UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Climate change impacts on living marine resources in the Eastern Tropical Pacific Clarke, Tayler McLellan 2020

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2021_may_clarke_tayler.pdf [ 7.41MB ]
Metadata
JSON: 24-1.0395020.json
JSON-LD: 24-1.0395020-ld.json
RDF/XML (Pretty): 24-1.0395020-rdf.xml
RDF/JSON: 24-1.0395020-rdf.json
Turtle: 24-1.0395020-turtle.txt
N-Triples: 24-1.0395020-rdf-ntriples.txt
Original Record: 24-1.0395020-source.json
Full Text
24-1.0395020-fulltext.txt
Citation
24-1.0395020.ris

Full Text

CLIMATE CHANGE IMPACTS ON LIVING MARINE RESOURCES IN THE EASTERN TROPICAL PACIFIC by  Tayler McLellan Clarke  B.A., Universidad de Costa Rica, 2008 M.Sc., Universidad de Costa Rica, 2013   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) November 2020  © Tayler McLellan Clarke, 2020     ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: CLIMATE CHANGE IMPACTS ON LIVING MARINE RESOURCES IN THE EASTERN TROPICAL PACIFIC  submitted by Tayler McLellan Clarke in partial fulfillment of the requirements for the degree of Doctor of Philosophy In Zoology  Examining Committee: Dr. William W. L. Cheung, Professor, Institute for the Oceans and Fisheries, UBC Supervisor  Dr. Daniel Pauly, Professor, Department of Zoology, UBC Supervisory Committee Member  Maria Maldonado-Pareja, Professor, Department of Earth, Ocean and Atmospheric Sciences, UBC University Examiner Colin Brauner, Professor, Department of Zoology, UBC University Examiner  Additional Supervisory Committee Members: Dr. Evgeny Pakhomov, Professor, Department of Earth, Ocean and Atmospheric Sciences, UBC Supervisory Committee Member Dr. Rashid Sumaila, Professor, Institute for the Oceans and Fisheries, UBC Supervisory Committee Member  iii  Abstract The impacts of climate change on fish stocks are heightened in the tropics, where catch losses are projected to be three to four times the global average. Yet, there are large sub-regional variations in the drivers and magnitudes of shifting species distributions and their implications for fisheries. In this thesis, I aim to better understand the impacts of climate change on fisheries in the Eastern Tropical Pacific Ocean, from Mexico to Peru. First, I applied a species distribution modeling approach to project future impacts on species caught by the main fisheries in the region. Species are projected to shift towards the equator, seeking the more favorable, cooler habitats associated with the Humboldt and equatorial upwelling systems, as well as towards more oxygenated, inshore waters, away from the expanding oxygen minimum zones. Second, I developed and evaluated the performance of a Biogeographically derived Metabolic Index (BDMI). The BDMI can be generalized to assess the combined effects of warming and deoxygenation on the habitat viability of marine fishes and invertebrates. Thirdly, I applied two catch-based indicators derived from the BDMI to analyze the sensitivity of pelagic fisheries in the Eastern Tropical Pacific to warming and deoxygenation between 1970 and 2009. Temperature was the main factor driving oxygen limitation in pelagic catches. In contrast, when I applied these indices to the demersal community along the oxygen minimum zones off the Costa Rican Pacific coast, ambient oxygen was the main factor driving the responses of the exploited community, although species distributions were sensitive to changes in both temperature and oxygen. In both pelagic and demersal environments, I identified potential temperature and oxygen thresholds that separate different exploited communities with different sensitivities to changing temperature and oxygen levels. Overall, I conclude that warming and deoxygenation will likely impact fisheries resources in the region, revealing the importance of expanding our capability and credibility in projecting future changes. By modeling the mechanisms underlying the non-linear responses of biological communities to ocean warming and deoxygenation, the analytical approaches developed in this thesis can facilitate the detection, attribution and projection of climate impacts, including biogeographical shifts and three-dimensional habitat compression.    iv  Lay Summary The ocean is warming and losing oxygen, forcing fish to move towards cooler waters with more oxygen. In this thesis, I study how a changing ocean may affect fish from Mexico to Peru. Using computer algorithms, I found that fish along the warm waters in the Central American Pacific will shift towards the cooler waters near the equator. Simultaneously, the expanding low oxygen areas will squeeze fish into shallower waters where they will be more susceptible to fisheries. Warming and oxygen losses are already affecting fish and fisheries, with warming causing most impacts at the ocean surface and oxygen loss causing most impacts at the ocean bottom. Warming and oxygen losses are probably causing similar impacts throughout the global ocean.  v  Preface  I was primarily responsible for the research included in this thesis. I designed the main research question, compiled the necessary data, conducted the analysis and am the first author of all chapters. Dr. William Cheung was my main advisor and contributed to this thesis with his expertise and guidance, developing ideas, analytical methods and actively discussing the interpretation of the results. Dr. Rashid Sumaila, Dr. Daniel Pauly, and Dr. Evgeny Pahomov were members of my thesis committee and contributed with their expertise, providing ideas, data interpretation and feedback. The four analytical chapters of this dissertation are published, in review, or are being prepared for submission. For chapter one, I wrote the first draft and received feedback from Dr. William Cheung. For chapter two, I defined the research question, compiled and analyzed the data and wrote the first draft of the manuscript. Dr. Gabriel Reygondeau instructed and guided me through the data analysis and continued to collaborate during the writing process. Throughout the entire study, Dr. William Cheung provided supervision and feedback, while Dr. Colette Wabnitz engaged in constant discussion and development of ideas. Dr. Ross Robertson was an active source of data and ideas, Dr. Manuel Ixquiac-Cabrera, Dr. Myrna López, Lic. Ana Rosa Ramírez Coghi, Dr. José Luis del Río Iglesias, Dr. Ingo Wehrtmann provided crucial information and reviewed the manuscript. Chapter two is published: Clarke, T.M., Reygondeau, G., Wabnitz, C., Robertson, R., Ixquiac-Cabrera, M., López, M., Ramírez Coghi, A.R., del Río Iglesias, J.L., Wehrtmann, I., Cheung. W.W.L. (2020). Climate change impacts on living marine resources in the Eastern Tropical Pacific. Diversity and Distributions.  For chapter three, I conducted the analysis and wrote the first draft of the manuscript. Dr. William Cheung developed the idea for the Biogeographically derived metabolic index, Dr. Colette Wabnitz and Dr. Gabriel Reygondeau provided guidance and advice throughout the study, while Dr. Sandra Striegel and Dr. Thomas Frölicher provided crucial oceanographic data analysis and expertise. Chapter three is currently in review: vi  Clarke, T.M., Colette C.C.W., Striegel, S., Frölicher, T.L., Reygondeau, G., Cheung, W.W.L. (in rev.). A new metabolic index to understand the impacts of ocean warming and deoxygenation on global marine fisheries resources.  For chapter four, I designed the concept, carried out the data analysis and wrote the first draft of the manuscript. Dr. William Cheung provided guidance, ideas and feedback, Dr. Colette Wabnitz and Dr. Gabriel Reygondeau collaborated with interpretation of the results and discussion. All co-authors greatly contributed to improving the first draft of the chapter, which I am preparing for submission. For chapter five, I carried out the design, data analysis and writing the first draft. I also participated in field work and data collection. The co-authors of chapter five are Dr. William Cheung, Dr. Colette Wabnitz, Dr. Gabriel Reygondeau, Dr. Thomas Frölicher, Dr. Ingo Wehrtmann, M.Sc. Fresia Villalobos and Dr. Daniel Pauly. Dr. William Cheung supervised the data analysis and manuscript preparation. Dr. Colette Wabnitz and Dr. Gabriel Reygondeau collaborated with interpretation of the results, discussing ideas and providing feedback. Dr. Ingo Wehrtmann and M.Sc. Fresia Villalobos were responsible for the design, field work, data collection, management of the sampling program and interpretation of the results. Dr. Thomas Frölicher contributed to the data analysis of oceanographic data and providing oceanographic expertise. All co-authors contributed to reviewing, editing and providing ideas to improve the first draft of the manuscript. For chapter six, I wrote the first draft and received feedback from Dr. William Cheung and Dr. Colette Wabnitz.  vii  Table of Contents  Abstract ....................................................................................................................................................... iii Lay Summary .............................................................................................................................................. iv Preface ......................................................................................................................................................... v Table of Contents ...................................................................................................................................... vii List of Tables ............................................................................................................................................ xiii List of Figures ........................................................................................................................................... xv Acknowledgements ............................................................................................................................... xviii Dedication ................................................................................................................................................. xix Chapter 1: Introduction .............................................................................................................................. 