….….……. 151 Feeding behavior observations 151 Trawl surveys 151 Sample Analysis 155 5.3.3 Visual Surveys for Plastic Debris …….….….….….…………………... 158 5.3.4 Manta Ray Egested Material Collection and Analysis ………………. 158 5.3.5 Calculating Theoretical Plastic Ingestion Rates ……………………….. 159 5.3.6 Statistical Analyses ……………………………………………………… 159 xxx 5.4 RESULTS …..……………………………………………………. 160 5.4.1 Local Rainfall ……………………………………………………………. 160 5.4.2 Manta Ray and Whale Shark Observations …………………………. 160 5.4.3 Trawl Surveys ……………………………………………………………. 161 5.4.4 Visual Surveys …………………………………………………………… 164 5.4.5 Characterization of Debris …………………………………………….. 165 5.4.6 Plastic Ingestion …………………………………………………………. 167 5.5 DISCUSSION ………..…….…….…….…….…….….…………… 172 5.5.1 Plastic Abundance Estimates …………………………………………… 172 5.5.2 Seasonal Trends in Plastic Debris Abundance ……………………….. 173 5.5.3 Geographic Differences in Abundance and Characteristics of Plastics …………………………………………………………………… 175 5.5.4 Theoretical Plastic Ingestion Rates …………………………………….. 178 5.5.5 Implications of Plastic Ingestion ……………………………………….. 181 5.5.6 Efforts to Stem the Plastic Tide …………………………………………. 184 5.6 CONCLUSION ………………………………………….……….… 184 CHAPTER 6. GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS 186 6.0.1 Habitat Use ………………………………….………………………….. 190 6.0.2 Threats to Manta Rays, Particularly Plastics ………………………….. 190 6.1 MANTA RAY ECOLOGY (CHAPTERS 3 AND 4): ……………….….…….…. 191 6.1.1 Nurseries and Manta Rays ………………………………....….……... 194 6.1.2 Citizen Science Wins and Shortcomings ……………………………... 199 6.1.3 Directions for Future Manta Ray Research …….……………………... 201 xxxi 6.2 PLASTIC POLLUTION IN FEEDING GROUNDS (CHAPTER 2 AND 5): .…….……. 204 6.2.1 Debates on the Extent of Risks Posed by Microplastics …………..…. 206 6.2.2 Directions for Future Microplastics Research ……………………….… 209 6.3 IMPLICATIONS FOR MANAGEMENT ……………….….….……….……. 216 6.3.1 Tourism Regulations …………………………………………………….. 216 6.3.2 Small-scale Fisheries ……………………………………………………. 219 6.3.3 Plastic Marine Debris …………………………………………………… 224 6.4 CLOSING STATEMENT ……………………………………………… 229 APPENDIX I. POLLUTION AND CONTAMINANTS 230 I.i IDENTIFYING TROPHIC INTERACTIONS WITH POLLUTANTS ………………… 230 I.ii QUANTIFYING IMPACTS FROM POLLUTANTS …………………….……… 232 APPENDIX II. SUPPLEMENTARY DATA 233 II.i CHAPTER 3 SIZE FREQUENCY DATA ….….………….….….….….….…. 233 II.i CHAPTER 3 SIZE FREQUENCY DATA ….….………….….….….….….…. 234 REFERENCES 238 xxxii 1 CHAPTER 1. GENERAL INTRODUCTION We find ourselves at a precarious time in history, racing to close the gaps in our understanding of marine ecosystems with unprecedented levels of research, all in effort to curb biodiversity loss and prevent ecosystem collapse (Borja, 2014; Parsons et al., 2014). Recent estimates indicate that up to 25% of assessed marine species are facing the risk of extinction (Webb et al., 2015). Over exploitation of natural resources (e.g. fisheries and their associated incidental bycatch), habitat loss and environmental degradation are some of the most common reasons for local, regional and global species extinction (Dulvy et al., 2003; Webb et al., 2015). Habitat loss and environmental degradation can result from coastal land and sea development, destructive fishing practices (i.e. bottom trawling and blast fishing), pollution, plastic debris, invasive species, or climate change amongst other stressors, such as increased marine traffic and water use. Conservation biology, the science focusing on mitigating biodiversity loss, ecosystem and species management, hinges on understanding habitat use patterns of species and their population demographics within their habitats (Hueter et al., 2005). In this thesis, I focus on the conservation biology of reef manta rays Mobula alfredi. I aim to increase the understanding of their habitat use, population demographics and dynamics of this threatened species in two key aggregation areas in Indonesia, Nusa Penida Marine Protected Area (MPA) and Komodo National Park (NP). Further, I focus on identifying and characterizing the remaining threats to manta rays within these conservation zones in an effort to improve their management. CHAPTER 1. GENERAL INTRODUCTION 2 1.1 THE FLAGSHIP APPROACH TO CONSERVATION Ever increasing limitations of resources and the escalating number of threatened species has resulted in conservation-centric projects with a focus on surrogates (Barua, 2011; Caro et al., 1999; Entwistle, 2000; Verissimo et al., 2011; Walpole et al., 2002). Commonly used surrogate species are ‘umbrella’ species, whose effective protection requires the protection of large tracts of land (e.g. tigers Panthera tigris and grizzly bears Ursus arctos) or seascapes (e.g. whales Mysticeti) and benefits other species with overlapping ranges (Caro & O’Doherty, 1999; Barua, 2011); and/or ‘flagship’ species, often defined as a species with desirable attributes that evoke certain emotions from the public that facilitate awareness, drive conservation action and garner financial support for conservation initiatives (e.g. orangutans Pongo spp, panda Ailouropda melanoleuca and polar bears Ursus maritimus) (Barua, 2011; Caro & O’Doherty, 1999; Verissimo et al., 2011; Walpole & Leader-Williams, 2002). Other categories of species of ecological significance include ‘indicator’, ‘keystone’, ‘foundation’, and ‘engineering’ species (Barua, 2011; Caro et al., 1999; Ducarme et al., 2013). The use of surrogates, particularly flagships, to drive conservation priorities has been debated (Caro & O’Doherty, 1999; Ducarme et al., 2012), particularly the issue of the surrogate gaining preference over the ultimate goal of preserving biodiversity (Barua, 2011; Entwistle, 2000). However, with limited resources for conservation initiatives, surrogates and especially flagships undeniably have an important role in facilitating biodiversity conservation (Smith et al., 2010; Verissimo et al., 2011; Walpole et al., 2002; Zacharias et al., 2001). Species that are considered both good umbrella and flagship species are those that have: 1) a large body size and long-life spans; 2) large home ranges and long-range movements, but with specialized habitats within their range; 3) smaller population sizes; and are 4) sensitive to human disturbance (Caro & O’Doherty, 1999; Zacharias & Roff, 3 2001). Species that are already in the public consciousness and are perceived as non-threatening to the public are often included in the criteria for choosing flagship species by international conservation organizations (Home et al., 2009). Many large marine filter-feeders fit well with the criteria for flagship and umbrella species and are used to drive conservation. However, those species with better characterized biology will make for better umbrella species as understanding how environmental factors are linked to population growth rates will make monitoring populations easier (Caro & O’Doherty, 1999). Manta rays (M. alfredi and the giant manta Mobula birostris) are perfectly poised to serve as flagship species based on their life history traits, iconic and charismatic status, and importance for blue economies and livelihoods in developing nations (O’Malley et al., 2013). Further, because of their feeding strategies manta rays and other large filter-feeders are especially susceptibile to accidental plastic ingestion. For these and other reasons (discussed in Chapter 2) large filter-feeders are ideal flagships to raise awareness about microplastics and marine plastic pollution as a whole, particularly when conservation and awareness campaigns engage all stakeholders, including local communities (Bowen-Jones & Entwistle, 2002). However, to increase the effectiveness of single species conservation programs that aim to also benefit habitats and other species under the same umbrella, it is important that we improve our understanding of how manta rays use the sites where they are commonly encountered. These very significant locations are often spatially limited areas of several hundred meters up to few kilometers, within the larger home ranges of manta rays (Graham et al., 2016; Hueter et al., 2005). Understanding this fine-scale manta ray habitat use is important in developing mitigation and management strategies from threat sources in the major aggregations across their range. For example, limiting fishing in key manta ray habitats takes the pressure of the targeted species in these areas and also protects manta rays and other species from the CHAPTER 1. GENERAL INTRODUCTION 4 risk of entanglement (see Chapters 3 and 4). In another example, if foraging grounds appear to be confined to specific areas (see Chapters 3 and 4), plastic abundance assessments need to be focused on these areas (see Chapter 5) and management strategies need to consider how to best reduce plastic pollution there (see Chapter 6). 1.1.1 Citizen Science With the benefit of public recognition, flagships can benefit from public engagement in another way – citizen science. Citizen science, the engagement of non-specialists in scientific research (Bonney et al., 2009; Miller-Rushing et al., 2012), can be seen as a two-pronged approach to conservation. First, citizen science can greatly enhance and potentially facilitate more cost-effective, rapid and widespread data collection (Chandler et al., 2017; Pierce et al., 2018). For example, citizen science, has been instrumental in providing basic, yet necessary, information such as occurrence, abundance, and movement for manta rays (Couturier et al., 2014; Germanov & Marshall, 2014) and whale sharks (Norman et al., 2017; Robinson et al., 2016). To streamline the process, online databases such as ‘MantaMatcher.org’ (Manta Matcher) and Whaleshark.org are used to assist in collecting data from the public and researchers (Germanov & Marshall, 2014; Norman et al., 2017, Robinson et al., 2016). These data have had significant impact on conservation policy and legislation e.g. the use of Manta Matcher has provided information on long-range (up to 450 km) manta ray movements between conservation zones and heavily fished waters (Germanov & Marshall, 2014) that ultimately helped influence manta ray protection laws in Indonesia. Second, citizen science encourages the involvement of stakeholder communities in the scientific process, resulting in enhanced community awareness, empowerment, confidence in scientific recommendations and builds trust between the stakeholders and 5 researchers (Bonney et al., 2009; Miller-Rushing et al., 2012). As general public knowledge and awareness about ecological issues increases through citizen science participation (Bonney et al., 2009; Miller-Rushing et al., 2012), it is likely that stakeholders increase their commitment to improve environmental stewardship for the target species and its broader habitat. Popular tourism destinations are ideal locations to implement citizen science initiatives to collect data on economically important flagship species and engage with the public to provide interpretation and facilitate awareness (Brightsmith et al., 2008; Germanov & Marshall, 2014; Kasmidi & Gunadharma, 2017; Mau, 2008; Walpole et al., 2002). This is especially so when encounters with the target species are a great attraction and public interest and the desire to learn is high (Mau, 2008; Ziegler et al., 2012), thus maximizing the potential benefits of the flagship approach to conservation. 1.2 CHALLENGES TO MARINE MEGAFAUNA CONSERVATION Conservation of marine megafauna has its challenges. Often these species cover extensive geographical areas, have conservative live-history traits, which make them very vulnerable to additional sources of mortality and pressures that might impact reproductive fitness, thus requiring specialized management considerations. Effective conservation is often further hampered by the limited general ecological knowledge, such as habitat use, population dynamics, and the extent of threats across their range. Perhaps the most challenging aspect of marine megafauna conservation is their inherent home range sizes. Animals with large body masses typically require access to abundant prey. As a result, most marine megafauna, such as marine mammals (Rasmussen et al., 2007), sharks (Bonfil et al., 2005; Eckert et al., 2001) and marine turtles (Bowen et al., 2007; Boyle et al., 2009), have large home ranges and undergo long CHAPTER 1. GENERAL INTRODUCTION 6 distance movements and migrations to access productive foraging grounds (e.g. seasonal poleward migrations of whales; (Dawbin, 1966) that in some species are in the range of 10,000 km (Alerstam et al., 2003; Block et al., 2011; Hays, 2008; Ogburn et al., 2017). Marine megafauna may also regularly undergo large-scale movements for reproductive purposes (e.g. marine mammals migrations to equatorial birthing grounds, Lockyer, 1984; and turtles laying eggs in natal grounds, Lohmann et al., 2013). Ultimately the home ranges of many marine megafauna species are vast and their large-scale movements often transverse national boundaries making broad conservation strategies complex to execute. As a result, management efforts have largely focused on the identification and protection of critical habitats within these larger home ranges (Chapman et al., 2015; Graham et al., 2016; Hooker et al., 2011; Hueter et al., 2005). It is common for marine conservation zones such as Marine Protected Areas (MPAs), national parks and marine sanctuaries to be created with the purpose of protecting key flagship species (Spalding et al., 2013) while also generally aiming to mitigate biodiversity loss (Borja, 2014). However, these areas are often designated without sufficient knowledge of how these habitats are really used by the target conservation species (Graham et al., 2016; Mora et al., 2006). Often additional information on population demographics, residency, movement and behavioral ecology are needed to create comprehensive conservation and management plans, especially when management resources are limited (see Chapters 3 and 4). A variety of terms describing site use have been used in the literature and I have summarized my use of these terms in Box 1 below. ‘Residency’ refers to the uninterrupted occupations of a limited area for a defined length of time (Chapman et al., 2015); ‘site fidelity’ to the return of an individual to a site of previous temporary residence after an absence at least equal to the time-frame of the previous site occupation (Chapman et al., 2015); and ‘site affinity’ to the repeated visitation of a site where individuals are moving through a large home range, but data are insufficient or do not support the designation of true site fidelity (Couturier et al., 2011; Jaine et al., 2014) (Box 1). 7 The life history of marine megafauna can also present inherent challenges to conservation. Most marine megafauna have long life-spans (e.g. leatherback turtles, Dermochelys coriacea > 40 years, Avens et al., 2009; white sharks, Carcharadon carcharias > 40 years, Bruce, 2008; humpback whales, Megaptera novaeangliae ≥ 95 years, Fleming et al., 2011; whale sharks ≥ 100 years, Perry et al., 2018;) and take many years to reach sexual maturity (e.g. humpback whales = 5 – 11 years; whale sharks > 20 years; white sharks > 12 years; leatherback turtles > 16 years). Further, fecundity in many marine mammals and some elasmobranchs is often low (e.g. humpback whales – one calf every two to three years; white sharks – up to 17 pups per litter every three years) with high maternal investment during gestation (e.g. humpback whale = 11 – 12 months; white shark = 18 months), and in some cases, also after birth (e.g. cetaceans nourishing young through lactation; escorting young to feeding grounds, Shane et al., 1986; and training young to hunt, Yeater et al., 2010). These life-history traits require conservationists and managers to enact conservation plans that address the different life stages, from the young and nurseries, to adults and reproductive and foraging grounds, for a rounded conservation strategy (Chapman et al., 2015; Lockyer, 1984; Ogburn et al., 2017). Different milestones along life-histories (e.g. parturition) and/or life stages in megafauna might occur in spatially diverse habitats (Chapman et al., 2015; Fleming et al., 2011; Heupel et al., 2007; Martins et al., 2018; Weishampel et al., 2003). In some megafauna species, reproductive (e.g. whale sharks, Rowat et al., 2012) and birthing grounds (e.g. manta rays, Stewart et al., 2018a), and where the young spend their early years in life (e.g. sea turtles Chelonioidea, Avens et al., 2009) still remain a mystery, hindering conservation planning. CHAPTER 1. GENERAL INTRODUCTION 8 Residency – the generally uninterrupted occupation of a limited area by an individual for a defined length of time (Chapman et al., 2015). A limited area, in this case, could be the home range of the individual or a core area that is regularly travelled. Home ranges can vary between species and life stages, thus descriptions of residency are more comprehensive when they are limited to a single species and ontogenic class, and are spatially and temporally specific (e.g. Caribbean reef sharks Carcharhinus perezihis of various life stages were typically year-round residents of Glover’s Reef Marine Reserve, Belize; Bond et al., 2012). Site fidelity – the return of an individual to a site of previous temporary residence after a periodic absence the same duration or longer than the residency period (Chapman et al., 2015). Site affinity – same site visitations where individuals are moving through a large home range or between core areas, but data are lacking/does not support true site fidelity as defined above (Couturier et al., 2011; Jaine et al., 2014). Box 1. Definitions and graphical depictions of the common site use terminology applied throughout this thesis. 9 1.3 MANTA RAY ECOLOGY All of the aforementioned conservation challenges apply to the world’s largest rays, the manta rays. They are large large-bodied, reaching up to 7 m in size (i.e. disk width), late to mature, have amongst the lowest fecundity of any elasmobranch species (Dulvy et al., 2014), are wide ranging and use varied habitats coastally and in offshore environments (reviewed in Stewart et al., 2018a). Thus, there is much to consider when developing comprehensive management plans for manta rays. These plans will require in situ data to understand local habitat use and identify key areas for prioritized protection. A lack of these data has slowed the identification of priority areas for protection and the creation and implantation of comprehensive management plans. While gaining momentum (Couturier at al., 2012), in-depth ecological studies for manta rays are in their infancy with less than 20 years of formal research. In fact, their taxonomic status is still being clarified with the genus Manta recently reclassified as Mobula (White et al., 2017). Under the former genus Manta, there were previously two recognized species Manta alfredi (Krefft, 1868) and M. birostris (Walbaum, 1792), with a potential third species proposed that is regionally endemic to the Gulf of Mexico, Caribbean and East Atlantic (Hinojosa-Alvarez et al., 2016; Hosegood et al., 2018; Marshall et al., 2009). Manta rays were subsequently grouped in the genus Mobula, where there are now eight recognized species (White et al., 2017) with manta rays reclassified as Mobula alfredi and M. birostris. Mobula alfredi is the better studied of the two formerly Manta species, partly because of their tendency to aggregate predictably in coastal waters. Thus M. alfredi are often referred to as ‘reef manta rays’. While displaying fidelity to key aggregation sites (Couturier et al., 2018; Deakos et al., 2011; Dewar et al., 2008; Stevens, 2016; van Duinkerken, 2010), this species is also wide CHAPTER 1. GENERAL INTRODUCTION 10 ranging, with long-range movements up to ~ 2,440 km (Arauz et al., 2019; Couturier et al., 2014; Germanov & Marshall, 2014; Jaine et al., 2014). The habitat ranges of both species of manta rays include the Coral Triangle, the world’s richest marine biodiversity region (Gray, 1997). Several aggregation areas for manta rays have been documented within the Coral Triangle (Couturier et al., 2012; Marshall et al., 2009; Figure 1.1), with Indonesia hosting a number of aggregations within designated conservation zones such as the Nusa Penida MPA, Komodo National Park (Komodo NP) and Raja Ampat (Conservation International, 2016; Heinrichs et al., 2011; Lawson et al., 2017). Additional aggregations within Indonesia are documented for Pulau Weh, North Sumatra; Sangalaki Island, East Kalimantan and Rote Island, Timor (Germanov & Marshall, 2014; Misool Foundation, pers. comm.). When manta ray fisheries were active in Indonesia, manta rays were landed in a number of additional sites (Lewis et al., 2015), some within potential movement corridors for manta rays (e.g. Tanjug Luar, Lombok and Lamakera, Solor, Germanov & Marshall, 2014). Although there is still much uncertainty around manta ray life history parameters, what is generally excepted is that both species are long-lived, can take a decade or more to become reproductively mature, and have exceptionally low fecundity (Dulvy et al., 2014; reviewed in Stewart et al., 2018a) – generally thought to give birth to a single pup every two to seven years. Manta rays are ovoviviparous, nourishing their young in utero with lipid-rich uterine milk during a 12-month gestation period (Dulvy et al., 1997; Marshall et al., 2010). While births in aquaria have been reported, the location of birthing grounds in the wild remains unknown. From observations in the wild and aquaria, and post-mortem analyses (compiled in Stewart et al., 2018a), full-term young of M. alfredi and M. birostris could be up to 46% the size of adults, signifying substantial maternal investment during gestation. For example, a 1.9 m pup was born to a 4.2 m sized adult in 11 Figure 1.1 The (A) global habitat range of Mobula alfredi as indicated by thatched lines and (B) known aggregation and fisheries landing sites sites within (orange) and adjacent to (green) the Coral Triangle region. Indonesian locations are listed from left to right as: a) Pulau Weh, b) Sangalaki Island c) Nusa Penida MPA, d) Tanjug Luar, e) Komodo National Park, f) Rote Island, g) Lamakera and h) Raja Ampat. the Okinawa Churaumi Aquarium, Japan (Reuters, 2007). In the wild, the smallest recorded are 1.3 m for M. alfredi and 2.0 m for M. birostris, while the size of maturity for females (depending on geographical location) are considered to be at least 3.2 and 4.1, respectively (Stewart et al., 2018a). Like most elasmobranchs (Carrier et al., 2004), there is no evidence that any post-natal parental care takes place. Pups are fully functional at CHAPTER 1. GENERAL INTRODUCTION 12 birth, are large and able to fend for themselves. Despite this, it is likely that manta rays are most vulnerable at this life stage (Heupel et al., 2007; Martins et al., 2018). Natural predators to manta rays include several large species of sharks (see Marshall et al., 2010) and orcas (Alava & Marlen, 2009). Given the low fecundity of manta rays, enhancing survivorship and limiting sources of mortality will be the most crucial conservation considerations (reviewed in Martins et al., 2018). Nurseries are thought to be crucial for extending the survival of young by providing an environment that reduces the time they spend at a smaller, more vulnerable size, such as by providing increased foraging opportunities and thermoregulation; and/or protection from predation (reviewed in Heupel et al., 2007). However, these benefits are not uniform across species (Heupel et al., 2007). The specific benefits provided by these proposed nurseries need to be considered on a case by case basis for each species/and perhaps for each distinct population across their range. Further, while many species of sharks and other elasmobranchs are now known to have nurseries, it must not be assumed that this is the case for all species or even for different populations of the same species (e.g. tiger sharks; Grubbs & Musick, 2007; McAuley et al., 2007). Caution is warranted in over-assigning the term ‘nursery’ to all areas where young are sighted (Beck, 2001; Heupel et al., 2007; Martins et al., 2018). From a conservation point of view, given that many species of sharks and rays have large home ranges, it is logistically complex and costly to protect large swaths of the marine environment, and attempting to do so might dilute resource and thus hamper conservation outcomes (Beck, 2001; Heupel et al., 2007; Martins et al., 2018). Further, larger areas would be more prone to resource use conflicts and as a result the political will to act for conservation may be lacking. Thus, it is considered a better conservation strategy to identify the most valuable habitats and narrow conservation strategies specifically to these areas (Heupel et al., 2007; Martins et 13 al., 2018). To that end a specific set of criteria have been proposed (Heupel et al., 2007; Martins et al., 2018) to better define nurseries for elasmobranchs. These criteria set out that areas constitute nurseries only where newborn or young-of-the-year (YOY) individuals (1) are more commonly encountered compared to other areas, (2) are found to be resident and/or return to for long time periods (3) and that this habitat use is consistent across years. In relation to manta rays, recent evidence suggests that juvenile M. birostris use shallow reef habitats as nurseries (Stewart et al., 2018b), and other reports hint that M. alfredi also likely use shallow protected areas, such as lagoons, as nurseries (Conservation International, 2015; McCauley et al., 2014; Stevens, 2016;). A greater understanding of whether nurseries exist and where potential nurseries are located is needed across the range of manta rays (see Chapters 3 and 4). 1.4 THREATS TO MANTA RAYS As of 2011 manta rays are listed as Vulnerable by the IUCN Redlist of Endangered Species based on global declining population trends (recently updated in Marshall et al., 2018a; Marshall et al., 2018b). Directed fisheries and by-catch are the biggest threat to manta ray populations worldwide (Couturier et al., 2012; Heinrichs et al., 2011; Stewart et al., 2018a; Ward-Paige et al., 2013) accounting for declines of as much as 71% in catch landings in Indonesia (combined M. alfredi and M. birostris) prior to the fishing ban (Lewis et al., 2015; White et al., 2006), and likely contributing to an 88% decline in M. alfredi sightings in Mozambique (Rohner et al., 2013). In 2013, in an effort to control this overexploitation driven by the high demand for mobulid brachial gill plates used in Chinese traditional medicine, manta rays were added to the Convention on International Trade of Endangered Species (CITES). The Convention on the Conservation CHAPTER 1. GENERAL INTRODUCTION 14 of Migratory Species (CMS) first listed M. birostris in 2011, and when M. alfredi were found to cross international borders (Couturier et al., 2014; Germanov & Marshall, 2014; Jaine et al., 2014), the species was listed in 2014. In the same year, following CITES listings and compelling economic evidence (O’Malley et al., 2013) to protect manta rays for manta ray tourism, Indonesia protected both species within its entire Exclusive Economic Zone (Ministry of Marine Affairs and Fisheries, 2014), effectively outlawing all directed fisheries, including small-scale and/or subsistence only fisheries (Ministry of Marine Affairs and Fisheries, 2016). However, enforcement is difficult throughout the waters of this vast archipelagic nation, and underreported manta ray catches either directed or from bycatch likely continues (Booth, 2016). While there are landing and recording regulations for licensed fishing boats that primarily target tuna (Ministry of Marine Affairs and Fisheries, 2016), many small-scale and artisanal fishers are unlikely to report incidental manta ray catches. Manta rays are also likely to come into contact with fishing gear, but not necessarily be lethally impacted (reviewed in Stewart et al., 2018a). In some cases, evidence of such entanglements might be seen (i.e. the manta ray is trailing a line or net, and/or have a resulting injury) and thus quantifiable. These observations could be used to gauge the extent of sub-lethal impacts that fisheries might be having on manta populations in specific aggregation areas (see Chapters 3 and 4). Other sub-lethal threats, as outlined by Stewart et al. (2018a), include those associated with habitat and individual disturbance and/or injury caused by divers/swimmers and tourism boats. Directed manta ray tourism has been actively promoted by both the Indonesian government (Ministry of Marine Affairs and Fisheries., 2016) and tourism industry with locations adjacent to manta ray aggregation sites seeing a spike in visitors (Barr et al., 2017; Hani et al., 2019). Closely monitoring the level of 15 tourism at manta ray aggregation sites for the benefit of adaptive management is necessary (see Chapters 3 and 4). Finally, plastic marine pollution is a major problem for Indonesia and the Coral Triangle region as a whole (Worm et al., 2017). This region is estimated to have high levels of plastic in its surface waters (Eriksen et al., 2014; van Sebille et al., 2015) and is adjacent to the majority of the world’s most polluting nations (Jambeck et al., 2015), with Indonesia ranking second to China. As large filter-feeders, manta rays are particularly susceptible to plastic ingestion (see Chapter 2). Estimating fine-scale plastic abundance levels in foraging habitats (see Chapters 3 and 4) is necessary to establish the potential implication of plastic pollution to manta rays and other filter-feeders, such as whale sharks (see Chapter 5). 1.5 AIMS AND OBJECTIVES With this thesis I aim to provide a basis of knowledge on M. alfredi (hereafter ‘manta rays’) ecology and identify and characterize existing sub-lethal threats to manta ray populations in two well-established conservation zones in Indonesia – Nusa Penida MPA and Komodo NP (Figure 1.2). Specifically, the objectives of this thesis are to: 1) Characterize manta ray habitat use and behavior and provide the first insights into population demographics and dynamics (Chapters 3 and 4), 2) Identify and characterize threats to manta rays within two conservation zones (Chapters 3, 4 and 5), 3) Identify and characterize the potential implication of microplastic ingestion by large filter-feeders and identify future research priorities (Chapter 2), 4) Characterize fine-scale plastic abundance in feeding grounds for manta rays and whale sharks within Indonesia (Chapter 5), and 5) Provide science-based recommendations for manta ray management (Chapter 6). 16 Figure 1.2 Schematic diagram of thesis structure. This schema will be repeated at the beginning of each chapter to orient the reader to the location of the particular chapter in the overall thesis framework. The position on the thesis roadmap is indicated in orange. 17 Briefly, the following approaches were used to meet the study objectives. Identification records of manta rays, which relied on substantial involvement of citizen scientists, were analyzed using maximum likelihood techniques, enabling me to gain further insight into manta ray ecology, population demographics, habitat use, movement and behavioral ecology in Nusa Penida MPA and Komodo NP (Figure 1.2). Additional observations allowed me to identify and characterize threats to manta rays within these zones. Plastic pollution is identified as a prominent threat to marine ecosystems in general, particularly in the Coral Triangle region, and was of particular interest to me to further characterize in the scope of its implications to large filter-feeders. I first explored the existing literature and compiled a synthesis of the known and potential implications of plastic ingestion – with a focus on microplastics. Using visual surveys and trawl data I then quantified plastic abundance in Indonesian manta ray and whale shark feeding grounds to a fine scale and provided theoretical ingestion estimates. Opportunistic collection of egested materials from manta rays provided evidence for plastic ingestion. 1.5.1 Thesis Structure The thesis is organized into six chapters, one is a synthesis and three are substantive data chapters (Figure 1.2), with the first and current chapter providing a general introduction to the thesis and the final chapter a general discussion. Each of the data chapters are written to be stand-alone, necessitating some repetition of the introductory material. Further, sections of this thesis were published prior to the completion of the thesis. With the rapid advances in the fields of both manta ray ecology and microplastics, the published chapters represent the available knowledge at that time. I have incorporated relevant recent literature in the yet unpublished Chapter 4, and the General Introduction (Chapter 1) and Discussion (Chapter 6). In Chapter 6, the final of CHAPTER 1. GENERAL INTRODUCTION 18 the thesis, I take the opportunity to discuss more recent findings and debates within the fields as it relates to the work presented. Earlier in this chapter I highlight the major challenges to marine megafauna conservation and management by focusing on manta rays. A lack of data regarding manta ray aggregations and fine scale ecology is a large barrier to effective conservation and management, one that I attempt to help bridge with Chapters 3 and 4 (Figure 1.2). Chapter 3 focuses on habitat use, population dynamics and behavioral ecology within the Nusa Penida MPA. The existence of nurseries, an outstanding and somewhat debated topic for manta rays, is discussed in relation to newly presented evidence in Chapters 3 and 4 indicating that nurseries are likely to exist within shallow coastal areas where there is reliable access to foraging opportunities. Chapter 4 builds on information from previous studies in the Komodo National Park (NP) expanding on the understanding of site use and providing insight into population demographics and dynamics. I have outlined the threats to manta rays in Indonesia above, and in this thesis focus on those that are sub-lethal or might occur within conservation zones. Contact between manta rays and fisheries continues with surviving individuals observed entangled in fishing gear or with injuries, some of them permanent. Chapters 3 and 4 provide information on anthropogenic injuries caused by fishing gear contact in the two conservation zones Nusa Penida MPA and Komodo NP. The growing popularity of manta ray tourism in Indonesia over the last decade or more, which until recently was largely unrestricted, is identified as a pressure on manta rays within the conservation zones studied in Chapters 3 and 4. The number of boats present at manta ray sites are presented to aid management considerations, which are discussed for the two regions in Chapters 3 and 4. Plastic marine debris is identified as a major threat to large marine filter-feeders in general, which is discussed in more detail in the synthesis Chapter 2. Chapter 2 identifies 19 microplastic hotspots as they relate to filter-feeding megafauna habitat ranges, and discusses the potential implications of microplastic ingestion based on the existing research, which thus far has focused on the potential of microplastic to transfer pollutants via ingestion. The format of this chapter is intended to be brief and provide an overview of some of the emerging themes on microplastics research as it relates to filter-feeding megafauna and act as a call to action for more dedicated research. Future research directions are outlined there and in Appendix I (published as part of a larger treatment on future research directions for mobulids). Some of these directions are investigated in Chapter 5, where I provide the first empirical assessments of plastic pollution in the feeding grounds for manta rays and whale sharks within three Indonesian locations (Nusa Penida MPA and Komodo NP for manta rays, and Pantai Bentar for whale sharks) (Figure 1.2). I explore the existence of seasonal and location trends in abundance and provide theoretical ingestion rates for manta rays and whale sharks. Recommendations for management and future research, especially with regards to microplastic, building upon newly available information in this quickly evolving field, are included as components of Chapters 3, 4 and 5 and an overall summary is provided in Chapter 6. The overview provided in Chapter 6 also assesses the strengths and weakness of citizen science, which played a substantial role in data collection supporting the findings of Chapters 3 and 4, for increasing knowledge on manta rays and enhancing their conservation. Finally, earlier in the present Chapter, I introduced and discussed the flagship approach to conservation, particularly as it as it relates to manta rays (and whale sharks) in Indonesia. This idea is further explored as it relates to marine plastic pollution in Chapter 2 and is a theme that features throughout the thesis. 20 CHAPTER 2. MICROPLASTICS: NO SMALL PROBLEM FOR FILTER-FEEDING MEGAFAUNA Right photo, © Simon Pierce 21 CHAPTER 2. MICROPLASTICS & MEGAFAUNA 22 2.1 ABSTRACT Microplastic pollution can impact filter-feeding marine megafauna, namely mobulid rays, filter-feeding sharks, and baleen whales. Emerging research on these flagship species highlights potential exposure to microplastic contamination and plastic-associated pollutants. Research and its wide communication are needed to understand the magnitude of the issue and improve marine stewardship. 2.2 INTRODUCTION Microplastics (Figure 2.1) have emerged as a major threat to the marine environment (Gall et al., 2015; Worm et al., 2017), highlighted at recent global marine health conferences (The Ocean Conference, held in June 2017 in New York, USA1 and the World Ocean Summit, held in February 2017 in Bali, Indonesia2). Microplastics are now present in every marine environment (van Sebille et al., 2015), easily permeate food webs, and are vectors for pollutants (Rochman et al., 2014; Worm et al., 2017). Despite the acceleration of research on microplastics in the marine environment, few reports focus on large organisms near the base of food webs and we have limited understanding of microplastic pollution threats to large filter-feeders (i.e. mobulid rays, filter-feeding sharks, and baleen whales). The first warning of this emergent threat was reported for baleen whales in 2012 (Fossi et al., 2012), while the ingestion of microplastics was confirmed only in 2015 (Besseling et al., 2015). Filter-feeding megafauna are particularly susceptible to high levels of microplastic ingestion and exposure to associated pollutants (Rochman et al., 2014; Worm et al., 2017) due to their feeding strategies (Paig-Tran et al., 2013), target prey (Setälä et al., 2014), and, for most, habitat overlap with 1 https://events.economist.com/events-conferences/asia/ocean-summit-2017/ 2 https://oceanconference.un.org/about 23 Figure 2.1 Classification, origins and trophic transfer of microplastics. Microplastic classification is based on (A) size in the environment, with microplastics defined as < 5 mm in size; and (B) size of manufacture and origin, with primary microplastics being those manufactured to be < 5 mm in size and secondary microplastics resulting from environmental degradation of larger plastic items. (C) Microplastics contain toxic plastic additives and adsorb persistent organic pollutants (POPs), such as such as dichloro-diphenyl-trichloroethanes (DDTs) and polychlorinated biphenyls (PCBs), that flow through marine food webs (depicted are the chemical structures of four common plastic-associated pollutants). Filter-feeding marine megafauna are particularly prone to microplastic ingestion and contamination by plastic-associated pollutants because of the large volumes of water they ingest during feeding. Modified, with permission, from The Ocean Cleanup, Copyright & Associates and designed by G.D. Amicis and M. Golasch. CHAPTER 2. MICROPLASTICS & MEGAFAUNA 24 microplastic pollution ‘hotspots’ or areas with relatively high levels of marine pollution (Figure 2.2, Table 2.1). Hotspot densities in this case refer to particle abundance of 105 or higher, which represent the levels estimated for the world’s five concentrating currents, the oceanic gyres, and the coastlines neighboring the world’s largest pollution emitters (Jambeck et al., 2015). Understanding the effects of microplastic pollution on filter-feeding megafauna is imperative because nearly half of the mobulid rays, two-thirds of filter-feeding sharks, and over a quarter of baleen whales are listed by the IUCN3 as globally threatened species and prioritized for conservation (Table 2.1). Many filter-feeding marine megafauna are charismatic and iconic species, with the potential to act as flagship species communicating awareness and stimulating community action (Bowen-Jones et al., 2002) to curb microplastic pollution. Here, we provide a synthesis of the demonstrated contamination of filter-feeding megafauna by microplastic pollution and its potential effects, and identify knowledge gaps, geographic hotspots for focused research, limitations, and challenges. Finally, we discuss how flagship species in conservation improve marine stewardship through the communication of scientific findings on threats to iconic marine megafauna and the broader marine environment. 2.3 SUSCEPTIBILITY TO MICROPLASTIC INGESTION Data-calibrated models have been developed to estimate the distribution of micro- plastics at the sea surface layer (van Sebille et al., 2015), with predictions of microplastic pollution hotspots, which include five oceanic gyres, semi enclosed basins (Gulf of Mexico, Mediterranean Sea, and Bay of Bengal) and biodiversity hotspots (Coral Triangle). In addition to these five oceanic gyres, four key regions stand out as high-priority areas for research and intervention based on our assessment of overlap between 3 www.iucnredlist.org 25 CHAPTER 2. MICROPLASTICS & MEGAFAUNA 26 (from previous page) Figure 2.2 Key buoyant microplastic hotspots overlap with habitat ranges of filte feeding marine megafauna. The habitat ranges for (A) Mobula (Manta) alfredi, (B) Rhincodon typus, and (C) Balaenoptera physalus, as indicated by thatched, lined, or dotted overlay, respectively, overlap with regions containing high levels of buoyant microplastic pollution. Adapted, with permission, from data-calibrated models of buoyant microplastic concentrations (van Sebille et al., 2015) and habitat range data from the IUCN (2017). Designed by V. Brakalova. Table 2.1 The world’s large marine vertebrate filter-feeders, their IUCN conservation status, and their presence in key microplastic pollution hotspots. Class and Species IUCN Status Population Trend Coral Triangle Gulf of Mexico Mediterranean Sea Rays Manta alfredi* Vulnerable Decreasing √ Manta birostris* Vulnerable Decreasing √ √ Mobula eregoodootenkee* Near Threatened Unknown √ Mobula hypostoma* Data Deficient Unknown √ Mobula japonica* Near Threatened Unknown √ Mobula kuhlii* Data Deficient Decreasing √ Mobula mobular* Endangered Decreasing √ Mobula munkiana Near Threatened Unknown Mobula rochebrunei* Vulnerable Unknown Mobula tarapacana Vulnerable Decreasing √ √ Mobula thurstoni Near Threatened Decreasing √ √ Sharks Cetorhinus maximus Vulnerable Decreasing √ √ √ Megachasma pelagios Least Concern Unknown √ Rhincodon typus Endangered Decreasing √ √ Marine Mammals Balaena mysticetus Least Concern Increasing √ Balaenoptera acutorostrata Least Concern Stable √ √ Balaenoptera bonaerensis Data Deficient Unknown √ Balaenoptera borealis Endangered Unknown √ √ Balaenoptera edeni Data Deficient Unknown √ √ Balaenoptera musculus Endangered Increasing √ √ Balaenoptera omurai Data Deficient Unknown √ Balaenoptera physalus Endangered Unknown √ Caperea marginate Data Deficient Unknown Eschrichtius robustus Least Concern Stable Megaptera novaeangliae Least Concern Increasing √ √ *Recent reclassifications within Mobulidae have resulted in major changes to the taxonomic arrangement of the family. The genus Mobula is now paraphyletic. Several species were synonomized, M. japonica with M. mobular, M. eregoodootenkee with M. kuhlii, and M. rochebrunei with M. hypostoma. For consistency and clarification purposes, the species names are kept as they appear in the current IUCN Red List. Data from The IUCN Red List of Threatened Species. Version 2017-1. . Accessed on 21 August 2017. 27 regions with estimated high levels of surface microplastics (van Sebille et al., 2015), mismanaged waste entering the ocean (Worm et al., 2017), and habitat ranges for filter-feeding megafauna (Figure 2.1, Table 2.1): the Mediterranean Sea, with high microplastic: plankton ratios (Collignon et al., 2012); the Gulf of Mexico, an emerging hotspot for microplastic pollution (Di Mauro et al., 2017); the Bay of Bengal, downstream of the Ganges river, one of many large rivers facilitating the transport of plastic to the marine environment from land-based sources (Jambeck et al., 2015); and the Coral Triangle, the region with the world’s richest biodiversity (Gray, 1997; Worm & Brach, 2012) and in close proximity to the world’s largest plastic pollution emitters (Jambeck et al., 2015). Therefore, filter-feeding megafauna resident in these areas have a high probability of ingesting microplastics, because they must filter hundreds to thousands of cubic meters of water daily to obtain adequate nutrition (Paig-Tran et al., 2013). They can ingest micro- plastics directly from polluted water or indirectly through contaminated plank- tonic prey (Setälä et al., 2014). The high plastic: plankton weight ratios (0.5) in the Mediterranean (Collignon et al., 2012) might lead to a significant reduction in nutritional uptake for filter-feeders, with animals feeding on the same quantities of particulate matter but receiving a lowered nutritional benefit. The estimated daily plastic ingestion rates for filter-feeding megafauna vary greatly, depending on location and feeding behavior, and range from as low as 100 pieces for whale sharks in the Gulf of California (Fossi et al., 2017) to as high as thousands of pieces for fin whales in the Pelagos Sanctuary (Fossi et al., 2012). Currently, the plastic ingestion rates by filter-feeding megafauna in the Gulf of Mexico, Bay of Bengal, and the Coral Triangle are unknown, as are the ingestion rates for mobulids anywhere in the world. CHAPTER 2. MICROPLASTICS & MEGAFAUNA 28 2.4 EFFECTS OF MICROPLASTIC INGESTION The effects of ingesting indigestible particles include blocking adequate nutrient absorption and causing mechanical damage to the digestive tract (reviewed in Kühn et al., 2015). Microplastics can also harbor high levels of toxic substances and persistent organic pollutants (POPs; Figure 2.1), and introduce these pollutants to organisms via ingestion (Rochman et al., 2014; Worm et al., 2017; Figure 2.1C). These pollutants can bioaccumulate over decades in long-lived filter-feeding megafauna, leading to a disruption of biological processes (e.g., endocrine disruption), and potentially altering reproductive fitness (Rochman et al., 2014). Furthermore, pollutants can be transferred or offloaded from mother to offspring (Lyons et al., 2013), potentially influencing the growth, survival, and reproduction of progeny. Recent research on large filter-feeders has found plastic additives and POPs in the muscle of basking sharks (Fossi et al., 2014), blubber of fin whales (Fossi et al., 2012; 2014), and skin of whale sharks (Fossi et al., 2017). Microplastic pollution has the potential to further reduce the population abundance of species that are already prioritized for conservation (Table 2.1). Although evaluating the full extent of microplastic (or any single contaminant) effects on population processes is challenging, Gall and Thompson (2015), in their review of the impacts of plastic debris on marine life, remind us that ‘a lack of evidence does not therefore necessarily imply a lack of effect’. 2.5 FUTURE DIRECTIONS Studies have yet to confirm directly that filter-feeding megafauna are exposed to POPs and other plastic-associated pollutants and additives through microplastic ingestion. Gaining direct evidence through conventional methods used to study diet, such as 29 stomach-content analysis or collecting egested material, relies on opportunistic observations and, thus, is not suitable for threatened marine megafauna. However, as the analytical approaches available to detect pollutants continue to increase, it is possible to analyze small amounts of tissue obtained nonlethally (i.e. via biopsies) and to test for plastic chemical tracers, such as phthalates, organobromines, or specific congeners of POPs (Fossi et al., 2012; 2014; 2017), allowing us to investigate correlations between microplastics in the feeding grounds of filter-feeders and the exposure of these organisms to pollutants. Understanding the effects of microplastic contamination through metabolomics studies and monitoring of biomarkers responses (Fossi et al., 2017) can help to shed light on the health of populations in response to plastic-associated pollutants. Long-lived species that are resident in specific regions can be monitored throughout their lives, providing an indication of pollutant exposure over time. The levels of pollutants, especially those that are unique to plastics, in resident large filter-feeders might provide indirect indicators, or sentinels, for microplastic pollution in local marine environments. Given that plastic production is projected to increase globally (MacArthur et al., 2016), the establishment of long-term monitoring programs is needed. Priority regions for monitoring the exposure and impact of microplastic ingestion on filter-feeding marine megafauna are the Gulf of Mexico, the Mediterranean Sea, the Bay of Bengal, and the Coral Triangle (Figure 2.2, Table 2.1), which contain populations of filter-feeding megafauna and their critical habitats, including feeding and breeding grounds. Many areas in these regions are biodiversity hotspots (Gray, 1997; Worm & Branch, 2012) and of economic importance (e.g., fisheries and megafauna-focused tourism), warranting increased prioritization for intervention. Intervention strategies might include dissemination of information gleaned from monitoring programs to governing bodies and raising the awareness of local communities in an effort to improve waste management practices. CHAPTER 2. MICROPLASTICS & MEGAFAUNA 30 2.6 VALUE OF FOCUSING ON MEGAFAUNA Many filter-feeding megafauna are charismatic and iconic species that can serve as flagship species for marine conservation. Filter-feeding megafauna are also ideal umbrella species candidates (Bowen-Jones & Entwistle, 2002) due to their wide habitat ranges, including coastal and pelagic environments. While umbrella species are useful for directing intervention strategies that cascade through marine ecosystems for the benefit of the systems as a whole, flagship species can provide a mechanism for communicating awareness and stimulating action to tackle marine plastic pollution. For conservation strategies to work with local communities, the flagship species chosen need to have significance for local people as well as for international bodies and potential financial donors (Setälä et al., 2014). Filter-feeding megafauna, such as manta rays and whale sharks, fit this criteria well because they are prioritized for conservation (Table 2.1), internationally protected4,5, and form the basis for significant, local tourism economies in developing countries (Higham et al., 2016) struggling with waste management (Worm et al., 2017). Over the past decade, marine tourism (of which marine megafauna tourism, including whale watching, snorkeling, and SCUBA diving with manta rays and whale sharks, is a large sector) generated more global revenue than aquaculture and fisheries combined (Higham et al., 2016). Thus, it is in the best interests of local communities and governments to protect these species and their habitats, to safeguard a sustainable ‘Blue Economy’ and future livelihoods. To ensure that the full benefit of focusing research and community engagement on flagship species is realized, researchers should work collaboratively with local communities and incorporate local outreach during their research. Social science studies to assess human attitudes to plastic waste and consumption could be run in parallel to 4 http://checklist.cites.org/ 5 www.cms.int/en/species 31 filter-feeding megafauna research programs. This will facilitate the engagement of communities in the conservation work, while raising awareness of the issues of marine plastics. Using iconic filter-feeding megafauna as flagships highlights the impacts of microplastics on marine life, including marine food resources, and is likely to increase public awareness and enhance stewardship of marine ecosystems. 32 CHAPTER 3. CONTRASTING HABITAT USE AND POPULATION DYNAMICS OF REEF MANTA RAYS WITHIN THE NUSA PENIDA MARINE PROTECTED AREA, INDONESIA Image: Giulia De Amicis 33 CHAPTER 3. NUSA PENIDA 34 3.1 ABSTRACT Manta rays (Mobula spp.) are highly valued in nature-based tourism globally. In Indonesia, although manta rays are protected, critical information is lacking on their habitat use, population dynamics and movements. We investigate the population structure and residency patterns of reef manta rays Mobula alfredi in the Nusa Penida Marine Protected Area (MPA). From photo-identification data logged by citizen scientists and trained observers (mantamatcher.org), we identified 624 reef manta rays from 5,913 sightings (January 2012 – April 2018) based on their unique ventral coloration patterns. Year-round records were collected from two shallow (< 20 m) reefs – Manta Bay (MB; n = 3,029 sightings) and Manta Point (MP; n = 3,058) – that are used frequently by tourism operators. Maximum likelihood techniques and a Markov movement analysis were used to model residency patterns and movement between these sites within the MPA. Manta rays at MB were predominantly male (64%, n = 261 individuals), with immature males (14%, n = 59) being sighted most frequently (39%, n = 1,170). In contrast, few immature individuals were sighted at MP (6%, n = 28), and they were sighted on few occasions (2%, n = 45), while mature female manta rays comprised 26% (n = 127) of the MP community and were the most frequently sighted (48%, n = 1,413). Lagged identification rates indicated high site fidelity at each location. However, 44% (n = 278) of individuals moved between the two sites and cumulative discovery curves showed a continued recruitment of individuals over the 6 years of the study. In addition, the behaviors displayed by the manta rays differed markedly between the two sites: MB appears to be a foraging ground, especially for juveniles, and potentially a nursery, while MP is used mainly for cleaning and courtship, indicating a social and reproductive site. Reproductive behavior coincided with the peak annual sightings in May. To prevent disturbance to this threatened species by tourism, regulations for the number of boats and interactions, 35 especially during key reproductive times should be considered. Further, strict fishing regulation in the area is recommended as fishing gear entanglement was identified as a threat to these aggregations. 3.2 INTRODUCTION Many marine megafauna species, such as sea turtles (Bowen et al., 2007), cetaceans (Baird et al., 2008), large teleost fish (Robichaud et al., 2001; Rooker et al., 2008), and elasmobranchs (Hueter et al., 2005), range widely, but within these large areas show high fidelity to defined areas for reproduction and feeding (Chapman et al., 2015). Further, a specific demographic within a population might depend on limited habitats during different stages in their development (e.g., coastal nurseries versus open ocean foraging grounds) (Chapman et al., 2015; Hueter et al., 2005). While Marine Protected Areas (MPAs) are often created to protect critical habitats of threatened species, these areas are sometimes designated without in-depth knowledge of how the habitats are used by the target species. An understanding of habitat use is needed to achieve conservation goals, which can also require that tourism and other threats are managed to avoid disruption to the biology and behavior of the threatened species in question (e.g., Hueter et al., 2005). Manta rays are large, pelagic, filter-feeding mobulid rays (Mobula spp.) found in equatorial and tropical waters (e.g. Couturier et al., 2012), that are threatened with extinction according to the IUCN (Marshall et al., 2018a; 2018b). Several known aggregation areas exist within the Coral Triangle region, particularly within Indonesia (Couturier et al., 2012; Germanov & Marshall, 2014; Marshall et al., 2009). Manta rays are long-lived, slow to mature, and exhibit low fecundity (Marshall & Bennett, 2010; Rambahiniarison et al., 2018; Stevens, 2016). Owing in part to these conservative life CHAPTER 3. NUSA PENIDA 36 history traits and to overfishing across their range (Dulvy et al., 2014), both Mobula alfredi (Krefft, 1868) and the larger M. birostris (Walbaum, 1792) are currently listed as vulnerable to extinction on the IUCN Red List of Endangered Species (Marshall et al., 2018a, b). Concerns over the population status of manta rays led to their listing on the Convention on the International Trade of Endangered Species (CITES) and the Convention on the Conservation of Migratory Species (CMS), which aim to curb their consumptive over-exploitation by fisheries (Ward-Paige et al., 2013). Until recently, Indonesia was ranked third for the highest annual mobulid catches, including manta rays (Heinrichs et al., 2011). In 2014, Indonesia protected both manta ray species within their entire exclusive economic zone (an area of over 6 million km2) through the Ministerial Decree of Marine Affairs and Fisheries No. 4/2014 (Ministry of Marine Affairs and Fisheries, 2014), in the hopes of slowing population declines. Currently, limited information is available for Indonesian manta ray populations to assist in tracking the current status of these slow-growing species. Manta ray protection within Indonesia came largely in response to the growth of the manta ray tourism industry. Indonesia is ranked second in the world for manta ray tourism, with an estimated value of USD $15 million per year (O’Malley et al., 2013). Prior to their national protection, manta rays were first protected within three sanctuaries east of Bali: The Nusa Penida MPA (200 km2), West Manggarai including Komodo National Park (7,000 km2), and Raja Ampat (11,655 km2). Together, these sites encompass the bulk of the manta ray tourism industry in Indonesia (O’Malley et al., 2013). In addition to CITES and CMS regulations in Indonesia, evidence-based conservation strategies are needed to protect the remaining manta ray populations from the potential threats associated with unregulated tourism (reviewed by Stewart et al., 2018a; Trave et al., 2017; Tyne et al., 2014). Globally, as awareness grows of the chronic 37 stresses to manta rays associated with an increase of tourism, measures aimed at curbing these impacts are underway or planned in some popular manta ray dive sites (e.g., Raja Ampat, Indonesia, Kasmidi & Gunadharma, 2017; and West Hawaii, United States, Division of Boating and Ocean Recreation, 2016). Other hazards to manta rays (reviewed by Stewart et al., 2018a) include entanglement in fishing lines and boat mooring lines (Croll et al., 2016). The Nusa Penida MPA, located 18 km south-east of Bali Island (Figure 1), encompasses the habitats for M. alfredi (hereafter manta rays). The waters of the Nusa Penida MPA, established in 2010 (District Fisheries and Marine Agency, 2010), are very productive (Ayers et al., 2014), with complex oceanographic conditions owing to their proximity to a major channel of the Indonesian Throughflow (ITF) current (Murray et al., 1988; Nyegaard, 2018). The islands of Ceningan, Lembongan, and Penida, located within the MPA, are home to 48,000 people and have been a tourist destination since the 1990s. In 2015, the MPA received an estimated 200,000 tourists per year, with an annual estimated increase of 7.7% (Barr et al., 2017). Manta ray watching is a major drawcard for tourist visits to the MPA, with annual estimates ranging from 4,200 (Barr et al., 2017) to 10,440 people (Aquatic Alliance, pers. comm.) or ~ 63,500 manta dives and ~ 15,000 manta swims (O’Malley et al., 2013). As manta rays are present in the MPA year-round (Germanov & Marshall, 2014), this manta ray population is subjected to chronic pressure from tourism. Currently, the number of boats allowed to enter manta ray habitats is unregulated, as is general boat traffic in the area (Barr et al., 2017). Further, codes of conduct for manta ray interactions are voluntary and are not strictly enforced, leading to concerns over low compliance (Barr et al., 2017). Moreover, basic information about the local manta ray population structure, habitat use, and movement patterns is lacking in the region, yet is required to refine spatial management within the MPA and promote CHAPTER 3. NUSA PENIDA 38 ecologically sustainable tourism (Trave et al., 2017). Here, we used a citizen science approach accessing over six years of publically contributed photographs and individual-based sighting records of manta rays within the Nusa Penida MPA, which were logged on the online database ‘Manta Matcher’ Wildbook – www.mantamatcher.org (Marshall & Holmberg, 2018), to investigate population dynamics and habitat use. The aim of this work is to inform evidence-based management plans to ensure the long-term sustainability of manta ray tourism within the Nusa Penida MPA for both operators and manta rays. This was done by comparing the habitat use and population dynamics at two, high-intensity tourism sites, located only 12 km apart, to better understand the fine-scale habitat use of the manta rays in the region. Given the proximity of the two sites, we predicted that they would share common aspects of population demography and behavior. With this study, we provide in-depth information on: (1) population structure, (2) injury rates, (3) fine scale habitat use, (4) local movement, and (5) behavior, and provide specific recommendations for effective manta ray management within the Nusa Penida MPA. 3.3 METHODS 3.3.1 Description of Study Sites The Nusa Penida MPA is located in a unique oceanographic region, with a shallow (200 m) continental shelf adjacent to deep water basins of up to 4,200 m depth. The three islands located within the Nusa Penida MPA (Figure 3.1, created using QGIS Development Team, 2016, v 2.18) are in the direct path of the ITF current due to their proximity to the Lombok Strait (Murray & Arief, 1988). This major current separates two 39 Figure 3.1 Location of study sites in the Indonesian archipelago along the west coastline of Nusa Penida, Manta Bay (MB), and Manta Point (MP) shown in relation to Bali and the Lesser Sunda Islands, Indonesia. Popular SCUBA diving locations around the Nusa Penida MPA with infrequent manta ray sightings are also marked as described in the legend. Sites marked as “other” are ‘Pontoon,’ ‘Toyapakeh,’ ‘Sekolah Desa,’ ‘Pura Ped,’ ‘Sental,’ and ‘Pura Mas Gading.’ Bathymetry information was available from: GEBCO_2014 Grid, version 20150318; www.gebco.net. significant marine and terrestrial biogeographical realms (i.e. the ‘Wallace Line,’ Barber et al., 2000). The current flows largely from north to south between the larger islands of Bali and Lombok, with its strength varying seasonally (i.e. the north-west vs. south-east windward seasons, (Mayer et al., 2012). The ITF brings warmer water (Tillinger, 2011) and nutrients (Ayers et al., 2014) from the Pacific to the Indian Ocean. Deep water upwelling (Ningsih et al., 2013) south of Bali further contributes nutrient rich waters to the region, creating productive foraging grounds within the Nusa Penida MPA for sunfish (Nyegaard, 2018; Thys et al., 2016; Tito et al., 2017), and potentially other megafauna CHAPTER 3. NUSA PENIDA 40 like manta rays. Although 11 sites are dived frequently (number of days > 20 per year) in the Nusa Penida MPA, the majority of manta ray sightings (~ 93%), based on year-round dive logs for 2016 (n = 809) and 2017 (n = 753), by one dive operator (Figure 3.2), were at two sites, Manta Bay (MB) and Manta Point (MP) (Figure 3.1). Thus, these two sites were selected for detailed investigation. Both are shallow bays (7 – 25 m depth) fringed by steep rocky cliffs and deep water, separated by a straight-line distance of ~ 12.4 km. A range of habitats used by manta rays, including cleaning stations, foraging aggregation areas, and reproductive grounds are found at the two sites (Germanov & Marshall, 2014). 3.3.2 Data Collection The online database ‘Manta Matcher’ Wildbook (Marshall & Holmberg, 2018) was used to collate data (date, time, location, and identifying photographs of manta rays) from the public and researchers (Germanov & Marshall, 2014) (Figure 3.3). Citizen science was an important part of this process and has been demonstrated to provide basic information when data are limited (e.g. Couturier et al., 2014; Germanov & Marshall, 2014; Jaine et al., 2012; Norman et al., 2017; Pierce et al., 2018; Robinson et al., 2016; Rohner et al., 2013) and engage local communities in marine stewardship (Miller-Rushing et al., 2012). Approximately 40 observers were trained to take photos of manta ray ventral spot patterns, record sex and maturity data, estimate size, and to note relevant behaviors (discussed below), as well as the maximum number of boats present at the sites. Tourists were regularly informed about ‘Manta Matcher’ through dive operator briefings and weekly educational presentations starting in 2012, resulting submissions from 276 different emails. A more traditional sampling approach with structured surveys (e.g. Couturier et al., 2014; Deakos et al., 2011) is required for a traditional mark- 41 recapture approach to population analyses to estimate abundance and survival. Photographs of the ventral surfaces of individual manta rays (Couturier et al., 2014; Deakos et al., 2011; Kitchen-Wheeler et al., 2011; Marshall et al., 2011) were logged year-round by visiting tourists to the Nusa Penida MPA and members of the local dive community (i.e. residents and dive professionals). The earliest available photographs and sighting records date back to June 2004, while year-round survey effort by trained observers started in 2012 (Table 3.1). Only records collected between January 2012 and April 2018, when greater than 400 sightings were logged in any one year, were used for the analyses. Further, data from 2012, 2015, and 2018 were excluded from seasonal analyses as year-round survey effort was lower, or not available, during these years. For the purposes of this study, we assign a survey day as the unit of survey effort, as the number of dives per day and hours of diving were not recorded on a regular basis. Figure 3.2 Dive operator logs of dives around the Nusa Penida MPA and those with manta ray sightings. The percent of total dives logged (top; n = 1,562) and of dives with manta ray sightings per location (bottom; n = 327) based on 2016 and 2017 logs from one dive operator. MB = Manta Bay; MP = Manta Point; BY = Buyuk; CB = Crystal Bay; MG = Mangrove; Other = six locations on the north coast of Nusa Penida. CHAPTER 3. NUSA PENIDA 42 Figure 3.3 The combined number of annual survey days at manta ray sites Manta Bay and Manta Point by trained observers (Logs) and the manta ray survey days deposited to ‘Manta Matcher’ between 2012 and 2018 in the Nusa Penida MPA. Note: 2018 = January to April only. Table 3.1 Individual manta rays, sightings and survey days, recorded between June 2004 and April 2018, within Nusa Penida overall, Manta Bay (MB) and Manta Point (MP). Daily duplicates are removed (n = 163). Year Individuals Sightings Survey Days All MB MP All MB MP All MB MP 2004 - 11 136 29 122 174 34 140 72 18 55 2012 186 146 56 506 448 58 99 89 20 2013 315 203 166 1092 846 246 149 125 46 2014 332 203 189 1219 833 386 173 137 62 2015 261 76 216 563 121 442 87 32 64 2016 336 104 278 990 262 728 136 51 96 2017 368 147 280 1131 354 777 200 107 132 2018* 239 87 165 412 131 281 49 22 36 TOTAL 624 497 494 6087 3029 3058 965 581 511 *To April only 43 All sighting records were validated manually for accuracy, completeness, and adequate photo quality by trained observers. Sightings lacking information on location or date and those with an indistinct ventral image of the manta ray were excluded. Photographs were manually matched to an ID catalog with the assistance of built-in (Germanov & Marshall, 2014; Town et al., 2013) and/or external software (‘MantaUtil,’ Winstanley, 2016) by trained observers and error-checked by a regional research manager, who assigned new individuals if no match was found. The sex and maturity status of manta rays were assessed based on the absence (female) or presence (male) of claspers, clasper size and/or clasper scarring for males, and pregnancy bulge and/or presence of pectoral fin mating scars for females (Marshall & Bennett, 2010; Marshall et al., 2011). Immature females were not assigned throughout the study because of the diffculty in definitively classifying females lacking maturity indicators in the absence of accurate size measurements. Manta ray sizes, based on disk width, were estimated relative to an entity of known size (i.e. swimmer or dive equipment) and placed into 0.25 m size classes. Only size estimates recorded by two observers that spanned the entire study period were considered (n = 260) in an effort to reduce observer bias. Injuries to the manta rays, including cephalic and pectoral fin truncations or disfigurement, and fishing line cuts and entanglements, were also noted (Figure 3.4). Manta ray behavior was assessed from sighting data based on mouth and cephalic fin positioning, the presence and absence of cleaner fish, and records of social interactions, including mating trains as described by Deakos, (2012), Jaine et al. (2012), Marshall and Bennett (2010), and Stevens et al. (2018). Based on these previous studies, behavior was categorized into four mutually exclusive categories: foraging, cleaning, cruising, and courtship (Figure 3.5). Two additional non-exclusive behavior categories were included – foraging/cleaning and courtship/cleaning to account for instances when CHAPTER 3. NUSA PENIDA 44 an individual was observed performing two behaviors within the same encounter (maximum 60-min dive time). ‘Foraging’ manta rays were defined as individuals with open mouths and unrolled cephalic fins near the surface of the water column, or those who were completing full vertical rotations (‘barrel-rolling’). ‘Cleaning’ manta rays were defined as individuals hovering over reef patches (‘cleaning stations’) with cleaner fish species (Chaetodon kleinii, Thalassoma lunare, Labroides bicolor, and Labroides dimidiatus) in close proximity. ‘Cruising’ manta rays were those sighted with closed mouths and rolled up cephalic fins. ‘Courtship’ behavior was recorded when several male manta rays pursue a mature female (in the absence of foraging behavior), or when two or more individuals gave acrobatic displays as described by Stevens et al. (2018). 3.3.3 Analyses of Residency and Movement We use the term ‘site fidelity’ to denote the return of an individual to a site of previous residency after a periodic absence greater or equal to the residency period, following Chapman et al. (2015). However, a caveat of solely using this terminology is that some individuals might regularly use both study areas within the Nusa Penida MPA, moving frequently between the two. Further, periods between sightings varied greatly between individuals, precluding us from assigning a standard residency and absence period. In these uncertain scenarios, where individuals could be moving between areas across the large home range, we use the term ‘site affinity’ to describe our results (Couturier et al., 2011; Jaine et al., 2014). 45 Figure 3.4 The number of individual manta rays observed with permanent injuries, including pregnant manta rays. Photos are copyright of the Marine Megafauna Foundation and the following individuals (from top): 1) Peter Bassett, 2) Jennifer Ward, 3) Stuart Ecob 4) Elias Sinderson, and 5) Lillian Ellevog. CHAPTER 3. NUSA PENIDA 46 Figure 3.5 Example photographs and observational attributes indicative of manta ray behavior. Photos are copyright of the Marine Megafauna Foundation and the following individuals (from top): 1) E. Germanov, 2) Lillian Ellevog, 3) E. Germanov, 4) Nicholas Longfellow, 5 -7) E. Germanov. 47 A modified maximum likelihood approach was used to compare manta ray re-sighting data against residency models, implemented in the program SOCPROG 2.8 (Whitehead, 2009). These statistical models were previously used for manta rays in Hawaii by Deakos et al. (2011) and in several more recent studies on whale sharks Rhincodon typus (Fox et al., 2013; McCoy et al., 2018; Robinson et al., 2016) and cetaceans (e.g. Chabanne et al., 2017). This approach determines the spatial and temporal distribution of sampling effort using the re-identification data itself, making this approach suitable for opportunistic and presence-only sighting data (Fox et al., 2013). Residency patterns for individuals within the study sites were investigated by using the ‘Movement Analyses’ module of SOCPROG 2.8 to calculate the lagged identification rate (LIR), the probability of re-sighting an individual manta ray after a variable lag time (Whitehead, 2009). These empirical LIR data, based on a per day time lag, were then compared to a series of model scenarios encompassing both closed and open populations. Open population models were developed by incorporating varying situations with emigration, immigration, re-immigration, and/or mortality (Table 3.2). The lowest value from the quasi-Akaike information criterion (QAIC), used to account for over-dispersion of the data (Whitehead, 2007), was used to select which model best approximated the residency characteristics of manta rays at each study site. The sole exception was for female manta rays at MP, where AIC was used to select a best-fit model, as the data were not over- dispersed. The QAIC difference (or ΔQAIC) between the best-fit model and any other indicates how well the data support the less favored model as follows: ∆QAIC < 2 = substantial support; ∆QAIC 4 – 7 = considerably less support; and ∆QAIC > 10 = essentially no support (Whitehead, 2007). The LIR analyses were extended to a ‘within/between’ analysis to estimate the probability that an individual manta ray, identified first within one site, will be re-sighted at the other site after a specified time lag CHAPTER 3. NUSA PENIDA 48 (in days). This analysis effectively provides a significance test for population-level mixing between sites. Model fits were bootstrapped 1,000 times to generate standard errors (SE). The transition probabilities between sites were calculated using a parameterized Markov movement model (Whitehead, 2009), in which an individual has a certain probability of moving from one area to another at the specified time lag (one year; Table 3.3). This model includes movements to and from both MB and MP and also accounted for movements to a third, hypothetical area (i.e. an area ‘outside’ of MB or MP). Optimized values of transition probabilities were bootstrapped 1,000 times to generate SEs and the maximum number of evaluations was set to 100,000. Mortality, including permanent emigration from all areas within the MPA, was considered in the model. 3.3.4 Statistical Analyses The correlation between the annual number of survey days and sightings was estimated using Pearson’s product-moment correlation, cor.test function of the R statistical software (R Core Team, 2018). Differences in the number of individuals and sightings between the sexes at MB and MP were tested using chisq.test function of the R statistical software. Chi-squared (χ2) goodness of fit tests (one-dimensional contingency table) were used to compare sex ratio data for each site, while a two-dimensional contingency table was used to compare sex ratios and behavior frequencies between the two sites. Behavior data were collapsed into four categories to facilitate χ2 testing where counts were less than five in either of the site categories (i.e. foraging/cleaning was reclassified to foraging; and courtship/cleaning was reclassified to courtship). The Yates’ continuity correction was applied to tests where there was one degree of freedom. 49 Table 3.2 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of (A) all manta rays, (B) males and (C) females at Manta Bay (MB) and Manta Point (MP) in the Nusa Penida MPA, Indonesia. Model Model description A Closed (1/a1 = N) B Closed (a1 = N) C Emigration/mortality (a1 = emigration rate; 1/a2=N) D Emigration/mortality (a1 = N; a2=mean residence) E Closed: emigration + re-immigration (a1 = emigration rate; a2/(a2 + a3) = proportion of population in study area at any time) F Emigration + re-immigration (a1 = N; a2 = res time in; a3 = res time out) G Emigration + re-immigration + mortality H Emigration + re-immigration + mortality (a1 = N; a2 = res time in; a3 = res time out; a4 = mort) A) All B) Males C) Females Location MB MP MB MP MB MP Model ΔQAIC ΔQAIC ΔQAIC ΔQAIC ΔQAIC ΔAIC A 4291.77 779.72 3543.97 104.11 782.86 662.31 B 4291.77 779.72 3543.97 104.11 782.86 662.31 C 1.92 21.90 63.25 6.46 11.08 21.32 D 1.92 21.90 63.25 6.46 11.08 21.32 E 4228.42 721.63 0 89.63 760.06 622.28 F 3.92 5.33 65.25 6.68 13.08 0 G 6881.22 25.9 62.24 10.46 0 25.32 H 0 0 61.28 0 13.27 3.76 N = population; res = residence Table 3.3 The estimated probability (± 1 SE) of an individual manta ray originally identified from one area being re-sighted within the same or different area. Movement probabilities as calculated by the Markov movement model are presented for (A) all manta rays (n = 624), (B) males (n = 342) and (C) females (n = 244). From Area: To Area: MB MP Outside A) All MB 64.7% 35.3% ± 9.9% 0.0% ± 1.6% MP 16.4% ± 6.5% 73.7% 9.9% ± 15.7% Outside 3.0% ± 3.2% 3.4% ± 9.8% 93.7% B) Males MB 53.1% 44.7% ± 5.0% 2.2% ± 1.8% MP 24.5% ± 2.8% 75.5% 0.0% ± 2.2% Outside 0.0% ± 0.2% 0.2% ± 13.0% 97.9% C) Females MB 59.5% 40.5% ± 5.6% 0.0% ± 0.8% MP 8.1% ± 2.2% 85.8% 6.1% ±5.6% Outside 1.5% ± 1.8% 0.1% ± 5.7% 97.9% CHAPTER 3. NUSA PENIDA 50 3.4 RESULTS 3.4.1 Sightings and Survey Effort Data from 6,087 sightings (excluding daily duplicate data for the same individuals; n = 163) of manta rays were collected within the Nusa Penida MPA from 25 June 2004 to 9 April 2018. Sightings came from 965 unique dates, with ~ 97% of the records (5,913 sightings, excluding daily duplicates) logged between January 2012 and April 2018 and from our two main focal sites of MB and MP (Table 3.1). Data prior to 2012 allowed 136 individual manta rays to be identified from 174 sightings within the MPA. After 2012, as education about manta rays amongst tour operators in the area improved, the number of manta ray identification photos and sighting information available increased (see Figure 3.3). The average annual sighting rate from 2012 until 2017 (i.e. the full survey years) was 917 sightings/year ± 125 (± 1 SE), while 412 sightings were recorded for the first quarter of 2018. The number of annual manta ray sightings varied across the study period (Figure 3.6A and Table 3.1), with sightings being positively correlated with survey effort (r = 0.92, p < 0.01) for all study years. As the days where no manta rays were sighted (i.e. absence data) are not logged with ‘Manta Matcher,’ we used survey dives logged by trained observers as a proxy for survey effort (Figure 3.3). The number of annual manta ray sightings submitted to ‘Manta Matcher’ varied between MB and MP. In the earlier study years (i.e. 2012 – 2014), more sightings were recorded (daily mean and total) at MB than MP, while there were more sightings at MP in the later years (i.e. 2015 – 2017) (Figure 3.6A and Table 3.1). These observations reflect the gradual increase in reported surveys at MP and the steep decrease of reported surveys at MB after 2014, influenced by a shift in site usage by tourism operators (i.e. more dive operators using MP and more snorkel operators using MB; E. Germanov, pers. obs.). 51 Figure 3.7 The number of survey days (n = 659) logged by trained observers for each month between 2012 and 2017 in the Nusa Penida MPA. *The average per month is pooled and calculated across years (excluding prior 2013 and 2015 due to lower year-round effort). Figure 3.6 The average annual (A) and monthly (B) manta ray sightings per unit effort (day) recorded between 2012 and 2017 in the Nusa Penida MPA at Manta Bay (MB) and Manta Point (MP) presented as the daily means (± 1 SE). Data from 2012 and 2015 is excluded from (B) due to lower year-round effort. (A) Sightings (MB = 2,995, MP = 2,918), days (MB = 581, MP = 511). (B) Sightings (MB = 2,295, MP = 2,137), days (MB = 420, MP = 336). CHAPTER 3. NUSA PENIDA 52 Manta ray sighting trends were considered in relation to survey effort. Logged presence/absence data indicated variable survey effort by trained observers between years and months (Figure 3.7), with a lack of complete year-round effort in 2012 and 2015. Further, annual weather patterns, such as the north-west monsoon season (December – January), that result in unfavorable weather conditions and limit access to the study sites, regularly contributed to a seasonal reduction in survey effort. Nevertheless, there was an approximately two-fold increase in manta ray sightings for MP in the month of May (Figure 3.6B). In contrast, sightings at MB were only marginally higher from February to April and in July than other months of the year. A total of 624 individual manta rays were identified from sightings within the Nusa Penida MPA between 2012 and 2018. Of these individuals, 407 were sighted at MB and 494 were sighted at MP, with 277 sighted at both sites. A discovery curve shows a steep rise in newly identified manta rays within the Nusa Penida MPA until approximately 400 individuals, but the curve kept increasing to 624 individuals, though at a slower rate, until 2,260 days, with no evidence of an asymptote (Figure 3.8). When treated separately, neither of the discovery curves for the two study sites show signs of approaching an asymptote. However, a brief period of slower discovery (i.e. a plateau) of new individuals was observed at MB after the identification of approximately 280 individuals at 800 days, until 1,600 days or between April 2014 and December 2015 (i.e. years two and three of data collection), after which the numbers of individuals identified increased to 407 after 2,260 days. 53 Figure 3.8 Discovery curves for newly identified manta rays against elapsed days. Discovery curves are presented for the Nusa Penida MPA (all) and broken down by sites Manta Bay (MB) and Manta Point (MP) from January 2012 until April 2018. The x-axis gray lines indicate survey years. NS = number of survey days and NI = number of individuals. 3.4.2 Population Structure Dataset allowed for characterization of the demographic composition at both MP and MB. Overall, more male (n = 344; 55%) than female (n = 243; 39%) manta rays were identified in the Nusa Penida MPA (χ21 = 17.378, p < 0.001; Figure 3.9), while the sex was unknown for 37 individuals (6%). There was a significant association between sex and site for individuals (χ21 = 13.864, p < 0.001), and between sex and site for sightings (χ21 = 1080.2, p < 0.001). Of the entire population, at least 53% (n = 129) of the females were considered to be sexually mature, based on their pregnant appearance or the CHAPTER 3. NUSA PENIDA 54 presence of mating scars. A total of 100 females (41%) were recorded as pregnant throughout the study. Among the males, 81% (n = 277) were considered sexually mature based on the size and shape of their claspers, with the remaining 19% (n = 66) being immature. In addition, 123 manta rays (~ 20% of the total identified) were classed as immature at some point throughout the study, with 57 known to reach maturity. The population structure of individuals and the demographics of sightings differed between MB and MP (Figure 3.9). At MB, males made up 64% (n = 261) of identified individuals, whereas only 29% (n = 120) were female (χ21 = 52.18, p < 0.001). Males were also more frequently encountered (n = 2,388; 80%) than females (n = 564; 19%) at MB (χ21 = 1127, p < 0.001) and immature males had the highest proportion of the total sightings (n = 1,170; 39%). At MB, three males and one female were estimated to be 1.5 m in width on their first recording and another thirteen individuals were estimated to be < 2 m, while 40 individuals were estimated to be 3 m (Appendix II.i Supplementary Figure 1A, B). At MP, the majority of the individuals were mature males (n = 240; 49%) and females (n = 127; 26%), with overall more males (n = 268; 54%) than females (n = 212; 43%; χ21 = 6.53, p = 0.011). However, females were more frequently encountered (n = 1,759; 60%) than males (n = 1,142; 39%; χ21 = 131.23, p < 0.001). Immature males were rarely encountered at MP, comprising just 2% of sightings (n = 45), a nearly 20-fold difference to MB. Five manta rays were estimated to be < 2 m in width on their first recording with the smallest being 1.75 m, while 130 were estimated to be 3 m (Appendix II.i Supplementary Figure 1C, D). The maturity status could not be determined (unknown) for 66 and 85 females (on 16 and 17% of total sightings) at MB and MP, respectively. 55 Figure 3.9 Population structure of manta rays in the Nusa Penida MPA, with (A) the number of individuals and (B) the number of sightings for Nusa Penida (All) and the two sites Manta Bay (MB) and Manta Point (MP). Based on sightings from January 2012 to April 2018. Percentages were calculated from the total number of individuals (A) or sightings (B) for each site. 3.4.3 Injury Rates Notably, 87 (~ 14%) of identified manta rays had either cephalic fin, pectoral fin, and/or fishing line injuries (Appendix II.ii Supplementary Table 1A). A total of 48 manta rays (7.7%) were observed entangled in fishing line or with fishing line marks (Figure 3.4) and 49 (7.9%) had permanent injuries, either cephalic (n = 29; 4.6%) and/or pectoral fin (n = 21; 3.3%) truncations or disfigurements. Nine individuals had more than one type of injury. There were 11 pregnant females observed with injuries (three of the individuals CHAPTER 3. NUSA PENIDA 56 were sighted pregnant two times with a span of at least one year in between) from a total on 17 injured females during the study period. To test for potential reproductive impairment caused by injuries, the proportion of pregnant manta rays with injuries (0.11) was compared to the overall proportion of females with injuries (0.12; χ21 = 0.676, p > 0.5), and was found to not be significantly different. 3.4.4 Residency and Movement Of the 624 uniquely identified manta rays, the majority were encountered more than once (82%), with 181 individuals (29%) sighted more than 10 times, and 30 individuals (5%) sighted 31 – 99 times (Figure 3.10). The mean number of re-sightings per individual was 9.5 (± 0.46). The majority (n = 514; 82%) of these individuals were re-sighted across multiple years and 18% (n = 110) were sighted in only 1 year. The longest time between re-sightings of an individual was 13.8 years (including pre-2012 data). Overall, of the ten most re-sighted individuals, nine were male and one was a female. The most re-sighted individual was a male that was seen 99 times, but at MB only, between February 2012 and October 2017. This individual was classified as a juvenile at the first encounter, in February 2012, but matured over the monitoring period and was reclassified as an adult in July 2016. Similarly, the other nine most re-sighted individuals at MB were all male, with six out of ten classified as juveniles upon first encounter (Figure 3.11A). The most re-sighted female was seen 56 times (54 times at MP and 2 at MB) between June 2015 and April 2018. Her maturity status was unknown until March 2017, at which point she was classified as mature as she appeared to be pregnant. Similarly, the other nine most re-sighted individuals at MP were females, and all were seen pregnant at least once during the study period (Figure 3.11B). The average period between re-sightings across all individuals (i.e. lag) was 118 days (± 3), ranging from 1 day to 5.3 years. For 57 the most sighted male and female, greater than 90% of re-sightings had a lag period of less than 2 months, with the longest lag periods of these individuals being 234 and 239 days, respectively. We found that 277 (44%) individuals moved between the two study sites, and tested whether the two sites were fully mixed (a single population) using LIR analysis (Figure 3.12A). The two LIR curves did not converge during the study period, indicating that, although there is interchange between MB and MP, most individuals have a high affinity for one site over the other (i.e. communities were not mixed), at least when considering time lags of 1,500 days (~ 4.1 years) or less. This notion is supported by the Markov movement model, which indicated high site affinity for MB (64.7%) and MP (73.7%) (Table 3.3A). A higher percentage of individuals moved from MB to MP (35.3 ± 9.9%) than vice versa (16.4 ± 6.5%) (Table 3.3A). Further, a higher percentage of individuals move away from MP (9.9 ± 15.7%) to ‘outside’ areas than from MB (0.0 ± 1.6%), indicating that the MP community is more mobile. Figure 3.10 The proportion of manta rays identified in the Nusa Penida MPA (All), and study sites Manta Bay (MB) and Manta Point (MP) plotted against the number of times individuals were sighted between January 2012 and April 2018. Total number of individuals for All, MB, MP, are shown for each number of sightings. CHAPTER 3. NUSA PENIDA 58 Figure 3.11 The top ten most sighted manta rays at Manta Bay (MB) and Manta Point (MP) are all males (A) and females (B), respectively. Colored circles show sighting locations. Solid lines indicate the first sighting where the individual is deemed mature, whereas dashed lines indicate observed pregnancy (female). 59 Figure 3.12 Lagged Identification Rates (LIR ± SE) of manta rays within the two study areas, Manta Bay (MB) and Manta Point (MP), Nusa Penida MPA, Indonesia. The mean probability of re-identifying an individual manta ray, from the time of first identification until approximately 6 years later within the same or different area in the Nusa Penida MPA (A), MB and MP (B); and broken down by sex within the sites MB (C); and MP (D). The predicted LIRs for models of best fit (Table 1) are shown for each group. Records from January 2012 to April 2018. CHAPTER 3. NUSA PENIDA 60 Results of the ‘within/between’ analysis showed that the sites should be treated separately for detailed analysis. Based on the ΔQAIC (Table 3.2A), the best fitting model for both MB and MP was Model H, which described a pattern of emigration, re- immigration and mortality. Models C and D, where emigration and mortality are included, were also supported (to a lesser degree) for MB (based on ∆QAIC < 2). Although a similar number of sightings were recorded for both areas, re-sighting rates are initially (up to two-year time lag) much higher at MB than MP, indicating that manta rays tended to reside longer in MB than MP (Figure 3.12B). Model H scenarios provide information on the proportion of time individuals spend within (residence time in) and out (residence time out) of an area, as well as mortality. Individual manta rays stayed approximately twice as long (2.0 days ± 187.7) at MB than at MP (0.9 days ± 0.2). Further, individuals sighted at MP spent an average of 2.9 (± 0.6) days outside the site, being absent nearly twice as long as those initially sighted at MB (1.6 ± 3.1 days). However, estimations of residence time within and out of MB were variable, suggesting that sex and/or demographic-specific (i.e. males vs. females, immature individuals vs. adults) or high individual differences in residence patterns exist. Mortality (which includes permanent emigration) was negligible for both sites (~ 0). To further explore sex specific differences in LIRs, we ran separate analyses for each site and sex (Tables 3.2B, C). Based on the ∆QAIC (except for females at MP, which was based on ∆AIC), the best fitting models were Models E (closed population with emigration and re-immigration) and H (emigration, re-immigration and mortality) for males and Model G (emigration, re-immigration and mortality) and F (emigration and re-immigration) for females at MB and MP, respectively (Tables 3.2B, C). We further explored maturity influenced differences in LIR for males (immature females were not assigned) at MB (Table 3.4) using sighting records (immature = 487; mature = 707) of 61 Table 3.4 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of (A) immature and (B) mature male manta rays at Manta Bay (MB) in the Nusa Penida MPA, Indonesia. Model Model description ΔQAIC ΔQAIC A) Immature males B) Mature males A Closed (1/a1 = N) 1379.76 19.62 B Closed (a1 = N) 1379.76 19.62 C Emigration/mortality (a1=emigration rate; 1/a2=N) 0 5.88 D Emigration/mortality (a1=N; a2=mean residence) 0 5.88 E Closed: emigration + re-immigration (a1=emigration rate; a2/(a2+a3)=proportion of population in study area at any time) 1374.56 19.71 F Emigration + re-immigration (a1=N; a2=res time in; a3=res time out) 2 1.06 G Emigration + re-immigration + mortality 92.19 8.57 H Emigration + re-immigration + mortality (a1=N; a2=res time in; a3=res time out; a4=mort) 4 0 N = population individuals (immature = 59; mature = 127) whose maturity status remained the same throughout the duration of the study (January 2012 to April 2018). The best fitting models (based on ΔQAIC) were Models C and D (emigration/mortality) for immature males and Model H for mature males. Site and sex-specific LIR indicated that, on average, female manta rays that use MB, and to a lesser extent MP, are two times more likely to be re-sighted at the site of first sighting during the first two years after their initial identification than males (Figure 3.12C, D and Tables 3.2B, C). However, immature males were twice as likely to be re-sighted at MB than mature males (Figure 3.13). The Model H scenario indicates the residence time in for mature males was 32.3 days (± 47.0), while Model D estimated the mean residence time for immature males to be ~ 23-fold longer (745.3 days ± 169.9). Markov movement models (Table 3.3) indicated differential movement patterns with a higher percentage of males, moving from MP to MB (24.5 ± 2.8% SE) than CHAPTER 3. NUSA PENIDA 62 females (8.1 ± 2.2% SE). Further, a modestly higher percentage of females are staying within MP than males (85.8% vs. 75.5%). Notably, the model maximum number of iterations was exceeded, suggesting that the estimated SEs are inaccurate and larger than expected (H. Whitehead, Dalhousie University, pers. comm.). Nevertheless, taken together these results show a lower probability of movement between the study sites (i.e. higher residence) for female manta rays than mature males; although, immature males display high residency to MB. Figure 3.13 Lagged Identification Rate (LIR ± 1 SE) of immature and mature male manta rays within Manta Bay, Nusa Penida MPA, Indonesia. The predicted LIRs for models of best fit (Table 3.4) are shown for each group. Only sighting records of individuals whose maturity status remained the same throughout the duration of the study (January 2012 to April 2018) are used. Immature male sightings = 487, individuals = 59. Mature male sightings = 707, individuals = 127. 3.4.5 Behavior A range of manta ray behaviors (Figure 3.5) were observed (n = 3,692) at both study sites, including foraging, cleaning, courtship, cruising and a mix of cleaning and courtship, and cleaning and foraging (Figure 3.14). However, the frequency of observed behaviors differed between the two sites, and a significant association between behavior and site was present (χ23 = 2303.7, p < 0.001). From the 1,211 sightings with recorded behavior at MB, manta rays were foraging in 79% of sightings (surface ram feeding and 63 near surface barrel-rolling), with cleaning and courtship behaviors observed in 7% and < 1.5% of sightings, respectively. In contrast, at MP (2,481 sightings with recorded behavior), manta rays were foraging in < 4% of sightings. The principal observed behaviors at MP were cleaning (59%), courtship (7%), and courtship combined with cleaning (8%) during single encounter sessions. Courtship behavior was observed year-round at MP but peaked in May (Figure 3.14B). More cruising behavior was observed at MP (22%) than MB (13%). Figure 3.14 (A) The percentage of observed manta ray behaviors determined from sighting logs and photographs from Manta Bay (MB, n = 1,211) and Manta Point (MP, n = 2,481) from January 2012 to April 2018. Foraging and cleaning, as well as courtship and cleaning behaviors by an individual manta ray were on occasion both witnessed during the same sampling event and identified as such. (B) The average monthly sightings per unit effort (day) of manta rays engaging in courtship behavior. Data recorded between 2013 and 2017 (excluding 2015 due to lower year-round effort) in MP and presented as the daily means (± 1 SE). CHAPTER 3. NUSA PENIDA 64 3.4.6 Boats The average annual number of boats (7.2) present at the two manta ray sites increased by 60% (Figure 3.15) from 2012 to 2017. This increase was greater at MP than at MB, with the average annual number of boats increasing from 5.3 in 2012 to 13.4 in 2017 at MP (153% increase) compared with virtually no increase at MB (5%, annual average number of boats = 6.2, range = 5.3 – 7.3) (Figure 3.15A). The monthly average number boats at MP when considering all years was higher from April – October (10.9 – 14.0), than the other months (< 8.3), with peaks in May and September (Figure 3.15B). In contrast, the average number of boats at MB was relatively constant throughout the year (5.4 – 7.9), peaking in August. Figure 3.15 The annual (A) and monthly (B) daily mean (± 1 SE) number of boats present at manta ray sites in the Nusa Penida MPA overall, and separately in Manta Bay (MB) and Manta Point (MP), between 2012 and 2017. 65 3.5 DISCUSSION We found the Nusa Penida MPA was an important habitat for a substantial number of manta rays (624 identifications in approximately 6 years), with manta rays present year- round and showing high site affinity to two sites, ~ 12 km apart. Individual re-sighting rates reported in the Nusa Penida MPA are higher than those reported for any other M. alfredi population in the world (Table 3.5), with 82% of the individuals re-sighted more than once. The manta rays display high site affinity to the west coast of Penida Island and are rarely reported from other parts of the MPA. Our analyses highlighted that the two main study sites, Manta Bay (MB) and Manta Point (MP), are used for different purposes by different life stages of manta rays. Manta Bay is a foraging ground, used primarily by immature individuals, whereas MP is an important site for adult social and reproductive activity. Female and male manta rays also used the sites differently, with more males frequenting MB than MP. As a whole, the sex bias found in the Nusa Penida MPA (males: females = 1.4:1) is the largest reported bias toward males worldwide (Table 3.5). On the other hand, the proportion of females confirmed as sexually mature (53%) is greater than anywhere else in the world (Table 3.5). This clear segregation in habitat use by different manta ray demographics at almost adjacent sites along a continuous coastline is noteworthy for marine spatial planners and conservationists globally. 3.5.1 Manta Bay Unlike other manta ray aggregation sites studied to date, MB was used mostly by males with a few resident females. The largest percentage of sightings at MB (38%) were of 59 immature males (one sighted 99 times during the study period). This is particularly noteworthy as few studies report such frequent sightings of juvenile manta rays (Stevens, 2016). Thus, the steady accumulation of new individuals being identified at this site may CHAPTER 3. NUSA PENIDA 66 Table 3.5 A comparison of manta ray M. alfredi population studies and demographics from the available literature. Location Location Size Study Years Estimate Method Population estimate Male: Female Maturity (% mature) Re-sighting (%, avg, max) Study Inhambane coast, Mozambique ~12 km 2003 – 2007 (4 yrs) Min Jolly–Seber Super population 449 149 – 454 (annual range) 802 (± 106 SE) 22% vs. 78% 1:3.55 Males: 90% Females: 50% 41% 2.2 (avg) 12 (max) Marshall et al., 2011 East Coast (Heron Island to Solitary Islands) Lady Eliot Island, Australia 845 km 0.45 km2 1982 – 2012 2007 – 2012 (4 yrs) Min Min Pollock’s Robust Design 716 636 456 42% vs. 53% 1:1.2 42% vs. 53% 1:1.3 Males: 74% Females: 18% 63% 66% 20 (max) Couturier et al., 2014 Maldives ~15,000 km2 2000 – 2009 (8.5 yrs) Min Jolly-Seber 1,835 9,677 (extrapolated) 36% vs. 64% 1:1.8 > 99% adults 27% Kitchen-Wheeler et al., 2010 2005 – 2015 (11 yrs) Min 4,000 49% vs. 50% 1:1 Males: 70.5% Females: 42% ≥ 159 (max) Stevens, 2016 West coast Maui, Hawaii 0.03 km2 2005 – 2009 (5 yrs) Min Cormack-Jolly-Seber 290 77 – 230 (annual range) 53% vs. 44% 1.2:1 Males: 72% Females: 44% 73% 41 (max) Deakos et al., 2011 Nusa Penida, Raja Ampat, Komodo, Indonesia 202.8 km2 40,000 km2 1,817 km2 2006 – 2014 (8 yrs) Min Min Min 417 100 303 N/A N/A 65%, 6.9 (avg) 30%, 3.1 (avg) 33%, 2.7 (avg) Germanov & Marshall, 2014 Nusa Penida, Indonesia 202.8 km2 2012 – 2018 (6.25 yrs) Min 624 55% vs. 39% 1.4:1 Males: 80% Females: 53% 82% 9.5 (avg) 99 (max) Current study yrs = years; Min = number of individuals identified; Re-sighting % = percent of individuals re-sighted; avg = average number of re-sightings; max = maximum number of re-sightings on record 67 reflect recruitment of juveniles to an important foraging ground, or potentially a nursery, following birth. The smallest manta rays recorded in the duration of the study were approximately 1.5 m in disk width, which is within the size range of newborn M. alfredi (1.3 – 1.6 m; Stewart et al., 2018a). Future research looking into the size of manta rays using standardized measurement techniques (Costa et al., 2006; Couturier et al., 2014; Deakos, 2010;) would further our understanding of how different age classes use these habitats. Continued recruitment of new, immature, individuals to MB suggests that females give birth nearby, which is a topic worthy of future research. The location of birthing grounds (i.e. pupping grounds) and (or existence of) nurseries for manta rays and other mobulids are major outstanding questions in mobulid ecology (Stewart et al., 2018a). The first nursery described for M. birostris in Flower Garden Banks National Marine Sanctuary in the northwestern Gulf of Mexico (Stewart et al., 2018b), followed the most recent re-description of the term ‘nursery’ for elasmobranchs, proposed by Heupel et al. (2007), and hinges on: (1) more immature individuals are found in this area relative to others nearby; (2) immature individuals reside for extended periods in this area relative to others nearby; and (3) immature individuals show long-term habitat use of this area relative to others. Thus, nurseries are expected to be areas that provide beneficial opportunities to the young, such as increased food availability (Heupel et al., 2007; McCauley et al., 2014), protection from predation, or areas for thermoregulation via basking behavior (Stevens, 2016; Stewart et al., 2018b). The high re-sighting rates of immature manta rays, the observation of foraging as the dominant behavior, and meeting the criteria proposed by Heupel et al. (2007) and others for nurseries (McCauley et al., 2014; Stevens, 2016; Stewart et al., 2018b), provide strong evidence that MB forms part of a nursery for the local M. alfredi population. The proposed nursery criteria are aligned with our study in that (1) more immature individuals CHAPTER 3. NUSA PENIDA 68 are found at MB than the nearby MP; (2) much longer residency of immature individuals is observed at MB than at MP; and (3) manta rays identified as immature on first sighting show long-term habitat use of MB relative to MP. In oligotrophic environments, immature and smaller sized manta rays (more often male because females reach a larger maximum size) possibly prefer to aggregate in shallow coastal areas that offer protection from predators (McCauley et al., 2014; Stevens, 2016). In the Nusa Penida MPA, smaller individuals might use MB as a prominent foraging ground because of reliable food availability. The semi-enclosed nature of the bay, located at the base of the ITF through the Badung Strait and neighboring channel between Ceningan and Penida Islands, and proximity to deep water to the south, could entrain plankton and thereby provide consistent shallow water foraging opportunities for these immature rays. 3.5.2 Manta Point The manta rays sighted at MP were typically larger than those at MB, with mature females making up 40% of sightings. Thus, the recruitment of new individuals to MP is probably due mainly to immigration, not births. Notably, we found the highest percentage of mature females at this site ever reported (63%) globally (Table 3.5). The most frequent behavior observed at MP was cleaning (67%). The removal of parasites and other external fouling and the promotion of wound healing is likely an important service provided by cleaner fish (Marshall, 2008; Stevens, 2016). Cleaning activity appears to be a daily ritual, with individuals sometimes spending hours at cleaning stations during the day (Dewar et al., 2008; Marshall, 2008; O’Shea et al., 2010). Visits to reefs serving as cleaning stations may also provide an opportunity for social interactions in elasmobranchs, including mobulids (Murie et al., 2016; Oliver et al., 69 2011; O’Shea et al., 2010; Stevens, 2016). Both cleaning and courtship behaviors were observed during 8% of sightings at MP. Courtship behavior is clearly initiated at predictable aggregation sites for mature individuals, with cleaning stations potentially acting as lekking sites (Stevens, 2016). Individuals might visit these aggregation sites during the breeding season in search of mates, rather than for cleaning (Deakos et al., 2011; Stevens, 2016). These suggestions are consistent with the peak in reported sightings at MP during the main reproductive time in May. The seasonal pattern of manta ray courtship behavior varies throughout the world (Couturier et al., 2014; Deakos et al., 2011; Marshall & Bennett, 2010; Stevens, 2016), with no single driver for initiating courtship identified to date. Reproductive behavior is likely to be linked to seasonal productivity in the water column, as fecundity has been linked to food availability (Stevens, 2016). 3.5.3 Site Affinity and Productivity Manta rays are very mobile species and range widely in both archipelago environments (Conservation International Indonesia, 2016; Germanov & Marshall, 2014) and along continental coastlines (Couturier et al., 2014; Jaine et al., 2014). However, recent studies (Couturier et al., 2018; Setyawan et al., 2018) show that they do not use the extent of their home range uniformly, and that individuals concentrate the majority of their activities (foraging, cleaning, and courtship) within specific critical habitats. While our study focused on sightings in two specific areas, it is evident that habitat use within the Nusa Penida MPA is not uniform. The year-round sightings of manta rays and the many individuals documented using the Nusa Penida MPA are likely to be linked to the sustained productivity in the region (Ayers et al., 2014; Nyegaard, 2018; Surinati, 2009; Tito et al., 2017). Food CHAPTER 3. NUSA PENIDA 70 availability and sustained foraging opportunities are thought to be responsible for dictating the size of manta ray populations aggregating in a particular area (Deakos et al., 2011) and the frequency of their pregnancies (Stevens, 2016). In most of the manta ray populations studied worldwide, seasonal increases in manta ray abundance coincide with increased oceanic productivity (Dewar et al., 2008; Jaine et al., 2012) and in prey density (Armstrong et al., 2016; Rohner et al., 2017). For example, the largest estimated manta ray population globally is found in the Maldives (Kitchen-Wheeler et al., 2011; Stevens, 2016), where alternating monsoons result in year-round productivity (Anderson et al., 2011). The fine-scale oceanography of the Nusa Penida area is poorly known at present, though a seasonal decline in water temperature, indicating regional upwelling and the potential for nutrient mixing and increased nutrient availability to support primary productivity, were recorded over distances of about 30 km from the northern coast to the south-western coast during research on sunfish populations in the region (Nyegaard, 2018; Tito & Susilo, 2017). At the end of the north-west monsoon season, declines in water temperature were recorded starting April for the south-west coast of Penida, similar to southern Komodo NP (Dewar et al., 2008). Manta ray visitations to the south Komodo increased in April, similar to what is observed at MP, suggesting that the broad season patterns that influence the Indian Ocean and the Lesser Sunda region might be a driver in manta ray abundance in Nusa Penida as they are for Komodo NP. Studies that evaluate both the presence/movements of filter-feeders and the concurrent biological and physical oceanographic characteristics (Jaine et al., 2012; Rohner et al., 2013) would enhance the understanding of manta ray ecology and the drivers of differential site use, increased seasonal abundance and reproductive behavior in this complex area. 71 3.5.4 Movement Patterns Although the manta rays of Nusa Penida display some site preference, emigration and interchange between MB and MP does occur. Almost half of the individual manta rays recorded (n = 277; 44%) visited both locations during the study period. However, the interchange between the two sites was not symmetric and differed between the sexes. Higher transition probabilities for movement were observed from MB to MP (~ 35%) than for the reverse movement (~ 16%), raising the possibility that a longer-term study may demonstrate that the two sites, taken together, constitute a single population (Figure 3.16). The higher percentage of movement from MB to MP could be individuals leaving their nurseries as they grow larger, to forage offshore and spend more time interacting with other mature individuals (McCauley et al., 2014; Stevens, 2016). Further, the sex-based differences in habitat use for manta rays might be mainly linked to reproductive behavior (Deakos et al., 2011; Stevens, 2016). Similar to other pelagic species of marine megafauna including sharks, sea turtles and cetaceans (Engelhaupt et al., 2009; Hueter et al., 2005; Lee et al., 2007), female manta rays, as the sex with the greater parental investment (i.e. the ‘limiting sex’), gain a greater choice of mates by residing in a popular aggregation area, assuming that ample foraging opportunities are available nearby. Males, in contrast, benefit from moving between aggregation areas in search of mates. Consistent with this hypothesis, female manta rays from the Nusa Penida MPA are likely to reside in either of the sites for a longer time than males. Moreover, the Nusa Penida MPA manta rays occasionally move to areas ‘outside’ MB and MP. Previous research (Conservation International Indonesia, 2016; Germanov & Marshall, 2014) shows that manta rays move long distances from the Nusa Penida MPA to locations such as the Gili Islands (80 km straight-line distance north-east), Sumbawa, and the Komodo National Park (up to 450 km straight-line distance east). CHAPTER 3. NUSA PENIDA 72 Figure 3.16 Movement patterns of male and female manta rays in the Nusa Penida MPA. 3.5.5 Implications for Management The Nusa Penida MPA serves as year-round critical habitat for a substantial number of manta rays making it a high priority area for conservation and management. MB is used year-round for foraging, particularly by juvenile manta rays, while MP is an important area for cleaning and reproductive behaviors that peak in May each year. This has implications for MPA management, zoning and the types of activities to be allowed in these areas during distinct times of year. Management actions should address the following threats to the local manta ray populations: (1) disruption of manta ray behavior through habitat crowding and human disturbance from excessive tourism (Barr et al., 2017; Trave et al., 2017; Venables et al., 2016) and (2) entanglement in fishing gear and injury from continuing fisheries in the area (present study, Aquatic Alliance, pers. comm.). 