1 1.1 Ocean warming and deoxygenation and their impacts on fish and fisheries .............................. 1 1.2 Eastern Tropical Pacific Ocean as ‘natural experiment’ .............................................................. 3 1.3 Models to project impacts and risks under climate change ......................................................... 5 1.4 Thesis structure ........................................................................................................................... 6 Chapter 2: Climate change impacts on living marine resources in the Eastern Tropical Pacific Ocean ......................................................................................................................................................... 10 2.1 Introduction ................................................................................................................................ 10 2.2 Methods ..................................................................................................................................... 12 2.2.1 Oceanographic setting of the Eastern Tropical Pacific ..................................................... 12 2.2.2 Fisheries of the Eastern Tropical Pacific........................................................................... 14 2.2.3 Biotic data ......................................................................................................................... 14 2.2.4 Abiotic data ....................................................................................................................... 16 2.2.5 Species distribution models (SDMs) ................................................................................. 17 2.3 Results ....................................................................................................................................... 20 2.3.1 Oceanographic section ..................................................................................................... 20 2.3.2 Species shifts .................................................................................................................... 20 viii  2.3.3 Habitat suitability ............................................................................................................... 22 2.4 Discussion.................................................................................................................................. 28 2.4.1 Species shifts .................................................................................................................... 29 2.4.2 Implications for fisheries and conservation ....................................................................... 31 2.4.3 Model robustness and uncertainty .................................................................................... 33 Chapter 3: A new metabolic index to understand the impacts of ocean warming and deoxygenation on global marine fisheries resources ........................................................................... 35 3.1 Introduction ................................................................................................................................ 35 3.2 Methods ..................................................................................................................................... 36 3.2.1 Biogeographically derived Metabolic index (BDMI)................................................................... 36 3.2.2 Illustrative examples .................................................................................................................. 40 3.2.3 Index comparisons .................................................................................................................... 43 3.3 Results and discussion .............................................................................................................. 44 3.3.1 BDMI predictions ........................................................................................................................... 44 3.3.2 Comparison between BDMI and physiologically derived metabolic index ................................ 44 3.3.3 Uncertainties.............................................................................................................................. 49 Chapter 4: Impact of warming and deoxygenation on pelagic fisheries of the Eastern Tropical Pacific ......................................................................................................................................................... 54 4.1 Introduction ................................................................................................................................ 54 4.2 Methods ..................................................................................................................................... 56 4.3 Catch data.................................................................................................................................. 56 4.3.1 Environmental data ........................................................................................................... 58 4.3.2 Mean Oxygen Demand of the Catch (MODC) and the Biogeographically derived metabolic index for the Catch (BDMC). .............................................................................................. 58 4.3.3 Spatial trends .................................................................................................................... 60 4.3.4 Temporal trends ................................................................................................................ 60 4.4 Results ....................................................................................................................................... 61 ix  4.4.1 Spatial trends .................................................................................................................... 61 4.4.2 Temporal trends ................................................................................................................ 61 4.5 Discussion.................................................................................................................................. 69 Chapter 5: Temperature and oxygen supply shape the demersal community in a tropical oxygen minimum zone ........................................................................................................................................... 73 5.1 Introduction ................................................................................................................................ 73 5.2 Methods ..................................................................................................................................... 75 5.2.1 Demersal species community survey ............................................................................... 75 5.2.2 Community-level indicators (Biogeographically derived Metabolic Index of the Catch, Mean Oxygen Demand of the Catch) ................................................................................................. 77 5.3 Results ....................................................................................................................................... 79 5.4 Discussion.................................................................................................................................. 88 Chapter 6: General discussion and conclusions................................................................................... 92 6.1 Synthesis of the main findings ................................................................................................... 92 6.2 Key uncertainties ....................................................................................................................... 94 6.3 Future research areas ............................................................................................................... 96 6.4 Recommendations for climate-smart fisheries in the Eastern Tropical Pacific Ocean .............. 98 Bibliography ............................................................................................................................................ 101 Appendices .............................................................................................................................................. 132 Appendix A Chapter 2: Climate change impacts on living marine resources in the Eastern Tropical Pacific ................................................................................................................................................... 132 A.1 Species caught by the coastal small-scale, large pelagics, small pelagics, shrimp-trawl bycatch and shrimp-trawl fisheries in the tropical Eastern Pacific. .................................................. 132 A.2 Geographic occurrence records for all the 505 modeled marine and invertebrate species……………………………………………………………………………………………………….151 A.3 Results of Generalized Linear Models utilized to downscale environmental parameters in the tropical Eastern Pacific.. ............................................................................................................. 