73 3.5.6 Manta Ray-Focused Tourism Impacts While tourism to the area is increasing, there are currently no regulations in place for manta ray interactions (Barr et al., 2017). A 2016 survey (Barr et al., 2017) directed at divers and snorkelers participating in manta ray watching activities, within the Nusa Penida MPA, reported participant conduct that is considered disruptive to the manta rays (e.g., closely following and touching manta rays was reported by 10% and 3.5% of participants, respectively). Disturbance stimuli to animals, such as loud boat engine noise, fast approaches by boats, in water chasing, touching, and crowding behavior by tourists (Anderson et al., 2010; Bejder et al., 2006a, b; Barr et al., 2017; Higham et al., 2016; Pierce et al., 2016; Venables et al., 2016), is argued to be analogous to a predation risk (reviewed by Frid et al., 2002). Thus, tourism pressure at MB might disrupt foraging behaviors, reducing growth rates of immature manta rays and the fitness of mature individuals (Stewart et al., 2018a). Further, higher boat numbers at MP in May coincide with the seasonal increase in sightings at MP and in reproductive behavior. With the Nusa Penida MPA serving as an important reproductive ground, disruption of these important social behaviors is a concern. Based on the criteria to evaluate marine wildlife tourism practices outlined by Trave et al. (2017), we recommend that: (1) science based carrying capacity calculations of tourism operations be carried out to estimate the acceptable number of tour boats and diver interactions for the area (Ríos-Jara et al., 2013; Zelenka et al., 2014) potentially limiting the number of boats/divers/swimmers allowed at one time (Division of Boating and Ocean Recreation, 2016; Kasmidi & Gunadharma, 2017); (2) codes of conduct for diving and snorkeling with manta rays become mandatory (see Garrud, 2016; Venables et al., 2016), akin to regulations for whale shark interactions in Ningaloo Reef, Australia (Catlin et al., 2010; Mau, 2008); (3) a licensing system for tour operators with penalties for breaches be implemented (Mau, 2008; Division of Boating CHAPTER 3. NUSA PENIDA 74 and Ocean Recreation, 2016); and (4) area-time closures be considered as a management option (Setyawan et al., 2018; Tyne et al., 2014) to protect the manta rays from disturbance during the peak time of mating, especially at MP. 3.5.7 Artisanal Fisheries Small-scale fishing often takes place in MB and nearby coastlines (E. Germanov, pers. obs., Figure 3.17). Over the course of the study, ~ 14% (n = 87) of manta rays were observed either trailing hooks and lines or with cephalic and pectoral fin injuries and/or amputations, which occur when fishing lines or nets cut through the skin and cartilage skeleton. The true scale of the issue is under-represented as only animals that have survived the entanglement event are counted. Currently, the long-term impacts of these effects on individuals that survive entanglement are not known, nor whether foraging and swimming efficiency is impaired and if there are any impacts on fecundity. However, observations of pregnant individuals within the Nusa Penida MPA, with single cephalic fin amputations, suggest that manta rays retain their reproductive fitness even with these sub-lethal injuries. Further, the proportion of pregnant manta rays from the Nusa Penida MPA with injuries was not significantly lower than the overall proportion of females with injuries, suggesting that injured manta rays are not substantially impaired reproductively. Nevertheless, it is unclear whether these injuries have an initial impact on the success of their pregnancies, as many related species (e.g., batoids) will abort their fetuses if their individual survival is threatened (entanglement, landing, predation; reviewed by Adams et al., 2018). Further, biomechanical modeling of manta ray pectoral fin movements indicate that the major thrust force comes from the distal portion of the pectoral fins (Liu et al., 2015), thus pectoral fin truncations could significantly impact manta ray swimming efficiency, energy consumption and ability to evade predation. While all fishing activities 75 are officially prohibited in both MB and MP, increased enforcement of this regulation is necessary. As a precautionary measure, management could prohibit all fishing activities along the west coast of Nusa Penida, which would require strong compliance if it was to be effective in contributing to the recovery of manta rays in the Nusa Penida MPA. Figure 3.17 A manta ray forages at Manta Bay while people fish with pole and line from Nusa Penida cliffs. CHAPTER 3. NUSA PENIDA 76 3.5.8 Limitations and Future Directions The data collection methods of this study had several limitations. There advantages to using photo identification and a sightings-based approach as used in this study, including that it is non-invasive, cost effective compared to electronic tracking (i.e. satellite tags and prescriptions; acoustic tags and receivers), and unchanging natural markings serve as life-long ‘tags’ that cannot be shed and lost (discussed in Pierce et al., 2018). Once coupled to citizen science, this approach allowed us to employ the site use and movement patterns of hundreds of individuals cost effectively. However, there are drawbacks to this approach that have posed several limitations on our study. Direct sightings-based approaches (sightings recorded by an observer rather than an unmanned camera) are inherently biased to when observers are present in the water. This study was supported by local dive operators who provided trained observers space on dive boats free of charge. Thus, surveys were only conducted in sites that are popular for tourism, potentially ignoring other aggregation sites within the area. Exploration of areas further afield to the southernmost and less visited parts of the Nusa Penida coastline might uncover other aggregation sites. The potential for other aggregation sites to existed is supported by tracking data (Conservation International Indonesia, 2016). Further, surveys were only performed when environmental conditions (i.e. wind, swell and currents) were conducive to observer safety, thus potentially skewing observations to only a limited range of environmental conditions. During surveys photographers would need to be successful in observing the individuals present and then adequately photographing them for identification purposes. Thus, not all individuals present were recorded at all times. Recorded sightings, especially those submitted by citizen scientists, at times lacked data (e.g. time of encounter) that would enable us to study environmental drivers on manta ray abundance at these sites (i.e. effects of tides, time of day, temperature, water clarity, wind 77 speed and direction, current speed and direction, etc.; e.g. Barr & Abelson, 2019; Jaine et al., 2012; Rohner et al., 2013). Further, even with trained observers, size estimates varied in their accuracy, limiting the data available to those collected by few highly skilled observers. Further, the study design, which relied on opportunistic data, limited the usefulness of the data for capture-recapture analyses for abundance and survival estimates, which require rigorous and systematic survey design. Future research on manta rays using the Nusa Penida MPA should focus on answer questions regarding abundance and survival (e.g. Couturier et al., 2014; Deakos et al., 2011; Kitchen- Wheeler et al., 2011, Marshall et al., 2009), environmental drivers of habitat use and abundance (e.g. Barr & Abelson, 2019; Jaine et al., 2012; Rohner et al., 2013), feeding ecology (e.g. Armstrong et al., 2016; Couturier et al., 2013; Peel et al., 2019b) and reproductive biology (e.g. Deakos, 2012; Marshall & Bennett, 2010; Stevens, 2016). Given the regular sightings of immature individuals at MB, investigation into the possibility of a discrete pupping ground in the area would be important as currently little knowledge exists on this aspect of manta ray reproduction (Stewart et al., 2018a). Accurate size estimates to track growth in conjunction with maturity indicators could be collected employing the current gold standard, stereo-photogrammetry (Santana-Garcon et al., 2014; Whitehead, 2014). Further, examining the potential impacts of tourism on feeding, cleaning and reproductive behavior are needed as tourism use of manta ray aggregation sites is increasing (e.g. Garrud, 2016; Venables et al., 2016; Whitehead, 2014). Lastly, barriers to effective enforcement and compliance to regulations on MPA zone use need to be identified and addressed through social studies (e.g. Booth, 2016, Jaiteh et al., 2016; 2017). CHAPTER 3. NUSA PENIDA 78 3.6 CONCLUSION Understanding localized habitat use, identifying nurseries, as well as key times of year for reproduction, provides enhanced information for developing effective management plans for manta rays in this region related to specific manta ray behaviors. Context-dependent and adaptive management solutions are important for manta rays as well as for people, as manta ray watching tourism supports many livelihoods through activities that rely on healthy manta ray populations (i.e. snorkeling and SCUBA diving). With their 2014 protection in Indonesia, manta rays have entered the spotlight for conservation initiatives and serve as ideal flagship species to refocus future research on the overarching challenges and opportunities facing ocean health in the Coral Triangle (see also Chapter 2), the world’s premier marine biodiversity hotspot. 79 CHAPTER 4. NO PLACE LIKE HOME? PHOTO-IDENTIFICATION REVEALS FINE-SCALE HABITAT USE OF REEF MANTAS IN KOMODO NATIONAL PARK, INDONESIA Image adapted from: Giulia De Amicis 80 81 4.1 ABSTRACT The reef manta rays (Mobula alfredi) of Komodo National Park are a threatened species and an an iconic tourist attraction for divers and snorkelers. Yet population-level work on the species has been limited within this region. Previous studies demonstrated high site fidelity to sites within the park by individual manta rays, and long-range movements of up to 450 km outside the park. The present study uses a long-term (July 2004 – March 2018) library of photo-identification data collected by researchers and citizen scientists and submitted to the global manta ray sightings database at www.mantamatcher.org. A total of 1,104 individual manta rays were identified within the park over this period, with most sightings coming from four sites that are frequented by tour operators: Cauldron – CL, Karang Makassar – KM, Mawan – MW, and Manta Alley – MA. Residency and movements of manta rays at the three sites with the highest survey effort, KM, MW and MA, were investigated with maximum likelihood analyses and Markov movement models. In general, individuals displayed higher affinity to one site over others. The highest re-sighting probabilities came from the comparatively remote MA, located south of Komodo Island. Although the two sites in the central park, KM and MW (~ 5 km apart), had a high level of connectivity between them, a degree of affinity to each site was still present. There was exchange between MA and these central sites (~ 35.5 km apart), likely on a seasonal basis, but more manta rays traveled from the south to central area than vice versa. In contrast to other studied populations, females appeared to be the more mobile of the sexes. Although there was a significant association between behavior and site (χ26 = 202.077, p < 0.001), similar demographic groups used KM, MW and MA for a mixture of foraging, cleaning, cruising and courtship activities. The northern site, CL, had both a higher proportion of immature manta rays present, and a higher prevalence of foraging activity than found at the other sites. Fishing gear related CHAPTER 4. KOMODO NATIONAL PARK 82 injuries were noted on 56 individuals (~ 5%), and predatory injuries were present on 32 individuals (~ 3%). Tourism within the park is increasing, with the number of dive boats at KM and MW increasing by a combined 34% from 2014 to 2017, and a daily mean of up to 12.3 ± 3.2 boats present in the busiest month (August). Increasing pressure from tourism at these important habitats supports the necessity for tourism limitations to minimize disturbance to manta rays. 4.2 INTRODUCTION Increasingly, it is being recognized that highly mobile marine megafauna species often return to, and spend a disproportionate amount of time in important sites within their broad range (e.g. Baird et al., 2008; Bowen & Karl, 2007; Dewar et al., 2008; Graham et al., 2016; Rooker et al., 2008). This realization is important for spatial planning and conservation efforts (Chapman et al., 2015; Heuter et al., 2005; Speed et al., 2010), as improved management of these habitats can have a disproportionate positive benefit for the species over a wide range. Such benefits can be maximized by considering the species’ demography and identifying key areas (e.g. juvenile habitats or adult reproductive areas) as conservation priorities to maximize population recovery for threatened species (Heupel et al., 2007; Norse, 2005; Martins et al., 2018). However, conservation considerations of movement corridors are also important (Hooker et al., 2011). Manta rays (Mobula alfredi – Krefft, 1868; M. birostris – Walbaum, 1792) are long-lived (Stewart et al., 2018) with low fecundity (Marshall & Bennet, 2010; Deakos et al., 2012), resulting in extremely low population growth rates (Dulvy et al., 2014; Rambahiniarison et al., 2018). The species are prized by fishers for the demand of their brachial gill plates in the Asian medicinal markets (O’Malley et al., 2017). Both manta 83 species were listed as Vulnerable to extinction by the IUCN Red List of Endangered Species (Marshall et al., 2018a, b) due to declining population trends (Dulvy et al., 2014; Ward-Paige et al., 2013), and subsequently listed on Appendix II of the Convention on the International Trade of Endangered Species (CITES) and Appendices I and II of the Convention on the Conservation of Migratory Species. Both manta species were given legal protection in Indonesia throughout the country’s entire exclusive economic zone (an area of over 6 million km2) in 2014 (Ministry of Marine Affairs and Fisheries, 2014). National protection for manta rays followed from the international listings mentioned above, but the Indonesian government was also motivated to protect the economic contribution of manta ray tourism (O’Malley et al., 2013). However, the species are still subject to some bycatch and illegal directed fisheries in the country (Booth, 2016; Croll et al., 2016), particularly in movement corridors (Germanov & Marshall, 2014). Designated conservation zones, such as Nusa Penida Marine Protected Area (MPA) and Komodo National Park (NP), offer an additional layer of protection due to general fishing restrictions and increased levels of compliance monitoring. Komodo NP was formally established in 1980, initially as a conservation zone for the endemic Komodo dragon (Varanus komodoensis) (Erdmann, 2004a). Reef manta rays (M. alfredi, hence referred to as manta rays) were also a significant beneficiary of the park’s creation. The first study on M. alfredi within Komodo NP (Dewar et al., 2008; this study predated the taxonomic split from M. birostris in Marshall et al., 2009), investigated their movements within the area (October 2000 – June 2003). Passive acoustic telemetry employed between September 2000 and May 2003 identified regular use and connectivity for manta rays within certain areas of the park. These were welcome findings at a time when manta ray fisheries were active through the Lesser Sunda region CHAPTER 4. KOMODO NATIONAL PARK 84 (Dewar et al., 2008; Germanov & Marshall, 2014; Heinrichs et al., 2011; Lewis et al., 2015; White et al., 2006), with the park affording resident manta rays an increased level of protection from fishers. Germanov and Marshall (2014), using photo-identification techniques, subsequently showed that some individuals do leave the park and moved through heavily fished waters to be re-sighted up to 450 km away at the Nusa Penida Marine Protected Area (MPA). This study also provided information on the minimum numbers of individuals using the park, which at the time was ~ 300 individuals. Population doubling times have not been specifically calculated for M. alfredi; however; estimates for M. birostris indicate long population doubling times of ~ 15 – 87 years (Rambahiniarison et al., 2018). These estimates highlight the importance of Komodo NP and other conservation zones as refuges for the species while Indonesia transitions from manta fishing to non-consumptive use through tourism. Identifying key habitats and presence/absence trends helps to improve tourism experiences by increasing the chances of encounters with manta rays (Barr & Abelson, 2019; Dewar et al., 2008). However, it is important to consider that times of mass gatherings may coincide with important feeding or reproductive events where behavior disturbance should be minimized (Chapter 3, Armstrong et al., 2016; Stevens 2016; Weeks et al., 2015). The rapid increase in marine tourism at other manta sites in the country (Chapter 3) and a major (~ 14-fold) increase in tourism in the Komodo NP (Komodo National Park, pers. comm.) emphasizes the need for effective tourism management to avoid potential negative impacts (i.e. reviewed by Trave et al., 2017; Tyne et al., 2014; Stewart et al., 2018). As of September 2019, the Komodo NP is trailing for the first time limits on the number of boats and divers/snorkelers allowed in the water simultaneously at the popular manta ray aggregation site Karang Makassar. While regulations on appropriate marine conduct within the park are presently being socialized, enforcement is minimal. 85 In the current study, we expand on earlier research in Komodo NP (Dewar et al., 2008; Germanov & Marshall, 2014) through the use of a long-term photo-identification library spanning 14 years to investigate habitat use of a large number of manta rays. The goals of this study are to assess: (1) residency and movement, (2) population demography, (3) behavior and (4) potential threats of manta rays inhabiting the Komodo NP. This research utilizes researcher and citizen science contributions to the online global database for manta ray sightings, ‘Manta Matcher’ – www.mantamatcher.org (Marshall & Holmberg, 2018). The currently available information from Manta Matcher, including anthropogenic and predatory injuries, as well as data on tourism use of manta ray aggregation areas, allows for a better understanding of manta rays ecology within Komodo NP, which is pertinent to effective manta ray management in the region. 4.3 METHODS 4.3.1 Study Area The Komodo National Park (Komodo NP; Figure 4.1 created using QGis v 2.18, 2016), covers 1,817 km2 of land and sea within the terrestrial Wallacea region. This area lies east of Bali and Borneo and west of Papua and contains high levels of endemic terrestrial species, and is on the southern side of the Flores Sea “Marine Wallace line” (Barber et al., 2000). This is a complex oceanographic region (marine area = 1,214 km2) characterized by three main large islands (Komodo, Rinca and Padar) and additional smaller islands within the park boundary, totaling 603 km2 of terrestrial habitat (Erdmann, 2004b). The three large islands create the large shallow (≤ 100 m) Lintah Strait through the center of the park. The park boundaries are flanked by Sape Strait, another large, slightly deeper (~ 100 – 200 m) strait to the west of Komodo Island, a CHAPTER 4. KOMODO NATIONAL PARK 86 Figure 4.1 Location of study sites within the north (Cauldron – CL), central (Karang Makassar – KM, Mawan – MA), and south (Manta Alley – MA) regions of the Komodo National Park (NP) zone, in the Lesser Sunda region of the Indonesian archipelago. Other popular SCUBA diving locations with infrequent manta ray sightings are marked as “other” sites. These locations include (from north to south): ‘Castle’ and ‘Crystal Rock’, ‘Lighthouse’, ‘Tatawa Besar’, ‘Tatawa Kecil’, ‘Batu Bolong’, ‘Siaba Besar’, and ‘German Flag’. Bathymetry information was obtained from: GEBCO_2014 Grid, version 20150318; www.gebco.net. narrow passage to the east of Rinca Island (Molo Strait), and deep water (> 800m) basins to the north and south (Figure 4.1). Water exchange through the park is driven by strong tidal flow currents (up to ~ 15 km/h in the Lintah Strait), with water from the Indian 87 Ocean flowing north through the straits on the incoming tide and water from the Pacific Ocean flowing south on the outgoing tide (Erdmann, 2004c) via the Indonesian Throughflow (ITF). Upwelling to the south of the Lesser Sunda region is strongest from June through October (coinciding with the south-east monsoon, active May - October) and serves to enrich the waters in southern Komodo NP during this time of year (Ningsih et al., 2013). Nutrient-rich waters flow throughout the straits, particularly with spring tides. This trend is reversed during the inverse north-west monsoon period (November – April), with nutrient-rich water flowing through Komodo NP from the Flores Sea located to the north. We focused our analyses on four main sites within the Komodo NP (from north to south: Cauldron – CL, Karang Makassar – KM, Mawan – MW, Manta Alley – MA) based on the consistent manta ray sightings at these locations (Figure 4.1; Table 4.1). Three of these sites, KM, MW and MA, were included in a previous acoustic telemetry study of manta ray movements within the park (Dewar et al., 2008). However, CL, located in the north area of the park, essentially a shallow channel (< 25 m) between two islands with complex bathymetric structure and strongly affected by tidal currents, has yet to be reported on. In the central area of the park, located close to a sandy island with fringing reef, KM refers to a gently sloping shallow (< 18 m) rubble field that runs ~1.5 km north to south, with patchy reef and coral heads where manta rays are commonly observed cleaning (i.e. cleaning stations; see Chapter 3, Figure 3.5 for a more detailed description). Mawan is an island, ~ 5 km to the east of KM, with a sandy slope from 3 to 20 m on its south-eastern tip with a shallow ‘cleaning station’ at 5 m. Shallow rubble reefs also serve as cleaning stations to the north and to the south of the sandy slope of MW, covering an overall distance of ~ 0.7 km. Both these central sites (i.e. KM and MW) are also exposed to strong tidal currents. Manta Alley is located ~ 35.5 km in straight-line CHAPTER 4. KOMODO NATIONAL PARK 88 distance from KM in south-west Komodo NP. This site encompasses several rocky islands in the center of a large bay lined with steep cliff walls and rocky shorelines. The islands form several channels, exposed to tidal current flows on the northernmost side with sloping reefs (to ~ 35 m) to the east and west of the island group, where several manta ray cleaning stations are located. Table 4.1 Individual manta rays, sightings and survey days recorded between July 2004 and April 2018. Data are reported for Komodo National Park (KNP) overall, Cauldron (CL), Karang Makassar (KM), Mawan (MW), Manta Alley (MA), other sites within KNP and for manta rays sighted in KNP as well as in the Nusa Penida Marine Protected Area (NP MPA). Daily duplicates are removed (n = 755). All KNP CL KM MW MA Other KNP NP MPA Year Individuals 2004 – 12 139 1 122 - 16 2 5 2013 308 3 186 58 103 3 4 2014 473 8 270 92 186 5 6 2015 371 5 192 127 102 2 7 2016 457 8 277 103 159 9 8 2017 688 34 354 348 116 24 4 2018* 167 1 103 72 - 1 1 TOTAL 1104 48 779 535 383 45 11 Sightings 2004 – 12 177 1 156 - 18 2 13 2013 463 3 246 67 144 3 19 2014 838 19 404 116 294 5 16 2015 535 11 229 147 146 2 13 2016 765 10 375 117 253 9 19 2017 1200 59 476 464 174 25 6 2018* 188 1 107 79 - 1 3 TOTAL 4166 104 1993 990 1029 47 89 Survey days 2004 – 12 73 1 66 - 8 2 13 2013 116 2 85 23 14 3 19 2014 150 12 94 39 39 3 15 2015 97 9 56 35 15 2 12 2016 114 7 77 33 21 9 18 2017 198 30 103 103 13 20 6 2018* 43 1 26 25 - 1 3 TOTAL 791 62 507 258 110 40 86 *To April only - no data available for this time 89 4.3.2 Data Collection and Processing Data collection, including citizen science contributed data, validation and processing followed the procedures described in Chapter 3. Data on manta ray sightings (date, time, location, and identifying ventral photographs of manta rays) logged by researchers and the public were accessed from the online database ‘Manta Matcher’ Wildbook (www.mantamatcher.org; Holmberg and Marshall, 2018; Table 4.1). Approximately 20 trained observers contributed manta ray data from 2013 – 2018. Briefly, their training included details of how to take manta ray identification photographs, identify sex, maturity and behavior, estimate size, and how to record injuries and/or entanglements with fishing gear, along with other data collection, such as the number of dive boats present on-site (Chapter 3). Photo contributions from tourists was encouraged through educational presentations, informative dive briefings and awareness materials about ‘Manta Matcher’ (i.e. posters and infographics) displayed at local dive centers and within ‘liveaboard’ dive boats. All sightings records included in the study were validated and identifying photographs were manually matched by researchers and trained observers to an ID catalogue, with the assistance of automated pattern matching algorithm (Germanov & Marshall, 2014; Town et al., 2013) and/or external software (‘MantaUtil’, Winstanley, 2016). The sightings were independently checked for errors by the lead author. Manta ray sex was assigned based on the absence (female) or presence (male) of claspers, and maturity status was assigned based on clasper size and/or clasper scarring in males, and the presence of a pregnancy bulge and/or pectoral fin mating scars in females. This methodology means no females could be classed as immature in the absence of accurate size estimates; instead, their maturity status was classed as “unknown” (Marshall & Bennett, 2010; Marshall et al., 2011). Behavior was classified into four mutually exclusive categories: foraging, cleaning, cruising and CHAPTER 4. KOMODO NATIONAL PARK 90 courtship. Two additional non-exclusive behavior categories were also used, foraging/cleaning and courtship/cleaning, when more than one behavior was observed for an individual manta ray within a single dive. Further details on behavioral categorization are provided in Chapter 3, Figure 3.5. Overall, data were received from 330 different email submitters, totaling 4,163 sightings spanning from 7 July 2004 to 31 March 2018 (Table 4.1). The core sightings records used for statistical analyses (outlined below) were those collected between January 2013 and April 2018. Annual logged sightings records exceeded 400 across those years, with near year-round coverage for the three of the four main sites considered here. Data from 2018 were excluded from seasonal analyses, as year-round survey effort was not available, and from CL (all years) as there were relatively low records. We considered a single survey day as the unit of effort for this study, as the number of hours and/or daily dives logged was not regularly available. Trained observer dives were used as a proxy for survey effort, which was assigned as per dive, as dive time was not recorded prior to 2016. However, dive times set by dive operators are a maximum of 60 mins and the means of available dive times (post 2016) across sites were relatively consistent (58.3 ± 7.1 min). However, as most dive operators complete two or more dives daily at the remote site MA (i.e. effectively doubling daily effort compared to other sites, E Germanov, pers. obs.), we have also presented mean monthly sightings data as a more suitable comparison between MA and KM/MW (Figure 4.2). Use of Manta Matcher also facilitated identification of manta ray movements between geographical regions, such as re-sightings between Komodo NP and the Nusa Penida MPA. Sightings of all individuals are publicly available at www.mantamatcher.org. Identification numbers for the most re-sighted individuals and those making long-large movements are provided in the results section. These 91 identification numbers can be input into the search function of the online database to reveal full sighting histories for the individual manta rays. 4.3.3 Statistical Analyses Pearson's product-moment correlations, using the cor.test function of the R statistical software (R Core Team, 2018), were used to investigate the relationship between the annual number of survey days and number of sightings. The chisq.test function (R Core Team, 2018) was used to test whether the numbers of individuals and sightings differed between males and females at the sites KM, MW, MA and CL. Sex ratio data for each site were compared using chi-squared (χ2) goodness of fit tests (one-dimensional contingency table), while sex ratios and behavior frequencies between the sites were compared using a multiple-dimensional contingency table (vassarstats.net, 1998). To facilitate χ2 testing for behavior, where counts were less than five per site, the data were condensed into four categories (i.e. foraging/cleaning was reclassified to foraging; and courtship/cleaning was reclassified to courtship). Data from CL were excluded from χ2 testing for behavior and data from both CL and MW were excluded from χ2 testing for seasonality in foraging, as the counts were less than five for several categories. A Fisher's Exact Probability Test was used for 2 × 2 contingency tables if counts were less than five. The Yates’ continuity correction was applied to tests where there was one degree of freedom. CHAPTER 4. KOMODO NATIONAL PARK 92 Figure 4.2 Mean daily (A, B) and monthly (C, D) sightings of identified manta rays for the Komodo National Park at Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) from 2013 – 2017. Data are presented as annual (A, C) and monthly (B, D) mean sightings (± 1 SE) per unit effort (day – A, B; month – C, D). KM = 1,730 sightings, 415 days; MW = 911 sightings, 233 days; MA = 1,011 sightings, 102 days. 93 4.3.4 Analyses of Residency and Movement Residency, as described by Chapman et al., (2017) is considered to be the generally uninterrupted occupation of a specific area by an individual for specified time period. Along those lines, individuals can be temporary residents of site, that they leave and come back to. The return of an individual to the site of previous occupancy after a periodic absence greater or equal to the original residency period, is described as ‘site fidelity’ (Chapman et al., 2015). Here, we use the term ‘site affinity’ to describe same site re-sightings (Couturier et al., 2011; Jaine et al., 2014) as with our presence-only sightings data we are unable to discern whether these same site visitations are true site fidelity, or, at least for some individuals, movements within a large home range. We also deemed this term to be more appropriate in instances where high variability in site use between individuals exists, and when at least some individuals use several sites to similar levels (see Chapter 3). A modified maximum likelihood approach was used to compare manta ray re-sighting data against residency models to investigate residency patterns, following Chapter 3. Data from CL were not included in these analyses, as sighting records at this site were sparse in comparison to those at KM, MW and MA (Table 4.1). The ‘Movement Analyses’ module of the program SOCPROG 2.8 (Whitehead, 2009) was used to calculate Lagged Identification Rate (LIR), defined as the probability of re-identifying an individual after a given time lag. Empirical results were compared to model scenarios of both closed and open populations to estimate movement parameters (Table 4.2 and Table 4.3). Emigration, immigration, re-immigration and/or mortality were evaluated in the various open population models. The model that best fit the residency characteristics for each site was selected based on the lowest quasi-Akaike information criterion (QAIC) value, which accounts for over-dispersion of the data (Whitehead, 2007). The LIR CHAPTER 4. KOMODO NATIONAL PARK 94 analyses were extended to ‘within/between’ analysis, which estimates the probability of re-sighting an individual in a site other than the one where it was first identified after a set time lag (days). These analyses test for population-level mixing between sites. Model fits were bootstrapped 1,000 times to generate standard errors (SE). Table 4.2 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of manta rays at Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) in the Komodo NP, Indonesia. Study Sites KM MW MA Model Model description ΔQAIC ΔQAIC ΔQAIC A Closed (1/a1 = N) 58178.48 13966.47 81.84 B Closed (a1 = N) 101.70 10.14 81.84 C Emigration/mortality (a1=emigration rate; 1/a2=N) 15.77 4.83 11.90 D Emigration/mortality (a1=N; a2=mean residence) 15.77 4.83 11.90 E Closed: emigration + re-immigration (a1=emigration rate; a2/(a2+a3)=proportion of population in study area at any time) 73.33 3.12 61.46 F Emigration + re-immigration (a1=N; a2=res time in; a3=res time out) 42.88 3.12 2.90 G Emigration + re-immigration + mortality 1547.52 8.95 781.20 H Emigration + re-immigration + mortality (a1=N; a2=res time in; a3=res time out; a4=mort) 0 0 0 N = population The annual transition probabilities between sites, i.e. the probability of an individual manta ray moving from one area to another within a year, were calculated using a parameterized Markov movement model (Tables 4.4 and 4.5; Whitehead, 2009). This model includes a hypothetical ‘outside’ area (i.e. leaving the study site/s). Movements between the three core sites (KM, MW, MA) were investigated, as well as by grouping KM and MW as a collective ‘central’ site and investigating movements to and from MA in the south. Optimized values of transition probabilities were bootstrapped 1,000 times to generate SE, and the maximum number of evaluations was set to 10,000. 95 Mortality, including permanent emigration from all core sites, was considered in the model. Table 4.3 Model parameters, fits (ΔQAIC) and comparison for the Lagged Identification Rate (LIR) of (A) male (n = 507) and (B) female (n = 498) manta rays at Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) in the Komodo NP, Indonesia. Model Model description A Closed (1/a1 = N) B Closed (a1 = N) C Emigration/mortality (a1 = emigration rate; 1/a2=N) D Emigration/mortality (a1 = N; a2=mean residence) E Closed: emigration + re-immigration (a1 = emigration rate; a2/(a2 + a3) = proportion of population in study area at any time) F Emigration + re-immigration (a1 = N; a2 = res time in; a3 = res time out) G Emigration + re-immigration + mortality H Emigration + re-immigration + mortality (a1 = N; a2 = res time in; a3 = res time out; a4 = mort) A) Males B) Females Location KM MW MA KM MW MA Model ΔQAIC ΔQAIC ΔQAIC ΔQAIC ΔAIC ΔQAIC A 21508.8767 7201.8311 54.9819 78.1299 2.7558 40.8905 B 16.2036 7.7572 54.9818 78.1298 2.7558 40.8905 C 5.2733 5.1817 4.3354 6.4744 ~ 0 23.6264 D 5.2733 5.1815 4.3349 6.4743 0 23.6261 E 8.8397 0.9351 2.4273 63.3293 3.9765 25.3614 F 1.2962 0.9178 2.4273 8.4743 3.9711 0 G 0 232.0128 1.514 0.3141 2.2387 496.5678 H ~ 0 0 0 0 3.0742 0.0272 N = population CHAPTER 4. KOMODO NATIONAL PARK 96 Table 4.4 The estimated probability (± 1 SE) of an individual manta ray originally identified from one area being re-sighted within the same or different area. Movement probabilities are presented for manta rays (n = 1,061 individuals) at (A) the three core sites Karang Makassar (KM), Mawan (MW) and Manta Alley (MA). (B) Sites grouped into central (KM + MW) and southern (MA) regions. To: KM MW MA Outside A) Core From: KM 39.0% 37.0% ± 16.1% 12.9% ± 3.2% 11.1% ± 7.0% MW 29.0% ± 17.0% 59.2% 9.8% ± 2.0% 2.0% ± 4.5% MA 19.5% ± 9.7% 22.6% ± 6.3% 58.0% 0.0% ± 0.3% Outside 9.6% ± 5.9% 11.7% ± 5.0% 0.0% ± 0.3% 78.8% To: Central South Outside B) Central, south From: Central 86.5% 10.8% ± 4.0% 2.8% ± 19.8% South 46.5% ± 11.2% 53.2% 0.3% ± 1.4% Outside 11.1% ± 19.8% 0.0% ± 1.8% 88.9% Table 4.5 The estimated probability (± 1 SE) of an individual manta ray originally identified from one area being re-sighted within the same or different area. Movement probabilities are presented for (A) males (n = 507) and (B) females (n = 498) at core sites Karang Makassar (KM), Mawan (MW) and Manta Alley (MA). From Area: To Area: KM MW MA Outside A) Males KM 53.1% 27.4% ± 15.0% 11.3% ± 2.3% 8.3% ± 4.6% (n = 1,682) MW 31.9% ± 12.3% 50.5% 7.8% ± 2.3% 9.8% ± 8.6% MA 21.0% ± 6.5% 15.2% ± 5.5% 63.8% 0.0% ± 0.0% Outside 1.5% ± 1.8% 2.2% ± 1.7% 0.1% ± 0.4% 96.3% B) Females KM 43.2% 27.7% ± 17.0% 17.3% ± 4.5% 11.1% ± 0.7% (n = 2,051) MW 31.9% ± 18.0% 41.0% 14.5% ± 4.4% 2.0% ± 4.5% MA 23.3% ± 11.4% 18.7% ± 9.8% 58.0% 0.0% ± 0.3% Outside 15.1% ± 8.9% 9.1% ± 4.0% 0.0% ± 0.1% 78.8% 4.4 RESULTS 4.4.1 Sightings and Survey Effort A total of 1,104 individual manta rays were identified from 4,163 sightings (excluding daily duplicate data for the same individuals; n = 795) across 791 unique dates (from 7 July 2004 to 31 March 2018) within the Komodo NP. Sighting records for four sites within the park, Cauldron (CL), Karang Makassar (KM), Mawan (MW) and Manta Alley (MA), and within core survey years (2013 – 2018) represented ~ 98.8% (n = 4,116) of total sightings (Table 4.1). Sightings records prior to 2013 (n = 177) included 139 97 individuals. In addition to sightings within the Komodo NP, 11 (1%) individual manta rays sighted within the Komodo NP were also sighted within the Nusa Penida MPA, on 89 separate days (including pre-2013 data; Figure 4.3). Annual manta ray sightings varied across study years (Table 4.1) and were positively correlated with survey effort (r = 0.94, p = 0.002). There were > 400 sightings logged on Manta Matcher in each of the core years. Data submissions to Manta Matcher gradually increased over the study period with increased awareness of the citizen science program within the dive community, surpassing data collected solely by trained observers in 2014 (Figure 4.4). Sightings rates averaged (mean ± 1 SE) 760 ± 130 sightings/year from 2013 – 2017, and January – March of 2018 had 188 sighting records. The highest number of sightings records was from 2017 (n = 1,200), with the second highest number from 2014 (n = 838; Table 4.1), although it should be noted that survey effort by trained observers was lower in 2015 and 2016. Daily and monthly mean sightings varied among years (Figures 4.2A, C). There was an overall increase in sightings from 2013 to 2017 across the sites, with the largest increase observed at MW. Sightings per dive recorded solely by trained observers also varied modestly for sites among years, with the exception of KM where there was a ~ two-fold increase in sightings per dive during 2015 and 2016 than in 2013 and 2014 (Figure 4.5A). Sightings per dive by trained observers also varied across months and sites and followed similar trends to the sightings overall (Figures 4.2B, D) with a decrease in sightings for the central sites (MW and KM) mid-year (Figure 4.5B). While a higher rate of sightings was recorded by trained observers for MA mid-year, data was not available for December - February and in June prohibiting us from commenting on seasonality trends based on this data alone. Although observer logs lacked complete year-round effort in 2015 and 2016, citizen science data was logged on Manta Matcher throughout the year (Figure 4.2B, D). CHAPTER 4. KOMODO NATIONAL PARK 98 Figure 4.3 Individual manta rays were sighted in both Komodo NP and Nusa Penida MPA. The sighting location (site) and date are plotted for each individual according to their identification (ID) codes. The sex of the individual is indicated within the brackets following the ID code, where (F) indicates female and (M) male. The ID codes, excluding the bracketed sex information, can be used to call up full sighting records for the individuals from ‘Manta Matcher’. Figure 4.4 The combined number of annual survey days at manta ray sites in Komodo National Park (Karang Makassar, Mawan, Manta Alley and Cauldron) by trained observers (Logs), and additional manta ray survey days submitted to ‘Manta Matcher’. 99 Figure 4.5 The mean individual sightings per dive (ndives = 617) logged by trained observers per (A) year and (B) month in Karang Makassar (KM), Mawan (MW), and Manta Alley (MA), Komodo NP. Annual weather patterns also influenced the results. Strong winds during December – February limit access to the north (CL) and central (KM, MW) sites, while there is limited access to the south (MA) during June – August. Survey effort is consequently reduced during those times in the respective sites. Sightings in central sites nonetheless displayed a seasonal trend, with less sightings in KM and MW during July – September than at MA (Figure 4.2B, D). Limited (three survey days) or no (January) sighting data were available for MA during December – February. CHAPTER 4. KOMODO NATIONAL PARK 100 A total of 1,085 individual manta rays were identified from data collected during the core study years (January 2013 to April 2018) within the Komodo NP. Data collected from the three core study sites KM, MW, MA yielded 1,061 individual manta rays. Of these individuals, 749 were sighted in KM, 535 in MW, and 376 in MA. Data collected from CL yielded 48 individuals, of which 22 were not sighted elsewhere, and two were sighted at all of the other three sites. A further 45 individual manta rays also had sighting records in sites other than the above four sites, of which two had no other records elsewhere.Newly identified individuals were plotted against the days elapsed since the start of the study to create discovery curves. Gaps in the discovery curves indicate a gap in surveys during these time periods. Considering KM, MW and MA sites combined (i.e. the three sites with substantial sighting records), the discovery curve of newly-identified individuals showed a steep rise until approximately 685 individuals (860 days), and continued to increase at a slightly lower rate until the end of the study period (1,061 individuals after 1,913 days) (Figure 4.6). Neither the combined discovery curve, nor any of the site-specific discovery curves, approached an asymptote. However, the rate of increase in the number of new individuals discovered at MW slowed after the identification of 224 individuals at 710 days, until it reached 254 individuals at ~ 1,200 days (between August 2015 and October 2016, or years 2 and 3 of data collection), at which point identifications increased steeply to 535 until ~ 1,700 days (Figure 4.6). 101 4.4.2 Population Structure The sex ratio of males (507) to females (498) identified within Komodo NP was not significantly different overall from 1:1 (χ21 = 0.1, p = 0.777), with the sex of 56 individuals remaining unknown. Likewise, the sex ratio did not differ significantly from 1:1 at KM (0.92:1, χ21 = 1.4, p = 0.234 ), MW (0.97:1, χ21 = 0.2, p = 0.692) or MA (0.9:1, χ21 = 1.0, p = 0.319) when the sites were considered separately (Figure 4.7A). While there were more males than females (23 vs. 16, with nine unknown individuals) identified at CL, this difference was not significant (χ21 = 1.3, p = 0.262). However, based on sightings, the population structures at KM, MW and MA were significantly different (χ21 = 32.1, p = 1.47 × 10-9; Figure 4.7B). The most divergent site was KM, where males represented 43% and females 57% of sightings (χ21 = 35.1, p = 3.176 × 10-9). Figure 4.6 Discovery curves for newly-identified manta rays over time (in days) from January 2013 until April 2018. The x-axis gray lines indicate survey years. Discovery curves are presented for the Komodo NP (Comb.) and broken down by sites Karang Makassar (KM), Mawan (MW), and Manta Alley (MA). NS = number of survey days and NI = number of individuals. CHAPTER 4. KOMODO NATIONAL PARK 102 Sightings at CL, although fewer in number (n = 103), were biased towards males (66% vs. 34% females, χ21 = 9.8, p = 0.002). There was no significant association between site and sex for individuals (χ22 = 0.3, p = 0.851), nor for sightings (χ22 = 4.9, p = 0.087) at the three main sites. However, there were significantly more male than female sightings at CL (χ23 = 21.5, p = 8.128 × 10-5), but not individuals (χ23 = 2.1, p = 0.549). Across all sites, maturity was assigned to all individual males. However, on 7% of the total sightings the maturity status could not be determined (i.e. unknown). Females are more difficult to assign a maturity status without accurate size estimates, which we lacked, and for 290 females representing 66% of total sightings the maturity status was unknown. Considering KM, MW and MA combined, at least 43% (n = 212) of the females were sexually mature, and just under half of those (n = 92, 18% of overall females) were observed to be pregnant during the study period. Approximately 90% (n = 454) of the males were sexually mature, with ~ 11% (n = 53) of males classed consistently as immature across the duration of the study. An additional 11 immature males were identified at CL and not sighted at other sites. Altogether, 96 individuals (all males) or 9% of the total were classed as immature at some point throughout the study, with 32 of these reaching maturity over this period. The maturity status of males and females was comparable among all sites (Figure 4.7A; χ23 = 1.2, p = 0.742), with more mature males than females based on individuals where the sex was known. Sightings based on maturity status were also comparable among sites (Figure 4.7B; χ23 = 5.3, p = 0.140), with mature males sighted more often than mature females. Looking at maturity status alone, there was relationship between the site and maturity status of individuals (χ23 = 78.7, p =2.200 × 10-16), with a larger proportion of immature than mature males sighted at CL (38% vs 21%; Figure 4.7A). Likewise, there was a relationship between site and maturity status for the number of sightings of individuals 103 (χ23 = 270.8, p = 2.200 × 10-16), with immature males sighted more often than mature males at CL (55% vs 11%; Figure 4.7B). Notably, of the manta rays identified at CL sighted more than once (range: 2 – 11 sightings), eight out of 17 individuals (47%) were classed as immature males at some stage during the study, with two maturing during the study (Figure 4.8). All females except one encountered more than once at CL were classed as unknown maturity status. Conversely, all manta rays (nfemales = 6 and nmales = 5; Figure 4.3) that were identified in both Komodo NP and Nusa Penida MPA were classified as mature, with estimated sizes ranging from 2.25 – 4 m for those individuals with available size estimate data (n = 6). 4.4.3 Residency and Movement The majority of manta rays were encountered more than once within the core sites (n = 768; 72%). A few individuals were sighted more than 10 times (n = 48; 5%), with up to a maximum of 21 sightings per individual (Figure 4.9). The mean re-sightings per individual (± 1 SE) was 3.6 ± 0.1. The majority (n = 805; 76%) of these individuals were re-sighted across multiple years, with the remainder only sighted in a single year. The mean and median periods between re-sightings across all individuals (i.e. lag) was 239 days (± 5), and 132 days, respectively (range = 1 day to 4.8 years). For the most sighted male (16 sightings) and female (21 sightings), greater than 60% of re-sightings had a lag period of less than 2 months, with the longest lag periods of these individuals being 635 and 620 days, respectively. The longest time between the initial and most current re-sighting for an individual (INKNP0154A) was 13.4 years. This female was first sighted in 2004 with unknown maturity status, but was observed with a mating scar in 2014. Overall, of the seven most re-sighted individuals (≥ 15 sightings), four females had the highest number of sightings (range: 17 – 21) followed by three males (range: 15 – 16). CHAPTER 4. KOMODO NATIONAL PARK 104 Figure 4.7 Population structure of manta rays in the Komodo NP from 2013 – 2018. (A) Population structure and (B) sightings for the combined (Comb.) core Komodo National Park sites and broken down by sites (from north to south): Cauldron (CL), Karang Makassar (KM), Mawan (MW) and Manta Alley (MA). 105 Figure 4.8 Manta rays sighted more than once at the Cauldron (CL) and their sighting locations. Colored circles show sighting locations. Solid vertical line indicates the first sighting where the individual is deemed mature. *indicates individuals that were immature throughout the duration of the study. The most re-sighted individual was a mature female (INKNP0150A) that was seen a total of 21 times and seen at all three core sites between April 2013 and February 2018, including while pregnant in April 2017. This female was recorded moving from MA to the central sites, and back again, five times during this almost six-year period. However, few of the most sighted individuals in the central area of the park were also sighted at MA. For example, only two out of 13 of the most sighted individuals at KM and three out of nine of the most sighted at MW were also sighted at MA (Figure 4.10A, B). Conversely, 11 of the 13 most re-sighted individuals at MA (≥ 10 sightings; nfemales = 6 and nmales = 5), were sighted at both MA and the central locations throughout the study (Figure 4.10C). The most re-sighted male (INKNP0111A), classed as immature throughout the study, was seen 16 times (13 times at MA and three times at central sites) between September 2013 and December 2017. This individual was only recorded in MA CHAPTER 4. KOMODO NATIONAL PARK 106 until November 2014, after which time no sightings of this individual were recorded until he was sighted at KM in August 2016. Approximately 91% (n = 961) of individuals identified in this study were sighted at least once in the central sites of the park highlighting the importance of this area. Only 9% (n = 100) of individuals were exclusively sighted at MA. There were 104 (9.8%) individuals recorded in all three core sites, and 275 (25.9%) were sighted in both the central area (either KM or MW) and MA. We thus tested whether the core sites were fully mixed (i.e. a single population) using LIR analysis (Figure 4.11A). The LIR curves for re-sightings within the same or different site did not converge during the study period. This result indicates that, although interchange between sites does occur, full interchange was not present and most individuals had a high affinity for one site over the others. Figure 4.9 The proportion of individual manta rays in Komodo NP plotted against the number of times individuals were sighted between January 2012 and April 2018. (A) Combined sightings from the three core sites (Comb.), (B) Karang Makassar (KM), (C) Mawan (MW) and (D) Manta Alley (MA). Total number of individuals are shown for each number of sightings. 107 Figure 4.10 The top most sighted manta rays at (A) Karang Makassar (KM), (B) Mawan (MW) and (C) Manta Alley (MA). Colored circles show sighting locations. CHAPTER 4. KOMODO NATIONAL PARK 108 Similarly, the Markov movement model showed that the re-sighting probability for each individual was highest within, rather than between, sites (39% at KM, 59.2% at MW and 58.0% at MA; Table 4.4A). However, there was a relatively high probability of movement from KM to MW (37.0 ± 16.1%) or vice versa (29.0 ± 17.0%), indicating sizeable interchange between the two adjacent central sites (~ 5 km apart). When these two central sites KM and MW are grouped (Table 4.4B), a substantial probability of re-sighting within the central area was documented (86.5%), with smaller, yet considerable re-sightings (53.2%) at MA. Further, the movement between central sites and MA was asymmetric, with far less movement from central sites to MA (10.8 ± 4.0%) than vice versa (46.5 ± 11.2%). More movement to the central sites from other, ‘outside’ sites was modelled (11.1 ± 19.8%) than from ‘outside’ sites to the south (0.0 ± 1.8%). Notably, the model maximum number of iterations was exceeded for these analyses, suggesting that the estimated SEs are inaccurate and larger than expected (H. Whitehead, Dalhousie University, pers. comm.). Considered together, however, the movement model results indicate that individuals routinely travel to the central sites from MA and other sites, but not necessarily vice versa. Movement from ‘outside’ the central sites could, in this case, also include re-sightings from CL and further areas such as Nusa Penida MPA, where there is documented exchange. We observed that 40% of individuals sighted at CL were also sighted in the central sites, and 13% of individuals sighted at CL were also sighted at MA. For long-range movements to Nusa Penida MPA, considering all available data (2008 – 2018), the days between recaptures in the different regions for the 11 individuals varied widely (range: 33 – 1,550 days) and averaged 391.4 ± 77.7 days. Based on the results of the ‘within/between’ and movement analyses, as described above, the three core sites were treated separately for LIR analysis (Table 4.2, Figure 109 4.11B). Although a high degree of exchange occurred between the two central sites, a ‘within/between’ analysis between just KM and MW did not support a fully mixed population within the central area (Figure 4.12) so these sites were also treated separately. The best-fit re-sighting model was Model H for all three sites (Table 4.2). This model describes a pattern of emigration and re-immigration with some mortality (which includes permanent emigration). The LIRs within the two central sites were similar and much lower (≤ 2 x) than those at MA, indicating that individuals have a higher probability of re-sightings at MA than either KM or MW (Figure 4.11B). Model H scenarios provide information on the proportion of time individuals spend within (residence time in) and out (residence time out) of an area, as well as mortality. The rates that individuals remained at each site differed, especially between MW and MA. Individuals stayed approximately twice as long at MA than at MW, although there was high variation among individuals (1.6 ± 12.9 days vs. 0.8 ± 0.3 days at MW). Individuals sighted in any of these three sites spent comparable time outside the sites (range: 5.8 – 6.9 days); however, the results for MA indicated more variability between individuals than at MW and KM (Table 4.6). Exploration into potential sex-linked differences in site use showed no substantial differences in LIRs between the sexes (Figure 4.13, Table 4.3). However, movement analyses indicated that females tended to move more from central sites to MA, and from ‘outside’ to the central sites, than males (Table 4.5) suggesting that there might be some sex-linked differences in site use. Mortality (which includes permanent emigration) was considered negligible for all sites and analyses (≤ 0.045). Table 4.6 Model H outputs for time spent within (res time in ± 1 SE) and outside (res time out ± 1 SE) of Karang Makassar (KM), Mawan (MW) and Manta Alley (MA) sites in the Komodo NP, Indonesia. Site KM MW MA Res time in (days) 1.3 ± 5.4 0.8 ± 0.3 1.6 ± 12.9 Res time out (days) 6.9 ± 5.3 5.8 ± 4.4 5.8 ± 16.7 CHAPTER 4. KOMODO NATIONAL PARK 110 Figure 4.11 Lagged Identification Rates (± 1 SE) of manta rays within and between the three core study areas in Komodo NP (2013 – 2018): Karang Makassar (KM), Mawan (MW), and Manta Alley (MA). (A) The probability of an individual manta ray being identified in the same or a different site within the Park over time (days). (B) The probability of an individual manta ray being identified over time (days) within each of the three core sites. Best-fit LIR models (Table 4.2; dotted lines) are shown for each site. Figure 4.12 The mean (± 1 SE) probability of re-identifying an individual manta ray within the same or different central core site (Karang Makassar and Mawan) within Komodo NP, Indonesia. Records from January 2013 to April 2018. 111 Figure 4.13 The mean probability (± 1 SE) of re-identifying an individual manta ray, broken down by sex, from the time of first identification until an average of approximately 3.5 years later within the three study areas, (A) Karang Makassar, (B) Mawan, and (C) Manta Alley, Komodo NP, Indonesia. The predicted LIRs for models of best fit (Table 4.5) are shown for each group. Records from January 2013 to April 2018. 4.4.4 Behavior Manta rays were observed foraging, cleaning, cruising, and engaging in courtship at KM, MW, and MA (n = 2,306, Figure 4.14A). The CL site was excluded from statistical analysis because of the low number of behavioral observations (n = 49), although some interpretations of these results are retained below. There was a significant CHAPTER 4. KOMODO NATIONAL PARK 112 association between behavior and site (χ26 = 202.1, p < 0.001), with MW having the highest proportion of cleaning behavior (74%), while CL had the lowest (6%). More cleaning was observed at KM (53%) than MA (39%). Overall, there were few records of mixed behaviors (i.e. cleaning and courtship = 20; cleaning and foraging = 1) for individuals within a single survey dive. Substantially less foraging behavior was observed at MW (1%) compared to the other sites, especially CL, where the highest proportion of foraging behavior was observed (20%). There were seasonal differences in the number of individuals observed foraging at KM and MA (χ21 = 28.9, p = 7.52 × 10-8; Figures 4.14B, C), the two sites where data on foraging behavior were available from both seasons. However, no significant differences were detected based on the number of days that feeding events were recorded (p = 0.058), which were few. However, the many individuals observed foraging at one time at KM (up to 30 individual identifications in one day) are indicative of mass feeding events, which were rare in the dataset. Similarly, foraging was recorded at CL during the south-east monsoon season. While foraging at MW was recorded during the north-west monsoon season. Courtship behavior was observed year-round (Figure 4.14D) at all sites except CL. Mean daily recording of courtship behavior was modestly higher in May. Cruising behavior was frequently observed at CL (74%) and, to a lesser extent, MA (45%). Both locations have strong currents flowing through narrow channels where manta rays are commonly observed swimming into the current, but not necessarily foraging or cleaning (E Germanov, pers. obs.). 113 Figure 4.14 (A) Manta ray behaviors observed between January 2013 and April 2018 from four sites within Komodo NP. Sightings at Cauldron (CL, 49), Karang Makassar (KM, 1,067), Mawan (MW, 741) and Manta Alley (MA, 498). The number of foraging individuals (B) and feeding events (C) broken down by site and season (November – April = north-west monsoon, NW; May – October = south-east monsoon, SE). (D) The average monthly sightings per unit effort (day) of identified manta rays engaging in courtship behavior. Data recorded between 2013 and 2017 across sites KM, MW and MA and presented as the daily means (± 1 SE). The textured and solid backgrounds indicate NW and SE seasons, respectively. CHAPTER 4. KOMODO NATIONAL PARK 114 4.4.5 Injury Rates A total of 56 individuals (~ 5%) had either cephalic fin, pectoral fin, and/or fishing line injuries (Appendix II.ii Supplementary Table 1B). Nine individuals had more than one injury. The breakdown of injuries was: 13 hook and line entanglements, 30 cephalic fin, and 26 pectoral fin truncations or disfigurements. Six of these injured individuals were also observed to be pregnant during the study period. Predatory injuries i.e. bite marks were present on 32 (~ 3%) individuals. 4.4.6 Boating Activity The two central sites, KM and MW, are the closest to the town of Labuan Bajo on Flores island, where the majority of diving/snorkeling tours operate from. Based on observer logs, these sites saw a combined 34% increase in the number of tour boats recorded from 2014 to 2017 (Figure 4.14A). This increase was greater at MW than at KM, with the daily average of boats at MW increasing from 1.8 (± 0.2) in 2014 to 3.4 (± 0.3) in 2017 (81% increase), while at KM boats increased from 5.8 (± 0.6) in 2014 to 7.7 (± 0.6) in 2017 (33% increase). The number of boats on-site during July and/or August were 102% (12.3 ± 3.2 boats) and 50% higher (4.1 ± 1.2 boats) than the September – June average for KM and MW, respectively (Figure 4.14B). Records for CL and MA were sparse and were not available year-round. However, for comparative purposes, and to establish a baseline, from 2014 – 2017 inclusive, the daily mean and median number of boats at CL were 2.6 ± 0.3 and 2 (range: 1 – 14), respectively (n = 57). While the number of boats are much lower than those of the central sites, daily numbers well above average (i.e. 10 and 14 boats) were reported in August 2017, suggesting that a significant increase in tourism is also occurring at this site. While, based on limited records (n = 27), it 115 appears that the daily mean (1.9 ± 0.2) and median (2; range: 1 – 4) boat numbers at MA have remained relatively stable over time. Figure 4.15 The daily average (± 1 SE) number of boats present at core manta ray sites in the central region of the Komodo NP overall, and separately for Karang Makassar (KM) and Mawan (MW), across year (A) and month (B) for 2014 and 2017. 4.5 DISCUSSION A substantial number of manta rays display site affinity to the Komodo National Park (~1,100 individuals from almost 4,200 sightings over ~ 5 years). The total number of identified individuals is noteworthy relative to other well-studied manta ray habitats around the world (see Table 3.5, Chapter 3). While less new individuals were recorded after the second year of data collection, new identifications were common throughout the study indicating that the number of individuals using the area is actually much larger than 1,100. Within the Komodo NP, four sites were identified manta rays were commonly sighted: Cauldron (CL), Karang Makassar (KM), Mawan (MW) and Manta Alley (MA). Overall, manta rays display a high degree of site affinity to sites within the Komodo NP, in agreement with the conclusions of an earlier acoustic telemetry study (Dewar et al., CHAPTER 4. KOMODO NATIONAL PARK 116 2008). However, around 1% of individuals have been re-sighted ~ 450 km away at the Nusa Penida MPA indicating that manta ray populations within the Lesser Sunda region are connected by long-range movements of individual rays. Interestingly, despite having accessed a much larger dataset, a proportionally equal number of individuals were uncovered to make these long-range movements as in the previously published study (Germanov & Marshall, 2014). The large number of manta rays re-sighted within the Komodo NP, including immature individuals, and the documentation of courtship behaviors emphasize the importance of this region for manta rays in Indonesia. The clear increase in tourist visitation to the two central sites, in particular, highlights the need for a comprehensive management strategy to mitigate any detrimental impacts on the manta rays by the growing tourism, inadvertently harming tourism itself. 4.5.1 Residency and Movement This is the first population-level study of manta rays in Komodo NP. The previous acoustic study by Dewar et al. (2008) provided a sighting-independent means to constantly monitor tagged individuals at multiple sites. Despite the temporal gap in these projects, there was substantial overlap in the specific reefs that were shown to be most frequented by manta rays. Dewar et al., (2008) revealed that manta rays typically used these habitats during the day, indicating that our sightings-based sampling during the day is an accurate record of site use, allowing the incorporation of substantial citizen-science data. Generally, the present study has revealed trends in site affinity and seasonal movements that are consistent with those reported earlier by Dewar et al. (2008). An exception is that Dewar et al. (2008) reported few visitations for manta rays to MW, while we found MW contained an important cleaning station that was visited by half of the known individuals in the region. Despite the lack of continual monitoring, the 117 similarity in results validates the use of sightings data and citizen scientists to make population-level observations. The potential for relatively consistent observations is key to assessing the potential for negative impacts by excessive tourism and other threats. In this study, many individual manta rays were frequently re-sighted within the Komodo NP. Individuals typically showed higher affinity to one site over the others. These trends have been conserved for over a decade (Dewar et al., 2008) and similar findings have been documented in manta rays elsewhere in Indonesia (Chapter 3; Perryman et al., 2019) and other countries (Couturier et al., 2018; van Duinkerken, 2011; Venables et al., 2020). Affinity to MA, in the south, was higher than to sites centrally located in the park although this varied between individuals. Estimations of residence time in and out of sites were most variable for MA, potentially a reflection of differential site use by a specific demographic, e.g. males being more mobile than females (van Duinkerken, 2011), mature individuals being more mobile than juveniles (Chapter 3), or social groups (Perryman et al., 2019). Importantly, Dewar et al. (2008) also reported frequent observations of manta rays feeding at German Flag, a site adjacent to MA (~ 1.8 km away). The present study lacked information from German Flag; however, based on the results from Dewar et al. (2008), it is possible that sustained foraging opportunities at this site contributed to the higher residency at MA, particularly as in the previous study 20 out of 24 individuals recorded at German Flag were also recorded at MA. While the results from our study indicate that connectivity exists throughout the park, within/between analysis also indicates that we cannot consider it as a single population for modelling purposes here. There was higher interchange (i.e. immigration and emigration) between the two nearby central (~ 4.5 km apart) sites than elsewhere, indicating substantial connectivity between the individuals using these locations. More individuals were identified from the two central sites, but these individuals had a lower CHAPTER 4. KOMODO NATIONAL PARK 118 probability of being re-sighted (lower LIR) than those observed to have site affinity to MA. Further, the resident times in central sites for individuals were also shorter, than those observed to have site affinity to MA, albeit with less variability. Taken together, these observations could be explained by repeated movements between the two central sites by individuals. Indeed, grouping KM and MW for movement analysis indicated that the majority of movements were within this central area rather than to MA in the south, although a certain degree of structure was still maintained. Conversely, nearly half of the individuals moved from MA to the central area of the park. A notable percentage (~ 10%) of movements were from other ‘outside’ areas, which might include CL, the Nusa Penida MPA, where there is documented exchange (Germanov & Marshall, 2014), or other yet to be investigated manta ray aggregation areas (Conservation International Indonesia, 2016). Some of this apparent site connectivity between CL, central sites and MA might be partially explained by seasonal changes and shifts in prey abundance. However, as movements do not appear to be reciprocal (Figure 4.16; Dewar et al., 2008), differences in sex, age and social hierarchy could also influence these results (Chapter 3; McCauley et al., 2014; Perryman et al., 2019; Stevens, 2016). The lower re-sighting rates and LIRs (i.e. higher degree of transience) in Komodo NP than reported elsewhere in Indonesia (Chapter 3; Perryman et al., 2019), coupled with documented broader movements of manta rays throughout the region (Conservation International Indonesia, 2016; Germanov & Marshall, 2014; present study), indicates a degree of dispersal from Komodo NP. Thus Komodo NP could be serving as a corridor that links manta populations throughout the Lesser Sunda region. Until recently, manta rays were actively targeted in fisheries < 400 km to the east and west of the Komodo NP (Heinrichs et al., 2011; Lewis et al., 2015; White et al., 2006), which suggests that additional adjacent aggregations might exist. Links between different populations enables 119 the exchange of genetic diversity (Bonfil et al. 2005; Skomal et al. 2009) bolstering the resilience of populations and ecosystems (Oliver et al., 2015; Sgrò et al., 2011). Thus, it is fortunate that the conservation area has been in place for 30 years prior to the declaration of nation-wide manta ray protection, as it has likely been acting as a refuge for manta rays from fishing over this period. Figure 4.16 Movement patterns of male and female manta rays in the Komodo NP. Pink arrows denote governing movements of females, blue of males, and gray arrows denote no observed differences between the sexes. Population demographics Population demographics, particularly sex-linked differences in movements appear to influence fine-scale habitat use in manta rays (Chapter 3; Deakos et al., 2001; van Duinkerken, 2011; Perryman et al., 2019; Stevens, 2016). In this study we CHAPTER 4. KOMODO NATIONAL PARK 120 investigated population structures by comparing the demographics of overall individuals visiting sites as well as the sightings of these individuals. While sighting demographics is more an indicator of site affinity than the overall population structure, to understand habitat use it is valuable to gain insight into what the population structure might look like at any given time, which is better represented by the sightings data. The three core sites, KM, MW and MA, were all dominated by mature manta rays, and had similar demographic profiles. However, the observed bias towards mature males over females might be attributed to the difficulty in assigning maturity to females in absence of accurate size estimates and visual maturity indicators (i.e. mating wing scars or pregnancy bulge). The sole northern site investigated, CL, was quite different – it appears to be used primarily as a foraging site for immature manta rays, and it possibly serves as a nursery for these smaller individuals in the earlier years of their lives (Chapter 3; Heupel et al., 2007; Martins et al., 2018; McCauley et al., 2014; Stevens, 2016). Fewer immature individuals were recorded in Komodo NP (n = 96) as a whole compared to Nusa Penida MPA (n = 123; Chapter 3). This discrepancy in the percentage of immature individuals identified in Komodo NP (9%), relative to Nusa Penida MPA (20%), could indicate insufficient survey effort at CL and at other sites not regularly visited by tour operators (e.g. German Flag) that may be important to immature individuals. Dewar et al. (2008) reported that German Flag had the highest visitation rates of any of the other sites investigated in this earlier study. This coupled with the observation that individuals were commonly feeding at German Flag, and that some immature individuals (as small as 1.8 m) were tracked, it is possible that immature individuals also disproportionally use German Flag. While the present study lacks sighting data at this site, the habitat use patterns of frequently sighted immature individuals (≥ 10 sightings) from other study sites within Komodo NP show that when 121 younger some individuals have higher site affinity than when they are older. This observation is supported by similar findings in Manta Bay, Nusa Penida MPA (Chapter 3) and in the Seychelles (Peel et al., 2019a). In contrast to many other well-studied populations (See Table 3.5, Chapter 3), the manta rays identified in Komodo had an equal ratio of males to females overall. Further, in contrast to observations within the Nusa Penida MPA (Chapter 3), the LIRs also did not differ between the sexes. Similarly, Couturier et al., (2018) did not consider that the sex of individuals was a major predictor on manta ray visitation rates to Lady Elliot Island in Australia. However, movement analysis suggests that females within the Komodo NP are slightly more mobile, and sites not specifically identified in the analysis (i.e. ‘outside’) were more commonly used by females than males. This differs from results at other locations where males are more mobile than females (Chapter 3; Deakos et al., 2011; Perryman et al., 2019; Stevens, 2016). However, sex-biased dispersal does not appear to be a factor in the long-range movements from Komodo NP to the Nusa Penida MPA, as individuals of both sexes were documented moving between these locations. Rather than natal philopatry by females or genetic dispersal by males, as reported for other elasmobranchs and species in general (discussed by Chapman et al., 2015; Li & Hanna, 2019; Trochet et al., 2016), perhaps these long-range movements might be better explained by food seeking behavior. Sex-biased dispersal in most species, including marine animals, is often associated with reproduction (Chapman et al., 2015; Hoffman et al., 2012; Lohmann et al., 2013; Rooker et al., 2008), inbreeding avoidance, kin competition and food (Li & Hanna, 2019). For example, some species of elasmobranchs return to natal grounds specifically for parturition or to aid in gestation (Chapman et al., 2015; Feldheim et al., 2014; Mourier & Planes, 2013). For the manta rays in the current study, given that both sexes CHAPTER 4. KOMODO NATIONAL PARK 122 tend to make long-range movements, natal philopatry by females for the purpose of parturition seems unlikely. Similarly, sex-biased movements could be linked with genetic dispersal (Daly-Engel et al., 2012; Deakos et al., 2011; Li & Hanna, 2019; Pardini et al., 2001; Stevens, 2016); however, while this might be occurring on a smaller spatial scale (e.g. Chapter 3; Deakos et al., 2011) our limited observations cannot confirm that this is the main reason for the long-range movements either. Sex-biased dispersal has also been linked to sexual dimorphism; however, these links are mostly in the instance that the males are the larger sex, which is thought to benefit them during intense competition for mates (Trochet et al., 2016). Manta rays do exhibit sexual size dimorphism; however, it is the females that reach the larger size. This suggests that females likely have higher energy demands, especially during gestation. These demands might require females to venture further or more often into highly productive pelagic zones to feed. In the Seychelles, isotopic concentrations of stable isotopes varied more for females than males, suggesting that at least some females have more diverse nutritional sources, or potentially forage in a greater variety of locations (Peel et al., 2019b). However, larger males would also have higher energy requirements than smaller males and thus these large-scale movements might be a necessity of size rather than sex. Indeed, Peel et al., 2019a noted that larger manta rays travelled further distances on average, while immature individuals displayed greater residency. While noting that data on the size of all travelling individuals were not available in this study, the observed interchange between Nusa Penida MPA and Komodo NP could be a symptom of these further offshore foraging movements, followed by re-sightings in nearby coastal aggregation areas as individuals return to orientate, clean and and engage in reproductive behavior. Within the Nusa Penida MPA, it was suggested that ontogenetic/age-related shifts in site use were linked to prey availability, or more likely prey density (Chapter 3). Once 123 mature, it appears that individuals shifted to a more mobile lifestyle, characterized by increased movements between nearby sites in Komodo NP (e.g. the most sighted male from this study was initially only sighted at MA and only once mature was sighted at KM), Nusa Penida MPA (Chapter 3) and elsewhere (Peel et al., 2019a). Thus, lower prey density might be able to sustain immature and/or smaller individuals (more often males than females because of the larger maximum size of females), whereas larger individuals might need to forage mostly offshore to meet their greater energy demands. However, in Komodo NP, essentially the same patterns of ontogenetic classes were observed at three of the study sites, suggesting that a variety of foraging opportunities exist within the area for all size classes. The exception to this was CL, which was frequented mostly by immature individuals, and relative to other observed behaviors, had a two-fold greater foraging activity than that recorded at the other sites. This indicates that conditions at CL, are at times ideal to meet the nutritional demands of younger individuals, similar to the characteristics of Manta Bay, Nusa Penida (Chapter 3). Further, it appears that regular foraging opportunities in shallow coastal waters might be an important factor for identifying manta ray nurseries, or at very least, for recurrent sightings of immature individuals (Chapter 3; Stevens, 2016). Influence of environment A concentration of year-round foraging opportunities and high prey density likely contribute to the high numbers of individuals that use the Komodo NP (Anderson et al., 2011; Armstrong et al., 2016). Site use and movements between sites are to some extent also likely driven by prey availability. For example, higher manta ray numbers were seen at cleaning stations when prey was locally abundant in Mozambique, indicating that they likely prefer cleaning on reefs neighboring good feeding grounds (Rohner et al., 2013). CHAPTER 4. KOMODO NATIONAL PARK 124 Further, the patchy dispersal of plankton leads to certain habitats, such as channels and coral ledges, to be ideal for manta ray aggregations (Papastamatiou et al., 2012). The unique geographical location of Komodo NP, located in an area of several oceanographic features (e.g. mixing of waters from the Pacific Ocean via the Indonesian Throughflow (ITF) and the Indian Ocean, tidal mixing and productive upwelling) allows for year-round productivity that shifts in response to the broader regional oceanographic processes (monsoon driven influence on ITF and upwelling strength; Ningsih et al., 2013). Furthermore, seasonal associated shifts in manta ray abundance are well documented elsewhere (Anderson et al., 2011; Couturier et al., 2018; Peel et al., 2019a). Large-scale shifts in oceanographic conditions could be monitored with remote sensing of sea surface temperature and surface chlorophyll-a concentration (Dewar et al., 2008; Jaine et al., 2012). These seasonal shifts in the prevailing winds, precipitation and sea temperature, broadly referred to as the north-west and south-east monsoons (Dewar et al., 2008; present study), coincide with manta ray visitations increasing in the south during the south-east monsoon and in the central region during the north-west monsoon (Dewar et al., 2008; present study). Previously, Dewar et al. (2008) noted that temperature was a predictor for visitations to sites, with the majority of site visits (≥ 83%) occurring between 25 and 28°C, with no visitations above 29°C. Elsewhere, manta rays were more abundant at water temperatures ≥ 21°C in Mozambique (Rohner et al., 2013), between 21 and 24°C in Lady Elliot Island, Eastern Australia (Jaine et al., 2012), at 28°C in the Amirante Islands, Seychelles (Peel et al., 2019a), and within the full temperature range of the Red Sea (21.6 – 34.2°C) (Braun et al., 2014). Thus, thermal thresholds alone are unlikely to be responsible for the observed temperature preference. Similar to the association between productivity and temperature at Lady Elliot Island (Couturier et al., 2018), the optimal 125 temperatures for manta ray visitations observed within the Komodo NP are more likely a reflection of the increased productivity in the region from deep, cold water upwelling waters. Seasonal weather patterns largely govern upwelling in the region (Ningsih et al., 2013), with the times and areas of increased productivity aligning with seasonal trends in manta ray abundance. Fine-scale oceanographic conditions, bathymetry, and tidal currents appear to effect MA and the central sites/CL differently, likely influencing prey availability and thus seasonal variations in site use. The southern site, MA, located in a large bay with a high degree of structure, facing the Indian ocean and adjacent to deep basins, is more directly exposed to the south-east monsoon seasonal upwelling (Ningsih et al., 2013) and the resulting increase in productivity, coinciding with increased manta ray sightings, during that time. Central sites (KM and MW) are exposed to increased nutrients and prey primarily via tidal transport. In line with this phenomenon, Dewar et al. (2008) found that on the spring tides (i.e. during full and new moon), when currents and tidal transport are highest, manta ray visitations to KM and neighboring sites increased. Tidal current and phase had a lesser effect on visitations in MA (Dewar et al., 2008), presumably because nutrient availability in the south is less dependent on tidal transport. Tidal dependent nutrient availability likely allows for manta rays to be present in the central area year-round, albeit at times in lower numbers as has been observed (present study, Dewar et al., 2008). Tidal transport of nutrients into the central area would hypothetically occur year-round, allowing for regular foraging opportunities there. Indeed, in this study we noted that a similar number of feeding events were recorded for KM throughout the year. However, during the north-west monsoon, the at times high numbers of individuals observed to feed simultaneously at KM signify high levels of productivity at this time. Additionally, in Chapter 5, on the north-west section of KM, where foraging behavior CHAPTER 4. KOMODO NATIONAL PARK 126 was observed under specific tidal conditions, we recorded fewer feeding events during the south-east monsoon. Interestingly, ~12 km to the north of KM, the site CL appears to support feeding for some individuals during the south-east monsoon season, when productivity in the north is generally lower. This site is essentially a channel between two islands where tidal current flows, and potentially plankton, is concentrated providing feeding opportunities for what appears to a be mostly a specific demographic of individuals (i.e. immature individuals). Of note, given the duration of the study, the number of feeding events recorded are few. Here it is important to mention that obtaining photographic evidence of sightings was also limited by environmental conditions, such as times when strong currents and/or swell prevailed, reducing the ability of photographers to obtain quality identification images which would then be used in this study (e.g. when manta rays are feeding in strong current flows). Thus what appears to be a low number of feeding events recorded in this study could be a reflection of effort and/or successful identifications. It is becoming evident that it is important to investigate local environmental drivers specific to each manta ray aggregation area as amassing data on these different areas throughout the world indicate that environmental predictors for manta ray abundance vary, and in some cases, show opposing trends. For example, Barr and Abelson (2019) recently documented that manta ray site visits to sea-mount cleaning stations for M. alfredi and M. birostris decreased in periods when light intensity was high (mid-day, sunny conditions, and full moon illumination). Similar sighting trends for M. alfredi were observed by Rohner et al. (2013) along the coast of Mozambique and by Jaine et al., (2012) at Lady Elliot Island, with less individual sightings at cleaning stations when the moon was more than half illuminated. However, time of day was not a significant predictor for Mozambique sightings and cleaning activity was higher at 127 midday at Lady Elliot Island. Within Komodo NP, site visits detected by receivers were highest between 11:00 and 14:00 (Dewar et al., 2019), and in the present study we show that cleaning is the most common observed behavior at KM, MW and MA. Thus, unique local factors likely dictate when conditions are most optimal for foraging and/or cleaning (Barr & Abelson 2019; Jane et al., 2012), necessitating in situ observations of manta rays and environmental conditions for each major aggregation area. However, in general, cleaning activity happens in the day and night time visitations to cleaning stations by manta rays are rare globally (Braun et al., 2014; Couturier et al., 2018; Deakos et al., 2011; Dewar et al., 2008; Peel et al., 2019a; Setyawan et al., 2018; Venables et al., 2020). One of the reasons for this are that cleaner fish are generally not active at night (Couturier et al., 2018; Marshall & Bennett 2010). Secondly, results from biochemical (Couturier et al., 2013; Peel et al., 2019b) and telemetry studies (Braun et al. 2014; Couturier et al., 2018; Setyawan et al., 2018; Stewart et al., 2016) suggest that at night, manta rays are feeding on emergent and benthic zooplankton, and/or travel offshore to feed in the deep scattering layer on mesopelagic zooplankton that emerge from the sediment or arise to shallower depths during the night, respectively (Hays 2003). Reproductive behavior The observed courtship activity during the study occurred at sites where cleaning behavior also occurred. These observations are similar to those from Manta Point – the site used predominantly for cleaning, social activity and courtship in the Nusa Penida MPA (Chapter 3) – and are supported by the idea that cleaning stations are acting as lekking sites i.e. aggregation areas where courtship and mate competition takes place (Deakos et al., 2011; Stevens, 2016). The seasonal peak in manta ray sightings at the central sites of the park coincided with a modest peak in courtship (April/May), similar to the pattern in the Nusa Penida MPA (Chapter 3). This time of year is also when there is a CHAPTER 4. KOMODO NATIONAL PARK 128 shift in the weather between the north-west to the south-east monsoon. Further, another more modest peak in recorded reproductive activity occurs in October, when the reverse monsoonal shift occurs. Taken together these observations suggest that reproductive behavior is governed by monsoonal shifts and probably more directly through the changes in productivity that these shifts bring. However, despite combining courtship sightings from the three core sites, the total courtship records were much lower overall in Komodo NP than in Nusa Penida MPA (~ 3-fold). There were also fewer days in which courtship was recorded. A study limitation, with our use of a largely citizen scientist dataset, is it is not always possible to collect more specialized information. Thus it is not possible to determine the cause of the apparent lower level of courtship activity, i.e. whether it is a reflection of the true level of courtship or due to a lack of observer knowledge and/or lack of photographic evidence (see Figure 3.4, Chapter 3). Further, the possibility that courtship activity might be occurring in locations not popular with divers cannot be excluded (Conservation International Indonesia, 2016). 4.5.2 Local Threats to Manta Rays Unregulated tourism Tourism in the Komodo NP has greatly increased since the Dewar et al. (2008) study, which aimed to increase the understanding of spatial distribution and habitat use of manta rays residing in the Komodo NP. While this initial study sought to help develop a viable manta ray tourism industry there, it is now clear that up to date information is needed to guide management to better regulate growing tourism, preventing undue pressure on the manta rays in this critical habitat (e.g. Chapter 3; Kasmidi & Gunadharma, 2017; Venables et al., 2016; Division of Boating and Ocean Recreation, 2016). Tourist boats commonly visit manta ray sites within the central area carrying 129 anywhere between 10 and 35 divers/snorkelers per day (Komodo Diver Operators Community, pers. comm.). Central area sites receive the bulk of the tourism visits, as these sites are accessible by both day trip and ‘liveaboard’ operations year-round, and these visits have increased by 34% percent in the last five years. Thus, manta rays with greater site affinity to the central area are particularly at risk of chronic disturbance. Because of the high site affinity to KM, the site used by most of the identified manta rays in Komodo NP, KM is considered a critical habitat and thus requires management considerations. To this aim, as of September 2019, the Komodo NP authority has introduced limits on the number of boats and people allowed in the water at KM each day to 32 boats carrying a maximum of 10 divers (Komodo National Park, 2019). At the time of writing, other manta ray aggregation sites remain without restrictions. The reef that manta rays use to clean at MW is particularly small and therefore has a lower tourism carrying capacity than KM or MA. At times, this area is already overcrowded by divers, particularly during peak tourist season (July and August; E Germanov, pers. obs), which might prevent/substantially reduce the length of cleaning station visits by manta rays similar to the disturbance of feeding rays by swimmers elsewhere (Garrud, 2016; Venables et al., 2016). As there are now management restrictions on the daily number of boats and divers at KM, the main manta tourism site in Komodo NP, it is foreseeable that some of the tourism pressure might be displaced to MW, which currently does not have capacity restrictions. As cleaning at these reefs provides manta rays with an important health service (i.e. removing parasites and facilitating wound healing), reduced visitations could potentially affect individual fitness (O’Shea et al., 2010). Further, mounting evidence suggests that cleaning stations serve as meeting places for important behavioral interactions, such as courtship, thus reduced visitations might have population-wide implications (Perryman et al., 2019; Stevens, 2016; Stewart et al., 2018). In addition, CHAPTER 4. KOMODO NATIONAL PARK 130 more divers visiting these reefs increases the risk of habitat destruction by poor diving techniques (Trave et al., 2017). Increased boat traffic in manta ray feeding aggregation areas likely also increase the risk of boat strikes and propeller injuries (McGregor et al., 2019). In the north of Komodo NP, tourism at CL appears also to be on the rise. This site is generally accessible year-round by daily trip and ‘liveaboard’ boats, and continued use of this site by as many as 10 or more boats at time (100 to 350 divers/snorkelers), without strict adherence to best practice codes of conduct, will likely have a substantial impact on foraging manta rays (Garrud, 2016; Venables et al., 2016). The narrow channel at CL is also used as a passage for boats and this general boat traffic further increases the pressure on manta rays that use this site. Tourism at MA appears to have remained stable, likely due to the greater distance of this site from the nearest ports of Labuan Bajo or Sape, Sumbawa (~ 60 km, half – full day journey). Larger, ‘liveaboards’ carrying up to 30 divers per boat, typically visit this site for repeated daily dives, yet only stay for 1 – 2 days. Continued monitoring of boat and tourist numbers are necessary to better advise managers. The recommendations for consideration by managers, identified for the Nusa Penida MPA in Chapter 3 also apply to the Komodo NP. In particular, carrying capacity estimates for tourism boats and people in the water for all popular manta ray sites, restrictions on boat speed and minimum distances of approach to foraging manta rays and known cleaning stations are strongly recommended to minimize disturbance to the manta rays and to ensure the safety of divers/snorkelers. General best practice codes of conduct developed specifically for SCUBA diving and snorkeling with manta rays are available online (Manta Trust, 2018; Marine Megafauna Foundation Indonesia, 2018), as well as an example of detailed guidelines for activities at a manta ray cleaning station in Raja 131 Ampat, Indonesia (Kasmidi & Gunadharma, 2017). Consideration should be given as to how these codes of best practice could be implemented effectively in the Komodo NP. Fisheries and natural predators Fishing activities continue to threaten manta ray populations in Indonesia to some extent (Chapter 3), even if only by sub-lethal injuries (Stewart et al., 2018). A better understanding of site use and limiting fisheries to specific gear types in high use areas and along movement corridors would enhance conservation measures (Graham et al., 2016). While all the current study sites in the Komodo NP are within no-take fishing zones, in practice these regulations are not strictly enforced and non-compliance is common (E Germanov, pers. obs.; Dive Operators Community of Komodo, pers. comm.). However, fewer individuals were recorded with fishing gear injuries within the Komodo NP than in the neighboring Nusa Penda MPA (5% vs. 14%; Chapter 3). Further, these observations indicate a higher anthropogenic injury rate than reported for similar injuries (truncated or cut cephalic and pectoral fins) in the Ningaloo Coast World Heritage Area (~ 1%; McGregor et al., 2019), but lower than the rate reported in Hawaii (10%; Deakos et al., 2011). In terms of the impact of injuries to manta rays, similar to observations within the Nusa Penida MPA (Chapter 3), sightings of pregnant manta rays with single cephalic fin truncations suggest that these injuries do not catastrophically impair fitness. However, more information on the reproductive ecology of manta rays in this region is needed to better understand these impacts. Approximately half as many predatory bite injuries were observed on manta rays compared to those caused by contact with fishing gear. The potential predators in this region include grey reef sharks Carcharhinus amblyrhynchos, silvertip sharks C. albimarginatus, Java sharks C. amboinensis, bull sharks C. leucas, oceanic whitetip C. CHAPTER 4. KOMODO NATIONAL PARK 132 longimanus, dusky sharks C. obscurus, tiger sharks Galeocerdo cuvier, great hammerhead Sphyrna mokarran, and bluntnose sixgill Hexanchus griseus (White et al., 2006), similar to those proposed for manta rays in Mozambique (Marshall et al., 2010). Of those mentioned above, only grey reef and silvertip sharks are regularly encountered by divers. To our knowledge, only a single report for a tiger shark sighting exists for the Komodo NP (Dive Operators Community of Komodo, pers. comm.), an indicator that the majority of manta rays in this region occupy areas with a low abundance of their potential predators as compared to some other locations across their range (Marshall & Bennet, 2010). It should be noted, however, that anthropogenic and predator injuries observe and count only those individuals that have survived the encounters and do not provide a full indication of the number of individuals impacted. Further, we are not able to determine whether these lower anthropogenic injury rates were a result of less contact between fishers and manta rays in this broader region compared to Nusa Penida MPA, or a product of fishers using different gear types in the two regions, such as gill nets and long-lines, contact with which are likely to lead to lower manta ray survival rates than entanglements with rudimentary fishing gear types such as hook and line. However, modeled mortality rates were very low (~ 0), similar to those of the Nusa Penida MPA, suggesting that in general, mortality due to fisheries or natural predators is low for the manta rays sighted in the Komodo NP. However, these analyses are limited in that they are not able to discern between true mortality or permanent immigration. The present study highlights that manta rays inhabiting conservation zones within the Lesser Sunda region are sustaining fisheries related injuries. It is thus important to enforce the ban on manta ray fisheries and zoning restrictions for fishing activity within conservation zones. In addition, focusing on understanding and reducing the impacts of tourism should now be a priority as well. 133 4.5.3 Limitations and Future Directions While we are able to look at the residency and movement patterns of hundreds of individuals, our sighting-based approach is limited temporally and spatially to when and where divers are present in the water (i.e. by observer/photographer effort). While there were few or no manta ray sightings recorded from December – March for MA, this may be an artifact of low sampling effort, as most liveaboard do not operate in and around MA at that time. However, via a sightings-independent acoustic telemetry approach, Dewar et al., 2008 also reported fewer manta visitations to MA during December – March. Further, the increase in sightings per dive (i.e. observer logs) in KM during 2015 and 2016, might be a reflection of the success of the fewer, yet dedicated, research trips during that period. These research trips were generally planned for what were considered the best conditions for increased opportunities of manta ray encounters (i.e. moon phases and tides; Dewar et al., 2008, E. Germanov, pers. obs.). Trained observer surveys were not available year-round in 2015 – 2016; however, when citizen science data were included, no substantial changes in the overall mean daily sightings were noted. In terms of spatial effort, the few sighting records for German Flag, and none for Padar Kecil, which the previous study (Dewar et al., 2018) identified as important sites, are likely a result of insufficient sampling effort as these sites are rarely visited by dive operators in recent years. For future research, an intensive sampling approach with structured surveys is necessary to provide the data needed to enable population abundance and survival estimation using mark-recapture methods (i.e. Couturier et al., 2014). Within the Komodo NP, the uneven interchange of individuals between north and south sites requires us to carefully consider how to best approach estimating populations for these areas so that model assumptions, such as all individuals having the same probability of re-capture, will be met. In all likelihood several different approaches will need to be presented and CHAPTER 4. KOMODO NATIONAL PARK 134 compared (Couturier et al., 2014; Kitchen-Wheeler et al., 2001), indicating that these analyses deserve focused treatment in future manta ray research. Gauging the impact of tourism on manta ray use of cleaning stations has yet to be empirically assessed. Underwater video cameras and aerial drones might be effective in monitoring various cleaning stations (Barr & Albelson, 2019; Stewart et al., 2018) to view manta ray cleaning behavior when people are and are not present (i.e. serving as a control) in the water. The study could compare aversion behavior (Garrud, 2016; Venables et al., 2016) and length of cleaning under different scenarios: 1) no people, 2) people adhering to codes of conduct, and 3) people not adhering to these codes. Differences in manta ray behavior and visitations lengths to cleaning stations could be evaluated relative to the number of people at the cleaning station at one time. Comparing observations on visitation patterns with longitudinal population abundance estimates would provide important insight into understanding the potential aversive effects of tourism on manta rays (reviewed in Stewart et al., 2018a). The results of these studies would serve to guide management in carrying capacity evaluations. To increase the current understanding of site use, movements, population demographics and dynamics within the Komodo NP, we also recommend expanding the survey area to investigate other locations where satellite tagged manta rays spent time (e.g. East Rinca and West Komodo; Conservation International Indonesia, 2016). Studies at these sites are more likely to require a targeted research sampling program as they are less accessible/less visited by tourism operators, and hence citizen science programs. We further recommend a renewed focus on surveys at German Flag to confirm previous observations of high site use and by small individuals (Dewar et al., 2008). Additional tracking studies would provide more information on movement corridors throughout the region, connectivity between conservation zones and highlight if there are any seasonality 135 trends in long-range movements. Further, better quality data on the size of manta rays will allow for the maturity of females to be assessed in more detail and the growth of immature individuals to be monitored. With additional, and detailed data on pregnancies, when and where they occur, we could explore whether there are sites more important for pregnant individuals and whether there is a seasonality and/or annual variability in pregnancy rates. Further, pinpointing whether there is a specific time of year when the appearance of new immature individuals is more likely could be indicative of a general pupping time and potential location of pupping areas. These data would allow for a dedicated follow-up study on the reproductive ecology of manta rays within the Komodo NP and provide vital information for conservation planning, particularly the establishment of closed areas for breeding females or foraging juveniles. In terms of better understanding the influence of environmental conditions on site use in this complex environment, calculating in situ prey density thresholds (e.g. Armstrong et al., 2016) and correlating this information with seasonal trends in manta ray abundance and environmental drivers (Anderson et al., 2010; Barr & Abelson, 2019; Dewar et al., 2008; Jaine et al., 2012) will be important. Further, inter-annual variability in manta ray residency and movements (Dewar et al., 2008) is probably influenced by variations in the prevailing environmental conditions (e.g. Indian Ocean Dipole and El Niño southern oscillation; Ningsih et al., 2013) that affect prey availability and prey dispersion. Aerial/drone surveys of mass feeding events would provide additional information on the number of individuals aggregating at KM (Weeks et al., 2015). Further, continuing the photo identification data collection for the Komodo NP would be valuable in monitoring these trends over a longer-term, particularly with projected CHAPTER 4. KOMODO NATIONAL PARK 136 climate-change induced global declines in zooplankton biomass (reviewed in Richardson, 2008; Stewart et al., 2018). Finally, genetic studies that include populations from the Komodo NP, Nusa Penida MPA and other known aggregation sites around Indonesia, such as Raja Ampat and eastern Kalimantan, and beyond, such as the Philippines and Australia, would be beneficial for understanding regional connectivity and the historical movements of manta rays in the region. Close-kin genetic population analyses can supplement traditional mark-recapture population analyses to effectively estimate manta ray populations throughout Indonesia (Dudgeon et al., 2012; Schwartz et al., 2007). 4.6 CONCLUSION Komodo National Park, is an important habitat for a large number of manta rays with several documented aggregation sites, including reproductive grounds and foraging grounds for immature individuals and adults. There is also sustained connectivity between the Komodo NP and the Nusa Penida MPA, another conservation zone and manta ray aggregation area. With the national ban on manta ray fisheries, the main threat of directed fisheries should now be much alleviated; however, undirected fisheries and by-catch in movement corridors will continue to put pressure on this threatened species. Further, increasing use of manta ray aggregations for tourism now requires that science-based management strategies are developed, implemented, monitored and adapted to maintain these important habitats from undue anthropogenic disturbance. These strategies need to include regulations when it comes managing tourism, fishing within important areas and corridors and other potential threats. An outreach and education component should be imbedded in tourism management so that conservation strategies are supported by dive operators and local communities. In the absence of empirical data, the 137 precautionary approach is currently the gold standard until such data are available (UNESCO 2005; Fennel & Ebert 2004; Venables et al., 2016; Garrud 2016). To that aim, when data and/or funding is scarce, citizen science can successfully provide important long-term data that can guide directions for conservation management and more targeted rigorous research. Thus, adding citizen science to the conservation toolset should be considered when managing threatened marine megafauna and protecting the livelihoods of people that depend on associated tourism well into the future. 138 CHAPTER 5. MICROPLASTICS ON THE MENU: PLASTICS POLLUTE INDONESIAN MANTA RAY AND WHALE SHARK FEEDING GROUNDS A manta ray feeds amidst plastic debris in Manta Bay, Nusa Penida MPA © Lillian Ellevog 139 CHAPTER 5. MICROPLASTICS ON THE MENU 140 5.1 ABSTRACT The implications of plastic pollution, including microplastics, on marine ecosystems and species are increasingly seen as an environmental disaster. Yet few reports focus on filter-feeding megafauna in regions heavily impacted by plastic pollution, such as Indonesia in the Coral Triangle, a global marine biodiversity hotspot. Here, we evaluate plastic abundance and characterize debris from feeding grounds for manta rays Mobula alfredi and whale sharks Rhincodon typus in three coastal locations in Indonesia: Nusa Penida Marine Protected Area, Komodo National Park, and Pantai Bentar, East Java. A 200 µm plankton net was used to sample the top 0.5 m of the water column (‘trawl survey’) and floating plastics were assessed along ~ 440 m long transects (‘visual survey’) during the Indonesian north-west (wet) and south-east (dry) monsoon seasons during 2016 – 2018. Microplastics were identified visually, measured and categorized from trawl samples, and larger floating plastics were counted and categorized visually from boats. Plastic abundance ranged widely from 0.04 to 0.90 pieces m3 (trawl survey) and 210 to 40,844 pieces km2 (visual survey). Results from linear models showed significant seasonal and location differences in estimated plastic abundance for trawl and visual surveys in Nusa Penida and Komodo. Plastic abundance was up to ~ 44 times higher in the wet than the dry season, with the largest seasonal effect observed in Nusa Penida. Overall, small pieces < 5 mm (78%), films and fragments (> 50% combined) were the most prevalent plastics. Theoretical plastic ingestion rates were calculated using estimated filtration volumes of manta rays and whale sharks and the mean plastic abundance in their feeding grounds. Upper plastic ingestion estimates for manta rays were ~ 63 and 25 pieces h-1 for Nusa Penida and Komodo locations, respectively, and ~137 pieces h-1 for whale sharks in Java. Analysis of manta ray egested material confirmed plastic ingestion, the consequences of which might include exposure to toxic plastic 141 additives and adhered persistent organic pollutants. Communicating this information to communities who stand to benefit from healthy megafauna populations might help local governments as they work toward reducing plastics in the marine environment. 5.2 INTRODUCTION The rising level of plastic debris in our oceans is a large- scale environmental problem with wide ranging impacts (Chapter 2; van Sebille et al., 2015; Worm et al., 2017). Perhaps most significantly, this form of pollution wreaks havoc on marine life (Gall & Thompson, 2015; Wilcox et al., 2015, 2018; Worm et al., 2017). While large debris also impacts marine organisms (e.g., turtles, birds, sharks, and mammals), once broken down to microscopic sizes through environmental exposure, microplastics are of growing concern as they easily enter food webs (Andrady, 2011; Worm et al., 2017). Microplastics, generally referred to as plastics < 5 mm in diameter (Andrady, 2011), are comparable in size to, or smaller than zooplankton, an integral component in marine ecosystems and the primary food for planktivores. These small plastic particles are ingested by a range of organisms, including zooplankton themselves (Cole et al., 2013; Setälä et al., 2014), small planktivorous fishes (Boerger et al., 2010) and filter-feeding megafauna (Besseling et al., 2015). In addition to the more obvious issues associated with the ingestion of foreign, potentially indigestible objects – such as digestive tract obstruction and perforation, dietary dilution and starvation – plastics, are significant carriers of toxic additives such as bisphenol A, phthalates, and flame retardants (Chapter 2; Andrady, 2011; Worm et al., 2017). Microplastic ingestion is proposed to impact gut biomes of host organisms, potentially leading to gut dysbiosis (reviewed by Fackelmann & Sommer, 2019). Further, the hydrophobic nature of some plastics and the increased surface area of microplastics, CHAPTER 5. MICROPLASTICS ON THE MENU 142 through pitting processes, create an ideal sink for persistent organic pollutants (POPs), such as dichloro-diphenyl-trichloroethanes (DDTs), polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs) (Rios et al., 2007, 2010; Ogata et al., 2009; Cole et al., 2011) and heavy metals (Ashton et al., 2010). Pollutants are concentrated up to a million-fold in microplastics compared to surrounding seawater (Mato et al., 2001). Upon plastic ingestion, these toxic chemicals can leach into organisms (Teuten et al., 2009) and may negatively affect the physiology (Browne et al., 2013; Rochman et al., 2013), development (Anselmo et al., 2011) and endocrine functions (Rochman et al., 2014) of marine organisms (reviewed in Browne et al., 2015; Worm et al., 2017), even if the plastic pieces eventually pass through the digestive system. Long-lived species may bioaccumulate these pollutants over decades to concentrations that lead to altered reproductive fitness (Reddy et al., 2001), which may further impact their young through maternal offloading (e.g. Genov et al., 2019; Lyons et al., 2013a,b, 2014) leading to population level impacts (Worm et al., 2017). Owing to their feeding strategies, filter-feeding megafauna are particularly susceptible to microplastic ingestion (Chapter 2), while their longevity (Stevens, 2016; Stewart et al., 2018) increases the risks of pollutant bioaccumulation. In a number of studies on filter-feeding megafauna, Fossi et al. (2014, 2016, 2017) documented trace levels of phthalates and POPs in the muscle of basking sharks Cetorhinus maximus, the blubber of fin whales Balaenoptera physalus and the skin of whale sharks Rhincodon typus. For phthalates, a plastic additive and chemical tracer, and probably with POPs, these chemicals are likely to have been acquired through plastic ingestion. Filter-feeding reef and giant manta rays (Mobula alfredi, and M. birostris, respectively) and whale sharks are likely to feed in a variety of habitats, including deep and pelagic environments (Couturier et al., 2013), and they are commonly observed 143 feeding near the surface in shallow coastal waters (Chapter 2; Armstrong et al., 2016; Jaine et al., 2012; Rohner et al., 2015). Some types of plastic polymers are buoyant in seawater and accumulate in the surface layer of the water column (Andrady, 2011). Additionally, microplastics might be acquired indirectly through ingestion of plankton that have ingested microplastics (Cole et al., 2013; Setälä et al., 2014). Habitat ranges for many filter-feeding megafauna also overlap with areas of heavy plastic pollution, such as the Coral Triangle (reviewed in Chapter 2; Eriksen et al., 2014; Maximenko et al., 2012; van Sebille et al., 2015). Indonesia is situated within the Coral Triangle, a significant region for global marine biodiversity (Gray, 1997; Worm & Branch, 2012). It is home to many endemic and threatened marine species and boasts globally significant populations of manta rays (Chapter 2; Germanov & Marshall, 2014) and whale sharks (Norman et al., 2017). Currently, Indonesia is also ranked as the world’s second largest plastic marine debris emitter (Jambeck et al., 2015). In a sample of fishes sold for human consumption from the Makassar fish market, Indonesia, over a quarter of the fishes, including over half of the species sampled, contained plastic debris in their digestive tracts (Rochman et al., 2015). The impacts of plastic pollution on megafauna in Indonesia and neighboring countries are also now being documented. Three recent examples of deceased marine mammals that have ingested substantial amounts of plastic have been documented and have prompted concern for other species that live and feed in these polluted waters: a sperm whale Physeter microcephalus in Indonesia (Victor, 2018); a pilot whale Globicephala macrorhynchus in southern Thailand (Zachos, 2018); and a juvenile Cuvier’s beaked whale Ziphius cavirostris in the Philippines (British Broadcasting Corporation [BBC], 2019). In Australia, sea turtles, well-established to be at risk of plastic ingestion, were recently evaluated to have a 50% risk of mortality upon ingesting CHAPTER 5. MICROPLASTICS ON THE MENU 144 as little as 14 pieces of plastic (Wilcox et al., 2018). In Indonesia, among the megafauna most likely impacted by plastic pollution are manta rays and whale sharks (Chapter 2). They are globally threatened (Marshall et al., 2018a, b; Pierce & Norman, 2016) with the main threats coming from direct and indirect fisheries. Protecting these species can be an economic benefit, as they are important drawcards for marine tourism in Indonesia (Gallagher & Hammerschlag, 2011; O’Malley et al., 2013). In Indonesia, despite directed fisheries being formally outlawed throughout the entire Exclusive Economic Zone in 2014 (Ministry of Marine Affairs and Fisheries, 2014), these species are still exposed to other anthropogenic threats including boat strikes, and habitat destruction and disturbance, and the rising levels of plastic pollution in coastal waters, particularly at feeding habitats. In these plastic pollution hotspots, filter-feeders would be at an elevated risk of ingesting microplastics and tainted zooplankton, and would be exposed to plastic associated pollutants (Chapter 2). To gain insight into the level of threat plastic debris poses to Indonesia’s manta rays and whale sharks, it is important to quantify the abundance of plastic debris in feeding grounds. While microplastic levels have been modeled globally (Eriksen et al., 2014; Maximenko et al., 2012; van Sebille et al., 2015), abundance estimates for microplastic pollution within the Coral Triangle and Indonesia vary considerably between models (van Sebille et al., 2015). Further, with a few exceptions (Manalu et al., 2017; Syakti et al., 2017), empirical data on microplastics are lacking for coastal areas in Indonesia. Plastic abundance and theoretical ingestion rates have been calculated for whale sharks feeding in La Paz, Mexico (Fossi et al., 2017). However, to our knowledge, no plastic abundance estimates for manta ray and whale shark feeding grounds currently exist for Indonesia or the greater Coral Triangle. Furthermore, while there is empirical evidence that whale sharks ingest plastic (Abreo et al., 2019; Donati et al., 2019; 145 Haetrakul et al., 2009; Sampaio et al., 2018;), confirmation that manta rays ingest plastics is yet unavailable. Plastic ingestion by manta rays might be confirmed from opportunistic stomach contents analysis of stranded individuals (e.g. Abreo et al., 2019; Besseling et al., 2015; Haetrakul et al., 2009; Sampaio et al., 2018) and could potentially also be assessed through non-invasive analysis of egested material (Donati et al., 2019). Our primary aim of the study was to quantify the surface plastic abundance in coastal feeding grounds for manta rays and whale sharks in Indonesia, within the Coral Triangle region, and establish if reef manta rays, hereafter ‘manta rays’ ingest plastics. The study region has a high but spatially variable human population density, is oceanographically complex, and has strong seasonal weather patterns that influence regional river pollution emissions (Lebreton et al., 2017). To that end, we examined whether geographic and seasonal variations in plastic debris abundance were evident. Based on these plastic abundance estimates, we have provided theoretical plastic ingestion rates for manta rays and whale sharks in the region. Egested material from manta rays was also analyzed for plastic content. By highlighting the implications of plastic pollution to manta rays and whale sharks, we are effectively employing a flagship approach to conservation (Bowen-Jones & Entwistle, 2002; Barua, 2011; Verissimo et al., 2011) in the hopes of creating flow-on effects from these iconic species to positively enhance the stewardship of Indonesia’s marine ecosystems. 5.3 MATERIALS AND METHODS 5.3.1 Study Locations and Oceanographic Conditions Three separate locations within Java and the Lesser Sunda Island chain of Indonesia that are frequented by feeding manta rays and whale sharks (see below) were chosen as study sites (Figures 5.1, 5.2, created using QGIS Development Team, 2016). CHAPTER 5. MICROPLASTICS ON THE MENU 146 Manta Bay (MB), within the Nusa Penida Marine Protected Area (MPA) located to the south east of Bali Island (Figure 5.2A); and Karang Makassar (KM), in the central region of the Komodo National Park (NP), Flores (Figure 5.2B) are year-round manta ray feeding grounds (Chapter 3; Dewar et al., 2008; Germanov & Marshall, 2014). Pantai Bentar (PB), southeast of Probolinggo, East Java (Figure 5.2C) is a seasonal (December – March) whale shark feeding ground (Kamal et al., 2016). The sites consist of: a shallow (5 – 20 m deep), steep cliff-lined bay with mixed sand, rock and reef substrate in the north-west section of Nusa Penida Island, referred to here as North and South MB (N-MB and S-MB); a shallow rubble and reef slope (6 – 20 m deep) east of Komodo Island (KM); and a large, shallow (0 – 20 m) mangrove fringed bay affected by river runoff and heavy sedimentation (E. Germanov, pers. obs.) near PB beach. Nusa Penida MPA and Komodo NP, located along an island chain, are situated on a shallow (< 200 m) continental shelf with deep water basins (> 1,000 m) > 20 km offshore. The Madura Strait, East Java, including PB, is entirely < 100 m in depth (Figure 5.1). Oceanography in the region The region is oceanographically complex. Strong seasonal conditions influence oceanographic conditions and rainfall, which varies between the study locations (Figure 5.3). Thus, we anticipated that there would be regional and seasonal differences in plastic abundance between the study sites and tested for these variations. Generally, the region experiences substantial influence from the Indonesian Throughflow (ITF) (Tillinger, 2011) via the Makassar Current, estimated to transport up to 50% of the ITF volume (Mayer & Damm, 2012). The majority of the Makassar Current flows southward through the Makassar Strait and passes through the Lombok Strait between the Islands of Lombok and Bali and the eastern side of Nusa Penida as a surface current (Mayer & Damm, 2012). A portion of the Makassar current also enters the Flores Sea, north of Komodo NP, 147 as a subsurface current (Mayer & Damm, 2012). East of Bali, the ITF pathway known as the Karimata Strait, located between the islands of Kalimantan and Sumatra and Java, connects the South China Sea to the Java Sea. This pathway serves to bring less saline water eastward toward the Flores Sea and into the Makassar Strait during the north-west monsoon season (Mayer & Damm, 2012). The prevailing winds of the north-west and south-east monsoons generally act to decelerate the Makassar Current flow (Mayer & Damm, 2012), leading to seasonal differences in intensity, with it being greatest during the south-east monsoon (Sprintall et al., 2009). Further oceanographic influences on the region come from the Indian Ocean (Hendrawan & Asai, 2011; Ningsih et al., 2013) and tidal currents that move north to south and vice versa through the straits surrounding the Komodo NP (Mehta-Erdmann, 2004). The regional oceanic productivity, that sustains the feeding grounds for megafauna (e.g., Dewar et al., 2008; Mayer & Damm, 2012; Thys et al., 2016), is owed to the warmer (Tillinger, 2011) and nutrient rich Pacific waters (Ayers et al., 2014) carried by the ITF mixing with Indian Ocean waters. The area is further enriched by deep water upwelling to the south of the island chain, especially during the south-east monsoon (Ningsih et al., 2013). The two alternating monsoon seasons also vary the amount of precipitation that falls on the region, which is higher during the north-west monsoon (November – April), hereafter ‘wet’ season, compared to the south-east monsoon (May – October), hereafter ‘dry’ season (Chang et al., 2005). CHAPTER 5. MICROPLASTICS ON THE MENU 148 Figure 5.1 Regional oceanographic details, river emission points and population density adjacent to study areas. Bathymetry information was available from GEBCO_2014 Grid, version 20150318; www.gebco.net. River emission points available from Lebreton and Reisser (2018). Population information in15 min × 15 min grids (~ 760 km2) using 2025 population projections available from Yetman et al., (2004). 