152 x  A.4 Surface and bottom environmental parameters selected by the Ecological Niche Factor Analysis (ENFA) as important in determining the species environmental niche. ............................. 153 A.5 Average Area Under the Curve for each species distribution model (SDM) and Earth system model (ESM) for 505 marine fish and invertebrate species in the Eastern Tropical Pacific. 154 A.6 Percentage of the 505 study species according to the proportional change of habitat suitability in the Pacific Exclusive Economic Zones between 2001-2020 and 2041-2060. .............. 155 Appendix B Chapter 3: A new metabolic index to understand the impacts of ocean warming and deoxygenation on global marine fisheries resources ........................................................................... 156 B.1 Relationship between the BDMI estimated with oxygen in atm using different temperature dependence parameters (j1,j2) and the physiologically derived metabolic index (Penn et al., 2018). ……………………... .......................................................................................................................... 156 B.2 Relationship between the BDMI estimated with oxygen in atm with a metabolic scaling parameter of 0.7 in red and of 0.9 in yellow based on Pauly and Cheung (2018) and the physiologically derived metabolic index (Penn et al., 2018). ........................................................... 157 B.3 Relationship between the BDMI estimated with oxygen in atm (red) and oxygen in mol/m3 (blue) and the physiologically derived metabolic index (Penn et al., 2018)………………………….158 B.4 Biological and ecological input parameters necessary to estimate the Biogeographically derived metabolic index (BDMI) for Gadus morhua, Diplodus puntazzo and Callinectes sapidus……………………………………………………………………………………………………….159 B.5 Comparing the habitat loss (%) and Φcrit projected by the physiologically derived metabolic index (MI, Penn et al., 2018) and different parametrizations of the BDMI ....................... 161 B.6 Linear regressions between the BDMI and the physiologically derived metabolic index (Penn et al., 2018) for the baseline period (1971-2000)………………………………………………..162 B.7 Linear regressions between the change in BDMI and the physiologically derived metabolic index (Penn et al., 2018) by end of century (2071-2100) relative to the baseline period (1971-2000). ..................................................................................................................................... 163 xi  B.8 Linear regressions between the physiologically derived metabolic index and the Biogeographically derived metabolic index (BDMI) for the baseline period (1971-2000). ............... 164 B.9 Linear regressions between the percent change in BDMI and the physiologically derived metabolic index (Penn et al., 2018) by end of century (2071-2100) relative to the baseline period (1971-2000). ..................................................................................................................................... 165 B.10 Percent habitat loss by end of century (2071-2100) relative to baseline conditions (1971-2000) and Φcrit values estimated by the Biogeographically derived metabolic index computed using six different percentiles to define the species oxygen threshold . .................................................... 166 Appendix C Chapter 4: Impact of warming and deoxygenation on pelagic fisheries of the Eastern tropical Pacific ....................................................................................................................................... 167 C.1 Temperature preferences and oxygen thresholds of all the species included in the computation of BDMC and MODC for the Eastern Tropical Pacific Ocean. . .................................. 167 C.2 Average sea surface temperature (°C) in Exclusive Economic Zones of the Eastern Tropical Pacific Ocean (1970-2009). ................................................................................................ 168 C.3 Average sea surface oxygen (mol/m3) in the Exclusive Economic Zones of the Eastern Tropical Pacific Ocean (1970-2009) ................................................................................................. 169 C.4 Biogeographically-Derived Metabolic Index of the Catch (BDMC) time series for Exclusive Economic Zones of the Eastern Tropical Pacific Ocean (1970-2009) ............................................. 170 C.5 Mean Oxygen Demand of the Catch (MODC, mol/m3) time series for EEZ in the Eastern Tropical Pacific Ocean (1970-2009) ................................................................................................. 171 C.6 Coherence between Mean Oxygen Demand of the Catch and the Oceanic Niño Index for pelagic fisheries in the Eastern Tropical Pacific. . ............................................................................ 172 C.7 Coherence between the Biogeographically derived Metabolic Index of the Catch of pelagic fisheries in the Eastern Tropical Pacific Ocean and the Oceanic Niño Index. . .................. 174 C.8 Catches of pelagic species in Pacific Exclusive Economic Zones between 1970 and 2009, represented as a percentage of its total pelagic catches. . .................................................... 176 xii  Appendix D Chapter 5: Temperature and oxygen supply shape the demersal community in a tropical Oxygen Minimum Zone ......................................................................................................................... 178 D.1 Area Under the Curve comparing presence with predicted presence based on a Φcrit threshold of the local minimum. ........................................................................................................ 178  xiii  List of Tables Table 2.1. Percent change in the habitat suitability projected by 2041-2060 relative to 2001-2020 for species groups caught in the four main fisheries in the Pacific Exclusive Economic Zones of Mexico to Peru. ............................................................................................................................................................ 23 Table 2.2. Projected species turnover (%) by 2041-2060 relative to 2001-2020 for species groups caught in the four main fisheries in the Pacific Exclusive Economic Zones from Mexico to Peru. ......................... 25 Table 3.1. Data requirements for the Biogeographically derived Metabolic Index, with the values of the constants and fixed coefficients, as well as the units for each parameter. ................................................. 40 Table 3.2. Biological and ecological input parameters necessary to estimate the Biogeographically derived metabolic index (BDMI) for Gadus morhua, Diplodus puntazzo and Callinectes sapidus... .......... 42 Table 3.3. Habitat loss by 2071-2100 relative to 1971-2000 according to the Biogeographically derived Metabolic Index (BDMI) and the Physiologically derived Metabolic Index. ................................................ 45 Table 4.1. Segmented regression analyses between sea surface oxygen concentration (mol/m3) and Mean Oxygen Demand of the Catch (MODC) as well as between sea surface temperature and MODC of pelagic fisheries in the Eastern Tropical Pacific (from Mexico to Ecuador) from 1970 to 2009. ................ 64 Table 4.2. Segmented regression analysis between sea surface oxygen (mol/m3) and mean Biogeographically derived Metabolic Index of the Catch (BDMC) as well as between sea surface temperature and BDMC of pelagic fisheries in the Eastern Tropical Pacific Ocean (from Mexico to Ecuador) from 1970 to 2009.. ..................................................................................................................... 65 Table 4.3. Linear regression analyses between the normalized Mean Oxygen Demand of the Catch (MODC) and the mean Biogeographically derived Metabolic Index of the Catch (BDMC) of pelagic fisheries in the Eastern Tropical Pacific Ocean from 1970 to 2009. ........................................................... 66 Table 5.1. Analysis of variances of the differences in temperature, dissolved oxygen concentration (mol/m3), BDMC, MODC (mol/m3) and depth centroids (m) across sampling years (2008-2011). ............ 81 Table 5. 2.Tukey tests to identify the years with differences in temperature, dissolved oxygen concentration (mol/m3), BDMC, MODC (mol/m3) and depth centroids (m) of demersal species along the Pacific coast of Costa Rica.. ....................................................................................................................... 82 xiv  Table 5.3. Break-points and adjusted R squared of the segmented regressions between BDMC and oxygen and temperature, as well as MODC, oxygen and temperature. ..................................................... 83 Table 5.4. Results of the segmented regression between each indicator (BDMC, MODC) and, temperature, oxygen and their interaction. ................................................................................................. 84 Table 5.5. Mean abundance of species in the communities in low oxygen and higher oxygen environments along the Pacific coast of Costa Rica. .................................................................................. 87  xv  List of Figures Figure 1.1. Map of the main oceanographic features in the Eastern Tropical Pacific taken from Fiedler and Lavin (2017). ................................................................................................................................................. 4 Figure 1.2. Four main chapters of the thesis. ............................................................................................... 9 Figure 2.1. Exclusive Economic Zones in the Eastern Tropical Pacific of interest in this study. ................ 11 Figure 2.2. Temperature, oxygen, pH, net primary production and salinity) at the ocean surface in 2001-2020 (left most panel) and anomalies by 2041-2060 under RCP 2.6 (middle panel) and RCP 8.5 (right most panel).. ............................................................................................................................................... 13 Figure 2.3. Oceanographic conditions at the ocean seafloor in 2001-2020 and anomalies by 2041-2060 under RCP 2.6 and RCP 8.5. ...................................................................................................................... 15 Figure 2.4. Direction and distance (km) of geographic shifts undergone by species groups (classified according to fisheries) by 2041-2060 relative to 2001-2020 under RCP 8.5. ............................................. 21 Figure 2.5. Mean shifts in depth centroids and associated standard deviation for demersal species by 2041-2060 relative to 2001-2020 under RCP 8.5. ...................................................................................... 21 Figure 2.6. Projected change in the habitat suitability of species groups caught in the four main fisheries by 2041-2060 relative to present (2001-2020) under RCP 8.5. .................................................................. 24 Figure 2.7. Projected local extirpation rate of species groups caught in the four main fisheries between for 2041-2060 relative to 2001-2020 under RCP 8.5.. ..................................................................................... 26 Figure 2.8. Projected local invasion rate of species groups caught in the four main fisheries between for 2041-2060 relative to 2001-2020 under RCP 8.5. ...................................................................................... 27 Figure 3.1. Relationship between the Biogeographically derived metabolic index and the physiologically derived metabolic index (Penn et al., 2018) for baseline conditions (1971-2000) for three species: crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua). .................................... 46 Figure 3.2. Linear regressions for the percent change by the end of the century (2071-2100) relative to the reference period (1971-2000) between the Biogeographically derived metabolic index and the physiologically derived metabolic index (Penn et al., 2018) for three species: crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua). ........................................................................ 48 xvi  Figure 3.3. Habitat loss by end of the century (2071-2100) relative to the baseline period (1971-2000) estimated by both indices (red), the Biogeographically derived metabolic index (light blue) and the physiologically derived metabolic index (green) for crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua). ......................................................................................................... 49 Figure 3.4. Range of temperature preferences and oxygen thresholds estimated from 10’000 subsamples randomly selected across a species' distribution ........................................................................................ 53 Figure 4.1. Hypothesized trends of mean oxygen demand of the catch (MODC) and biogeographic-derived metabolic index of the catch (BDMC) in subtropical and tropical pelagic catches.. ...................... 57 Figure 4.2. Average sea surface temperature (°C) and sea surface oxygen concentration (mol/m3) (left panel) and their rate of change (right panel) in exclusive economic zones of the Eastern Tropical Pacific Ocean from 1970 to 2009 ........................................................................................................................... 62 Figure 4.3. Segmented regressions between A) sea surface temperature (°C) and the Mean Oxygen Demand of the Catch (MODC); B) sea surface temperature (°C) and the mean Biogeographically derived metabolic index of the catch (BDMC); C) sea surface oxygen (mol/m3) and MODC; D) sea surface oxygen (mol/m3) and BDMC. ...................................................................................................................... 63 Figure 4.4. Correlation between the Mean Temperature of the Catch (MTC), the Mean Oxygen Demand of the Catch (MODC), the Biogeographically derived Metabolic Index (BDMC), sea surface oxygen (mol/m3), sea surface temperature (°C), Oceanic Niño Index (ONI) and year. .......................................... 67 Figure 4.5. Standardized time series of the Mean Oxygen Demand of the Catch (MODC) and Mean Biogeographically derived Metabolic Index of the Catch (BDMC) of pelagic fisheries in the exclusive economic zones in the Eastern Tropical Pacific Ocean. ............................................................................. 68 Figure 5.1. Trawl sample locations along the Pacific coast of Costa Rica, 2008-2011. Sampling was carried out at 145-350 m depths. ................................................................................................................ 76 Figure 5.2. Temperature (A), dissolved oxygen concentration (B), Biogeographically derived Metabolic Index of the Catch (BDMC) (C), and Mean Oxygen Demand of the Catch (MODC) (D) across sampling years along the Pacific coast of Costa Rica (150 – 350 m). ....................................................................... 80 xvii  Figure 5.3. Depth centroids of demersal species along the Pacific coast of Costa Rica (150 – 350 m) from 2008 to 2011. .............................................................................................................................................. 81 Figure 5.4. Segmented regression between Biogeographically derived Metabolic Index of the Catch (BDMC) and temperature, oxygen and its interaction. ................................................................................ 85 Figure 5.5. The segmented regressions between Mean Oxygen Demand of the Catch (MODC) and temperature, oxygen and its interaction. ..................................................................................................... 86                xviii  Acknowledgements  I am grateful to Dr. William Cheung for teaching me so much with patience and humility. Thanks to my supervisory committee Dr. Daniel Pauly, Dr. Rashid Sumaila and Dr. Evgeny Pakhomov for your guidance and constructive feedback. I also appreciate the mentorship of Dr. Colette Wabnitz and Dr. Gabriel Reygondeau. Thank you to Dr. Muhammed Oyinlola and Dr. Ravi Maharaj, who were always one step ahead and there to answer questions and discuss ideas. I am grateful for the support and discussions with all the members of my lab and beyond PhD. Cand.  Juliano Palacios, Dr. Travis Tai, MSc. Patricia Angkiriwang, Dr. Juan Jose Alava, Dr. Oai Li Chen, PhD. Cand. Virginie Bornarel, MSc. Emilie Stump, PhD. Cand. Tanvi Vaidyanathan, PhD. Cand. Santiago de la Puente, PhD. Cand. Rocio Lopez de la Lama, Kristen Sora, Ambre Soszynsky, Dr. Lydia Teh, Dr. Andres Cisneros Montemayor and MSc. Melanie Ang. I also thank the UNIP-CIMAR, GIACT-CIMAR, my masters supervisor Dr. Ingo Wehrtmann, and lab mates Dr. Mario Espinoza, MSc. Fresia Villalobos-Rojas, MSc. Raquel Romero, MSc. Juan Carlos Azofeifa, MSc. Juliana Herrera, MSc Jaime Nivia, Dr. Patricio Hernáez, MSc. Catalina Benavides and so many more for preparing me for a PhD. I could not have done this without funding from the Fondo de Incentivos of the Consejo Nacional para Investigaciones Científicas y Tecnológicas (CONICIT) and the Ministerio de Ciencia, Tecnología y Telecomunicaciones (MICITT) of Costa Rica, the University of British Columbia and the Natural Sciences and Research Council of Canada (Discovery Grant). I am grateful to my partner Esteban Adams, for always being there with encouragement, lots of patience, help and love. I thank my very large family for showing me the importance of education and hard work, as well as for the support along the way, Mom, Dad, Audrey, Emily, Justin, Bastian, Vivian, Alexander, Kyla, Jimmy, Alison, Henry, Betsy and Lauren. xix  Dedication  To nature and the unknown for inspiring me. To my partner, family and friends for the love and support.1  Chapter 1: Introduction Marine ecosystems and fisheries play an important role in providing food, supporting livelihoods and generating income for the growing world population (FAO, 2020). Yet, climate change and unsustainable fisheries are affecting the ocean´s ability to continue providing these benefits (Bindoff et al., 2019; FAO, 2020; Sumaila et al., 2011). The impacts of climate change on fish stocks are heightened in the tropics, where catch losses are projected to be three to four times the global average (Bindoff et al. 2019; FAO 2020). In addition, a large percentage of fish stocks are unmanaged and overfished, making them particularly sensitive to climate impacts (FAO, 2020). On the other hand, effective fisheries management and climate change mitigation can reduce potential catch losses (Sumaila & Tai, 2020). Rebuilding over-exploited fish stocks could be a cost-effective way to recover some benefits derived from fisheries and compensate for the negative impacts of climate change on potential catches (Bindoff et al., 2019; Cheung et al., 2018; Gaines et al., 2018; Sumaila & Tai, 2020). However, effective policies designed to increase fisheries resilience must be informed by local estimates of potential climate impacts, as their drivers and magnitude vary greatly between regions (Morley et al., 2018).  