149 Figure 5.2 Map of study locations at (A) North and South Manta Bay (N-MB and S-MB, respectively), Nusa Penida Marine Protected Area; (B) Karang Makassar (KM), Komodo National Park; and (C) Pantai Bentar (PB), East Java in Indonesia. Different color lines in (A – C) represent individual survey GPS tracks. CHAPTER 5. MICROPLASTICS ON THE MENU 150 Figure 5.3 Total monthly rainfall for study regions (A) Bali, (B) Komodo Island, and (C) East Java during study years. 0 = no precipitation recorded; nd = no data available. Precipitation data was available from the Indonesian Meteorology, Climatology and Geophysics Agency Data online, Pusat Database BMKG; http://dataonline.bmkg.go.id/. 151 5.3.2 Surveys and Sample Analysis Feeding behavior observations Feeding behavior was assessed by observing manta rays and whale sharks both from a boat and directly from underwater. Manta rays display different feeding behaviors (Stevens, 2016) and we observed surface feeding in our study (for a description of manta ray behaviors see Chapter 3). This behavior is characterized by manta rays swimming near the surface (~ 0.5 m), either in a straight line and periodically completing tilting ‘U-turns,’ or infrequently, ‘barrel-rolling’ by completing full vertical rotations (Chapter 3; Jaine et al., 2012; Stevens, 2016). Feeding manta rays had open mouths and their cephalic fins unrolled in a funnel shape. Whale sharks have several different feeding modes (Nelson & Eckert, 2007), and those at PB were actively surface feeding. This feeding mode is characterized by whale sharks swimming fast at the surface, often with the upper jaw, first dorsal and upper caudal lobe out of the water, and actively intaking water. Trawl surveys Plastic sampling took place between January 2016 and February 2018. Samples were collected during three wet seasons (2016, 2017, 2018) and two dry seasons (2016, 2017) from N-MB and S-MB; and during two of each season from KM (2016, 2017) (Table 5.1). Swimmer towed sampling occurred only when manta rays or whale sharks were feeding. Samples were collected by towing a small round net (0.2 m diameter mouth × 0.60 m length, 200 µm mesh size) adjacent to feeding animals (min. distance of 3 m) by a swimmer (Figure 5.4) for 5 to 10 min (depending on conditions and animal behavior). The results were normalized to the tow times (i.e., results are presented in pieces min-1). CHAPTER 5. MICROPLASTICS ON THE MENU 152 Table 5.1 Research locations, sampling seasons and times, and the numbers of (A) feeding, (B) sub-surface, (C) surface, (D) sampling times in Indonesia from January, 2016 to February, 2018 and (E) egested material samples collected from 2016 – 2018. Locations N-MB days/samples S-MB days/samples KM days/samples PB days/samples Year Season A) Feeding (swimmer trawl) 2016 Wet1 1/1 2/2 2/2 - Dry2 2/2 3/3 1/1 - 2017 Wet1 2/3 3/3 6/12 6/12 Dry2 2/2 3/3 1/1 - 2018 Wet1 3/6 3/5 - - Dry2 - - - - Total Wet 6/10 8/10 8/14 6/12 Total Dry 4/4 6/6 2/2 - Grand Total 10/14 14/16 10/16 6/12 B) Sub-surface (boat trawl) 2016 Wet1 6/12 6/12 6/12 - Dry2 6/12 6/12 6/12 - 2017 Wet1 6/12 6/12 7/14 6/12 Dry2 6/12 6/12 7/14 - 2018 Wet1 6/12 6/12 - - Dry2 - - - - Total Wet 18/36 18/36 13/26 6/12 Total Dry 12/24 12/24 13/26 - Grand Total 30/60 30/60 26/52 6/12 C) Surface (visual survey) 2016 Wet1 6/18 6/18 6/12 - Dry2 6/12 6/12 6/12 - 2017 Wet1 6/12 6/12 7/14 6/12 Dry2 6/12 6/12 7/14 - 2018 Wet1 6/12 6/12 - - Dry2 - - - - Total Wet 18/42 18/42 13/ 26 6/12 Total Dry 12/24 12/24 13/ 26 - Grand Total 30/66 30/66 26/ 52 6/12 D) Sampling times 1 January – March January – March April – May February 2 August – October August – October August – October n/a E) Egested material Type Location Samples Fecal Nusa Penida MPA 21 Vomit Nusa Penida MPA 1 N-MB = North Manta Bay (-8.735, 115.451), S-MB = South Manta Bay (-8.742, 115.451), Nusa Penida MPA; KM = Karang Makassar (-8.535, 119.593), Komodo NP; PB = Pantai Bentar (-7.774, 113.281), East Java; “-” indicates no samples were available for this time 153 Figure 5.4 Samples collected by trawling the near surface water adjacent to feeding animals. Boat towed trawl sampling occurred in two consecutive net tows (replicates) per day taken on six or seven individual sampling days per season for each year at MB and KM (Table 5.1). Sampling at PB took place during six sampling days in 2017 only during the wet season as, generally, whale sharks are absent during the dry season (Kamal et al., 2016) (Table 5.1). Samples were collected by towing a round net (0.5 m diameter mouth × 1.5 m length, 200 µm mesh size), with a mechanical flow meter (Sea Gear) affixed in the center of the net mouth. The results were normalized to the filtered volume (i.e., results were presented in pieces m-3 ). The net was rigged to sample from the sub-surface layers of the water column (up to 0.5 m depth), representative of where local manta rays and whale sharks surface feed (E. Germanov and M. Himawan, pers. obs.). The net was towed for up to 10 min (mean ± 1 SE = 8.1 ± 0.2) between two standard start and stop CHAPTER 5. MICROPLASTICS ON THE MENU 154 positions at a mean speed of 3.28 ± 0.1 km h-1, covering a mean distance of 438.6 ± 15.8 m. Overall the mean sampled volume was 35.8 ± 1.4 m-3. Care was taken to ensure sampling took place outside the wake of the boat to limit surface vertical mixing. Furthermore, to minimize any potential disturbance to animals, the boat was kept 15 m from any animal present. To limit potential tidal influences caused by tidally forced internal waves affecting the transport of surface slicks and maximize sampling opportunities while animals were feeding (E. Germanov, pers. obs.), we aimed to complete sampling and surveys at MB and PB within ~ 2 h of high tide, and within 1 h prior to low tide at KM. To limit wind-mixing variability (Collignon et al., 2012; Reisser et al., 2013), surveys were completed under Beaufort wind scale conditions of ≤ 4. Our trawl methodology is appropriate for manta ray and whale shark feeding ecology studies (Couturier et al., 2013; Rohner et al., 2015), but differs from the standard microplastic methodology using neuston or ‘manta’ nets (Eriksen et al., 2014; Reisser et al., 2013 and reviewed by Hidalgo-Ruz et al., 2012). Neuston or ‘manta’ nets, rigged to sample from the surface- water interface will sample proportionally sample more from the surface than a similarly rigged net with a round opening, as the rectangular mouth opening of the net would sample equally from the surface-water interface as below the surface (see Figure 5.5A). Our plastic abundance estimates are based on samples collected mostly from sub-surface waters, and as a result, might not be directly comparable to those obtained with standard microplastic methodology. However, we converted our results to be presented in comparable units (km-2) to those of modeled estimates (as per van Sebille et al., 2015) by multiplying by the submerged height of the net (0.5 m) to obtain the data in m-2 and then converted to km-2 (see Figure 5.5B). Results are expressed two ways (Figure 5.6 and Table 5.2): (1) the number of plastic pieces m-3 and (2) the number of pieces converted to plastic km-2. 155 Figure 5.5 (A) Sampling from the surface-water interface using nets with round or rectangular openings. (B) Example calculations for converting plastic per volume to plastic per surface area and unit conversion. Sample analysis After each tow, the net was washed with filtered sea water to collect all organic and inorganic material stuck on the side of the mesh. Any large debris was removed manually and placed in a separate container. Large natural debris, such as leaves and sticks, were carefully rinsed before being discarded. The resulting sample was decanted from the collecting jar, reduced to a final volume of 50 or 100 ml (depending on the density of plankton) and preserved in a 5% formaldehyde-seawater buffered solution. Samples were stored in the 5% formaldehyde solution or were transferred to 70% ethanol for storage prior to analysis. CHAPTER 5. MICROPLASTICS ON THE MENU 156 Table 5.2 Floating plastic debris abundance estimates for each sampling location and season and pooled across years. Trawl Survey Visual Survey Location Season Plastic m-3 Mean ± SE Plastic m-3 Median Plastic km-2 Mean ± SE Plastic km-2 Median Plastic (g) km-2 Mean ± SE Plankton (g) km-2 Mean ± SE Plastic: Plankton (g) Plastic km-2 Mean ± SE Plastic km-2 Median Nusa Penida N-MB Wet 0.56 ± 0.12 0.35 275,210 ± 57,504 175,000 399 ± 130 7,500 ± 2,438 0.11:1 17,520 ± 7,093 6,565 S-MB Wet 0.90 ± 0.23 0.41 449,381 ± 115,545 202,500 22,190 ± 17,997 13,333 ± 3,259 0.98:1 40,844 ± 19,909 11,606 N-MB Dry 0.06 ± 0.01 0.05 28,261 ± 6,270 22,500 7 ± 5 10,000 ± 3,989 0:1 608 ± 190 401 S-MB Dry 0.04 ± 0.02 0.02 20,200 ± 8,243 10,000 4 ± 4 7,917 ± 2,083 0:1 649 ± 254 228 Komodo KM Wet 0.29 ± 0.07 0.21 144,073 ± 36,383 105,000 4 ± 3 23,846 ± 6,102 0:1 2,004 ± 647 851 KM Dry 0.24 ± 0.10 0.05 119,275 ± 48,994 25,000 2 ± 2 26,923 ± 5,446 0:1 559 ± 182 262 East Java PB Wet 0.42 ± 0.15 0.39 207,806 ± 74,217 192,500 0 ± 0 35,833 ± 7,683 0:1 210 ± 126 135 N-MB = North Manta Bay, S-MB = South Manta Bay, Nusa Penida MPA; KM = Karang Makassar, Komodo NP; PB = Pantai Bentar, East Java 157 The plastics were separated from the organic material gravimetrically (Collignon et al., 2012) in filtered sea water for 10 – 16 h (overnight) in 1 L graduated cylinders (organic matter settled on the bottom of the cylinders while plastics floated). Visible pieces of plastic were separated from the solution using forceps and the top water layer containing microplastics was siphoned, transferred to filter paper (Whatman No.1) and dried at room temperature. The organic layer was thoroughly checked for any remaining plastics, which were added to the filter paper. Once dry, the filter paper was examined for plastic using a dissecting microscope. Plastics were counted, measured (length = longest side), and sorted by size class (< 1, 1.01 – 5, 5.01 – 200, and > 200 mm; Eriksen et al., 2014), and type (‘film,’ ‘foam,’ ‘fragment,’ ‘line,’ and ‘other’; Rochman et al., 2015). Plastic pieces as small as 0.1 mm were considered with no upper size limit. For this study, we considered all plastics that were collected in our trawls and going forward we use the term ‘microplastics’ to distinguish plastic pieces < 5 mm as is the commonly accepted definition (e.g., Andrady, 2011). A plastic particle was considered to be ‘film’ if it was thin and bent easily; ‘foam’ if it was thicker and compressed under light pressure; ‘fragment’ if it was hard, often brittle under light pressure; ‘line’ if it had an elongated, cylindrical shape, bent easily, but was hard under light pressure; and ‘other’ included paint fragments, which were opaque, thin pieces that fragmented easily under light pressure, and tar, which was black in color and viscous. Fibers (identified as microscopic, thin, same colored particles), apart from fishing line and monofilament rope (thicker, with several intertwining lines), were excluded from the results, as we could not guarantee that samples were free from airborne fiber contamination during sample analysis (Rochman et al., 2015). When it was unsure if a particle was composed of plastic it was touched with a hot needle point, and particles that melted were confirmed as plastic (Roch & Brinker, 2017; Shim et al., 2017; Vandermeersch et al., 2015). Plastic and biological material CHAPTER 5. MICROPLASTICS ON THE MENU 158 were dried (24 h at 60°C) and weighed separately to calculate the weight ratio of plastic to plankton for each sample. 5.3.3 Visual Surveys for Plastic Debris Visual survey transects of floating plastic debris were carried out simultaneously with the net tows. Additionally, during the 2016 wet season in MB a third daily replicate of the visual surveys was recorded (Table 5.1). Dedicated observers stood on the bow of the small motorized boats and noted debris items within 5 m of either side of the boats during timed observation periods, with GPS points taken for the start and end positions of each transect. These positions were used to denote the length of the transect and calculate the area surveyed (Eriksen et al., 2014). The results are expressed in number of plastic pieces km-2. Debris observations were broken down into five categories: plastic ‘films’ (i.e., bags and wrappers), plastic ‘fragments’ (i.e., hard plastic), ‘foam’ (i.e., foamed polystyrene), ‘bottle’ (i.e., beverage bottles and single-use plastic cups) and ‘other’ items such as plastic drinking straws, cigarette butts, sandals and ropes, representing commonly found items and plastic polymer types found in the marine environment (Andrady, 2011). 5.3.4 Manta Ray Egested Material Collection and Analysis Manta ray fecal matter or vomit (i.e., egested material) was collected opportunistically during recreational dives by dive professionals in the local dive community. Once egested material was released into the open water column by the manta ray, any densely packed matter was immediately collected into 50 ml conical polypropylene containers with a screw cap, keeping the material as intact as possible. Overall, a total of 22 samples of egested material (21 fecal and one vomit sample) were collected from within the Nusa Penida MPA. Samples were stored at -20°C until analysis. Prior to analysis, the samples were 159 thawed at 4°C before examining the contents carefully at 40× magnification in a glass Petri dish using a stereomicroscope. Plastic pieces were removed using forceps and placed on filter paper. The remaining organic material was separated gravimetrically and floating plastic was separated as outlined above. Subsequently, the remaining biological material and debris transferred to filter paper were carefully examined and all plastics were counted and classified according to the criteria outlined above. Fibers (apart from fishing line and monofilament rope) were excluded from the final plastic tally for the reasons outlined above. Finally, both organic and plastic material were dried and weighed and the ratio of plastic to organic material was calculated as outlined above. 5.3.5 Calculating Theoretical Plastic Ingestion Rates Based on the estimated abundance of microplastics and the estimated water filtration rates for manta rays and whale sharks of approximately 86.4 m3 and 326 m3 h-1, respectively (Motta et al., 2010; Paig-Tran et al., 2013), we calculated theoretical plastic ingestion rates for manta rays and whale sharks feeding at the different study locations from the number of pieces of plastics per unit volume (i.e. ingestion rate = filtration rate × number of pieces of plastic per unit volume). 5.3.6 Statistical Analyses Daily site replicate samples for visual, boat and hand trawl survey data were pooled, and the resulting daily site means were used in analyses, which were all carried out using the R statistical software (R Development Core Team, 2018). The correlation between the levels of plastic documented by visual and trawl surveys was estimated using Kendall’s tau. Annual, seasonal and location effects were evaluated. Initial analysis highlighted CHAPTER 5. MICROPLASTICS ON THE MENU 160 the importance of incorporating a location × season interaction, as well as the need to log-transform the dependent variable (number of floating plastic pieces assessed by visual and trawl surveys). No plastics were detected in 11% (n = 172) of trawl and 12% (n = 184) of visual samples (Table 5.1). For these cases, we adjusted the 0 values with a constant equal to approximately 1/2 of the lowest non-zero value (30 for visually assessed floating plastic data and 0.005 for trawl survey floating plastic data). The value of the constant was chosen to minimize its influence on the results, while ensuring the greatest compliance of residuals for a homogenous distribution. Back transformations of coefficients from linear models ignored these constants, as a sensitivity analysis found that their impact was negligible. Data from PB were excluded from these analyses as we only collected data in the 2017 wet season, so annual and season effects could not be estimated. Data from PB, and, because of the small sample sizes, all data resulting from swimmer towed trawl surveys were analyzed using descriptive statistics. 5.4 RESULTS 5.4.1 Local Rainfall Remote sensing rain data available from the Indonesian Meteorology, Climatology and Geophysics Agency (BMKG) illustrates seasonality in rainfall and annual and regional variations (Figure 5.3). 5.4.2 Manta Ray and Whale Shark Observations Manta rays were present on 80% of the surveys (n = 60) at north and south Manta Bay (N-MB, S-MB) combined and were feeding 73% of the time with no seasonal variation in presence. Manta rays were present and feeding on 58% of surveys (n = 26) at 161 Karang Makassar (KM); however, most (80%) of the feeding events at KM were during the wet season. Whale sharks were present during every survey (n = 6) at Pantai Bentar (PB) and feeding on 83.3% of surveys. 5.4.3 Trawl Surveys Swimmer towed trawl samples collected whilst manta rays were feeding confirmed that microplastics were present in their immediate feeding areas (Figure 5.6A). A total of 43% of all samples including daily replicates (25 out of 58) contained plastics corresponding to 45% of sample days (17 out of 38 days). The total mean number (± 1 SE) of plastic pieces min-1 of sampling by swimming in Nusa Penida during the wet season was 0.68 ± 0.23 pieces min-1 for N-MB and 0.48 ± 0.47 pieces min-1 for S-MB. Little (0.17 ± 0.17 pieces min-1) to no plastic was found in the dry season samples from S-MB and N-MB, and respectively. The total mean plastics by swimming in KM was 0.11 ± 0.09 pieces min-1 for the wet season, while no plastic was found in the dry season. Swimmer towed samples collected amidst feeding whale sharks at PB also confirmed microplastics in their immediate feeding area. The total mean number of plastic pieces min-1 swimming there was 0.33 ± 0.16. To assess annual (2016 – 2018), seasonal (wet vs. dry), and location (N-MB, S-MB, or KM) effects on boat trawl sampled plastic abundance, we fitted a linear model of log-transformed plastic abundance on year, season, location and a season-location interaction with N-MB wet season data as the intercept. The results from these models (Table 5.3) based on the quantitative boat trawl surveys of floating plastic (Figure 5.6B), indicate that there is a seasonal effect on plastic abundance that is dependent on location (p = 2.4 × 10-6), but did not detect a year effect after adjusting for location, season, and the season-location interaction (p = 0.073) (Table 5.3). Plastic at N-MB was 12.6 times CHAPTER 5. MICROPLASTICS ON THE MENU 162 more abundant (p = 2.4 × 10-6) during the wet season than the dry season for a given year (Table 5.3). Plastic abundance at S-MB in the wet season is estimated to be 28.2 times more than plastic abundance in the dry season for a given year; however, this seasonal effect was not significantly different from the seasonal effect for N-MB (p = 0.246). The seasonal effect was also significant at KM (p = 0.037), but the size of the effect was smaller than at MB – plastic was 2.8 times more abundant during the wet season than the dry season for a given year (Table 5.3). Plastic abundance was similar for nearby N-MB and S-MB during the wet season (p = 0.498), but 3.5 times greater (p = 0.012) at N-MB than at KM during the wet season (Table 5.3), indicating geographic differences in plastic abundance between these two locations over 450 km apart. Testing for location effects excluded PB due to the availability of a single season of data (the 2017 wet season only). However, the mean and median abundance of plastic during the 2017 wet season at PB (207,806 ± 74,217) was similar in magnitude to those for MB and KM during the wet season (range of means: 144,073 – 449,381; see Figure 5.6B and Table 5.2). Table 5.3 Estimated coefficients, t-values and significance levels (p), for linear model of log-transformed trawl sampled anthropogenic debris (plastic m-3) on year, season, location and a season-location interaction. Coefficients Estimate SE t value p value Intercept (N-MB, Wet) -0.512 0.376 -1.362 0.177 Year -0.393 0.216 -1.820 0.073 Season, Dry -2.535 0.498 -5.087 2.4 × 10-6 Location, S-MB 0.297 0.435 0.681 0.498 Location, KM -1.251 0.486 -2.577 0.012 Location S-MB: Season, Dry -0.804 0.688 -1.169 0.246 Location KM: Season, Dry 1.520 0.714 2.128 0.037 Residual standard error: 1.305 on 79 degrees of freedom; Multiple R-squared: 0.4913; Adjusted R-squared: 0.4526; F-statistic: 12.71 on 6 and 79 DF, p-value: 5.445 × 10-10; N-MB = North Manta Bay, S-MB = South Manta Bay, Nusa Penida MPA; KM = Karang Makassar, Komodo NP; PB = Pantai Bentar, East Java 163 CHAPTER 5. MICROPLASTICS ON THE MENU 164 5.4.4 Visual Surveys Similar to trawl surveys, the linear model (Table 5.4) for visual surveys of floating plastics (Figure 5.6C) identified a strong seasonal effect of plastic abundance that is dependent on location, but failed to find an annual temporal effect (p = 0.131). Plastic was an estimated 23 times more abundant (p = 7.83 × 10-7) during the wet season than the dry season at N-MB for a given year. Plastic abundance at S-MB in the wet season is estimated to be 44.5 times greater than plastic abundance during the dry season. Although this seasonal effect for S-MB was significant, it was not significantly different than the seasonal effect for N-MB (p = 0.417), in line with what was observed with the trawl survey. In comparison to N-MB, a lesser (p = 0.008) seasonal effect was observed for KM with plastic abundance in the wet season an estimated 2.4 times greater than in the dry season for a given year (Table 5.4). Like trawl sampled plastic abundance, visually assessed plastic abundance during the wet season differed between N-MB and KM (p = 9.83 × 10-5), but not between S-MB and N-MB (p = 0.313) (Table 5.4), further indicating a geographic effect on plastic abundance. Moreover, the amounts of visually assessed and trawl sampled plastic were dependent and concordant (Kendall’s tau coefficient = 0.42, z = 5.83, p = 5.60 × 10-9, n = 184) for all locations (including PB), indicating that visually monitoring plastic abundance might be sufficient to capture the broader seasonal and geographic trends. However, the mean abundance of visually observed floating plastic at PB during the wet season (210 ± 126 plastic km-2) appears lower than that of N-MB, S-MB, and KM during the same season (range of means: 2,004 – 40,844, see and Table 5.2). (Continued from previous page) Figure 5.6 Boxplot depicting plastic abundance estimated through (A) swimmer towed trawl surveys (plastic min-1), (B) boat trawl surveys (plastic m-3), and (C) visually counted anthropogenic debris (plastic km-2) from north (N-MB) and south (S-MB) Manta Bay, Nusa Penida MPA; Karang Makassar (KM), Komodo NP; and Pantai Bentar (PB), East Java. 165 Table 5.4 Estimated coefficients, t-values and significance levels (p), for linear model of log-transformed visually counted anthropogenic debris (plastic km-2) on year, location, season, and a location-season interaction. Coefficients Estimate SE t value p value Intercept (N-MB, Wet) 8.551 0.441 19.386 < 2 × 10-16 Year 0.386 0.254 1.523 0.132 Season, Dry -3.137 0.585 -5.366 7.83 × 10-7 Location, S-MB 0.518 0.510 1.016 0.313 Location, KM -2.337 0.569 -4.104 9.83 × 10-5 Location S-MB: Season, Dry -0.658 0.807 -0.815 0.417 Location KM: Season, Dry 2.272 0.838 2.711 0.008 Residual standard error: 1.531 on 79 degrees of freedom; Multiple R-squared: 0.5847; Adjusted R-squared: 0.5531; F-statistic: 18.54 on 6 and 79 DF, p-value: 2.512 × 10-13; N-MB = North Manta Bay, S-MB = South Manta Bay, Nusa Penida MPA; KM = Karang Makassar, Komodo NP; PB = Pantai Bentar, East Java 5.4.5 Characterization of Debris Most (range: 78 – 98%) trawl sampled plastic pieces were < 5 mm long in all study locations (Figure 5.7A). The majority of plastic pieces at N-MB, S-MB, and KM were in the second smallest category (1 – 5 mm; 53 – 66%), while the smallest category (< 1 mm) dominated at PB (53%). Larger plastic debris > 5 mm long was present at all locations, but its contribution to the total plastics was generally small and varied among locations (2 – 22%). Overall, the trawl sampled debris was mostly composed of plastic ‘films’ (range: 28 – 55%), with the exception of the dry season in KM (13%), and plastic ‘fragments’ (range: 20 – 65%) (Figure 5.7B). At N-MB and S-MB, 35 – 55% of debris were plastic ‘films’ and 20 – 43% were plastic ‘fragments.’ At KM, plastic ‘fragments’ were most common (63 – 65%) during both seasons; whereas, during the wet season, 28% of debris were plastic ‘films’ compared to 13% in the dry season (Figure 5.7B). Plastic debris at PB during the wet season was approximately equally composed of ‘films’ (33%), ‘fragments’ CHAPTER 5. MICROPLASTICS ON THE MENU 166 Figure 5.7 The size (A) and type (B) of anthropogenic debris recorded through trawl and visual (C) surveys from South (MB-S) and North (MB-N) Manta Bay, Nusa Penida MPA, Karang Makassar (KM), Komodo NP and Pantai Bentar (PB), East Java pooled across years and presented as a fraction of the total amount. 167 (35%) and ‘other’ debris (32%), which was mostly paint fragments. Debris classified as ‘other’ (range: 1 – 25%) included paint fragments and tar for N-MB, S-MB, and KM. A total of 458 fibers were found in all samples and were excluded from debris counts. The types of debris found in the swimmer towed samples amidst feeding manta rays from MB and the number of pieces (in parentheses) included films (9), fragments (5), foam (1), lines (2) and other items, including paint fragments and tar (3). The types of debris collected by swimmer towed samples from KM included fragments (3) and tar (1). The types of debris found in swimmer towed samples from PB were films (2), fragments (3) and other items (6), including paint fragments and tar. Fibers were found in 43% (25) of all samples but were excluded from plastic calculations. The visually counted plastic debris was mostly plastic ‘films’ (range: 45 – 87%) and plastic ‘fragments’ (range: 13 – 24%; Figure 5.7C) corresponding to the most common types of trawl sampled debris (Figure 5.7B). Foamed polystyrene was commonly observed at N-MB, S-MB, and KM making up 3 – 13% of debris. Whole bottles and single-use cups (counted as ‘bottle’) were also commonly observed at the sites in Nusa Penida, regardless of season, and in Komodo during the dry season, making up 9 – 18% of floating debris. Items classified as ‘other’ in visual surveys included plastic straws, cigarette butts, ropes, sandals, shoe soles, and other rubber items. 5.4.6 Plastic Ingestion In MB, based on mean plastic abundance estimates for the north and south sections of the bay combined, the estimated plastic ingestion by manta rays was 62.7 pieces h-1 during the wet season and 4.4 pieces h-1 during the dry season (Table 5.5). These values compare with estimated ingestion rates for KM of 25.1 pieces h-1 during the wet season and 20.7 pieces h-1 during the dry season. In PB, the estimated wet season plastic ingestion by whale sharks was ~ 136.9 pieces h-1. Based on the ratio of plastic to CHAPTER 5. MICROPLASTICS ON THE MENU 168 Table 5.5 A comparison of plastic abundance estimates in filter-feeding megafauna foraging grounds, theoretical ingestion estimates and plastic associated pollutant levels (when available). Location Region Species Season Plastic pieces (m-3) Ingestion Rate (m3h-1 or *d-1) Theoretical Plastic Ingestion (h-1 or *d-1) MEHP (ng/g) HCB (ng/g) PCB (ng/g) DDT (ng/g) PBDE (ng/g) OCs (ng/g) Reference MB, Nusa Penida MPA CT Mobula alfredi Wet 0.55 - 0.90 86.4 h-1 62.7 h-1 / / / / / / present study MB, Nusa Penida MPA CT Mobula alfredi Dry 0.04 - 0.06 86.4 h-1 4.4 h-1 / / / / / / present study KM, Komodo NP CT Mobula alfredi Wet 0.29 86.4 h-1 25.1 h-1 / / / / / / present study KM, Komodo NP CT Mobula alfredi Dry 0.24 86.4 h-1 20.7 h-1 / / / / / / present study PB, East Java CT Rhincodon typus Wet 0.42 326 h-1 136.9 h-1 / / / / / / present study La Paz Bay SoC Rhincodon typus January - February 0.0 - 0.14 326 h-1 22.8 h-1 171d-1*# / 0.19 ± 0.19a 11.4 ± 8.6a 1.3 ± 1.8a 0.29 ± 0.32a / Fossi et al., 2017 La Paz Bay SoC Balaenoptera physalus / 0.0 - 0.14 5,893 d-1* 825d-1* 40.0 ± 23.2b 38.5 ± 33.6c 8,754 ± 6543c 3,110 ± 2246c / 11,897 ± 6763c Fossi et al., 2014, 2016 Pelagos Sanctuary Med Balaenoptera physalus June - September 0.62/0.31 5,893 d-1* 1,827/ 3,654d-1* 54.8 ± 27.7b 29.9 ± 12.0c 13,327 ± 8548c 10,477 ± 7477c / 23,835 ± 15,057c Fossi et al., 2014, 2016 Pelagos Sanctuary Med Cetorhinus maximus / 0.62 881 h-1 540 h-1, 13,110 d-1* / 20.6c 1,483c 1,668c / / Fossi et al., 2014 CT = Coral Triangle; SoC = Sea of Cortez; Med = Mediterranean Sea. Wet = November – April; Dry = May – October * daily ingestion estimate; # based on 7.5 hours of feeding; a) ng/g wet weight; b) ng/g frozen weight; c) ng/g lipid basis. Abundance ± Standard Error. Pollutant concentrations ± Standard Deviation. MEHP = Monoethylhexyl phthalic acid; HCB = Hexachlorobenzene; PCB = Polychlorinated biphenyl; DDT = Dichlorodiphenyltrichloroethane; PBDE = Polybrominated diphenyl ethers; OCs = organochlorines 169 plankton by weight (Table 5.2), in the wet season, manta rays in MB might be ingesting between 110 and 980 g of plastic for every kg of plankton. However, unlike plastic, plankton abundance by weight did not appear to vary seasonally for MB and KM (Figure 5.8). Figure 5.8 Plankton abundance pooled by season for south (S-MB) and north (N-MB) Manta Bay, Nusa Penida NPA; Karang Makassar (KM), Komodo NP; and Pantai Bentar (PB), East Java. The amount of manta ray egested fecal material opportunistically collected from free swimming manta rays varied considerably, ranging from ~ 0 to 3.17 g dry weight (n = 21), with three samples weighing > 0.5 g dry weight. One of these fecal samples (0.58 g), collected at Manta Point, Nusa Penida, ~ 12 km south of S-MB, contained 26 pieces of plastic (Figure 5.9). The overall mean length of the plastics was 7.8 mm, with 54% of plastics being < 5 mm, and the rest being > 5 up to 30 mm. Two fiber pieces, 3 and 5.5 mm, were excluded. The ratio of plastic dry weight (0.05 g) to that of organic material (0.58 g) for this sample was 0.09:1. Further, a single sample of manta ray vomit, also collected at Manta Point, contained 66 pieces of plastic (Figure 5.10). The overall mean length of the plastics was 7.7 mm, with 77% of plastics being < 5 mm, and the rest being > 5 up to 94 mm. No fibers were found in this sample. The ratio of plastic dry weight (0.25 g) to that of organic material (4.12 g) for this sample was 0.06:1. Although only two of the samples of egested material were confirmed to contain plastics (other than fibers), these findings provide proof of plastic ingestion by manta rays. CHAPTER 5. MICROPLASTICS ON THE MENU 170 Figure 5.9 Low (A) and high magnification photographs of smaller (B) and larger (C) plastic debris found in manta ray fecal material (n = 1) from Nusa Penida MPA. 171 Figure 5.10 Various sizes and types of plastic debris found in manta ray vomit material (n = 1) from Nusa Penida MPA. Larger debris are photographed under high (A, B) and low (C) magnification, and smaller debris (D) under high magnification. CHAPTER 5. MICROPLASTICS ON THE MENU 172 5.5 DISCUSSION This study has confirmed that manta rays are able to ingest plastic marine debris. Furthermore, plastic debris and microplastics were found in all of the examined feeding grounds for manta rays and whale sharks in Indonesia, within the Coral Triangle. Our observations confirm that manta rays and whale sharks are feeding amongst plastic debris in these locations and that plastic materials are likely being ingested from surface waters. These preliminary findings suggest that manta rays and likely whale sharks are unable to select out plastic debris during the feeding process, that it is feasible that they are regularly ingesting plastic material in areas with high densities of surface plastics, but that they may also be able to expel at least some of what is ingested through regurgitation or passing it in fecal matter. Further, quantitative trawl sampling enabled us to estimate the theoretical plastic ingestion rates for manta rays and whale sharks at these study locations. Plastic debris abundance estimates fit within the current modeled estimates (Eriksen et al., 2014; Maximenko et al., 2012; van Sebille et al., 2015), but differed between locations and were higher during the wet than the dry season. Across locations, the most common types of plastics were ‘films’ and ‘fragments,’ which were mostly < 5 mm. 5.5.1 Plastic Abundance Estimates Overall, our plastic abundance estimates (Table 5.2) from trawls (range: 20,000 –449,000 pieces km-2) and visual surveys (range: 100 – 10,000 pieces km-2) were in the upper end of the modeled estimates for the Coral Triangle region (Eriksen et al., 2014; van Sebille et al., 2015). Plastic by weight estimates also fit within the models with Manta Bay (MB) wet season estimates resembling those by Eriksen et al. (2014) and van Sebille et al. (2015) and the rest the Maximenko et al. (2012) model. 173 However, the three most recent modeled predictions for microplastics (Eriksen et al., 2014; Maximenko et al., 2012; van Sebille et al., 2015) vary considerably (up to a factor > 100) for Indonesia and the Coral Triangle. Variations between models are likely in part due to differences in methodology (i.e., net mesh sizes, sample processing, and model approach; van Sebille et al., 2015). Complex regional oceanographic and weather conditions are also likely to contribute to these variations. Further, fine scale oceanographic features are likely not detected in these large scale oceanographic models; however, such small-scale variations are critical when considering the feeding ecology of large filter-feeders. For instance, previous small scale investigations of fin whale feeding grounds, in two sections (Ligurian and Sardinian Seas) of the Pelagos Sanctuary, Mediterranean Sea, where the coastal geography is complex, show variability in plastic abundance on the scale of 101 km (Fossi et al., 2016). Plankton within manta ray and whale shark feeding areas vary often on an even finer scale, with patches of high prey density ranging 101 – 102 m (Armstrong et al., 2016; Rohner et al., 2015). So while the global models are useful, in the case of megafauna feeding implications fine scale investigations are crucial. 5.5.2 Seasonal Trends in Plastic Debris Abundance In N-MB and S-MB, and to a lesser extent, Karang Makassar (KM), plastic abundance was greater in the wet than dry seasons. The magnitude of this seasonal difference was much greater at S-MB and N-MB than at KM, where these estimates were highly variable and the seasonal patterns of change were less clear. Nevertheless, an increase in plastic abundance during the wet season is consistent with recent evaluations of marine debris sources that point toward rivers contributing the majority (88 – 94%) of marine debris (Lebreton et al., 2017; Schmidt et al., 2017). Asia hosts the majority of the CHAPTER 5. MICROPLASTICS ON THE MENU 174 world’s most polluted rivers (103 out of 122), including the top 20, which account for 67% of the annual river debris input globally (Lebreton et al., 2017). Indonesia itself was identified as a major contributor with rivers from Java and Sumatra contributing ~14% of the total global annual marine plastic emissions (Lebreton et al., 2017). Seasonally, Indonesian waste inputs from rivers are highest during the wet season – an estimated 35,000 t in February compared with 1,800 t in August (Lebreton et al., 2017). Thus, our findings of increased plastic abundance during the wet season are supported by the seasonal increase in regional river pollution emissions. Similar findings from the other side of the world indicate that the increase in rainfall during the hurricane season in Banderas Bay, Mexico also likely leads to an increase in the contribution of plastic debris from land-based sources via river emissions (Pelamatti et al., 2019). Further, our findings suggest that most of the debris observed at the study sites is generated regionally, as plastic volumes emitted from heavily polluted Chinese rivers are highest during the Indonesian dry season (June – September; Lebreton et al., 2017). However, it is conceivable that long-range drifting debris (Maximenko et al., 2012; van Sebille, 2014), potentially brought via the southward currents (Mayer & Damm, 2012) of the Indonesia Throughflow (ITF), also makes a contribution to the plastic pollution in the region. Nevertheless, in many parts of Indonesia, and other nearby developing countries, common practice is to dispose of waste in open dumping grounds close to river beds (Hoornweg & Bhada-Tata, 2012). With the onset of the wet season, river beds swell to accommodate the increased water volume and in the process deliver debris to the marine environment. Thus waste management efforts should be focused on preventing waste from entering the marine environment prior to the start of wet season. Such intervention could be in the form of river bed cleanups, prohibition and enforcing a moratorium on 175 waste disposal nearby water sources. In addition, the introduction of contained waste disposal facilities, located away from rivers and coastlines is needed throughout Indonesia and other developing countries, especially in more remote and rural areas that currently lack access to any waste disposal facilities. 5.5.3 Geographic Differences in Abundance and Characteristics of Plastics Both trawl sampled and visually assessed plastic abundance estimates were greater at MB than at KM, which is likely a factor of the distance of these locations from densely populated areas (Pedrotti et al., 2016) and rivers with substantial plastic emissions (Lebreton et al., 2017). Nusa Penida MPA and the MB sites, for instance, are in close proximity to some of Indonesia’s most populous islands, including Java, Bali and Lombok, while Komodo National Park (NP) is surrounded by the relatively sparsely populated islands of Sumbawa, Flores, and Sumba (Gaffin et al., 2004; Yetman et al., 2004). The study locations adjacent to densely populated areas are likely to have a higher abundance of plastic debris (Pedrotti et al., 2016); however, local oceanographic conditions are likely to strongly influence the trajectories of drifting plastic and thus the local abundance. Fine- scale oceanographic conditions in this region, such as extreme tidal currents (Field & Gordon, 1996), complex shorelines (Sugimoto, 1975) and bathymetry could generate local gyres and traps/sinks for plastics. There was notable variability in daily plastic abundance estimates. Particularly at MB, this might be an effect of the surface debris coming into sampling areas in large slicks (E. Germanov, pers. obs.), as it is transported shoreward from the pelagic environment akin to natural debris and plankton (Shanks, 1983). Sampling through these slicks, when present, might result in substantially higher abundance estimates on those days. While efforts were made to reduce variability by sampling at similar tide and wind CHAPTER 5. MICROPLASTICS ON THE MENU 176 conditions, the daily variability in plastic abundance that we observed highlights the need for more focused and fine scale research of the coastal oceanographic and weather influences on plastic debris abundance and drifter trajectories. The higher percentage of smaller sized microplastics sampled in KM compared to those sampled from MB, suggest that the plastic pieces are either traveling from further locations and undergoing additional degradation through ultra-violet radiation and other environmental influences, or entering the marine environment at more advanced stages of fragmentation. The remoteness of Komodo NP from major urban centers and large, plastic-emitting rivers suggests that plastic size is more likely to be related to the distance and extended time for degradation than size at entrance to the marine environment. Plastic and anthropogenic debris abundance estimated from trawl surveys at Pantai Bentar (PB) during the 2017 wet season was comparable to those estimated for MB and KM during the wet season, as well as for the Cilacap coast of Java (0.27 – 0.54 pieces m-3), Indonesia, adjacent to the emission points of the Donan and Serayu rivers (Syakti et al., 2017). Interestingly, the amount of visually observed floating plastics at PB was markedly lower and more similar to levels detected in MB and KM during the dry season. Further, microplastics sampled from PB had the highest percentage of particles < 1 mm compared to the other study locations. Given the proximity of PB to densely populated areas (e.g., Surabaya city with ~ 3,457,00 people as of 2015; Indonesia Population, 2019) and river emission points, it is surprising that the larger debris size classes did not comprise a greater portion of the total plastic debris. Generally, there are more river outflows on the east Java and Bali coastlines than further east, like those adjacent to the Komodo NP. Thus we would expect to find higher or comparable levels of plastic at PB and MB, to those at KM during the wet season. However, while the overall rainfall patterns are generally similar for the three study sites during our study period (see 177 Figure 5.3), regional variations and inter-annual differences (Hendrawan et al., 2019) likely influence the strength of river flows and thus potentially plastic inputs. A striking difference between the study locations was the water clarity (E. Germanov, pers. obs.), with PB having high levels of algae and sediment in the water, which could contribute to faster sinking of marine debris (i.e., vertical migration) and thus reduced detection at the surface. A faster vertical decent of debris might occur due to biofouling of larger buoyant items, such as plastic bags (i.e., film plastic) (reviewed in Bond et al., 2018; Cole et al., 2011; Lobelle & Cunliffe, 2011), as well as the attachment of foreign particles such as clay minerals (Kowalski et al., 2016), which both would decrease the buoyancy of plastics. Further, the inflow of less saline water via the Karimata Strait (Mayer & Damm, 2012) and nearby rivers emissions might decrease the buoyancy of larger plastics and accelerate their migration into the sediments (Cole et al., 2011). Follow- up investigation of plastic abundance in the bottom sediments at PB is warranted. The most abundant types of debris by both trawl sampling and visual observation, as well as those present in the manta ray egestion material, were plastic films. This category of plastics corresponds to items commonly composed of Polyethylene (PE) polymers that have a lower density than sea water and are thus buoyant (Andrady, 2011). This observation coincides with PE being the most abundant plastic polymer in the marine environment (reviewed in Bond et al., 2018; Hidalgo-Ruz et al., 2012). Further, this is likely a reflection of the high level of packaging on consumer goods and single-use items, including ‘sachets’ that are commonly used in Indonesia (Merkl et al., 2015). Film plastics are considered low value plastics for sales into recycling and are rarely salvaged from waste streams (Merkl et al., 2015). Although more valuable and readily recycled (Merkl et al., 2015), other single-use plastics commonly encountered in our surveys were CHAPTER 5. MICROPLASTICS ON THE MENU 178 plastic bottles. Plastic bottles, composed of polyethylene terephthalate (PET), will eventually break up to plastic fragments, but are much more durable than PE (Bond et al., 2018; Muthukumar et al., 2011) and thus persist longer and disperse further into the marine environment. The durability of plastic items, and the distance and time taken to ‘arrive’ at the sampling locations, as well as the rates of vertical migrations through biofouling, might explain the differences in the plastic ‘signatures’ of the different locations that we investigated (e.g., more hard plastic fragments in KM than MB). Follow-up studies characterizing the polymers types present in the microplastics can help link the plastic debris to consumer goods and help to identify their origin (Reisser et al., 2013; Rios et al., 2010; Pedrotti et al., 2016; Pelamatti et al., 2019). Further, as different polymers have different affinity properties (Bond et al., 2018), and contain different additives with variable toxicity (Endo et al., 2005; Teuten et al., 2009), understanding the chemical composition of microplastics in this region will allow us to better evaluate the potential implications of plastic ingestion on manta rays, whale sharks and other filter-feeding megafauna (Chapter 2). 5.5.4 Theoretical Plastic Ingestion Rates The plastic abundance and ratios of plastic to plankton calculated from this study were compared with those reported for filter-feeders elsewhere. The plastic to plankton weight ratios estimated for manta ray and whale shark feeding grounds in the current study, apart from S-MB, were lower than the 0.5:1 ratio estimated for the Mediterranean Sea (Collignon et al., 2012). As well as an indicator of the high pollution levels in the Mediterranean Sea, perhaps the lower ratios we observed are also a function of the regional productivity and the high plankton abundance in megafauna feeding hotspots (Putra et al., 2016). Manta rays and whale sharks feed in high density prey patches to 179 attain a positive energy balance, as feeding behaviour is energetically costly (e.g. Armstrong et al., 2016; Motta et al., 2010; Rohner et al., 2015). The necessity for high density plankton in areas where manta rays and whale sharks feed renders the plankton to plastic ratio less informative than in other less productive regions. The amount of plastic in Indonesian feeding grounds during the wet season was comparable to the levels found in fin whale feeding grounds within the Pelagos Sanctuary, Mediterranean Sea (≤ 0.62 items m-3; Fossi et al., 2014, 2016) and higher than those in whale shark feeding habitats within La Paz Bay, Sea of Cortez (≤ 0.14 items m-3; Fossi et al., 2017). The estimated plastic ingestion rates for manta rays (~ 63 items h-1 in the wet season, ~ 4 items h-1 in the dry season at MB; ≤ 25 items h-1 at KM) and the ratio of plastic to plankton indicate that in some locations during times of the highest plastic abundance (wet season), manta rays might be ingesting up to 980 g of plastic per kg of plankton. The ratios of plastics to plankton noted from the egested material samples are 10 – 16-fold lower, but nevertheless considerable. Further, we estimated that the whale sharks feeding at PB are ingesting up to 6-fold higher amounts of plastic (~ 137 items h-1) than the those feeding in La Paz Bay (see Table 5.5; Fossi et al., 2017). Even with these higher estimates, we are likely to be underestimating the amount of plastic intake during surface feeding, as our abundance estimates are limited to free-floating surface and near surface plastics and do not include any plastics ingested by plankton (Cole et al., 2013; Setälä et al., 2014). Further, these estimates do not account for any mid-water or bottom feeding. Plastic levels have not been characterised in many other manta ray and whale shark feeding grounds, but despite this general lack of comparisons, we think that large filter-feeders in Indonesia are likely to be feeding in more heavily polluted waters than most other locations throughout their range (Chapter 2). The current understanding of local manta ray feeding ecology and habit use is CHAPTER 5. MICROPLASTICS ON THE MENU 180 considered in light of plastic estimates in known feeding grounds. Information on the total daily time that individual whale sharks and manta rays spend feeding at PB and MB/KM, respectively, is not currently available. However, manta rays display a high degree of site affinity to MB, especially immature manta rays, and are most commonly observed feeding year-round (Chapter 3). Further, high levels of habitat use are documented for KM, with median individual visitations lasting two hours (Dewar et al., 2008). Returning to the same polluted area to feed for several hours each day would put these manta rays at high risk of ingesting substantial quantities of plastic. Interestingly, a sample of vomit containing plastics was collected from a manta ray that had not been sighted at MB since 2015, which might indicate that feeding in other polluted areas is occurring. Manta rays are likely feeding in other sites within the greater region other than MB (Chapter 3) and KM (Chapter 4), and at depth (Couturier et al., 2013). Particularly for KM, seasonality appears to have an effect on feeding behaviour with less feeding behaviour observed at KM during the dry season, suggesting that manta rays are preferring to feed elsewhere at this time. However, based on the weight of plankton collected at both MB and KM, which was similar for both wet and dry seasons, we were unable to relate plankton availability with plastic abundance. To fully grasp the level of threat from plastic ingestion it is necessary to gain a greater understanding of feeding ecology for manta rays and the variety of locations used in the region. Plastic abundance is yet to be evaluated in other known feeding grounds within the region (e.g., Raja Ampat, Perryman et al., 2019; Setyawan et al., 2018), and at different water depths where feeding might also occur. Tagging studies using transmitters with depth and acceleration sensors (e.g., daily diary tags, Gleiss et al., 2013; Wilson et al., 2008) would enable studies of fine scale manta ray habitat use, visitation lengths, and behavior (e.g. feeding at the surface, at depth vs. cleaning), thus improving estimates of plastic exposure. 181 5.5.5 Implications of Plastic Ingestion Plastic pieces in two samples of manta ray egested material collected within the Nusa Penida MPA confirm that plastic ingestion by manta rays, including plastic > 5 mm, is feasible. Observing plastic in a fecal sample from a manta ray indicates that this manta ray has ingested plastic and that at least some of this plastic has passed through the entirety of its digestive tract and was excreted. Observing plastic in a vomit sample from a manta ray indicates that while plastic was ingested, it is also possible for manta rays to regurgitate plastic before it passes through the digestive system. It is not surprising that manta rays cannot selectively exclude microplastics from zooplankton in their filter-feeding process, but empirical evidence until now has been lacking. Whale sharks are also confirmed to ingest plastics ranging from 15 to 200 mm in length, as shown from necropsies performed in Brazil (Sampaio et al., 2018), Thailand (Haetrakul et al., 2009), Philippines (Abreo et al., 2019), and Malaysia (Lee, 2019), by direct observations by Fossi et al. (2017) in La Paz Bay, Sea of Cortez, and analyzed from fecal samples (Donati et al., 2019). Ingestion of large plastic pieces is a concern, as the accumulation of plastic in the digestive system is more likely. Plastic accumulation in the digestive tracts of marine mammals (de Stephanis et al., 2013; Jacobsen et al., 2010) and sea turtles (Santos et al., 2015; Stamper et al., 2009) is linked to mortality and/or morbidity (either through dietary dilution, blockage or mechanical damage of intestinal tracts). While the levels of plastic debris found in the stranded and deceased whale sharks mentioned above were not as high as those reported for some stranded whales (e.g., 73.6 kg; Jacobsen et al., 2010), internal perforations from what appeared to be a single hardened plastic straw are believed to have caused the death of at least one whale shark (Haetrakul et al., 2009). The ability of manta rays and whale sharks (Donati et al., 2019) to regurgitate CHAPTER 5. MICROPLASTICS ON THE MENU 182 ingested plastic or pass it through their digestive systems and eventually expel it is a positive finding. However, even if eventually expelled, while plastic is within the body it has the potential to disrupt regular gut functioning (Fackelmann & Sommer, 2019) and to leach and desorb associated toxic chemicals, such as POPs and heavy metals (Rochman et al., 2013, 2014; Tanaka et al., 2013; Teuten et al., 2009), allowing these chemicals to be absorbed by the organisms. Polyethylene (PE), the key component of films and fragments, is particularly adsorbent for POPs (Bond et al., 2018; Teuten et al., 2009). Smaller particles, because of increased surface area, might harbor higher levels of POPs (Cole et al., 2011; Mato et al., 2001). Thus finding that the majority of plastics in feeding grounds are small (< 5 mm) indicates an increased risk of pollutant exposure for filter-feeding megafauna feeding at these sites. Plastic associated pollutants, including phthalates and brominated flame retardants (e.g., polybrominated diphenyl ethers [PBDEs]) and POPs (DDTs, PCBs, and other organochlorine compounds) were detected in biopsy sampled skin from whale sharks feeding in La Paz Bay and blubber from fin whales feeding in La Paz Bay and the Mediterranean Sea (Fossi et al., 2016, 2017). Microplastic abundance estimates in PB and the theoretical estimates of plastic ingestion by whale sharks at this location are markedly greater than those for La Paz Bay (Table 5). However, whether whale sharks, manta rays and other large filter-feeders are exposed to comparable or higher levels of POPs as those reported for La Paz Bay and the Mediterranean is less certain. Based on a 2009 global assessment of POPs in coastal waters via the analysis of PE pellets collected from beaches (Ogata et al., 2009), pollution levels of PCBs and DDTs were markedly lower in Jakarta Bay, Indonesia than those in the American West Coast and the Mediterranean Sea. However, more recent pellet analyses indicate that Indonesian pollution levels have increased (International Pellet Watch, 2019). Heavy metal contamination of mobulids has 183 been confirmed (Essumang, 2010; Ooi et al., 2015); however, this contamination is yet to be linked to heavy metal exposure via plastic ingestion. An understanding of the implications of microplastic ingestion on the population level, requires knowledge of the physiological response of individuals, and how likely it is to influence their reproduction. As filter-feeders are likely to feed from several locations, and plastic abundance is likely to vary along these planes, quantifying their exposure to pollutants that serve as a proxy for plastic exposure (e.g. Fossi et al., 2014, 2017), might better capture the level of risk to individuals and populations. Thus, future investigations should focus on quantifying the plastic associated pollutant contamination levels of manta rays, whale sharks, and other filter-feeding megafauna within Indonesia and the greater Coral Triangle (see Chapter 3 and Appendix I for recent discussions on future direction recommendations). These inquiries could be achieved through minimally invasive and non-lethal biopsy sampling (e.g. Fossi et al., 2017; 2000). Further, monitoring individual health might be achieved through the analysis of biomarker responses (Dove et al., 2012; Fossi et al., 2017). For example, some serum metabolites have been associated with the health of whale sharks in aquaria providing a starting point for assessing the health of wild individuals and populations (Dove et al., 2012). Further, the first investigation using CYP1A-like protein responses in whale sharks in the wild shows a positive correlation with some common plastic additives (i.e. brominated flame retardants), suggesting that induction of this biomarker may serve as a warning indicator of pollutant exposure in this species (Fossi et al., 2017). Long-term monitoring of health indexes for high risk populations will provide insight into the implications of microplastic ingestion and exposure to associated pollutants, while broad regional studies will assist in identifying pollution hotspots, necessary for effective management of these threatened species. Focusing further research on these species is CHAPTER 5. MICROPLASTICS ON THE MENU 184 warranted, particularly as large filter-feeders are widely considered important sentinels for the implications of plastic pollution in marine environments (Chapter 2; Nelms et al., 2019). 