1.1 Ocean warming and deoxygenation and their impacts on fish and fisheries Warming and deoxygenation are amongst the most important climate-related stressors for marine ecosystems and fisheries (Breitburg et al., 2018; Poloczanska et al., 2016). Globally, the ocean has warmed at a rate of 4.83-5.79 (0-700m) and 3.05-4.99 (700-200m) ZJ yr–1 (energy accumulation in ZJ, where1 ZJ = 1021 J) between 2005 and 2017 (Bindoff et al., 2019). Ocean warming and changes in water column ventilation have lowered the global ocean oxygen content by 1-3% and expanded oxygen minimum zones by 3-8% over the past 50 years (Bindoff et al., 2019; Breitburg et al., 2018; Schmidtko et al., 2017). Ocean warming and deoxygenation are projected to continue under climate change, warming 2150 ZJ (high emissions scenario, RCP 8.5) or 900 ZJ (low emissions scenario, RCP 2.6), and losing 3.2-3.7% (RCP 8.5) or 1.6–2.0% (RCP 2.6) of the global oxygen content by 2100 (Bindoff et al., 2019). Oxygen minimum zones are also projected to keep growing 1.4-12.6% by 2100 (RCP 8.5) relative to 1850–1900 (Bindoff et al., 2019).  2  Temperature and oxygen are key factors that determine the distribution, abundance and size of commercial species (Bindoff et al., 2019). Temperature and oxygen are closely linked with physiological responses at the organismal level, which then propagate through the population (Breitburg et al., 2019; Cheung, 2018). Temperature and oxygen affect the physiological performance of marine ectothermic organisms by limiting the amount of oxygen demanding processes that can occur simultaneously, such as survival, growth, movement, defense, feeding and reproduction. Organisms have a higher aerobic scope (difference between the maintenance metabolic rate and the maximum metabolic rate) within a certain temperature range, where physiological performance is maximized (Forster et al., 2012; Hoefnagel & Verberk, 2015; Pörtner et al., 2017; Rollinson & Rowe, 2018). Species follow the thermal window they are adapted to in both time and space, by shifting the onset of biological events (e.g. spawning, migration) earlier in the year (Poloczanska et al., 2013), and shifting their geographic distributions towards cooler waters (Poloczanska et al., 2016). When temperature increases above the maximum performance range, fish body size and reproductive output decline (Baudron et al., 2014; Cheung et al., 2013a; Pörtner et al., 2017).  The impacts of warming are intensified by deoxygenation. Combined, both stressors narrow an organism’s aerobic scope by increasing its maintenance metabolic rate and reducing its maximum metabolic rate (Leiva et al., 2019). An inability to maintain the metabolic scope will lead to trade-offs across growth, reproduction, digestion, defense, among others, that can eventually lower population abundance and alter population size structure. Under continued warming and deoxygenation, the oxygen available in the organism’s current ocean environment may not be enough to sustain its resting metabolic rate, forcing the species distribution to shift towards cooler, more oxygenated waters (Pauly & Cheung, 2018). Documented impacts of deoxygenation on commercial fish and invertebrates include reductions in species richness, body size, growth rates, catches and catch per unit effort (Arntz et al., 2006; Keller et al., 2015; Levin et al., 2009; Sato et al., 2017). Oxygen limitation can also change the way species interact with fisheries, for example, expanding oxygen minimum zones can compress hypoxia-intolerant fish closer to the surface, increasing their overlap with fishing grounds. Expanding oxygen minimum 3  zones can also squeeze demersal species closer to the coastline, where they would be more accessible to fisheries (Prince et al., 2010; Prince & Goodyear, 2006; Stramma et al., 2012). Despite the importance of oxygen in maintaining global fishery resources, most existing global models used to project climate impacts on commercial species do not include oxygen (Tittensor et al., 2018).   1.2  Eastern Tropical Pacific Ocean as ‘natural experiment’ Warming, deoxygenation and acidification are expected to be key drivers of climate impacts on fisheries in the Eastern Tropical Pacific Ocean (Lluch-Cota et al., 2019). Currently, El Niño Southern Oscillation causes a high interannual variability in temperature and oxygen levels that obscures long-term trends within the Eastern Tropical Pacific Ocean (Hameau et al., 2019; Rodgers et al., 2015). The El Niño Southern Oscillation has a three to seven-year cycle with a neutral phase, a warm El Niño phase and a cool La Niña phase (Fiedler & Lavín, 2017). During El Niño, upwelling intensity decreases, shoving the thermocline deeper and increasing the amount of heat and oxygen in the top mixed layer of the ocean (Leung et al., 2019). During La Niña, upwelling intensifies, bringing the thermocline closer to the surface, cooling waters and expanding oxygen minimum zones (Leung et al., 2019).  During the warm El Niño periods, species within the Eastern Tropical Pacific Ocean shift towards the cooler waters produced by upwelling systems to the north (California Current) or the south (Humboldt Current, equatorial upwelling) (Funes‐Rodríguez et al., 2011; McClatchie et al., 2018; Schwing et al., 2005; Sielfeld et al., 2010). Local losses, population recoveries and range extensions are common responses to the inter-annual temperature variability in the Eastern Tropical Pacific Ocean (Mora & Robertson, 2005). Reports from as early as 1982 show that tropical shrimp shift south towards Peru during El Niño (Barber & Chavez, 1983). More recent studies documented the presence of warm-adapted species in the cooler waters of Chile and the California Current during El Niño years (Funes‐Rodríguez et al., 2011; McClatchie et al., 2018; Schwing et al., 2005; Sielfeld et al., 2010). Long-term climate change in the Eastern Tropical Pacific Ocean (Fiedler & Lavín, 2017) has increased sea surface temperature by 0.4–1.0 °C since 1900, with proportional declines in sea surface oxygen solubility (Fiedler & Lavín, 2017). Although dissolved oxygen in the surface mixed layer is fully 4  saturated, below there are large areas with extremely low dissolved oxygen concentrations (below 80 mmol/m3) (Hameau et al., 2020; Oschlies et al., 2018). These oxygen minimum zones form near eastern boundary upwelling systems where the high respiration rates deplete oxygen levels, and ventilation is low. Oxygen minimum zones have expanded throughout the Eastern Tropical Pacific and are projected to continue expanding under global warming (Fiedler & Lavín, 2017). Coastal upwelling plays an important role in shaping the temporal trends and spatial patterns of temperature and dissolved oxygen in the Eastern Tropical Pacific Ocean (Fig. 1.1; Fiedler & Lavín, 2017).    Figure 1.1. Map of the main oceanographic features in the Eastern Tropical Pacific taken from Fiedler and Lavin (2017). Shades represent sea surface temperatures, from dark (18°C) to light (29°C). STSW: subtropical surface water, TSW: tropical surface water, ESW: equatorial surface water.  The California Current, Humboldt Current and equatorial upwelling systems cool water in the northern and southern margins of the region (Fiedler & Lavín, 2017). There is a large uncertainty surrounding how climate change may affect upwelling systems (Bindoff et al., 2019; Fiedler & Lavín, 2017; Lluch-Cota et al., 2019). On one hand, global trends show an increase in ocean stratification, which would reduce upwelling intensity and lead to a warming trend. On the other hand, wind strength is increasing, which would intensify upwelling and lead to a cooling trend (Bakun et al., 2015; Bakun, 1990). Consequently, 5  the future of upwelling will depend on whether the increasing wind strength can compensate for the additional energy required to mix the increasingly stratified waters (Bindoff et al., 2019). In the past decades, wind strength and upwelling systems have increased in both the California and the Humboldt currents, producing strong deoxygenation and acidification trends, as well as a cooling of the Gulf of California. Current projections indicate deoxygenation and acidification will strengthen, with impacts on the very productive fisheries associated with these upwelling systems (Bakun et al., 2015).  The cool, Humboldt and California upwelling regions enclose the Eastern Pacific warm pool along Central America, where temperatures have increased around 1 °C since 1900 (Lluch-Cota et al., 2019). A further increase of 2 °C is projected with a doubling of CO2, which may be reached by 2080 (Fiedler & Lavín, 2017). In general, losses in catch potential are expected along Central America, as species shift north or south, following the water temperatures they are adapted to. Global projections show that in general, species are shifting towards cooler waters, leading to losses in diversity, biomass and potential catches in the tropics (Cheung et al., 2010, 2016; Jones & Cheung, 2015; Lotze et al., 2019). Potential catch for exclusive economic zones in the Eastern Tropical Pacific Ocean may decline 13 - 30% by mid-century under a high emissions scenario (Cheung et al., 2018), much lower than projected catch losses for the Western Tropical Pacific Ocean (Cheung et al., 2018). Seasonal upwelling in Tehuantepec (Mexico), the Costa Rica Dome and Panama temporarily cool waters and enhance productivity along Central American Pacific, which may act as a small refuge for marine species. The degree to which these upwelling systems may function as climate refuges will depend on the magnitude of oxygen minimum zone expansions, as well as the balance between wind strength and stratification.  1.3 Models to project impacts and risks under climate change Species distribution modeling is the most common approach taken to model the impacts of climate change on marine biogeography and biodiversity (Evans et al., 2015; Koenigstein et al., 2016; Urban, 2019). Their low data requirements allow species’ distribution models to broadly and rapidly inform climate change policy (Evans et al., 2015). However, the correlative approaches applied in most species’ distribution models are less accurate when projecting into the future under novel climates. This 6  uncertainty may be reduced by identifying and modeling the physiological mechanisms that drive changes in species distributions. Yet, this type of mechanistic modeling has high data requirements that prevent its widespread application, so hybrid models are being developed based on generalized biological principles that define how marine ectothermic organisms relate to their environment (Urban, 2019). For example, previous studies have linked metabolism to temperature and oxygen in order to project changes in growth and body size (Cheung et al., 2013a; Evans et al., 2015; Koenigstein et al., 2016).  Projecting the future impacts of climate change comes with uncertainty surrounding future emissions, model structure, and species responses to climate change. Uncertainty in future emissions can be accounted for by projecting models into the future under different carbon emission scenarios. In this thesis, I use two climate change scenarios, the high mitigation/low emissions scenario (Representative Concentration Pathway, RCP 2.6) and the low mitigation/high emissions scenario (Representative Concentration Pathway, RCP 8.5). In RCP 2.6 radiative forcing reaches 2.6 W/m2 by end of the 21st century, while in RCP 8.5 it reaches 8.5 W/m2 (van Vuuren et al., 2011). These climate change scenarios are used as input for Earth system model runs, allowing for better comparisons across modeled climate, biological, ecological and socioeconomic impacts. Uncertainty in model structure can be accounted for through multi-model approaches that represent the range of projections from different correlative and mechanistic models (Eddy, 2019; Guo et al., 2015). This approach can increase our confidence in regions where model projections converge (Eddy, 2019; Evans et al., 2015).  1.4 Thesis structure In this thesis, I assess the effects of climate change on marine fish and invertebrates caught throughout the Eastern Tropical Pacific Ocean, in order to infer climate-impacts on fisheries. I begin with a correlative approach to identify the main climate stressors on regional biogeography: warming and deoxygenation. While the impacts of warming on fish and fisheries have been widely studied, the impacts of deoxygenation have only been recently recognized (Limburg et al., 2017). This thesis proposes a framework that can be used to assess the synergistic effects of warming and deoxygenation on living marine resources. This framework builds on existing knowledge about the biological principals that 7  characterize the relationship between temperature, oxygen and the physiology of marine organisms (Cheung et al., 2013a; Cheung et al., 2011; Pauly, 2010; Pauly & Cheung, 2018). I then applied this framework to better understand how warming and deoxygenation has impacted marine living resources in the Eastern Tropical Pacific Ocean. Specifically, this thesis will examine the following hypotheses:  (1) Climate change will drive species in the Eastern Tropical Pacific Ocean to shift towards cooler waters that are brought up to the surface through upwelling (see Chapter 2);  (2) The impacts of warming and deoxygenation on marine ectothermic organisms can be modeled using biological principles that relate organisms´ physiology to their environment (see Chapter 3);  (3) Warming drives changes in physiological traits and metabolic indices of commercial pelagic species in the Eastern Tropical Pacific Ocean (see Chapter 4) (4) Changing ambient oxygen and temperature contributed to ENSO-driven shifts in the demersal community along the oxygen minimum zone in the Pacific coast of Costa Rica (see Chapter 5). In Chapter 2, I examine how changes in temperature, oxygen level and other climatic factors may affect commercial fish and invertebrate populations in the Eastern Tropical Pacific Ocean (from Mexico to Peru). I apply a species distribution modeling approach, which is based on the concept of environmental niche, defined as a set of environmental conditions that would allow a species to exist indefinitely (Hutchinson, 1957). This approach can help elucidate the complexity of biogeographical responses of marine species in sub-regional scales even in data-limited areas (Guisan & Thuiller, 2005).  In Chapter 3, I develop a tool that is based on mechanistic understanding of the linkages between temperature, oxygen and ecophysiology of marine ectotherms to further assess the impacts of warming and deoxygenation on marine fishes and invertebrates. I develop and examine an index called the Biogeographically derived Metabolic Index (BDMI). The BDMI is a ratio of oxygen supply to oxygen demand that aims to provide estimates of habitat viability for marine fishes and invertebrates. In this approach, oxygen demand is derived from a mathematical model developed by Cheung et al. (2013a), Cheung et al. (2011) and Pauly (2010). The model is built on the von Bertalanffy´s bioenergetic growth model (von Bertalanffy, 1957), where growth is represented as a function of anabolism and catabolism. 8  Both metabolic processes scale with temperature, while anabolism also depends on oxygen supply. Anabolic processes are limited by oxygen supply because the body and its demand for oxygen grows in three dimensions (i.e., as a volume), while the gills that supply oxygen to the body grow as a surface (Pauly & Cheung, 2018). Catabolism increases at a faster rate than anabolism, until eventually catabolism equals anabolism and growth stops (Cheung et al., 2013a; Pauly & Cheung, 2018). Here I assume that a population cannot be sustained if growth is not possible.  In Chapters 4 and 5, I use the BDMI to develop two catch-based indicators: the Biogeographically derived Metabolic Index of the Catch (BDMC), and Mean Oxygen Demand of the Catch (MODC), to represent changes in the potential physiological constraints and sensitivity to ocean biogeochemistry for species in fisheries catches. I apply these indicators to better understand the impacts of warming and deoxygenation on pelagic fisheries in the Eastern Tropical Pacific Ocean (Chapter 4), and on a demersal marine community in the oxygen minimum zones margins off the Pacific coast of Costa Rica (Chapter 5). I expect that as the ocean warms and oxygen minimum zones expand, species with high oxygen demands will be impacted by oxygen limitation, so their growth, reproduction and abundance will decline, until eventually they are no longer present in the catch. The BDMC allows us to detect the role of warming in increasing oxygen demands and of deoxygenation in reducing supply, while the MODC allows us to detect the resulting decline in abundance or fisheries yields of species with high oxygen demands.  I conclude with a synthesis of the main findings and a discussion of their implications for fisheries in Chapter 6. Overall, this thesis contributed towards an understanding of how environmental change may transform marine biogeography in the Eastern Tropical Pacific, using empirical data and model simulations. The BDMI offers the potential to further understand the impacts of warming and deoxygenation on marine fisheries in regions without fisheries and oceanographic monitoring programs. 9   Figure 1.2. Four main chapters in this thesis. Chapter 2 includes an exploratory analysis of climate change impacts on fisheries resources in the Eastern tropical Pacific Ocean. This chapter revealed the need to expand on our knowledge of the impacts of warming and deoxygenation on fisheries and marine ecosystems in the region. Chapters 3 through 5 used empirical data to examine the mechanisms linking warming and deoxygenation with changes in fisheries and marine ecosystems in the Eastern Tropical Pacific region. 10  Chapter 2: Climate change impacts on living marine resources in the Eastern Tropical Pacific Ocean  2.1  Introduction Climate change is causing a global redistribution of species as they track their shifting environmental niches (Jones & Cheung, 2015; Poloczanska et al., 2016). Rapid changes in ocean conditions are exposing ectotherms to unfavorable conditions (Pauly & Cheung, 2018; Pörtner et al., 2017) that lower their growth, reproduction and survival rates. These physiological impacts lead to shifts in species’ biomass and distribution (Cheung & Pauly, 2016; Gattuso et al., 2015). The vulnerability of a species to climate change mainly depends on the scope between current conditions and the species’ physiological tolerance limits (Pörtner et al., 2017). Tropical species generally exhibit narrow thermal tolerances relative to temperate species and therefore, tend to be more sensitive to changes in environmental temperature (Nguyen et al., 2011; Payne & Smith, 2017; Pörtner et al., 2017). Under the high emissions scenario (RCP 8.5), species richness in the tropics is projected to decrease by more than 20% by 2050 relative to the 2000s (Jones & Cheung, 2015). Maximum catch potential (a proxy for maximum sustainable yield) is also projected to decrease globally by 3.4 million tonnes per degree Celsius of atmospheric warming, with the tropics bearing most of such impacts (Cheung et al., 2018; Cheung et al., 2010). These climate-driven declines in the availability of tropical living marine resources have the potential to affect national economies and food security, especially in developing countries with low adaptive capacity (Allison et al., 2009; Blasiak et al., 2017; Lam et al., 2020).  