5.5.6 Efforts to Stem the Plastic Tide In 2017, in response to being ranked as the world’s second largest plastic marine debris emitter (Jambeck et al., 2015), mostly through river emissions (Lebreton et al., 2017), the Indonesian government pledged 1 billion USD per annum towards reducing plastic marine debris in their waters by 70% before 2025 (Langheim, 2017). Recently, in December 2018, several major Indonesian cites, including those in the Bali province (Aqil, 2019), enacted bans on frequently used single-use plastics such as bags, straws and polystyrene containers, demonstrating the commitment of government to addressing the plastic issue. To this end, our multi-year dataset can serve as a baseline for monitoring the plastic pollution in these coastal environments within Bali province and important habitats for threatened species. Continuing to monitor these coastal environments and gaining an understanding of how local oceanography contributes to the accumulation of plastic debris along coastlines will be crucial as Indonesia’s government works to curb this threat. 5.6 CONCLUSION We estimated plastic marine debris abundance, and characterized the debris in three feeding grounds for manta rays and whale sharks in Indonesia, thus providing the the first empirical estimates of plastics in the region. Our plastic abundance estimates are consistent with overall modeled microplastic abundance estimates for the greater region. Estimated pollution emission from regional rivers support the relatively higher plastic 185 abundance we observed during the Indonesian wet season (November – April). Regional differences in plastic abundance, prevalent plastic sizes and types indicate that factors in addition to season, such as population density and local oceanographic processes, influence local plastic debris abundance. Nevertheless, the seasonal variation in plastic abundance highlights that concentrating marine debris intervention and prevention measures from land-based sources, such as rivers, prior to the onset of the wet season would maximize efforts to reduce plastic levels at the source. This is the first study to provide theoretical estimates of plastic marine debris ingestion for manta rays, as well as for whale sharks in the Coral Triangle. Although, the extent of the impacts on individuals and populations remain unknown, manta rays ingested plastic marine debris, and it appears that whale sharks are probably also ingesting plastics in this region. The urgency to reduce marine plastic pollution has become a prominent environmental issue, and many countries are joining global initiatives (e.g., Clean Seas Campaign, 2019) to reduce plastic waste and marine pollution. Incorporating the information gleaned from this study into educational campaigns that bring awareness about the potential impacts of plastic marine debris to flagship species, like manta rays, whale sharks and other charismatic megafauna (Chapter 2; Bowen-Jones & Entwistle, 2002; Panti et al., 2019), might assist local governments as they attempt to socialize increasing regulations on plastic usage and disposal. These efforts could have a positive effect in regions with significant marine tourism industries, such as the Nusa Penida MPA and Komodo NP. Gaining the support of the public and coastal communities will be key in tackling plastic marine debris in the region, one of Indonesia’s, and the whole of the Coral Triangle’s, biggest challenges in protecting its marine environments. 186 CHAPTER 6. GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS 187 CHAPTER 6. GENERAL DISCUSSION 188 Prior to the start of my PhD research, manta rays had very recently emerged as a priority species of great conservation significance to Indonesia. Both manta ray species, Mobula alfredi and M. birostris, were legislatively protected throughout Indonesia’s entire exclusive economic zone in 2014 (Ministry of Marine Affairs and Fisheries, 2014), after their Appendix listings on both the Convention on International Trade of Endangered Species (CITES) in 2013 and the Convention on the Conservation of Migratory Species (CMS) for M. birostris (2011) and M. alfredi (2014). These increased protections, coupled with the increased understanding of the significant economic importance of these species for nature-based tourism (O’Malley et al., 2013), resulted in manta rays becoming flagship species and creating greater awareness for improving marine conservation within Indonesia (Chapter 2; Perryman & Tapilatu, 2018). However, a paucity of information on local manta populations has hampered the development of science-based conservation and management strategies within the country. The most closely monitored populations are those within Marine Protected Areas (MPAs), National Parks (NPs) and conservation zones, such as Nusa Penida MPA, Bali (Chapter 3); Komodo NP, Flores (Chapter 4); Sengalaki Island, East Kalimantan (E. Germanov, pers. obs.); and Raja Ampat; West Papua (Perryman et al., 2019). In these locations, M. alfredi displayed high site fidelity to sites within areas such as the Komodo NP (Dewar et al., 2008), but were also seen to make large-scale movements up to 450 km between protected areas, e.g. the Komodo NP and the Nusa Penida MPA (Germanov & Marshall, 2014). These movements highlighted the migratory ability of manta rays and indicated the existence of movement corridors between major aggregations. These corridors pass through heavily fished areas that until recently targeted mobulids (Heinrichs et al., 2011; Lewis et al., 2015; White et al., 2006), underscoring the need to better understand their habitat use, movement patterns and 189 behavior. Although the recent ban on directed fisheries (2014), and limitations on the international trade of manta ray materials in effect (2014) likely helped to ease pressure on manta ray populations, other anthropogenic pressures remained, which I aimed to clarify. I focused on the most apparent and immediate locally relevant pressures (Couturier et al., 2012). These included the growing manta ray directed tourism as a potential stressor to individuals and habitats through increased boats and divers (Anderson et al., 2011; Bejder et al., 2006a, b; Norman, 1999); anthropogenic injuries from fisheries; and the implications of marine plastic pollution in feeding grounds. Particularly, plastic marine pollution was highlighted (Chapter 2) as major threat in the region as Indonesia has both high levels of plastic emissions (2nd in the world; Jambeck et al., 2015) and high plastic abundance, estimated from models, in the marine environment (Eriksen et al., 2014; van Sebille et al., 2015). To clarify the extent of threats to manta rays in the region, a greater understanding of habitat use (e.g. Cooke, 2008; Dewar et al., 2008; Graham et al., 2016; Speed et al., 2010), population dynamics (e.g. Couturier et al., 2014; McCoy et al., 2018) and behavioral ecology were necessary (e.g. Ashe et al., 2010; Greggor. et al., 2016; Jaine et al., 2012). Citizen science emerged as a promising tool to gather large amounts of data on manta ray sightings in short time periods (Bonney et al., 2009; Pierce et al., 2018) helping to effectively close some of the knowledge gaps (see e.g. Germanov & Marshall, 2014). The important findings resulting from the research in this thesis are summarized below and further discussed in the following sections on manta ray ecology (Chapters 3 and 4), microplastics in the marine environment and large filter-feeders (Chapter 2) and microplastics in manta ray and whale shark feeding grounds (Chapter 5): CHAPTER 6. GENERAL DISCUSSION 190 6.0.1. Habitat Use A substantial number of manta rays used the Nusa Penida MPA (> 600 individuals) and showed high site affinity to two nearby sites (~12 km apart) on the west coast of Penida Island (Chapter 3). The two sites were used by different sex and age classes of individuals to meet different needs. One site was predominantly used by males, and especially immature males, for foraging suggesting that this site is used as a nursery. The other site appeared to be an important cleaning station, social and reproductive ground primarily visited by mature individuals. A larger number of manta rays, than that of the Nusa Penida MPA (> 1,100 individuals), use the Komodo National Park (Chapter 4). About 1% of the individuals are known to make long-range (~ 450 km) movements to the Nusa Penida MPA (Germanov & Marshall, 2014). Manta rays were predominately sighted at four sites stretching from the north to the south regions of the park, a distance of ~ 45 km. Residency rates, population dynamics and the behaviors observed differed between some sites more than others. Seasonal influences played a role in site use in the southernmost site, as was found in Dewar et al. (2008). 6.0.2 Threats to Manta Rays, Particularly Plastics Substantial increases in tourism were observed at manta ray sites in Nusa Penida MPA (Chapter 3) and Komodo NP (Chapter 4). Fishing gear injuries were present in both populations (14% – Nusa Penida MPA; 5% – Komodo NP) highlighting persistent threats from non-directed fisheries in the region. The implications and suggestions for improved management within and outside of conservation zones are discussed further below. Owing to their feeding ecology, habitat range overlap with plastic pollution hotspots, and life history characteristics, manta rays and other filter-feeding megafauna 191 (e.g. whale sharks) were identified as at high risk for plastic ingestion, particularly for microplastics (Chapter 2). The potential consequences of plastic ingestion, including exposure to toxic plastic associated pollutants and the potential for population wide impacts are outlined. Filter-feeding megafauna are proposed as ideal flagship species to raise awareness and engage communities in bettering marine stewardship in regions heavily dependent on marine resources and simultaneously impacted by marine pollution, like Indonesia. The first fine-scale, empirical estimates of plastic abundance from areas within marine conservation zones and feeding grounds for manta rays (Nusa Penida MPA and Komodo NP) and whale sharks (Pantai Bentar, East Java) (Chapter 5) confirmed that plastic is present in levels comparable to modeled estimates for the region (Eriksen et al., 2014; van Sebille et al., 2015). Plastic abundance differed among study sites and had a strong seasonal influence, particularly at Nusa Penida where plastic abundance was much greater in the wet (November - April) than dry (May - October) season. Plastic abundance estimates were used to provide theoretical plastic ingestion estimates and both ingestion and egestion of plastic by manta rays was confirmed. 6.1 MANTA RAY ECOLOGY (CHAPTERS 3 AND 4): The Lesser Sunda island chain, encompassing both the Nusa Penida MPA and the Komodo NP, contains critical habitats with repeated year-round use and high site affinity for collectively at least 1,700 identified manta rays (Chapters 3, 4). Nearly 1,100 individuals were identified in the Komodo NP and ~ 620 in the Nusa Penida MPA. Both of these aggregations are significant on global terms (see Table 3.5 Chapter 3). Manta rays sighted in the two regions differed in their affinity to the regions, with individuals displaying higher affinities to sites in Nusa Penida MPA than those in the Komodo NP CHAPTER 6. GENERAL DISCUSSION 192 suggesting a higher degree of transience might be occurring at Komodo NP as a whole. The average time individuals spend within the sites also differed across the region with the longest being for Manta Bay (MB), Nusa Penida MPA (2.0 days ± 187.7) and Manta Alley (MA), Komodo NP (1.6 days ± 12.9). Individuals appeared to spend less time at sites predominantly used by adults for cleaning and courtship (e.g. Manta Point – MP, 0.9 ± 0.2; Mawan – MW, 0.8 ± 0.3) compared to those also used for foraging (e.g. MB, MA, and Karang Makassar – KM; 1.3 ± 5.4). However, the times spent within some sites were highly variable indicating heterogeneity in site affinities that might be due to individuals being in different demographic groups or different social hierarchies (Perryman et al., 2019). My research showed that population dynamics can differ significantly for manta rays in areas as close as ~ 12 km apart, as we observed in the Nusa Penida MPA, where predominantly males use MB; whereas mature individuals of both sexes typically use nearby MP (Chapter 3). Within the Komodo NP, variations in habitat use, population dynamics and demographics of neighboring sites are more subtle. Seasonality, potentially coupled with local oceanographic influences, appears to play a greater role in determining site use, particularly at the southernmost site, MA (Dewar et al., 2008; Chapter 4). Taken together these findings indicate that manta ray site use is heterogeneous throughout the region and different sites might be more important for certain behaviors (Jaine et al., 2012; Stevens, 2016) or demographic groups (Deakos et al., 2011; Marshall et al., 2011) that changes with life stage (Heupel et al., 2007; Speed et al., 2010; Stewart et al., 2018b). Within the Nusa Penida MPA, it is proposed that Manta Bay (MB) specifically qualifies as a nursery for immature manta rays (Chapter 3 and discussed below). While the data available on immature manta rays at the Cauldron (CL), Komodo NP was too 193 sparse to reach a conclusion, available data show that it fits some of the criteria for a nursery and warrants further investigation. With so few manta ray nurseries identified globally (Stewart et al., 2018a; 2018b), these findings are especially important to furthering our understanding of manta ray ecology. In depth studies on reef manta rays, such as the ones undertaken for this thesis, are revealing that: 1) habitat use varies widely between locations with some sites explicitly used by defined groups of individuals for specific behaviors (present study, Barr & Abelson, 2019; Deakos et al., 2011; van Diunkerken, 2011; Jaine et al., 2012; Stewart et al., 2018b; Stevens, 2016), 2) relatively few sites contain high proportions of immature individuals, and those that do appear to be primarily foraging and feeding grounds, at least for M. alfredi (Chapters 3 and 4, McCauley et al., 2014; Stevens, 2016); 3) a degree of structure within groups is becoming apparent (Chapters 3 and 4, Perryman et al., 2019); 4) while high degrees of site affinity exist, there is substantial mixing between neighboring aggregations and long-range movements of several hundred kilometers are occurring regularly (Chapters 3 and 4, van Diunkerken, 2011; Germanov & Marshall, 2014; Jaine et al., 2014). Thus larger conservation zones that are well enforced, and gear restrictions/modifications for fisheries in movement corridors will offer better protection for this species (e.g. Graham et al., 2016; Venables et al., 2020), especially as enforcement is likely unattainable throughout their entire home range (e.g. Beck et al. 2001; Heupel et al., 2007; Martins et al., 2018). However, some areas within these large home ranges (i.e. reproductive and foraging grounds) will be more important than others and should be prioritized for additional management considerations (e.g Heupel et al., 2007; Hueter et al., 2005; Martins et al., 2018). Further, conservation zones often have additional conservation measures designed to protect habitats and limit destructive practices (e.g. no take zones, fishing gear limitations, vehicle speed regulations, etc.; e.g. CHAPTER 6. GENERAL DISCUSSION 194 Erdmann, 2004a; Kasmidi & Gunadharma, 2017) that go beyond single-species conservation plans. Thus, in addition to the ban on targeted fisheries, these conservation measures add additional benefits to manta rays, such as critical habitat protection. 6.1.1 Nurseries and Manta Rays Repeated sightings of immature individuals were more common in some of the study sites than others prompting me to consider that these sites might serve as nurseries. The criteria developed for what constitutes an elasmobranch nursery has largely focused on sharks (Heupel et al., 2007; 2018; see Chapter 1). Briefly, the criteria for an area to be designated as a nursery hinge on the premise that neonates or young-of-the-year (YOY) sharks are (1) abundant; (2) persistent; and (3) consistently use the area across years (Heupel et al., 2007; 2018). More recently, an attempt was made to classify ray nurseries (Martins et al., 2018); however, mobulids were overlooked. This is likely because few studies describe potential mobulid nurseries (McCauley et al., 2014; Stewart et al., 2018a; 2018b; Stevens, 2016), and there is yet to be a consensus as to whether areas where young manta rays are commonly observed constitute a nursery as defined by Huepel et al., (2007). Here I review how the present study contributes to these discussions and what benefits nurseries are likely to convey to manta rays. First, a note on terminology and clarification is necessary. In Chapters 3 and 4 the term ‘immature’ is used to collectively refer all immature individuals i.e. neonates, YOY, juveniles and sub-adults, as accurate size estimates were sparse or unavailable requiring me to mostly rely on clasper size to designate immature males. Thus, while no distinctions were made between these different sub-categories of immature life-stages when applying the criteria for designating a nursery proposed by Heupel et al., (2007), below I outline how manta rays provide a unique case compared to other elasmobranchs 195 and it might be appropriate to broaden these criteria to include other immature classes alongside neonates and YOY. Additionally, I outline how the absence of absolute age class separation for manta rays might not necessarily preclude the designation of nurseries for manta rays (Beck et al., 2001) as competition between individuals of different age classes is likely low or non-existent. Thus, the criteria currently used to designate nurseries might need to be adjusted to be context specific within species (Heupel et al., 2018), especially when applying them to an ecologically diverse group of animals as is the elasmobranchs. Proposed nurseries appear to convey benefits to immature manta rays beyond the YOY mark. The low fecundity of manta rays means that there will be few new additions through births each year i.e. low numbers of neonates and YOY would be expected (Stevens, 2016). In the Maldives, growth rates for M. alfredi were estimated to be 11 cm per year indicating that it will take close to a decade for a neonate (130 – 150 cm in disc width; Kashiwagi 2014; Marshall & Bennett 2010a) or YOY (160 cm; Stevens, 2016) to reach the size of maturity (260 – 300 cm for males and 320 – 390 cm for females; Stewart et al., 2018a). Thus, despite manta rays being born at a larger size relative to adults compared to most other elasmobranch species (Carrier et al., 2004), manta rays likely will benefit from nurseries for far longer than the one-year period cut-off designating a YOY as proposed by Heupel et al., (2007). For example, in Nusa Penida immature individuals (including estimated size classes that overlap with newborns and YOY) used the proposed nursery Manta Bay (MB) repeatedly for up to five years before being sighted elsewhere (Chapter 3). Further, lagged identification rates (LIRs) for immature individuals at MB were higher than those of mature individuals for the first two years of sightings indicating that while there is an overlap in habitat use by immature individuals and adults, immature individuals have a higher site affinity. CHAPTER 6. GENERAL DISCUSSION 196 Proposed nurseries appear to convey a nutritional benefit to manta rays (M. alfredi). The sites where immature individuals had higher site affinities (e.g. MB), or were more often sighted (e.g. MB and Cauldron - CL), were locations where they were observed foraging (Chapters 3 and 4). These observations are similar to those in the Maldives (Stevens, 2016). Foraging biased behavior by immature individuals in commonly used sites aligns with one of the proposed functions for shark nurseries – easy access to food (Heupel et al., 2007; Martins et al., 2018). While this proposed benefit of nurseries does not appear to apply to all shark (Lowe, 2002) and ray species (Davy et al., 2015), the observation that immature manta rays are more commonly observed foraging than any other behavior in the sites where they display higher affinity (Chapters 3 and 4; McCauley et al., 2014; Stevens, 2016) suggests that, at least for M. alfredi, these proposed nurseries do offer a nutritional benefit. Further, despite their large size at birth, immature individuals will have a lower swimming efficiency than adults, and thus shorter daily travel distances (Nøttestad et al., 1999; Parsons, 1990; Peel et al., 2019a). Immature individuals will also be less experienced in offshore foraging compared to adults and, as a result, their ability to forage offshore or in deeper, colder waters is likely to be restricted (Peel et al., 2019a). Thus, perhaps rather than being abundant in prey, nurseries for manta rays might instead offer sufficient foraging opportunities for smaller individuals, which would have lower caloric demands, until such a time as they have grown to a size necessitating further/more frequent offshore foraging excursions (Carrier et al., 2004). In addition to providing a nutritional benefit, nurseries might provide other benefits to developing young that reduce the time they spend in more vulnerable size classes (Heupel et al., 2007). For ectothermic organisms, like most elasmobranchs, nurseries might allow for easier thermoregulation by providing optimal temperatures that are thought to aid in growth and digestion (Papastamatiou et al., 2015). For example, 197 some species of sharks change nursery locations throughout the year in response to the presence of more temperate environmental conditions (Castro, 1993), suggesting that these habitats might offer some benefits for thermoregulation. Further, as observed for M. birostris, the proximity to pelagic food sources is likely also an important component of potential nurseries for mobulids, even when no foraging activities were directly observed in the proposed nurseries (Stewart et al., 2018b). In this case, perhaps the nurseries provide manta rays the opportunity to bask in warmer water after deeper, colder water excursions (Thorrold et al., 2014). The life-history and foraging ecology parameters of manta rays do not appear to prevent adults and younger age classes to inhabit proposed nurseries as competition between age class is likely low or non-existent. However, intra-specific competition for resources is found in many other elasmobranch species (Carrier et al., 2004) between similar and consecutive size classes. In these cases, the spatial separation between young and adults that nurseries tend to offer would provide an advantage to the young by limiting size-based intra-specific competition (Beck et al., 2001; Heupel et al., 2007). Whether intraspecific resource competition occurs with filter-feeders like manta rays, that appear to meet their energy demands through a mosaic diet of pelagic (Armstrong et al., 2016; Peel et al., 2019b), mesopelagic (Braun et al., 2014; Couturier et al., 2013; Jaine et al., 2014) and demersal zooplankton (Couturier et al., 2013; McCauley et al., 2014) is yet to be directly explored. However, the lower caloric requirements of smaller individuals likely allow them to be sustained when prey types/densities are too poor to sustain larger individuals (Armstrong et al., 2016; Peel et al., 2019b). Larger individuals would need to forage elsewhere to meet their higher energy demands, thus removing them as a source of competition to younger age classes. Exceptions, where the young and larger adults are foraging simultaneously, would occur during exceptionally prey dense occasions when CHAPTER 6. GENERAL DISCUSSION 198 prey would likely be ample anyhow. The densities of rays at feeding aggregation sites i.e. the potential for intraspecific resource competition, is likely also a factor in this dynamic (e.g. Hines et al., 1997; McAuley et al., 2007), highlighting the need for manta ray abundance estimates for these aggregations, as well as characterization of the plankton communities and prey densities during feeding events (Armstrong et al., 2016; Rohner et al., 2015). These data would allow the relationship between size and age class feeding individuals with prey availability to be examined. For some species, nurseries provide a refuge from adults or other predators while young are small and vulnerable (Cerutti-Pereyr et al., 2014; Heupel et al., 2007; 2018; Martins et al., 2018). Large sharks and orcas are known to prey on adult manta rays (Couturier et al., 2012; Marshall et al., 2010) indicating that even though manta ray neonates are large compared to many other elasmobranch species, they would not be outside the gape size of their predators. If anything, predatory attacks on smaller manta rays might more often result in mortality than those on larger individuals (Stevens, 2016) as their swimming efficiency is lower and their smaller size means strikes are more likely to result in damage to critical internal organs. Given that smaller individuals are less commonly observed with predatory wounds (Deakos et al., 2011; Marshall & Bennett, 2010), might indicate that younger individuals have higher mortality rates or that potential nurseries do offer protection from predatory interactions for manta rays (Stewart et al., 2018b). The latter fits with the idea (Heupel et al., 2007) that species that are less fecund and are required to have high survival rates are likely to have nurseries that protect their young against predation. Further, especially with shark species, nurseries offer a refuge from intra-specific predation (i.e. cannibalism; Carrier et al., 2004). For planktivores like mobulids, intra-specific predation is not a consideration. In that respect, proposed nurseries can be shared by all age classes without a disadvantage to the young. 199 Thus age class separation is likely less important as a feature of mobulid nurseries. As documented in Chapter 3, when males mature, they tend to move to habitats used primarily by adults and distinct from those used by immature individuals, but return to the proposed nurseries periodically. Thus, even though there is overlap in habitat use between different age classes of the same species, manta rays could still be considered to have nurseries. More detailed investigations using acoustic telemetry would be beneficial in confirming differences in habitat use between age classes (Heupel et al., 2018). In closing, it is generally excepted that areas designated as nurseries are important for the survival of elasmobranchs (Heupel et al., 2007; 2018) and many other marine species (Beck et al., 2001; Dahlgren et al., 2006; Nagelkerken et al., 2015). However, the general premise of a nursery (i.e. distinct habitats that disproportionally assist the growth and maturation of young) likely apply to other distinct habitats that currently do not meet the proposed nursery criteria exactly (Heupel et al., 2007). Ultimately, whether manta rays are accepted to have nurseries, as the current literature has defined them, might not necessarily translate to substantial conservation benefits (Heupel et al., 2018). However, conservation benefits will occur so long as the spatially limited, proposed nurseries are recognized as being important within the scope of the species’ large home ranges, and are thus prioritized for management (Heupel et al., 2018). This does not discount the conservation significance of habitats that are predominantly used by adults, especially for reproductive behavior, like the ones identified by this study (Chapters 3 and 4). 6.1.2 Citizen Science Wins and Shortcomings The long-term data collection and monitoring needed to adequately monitor the health and stability of threatened marine megafauna populations is very costly, yet necessary for successful conservation programs (Trave et al., 2017). Thus, citizen science CHAPTER 6. GENERAL DISCUSSION 200 based data collection through organized initiatives like the Wildbooks (e.g. Manta Matcher for manta rays) are proving to be valuable tools to bolster datasets and provide additional information on larger areas than what could be achieve by independent researchers alone (e.g. present study, Couturier et al., 2014; Germanov & Marshall, 2014; Jaine et al., 2012; Norman et al., 2017, Pierce et al., 2018; Robinson et al., 2016; Rohner et al., 2013). The opportunities should be maximized going forward through developing structured and coordinated programs with the tourism sector following the examples from Ningaloo Reef, Australia (Mau, 2008) for whale sharks and Raja Ampat, Indonesia for manta rays (Kasmidi & Gunadharma, 2017). The citizen science data collection initiative in the Nusa Penida MPA and the Komodo NP has been invaluable for providing clear evidence of site affinity, long-term site use, and variability in site use by sex, age class and behavior. However, there were initial challenges to encourage participation by the public during the early years (2012 – 2013) of the programs, resulting in time delays from the initial launch of the program to the time that a there was a substantial uptake of the initiative by the local dive communities. Further, during the years (2015 – 2016) where no on-site manager was present to coordinate, inform and encourage citizen science contributions, participation declined as reflected by fewer survey days and sighting records. Thus, good oversight is needed to ensure the continuity of the programs over time and for sustaining effective participation by communities. Leveraging the increased public recognition that manta rays were gaining as a flagship species of conservation importance, I found that continued feedback to communities regarding preliminary findings was successful in sustaining the interest in the project and engage and empower the public (Bonney et al., 2009; Miller-Rushing et al., 2012). This was achieved by communicating broader findings and interesting anecdotes about individual manta rays through public presentations, info- 201 graphs, newsletters, and through the online database itself (e.g. a data submitter is automatically notified via email if their sighting has resulted in a positive or new identification). Going forward, citizen science data collection should be continued as it has the potential to serve as an important tool for longitudinal population monitoring. 6.1.3 Directions for Future Manta Ray Research There are a number of key outstanding research aims that are necessary to inform manta ray management and conservation strategies in Indonesia. Foremost and of highest priority is calculating abundance and survival estimates (e.g. Couturier et al., 2014; Tyne et al., 2016) for the populations that inhabit Nusa Penida MPA, Komodo National Park, Raja Ampat and those in other lesser studied locations. These metrics are necessary to support all future research and monitoring the success of management strategies put in place to safeguard and improve population health. Secondly, gaining a better understanding of the reproductive biology (e.g. Deakos, 2012; Marshall & Bennett, 2010; Rambahiniarison et al., 2018) of manta populations is an important metric that ties in with population health. Thirdly, a better understanding of the environmental drivers that dictate manta ray site use, movement and reproduction cycles are necessary (e.g. Barr & Abelson, 2019; Dewar et al., 2008; Jaine et al., 2012; Stevens, 2016). Understanding the influence of the environment is especially pertinent now to help predict population responses to climate change associated shifts. Additionally, increased research into the non-direct anthropogenic threats to manta rays is also necessary, especially the impacts of tourism, microplastic ingestion and small-scale fisheries. To achieve the outstanding primary aim for monitoring manta ray population health within Indonesia, estimating population abundance and survival, requires rigorous data collection. Standardized systematic survey methods (Couturier et al., 2014; Tyne et CHAPTER 6. GENERAL DISCUSSION 202 al., 2016) rather than opportunistic surveys, as are generally those collected through citizen science (Chapters 3 and 4), are required for the capture-recapture data collection that underpins population estimates. Ideally, the survey design will consist of standard surveys (i.e. same length, time of day, dive profile), at the same time of year for the same locations, regardless if animals are present on all surveys (Couturier et al., 2014; Tyne et al., 2014; Tyne et al., 2016). However, rigorous study designs can be costly. One potential solution that allows for fewer survey designs is to marginally lower the thresholds of sampling power precision (e.g. from 95% to 80%) that are used to detect changes in abundance (Tyne et al., 2016). This solution decreases study costs while still being able to monitor changes in abundance within relatively similar time frames (Tyne et al., 2016). Once suitable data is available, various capture-recapture models can be implemented in the program MARK (White & Burnham, 1999). Several models could then be compared that will each have different assumptions depending on ‘open’ or ‘closed’ systems and whether individuals will have equal probabilities of recapture (Kitchen-Wheeler et al., 2001; Tyne et al., 2014). The suite of POPAN models available (Schwartz & Arnason, 1996), based on Jolly-Seber models (Jolly, 1965; Seber, 1965), provide super population sizes (N), entry probabilities, survival rates, and capture probabilities. The models were used recently to test for differences in survival probabilities between melanistic (black color morph) and ‘normal’ manta rays within the Indo-Pacific region, including populations in Nusa Penida MPA and Komodo NP (Venables et al., 2019). While this study did not focus on estimates of abundance, population closure tests revealed that open population models would best apply to these populations. Further, as shown in Chapters 3 and 4 of this thesis, for both the Nusa Penida MPA and Komodo NP, manta ray affinities varied among sites. Further, there was an 203 uneven interchange between sites that, in some cases, was influenced by the sex and maturity status of individuals. These variances in site use will impact the ability to assume equal recapture probabilities for all individuals within these populations. Thus, careful consideration is needed to determine how to best estimate abundance in a way that allows for heterogeneity in recapture probabilities. Pollock’s robust design (Pollock et al., 1990), used previously to estimate abundance and survival of manta rays aggregating at Lady Eliot Island, Great Barrier Reef, Australia (Couturier et al., 2014), could be considered here, as this model allows for temporary emigration and heterogeneity in capture probabilities. Lastly, newer approaches using Bayesian binomial mixture models that provide results in spatially referenced point count data could also be tested (Wu et al., 2015). These models have the benefit of being simpler to execute than traditional capture-recapture studies as they are based on animal counts rather than recaptures. The additional research aims for Indonesia’s manta ray populations include gaining a better understanding of their reproductive biology and in-depth assessments of environmental drivers of abundance (discussed in greater detail in Chapters 3 and 4). Functionally, these research aims can be accomplished in tandem with abundance and survival estimates by using a study design that collects the required field data in tandem, thereby reducing study costs. Suggested research methods to investigate the potential impacts of tourism on manta rays are outlined in Chapter 4 (Section 4.5.3). A better understanding of these impacts will serve to guide tourism carrying capacity evaluations and conservation zone management (Section 6.3.1). Further, as discussed in the next section (Section 6.2.2), understanding the implications of plastic ingestion to manta rays, whale sharks, and other large filter-feeders is an important area for future research. Lastly, evaluation of the continuing threats from fisheries and the social implications that influence the success of these conservation measures are also required (discussed in Section 6.3.2). CHAPTER 6. GENERAL DISCUSSION 204 6.2 PLASTIC POLLUTION IN FEEDING GROUNDS (CHAPTER 2 AND 5): The threat of plastic pollution to many marine species (Gall & Thompson, 2015; Worm et al., 2017) is clear. Filter-feeding marine megafauna are highlighted in Chapter 2 as being particularly susceptible to microplastic ingestion through their feeding strategies (Motta et al., 2010; Paig-Tran et al., 2013) and extensive habitat range overlap with microplastic hotspots (Eriksen et al., 2014; van Sebille et al., 2015). The implications of plastic ingestion, particularly the exposure to toxic pollutants (Andrady, 2011; Teuten et al., 2009) with known endocrine disrupting properties (Rochman et al., 2014), are considered in the light of the potential for population-wide impacts for megafauna, many being long-lived species (Dulvy et al., 2014; Lockyer, 1984; Stevens, 2007), some with low reproductive outputs (Deakos, 2012; Marshall & Bennett, 2010) and the potential to offload pollutants from mothers to offspring i.e. maternal offloading (Lyons et al., 2013a; 2013b; 2014; Mull et al., 2013; Ross et al., 2000; Yordy et al., 2010). Chapter 2 reviewed current knowledge on the implications of plastic contamination on filter-feeders and their feeding grounds, which in 2018 was limited to fin whales and basking sharks in the Mediterranean (Fossi et al., 2014; 2016) and fin whales and whale sharks in the Sea of Cortez (Fossi et al., 2016; 2017). Estimates ranged from the daily ingestion of 100 pieces for whale sharks, just under 1000 pieces for basking sharks to 1000s of pieces for fin whales. These three species all tested positive for contamination with plastic associated pollutants, heightening concerns about the implications of plastic pollution in their feeding grounds. However, knowledge gaps remain and future research avenues include establishing the direct links between microplastic ingestion and plastic associated pollutant exposure, the levels of contamination that arise from plastic ingestion itself, and the lethal or sub-lethal effects of this contamination (see section 6.2.2 and Appendix I). Manta rays and whale sharks, as 205 filter-feeders, are highlighted as being vulnerable to plastic ingestion. Further, these species form the basis of local tourism economies in developing countries, like in Indonesia, and as such are prioritized for conservation. Thus, manta rays, whale sharks and other large filter-feeders, are ideal flagship species to raise awareness about plastic pollution and other issues of marine health, and garner the cooperation of local communities in effort to improve marine stewardship (Bowen-Jones & Entwistle, 2002). Chapter 5 provides empirical abundance estimates of plastic debris, including microplastics, for three different megafauna feeding grounds (two manta ray and one whale shark) within Indonesia. The estimates fall within the range of recent broad-scale modelled estimates (Eriksen et al., 2014; van Sebille et al., 2015). There were regional differences in plastic abundance with higher estimates generally coinciding with regions of higher population density and river outputs. Plastic abundance estimates in Java were lower than expected for a densely populated region nearby to several large river outputs. Further investigation into the possibility of increased vertical migration of plastic debris and plastic debris accumulation in bottom sediments are warranted there. Further, seasonal trends were evident, with higher abundance during the wet season than the dry, particularly for the Nusa Penida MPA sites. Seasonal and regional trends in plastic abundance underline the importance of fine-scale and temporally defined research in oceanographically complex locations, as well as, targeted mitigation measures (see below). Further, in situ investigations in megafauna feeding grounds are necessary to achieve realistic plastic ingestion estimates and better quantify risks to different populations. Plastic ingestion by manta rays and whale sharks has been proposed (Chapter 2; Fossi et al., 2014; 2017); however, estimates of plastic ingestion rates and confirmation of ingestion are limited. Local plastic abundance estimates in feeding grounds allowed for CHAPTER 6. GENERAL DISCUSSION 206 the first estimates of hourly plastic ingestion rates for filter-feeding megafauna within the Coral Triangle region, a globally significant microplastic hotspot (Chapter 2). Previous reports show that whale sharks do ingest (Abreo et al., 2018; Haetrakul et al., 2009; Sampaio et al., 2018) and are capable of egesting plastic (Donati et al., 2019). The present study confirms that this is also the case for manta rays (Chapter 5). Observing plastic particles ranging from 0.5 to 30 mm in manta ray egested material also confirms that manta rays ingest larger plastic debris, indicating that they are not efficiently selecting out larger plastics and are impacted by larger debris in feeding grounds, as well as microplastics. Although the plastic pieces observed in the egested material indicate that at least some ingested plastic is able to pass through the body, it also raises concerns that some plastic might remain for a prolonged period, increasing the risk of digestive tract obstruction, dietary dilution, and desorption of pollutants. Continued opportunistic investigations of egested material and digestive tracts of stranded, naturally deceased, and accidentally caught manta rays and whale sharks might provide further evidence of the extent of plastic ingestion and accumulation within the digestive tracts of these filter-feeders. 6.2.1 Debates on the Extent of Risks Posed by Microplastics The debate centered around the potential gap between the perceived versus the actual magnitude of the microplastics problem (Backhaus & Wagner, 2018; Burton, 2017; Hale, 2018; Rist et al., 2018) warrants discussion. The recent increase and expansion in research on plastic pollution (Barboza & Gimenez, 2015) and the growing public attention to the issue (Burton, 2017; Hale, 2018; Rist et al., 2018) have led to debates in the field. The main criticism towards much of the literature is centered around Paracelsus’ theory of “the dose determines the poison” (Backhaus & Wagner, 2018; Burton, 2017). 207 The issue being that ‘real-world’ doses are not used when assessing the risks of microplastics to individuals, populations and ecosystems (Backhaus & Wagner, 2018; Burton, 2017; Koelmans et al., 2017). Many studies assume that plastic ingestion itself is a risk, without considering whether the in situ exposure levels to organisms will lead to adverse effects (Backhaus & Wagner, 2018; Burton, 2017; Koelmans et al., 2017; Lenz et al., 2016). It is argued that, especially when it comes to direct impacts on humans, the perceived negative effects from ingesting food and beverages tainted with microplastics might be negligible compared to contamination resulting from everyday interactions with plastic items, such as the food packaging, and airborne plastic contamination (Rist et al., 2018). Further, although plastics are capable of adsorbing toxic pollutants from the surrounding environment (Cole et al., 2011; Ogata et al., 2009; Rios et al., 2007, 2010), and transferring them to biota (reviewed in Rochman, 2016; Teuten et al., 2009; Worm et al., 2017), whether microplastics transfer these pollutants to biota in substantially higher levels than natural exposure from water or prey is challenged (Koelamns et al., 2016) through thermodynamic (Gouin et al., 2011) and bioaccumulation modelling (Koelmans et al., 2013) in experimental systems. Given that resources, public attention and willpower have their limits, the pitfalls of making assumptions about the toxicity of plastics are that any overestimation of risks might potentially skew the importance of studying microplastics and creating policies aimed at limiting them, thus diverting resources away from other equally/more pressing environmental concerns (Backhaus & Wagner, 2018; Burton, 2017; Koelmans et al., 2017). However, as this field is still in its infancy, and many uncertainties remain, to answer the ultimate question of the true impacts of microplastics (and nanoplastics) on organisms will require continued research focus (Koelmans et al., 2016). To that end, Koelman et al. (2016) acknowledge the value of studying plastic, pollutants and exposure of biota under real world scenarios (e.g. Fossi CHAPTER 6. GENERAL DISCUSSION 208 et al., 2016) as these studies “represent the ultimate reality of nature”. However, they caution that with these studies alone, it is difficult to discern the exact cause of pollutant bioaccumulation, as there are many dynamic mechanisms in simultaneous effect. Thus, to be able to evaluate the proposed implications associated with plastic ingestion will require long-term studies with a variety of organisms that consider local environmental conditions (Koelmans et al., 2016). Risk management solutions for microplastics often call on the precautionary principle/approach (UNESCO, 2005), which aims to reduce the costs associated with inaction (Backhaus & Wagner, 2018). However, there are noteworthy strategies from the evidence-based approach, which should be incorporated to reduce inefficiencies in risk management measures. These strategies include identifying and filling knowledge gaps and conducting real-world risk assessments (Backhaus & Wagner, 2018; Koelmans et al., 2017). The knowledge gaps as they pertain to many filter-feeding megafauna, and how to fill them, are discussed in Section 6.2.2. Lastly, it is worth highlighting that the risks associated with the greater issue of plastic overconsumption, poor waste practices and marine debris are larger than the potential toxicology risks that microplastics pose. The plethora of issues that marine plastics cause are well described in a number of syntheses (Alimba & Faggio 2019; Browne et al., 2015; Derraik, 2002; Gall & Thompson, 2015; Gregory et al., 2009; UNEP, 2016) and include: ingestion of larger plastics, entanglement, rafting for invasive species and pathogens, habitat alteration, economic costs associated with cleanups and lost revenues from businesses like tourism and shipping. Plastic use is projected to increase (Geyer et al., 2017), and without intervention, this will result in more plastic waste in the environment (Jambeck et al., 2015). Considering all of these well documented issues with marine plastics, in combination with those documented and 209 postulated to be associated with microplastics, I stand by the ideas presented in Chapter 2 that identify filter-feeding megafauna as ideal flagship species to mobilize the public and policy maker will to prioritize tackling marine plastics in heavily afflicted areas. In the present day, where environmental problems are on a global scale, mobilizing a global effort to tackle the plastics issue might serve as the ‘gateway’ (Backhaus & Wagner, 2018) to foster collaboration across stakeholders on other pressing issues such as climate change and overfishing (Parsons et al., 2014). 6.2.2 Directions for Future Microplastics Research The next steps for future microplastic (and nanoplastic) research, as it pertains to filter-feeding megafauna, outlined below were considered in light of the arguments outlined above. Compared to entanglement threats, the implications of microplastic (especially, nanoplastic) ingestion and associated sub-lethal threats are not visually obvious and their effects are likely cumulative over long time periods and not readily detectable. These challenges are reflected in the literature, in that potential impacts are often quoted, but definitive conclusions on the magnitude of the risks are rarely, if ever, specified. Indeed, many studies, including those in this thesis, have reported plastic ingestion in a myriad of species (reviewed in Gall & Tompson, 2015; Gregory, 2009; Worm et al., 2017), and/or calculated theoretical exposure levels to microplastics based on in situ observations (e.g. Fossi et al., 2017), without being able to definitively evaluate the risks associated with exposure to the individuals and populations (i.e. does ingesting plastic cause mortality or other sub-lethal impacts). In some of the better studied model animals (e.g. seabirds and sea turtles) it is clear that microplastic ingestion has real-word consequents (reviewed by Worm et al., 2017; e.g. Pierce et al., 2004; Santos et al., 2015; CHAPTER 6. GENERAL DISCUSSION 210 Stamper et al., 2009; Wilcox et al., 2015). Future research needs to assess the magnitude of risks regarding impacts on biota with better study designs. Separating the direct particle effects associated with microplastic ingestion from the toxicology effects of plastic associated pollutants (i.e. separating the physical from the chemical effects; reviewed in Alimba & Faggio, 2019; Rist et al., 2018) is important for risk assessments (Koelmans et al., 2017). However, functionally this separation is likely impractical for megafauna in the wild. Plastic that is ingested by these megafauna would have spent time in the environment and would not be devoid of additives and pollutants, making it difficult to separate the physical from the chemical effects. Further, at least in experimental systems translocation of micro and nano size plastics into animal tissues (Browne et al., 2008) were documented to cause a myriad of negative outcomes (i.e. oxidative stress, immunological and genetic responses, endocrine and reproductive disruption; reviewed by Alimba & Faggio, 2019; Rist et al., 2018) that would present as/similar to chemical effects until more specific information is available. Additionally, real world microplastics would originate from variety of sources and contain particles of heterogeneous size, type, and polymer composition making standardization across specimens and studies for physical effects complex. Nevertheless, cadaver access through strandings and accidental mortality of filter-feeding megafauna provide the potential to begin to unravel the effects of environmentally relevant levels of exposure to plastics and associated pollutants (discussed below; e.g. Fossi et al., 2014). Although relatively rare to scientists, when they present, these opportunities shed light on plastic caused mechanical damages (e.g. Haetrakul et al., 2009) and obstructions (e.g. Pierce et al., 2004; Santos et al., 2015; Stamper et al., 2009); effects on gut health (Fackelmann & Sommer, 2019), and translocation of plastic into other tissues (Browne et al., 2008). 211 To estimate real world risks, in addition to estimating plastic abundance in feeding aggregation areas and ingestion rates for biota, the in situ linkages between plastic, additives, and pollutants need to be clarified (Chapter 2; Appendix I; Koelmans et al., 2016). In addition, the ‘background’ environmental contamination by pollutants linked to plastic exposure needs to be assessed to establish if microplastics are a significant source of these pollutants, which several studies now contest (reviewed Koelmans et al., 2016; 2017; Lohmann, 2017). To establish the mechanisms of biota contamination, different potential sources will need to be tested for the contaminating pollutants (Koelmans et al., 2017), as well as the relative ratio of ingestion of each. For example, biopsy tissue samples from filter-feeders (Fossi et al., 2016; 2017), their common food sources (e.g. Fossi et al., 2014), microplastics (e.g. Ogata et al., 2009), and other potentially ingested matter (water, organic materials, etc.) sampled from a known habitat could be simultaneously analyzed for plastic additives and pollutants. The ratios of each source in the water (e.g. plastic vs plankton, biological material, see Chapter 5; Collignon et al., 2012) where feeding takes place can be used to calibrate bioaccumulation models. Understanding the dynamics of pollutant uptake by megafauna from the various potential sources, and the levels of exposure, will allow us to better gauge the extent of risk to megafauna and ecosystems as a whole (Backhaus & Wagner, 2018). There is a spatial component to this risk with areas with high plastic pollution to prey ratios, and/or plastics that have high levels of pollutants relative to other local sources (i.e. higher than equilibrium), posing a higher risk than others (Chen et al., 2018). Even if microplastics prove not be a significant source of pollutants, it will not take away from other ways their ingestion might be detrimental to the health of fauna (e.g. nutritional dilution, mechanical damage, toxicity of particle effects, etc.; Koelmans et al., 2017; Lohmann, 2017). Further, assessing the levels and types of pollutants that threatened species are exposed to CHAPTER 6. GENERAL DISCUSSION 212 throughout their range, regardless if microplastics play a substantial vector role, are also important investigations to better understand this sub-lethal threat (Appendix I). Thus far, toxicology analysis on biopsy collected skin samples from free swimming whale sharks in La Paz, Mexico have confirmed plastic associated pollutants, verifying the feasibility of this methodology for testing pollutant exposure in elasmobranchs (Fossi et al., 2017). Non-lethal sampling (i.e. biopsy) is the more ethically sound method to investigate pollutant exposure on threatened species (Fossi et al., 2000). However, as above, access to opportunistically sourced cadavers would allow for the testing of pollutants and chemical effects across tissue types (Fossi et al., 2014; Lyons et al., 2013a; 2013b; 2014; Mull et al., 2012; 2013). These sorts of validations are important for understanding the relationship between pollutants accumulated in the skin in relation to other organs that likely accumulate pollutants to a much higher degree (Mull et al., 2012). Going forward these metrics might allow researchers to translate pollutant levels found in the skin samples (Fossi et al., 2016; 2017) to overall bodily pollutant burdens. Monitoring long-lived, wide-ranging megafauna provides an opportunity to examine responses to microplastics and pollutants over temporal and spatial scales in effort to establish risks to populations (Browne et al., 2015). This could be achieved with repeated testing for pollutants via biopsy samplings of the same individuals over time, across populations and pollution gradients (see Appendix I). How pollutant exposure translates to potential risks could be first investigated through associated biomarker responses to pollutants on the individual level (see Chapter 5; Barboza et al., 2018; Dove et al., 2012; Fossi et al., 2017; Pillet & Nicolas, 2005). Biomarker responses are indicative of underlying disease and malfunctions in regular bodily processes (Pillet & Nicolas, 2005). However, appropriate biomarkers for pollutant exposure that are indicative of diseased states will need to first be identified (e.g. Dove et al., 2012; Fossi et 213 al., 2017; Lehnert et al., 2019). Further, while it is difficult to identify which pollutant the biomarker response is attributed to (Lehnert et al., 2019), an understanding of pollution sources, as well as the locations where individuals are most likely to be exposed to pollution will assist in risk assessments (e.g. Wilson et al., 2012). In well-studied populations, the exposure level of an individual could be compared with its health metrics (e.g. biomarkers, body condition, reproductive output, and survival) and to those indicative of population health (e.g. decreasing population trends) (Appendix I). These health metrics could then be compared across pollution gradients and time to establish real world risks. Further, comparing the pollution burdens of younger individuals (e.g. manta rays observed at Manta Bay) to those of adults might indicate whether the young are exposed to higher pollutant loads through the process of maternal transfer (Lyons et al., 2013a; 2013b; 2014). These modes of intergenerational transfer are important to understand as they might enable pollutants to persist in populations, despite any immediate mitigation efforts (Mull et al., 2013; Murphy et al., 2015). Using large filter-feeders as sentinels to monitor the presence of plastic pollution in localized environments was suggested in Chapter 2. This monitoring method has advantages over direct water sampling for microplastics. Microplastic abundance in marine environments can change drastically with seasons and even day to day or within sample replicates (E. Germanov, pers. obs.), making it difficult to capture the full extent of the level of pollution without repeated sampling over time. As a result, these investigations can become overall costly and time consuming. Further, the complex toxicity of microplastics and associated pollutants, coupled with the limitations on the size of particles captured through the standard sampling methods, makes establishing microplastic abundance in environments just one of the steps in the sequence needed to establish the risks to biota and ecosystems. Similar to how seabirds are used to monitor CHAPTER 6. GENERAL DISCUSSION 214 for microplastic presence and composition (Ivar do Sol & Costa, 2014), and cetaceans and large-bodied fish for POPs (Kucklick et al., 2011; Ueno et al., 2004), long-lived filter-feeders can provide an indicator of the general level of plastic associated pollutants (via biopsies; e.g. Fossi et al., 2014; 2017) and microplastics (via egested materials; e.g. Chapter 5; Kamal et al., 2019) throughout their home-range. Thus, evaluating the actual exposure to plastic associated pollutants and the health indices for filter-feeders in these locations will give a better overall indicator of the levels of pollution exposure (and associated risks), in a way that sampling for microplastics (even repeatedly) in a limited area cannot establish. Further, where long-lived filter-feeders aggregate to forage is likely indicative of the health at the base of the food chain in this general area (Armstrong et al., 2016). Thus, by assessing pollutant exposure to filter-feeders, it might be possible to make general ecosystem risk evaluations. In this way scientists and managers can better evaluate plastic pollution not only in terms of the biological risks to the model organisms, and to the lucrative tourism industries they might support, but also to commercial or subsistence fisheries in the region. Studying the effects of microplastics on threatened megafauna certainly possess challenges. As, discussed above, access to cadavers will help to move this field forward and establish baselines for non-lethal sampling; however, for ethical and legal reasons cadavers should only be accessed from natural deaths and from accidental catches and mortalities, making then inherently rare to access. As such, stranding networks, as those set up for marine mammals (e.g. www.whalestrandingindonesia.com), and collaborations with fisheries for access to bycatch, would be beneficial in that respect. In addition to this, working in low to middle income (LMI) countries, often limits access to laboratory facilities needed for the required analyses, which often involve highly specialized, sophisticated and costly techniques (e.g. sensitive mass spectrometry analysis 215 capabilities). The alternative, to export samples to well-established laboratories in countries other than those where the samples were collected, is often faced with impeding cross-border sample export restrictions and difficulties with sample preservation during transport, especially when frozen tissue is required. To improve the feasibility of these types of studies in locations where they are most needed will require significant capacity building of existing local institutions and/or improved sample export and transport pathways. Finally, advance knowledge of the focal populations, such as base-line estimates, aggregation area uses and regional connectivity, are necessary when analyzing population wide implications on megafauna with large home ranges. Despite these challenges, in order to better understand the scale of the microplastic issue to filter-feeding megafauna in an effort to better protect them, we encourage prioritizing research on threatened species found in regions with high levels of microplastic pollution (Chapter 2). Increasing scientific capacity within the countries where the research takes place is key to improving local knowledge, research outcome visibility and uptake of research findings by local managers (Bennett et al., 2017; Minasny & Fiantis, 2018). Foreign researchers working in LMI countries should thus consider including a capacity building component alongside their primary research (Christie et al., 2017). To that aim, throughout the duration of the present study, seven undergraduate students or recent marine science graduates were trained and participated in the research on microplastics and manta rays. Future research planned on this topic aims to continue baseline data collection as part of a long-term collaboration with Indonesian higher learning institutes and conservation organizations. CHAPTER 6. GENERAL DISCUSSION 216 6.3 IMPLICATIONS FOR MANAGEMENT The current study has enhanced the understanding of manta ray habitat use within Indonesia. Areas where individuals aggregate within the lesser Sunda Island region are now better characterized and there is an increased understanding of how and when these habitats are used and by which demographic groups. This study has also highlighted some of the pressures to these populations. Based on the findings in this study and those from others, recommendations for management to enhance the conservation of manta populations and habitats are presented. Above all, monitoring of the successes (and lack thereof) of management strategies need to be ongoing and the strategies themselves adapted to include new relevant information (Abecasis et al., 2015; Rigby et al., 2019; Tyne et al., 2014). 6.3.1 Tourism Regulations As outlined by Venables et al. (2016), we advocate adopting a precautionary approach for the conservation of manta rays (UNESCO, 2005). The precautionary approach dictates that if harm to species and their habitats is scientifically plausible, but not yet proven, pre-emptive action to protect the species and their habitats would be the recommended action to ensure the viability of their population. We echo Venables et al. (2016) and recommend that codes of conduct for diving and snorkeling with manta rays become mandatory in Indonesia and elsewhere, akin to mandatory codes of conduct for whale shark interactions in Ningaloo Reef, Australia (Catlin & Jones, 2010; Mau, 2008). Science-based regulations that place a limit on tourism activities, i.e. a tourism carrying capacity, should be carried out in all areas where there is manta ray focused tourism to estimate what the acceptable number of boats/people a specific site can accommodate prior to a significant effect on the manta populations occurs (Ríos-Jara et al., 2013; 217 Zelenka & Kacetl, 2014). If at all possible, these calculations should be carried out prior to the establishment of manta ray tourism. In areas where tourism already exists, tourism levels might have already surpassed the estimated tourism carrying capacity and in these cases, management regulation might involve reducing the number of boats and swimmers allowed at one time. In the present study, both study locations exhibited marked increases in tourism over the duration of the study (Chapters 3 and 4). Science-based knowledge of how local manta populations are using critical habitats should underpin management strategies to ensure disturbance from tourism is minimized (Ashe et al., 2010). For example, if seasonal peaks in reproductive behavior are consistent as in the Nusa Penida MPA (Chapter 3) and other regions (Couturier et al., 2014; Deakos et al., 2012; Marshall & Bennett, 2010; Stevens, 2016), area-time closures or reductions in traffic could be considered for these times (Tyne et al., 2014, Setyawan et al., 2018; Stewart et al., 2018a). Important foraging areas for immature manta rays or nurseries are rare. When these areas are identified, management strategies need to adapt to increase protection for these small-scale habitats, which are likely critical for the fitness of these smaller rays. Manta rays feeding near or at the surface are liable to be struck by boat propellers, thus mandated speed limits and no-wake zones are crucial (Stewart et al., 2018a). Importantly any imposed regulations need strict enforcement and penalties for breaches of conduct (e.g. Colman, 1997; Tyne et al., 2014) to ensure uptake by local operators. It is also important to consider the benefits that tourism regulations could have on tourism itself. Less crowded locations will allow for better wildlife viewing experiences and less disturbances to wildlife. Evaluations of ecotourism experiences by visitors were negatively impacted by overcrowding (Hausmann et al., 2017). Further, the ability to experience solitude is a positive psychological component in how people perceive a place CHAPTER 6. GENERAL DISCUSSION 218 or experience (Manzo, 2005). A significant number of visitors were dissatisfied with the number of boats (23.4%) and other snorkelers (18.8%) present in the water during whale shark encounters tours in Mexico (Ziegler et al., 2012). Similarly, managing overcrowding at manta ray sites was listed as the top preferred management strategy in an economic survey of visitors to the Nusa Penida MPA (Barr & Abelson, 2017). Visitors ranked reducing the number of divers as one of the top three management strategies for manta ray conservation initiatives within the MPA and 20.7% ranked it as the number one option. Visitors also preferred management strategies to include patrolling to enforce regulations, making codes of conduct for manta ray encounters available and having on-site compliance officers to ensure proper etiquette is adhered to. Based on this information, regulations of the number of people allowed in the water at one time could improve the overall satisfaction of the divers/swimmers. Thus carrying capacity estimates need to consider the participant satisfaction thresholds, as well as potential negative impacts to megafauna and the environment. Failure to include these considerations into management startegies might have substantial negative consequences for the tourism businesses. Restrictions on the number of boats and swimmers have recently been put in place for two manta ray aggregation sites in Indonesia, Karang Makassar (KM) in Komodo NP (Komodo National Park, pers. comm.) and Manta Sandy in Raja Ampat (Kasmidi & Gunadharma, 2017). The daily number of boast allowed to visit KM, an area of about 1 km2, is currently set at 32, with each boat allowed 10 swimmers, for a total of 320 people per day (Komodo National Park, pers. comm.). No specific speed limits on boats are in effect; however, as per the newest regulations, operation of boats at diving/snorkeling areas or close to marine life should be below wake speed (Komodo National Park, 2019). At Manta Sandy, Raja Ampat, an area of about 0.06 km2, that is essentially a cleaning 219 station, there can be a maximum of 20 swimmers at once. Boat access is restricted to drop off and pickup of swimmers, with speed limits of 5 – 8 knots (depending on proximity to cleaning station) in effect within an area of approximately 0.5 km2 (Kasmidi & Gunadharma, 2017). As of writing, carrying capacity regulations are yet to be created and implemented for the Nusa Penida MPA or other manta ray sites within Indonesia. Codes of Conduct have been developed for manta ray interactions (examples are available from Manta Trust, 2018; Marine Megafauna Foundation Indonesia, 2018) that some businesses in Indonesia have adopted on a volunteer basis (Diver Operators Community of Komodo, pers. comm.; Lembongan Marine Association, pers. comm.). These codes describe the appropriate distance that divers/snorkelers need to maintain from manta rays, cleaning stations and the etiquette in the water (i.e. do not approach from front, preventing crowding, no chasing, touching or riding, practicing good buoyancy, limiting flash photography). These general codes could essentially be adopted for use at all manta ray sites in Indonesia and worldwide, but should also include locally relevant information, such as boat speed restrictions and areas prohibited for boat traffic (i.e. cleaning stations, feeding areas, etc.). Conservation zones, such as Nusa Penida MPA and Komodo NP, should require tour operators to adhere to these codes (Garrud, 2016; Venables et al., 2016) and enforce compliance (Catlin & Jones, 2010; Mau, 2008). 6.3.2 Small-scale Fisheries Large-scale movements along heavily fished corridors in the Lesser Sunda region region (present study Chapters 3 and 4; Germanov & Marshall, 2014), and high site affinity to areas where small-scale and artisanal fisheries occur, means that despite manta rays officially being protected from fisheries by law, illegal targeted fishing (Booth, 2016), accidental catches and otherwise contact with fisheries continue to occur. Given CHAPTER 6. GENERAL DISCUSSION 220 the enormity, complexity and cost of the task to patrol and enforce these fishing regulations throughout Indonesia’s entire EEZ (6,159,032 km2), or even within better known movement corridors, it is not unforeseen that illegal activities might continue. Another problem, is that the reporting of accidental manta ray catches is currently not required by law in Indonesia and reliable recent data on manta ray bycatch is not available (Booth, 2016). However, historical anecdotal reports indicate that even non-targeted fisheries e.g. using trap and drift gill nets for pelagic species, have led to manta ray extirpations in a number of locations throughout Indonesia (Lewis et al., 2015) underscoring the necessity to better understand remaining catches. Mandatory bycatch reporting and landing site enumeration would assist in quantifying and monitoring this hidden threat. A potential solution for reducing mobulid bycatch could be through gear modifications, such as deterrent lighting for mobulids on gill nets (Jordan et al., 2013; V. Rosandy, Manta Watch, 2018, pers. comm.). If successful, these modifications could be mandated for use on all gill net vessels within the Lesser Sunda region and other known manta ray movement corridors. The Nusa Penida MPA and the Komodo NP are partitioned allowing for different resource uses within dedicated areas of the conservation zones. In Nusa Penida (Coral Triangle Center, 2013) specific sections of the MPA are no-take-zones where fisheries and tourism are prohibited (core zone), other sections allow for tourism use while fisheries and other extractive activities are prohibited (e.g. Manta Bay and Manta Point), and other sections allow for traditional subsistence fisheries and pelagic fishing. The Komodo NP similarly has sections where traditional fisheries and pelagic use is sanctioned, and other sections where fishing and extraction is prohibited and are dedicated for tourism use (e.g. Karang Makassar, Mawan Island and Manta Alley; Komodo National Park, 2019). Despite regulations in place prohibiting fishing in sites 221 commonly inhabited by manta rays, at least ~ 14% and 5% of manta rays sighted in the Nusa Penida MPA and Komodo NP, come in contact with fisheries, respectively (Chapters 3 and 4). Direct evidence of fishing gear contact (e.g. trailing lines, hooks or nets) and/or disfigurements that likely occurred in response to entanglements in fishing line (e.g. cephalic and pectoral fin truncations) were present in both locations; however, where these contacts actually occur is unknown. In Nusa Penida MPA, the fisheries that occur at manta ray sites are generally local subsistence fisheries and mostly pole and line fishing for tuna and jacket fish (Acipenseridae spp). Some of the manta rays frequently sighted feeding in commonly fished waters were observed to be entangled in or trailing lines, indicating that at least some of these injuries occur locally. In contrast, fishing activity directly at manta ray sites was rarely observed in Komodo NP (Chapter 4). Fisheries utilizing indiscriminate fishing gear (e.g. longlines and gillnets) that operate in the pelagic zones within Komodo NP and the greater Lesser Sunda region are likely to pose the greatest fishing gear related threat for manta rays. Manta rays are less likely to survive contact with these types of fishing gear, which might explain the lower proportion of individuals with anthropogenic injuries noted in Komodo NP relative to Nusa Penida MPA. The success of conservation programs is gauged by the ability to enforce regulations and the levels of compliance (Arias, 2015; Christie et al., 2017). In terms of achieving compliance on directed fisheries, enforcement of no-take-zones, and regulations on gear types requires that the communities that might be negatively impacted as a result of these regulations be consulted early on (Bennett et al., 2017; Christie et al., 2017; Hoyt, 2014). In interviews with fishers following the implementation of the relatively recent laws and regulations regarding manta ray and shark fisheries in Indonesia, support for these conservation regulations was, at best, mixed. Many fishers CHAPTER 6. GENERAL DISCUSSION 222 thought that the regulations were created without consideration for their livelihoods, and as a result were negatively impacted (Booth, 2016, Jaiteh et al., 2016; 2017). In regards to manta ray targeted fishing, most, but not all fishers, cited compliance with new regulations, and some interviewees indicated that annual catches might still be as high as 400 individuals in one village (Booth, 2016). Providing alternative income streams for communities impacted by fishing regulations might be one way to ensure compliance with these regulations that are likely to impact their livelihoods (Arias, 2015; Hoyt, 2014). Studies evaluating the impact of shark no-take-zones and a decreased demand for sharks on remote Indonesian coastal communities have provided some insight into the drivers of compliance and the social repercussions of abrupt removal of income sources without alternative, equally profitable, legal, long-term opportunities available (Jaiteh et al., 2016; 2017). The studies found that while the shark-specific closures had a positive effect on shark abundance (Jaiteh et al., 2016), fishers in some of these communities went on to engage in riskier (e.g. going further out to sea for longer periods and/or during unfavorable conditions), and in cases, illicit behavior (e.g. switching to more destructive fishing practices like dynamite fishing, illegal transboundary fishing, fuel and people smuggling; Jaiteh et al., 2017), or shifted their fishing focus to areas where shark fishing was not restricted (Jaiteh et al., 2016), i.e. essentially relocating the problem elsewhere. Interviews with members of these communities indicated that if sustainable livelihood options are presented there is a will to change practices and occupation; however, self-initiated alternative programs had limited to no success (Jaiteh et al., 2017). For alternative livelihood programs (e.g. mariculture and agriculture; legal and sustainable fisheries; involvement in eco-tourism and/or conservation zone management) to become viable, substantial assistance in the form of subsidies and re-training opportunities is 223 needed from federal, provincial and local governments, non-governmental organizations, local community organizations and/or partnerships with the tourism industry (Jaiteh et al., 2016; 2017). In particular, employment through tourism could be a suitable alternative income in places with developed, but growing tourism industries e.g. Nusa Penida MPA and Komodo NP. Another key component to compliance is for those effected to understand what the incentives of new regulations are e.g. improved or preserved ecosystem services for sustainable livelihoods (Arias, 2015). Failing to communicate these benefits with communities is where many of these programs have so far failed (Booth, 2015; Jaiteh et al., 2016). In the Nusa Penida MPA and the Komodo NP, both conservation zones have designated areas that allow for traditional subsistence use and pelagic fisheries (Coral Triangle Center, 2013; Komodo National Park, 2019). However, non-compliance occurs regularly (E. Germanov, pers. obs.; Lembongan Marine Association and Diver Operators Community of Komodo, pers. comm.) in designated no-take-zones, including manta ray aggregation sites. Currently, enforcement of these regulations is minimal or non-existent, although at least for Komodo NP it has been better in the past (Divers Operators Community of Komodo, pers. comm.). It is important to determine whether the benefits of no-take-zones and gear restrictions were sufficiently communicated to local fishers and if there was adequate community involvement in developing the existing management strategies that would lead to better community compliance with regulations. Further, understanding the drivers of non-compliance through interviews with fishers to identify their reasons for not complying and their responses to other initiatives would be valuable. This would include understanding 1) their knowledge and understanding of the currently existing regulations; 2) the purpose of fishing at manta sites (subsistence and/or income generation); 3) if there is willingness to relocate fishing activities to other sites where CHAPTER 6. GENERAL DISCUSSION 224 fishing is permitted; 4) if an awareness exists about manta ray habitats, behavior and their value to tourism related livelihoods; and 5) is there a willingness to stop fishing if manta rays are present near fishing activities and are likely to have contact with gears. In summary, offering communities opportunities for stable, safer jobs and adequately communicating the benefits of no-take-zones, gear restrictions and recommended modifications alongside species directed fishing bans are incentives that can work hand-in hand with enforcement of noncompliance to pave the way towards voluntary compliance (Arias, 2015). 6.3.3 Plastic Marine Debris Plastic marine debris is ubiquitous, with sources that can be hundreds or thousands of kilometers away from the focus marine environment (Reisser et al., 2013; van Sebille, 2014); however, local contributions and prevention efforts for plastic of all sizes should not be underestimated. For example, many large-sized film plastics observed in this study contained Indonesian writing and originated from locally sold consumer products (E. Germanov, pers. obs.). Further, the majority of the microplastic and other small sized plastics in the environments surveyed in this study (Chapter 5) would be considered secondary plastics (i.e. resulting from the breakdown of larger debris). Thus, mitigation strategies aimed at plastic debris in general should address pollution across the size spectrum. A number of intervention points to stem the flow of plastic into the marine environment have been identified by Worm et al. (2017). Much of these interventions are at the government/industry level and span from reducing plastic production, better product and material design and substitution to reduce waste (i.e. ‘Cradle to Cradle’, see Braungart et al., 2007), and improved waste management and litter capture. Further, to achieve meaningful reductions in single-use plastics by the public, collaborations between 225 scientists, management authorities, government and non-governmental organizations will be beneficial to educate on reducing plastic use and its appropriate disposal. Barriers to adopting bans on single-use plastics, reducing waste and improving waste management practices will also need to be identified so they can be adequately addressed and overcome. Reducing single-use plastics will require a substantial push from governing bodies to improve infrastructure and impose regulations on plastic manufacturing to reduce the amount of single-use plastic created. Large-scale infrastructure improvements, such as contained waste sites, regular pickup services for household waste, and recycling can only help prevent plastic from entering the environment, but do not reduce the amount of waste created. Furthermore, these improvements are more difficult logistically on the smaller and scattered islands of the Indonesian archipelago, and are thus less likely to be effective. In these communities it is particularly important that waste reduction is emphasized. The majority of manufactured plastic is used for packaging and other single uses, and once discarded makes up the majority of secondary microplastics released into the world’s oceans (Andrady, 2011). For example, similar to other developing nations (Stoler et al., 2012), one infrastructure improvement in Indonesia that would reduce waste is providing access to clean public drinking water. A widespread practice in Indonesia is to provide drinking water in single-use bottles or worse, sealed low grade plastic cups with straws that are difficult to recycle. Another example, are plastic bag bans or levies, which have achieved immediate reductions in plastic bag use of up to 90% (Jambeck et al., 2018; Xanthos & Walker, 2017), including in some developing countries. Encouraging the uptake of similar reduction efforts in Indonesia is likely to result in a similar reduction in plastic use (Aqil, 2019; Gokkon, 2018). Lastly, incorporating new materials, such as compostable bioplastics and polyhydroxyalkanoates (PHAs; Jain et al., CHAPTER 6. GENERAL DISCUSSION 226 2015) to substitute more difficult to avoid single-use, throwaway packaging and packing material could be incentivized by governments in the form of tax breaks to industries and consumers. Alternatively, levies could be imposed for using traditional plastic materials. Indonesia has much to improve in its basic waste management infrastructure, in both cities and rural communities. In large to mid-sized cities, sites are designated for centralized waste disposal and pickup services exist. However, these designated sites might be poorly placed nearby water sources (e.g. Benoa harbour, Bali). Lightweight plastics can easily be dispersed by the wind to water sources (Ryan et al., 2009) from these waste disposal sites. Often these sites are not up to international landfill standards and leachates from these sites pollute ground water and smoke from open burning pollutes the air (Aye & Widjaya, 2006). In rural areas where there are no designated waste management sites, waste disposal is not centralized and will enter the environment more easily through informal dumping. In these cases, contained waste management facilities are needed that are located away from coastal areas and water sources. A number of waste management solutions have been considered for solid waste e.g. collaborations with the informal sectors for recyclable solid, non-organic waste, including plastics, and composting and biogas production of organics (Aye & Widjaya, 2006; Sicular, 1992). In some developing countries a circular economy for plastic waste has been informally in effect for decades, where people create livelihoods from obtaining valuable waste and reselling it as raw materials (Aye & Widjaya, 2006; Sicular, 1992; Wilson et al., 2006). Informal recycling comes in the form of community waste banks, ‘scavengers’, and others who store, trade, sell and process the materials. In these schemes there are a number of ‘leakage’ points where waste residue is disposed in the environment (Putri et al., 2018). The quantity of plastic recycled through these schemes and the amount of leftover residue are difficult to quantify in this decentralized model. To 227 increase material recovery, reduce the amount of waste going to landfill and leaking to the environment, requires better integration between the informal recycling sector and government waste collection. For example, scavengers can be hired to work with communities to improve plastic sorting at the source, once it arrives to collection centers and/or sorting facilities. Sorted plastic is sent for processing and all non-recyclable residues are then collected and disposed of properly in municipal landfill sites as a last resort (Putri et al., 2018). These integrated models will allow for better efficiency and quantification of waste recovery, and highlight areas for future improvements. In addition to the waste generated in country, Indonesia imports waste from other countries for recycling. Much of this waste could be contaminated with scrap plastic that is unsalvageable, leaving the burden of its disposal to Indonesia (Graham, 2019). Further, most film plastics and packaging are considered low value and are often openly discarded or burned, and in the process release toxic fumes and green-houses gases (Aye & Widjaya, 2006; Wilson et al., 2006). In some regions, low value plastics are sold to be burned as fuel in food production (Graham, 2019). Thus international trade on waste needs regulations that take into consideration the capabilities of importing nations to handle the waste. Standard international guidelines on what can be exported could be considered to protect nations that lack the abilities to deal with this waste. There are recent cases of some countries (e.g. China, Philippines) refusing to accept hard to recycle, low value plastics, mandating exporting countries to have to rethink their waste management strategies (Freyatas-Tamura, 2018; Leung, 2019). This study (Chapter 5) and others (Jambeck et al., 2015; Lebreton et al., 2017; Schmidt et al., 2017) show that Indonesia, and the region as whole, would benefit from focused prevention and capture at the source, prior to the onset of the wet season. Mitigation strategies could include river bed and coastal cleanups. Specifically, focused CHAPTER 6. GENERAL DISCUSSION 228 prevention strategies need to be put in place in areas nearby to known feeding grounds for filter-feeding megafauna. For example, the coast around Manta Bay, Nusa Penida MPA is being developed for tourism and small food stalls selling plastic packed drinks and food items to tourists have been erected. Without effective disposal and removal mechanisms in place, waste items, dispersed by the wind, might directly enter manta ray feeding grounds from the adjacent coastlines. Strategies to reduce the amount of waste could also be put in place in MPAs and national parks. Examples include, mandating tourism boats to provide drinking water from refillable dispensers rather than single-use water bottles, and food packaged in reusable containers rather than polystyrene or other single-use plastics. Responsible waste management should be incorporated into the general codes of conduct for marine conservation zone use (e.g. Komodo National Park, 2019). Lastly, to encourage the public to quickly adopt new schemes, continued public education and awareness campaigns are needed in tandem with improvements to infrastructure, capture and materials. Incorporating the information gleaned from this thesis and others (e.g. Fossi et al., 2014; 2017; Panti et al., 2019; Pelamatti et al., 2019) into educational campaigns that bring awareness about the potential impacts of plastic marine debris to flagship species, like manta rays, whale sharks and other charismatic megafauna (Chapter 2; Bowen-Jones & Entwistle, 2002; Panti et al., 2019), as well as to important fisheries (e.g. Lusher et al., 2013; Rochman et al., 2015; Romeo et al., 2015), has the potential to serve as a powerful awareness tool. 229 6.4 CLOSING STATEMENT The threats to iconic megafauna like manta rays and whale sharks characterized in this study are by no means exhaustive, nor are they unique to Indonesia. Most of these threats exist throughout the global ranges of these species (e.g. Philippines, Thailand, Maldives, Mozambique, and Mexico) and extend to other megafauna with similar flagship status, such as basking sharks, mobula rays, turtles, whales and dolphins. Continuing to increase our understanding of these overarching threats to vulnerable marine megafauna and effectively communicating these threats to the wider public are important components of increasing marine stewardship. It is through increasing the profile and ecological understanding of these flagships, that conservationists might evoke greater empathy and a desire from the public to act on behalf of healthier marine environments and sustainable industries. 230 APPENDIX I. POLLUTION AND CONTAMINANTS Mobulid rays, as filter-feeders, may be susceptible to ingesting marine pollutants and contaminants such as persistent organic pollutants (POPs), heavy metals, and microplastics. Pollutants enter the marine environment through wastewater, poor industry practices, and degradation of marine debris, among other sources. Microplastics contain added pollutants such as phthalates, bisphenol A, flame retardants, styrenes, and adsorb and concentrate persistent organic pollutants (POPs) and heavy metals from the marine environment (Worm et al., 2017). Many dissolved pollutants are lipophilic, persistent in the environment, and have the capacity to bio accumulate up the food chain, with large bodied, long-lived species at high risk to exposure (Niimi, 1996). Mobulid habitats, including foraging grounds, overlap with microplastic pollution hotspots in many cases (Chapter 2). However, rates of plastic ingestion by mobulids, bioaccumulation of pollutants, and the impacts of plastic pollution on mobulid biology, ecology, and population viability have not been studied. Future research should work toward identifying the scale of these problems and quantifying impacts so they can be evaluated in a management context. APPENDIX I.i. IDENTIFYING TROPHIC INTERACTIONS WITH POLLUTANTS In locations with high densities of floating microplastics, mobulids may directly ingest microplastics. Additionally, zooplankton can be contaminated with pollutants (Fossi et al., 2014) as well as ingest microplastics and nanoplastics (Cole et al., 2013; Setälä et al., 2014). This suggests that mobulids may be secondary consumers of microplastics and associated pollutants even if they are foraging in locations (or at depths) that do not have high densities of floating microplastics. Future research should 231 seek to quantify ingestion rates of microplastics in mobulids. This could be accomplished by evaluating microplastic densities and contamination in prey assemblages by sampling prey alongside feeding mobulids. Additionally, measuring the abundance of microplastic particles in stomach contents of mobulids landed in fisheries could provide estimates of direct ingestion. In order to evaluate whether ingestion leads to bioaccumulation in tissues, researchers should measure phthalate, heavy metal, and POP levels in mobulid tissue biopsies from wild populations and samples collected from fisheries. Previous studies found elevated levels of some heavy metals in mobulid tissues (Essumang, 2009, 2010; Ooi et al., 2015), but low levels of POPs (E. Germanov, pers. obs.; D. Fernando, Blue Resources, pers. comm.). Phthalates and/or POPs have been recorded in tissue samples of baleen whales, basking sharks and whale sharks in areas with high levels of microplastic pollution (Fossi et al., 2014, 2016, 2017), indicating that filter-feeding organisms are likely bioaccumulating these pollutants as a result of plastic ingestion. Future research in mobulids could evaluate levels of phthalates, POPs and heavy metals in mobulid tissues across gradients of microplastic pollution to determine how bioaccumulation scales with pollutant levels in the environment and to identify heavily impacted populations. Sources and aggregating mechanisms of plastic pollution vary between coastal and offshore environments, leading to differing levels of overlap between microplastic hotspots and mobulid habitats (Chapter 2). Researchers could investigate differences in pollutant contamination between coastal mobulids (e.g., Mobula alfredi, M. munkiana) and offshore species, as well as species with different foraging strategies to identify risk factors for plastic ingestion. Once trophic interactions between mobulids and pollutants are better understood, researchers could endeavor to identify high-risk areas for mobulids following existing approaches for other taxa such as sea turtles (Schuyler et al., 2016). 232 APPENDIX I.ii. QUANTIFYING IMPACTS FROM POLLUTANTS A number of recent studies have demonstrated that microplastics, POPs and heavy metals impact regular cellular and system functioning, including endocrine disruption, leading to knock-on negative impacts on reproductive output with the potential to alter populations and ecological assemblages of marine species (Chapter 2; Galloway & Lewis, 2016; Jakimska et al., 2011; Rochman, 2013; Rochman et al., 2014; Sussarellu et al., 2016). The implications of exposure to pollution and contaminants in mobulids is still speculative, especially at the level of individual fitness and population viability. Demonstrating that mobulids ingest microplastics and other pollutants is important for assessing the extent of these threats, but will be insufficient for evaluating the true impact of pollutants and their importance as a management consideration. Future studies should seek to quantify the impacts of plastic pollution on mobulids by, for example, comparing pollutant concentrations in the environment and in mobulid tissues with metrics of individual and population health such as body condition, reproductive output, population trends, and individual survival. Studies that compare populations across a pollutant gradient, or evaluate a single population through time as pollutant levels fluctuate could provide insights into the true impacts of plastics and other pollutants and their importance as an emerging management consideration. 233 APPENDIX II. SUPPLEMENTARY DATA APPENDIX II.i. Chapter 3 Size Frequency Data Supplementary Figure 1. The available estimated size classes of manta rays using Manta Bay (A, B) and Manta Point (C, D). Size classes are grouped by estimated disk width size (m) in 0.5 m bins and presented by sex (A, C) and for males according to maturity status (B, D). J = juvenile, SA = sub-adult, Mat = mature. 234 APPENDIX II.ii. Chapter 3 and 4 Injury Data Supplementary Table 1. The following data tables catalogue the identification (ID) numbers and types of injuries recorded as of April 2018 for individual manta rays that have been sighted with injuries in the A) Nusa Penida Marine Protected Area (MPA) and B) Komodo National Park (NP). ID numbers can be used to search for full sightings information at www.mantamatcher.org. Ceph Inj = Cephalic injury; Pec Inj = Pectoral injury; Line Inj = Line injury 235 236 237 238 REFERENCES Abecasis, D., Afonso, P., & Erzini, K. (2015). Toward adaptive management of coastal MPAs: The influence of different conservation targets and costs on the design of no-take areas. Ecological Informatics, 30, 263-270. Abreo, N.A.S., Blatchley, D., & Superio, M. D. (2019). Stranded whale shark (Rhincodon typus) reveals vulnerability of filter-feeding elasmobranchs to marine litter in the Philippines. Marine Pollution Bulletin, 141, 79-83. Adams, K.R., Fetterplace, L.C., Davis, A.R., Taylor, M. D., & Knott, N.A. (2018). Sharks, rays and abortion: The prevalence of capture-induced parturition in elasmobranchs. Biological Conservation, 217, 11-27. Alava, J., & Merlen, G. (2009). Video-documentation of a killer whale (Orcinus orca) predatory attack on a giant manta (Manta birostris) in the Galápagos islands. Latin American Journal of Aquatic Mammals, 7(1-2), 81-84. Alerstam, T., Hedenström, A., & Åkesson, S. (2003). Long-distance migration: Evolution and determinants. Oikos, 103(2), 247-260. Alimba, C. G., & Faggio, C. (2019). Microplastics in the marine environment: Current trends in environmental pollution and mechanisms of toxicological profile. Environmental toxicology and pharmacology. 68, 61-74. Anderson, R. C., Adam, M. S., & Goes, J. I. (2011). From monsoons to mantas: Seasonal distribution of manta alfredi in the Maldives. Fisheries Oceanography, 20(2), 104-113. Anderson, R.C., Adam, M.S., Kitchen-Wheeler, A., & Stevens, G. (2010). Extent and economic value of manta ray watching in Maldives. Tourism in Marine Environments, 7(1), 15-27. Andrady, A. L. (2011). Microplastics in the marine environment. Marine Pollution Bulletin, 62, 1596–1605. Anselmo, H. M., Koerting, L., Devito, S., van den Berg, Johannes HJ, Dubbeldam, M., Kwadijk, C., & Murk, A. J. (2011). Early life developmental effects of marine 239 persistent organic pollutants on the sea urchin psammechinus miliaris. Ecotoxicology and Environmental Safety, 74(8), 2182-2192. Arauz, R., Chávez, E. J., Hoyos-Padilla, E. M., & Marshall, A. D. (2019). First record of the reef manta ray, mobula alfredi, from the eastern pacific. Marine Biodiversity Records, 12(1), 3. Arias, A. (2015). Understanding and managing compliance in the nature conservation context. Journal of Environmental Management, 153, 134-143. Armstrong, A.O., Armstrong, A.J., Jaine, F.R., Couturier, L.I., Fiora, K., Uribe-Palomino, J., Weeks, S.J., Townsend, K.A., Bennett, M.B., & Richardson, A.J. (2016). Prey density threshold and tidal influence on reef manta ray foraging at an aggregation site on the great barrier reef. PLoS One, 11(5), e0153393. Ashe, E., Noren, D. P., & Williams, R. (2010). Animal behaviour and marine protected areas: Incorporating behavioural data into the selection of marine protected areas for an endangered killer whale population. Animal Conservation, 13(2), 196-203. Ashton, K., Holmes, L., & Turner, A. (2010). Association of metals with plastic production pellets in the marine environment. Marine Pollution Bulletin, 60(11), 2050-2055. Aqil, A.M.I. (2018) Ban against single-use plastic bags, half-hearted effort. The Jakarta Post. Available at: https://www.thejakartapost.com/news/2018/12/28/ban-against-single-use-plastic-bags-half-hearted-effort.html (accessed February 11, 2019). Avens, L., Taylor, J.C., Goshe, L.R., Jones, T.T., & Hastings, M. (2009). Use of skeletochronological analysis to estimate the age of leatherback sea turtles Dermochelys coriacea in the western north Atlantic. Endangered Species Research, 8(3), 165-177. Aye, L., & Widjaya, E. (2006). Environmental and economic analyses of waste disposal options for traditional markets in indonesia. Waste Management, 26(10), 1180-1191. Ayers, J.M., Strutton, P.G., Coles, V.J., Hood, R.R., & Matear, R.J. (2014). Indonesian throughflow nutrient fluxes and their potential impact on Indian ocean productivity. Geophysical Research Letters, 41(14), 5060-5067. 240 Backhaus, T., & Wagner, M. (2020). Microplastics in the environment: Much ado about nothing? A debate. Global Challenges, 4(6), 1900022. Baird, R.W., Gorgone, A.M., McSweeney, D.J., Webster, D.L., Salden, D.R., Deakos, M.H., Ligon, A.D., Schorr, G.S., Barlow, J., & Mahaffy, S.D. (2008). False killer whales (Pseudorca crassidens) around the main Hawaiian islands: Long-term site fidelity, inter-island movements, and association patterns. Marine Mammal Science, 24(3), 591-612. Barboza, L.G.A., Vieira, L. R., Branco, V., Carvalho, C., & Guilhermino, L. (2018). Microplastics increase mercury bioconcentration in gills and bioaccumulation in the liver, and cause oxidative stress and damage in Dicentrarchus labrax juveniles. Scientific Reports, 8(1), 15655. Barber, P. H. (2009). The challenge of understanding the Coral Triangle biodiversity hotspot. Journal of Biogeography, 36(10), 1845-1846. Barr, Y., & Abelson, A. (2019a). Feeding–Cleaning trade off: Manta ray ‘decision making’ as a conservation tool. Frontiers in Marine Science, 6, 88. Barr, R., Yudiarso, P., Welly, M., & Pratama, C.D. (2017). Improving manta protection in Nusa Penida Marine Protected Area: A willingness-to- pay exploration of potential permit scheme. Bali: Conservation Strategy Fund. Available at: https://www.conservation-strategy.org/en/publication/improving- manta-protection-nusa-penida-marine-protected-area-willingness-pay-exploration#.XCxa8M8zbL8 (accessed January 2, 2019). Beck, M. W., Heck, K. L., Able, K. W., Childers, D. L., Eggleston, D. B., Gillanders, B. M., Halpern, B., Hays, C. G., Hoshino, K., Minello, T. J., Orth, R. J., Sheridan, P. F., & Weinstein, M. R. (2001). The identification, conservation, and management of estuarine and marine nurseries for fish and invertebrates. Bioscience, 51(8), 633-641. Barua, M. (2011). Mobilizing metaphors: The popular use of keystone, flagship and umbrella species concepts. Biodiversity and Conservation, 20(7), 1427. 241 Bejder, L., Samuels, A., Whitehead, H., & Gales, N. (2006a). Interpreting short-term behavioural responses to disturbance within a longitudinal perspective. Animal Behaviour, 72(5), 1149-1158. BBC (2019) Dead Philippines whale had 40kg of plastic in stomach. BBC. Available at: https://www.bbc.com/news/world-asia-47608949. (accessed April 2, 2019). Bejder, L., Samuels, A., Whitehead, H., Gales, N., Mann, J., Connor, R., Heithaus, M., Watson-Capps, J.A., Flaherty, C., & Kruetzen, M. (2006b). Decline in relative abundance of bottlenose dolphins exposed to long-term disturbance. Conservation Biology, 20(6), 1791-1798. Bennett, N.J., Teh, L., Ota, Y., Christie, P., Ayers, A., Day, J.C., Franks, P., Gill, D., Gruby, R.L., Kittinger, J.N., & Koehn, J.Z. (2017). An appeal for a code of conduct for marine conservation. Marine Policy, 81, 411-418. Besseling, E., Foekema, E., Van Franeker, J., Leopold, M. F., Kühn, S., Rebolledo, E. B., Heße, E., Mielke, L.J., IJzer, J., Kamminga, P., & Koelmans, A.A. (2015). Microplastic in a macro filter feeder: Humpback whale Megaptera novaeangliae. Marine Pollution Bulletin, 95(1), 248-252. Block, B. A., Jonsen, I. D., Jorgensen, S. J., Winship, A. J., Shaffer, S.A., Bograd, S. J., Hazen, E.L., Foley, D.G., Breed, G.A., Harrison, A.L., & Ganong, J.E. (2011). Tracking apex marine predator movements in a dynamic ocean. Nature, 475(7354), 86. Boerger, C. M., Lattin, G. L., Moore, S. L., & Moore, C. J. (2010). Plastic ingestion by planktivorous fishes in the north pacific central gyre. Marine Pollution Bulletin, 60(12), 2275-2278. Bond, M. E., Babcock, E. A., Pikitch, E. K., Abercrombie, D. L., Lamb, N. F., & Chapman, D. D. (2012). Reef sharks exhibit site-fidelity and higher relative abundance in marine reserves on the Mesoamerican Barrier Reef. PloS one, 7(3). Bond, T., Ferrandiz-Mas, V., Felipe-Sotelo, M., & van Sebille, E. (2018). The occurrence and degradation of aquatic plastic litter based on polymer physicochemical 242 properties: A review. Critical Reviews in Environmental Science and Technology, 1-38. Bonfil, R., Meyer, M., Scholl, M. C., Johnson, R., O'Brien, S., Oosthuizen, H., Swanson, S., Kotze, D., & Paterson, M. (2005). Transoceanic migration, spatial dynamics, and population linkages of white sharks. Science (New York, N.Y.), 310(5745), 100-103. Bonney, R., Cooper, C.B., Dickinson, J., Kelling, S., Phillips, T., Rosenberg, K.V., & Srkhi, J. (2009). Citizen science: A developing tool for expanding science knowledge and scientific literacy. Bioscience, 59, 977. Booth, H. (2016). Using the Case of Illegal Manta Ray Trade in Indonesia to Evaluate the Impact of Wildlife Trade Policy. (Dissertation, Imperial College London). Borja, A. (2014). Grand challenges in marine ecosystems ecology. Frontiers in Marine Science, 1, 1. Bowen, B.W., & Karl, S. (2007). Population genetics and phylogeography of sea turtles. Molecular Ecology, 16(23), 4886-4907. Bowen-Jones, E., & Entwistle, A. (2002). Identifying appropriate flagship species: The importance of culture and local contexts. Oryx, 36(02), 189-195. Boyle, M., FitzSimmons, N., Limpus, C., Kelez, S., Velez-Zuazo, X., & Waycott, M. (2009). Evidence for transoceanic migrations by loggerhead sea turtles in the southern Pacific ocean. Proceedings of the Royal Society B: Biological Sciences, 276(1664), 1993-1999. Braun, C. D., Skomal, G.B., Thorrold, S.R., & Berumen, M.L. (2014). Diving behavior of the reef manta ray links coral reefs with adjacent deep pelagic habitats. Plos One, 9, e88170. Braungart, M., McDonough, W., & Bollinger, A. (2007). Cradle-to-cradle design: Creating healthy emissions–a strategy for eco-effective product and system design. Journal of Cleaner Production, 15(13), 1337-1348. Brightsmith, D. J., Stronza, A., & Holle, K. (2008). Ecotourism, conservation biology, and volunteer tourism: A mutually beneficial triumvirate. Biological Conservation, 141(11), 2832-2842. 243 Browne, M. A., Dissanayake, A., Galloway, T. S., Lowe, D. M., & Thompson, R. C. (2008). Ingested microscopic plastic translocates to the circulatory system of the mussel, Mytilus edulis (L.). Environmental Science & Technology, 42(13), 5026-5031. Browne, M.A., Niven, S.J., Galloway, T.S., Rowland, S.J., & Thompson, R.C. (2013). Microplastic moves pollutants and additives to worms, reducing functions linked to health and biodiversity. Current Biology, 23(23), 2388-2392. Browne, M. A., Underwood, A. J., Chapman, M. G., Williams, R., Thompson, R. C., & van Franeker, J. A. (2015). Linking effects of anthropogenic debris to ecological impacts. Proceedings of the Royal Society B: Biological Sciences, 282(1807), 20142929. Bruce, B. D. (2008). The biology and ecology of the white shark, Carcharodon carcharias. Sharks of the Open Ocean: Biology, Fisheries and Conservation. Oxford, UK: Blackwell Publishing, 69-81. Burton Jr, G. A. (2017). Stressor exposures determine risk: so, why do fellow scientists continue to focus on superficial microplastics risk? Caro, T., & O'Doherty, G. (1999). On the use of surrogate species in conservation biology. Conservation Biology, 13(4), 805-814. Carrier, J. C., Pratt, H. L., & Castro, J. I. (2004). Reproductive biology of elasmobranchs. In Biology of Sharks and Their Relatives, CRC, Boca Raton, FL, USA, 269-286. Castro, J.I. (1993). The shark nursery of Bulls Bay, South Carolina, with a review of the shark nurseries of the south-eastern coast of the United States. Environmental Biology of Fishes, 38, 37-48. Catlin, J., & Jones, R. (2010). Whale shark tourism at Ningaloo marine park: A longitudinal study of wildlife tourism. Tourism Management, 31(3), 386-394. Cerutti-Pereyra, F., Thums, M., Austin, C.M., Bradshaw, C.J.A., Stevens, J.D., Babcock, R.C., Pillans, R.D., & Meekan, M.G. (2014). Restricted movements of juvenile rays in the lagoon of Ningaloo Reef, Western Australia-evidence for the existence of a nursery. Environmental Biology of Fishes, 97(4), 371-383. 244 Chabanne, D.B., Finn, H., & Bejder, L. (2017). Identifying the relevant local population for environmental impact assessments of mobile marine fauna. Frontiers in Marine Science, 4, 148. Chandler, M., See, L., Copas, K., Bonde, A. M., López, B. C., Danielsen, F., Legind, J.K., Masinde, S., Miller-Rushing, A.J., Newman, G., & Rosemartin, A. (2017). Contribution of citizen science towards international biodiversity monitoring. Biological Conservation, 213, 280-294. Chang, C., Wang, Z., McBride, J., & Liu, C. (2005). Annual cycle of southeast Asia—Maritime continent rainfall and the asymmetric monsoon transition. Journal of Climate, 18(2), 287-301. Chapman, D.D., Feldheim, K.A., Papastamatiou, Y.P., & Hueter, R.E. (2015). There and back again: A review of residency and return migrations in sharks, with implications for population structure and management. Annual Review of Marine Science, 7, 547-570. Chen, Q., Reisser, J., Cunsolo, S., Kwadijk, C., Kotterman, M., Proietti, M., Slat, B., Ferrari, F.F., Schwarz, A., Levivier, A., & Yin, D. (2018). Pollutants in plastics within the north Pacific subtropical gyre. Environmental Science & Technology, 52(2), 446-456. Christie, P., Bennett, N.J., Gray, N.J., Wilhelm, T.A., Lewis, N.A., Parks, J., Ban, N.C., Gruby, R.L., Gordon, L., Day, J., & Taei, S. (2017). Why people matter in ocean governance: Incorporating human dimensions into large-scale marine protected areas. Marine Policy, 84, 273-284. Cole, M., Lindeque, P., Fileman, E., Halsband, C., Goodhead, R., Moger, J., & Galloway, T. S. (2013). Microplastic ingestion by zooplankton. Environmental Science & Technology, 47(12), 6646-6655. Collignon, A., Hecq, J., Glagani, F., Voisin, P., Collard, F., & Goffart, A. (2012). Neustonic microplastic and zooplankton in the north western Mediterranean Sea. Marine Pollution Bulletin, 64(4), 861-864. 245 Colman, J. (1997). A review of the biology and ecology of the whale shark. Journal of Fish Biology, 51(6), 1219-1234. Conservation International. (2015) Manta ray study reveals Southeast Asia's first nursery, confirms risky migration across hunting grounds. Jakarta: Conservation International Indonesia. Available at: https://www.rwsentosa.com/-/media/project/non-gaming/rwsentosa/press-release/pdf/en/2015/20151028-rws-conservationinternational-mantaraystudy.pdf (Accessed February 11, 2020). Conservation International. (2016). Laporan Program Kerjasama Tagging Satelit Pari Manta. Jakarta: Conservation International Indonesia. Cooke, S. J. (2008). Biotelemetry and biologging in endangered species research and animal conservation: Relevance to regional, national, and IUCN Red List threat assessments. Endangered Species Research, 4(1-2), 165-185. Costa, C., Loy, A., Cataudella, S., Davis, D., & Scardi, M. (2006). Extracting fish size using dual underwater cameras. Aquacultural Engineering, 3(35), 218-227. Coral Triangle Center. (2013). Nusa Penida – A Tropical Marine Paradise. Available at: https://www.coraltrianglecenter.org/nusa-penida-mpa/ (Accessed February 11, 2020). Couturier, L.I.E., Marshall, A.D., Jaine, F.R.A., Kashiwagi, T., Pierce, S.J., Townsend, K.A., Weeks, S.J., Bennett, M.B., & Richardson, A. J. (2012). Biology, ecology and conservation of the Mobulidae. J. Fish Biol., 80, 1075. Couturier, L.I.E., Dudgeon, C.L., Pollock, K.H., Jaine, F.R.A., Bennett, M.B., Townsend, K.A., Weeks, S.J., &Richardson, A. J. (2014). Population dynamics of the reef manta ray Manta alfredi in eastern Australia. Coral Reefs, 33(2), 329-342. Couturier, L., Newman, P., Jaine, F., Bennett, M., Venables, W., Cagua, E., Townsend, K.A., Weeks, S.J., & Richardson, A. (2018). Variation in occupancy and habitat use of Mobula alfredi at a major aggregation site. Marine Ecology Progress Series, 599, 125-145. 246 Couturier, L. I., Rohner, C. A., Richardson, A. J., Marshall, A. D., Jaine, F., Bennett, M. B., Townsend, K.A., Weeks, S.J., & Nichols, P. D. (2013). Stable isotope and signature fatty acid analyses suggest reef manta rays feed on demersal zooplankton. PLoS One, 8(10), e77152. Croll, D. A., Dewar, H., Dulvy, N. K., Fernando, D., Francis, M. P., Galván-Magaña, F., Hall, M., Heinrichs, S., Marshall, A., McCauley, D., & Newton, K.M. (2016). Vulnerabilities and fisheries impacts: The uncertain future of manta and devil rays. Aquatic Conservation: Marine and Freshwater Ecosystems, 26(3), 562-575. Dahlgren, C. P., Kellison, G. T., Adams, A. A., Gillanders, B. M., Kendall, M. S., Layman, C. A., Ley, J. A., Nagelkerken, I., & Serafy, J. E. (2006). Marine nurseries and effective juvenile habitats: concepts and applications. Marine Ecology Progress Series, 312, 291–295. Daly-Engel, T. S., Seraphin, K. D., Holland, K. N., Coffey, J. P., Nance, H. A., Toonen, R. J., & Bowen, B. W. (2012). Global phylogeography with mixed-marker analysis reveals male-mediated dispersal in the endangered scalloped hammerhead shark (Sphyrna lewini). PLoS One, 7(1). Davy, L. E., Simpfendorfer, C. A., & Heupel, M. R. (2015). Movement patterns and habitat use of juvenile mangrove whiprays (Himantura granulata). Marine and Freshwater Research, 66(6), 481-492. Dawbin, W.H. (1966). The seasonal migratory cycle of humpback whales. Whales, Dolphins and Porpoises, 45-170. Deakos, M.H., Baker, J.D., & Bejder, L. (2011). Characteristics of a manta ray manta alfredi population off Maui, Hawaii, and implications for management. Marine Ecology Progress Series, 429, 245. Deakos, M.H. (2012). The reproductive ecology of resident manta rays (Manta alfredi) off Maui, Hawaii, with an emphasis on body size. Environmental Biology of Fishes, 94(2), 443-456. Derraik, J. G. (2002). The pollution of the marine environment by plastic debris: a review. Marine Pollution Bulletin, 44(9), 842-852. 247 de Stephanis, R., Giménez, J., Carpinelli, E., Gutierrez-Exposito, C., & Cañadas, A. (2013). As main meal for sperm whales: Plastics debris. Marine Pollution Bulletin, 69(1), 206-214. Dewar, H., Mous, P., Domeier, M., Muljadi, A., Pet, J., & Whitty, J.V. (2008). Movements and site fidelity of the giant manta ray, Manta birostris, in the Komodo marine park, Indonesia. Marine Biology, 155, 121-133. District Fisheries and Marine Agency. (2010). Peraturan Bupati Klungkung No/12/2010 tentang Kawasan Konservasi Perairan Daerah. Klungkung. Bali: District Fisheries and Marine Agency. Available at: http://kkji.kp3k.kkp.go. id/index.php/en/marine- protected- area- data/details/7/84 (accessed January 2, 2019). Division of Boating, and Ocean Recreation (2016). Manta Viewing within the West Hawaii Ocean Recreation Management Area: Introduction of Proposed Administrative Rules. Kona: Department of Land and Natural resources. Available at: https://dlnr.hawaii.gov/dobor/files/2013/08/ MantaDiveSitesManagementPlan-9.9.16.pdf (accessed January 2, 2019). Di Mauro, R., Kupchik, M.J., & Benfield, M.C. (2017). Abundant plankton-sized microplastic particles in shelf waters of the northern Gulf of Mexico. Environmental Pollution, 230, 798-809. Donati, G.F.A., Shameel, I., Hindle, K., Zareer, I. & Wieczorek, A.M. (2019) Microplastics a macro-disaster: a threat to the largest fish of our seas? Paper presented at the IWSC5, Ningaloo, Australia. Dove, A. D., Leisen, J., Zhou, M., Byrne, J. J., Lim-Hing, K., Webb, H. D., Gelbaum, L., Viant, M.R., Kubanek, J., & Fernández, F. M. (2012). Biomarkers of whale shark health: A metabolomic approach. PLoS One, 7(11), e49379. Ducarme, F., Luque, G. M., & Courchamp, F. (2013). What are “charismatic species” for conservation biologists. BioSciences Master Reviews, 10(2013), 1-8. Dudgeon, C.L., Blower, D.C., Broderick, D., Giles, J.L., Holmes, B.J., Kashiwagi, T., Krück, N.C., Morgan, J.A.T., Tillett, B.J., & Ovenden, J. R. (2012). A review of the 248 application of molecular genetics for fisheries management and conservation of sharks and rays. Journal of Fish Biology, 80(5), 1789-1843. Duinkerken, D. (2010). Movements and site fidelity of the reef manta ray, manta alfredi, along the coast of southern mozambique. MSc Utrecht University, the Netherlands, Dulvy, N.K., & Reynolds, J.D. (1997). Evolutionary transitions among egg–laying, live–bearing and maternal inputs in sharks and rays. Proceedings of the Royal Society of London.Series B: Biological Sciences, 264(1386), 1309-1315. Dulvy, N.K., Sadovy, Y., & Reynolds, J.D. (2003). Extinction vulnerability in marine populations. Fish and Fisheries, 4(1), 25-64. Dulvy, N.K., Pardo, S.A., Simpfendorfer, C.A., & Carlson, J.K. (2014). Diagnosing the dangerous demography of manta rays using life history theory. Peerj, 2, e400. Eckert, S.A., & Stewart, B.S. (2001). Telemetry and satellite tracking of whale sharks, rhincodon typus, in the Sea of Cortez, Mexico, and the north Pacific Ocean. The behavior and sensory biology of elasmobranch fishes: An anthology in memory of Donald Richard Nelson (299-308) Springer. Endo, S., Takizawa, R., Okuda, K., Takada, H., Chiba, K., Kanehiro, H., Ogi, H., Yamashita, R., & Date, T. (2005). Concentration of polychlorinated biphenyls (PCBs) in beached resin pellets: Variability among individual particles and regional differences. Marine Pollution Bulletin, 50(10), 1103-1114. Engelhaupt, D., Rus Hoelzel, A., Nicholson, C., Frantzis, A., Mesnick, S., Gero, S., Whitehead, H., Rendell, L., Miller, P., De Stefanis, R., & Canadas, A. (2009). Female philopatry in coastal basins and male dispersion across the north Atlantic in a highly mobile marine species, the sperm whale (Physeter macrocephalus). Molecular Ecology, 18(20), 4193-4205. Entwistle, A. (2000). Flagships for the future? Oryx, 34(4), 239-240. Erdmann, A. (2004a). A natural history guide to Komodo National Park: Management. Nature conservancy, Indonesia, Coastal and Marine Program. Erdmann, A. (2004b). A natural history guide to Komodo National Park: Terrestrial. Nature conservancy, Indonesia, Coastal and Marine Program. 