Despite the profound impacts that climate change will have on tropical marine ecosystems, most of the available information is for northern, temperate regions (Lam et al., 2020; Poloczanska et al., 2016). Given the widespread lack of access to scientific vessels in tropical countries, fishing vessels represent useful platforms to gain information on species' distributions (Wehrtmann et al., 2012). Consequently, patterns in species caught by fisheries over time can help us understand regional biogeographic shifts occurring in response to climate change (Maestri & Duarte, 2020). Within the 11  Eastern Tropical Pacific Ocean (Fig.2.1), climate change effects on fisheries have only been documented for the Gulf of California  (Páez-Osuna et al., 2016).     Figure 2.1. Exclusive Economic Zones in the Eastern Tropical Pacific of interest in this study. 1: Mexico, 2: Guatemala, 3: El Salvador, 4: Nicaragua, 5: Costa Rica, 6: Panama, 7: Colombia, 8: Ecuador, 9: Galapagos, 10: Peru. The Galapagos EEZ belongs to Ecuador.  However, climate impacts on fishery species may go undetected due to the confounding effects of climate change, climate variability and overfishing (Hsieh et al., 2006), which are hard to disentangle due to the absence of long-term fisheries and oceanographic time series (Pauly & Zeller, 2016). Moreover, while global projections show that, in general, living marine resources will shift pole-wards (Jones & Cheung, 2015; Poloczanska et al., 2013; Pörtner et al., 2017), observations have underscored the variability in climate change responses across species and space (Lenoir & Svenning, 2015; Morley et al., 2018), with consequences for fisheries management and conservation actions (Arafeh-Dalmau et al., 2020; Frazão Santos et al., 2020; Selden & Pinsky, 2019). 12  This study aims to examine the sub-regional range shift patterns of key fishery species in the Eastern Tropical Pacific and the implications of such shifts for key fisheries and regional biogeography. I hypothesized that climate change would force species to shift towards cooler waters brought up to the surface through upwelling. I used an ensemble of species distribution models to project the future geographic distribution of living marine resources across the Eastern Tropical Pacific under two climate change scenarios. This approach is based on the concept of environmental niche, defined as a set of environmental conditions that allow a species to persist in time and space (Hutchinson, 1957). Species distribution models can help elucidate the complexity of biogeographical responses of marine species at sub-regional scales (Guisan & Thuiller, 2005; Guisan & Zimmermann, 2000) even in data-limited areas such as the Eastern Tropical Pacific (Hernandez et al., 2006; Wisz et al., 2008). Based on the model projections, I discuss the potential impacts of climate change on resource availability and regional marine biogeography.   2.2 Methods 2.2.1 Oceanographic setting of the Eastern Tropical Pacific The Eastern Tropical Pacific Ocean is defined here as the area between 31 °N and 5 °S, from the northern Gulf of California to northern Peru, with the East Pacific Barrier to the west and the Central American isthmus to the east (Fig.2.1). Upwelling systems within the Eastern Tropical Pacific play an important and complex role in driving the spatial patterns of temperature, primary productivity, pH and oxygen (Fiedler & Lavín, 2017). As such, the Eastern Tropical Pacific is delineated by the California Current Eastern Boundary Upwelling System in the north and the Humboldt Current Eastern Boundary Upwelling System, South Equatorial Current and equatorial upwelling system to the south (Fig. 2.2; Fiedler & Lavín, 2017). A ‘warm pool’ along Central America separates these cooler regions. This warm pool has three seasonal upwelling systems (Tehuantepec, Papagayo, Panama) produced by wind jets across Central America (Fiedler & Lavín, 2017). Oxygen minimum zones are formed below the shallow thermocline (< 80 m) as a result of the high primary productivity in upwelling systems, strong stratification   13    Figure 2.2. Temperature, oxygen, pH, net primary production and salinity) at the ocean surface in 2001-2020 (left most panel) and anomalies by 2041-2060 under RCP 2.6 (middle panel) and RCP 8.5 (right most panel). For current conditions, lower values are depicted in red and higher values in blue, except for temperature. For anomalies, warmer colors denote declines while cooler colors indicate positive differences, except temperature, and pH where the largest anomaly = 0.  14   and sluggish circulation (Fig. 2.3). The southern Eastern Tropical Pacific is characterized by low salinity (< 34), while the northern limits are characterized by high salinity (> 34), except for an area off Baja California with low salinity waters transported by the California Current (Fiedler & Lavín, 2017).  2.2.2 Fisheries of the Eastern Tropical Pacific Commercial fisheries catch in the region averaged 11 million tonnes per year during 2005-2014, contributing to approximately 10% of global catches (based on Sea Around Us catch database, Pauly et al., 2020). Marine fisheries in the Eastern Tropical Pacific are diverse in terms of species caught, gears used and fleet sizes (Lluch-Cota et al., 2019). This study focused on species caught by four key fisheries in the Eastern Tropical Pacific: small-scale fisheries, shrimp trawl fisheries, small pelagics and large pelagics fisheries. Small-scale fisheries generally operate in coastal areas within the continental platform, using a wide variety of gears to capture a large diversity of species. In contrast, the fisheries for shrimp, as well as small and large pelagics are predominantly large-scale. Shrimp trawl fisheries mainly target penaeid, pandalid and solenocerid shrimp species. The small pelagics fisheries mainly operate within the continental shelf and target engraulids and clupeids. Large pelagic fisheries generally operate in more off-shore oceanic waters and target tuna, mahi mahi, swordfish and sharks (Lluch-Cota et al., 2019).  2.2.3 Biotic data I modelled species caught by the four major fisheries in the Eastern Tropical Pacific, irrespective of the species’ commercial value. While all fisheries within the region catch a wide variety of species, only a subset of them are landed and sold. However, I included most species reportedly caught by these fisheries, since low-valued species may become valuable in the future (Appendix A. 1). Landings records have a low degree of taxonomic resolution and do not include discards. To obtain an accurate representation of species caught by fisheries in the region, inclusive of bycatch and discards, I conducted a thorough scientific literature review that yielded a list of 652 species, comprised of 512 bony fish spp., 74 elasmobranch species, 47 crustacean species, 16 mollusk species and three echinoderm species.  15    Figure 2.3. Oceanographic conditions at the ocean seafloor in 2001-2020 and anomalies by 2041-2060 under RCP 2.6 and RCP 8.5. Ensembles were created for each variable using the model using the model outputs for GFDL-ESM-2G, IPSL-CM5-MR, MPI-ESM-MR.  16   I compiled species occurrence data for all species on this list (Appendix A. 2; latitude, longitude, and when available, sampling date) from online databases, museum collections and reports (Centro Interdisciplinario de Ciencias Marinas, 2002; Facultad de Ciencias Biológicas, 2001; Froese & Pauly, 2019; GBIF, 2017; Gutiérrez García, 2003, 2004, 2006; Instituto de Biología, 2003; Instituto de Ciencias del Mar y Limnología, 2001a, 2001b; INVEMAR, 2017; IUCN, 2018; Ixquiac, 1998; OBIS, 2017; Robertson & Allen, 2015; Tapia García, 1997). In addition, data from the following databases were accessed through the Fishnet2 Portal (www.fishnet2.net, 2017-01-14): Australian Museum, California Academy of Sciences, Cornell University Museum of Vertebrates, Florida Fish and Wildlife Conservation Commission, Universidad Nacional Autónoma de México, IBiologia - CNPE/Colección Nacional de Peces; Los Angeles County Museum of Natural History, UNELLEZ Museo de Zoología, Colección de Peces, MCZ-Harvard University, Oregon State University, Texas Natural History Science Center - Texas Natural History Collections, Tulane University Museum of Natural History - Royal D. Suttkus Fish Collection, University of Arkansas Collections Facility, Yale University Peabody Museum.  I eliminated duplicates, points on land and points outside of the known species' biome from the compiled species occurrence data set (Froese & Pauly, 2019; Robertson & Allen, 2015). The data was then gridded into a raster of the global oceans (0.5° of longitude per 0.5° degree of latitude) indicating historical presence of each species. The 547 species with occurrence records in more than 30 cells were selected for further analysis (Hernandez et al., 2006).   