249 Erdmann, A. (2004c). A natural history guide to Komodo National Park: Marine. Nature conservancy, Indonesia, Coastal and Marine Program. Eriksen, M., Lebreton, L. C., Carson, H. S., Thiel, M., Moore, C. J., Borerro, J. C., Galgani, F., Ryan, P.G., & Reisser, J. (2014). Plastic pollution in the world's oceans: More than 5 trillion plastic pieces weighing over 250,000 tons afloat at sea. PloS One, 9(12), e111913. Essumang. (2010). First determination of the levels of platinum group metals in Manta birostris (manta ray) caught along the Ghanaian coastline. Bull Environ Contam Toxicol, 84, 720–725. Feldheim, K.A., Gruber, S.H., DiBattista, J.D., Babcock, E.A., Kessel, S.T., Hendry, A.P., Pikitch, E.K., Ashley, M.V., & Chapman, D.D. (2014). Two decades of genetic profiling yields first evidence of natal philopatry and long-term fidelity to parturition sites in sharks. Molecular Ecology, 23(1), 110-117. Field, A., & Gordon, A. L. (1996). Tidal mixing signatures in the Indonesian seas. Journal of Physical Oceanography, 26(9), 1924-1937. Fleming, A., & Jackson, J. (2011). Global review of humpback whales (Megaptera novaeangliae). Fossi, M.C., Marsili, L., Neri, G., Casini, S., Bearzi, G., Politi, E., Zanardelli, M., & Panigada, S. (2000). Skin biopsy of Mediterranean cetaceans for the investigation of interspecies susceptibility to xenobiotic contaminants. Marine Environmental Research, 50(1-5), 517-521. Fossi, M. C., Panti, C., Guerranti, C., Coppola, D., Giannetti, M., Marsili, L., & Minutoli, R. (2012). Are baleen whales exposed to the threat of microplastics? A case study of the Mediterranean fin whale (Balaenoptera physalus). Marine Pollution Bulletin, 64(11), 2374-2379. Fossi, M.C., Coppola, D., Baini, M., Giannetti, M., Guerranti, C., Marsili, L., Panti, C., de Sabata, E., & Clò, S. (2014). Large filter feeding marine organisms as indicators of microplastic in the pelagic environment: The case studies of the Mediterranean 250 basking shark (Cetorhinus maximus) and fin whale (Balaenoptera physalus). Marine Environmental Research, 100, 17-24. Fossi, M. C., Marsili, L., Baini, M., Giannetti, M., Coppola, D., Guerranti, C., Caliani, I., Minutoli, R., Lauriano, G., Finoia, M.G., & Rubegni, F. (2016). Fin whales and microplastics: The Mediterranean Sea and the Sea of Cortez scenarios. Environmental Pollution, 209, 68-78. Fossi, M.C., Baini, M., Panti, C., Galli, M., Jiménez, B., Muñoz-Arnanz, J., Marsili, L., Finoia, M.G., & Ramírez-Macías, D. (2017). Are whale sharks exposed to persistent organic pollutants and plastic pollution in the gulf of California (Mexico)? First ecotoxicological investigation using skin biopsies. Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology, 199, 48-58. Fox, S., Foisy, I., De La Parra Venegas, R, Galván Pastoriza, B., Graham, R., Hoffmayer, E., Holmberg, J., Pierce, S. (2013). Population structure and residency of whale sharks Rhincodon typus at Utila, Bay islands, Honduras. Journal of Fish Biology, 83(3), 574-587. Frid, A., & Dill, L. (2002). Human-caused disturbance stimuli as a form of predation risk. Conservation Ecology, 6(1) De Freyatas-Tamura, K. (2018). Plastics Pile Up as China Refuses to Take the West’s Recycling. Time. Available at: https://www.nytimes.com/2018/01/11/world/china-recyclables-ban.html (accessed February 11, 2019). Gall, S.C., & Thompson, R. C. (2015). The impact of debris on marine life. Marine Pollution Bulletin, 92(1–2), 170-179. Gallagher, A.J., & Hammerschlag, N. (2011). Global shark currency: The distribution, frequency, and economic value of shark ecotourism. Current Issues in Tourism, 14(8), 797-812. Galloway, T.S., & Lewis, C.N. (2016). Marine microplastics spell big problems for future generations. Proceedings of the National Academy of Sciences, 113(9), 2331-2333. 251 Garrud, E. (2016). Does tourist behaviour affect reef manta ray feeding behaviour. an analysis of human and Manta alfredi interactions in baa atoll, the Maldives. (Unpublished University of York., York, UK.) Gaffin, S., Rosenzweig, C., Xing, X. & Yetman, G. (2004). Downscaling and Geo-spatial Gridding of Socio-economic Projections from the IPCC Special Report on Emissions Scenarios (SRES). Global Environmental Change 14, 105-123. Genov, T., Jepson, P.D., Barber, J.L., Hace, A., Gaspari, S., Centrih, T., Lesjak, J., & Kotnjek, P. (2019). Linking organochlorine contaminants with demographic parameters in free-ranging common bottlenose dolphins from the northern Adriatic Sea. Science of the Total Environment, 657, 200-212. Germanov, E.S., & Marshall, A.D. (2014). Running the gauntlet: Regional movement patterns of manta alfredi through a complex of parks and fisheries. Plos One, 9, e110071. Gleiss, A.C., Wright, S., Liebsch, N., Wilson, R.P., & Norman, B. (2013). Contrasting diel patterns in vertical movement and locomotor activity of whale sharks at Ningaloo Reef. Marine Biology, 160(11), 2981-2992. Gokkon, B. (2018) As planned excise flops, Indonesia ponders how to give up plastic bags. Mongabay. Available at: https://news.mongabay.com/2018/07/as-planned-excise-flops-indonesia-ponders-how-to-give-up-plastic-bags/ (accessed April 5, 2019). Gouin, T., Roche, N., Lohmann, R., & Hodges, G. (2011). A thermodynamic approach for assessing the environmental exposure of chemicals absorbed to microplastic. Environmental Science & Technology, 45(4), 1466-1472. Graham, F., Rynne, P., Estevanez, M., Luo, J., Ault, J.S., & Hammerschlag, N. (2016). Use of marine protected areas and exclusive economic zones in the subtropical western north Atlantic Ocean by large highly mobile sharks. Diversity and Distributions, 22(5), 534-546. Graham, B. (2019) Indonesian village being inundated with illegal plastic waste from Australia. News.com.au. Available at: 252 https://www.news.com.au/technology/environment/indonesian-village-being-inundated-with-illegal-plastic-waste-from-australia/news-story/9a03e5a78d67f25994f4176731a30822 (accessed June 5, 2019). Gray, J. S. (1997). Marine biodiversity: Patterns, threats and conservation needs. Biodiversity & Conservation, 6(1), 153-175. Greggor, A.L., Berger-Tal, O., Blumstein, D. T., Angeloni, L., Bessa-Gomes, C., Blackwell, B.F., St Clair, C.C., Crooks, K., de Silva, S., Fernández-Juricic, E., & Goldenberg, S. Z. (2016). Research priorities from animal behaviour for maximising conservation progress. Trends in Ecology & Evolution, 31(12), 953-964. Gregory, M. R. (2009). Environmental implications of plastic debris in marine settings —entanglement, ingestion, smothering, hangers-on, hitch-hiking and alien invasions. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1526), 2013-2025. Grubbs, R. D., & Musick, J. A. (2007). Spatial delineation of summer nursery areas for juvenile sandbar sharks in Chesapeake Bay, Virginia. In C. T. McCandless, N. E. Kohler, and H. L. Pratt Jr. (Eds), Shark Nursery Grounds of the Gulf of Mexico and the East Coast Waters of the United States, American Fisheries Society Symposium, 19 August 2002, Baltimore, MD, USA. 50, 63-86. Haetrakul, T., Munanansup, S., Assawawongkasem, N., & Chansue, N. (2009). A case report: Stomach foreign object in whale shark (Rhincodon typus) stranded in Thailand. Hale, R. C. (2018). Are the risks from microplastics truly trivial?. Environmental science & technology, 52(3), 931-931. Hani, M., Jompa, J., Nessa, M., & White, A. (2019). Manta ray watching tourism in eastern indonesia: Is it sustainable? Paper presented at the IOP Conference Series: Earth and Environmental Science, 253(1) 012042. Hausmann, A., Slotow, R., Fraser, I. & Di Minin, E. (2017). Ecotourism marketing alternative to charismatic megafauna can also support biodiversity conservation. Animal Conservation, 20(1), 91-100. 253 Hays GC. (2003). A review of the adaptive significance and ecosystem consequences of zooplankton diel vertical migrations. Hydrobiologia, 503, 163-170. Hays, G. C. (2008). Sea turtles: A review of some key recent discoveries and remaining questions. Journal of Experimental Marine Biology and Ecology, 356(1-2), 1-7. Heinrichs, S., O'Malley, M., Medd, H., & Hilton, P. (2011). Manta ray of hope 2011 report: The global threat to manta and mobula rays. WildAid, San Francisco, CA. Hidalgo-Ruz, V., Gutow, L., Thompson, R.C., & Thiel, M. (2012). Microplastics in the marine environment: A review of the methods used for identification and quantification. Environmental Science & Technology, 46(6), 3060-3075. Higham, J.E., Bejder, L., Allen, S.J., Corkeron, P. J., & Lusseau, D. (2016). Managing whale-watching as a non-lethal consumptive activity. Journal of Sustainable Tourism, 24(1), 73-90. Hines, A. H., Whitlatch, R. B., Thrush, S. F., Hewitt, J. E., Cummings, V. J., Dayton, P. K., & Legendre, P. (1997). Nonlinear foraging response of a large marine predator to benthic prey: eagle ray pits and bivalves in a New Zealand sandflat. Journal of Experimental Marine Biology and Ecology, 216(1-2), 191-210. Hinojosa-Alvarez, S., Walter, R.P., Diaz-Jaimes, P., Galván-Magaña, F., & Paig-Tran, E. M. (2016). A potential third manta ray species near the yucatán peninsula? evidence for a recently diverged and novel genetic manta group from the Gulf of Mexico. Peerj, 4, e2586. Home, R., Keller, C., Nagel, P., Bauer, N., & Hunziker, M. (2009). Selection criteria for flagship species by conservation organizations. Environmental Conservation, 36(2), 139-148. Hooker, S.K., Cañadas, A., Hyrenbach, K.D., Corrigan, C., Polovina, J.J., & Reeves, R.R. (2011). Making protected area networks effective for marine top predators. Endangered Species Research, 13(3), 203-218. Hoornweg, D., & Bhada-Tata, P. (2012). What a waste: A global review of solid waste management. World Bank, Washington, DC. 254 Hosegood, J., Humble, E., Ogden, R., de Bruyn, M., Creer, S., Stevens, G., Abudaya, M., Bassos-Hull, K., Bonfil, R., Fernando, D., & Foote, A.D. (2018). Phylogenomics and species delimitation of mobulid rays reveals cryptic diversity and a new species of manta ray. Biorxiv, 458141. Hoyt, E. (2014). The role of marine protected areas and sanctuaries. In E. J. Techera and N. Klein (Eds), Sharks: Conservation, Governance and Management, New York, NY: Routledge, 263-285. Heupel, M. R., Carlson, J. K., & Simpfendorfer, C. A. (2007). Shark nursery areas: concepts, definition, characterization and assumptions. Marine Ecology Progress Series, 337, 287-297. Heupel, M. R., Kanno, S., Martins, A. P., & Simpfendorfer, C. A. (2019). Advances in understanding the roles and benefits of nursery areas for elasmobranch populations. Marine and Freshwater Research, 70(7), 897-907. Hueter, R., Heupel, M., Heist, E., & Keeney, D. (2005). Evidence of philopatry in sharks and implications for the management of shark fisheries. Journal of Northwest Atlantic Fishery Science, 35, 239-247. Indonesian Meteorology, Climatology and Geophysics Agency (2019) BMKG, Pusat Database BMKG. Indonesian Meteorology, Climatology and Geophysics Agency. Available at: http://dataonline.bmkg.go.id/. (accessed: January, 2019.) International Pellet Watch (2019) Google Pollution Map: Google Map. International Pellet Watch. Available at: www.pelletwatch.org (accessed: February 13, 2019). Jacobsen, J.K., Massey, L., & Gulland, F. (2010). Fatal ingestion of floating net debris by two sperm whales (Physeter macrocephalus). Marine Pollution Bulletin, 60(5), 765-767. Jain, R., & Tiwari, A. (2015). Biosynthesis of planet friendly bioplastics using renewable carbon source. Journal of Environmental Health Science and Engineering, 13(1), 11. Jaine, F.R.A., Couturier, L.I.E., Weeks, S.J., Townsend, K.A., Bennett, M.B., Fiora, K., & Richardson, A. J. (2012). When giants turn up: Sighting trends, environmental 255 influences and habitat use of the manta ray Manta alfredi at a coral reef. PLOS One, 7, e4610. Jaine, F.R.A., Rohner, C.A., Weeks, S.J., Couturier, L.I.E., Bennett, M.B., Townsend, K.A., & Richardson, A. J. (2014). Movements and habitat use of reef manta rays off eastern Australia: Offshore excursions, deep diving and eddy affinity revealed by satellite telemetry. Marine Ecology Progress Series, 510, 73-86. Jaiteh, V. F., Lindfield, S. J., Mangubhai, S., Warren, C., Fitzpatrick, B., & Loneragan, N. R. (2016). Higher abundance of marine predators and changes in fishers' behavior following spatial protection within the world's biggest shark fishery. Frontiers in Marine Science, 3, 43. Jaiteh, V. F., Loneragan, N. R., & Warren, C. (2017). The end of shark finning? Impacts of declining catches and fin demand on coastal community livelihoods. Marine Policy, 82, 224-233. Jakimska, A., Konieczka, P., Skóra, K., & Namieśnik, J. (2011). Bioaccumulation of Metals in Tissues of Marine Animals, Part I: the Role and Impact of Heavy Metals on Organisms. Polish Journal of Environmental Studies, 20(5). Jambeck, J., Hardesty, B.D., Brooks, A.L., Friend, T., Teleki, K., Fabres, J., Beaudoin, Y., Bamba, A., Francis, J., Ribbink AJ, & Baleta, T. (2018). Challenges and emerging solutions to the land-based plastic waste issue in Africa. Marine Policy, 96, 256-263. Jambeck, J.R., Geyer, R., Wilcox, C., Siegler, T.R., Perryman, M., Andrady, A., Narayan, R., & Law, K.L. (2015). Marine pollution. plastic waste inputs from land into the ocean. Science, 347(6223), 768-771. Jolly, G.M. (1965). Explicit estimates from capture-recapture data with both death and immigration-stochastic model. Biometrika, 52, 225-247. Kamal, M. M., Wardiatno, Y., & Noviyanti, N. S. (2016). Habitat conditions and potential food items during the appearance of whale sharks (Rhincodon typus) in Probolinggo waters, Madura strait, Indonesia. QScience Proceedings, 27. 256 Kashiwagi, T. (2014). Conservation biology and genetics of the largest living rays: manta rays. (University of Queensland). Kasmidi, M., & Gunadharma, A. (2017). Diving in manta sandy? Here’s what you need to know! Bird’s Head Seascape. Available at: http://birdsheadseascape.com/diving/diving- manta- sandy- heres- need- know-meidiarti-kasmidi-nikka-amandra-gunadharma/ (accessed January 2, 2019) Kinkade, C., Marra, J., Langdon, C., Knudson, C., & Ilahudet, A.G. (1997). Monsoonal differences in phytoplankton biomass and production in the Indonesian seas: Tracing vertical mixing using temperature. Deep Sea Research Part I: Oceanographic Research Papers, 44(4), 581-592. Kitchen-Wheeler, A., Ari, C., & Edwards, A. J. (2011). Population estimates of alfred mantas (manta alfredi) in central maldives atolls: North male, Ari and Baa. Environmental Biology of Fishes, 93, 557-575. Koelmans, A. A., Besseling, E., Wegner, A., & Foekema, E. M. (2013). Plastic as a carrier of POPs to aquatic organisms: a model analysis. Environmental science & technology, 47(14), 7812-7820. Koelmans, A. A., Bakir, A., Burton, G. A., & Janssen, C. R. (2016). Microplastic as a vector for chemicals in the aquatic environment: critical review and model-supported reinterpretation of empirical studies. Environmental science & technology, 50(7), 3315-3326. Koelmans, A.A., Besseling, E., Foekema, E., Kooi, M., Mintenig, S., Ossendorp, B.C., Redondo-Hasselerharm, P.E., Verschoor, A., Van Wezel, A.P. & Scheffer, M. (2017). Risks of plastic debris: unravelling fact, opinion, perception, and belief. Komodo National Park (2019). Komodo National Park Rules, Regulations and Zoning. Available at: https://drive.google.com/file/d/1Bu3ayaG0UeR6dRPYd5xEMKCgr0sOu6Sx/view?pli=1 (accessed February 11, 2020). 257 Kowalski, N., Reichardt, A. M., & Waniek, J. J. (2016). Sinking rates of microplastics and potential implications of their alteration by physical, biological, and chemical factors. Marine Pollution Bulletin, 109(1), 310-319. Kucklick, J., Schwacke, L., Wells, R., Hohn, A., Guichard, A., Yordy, J., Hansen, L., Zolman, E., Wilson, R., Litz, J., & Nowacek, D. (2011). Bottlenose dolphins as indicators of persistent organic pollutants in the western North Atlantic Ocean and northern Gulf of Mexico. Environmental science & technology, 45(10), 4270-4277. Kühn, S., Rebolledo, E. L. B., & van Franeker, J. A. (2015). Deleterious effects of litter on marine life. In Marine Anthropogenic litter, Springer, Cham., 75-116. Lagheim, J. (2017) Indonesia pledges $1bn a year to curb ocean waste. The Guardian. Available at: https://www.theguardian.com/environment/the-coral-triangle/2017/mar/02/indonesia-pledges-us1-billion-a-year-to-curb-ocean-waste (accessed February 11, 2019). Lawson, J. M., Fordham, S. V., O’Malley, M. P., Davidson, L. N., Walls, R. H., Heupel, Stevens, G., Fernando, D., Budziak, A., Simpfendorfer, C.A., & Ender, I. (2017). Sympathy for the devil: A conservation strategy for devil and manta rays. Peerj, 5, e3027. Lebreton, L.C.M., van der Zwet, J., Damsteeg, J. W., Slat, B., Andrady, A., & Reisser, J. (2017). River plastic emissions to the world's oceans. Nature Communications, 8, 15611. Lee, P.L., Luschi, P., & Hays, G.C. (2007). Detecting female precise natal philopatry in green turtles using assignment methods. Molecular Ecology, 16(1), 61-74. Lehnert, K., Siebert, U., Reißmann, K., Bruhn, R., McLachlan, M.S., Müller, G., van Elk, C.E., Ciurkiewicz, M., Baumgärtner, W., & Beineke, A. (2019). Cytokine expression and lymphocyte proliferative capacity in diseased harbor porpoises (Phocoena phocoena)–Biomarkers for health assessment in wildlife cetaceans. Environmental Pollution, 247, 783-791. Leung, H. (2019) Southeast Asia Doesn't Want to Be the World's Dumping Ground. Here's How Some Countries Are Pushing Back. Time. Available at: 258 https://time.com/5598032/southeast-asia-plastic-waste-malaysia-philippines/ (accessed February 11, 2019). Lenz, R., Enders, K., & Nielsen, T. G. (2016). Microplastic exposure studies should be environmentally realistic. Proceedings of National Academy of Sciences, 113, E4121-E4122. Lewis, S. A., Setiasih, N., Fahmi, F., Dharmadi, D., O'Malley, M. P., Campbell, S. J., Yusuf, M., & Sianipar, A.B. (2015). Assessing Indonesian manta and devil ray populations through historical landings and fishing community interviews. Peerj Preprints. Li, X. Y., & Kokko, H. (2019). Sex-biased dispersal: a review of the theory. Biological Reviews, 94(2), 721-736. Liu, G., Ren, Y., Li, C., Dong, H., Bart-Smith, H., & Fish, F. (2015). “Video: fin flexion and flow modulation in Manta’s forward swimming,” in Proceedings of the 68th Annual Meeting of the APS Division of Fluid Dynamics, Boston, MA, 22–24. Lobelle, D., & Cunliffe, M. (2011). Early microbial biofilm formation on marine plastic debris. Marine Pollution Bulletin, 62(1), 197-200. Lockyer, C. (1984). Review of baleen whale (Mysticeti) reproduction and implications for management. Report of the International Whaling Commission, (6), 27-50. Lohmann, K.J., Lohmann, C.M., Brothers, J.R., & Putman, N.F. (2013). Natal homing and imprinting in sea turtles. The Biology of Sea Turtles, 3, 59-77. Lohmann, R. (2017). Microplastics are not important for the cycling and bioaccumulation of organic pollutants in the oceans—but should microplastics be considered POPs themselves?. Integrated Environmental Assessment and Management, 13(3), 460-465. Lowe, C. G. (2002). Bioenergetics of free-ranging juvenile scalloped hammerhead sharks (Sphyrna lewini) in Kāne'ohe Bay, Ō'ahu, HI. Journal of Experimental Marine Biology and Ecology, 278(2), 141-156. Lyons, K., Carlisle, A., Preti, A., Mull, C., Blasius, M., O'Sullivan, J., Winkler, C., & Lowe, C.G. (2013a). Effects of trophic ecology and habitat use on maternal transfer 259 of contaminants in four species of young of the year lamniform sharks. Marine Environmental Research, 90, 27-38. Lyons, K., Lowe, C.G., & Gillanders, B. (2013b). Mechanisms of maternal transfer of organochlorine contaminants and mercury in the common thresher shark (Alopias vulpinus). Canadian Journal of Fisheries and Aquatic Sciences, 70(12), 1667-1672. Lyons, K., Lavado, R., Schlenk, D., & Lowe, C.G. (2014). Bioaccumulation of organochlorine contaminants and ethoxyresorufin-o-deethylase activity in southern california round stingrays (urobatis halleri) exposed to planar aromatic compounds. Environmental Toxicology and Chemistry, 33(6), 1380-1390. Lagheim, J. (2017). Indonesia pledges $1bn a year to curb ocean waste. The Guardian. Available at: https://www.theguardian.com/environment/the-coral-triangle/2017/mar/02/indonesia-pledges-us1-billion-a-year-to-curb-ocean-waste (accessed February 11, 2019). Lehnert, K., Siebert, U., Reißmann, K., Bruhn, R., McLachlan, M.S., Müller, G., van Elk, C.E., Ciurkiewicz, M., Baumgärtner, W., & Beineke, A. (2019). Cytokine expression and lymphocyte proliferative capacity in diseased harbor porpoises (Phocoena phocoena)-Biomarkers for health assessment in wildlife cetaceans. Environmental Pollution, 247, 783-791. Lusher, A. L., Mchugh, M., & Thompson, R. C. (2013). Occurrence of microplastics in the gastrointestinal tract of pelagic and demersal fish from the English Channel. Marine Pollution Nulletin, 67(1-2), 94-99. MacArthur, D.E., Waughray, D., & Stuchtey, M.R. (2016). The New Plastics Economy, Rethinking the Future of Plastics. In World. Manalu, A.A., Hariyadi, S., & Wardiatno, Y. (2017). Microplastics abundance in coastal sediments of Jakarta Bay, Indonesia. Aquaculture, Aquarium, Conservation & Legislation, 10(5), 1164-1173. Manta Trust (2018). How to Swim with Manta Rays. Manta Trust. Available at: https://swimwithmantas.org/ 260 Manzo, L.C. (2005). For better or worse: Exploring multiple dimensions of place meaning. Journal of Environmental Psychology, 25(1), 67-86. Marine Megafauna Foundation (2018). Codes of Conduct. Yayasan Megafauna Laut. Available at: https://mmfindonesia.wordpress.com/resources/ Marshall, A.D. (2008). Biology and population ecology of Manta birostris in southern Mozambique. (University of Queensland). Marshall, A.D., Compagno, L.J.V., & Bennett, M.B. (2009). Redescription of the genus manta with resurrection of Manta alfredi (Krefft, 1868) (Chondrichthyes; Myliobatoidei; Mobulidae). Zootaxa, 230(1), 1-28. Marshall, A.D., & Bennett, M.B. (2010). Reproductive ecology of the reef manta ray Manta alfredi in southern Mozambique. Journal of Fish Biology, 7, 169-190. Marshall, A.D., Dudgeon, C.L., & Bennett, M.B. (2011). Size and structure of a photographically identified population of manta rays Manta alfredi in southern Mozambique. Marine Biology, 15(8), 1111–1124. Marshall, A.D., & Holmberg, J. (2018). MantaMatcher Photo-identification Library. Available at: http://www.mantamatcher.org/ (accessed May 16, 2018). Marshall, A., Bennett, M. B., Kodja, G., Hinojosa-Alvarez, S., Galvan-Magana, F. Harding, M., Stevens, G. & Kashiwagi, T. (2018a). Manta birostris. in: The IUCN Red List of Threatened Species 2018 (amended version of 2011 assessment). Marshall, A., Kashiwagi, T., Bennett, M.B., Deakos, M., Stevens, G., McGregor, F., Clark, T., Ishihara, H. & Sato, K. (2018b). Manta alfredi. in: The IUCN Red List of Threatened Species 2018 (amended version of 2011 assessment). Martins, A., Heupel, M., Chin, A., & Simpfendorfer, C. (2018). Batoid nurseries: Definition, use and importance. Marine Ecology Progress Series, 59(5), 253-267. Mato, Y., Isobe, T., Takada, H., Kanehiro, H., Ohtake, C., & Kaminuma, T. (2001). Plastic resin pellets as a transport medium for toxic chemicals in the marine environment. Environmental Science & Technology, 35(2), 318-324. Mau, R. (2008). Managing for conservation and recreation: The Ningaloo whale shark experience. Journal of Ecotourism, 7(2-3), 213-225. 261 Maximenko, N., Hafner, J., & Niiler, P. (2012). Pathways of marine debris derived from trajectories of lagrangian drifters. Marine Pollution Bulletin, 65(1), 51-62. Mayer, B., & Damm, P. (2012). The Makassar strait throughflow and its jet. Journal of Geophysical Research: Oceans, 11(7). McAuley, R. B., Simpfendorfer, C. A., Hyndes, G. A., & Lenanton, R. C. J. (2007). Distribution and reproductive biology of the sandbar shark, Carcharhinus plumbeus (Nardo), in Western Australian waters. Marine and Freshwater Research, 58(1), 116-126. McCauley, D.J., DeSalles, P.A., Young, H.S., Papastamatiou, Y.P., Caselle, J.E., Deakos, M.H., Gardner, J.P., Garton, D.W., Collen, J.D., & Micheli, F. (2014). Reliance of mobile species on sensitive habitats: A case study of manta rays (Manta alfredi) and lagoons. Marine Biology, 161(9), 1987-1998. McCoy, E., Burce, R., David, D., Aca, E. Q., Hardy, J., Labaja, J., Snow, S.J., Ponzo, A., & Araujo, G. (2018). Long-term photo-identification reveals the population dynamics and strong site fidelity of adult whale sharks to the coastal waters of Donsol, Philippines. Frontiers in Marine Science, 5, 271. McGregor, F., Richardson, A.J., Armstrong, A.J., Armstrong, A.O., & Dudgeon, C.L. (2019). Rapid wound healing in a reef manta ray masks the extent of vessel strike. PloS one, 14(12). Merkl, A., Stuchtey, M., Russel, S., Crow, A., Simon, E., Wooster, J., & Erdmann, H. (2015). Stemming the tide: Land-based strategies for a plastic-free ocean. Ocean Conservancy and McKinsey Center for Business and Environment. Available at: https://oceanconservancy.org/wp-content/uploads/2017/04/full-report-stemming-the.pdf (accessed June 5, 2019). Miller-Rushing, A., Primack, R., & Bonney, R. (2012). The history of public participation in ecological research. Frontiers in Ecology and the Environment, 10(6), 285-290. Minasny, B., & Fiantis, D. (2018) ‘Helicopter research’: who benefits from international studies in Indonesia?. The Conversation. Available at: 262 https://theconversation.com/helicopter-research-who-benefits-from-international-studies-in-indonesia-102165 (accessed June 5, 2019). Ministry of Marine Affairs and Fisheries (2014). Keputusan Menteri Kelautan dan Perikanan Republik Indonesia No4/Kepmen-KP/2014 tentang Penetapan Status Perindungan Penuh Ikan Pari Manta. Jakarta: Ministry of Marine Affairs and Fisheries. Available at: http://kkji.kp3k.kkp.go.id/index.php/ dokumen/regulasi- hukum/keputusan- menteri/finish/14- keputusan- menteri/ 516- kepmen- kp- no- 4- tahun- 2014- tentang- penetapan- status- perlindungan- penuh-ikan-pari-manta (accessed January 2, 2019). Ministry of Marine Affairs and Fisheries. (2016). National Plan of Action: Conservation Management of Sharks and Rays. Jakarta: Ministry of Marine Affairs and Fisheries. Available at: http:// http://www.fao.org/3/a-bt661e.pdf (accessed January 2, 2019). Mora, C., Andrefouet, S., Costello, M.J., Kranenburg, C., Rollo, A., Veron, J., Gaston K.J., & Myers, R. A. (2006). Ecology. coral reefs and the global network of marine protected areas. Science, 312(5781), 1750-1751. Motta, P.J., Maslanka, M., Hueter, R.E., Davis, R.L., De la Parra, R., Mulvany, S.L., Habegger, M.L., Strother, J.A., Mara, K.R., Gardiner, J.M., & Tyminski, J.P. (2010). Feeding anatomy, filter-feeding rate, and diet of whale sharks Rhincodon typus during surface ram filter feeding off the Yucatan peninsula, Mexico. Zoology, 113(4), 199-212. Mourier J, & Planes S. (2013). Direct genetic evidence for reproductive philopatry and associated fine-scale migrations in female blacktip reef sharks (Carcharhinus melanopterus) in French Polynesia. Molecular Ecology, 22, 201-14. Mull, C.G., Blasius, M.E., O’Sullivan, J.B., & Lowe, C.G. (2012). Heavy metals, trace elements, and organochlorine contaminants in muscle and liver tissue of juvenile white sharks, Carcharodon carcharias, from the southern California bight. Global Perspectives on the Biology and Life History of the White Shark. CRC Press, Boca Raton, Florida, 59-75. 263 Mull, C.G., Lyons, K., Blasius, M.E., Winkler, C., O’Sullivan, J.B., & Lowe, C.G. (2013). Evidence of maternal offloading of organic contaminants in white sharks (Carcharodon carcharias). Plos One, 8, e62886. Murie, C.J., & Marshall, A.D. (2016). Mobula kuhlii cleaning station identified at an inshore reef in southern Mozambique. Peerj Preprints, 4:e1724v1 d Murphy, S., Barber, J.L., Learmonth, J.A., Read, F.L., Deaville, R., Perkins, M.W., Brownlow, A., Davison, N., Penrose, R., Pierce, G.J., & Law, R.J. (2015). Reproductive failure in UK harbour porpoises Phocoena phocoena: Legacy of pollutant exposure?. PLoS One, 10(7). Murray, S.P., & Arief, D. (1988). Throughflow into the Indian ocean through the Lombok strait, January 1985-January 1986. Nature, 333(6172), 444. Muthukumar, T., Aravinthan, A., Lakshmi, K., Venkatesan, R., Vedaprakash, L., & Doble, M. (2011). Fouling and stability of polymers and composites in marine environment. International Biodeterioration & Biodegradation, 65(2), 276-284. Nagelkerken, I., Sheaves, M., Baker, R., & Connolly, R. M. (2015). The seascape nursery: a novel spatial approach to identify and manage nurseries for coastal marine fauna. Fish and Fisheries, 16(2), 362-371. Nelms, S.E., Barnett, J., Brownlow, A., Davison, N.J., Deaville, R., Galloway, T.S., Lindeque, P.K., Santillo, D., & Godley, B.J. (2019). Microplastics in marine mammals stranded around the British coast: Ubiquitous but transitory?. Scientific Reports, 9(1), 1075. Nelson, J.D., & Eckert, S.A. (2007). Foraging ecology of whale sharks (Rhincodon typus) within Bahía de Los Angeles, Baja California Norte, Mexico. Fisheries Research, 84(1), 47-64. Niimi, A.J. (1996). Evaluation of PCBs and PCDDFs retention by aquatic organisms. Science of the Total Environment, 192(2), 123-150. Ningsih, N.S., Rakhmaputeri, N., & Harto, A.B. (2013). Upwelling variability along the southern coast of Bali and in Nusa Tenggara waters. Ocean Science Journal, 48(1), 49-57. 264 Nøttestad, L., Giske, J., Holst, J. C., & Huse, G. (1999). A length-based hypothesis for feeding migrations in pelagic fish. Canadian Journal of Fisheries and Aquatic Sciences, 56(S1), 26-34. Norman, B.M., Holmberg, J.A., Arzoumanian, Z., Reynolds, S.D., Wilson, R.P., Rob, D., Pierce, S.J., Gleiss, A.C., De la Parra, R., Galvan, B., & Ramirez-Macias, D. (2017). Undersea constellations: The global biology of an endangered marine megavertebrate further informed through citizen science. Bioscience, 67(12), 1029-1043. Norse, E. (2005). Pelagic protected areas: the greatest parks challenge of the 21st century. Parks, 15, 32-39. Nyegaard, M. (2018). There be giants! the importance of taxonomic clarity of the large ocean sunfishes (genus mola, family molidae) for assessing sunfish vulnerability to anthropogenic pressures. (Murdoch University) O’Malley, M. P., Lee-Brooks, K., & Medd, H.B. (2013). The global economic impact of manta ray watching tourism. Plos One, 8, e65051. O'Malley, M.P., Townsend, K.A., Hilton, P., Heinrichs, S., & Stewart, J. D. (2017). Characterization of the trade in manta and devil ray gill plates in china and South-east Asia through trader surveys. Aquatic Conservation: Marine and Freshwater Ecosystems, 27(2), 394-413. O’Shea, O.R., Kingsford, M.J., & Seymour, J. (2010). Tide-related periodicity of manta rays and sharks to cleaning stations on a coral reef. Marine and Freshwater Research, 61(1), 65-73. Ogata, Y., Takada, H., Mizukawa, K., Hirai, H., Iwasa, S., Endo, S., Mato, Y., Saha, M., Okuda, K., Nakashima, A., & Murakami, M. (2009). International pellet watch: Global monitoring of persistent organic pollutants (POPs) in coastal waters. 1. initial phase data on PCBs, DDTs, and HCHs. Marine Pollution Bulletin, 58(10), 1437-1446. Ogburn, M.B., Harrison, A., Whoriskey, F.G., Cooke, S.J., Mills Flemming, J.E., & Torres, L. G. (2017). Addressing challenges in the application of animal movement 265 ecology to aquatic conservation and management. Frontiers in Marine Science, 4, 70. Oliver, S.P., Hussey, N.E., Turner, J.R., & Beckett, A.J. (2011). Oceanic sharks clean at coastal seamount. PLoS One, 6(3), e14755. Oliver, T.H., Heard, M.S., Isaac, N.J., Roy, D.B., Procter, D., Eigenbrod, F., Freckleton, R., Hector, A., Orme, C.D.L., Petchey, O.L., & Proença, V. (2015). Biodiversity and resilience of ecosystem functions. Trends in Ecology & Evolution, 30(11), 673-684. Ooi, M.S., Townsend, K.A., Bennett, M.B., Richardson, A.J., Fernando, D., Villa, C.A., & Gaus, C. (2015). Levels of arsenic, cadmium, lead and mercury in the branchial plate and muscle tissue of mobulid rays. Marine Pollution Bulletin, 94(1-2), 251-259. Paig-Tran, E., Kleinteich, T., & Summers, A.P. (2013). The filter pads and filtration mechanisms of the devil rays: Variation at macro and microscopic scales. Journal of Morphology, 274(9), 1026-1043. Panti, C., Baini, M., Lusher, A., Hernandez-Milan, G., Rebolledo, E.L.B., Unger, B., Syberg, K., Simmonds, M.P., & Fossi, M. C. (2019). Marine litter: One of the major threats for marine mammals. outcomes from the European cetacean society workshop. Environmental Pollution, 247, 72-79. Papastamatiou, Y. P., DeSalles, P. A., & McCauley, D. J. (2012). Area-restricted searching by manta rays and their response to spatial scale in lagoon habitats. Marine Ecology Progress Series, 456, 233-244. Papastamatiou, Y. P., Watanabe, Y. Y., Bradley, D., Dee, L. E., Weng, K., Lowe, C. G., & Caselle, J. E. (2015). Drivers of daily routines in an ectothermic marine predator: hunt warm, rest warmer?. PLoS One, 10(6). Pardini, A.T., Jones, C.S., Noble, L.R., Kreiser, B., Malcolm, H., Bruce, B.D., Stevens, J.D., Cliff, G., Scholl, M.C., Francis, M., & Duffy, C.A. (2001). Sex-biased dispersal of great white sharks. Nature, 412(6843), 139-140. Parsons, G. R. (1990). Metabolism and swimming efficiency of the bonnethead shark Sphyrna tiburo. Marine Biology, 104(3), 363-367. 266 Parsons, E., Favaro, B., Aguirre, A. A., Bauer, A. L., Blight, L. K., Cigliano, J. A., Fletcher, S. & Foley, M.M. (2014). Seventy-One important questions for the conservation of marine biodiversity. Conservation Biology, 28(5), 1206-1214. Peel, L.R., Stevens, G.M., Daly, R., Daly, C.A.K., Lea, J.S., Clarke, C.R., Collin, S.P., & Meekan, M.G. (2019a). Movement and residency patterns of reef manta rays Mobula alfredi in the Amirante Islands, Seychelles. Marine Ecology Progress Series, 621, 169-184. Peel, L. R., Daly, R., Keating Daly, C. A., Stevens, G. M., Collin, S. P., & Meekan, M. G. (2019b). Stable isotope analyses reveal unique trophic role of reef manta rays (Mobula alfredi) at a remote coral reef. Royal Society Open Science, 6(9), 190599. Perry, C.T., Figueiredo, J., Vaudo, J.J., Hancock, J., Rees, R., & Shivji, M. (2018). Comparing length-measurement methods and estimating growth parameters of free-swimming whale sharks (Rhincodon typus) near the south Ari atoll, Maldives. Marine and Freshwater Research, 69(10), 1487-1495. Perryman, R., & Tapilatu, R.F. (2018) Mantas and leatherbacks-flagship species for conservation in Indonesia. Available at: https://www.researchgate.net/publication/325644411_Mantas_and_Leatherbacks-flagship_species_for_conservation_in_Indonesia. (accessed January 2020). Pierce, K. E., Harris, R. J., Larned, L. S., & Pokras, M. A. (2004). Obstruction and starvation associated with plastic ingestion in a Northern Gannet Morus bassanus and a Greater Shearwater Puffinus gravis. Marine Ornithology, 32, 187-189. Pierce, S.J., Holmberg, J., Kock, A.A., & Marshall, A.D. (2018). Photographic identification of sharks. Shark research: emerging technologies and applications for the field and laboratory. CRC, Boca Raton, FL, USA, 220-234. Pierce, S., & Norman, B. (2016). Rhincodon typus. The IUCN Red List of Threatened Species, e. T19488A2365291. Pillet, S., & Nicolas, J. M. (2005). Biomarkers of immunotoxic effects in the context of environmental risk assessment. Investigative Immunotoxicology. CRC, Boca Raton, FL, USA, 385-405. 267 Pollock, K.H., Nichols, J.D., Brownie, C. & Hines, J.E. (1990). Statistical inference for capture-recapture experiments. Wildlife monographs, 3-97. Putra, M., Lewis, S., Kurniasih, E., Prabuning, D., & Faiqoh, E. (2016). Plankton biomass models based on GIS and remote sensing technique for predicting marine megafauna hotspots in the Solor waters. Paper presented at the IOP Conference Series: Earth and Environmental Science, 47(1) 012015. Putri, A. R., Fujimori, T., & Takaoka, M. (2018). Plastic waste management in Jakarta, Indonesia: evaluation of material flow and recycling scheme. Journal of Material Cycles and Waste Management, 20(4), 2140-2149. United Nations Environment Clean Seas campaign. (2019) Clean Seas. “Clean Seas Countries.” United Nations Environment. Available at: https://www.cleanseas.org/tide-turners#governments (accessed April 17, 2019). QGIS Development Team (2016). (accessed November 1, 2018). R Core Team (2018). R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing. Available at: https://www.R- project.org/ (accessed April 1, 2018). Rambahiniarison, J., Lamoste, M. J., Rohner, C., Murray, R., Snow, S., Labaja, J., Araujo, G., & Ponzo, A. (2018). Life history, growth, and reproductive biology of four mobulid species in the Bohol sea, Philippines. Frontiers in Marine Science, 5, 269. Rasmussen, K., Palacios, D.M., Calambokidis, J., Saborío, M.T., Dalla Rosa, L., Secchi, E.R., Steiger, G.H., Allen, J.M., & Stone, G.S. (2007). Southern hemisphere humpback whales wintering off central America: Insights from water temperature into the longest mammalian migration. Biology Letters, 3(3), 302-305. Reddy, M.L., Reif, J., Bachand, A., & Ridgway, S. (2001). Opportunities for using navy marine mammals to explore associations between organochlorine contaminants and unfavorable effects on reproduction. Science of the Total Environment, 274(1-3), 171-182. 268 Reisser, J., Shaw, J., Wilcox, C., Hardesty, B. D., Proietti, M., Thums, M., & Pattiaratchi, C. (2013). Marine plastic pollution in waters around Australia: characteristics, concentrations, and pathways. PloS one, 8(11). Richardson, A.J. (2008). In hot water: zooplankton and climate change. ICES Journal of Marine Science, 65(3), 279-295. Rigby, C.L., Simpfendorfer, C.A., & Cornish, A. (2019). A practical guide to the effective design and management of MPAs for sharks and rays. Gland, Switzerland: World Wide Fund for Nature. Rios, L.M., Moore, C., & Jones, P.R. (2007). Persistent organic pollutants carried by synthetic polymers in the ocean environment. Marine Pollution Bulletin, 54(8), 1230-1237. Rios, L. M., Jones, P. R., Moore, C., & Narayan, U. V. (2010). Quantitation of persistent organic pollutants adsorbed on plastic debris from the Northern Pacific Gyre's “eastern garbage patch”. Journal of Environmental Monitoring, 12(12), 2226-2236. Ríos-Jara, E., Galván-Villa, C.M., Rodríguez-Zaragoza, F.A., López-Uriarte, E., & Munoz-Fernández, V.T. (2013). The tourism carrying capacity of underwater trails in Isabel Island national park, Mexico. Environmental Management, 52(2), 335-347. Rist, S., Almroth, B. C., Hartmann, N. B., & Karlsson, T. M. (2018). A critical perspective on early communications concerning human health aspects of microplastics. Science of the Total Environment, 626, 720-726. Robichaud, D., & Rose, G. (2001). Multiyear homing of atlantic cod to a spawning ground. Canadian Journal of Fisheries and Aquatic Sciences, 58(12), 2325-2329. Robinson, D. P., Jaidah, M. Y., Bach, S., Lee, K., Jabado, R. W., Rohner, C. A., March, A., Caprodossi, S., Henderson, A.C., Mair, J.M., & Ormond, R. (2016). Population structure, abundance and movement of whale sharks in the Arabian gulf and the Gulf of Oman. PloS One, 11(6), e0158593. Rochman, C.M. (2013). Plastics and priority pollutants: a multiple stressor in aquatic habitats. 269 Rochman, C.M., Kurobe, T., Flores, I., & Teh, S.J. (2014). Early warning signs of endocrine disruption in adult fish from the ingestion of polyethylene with and without sorbed chemical pollutants from the marine environment. Science of the Total Environment, 493, 656-661. Rochman, C. M., Tahir, A., Williams, S. M., Baxa, D. V., Lam, R., Miller, J. T, The, F.C., Werorilangi, S., The, S.J. (2015). Anthropogenic debris in seafood: Plastic debris and fibres from textiles in fish and bivalves sold for human consumption. Nature Science Reports, 5, 14340. Rochman, C. M., Browne, M. A., Underwood, A. J., Van Franeker, J. A., Thompson, R. C., & Amaral-Zettler, L. A. (2016). The ecological impacts of marine debris: unraveling the demonstrated evidence from what is perceived. Ecology, 97(2), 302-312. Rohner, C. A., Pierce, S. J., Marshall, A. D., Weeks, S. J., Bennett, M. B., & Richardson, A. J. (2013). Trends in sightings and environmental influences on a coastal aggregation of manta rays and whale sharks. Marine Ecology Progress Series, 482, 153-168. Rohner, C.A., Armstrong, A.J., Pierce, S.J., Prebble, C.E., Cagua, E.F., Cochran, J. E., Berumen, M.L., & Richardson, A. J. (2015). Whale sharks target dense prey patches of Sergestid shrimp off Tanzania. Journal of Plankton Research, 37(2), 352-362. Rohner, C.A., Burgess, K.B., Rambahiniarison, J.M., Stewart, J.D., Ponzo, A., & Richardson, A. J. (2017). Mobulid rays feed on Euphausiids in the Bohol sea. Royal Society Open Science, 4(5), 161060. Romeo, T., Pietro, B., Pedà, C., Consoli, P., Andaloro, F., & Fossi, M. C. (2015). First evidence of presence of plastic debris in stomach of large pelagic fish in the Mediterranean Sea. Marine Pollution Bulletin, 95(1), 358-361. Rooker, J.R., Secor, D.H., De Metrio, G., Schloesser, R., Block, B.A., & Neilson, J.D. (2008). Natal homing and connectivity in Atlantic bluefin tuna populations. Science, 322(5902), 742-744. 270 Ross, P.S., Ellis, G., Ikonomou, M., Barrett-Lennard, L., & Addison, R. (2000). High PCB concentrations in free-ranging pacific killer whales, orcinus orca: Effects of age, sex and dietary preference. Marine Pollution Bulletin, 40(6), 504-515. Rowat, D., & Brooks, K. (2012). A review of the biology, fisheries and conservation of the whale shark rhincodon typus. Journal of Fish Biology, 80(5), 1019-1056. Ryan, P. G., Moore, C. J., van Franeker, J. A., & Moloney, C. L. (2009). Monitoring the abundance of plastic debris in the marine environment. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1526), 1999-2012. Sampaio, C.L., Leite, L., Reis-Filho, J.A., Loiola, M., Miranda, R.J., José de Anchieta, C., & Macena, B.C. (2018). New insights into whale shark Rhincodon typus diet in brazil: An observation of ram filter-feeding on crab larvae and analysis of stomach contents from the first stranding in Bahia state. Environmental Biology of Fishes, 101(8), 1285-1293. Santana-Garcon, J., Braccini, M., Langlois, T. J., Newman, S. J., McAuley, R. B., & Harvey, E. S. (2014). Calibration of pelagic stereo-BRUVs and scientific longline surveys for sampling sharks. Methods in Ecology and Evolution, 5(8), 824-833. Santos, R.G., Andrades, R., Boldrini, M.A., & Martins, A.S. (2015). Debris ingestion by juvenile marine turtles: An underestimated problem. Marine Pollution Bulletin, 93(1), 37-43. Schmidt, C., Krauth, T., & Wagner, S. (2017). Export of plastic debris by rivers into the sea. Environmental Science & Technology, 51(21), 12246-12253. Schuyler, Q.A., Wilcox, C., Townsend, K.A., Wedemeyer-Strombel, K.R., Balazs, G., Van Sebille, E., & Hardesty, B.D. (2016). Risk analysis reveals global hotspots for marine debris ingestion by sea turtles. Global Change Biology, 22(2), 567-576. Schwartz, C.J. & Arnason, A.N. (1996). A General Methodology for the Analysis of Capture-Recapture Experiments in Open Populations. Biometrics, 52, 860-873. Schwartz, M. K., Luikart, G., & Waples, R. S. (2007). Genetic monitoring as a promising tool for conservation and management. Trends in Ecology & Evolution, 22(1), 25-33. Seber, G.A.F. (1965). A note on the multiple-recapture census. Biometrika, 52, 249-259. 271 Setälä, O., Fleming-Lehtinen, V., & Lehtiniemi, M. (2014). Ingestion and transfer of microplastics in the planktonic food web. Environmental Pollution, 185, 77-83. Setyawan, E., Sianipar, A. B., Erdmann, M. V., Fischer, A. M., Haddy, J. A., Beale, C. S., Lewis, S.A., & Mambrasar, R. (2018). Site fidelity and movement patterns of reef manta rays (Mobula alfredi): Mobulidae using passive acoustic telemetry in northern Raja Ampat, Indonesia. Nature Conservation Research, 3(4), 1-15. Sgrò, C. M., Lowe, A. J., & Hoffmann, A. A. (2011). Building evolutionary resilience for conserving biodiversity under climate change. Evolutionary Applications, 4(2), 326-337. Shane, S.H., Wells, R.S., & Würsig, B. (1986). Ecology, behavior and social organization of the bottlenose dolphin: A review. Marine Mammal Science, 2(1), 34-63. Shanks, A.L. (1983). Surface slicks associated with tidally forced internal waves may transport pelagic larvae of benthic invertebrates and fishes shoreward. Marine Ecology Progress Series. Oldendorf, 13(2), 311-315. Sicular, D.T. (1992). Scavengers, recyclers, and solutions for solid waste management in indonesia. University of California at Berkeley.: Center for Southeast Asia Studies. Skomal, G.B., Zeeman, S.I., Chisholm, J.H., Summers, E.L., Walsh, H.J., McMahon, K.W., & Thorrold, S. R. (2009). Transequatorial migrations by basking sharks in the western Atlantic Ocean. Current biology, 19(12), 1019-1022. Smith, R.J., Veréssimo, D., & MacMillan, D. C. (2010). Marketing and conservation: How to lose friends and influence people. Trade-offs in Conservation: Deciding what to Save, 215-232. Spalding, M.D., Meliane, I., Milam, A., & Fitzgerald, C. (2013). Protecting marine space: Global targets and changing approaches. Ocean YB, 27, 213. Speed, C.W., Field, I.C., Meekan, M.G., & Bradshaw, C.J. (2010). Complexities of coastal shark movements and their implications for management. Marine Ecology Progress Series, 408, 275-293. 272 Sprintall, J., Wijffels, S.E., Molcard, R., & Jaya, I. (2009). Direct estimates of the Indonesian Throughflow entering the Indian ocean: 2004–2006. Journal of Geophysical Research: Oceans, 114(C7) Stamper, M.A., Spicer, C. W., Neiffer, D.L., Mathews, K.S., & Fleming, G.J. (2009). Morbidity in a juvenile green sea turtle (Chelonia mydas) due to ocean-borne plastic. Journal of Zoo and Wildlife Medicine, 40(1), 196-198. Stevens, G. M. W. (2016). Conservation and population ecology of manta rays in the Maldives. (York University) Stevens, G., Hawkins, J., & Roberts, C. (2018). Courtship and mating behaviour of manta rays Mobula alfredi and M. birostris in the Maldives. Journal of Fish Biology, Stewart, J.D., Jaine, F.R., Armstrong, A.J., Armstrong, A.O., Bennett, M.B., Burgess, K.B., Couturier, L.I.E., Croll D.A., Cronin, M.R., Deakos, M.H., Dudgeon, C.L., Fernando, D., Froman, N., Germanov, E.S., Hall., M.A, Hinojosa-Alvarez, S., Hosegood, J.E., Kashiwagi, T., Laglbauer. B.J.L., Lezama-Ochoa., N., Marshall, A.D., McGregor, F., Notarbartolo di Sciara, G.N., Palacios, M.D., Peel, L.R., Richardson, A.J, Rubin, R.D., Townsend, K.A., Venables, S.K., & Stevens, G.M. (2018a). Research priorities to support effective manta and devil ray conservation. Frontiers in Marine Science, 5, 314 Stewart, J. D., Nuttall, M., Hickerson, E. L., & Johnston, M. A. (2018b). Important juvenile manta ray habitat at flower garden banks national marine sanctuary in the northwestern gulf of mexico. Marine Biology, 165(7), 111. Stoler, J., Weeks, J. R., & Fink, G. (2012). Sachet drinking water in ghana's accra-tema metropolitan area: Past, present, and future. Journal of Water, Sanitation and Hygiene for Development, 2(4), 223-240. Sugimoto, T. (1975). Effect of boundary geometries on tidal currents and tidal mixing. Journal of the Oceanographical Society of Japan, 31(1), 1-14. Surinati, D. (2009). Upwelling dan efeknya terhadap perairan laut. Oseana, 34(4), 35-42. Sussarellu, R., Suquet, M., Thomas, Y., Lambert, C., Fabioux, C., Pernet, M.E.J., Le Goïc, N., Quillien, V., Mingant, C., Epelboin, Y. & Corporeau, C. (2016). Oyster 273 reproduction is affected by exposure to polystyrene microplastics. Proceedings of the National Academy of Sciences, 113(9), 2430-2435. Syakti, A. D., Bouhroum, R., Hidayati, N. V., Koenawan, C. J., Boulkamh, A., Sulistyo, I., Lebarillier, S., Akhlus, S., Doumenq, P., & Wong-Wah-Chung, P. (2017). Beach macro-litter monitoring and floating microplastic in a coastal area of indonesia. Marine Pollution Bulletin, 122(1-2), 217-225. Tanaka, K., Takada, H., Yamashita, R., Mizukawa, K., Fukuwaka, M., & Watanuki, Y. (2013). Accumulation of plastic-derived chemicals in tissues of seabirds ingesting marine plastics. Marine Pollution Bulletin, 69(1), 219-222. Teuten, E.L., Saquing, J.M., Knappe, D.R., Barlaz, M.A., Jonsson, S., Bjorn, A., Rowland, S.J., Thompson, R.C., Galloway, T.S., Yamashita, R., & Ochi, D. (2009). Transport and release of chemicals from plastics to the environment and to wildlife. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 364(1526), 2027-2045. Thys, T., Ryan, J.P., Weng, K.C., Erdmann, M., & Tresnati, J. (2016). Tracking a marine ecotourism star: Movements of the short ocean sunfish Mola ramsayi in Nusa Penida, Bali, Indonesia. Journal of Marine Biology. 2016, 8750193. Trochet, A., Courtois, E.A., Stevens, V.M., Baguette, M., Chaine, A., Schmeller, D.S., Clobert, J., & Wiens, J.J. (2016). Evolution of sex-biased dispersal. The Quarterly Review of Biology, 91(3), 297-320. Thorrold, S. R., Afonso, P., Fontes, J., Braun, C. D., Santos, R. S., Skomal, G. B., & Berumen, M. L. (2014). Extreme diving behaviour in devil rays links surface waters and the deep ocean. Nature Communications, 5, 4274. Tillinger, D. (2011). Physical oceanography of the present day Indonesian Throughflow. Geological Society, London, Special Publications, 355(1), 267-281. Tito, C., & Susilo, E. (2017). The correlation of upwelling phenomena and ocean sunfish occurrences in nusa penida, bali. Paper presented at the IOP Conference Series: Earth and Environmental Science,55(1) 012031. 274 Town, C., Marshall, A., & Sethasathien, N. (2013). Manta matcher: Automated photographic identification of manta rays using keypoint features. Ecology and Evolution, 3(7), 1902-1914. Trave, C., Brunnschweiler, J., Sheaves, M., Diedrich, A., & Barnett, A. (2017). Are we killing them with kindness? evaluation of sustainable marine wildlife tourism. Biological Conservation, 209, 211-222. Tyne, J., Loneragan, N., & Bejder, L. (2014). The use of area–time closures as a tool to manage cetacean-watch tourism. Whale-Watching: Sustainable Tourism and Ecological Management, 242-260. Tyne, J.A., Loneragan, N.R., Johnston, D.W., Pollock, K.H., Williams, R., & Bejder, L. (2016). Evaluating monitoring methods for cetaceans. Biological Conservation, 201, 252-260. Ueno, D., Kajiwara, N., Tanaka, H., Subramanian, A., Fillmann, G., Lam, P.K., Zheng, G.J., Muchitar, M., Razak, H., Prudente, M., & Chung, K.H. (2004). Global pollution monitoring of polybrominated diphenyl ethers using skipjack tuna as a bioindicator. Environmental Science & Technology, 38(8), 2312-2316. UNEP (2016). Marine Debris: Understanding, Preventing and Mitigating the Significant Adverse Impacts on Marine and Coastal Biodiversity. Technical Series No.83. Secretariat of the Convention on Biological Diversity, Montreal. UNESCO (2005). The precautionary principle. VassarStats (1998). VasserStats: A Website for Statistical Computation. USA. Available at: http://vassarstats.net/ (accessed December 1, 2018). van Duinkerken (2011) Movements and site fidelity of the reef manta ray, Manta alfredi, along the coast of southern Mozambique. (Utrecht University) van Sebille, E. (2014). Adrift. org. au—A free, quick and easy tool to quantitatively study planktonic surface drift in the global ocean. Journal of Experimental Marine Biology and Ecology, 461, 317-322. 275 van Sebille, E., Wilcox, C., Lebreton, L., Maximenko, N., Hardesty, B. D., Van Franeker, J. A., Eriksen, M., Siegel, D., Galgani, F., & Law, K.L. (2015). A global inventory of small floating plastic debris. Environmental Research Letters, 10(12), 124006. Venables, S., McGregor, F., Brain, L., & van Keulen, M. (2016). Manta ray tourism management, precautionary strategies for a growing industry: A case study from the Ningaloo marine park, western Australia. Pacific Conservation Biology, 22(4), 295-300. Venables, S.K., Marshall, A.D., Germanov, E.S., Perryman, R.J., Tapilatu, R.F., Hendrawan, I.G., Flam, A.L., Van Keulen, M., Tomkins, J.L., & Kennington, W.J. (2019). It's not all black and white: investigating colour polymorphism in manta rays across Indo-Pacific populations. Proceedings of the Royal Society B, 286(1912), 20191879. Venables, S. K., van Duinkerken, D. I., Rohner, C. A., & Marshall, A. D. (2020). Habitat use and movement patterns of reef manta rays Mobula alfredi in southern Mozambique. Marine Ecology Progress Series, 634, 99-114. Verissimo, D., MacMillan, D.C., & Smith, R. J. (2011). Toward a systematic approach for identifying conservation flagships. Conservation Letters, 4(1), 1-8. Victor, D. (2018) 1,000 Pieces of Plastic Found Inside Dead Whale in Indonesia. The New York Times. Available at: https://www.nytimes.com/2018/11/21/world/asia/whale-plastics-indonesia.html (accessed February 11, 2019). Walpole, M.J., & Leader-Williams, N. (2002). Tourism and flagship species in conservation. Biodiversity and Conservation, 11(3), 543-547. Ward-Paige, C.A., Davis, B., & Worm, B. (2013). Global population trends and human use patterns of manta and mobula rays. Plos One, 8(9), e74835. Webb, T.J., & Mindel, B.L. (2015). Global patterns of extinction risk in marine and non-marine systems. Current Biology, 25(4), 506-511. 276 Weishampel, J.F., Bagley, D.A., Ehrhart, L.M., & Rodenbeck, B.L. (2003). Spatiotemporal patterns of annual sea turtle nesting behaviors along an east central Florida beach. Biological Conservation, 110(2), 295-303. Winstanley, G.W. (2016). MantaUtil: manta database utility. Available: http://www.pelagicon.com/software/mantautil/. (accessed March 1, 2017). White, W.T., Giles, J., Dharmadi, & Potter, I.C. (2006). Data on the bycatch fishery and reproductive biology of mobulid rays (myliobatiformes) in indonesia. Fisheries Research, 82, 65-73. White, G.C., & Burnham, K.P. (1999). Program MARK: survival estimation from populations of marked animals. Bird Study, 46, S120-S139. White, W.T., Corrigan, S., Yang, L., Henderson, A.C., Bazinet, A. L., Swofford, D.L., & Naylor, G.J. (2017). Phylogeny of the manta and devilrays (chondrichthyes: Mobulidae), with an updated taxonomic arrangement for the family. Zoological Journal of the Linnean Society, 82(1), 50-75. Whitehead, H. (2007). Selection of models of lagged identification rates and lagged association rates using AIC and QAIC. Communications in Statistics—Simulation and Computation®, 36(6), 1233-1246. Whitehead, H. (2009). SOCPROG: Programs for analyzing social structure. Whitehead, D. A. (2014). Establishing a quantifiable model of whale shark avoidance behaviours to anthropogenic impacts in tourism encounters to inform management actions. (University of Hertfordshire). Wilcox, C., Van Sebille, E., & Hardesty, B. D. (2015). Threat of plastic pollution to seabirds is global, pervasive, and increasing. Proceedings of the National Academy of Sciences. 112(38), 11899-11904. Wilcox, C., Puckridge, M., Schuyler, Q.A., Townsend, K., & Hardesty, B. D. (2018). A quantitative analysis linking sea turtle mortality and plastic debris ingestion. Scientific Reports, 8(1), 12536. Wilson, D.C., Velis, C., & Cheeseman, C. (2006). Role of informal sector recycling in waste management in developing countries. Habitat International, 30(4), 797-808. 277 Wilson, R.P., Shepard, E.L.C., & Liebsch, N. (2008). Prying into the intimate details of animal lives: Use of a daily diary on animals. Endangered Species Research, 4(1-2), 123-137. Wilson, R. M., Kucklick, J.R., Balmer, B. C., Wells, R. S., Chanton, J.P., & Nowacek, D.P. (2012). Spatial distribution of bottlenose dolphins (Tursiops truncatus) inferred from stable isotopes and priority organic pollutants. Science of the Total Environment, 425, 223-230. Worm, B., & Branch, T. A. (2012). The future of fish. Trends in Ecology & Evolution, 27(11), 594-599. Worm, B., Lotze, H. K., Jubinville, I., Wilcox, C., & Jambeck, J. (2017). Plastic as a persistent marine pollutant. Annual Review of Environment and Resources, 42, 1-26. Wu, G., Holan, S.H., Nilon, C.H. & Wikle, C.K. (2015). Bayesian binomial mixture models for estimating abundance in ecological monitoring studies. The Annals of Applied Statistics, 9(1), 1-26. Xanthos, D., & Walker, T. R. (2017). International policies to reduce plastic marine pollution from single-use plastics (plastic bags and microbeads): A review. Marine Pollution Bulletin, 118(1-2), 17-26. Yeater, D., Kuczaj, I., & Stan, A. (2010). Observational learning in wild and captive dolphins. International Journal of Comparative Psychology, 23, 379-385. Yetman, G., Gaffin, S. & Xing, X. Global 15 x 15 minute grids of the downscaled population based on the SRES B2 scenario, 1990 and 2025. NASA Socioeconomic Data and Applications Center (SEDAC). Yordy, J. E., Pabst, D., McLellan, W. A., Wells, R. S., Rowles, T. K., & Kucklick, J. R. (2010). Tissue-specific distribution and whole-body burden estimates of persistent organic pollutants in the bottlenose dolphin (Tursiops truncatus). Environmental Toxicology and Chemistry, 29(6), 1263-1273. Zacharias, M.A., & Roff, J.C. (2001). Use of focal species in marine conservation and management: A review and critique. Aquatic Conservation: Marine and Freshwater Ecosystems, 11(1), 59-76. 278 Zachos, E. (2018) How This Whale Got Nearly 20 Pounds of Plastic in Its Stomach. National Geographic. Available at: https://news.nationalgeographic.com/2018/06/whale-dead-plastic-bags-thailand-animals/ (accessed February 11, 2019). Zelenka, J., & Kacetl, J. (2014). The concept of carrying capacity in tourism. Amfiteatru Economic Journal, 16(36), 641-654. Ziegler, J., Dearden, P., & Rollins, R. (2012). But are tourists satisfied? importance-performance analysis of the whale shark tourism industry on Isla Holbox, Mexico. Tourism Management, 33(3), 692-701. Zelenka, J., & Kacetl, J. (2014). The concept of carrying capacity in tourism. Amfiteatru Economic Journal, 16(36), 641-654. Ziegler, J., Dearden, P., & Rollins, R. (2012). But are tourists satisfied? importance-performance analysis of the whale shark tourism industry on Isla Holbox, Mexico. Tourism Management, 33(3), 692-701.

Indonesia: hooray for manta rays

Abstract

Extracted data

Indonesia: hooray for manta rays

Abstract

Extracted data

Related items

Related items