2.2.4 Abiotic data I applied Generalized Linear Models (GLM - identity link, Gaussian distribution) to statistically bias-correct (Kilsby et al., 1998) average annual climatologies for bottom and surface temperature, oxygen, salinity, pH, surface primary productivity and mixed layer depth for the Eastern Tropical Pacific (Appendix A. 3). The dependent variables for each GLM were the annual climatology of observed surface and bottom temperature, salinity, dissolved oxygen concentration (1955–2012) (World Ocean Atlas 2013, http://www.nodc.noaa.gov/OC5/woa13/), surface chlorophyll-a concentration (1998 to 2012) 17  (http://oceancolor.gsfc.nasa.gov), and mixed layer depth (1998 to 2012) (http://oceancolor.gsfc.nasa.gov). The independent variables for each GLM were the annual climatology (1970-2000) of modeled surface and bottom temperature, salinity, dissolved oxygen concentration, surface chlorophyll-a concentration, and mixed layer depth. Data were obtained from three different commonly used Earth system models (ESM): Geophysical Fluid Dynamic Laboratory model (GFDL-ESM-2G) (Dunne et al., 2013), the Institut Pierre Simon Laplace model (IPSL-CM5A-MR) (Dufresne et al., 2013) and the Max Planck Institute for Meteorology model (MPI-ESM-MR) (Giorgetta et al., 2013). The performance of these ESMs has been extensively examined and tested for applications to the marine realm (Kwiatkowski et al., 2017; Laufkötter et al., 2015). To account for spatial-autocorrelation I also included the interaction between latitude and longitude as an independent variable in all models. Depth was included as an independent parameter for models of bottom environmental conditions. I did not bias-correct pH, because time series for observed pH data do not exist. All environmental parameters were re-gridded and interpolated on to a global 0.5° longitude x 0.5 latitude raster using the bilinear interpolation method, before the bias-correction (Lam et al., 2016). I selected the models with the strongest correlation between variables (R2, Appendix A. 3) and used them to produce regional annual climatology for each parameter based on outputs for each ESM. I assumed for the statistical relationships between ESM and observed climatology to hold in the future and, therefore, used them to project future observed environmental conditions given a set of ESM projections (annual averages of each environmental parameter for RCP 2.6 and RCP 8.5 from 2001 to 2060).   2.2.5 Species distribution models (SDMs) The current and future distributions of the 547 focal species were projected using species distribution models (SDMs). I used a multi-model approach (Jones & Cheung, 2015) to account for the variability across different Earth system models and SDM outputs and increase the accuracy of the projections (Guo et al.,2015; Jones & Cheung, 2015). I applied four SDMs to quantify the environmental niche of each species: Surface Range Envelope (Thuiller et al., 2016), maximum entropy method (Maxent) (Jane Elith et al., 2011), Generalized Boosting Model (Elith et al., 2008) and Artificial Neural Networks (Lek & 18  Guégan, 1999). The input data for each SDM were the species occurrence raster (75% of the original presence data were used to train the model) and the bias-corrected climatology of the environmental conditions. I selected variables representing surface water conditions for benthic and demersal species, and surface water conditions for pelagic or coastal species (Froese & Pauly, 2019; Robertson & Allen, 2015). I defined coastal species as those with common maximum distributions above the mixed layer depth, and assumed they will be more sensitive to changes in the environment at the surface.I avoided over-parameterization by selecting the subset of the environmental parameters that resulted in the highest specialization (narrowness of the niche) and marginality (difference between the niche and the available environment) values (Appendix A. 4) produced by the Ecological Niche Factor Analysis (ENFA, (Basille et al., 2008; Calenge, 2006).  For each species, I ran the four SDMs using the three bias-corrected climatologies (GFDL-ESM-2G, IPSL-CM5-MR, MPI-ESM-MR), resulting in outputs from a total of 12 models per species. All SDMs were run with the Biomod2 package in R (Thuiller et al., 2016). Each SDM calculated a Habitat Suitability Index (HSI) value for each spatial cell in the Eastern Tropical Pacific region, ranging between 0 (not suitable) and 1 (very suitable). I evaluated the accuracy of each model using and Area Under the Curve (AUC) analysis of the Receiver Operating Characteristic (ROC; Appendix A. 5). I used the ROC to compare the fitted HSI with the species occurrence raster reserved for testing the model fit (25% of the original presences). Models with an AUC below 0.5 were eliminated, as predictions were no better than random. This analysis was performed with the pROC package in R (Robin et al., 2011). To be consistent with the approximate time frame represented by the occurrence records and for the development of the models, I used the average SDM predictions for 1970 to 2000.  I then projected changes in the geographic distribution of the species’ environmental niche for each year between 2001 and 2060 under the ‘high emission’ (RCP 8.5) and ‘low emission’ (RCP 2.6) scenarios. I built an ensemble of model outputs for each species and climate change scenario. Specifically, for each ESM, I first calculated the average HSI weighted by the AUC values of each species distribution model, and then averaged HSI values across ESMs, to produce one HSI value per cell. If the 19  habitat suitability was higher than the species prevalence (i.e., the fraction of cells in which the species was present), I considered the species to be present in the cell (Phillips et al., 2009). I averaged projections over a 20-year timeframe to reduce the effect of inter-annual variability of climatic conditions on species distributions (Stock et al., 2011).  I calculated indicators of biogeographic shifts for the species assemblages caught by each of the four fisheries. I present the results for the shrimp trawl fishery separated in target and bycatch species. These indicators include the shift in geographic and depth centroids, local species loss rates, local invasion rates, species turnover (Cheung et al., 2009) and change in habitat suitability for 2041-2060 relative to 2001-2020. Centroids were defined as  𝐶 =  ∑ 𝑋𝑖 ∗ 𝐻𝑆𝐼𝑖𝑛𝑖=1∑ 𝐻𝑆𝐼𝑖𝑛𝑖=1 where 𝐶 is a latitudinal, longitudinal or depth centroid, 𝑋𝑖 is latitude, longitude or depth in each cell and 𝐻𝑆𝐼𝑖 is the HSI for the ensemble model in each cell.  Depth shifts were calculated as the difference between the depth centroid in 2041-2060 and the depth centroid in 2001-2020. Latitudinal and longitudinal shifts were estimated as the shortest distances between the centroids according to the haversine method, which assumes a spherical earth. Local invasion and local loss rates were estimated as 𝐿𝐼𝑖,𝑦 =  𝑛𝑖,𝑦𝐿𝐼𝑛𝑖 + 1 𝐿𝐿𝑖,𝑦 =  𝑛𝑖,𝑦𝐿𝐿𝑛𝑖 + 1 where n is the number of species per cell at the beginning of the century, and 𝑛𝑖,𝑦𝐼  and 𝑛𝑖,𝑦𝐸  are the number of species invading or going extinct in each cell, respectively, by the end of the study's timeframe. Finally, species turnover is the sum of invasion and extirpation rates.  20  2.3 Results  2.3.1 Oceanographic section The bias-corrected surface-level projections show that anomalies by the mid-21st century are much larger under RCP 8.5 than RCP 2.6, especially for surface temperature, oxygen and pH. Differences between RCPs are much lower for seafloor conditions (Fig. 2.2, 2.3). Present (2001-2020) temperatures at depth remained between 0 and 5 °C throughout most of the region, bordered by warmer waters of 6-10 °C in the narrow shallow areas along the coastline (Fig.2.3). By the mid-21st century, surface temperatures are projected to increase by 0.96 °C, on average, relative to the present, under RCP 8.5 scenario, except in upwelling regions. Surface oxygen follows the same spatial patterns, which is not surprising considering the negative correlation between temperature and oxygen solubility. Amongst the 547 species with sufficient data for species distribution modelling, 505 species were used for further analysis because at least three of the four species distributions models had AUCs above 0.5 (Appendix A. 5). Variability in AUC values was high across SDMs and low across Earth system models (Appendix A. 5). There was significant overlap between species caught by coastal small-scale fisheries (371 spp.) and shrimp trawl fisheries (441 spp. of bycatch, 19 spp. of shrimp; Appendix A. 1). The environmental variables used to model most species distributions were temperature (505 spp.), oxygen (473 spp.), salinity (358 spp.) and pH (356 spp.) due to their high marginality and specificity values estimated from Ecological Niche Factor Analysis (Appendix A. 4).   2.3.2 Species shifts In the northern region (> 15⁰ N), the centroids of species' distributions were projected to shift towards the northwest at an average rate of 71 km decade-1, while in the northern equatorial region (0⁰ to 15⁰ N) and the southern Eastern Tropical Pacific (0⁰ to 20⁰ S) species were projected to shift southeast at an average rate of 59 and 30 km decade-1, respectively (Fig. 2.4). The direction of projected species’ geographic shifts was similar across fisheries. Demersal species in most countries were found to move towards shallower waters by an average rate of shifts in depth-centroid of approximately 1 – 13 m decade-1; 21  however, there was considerable variability in the direction and magnitude of depth shifts across species (Fig. 2.5).  Figure 2.4. Direction and distance (km) of geographic shifts undergone by species groups (classified according to fisheries) by 2041-2060 relative to 2001-2020 under RCP 8.5. L-PEL: large pelagics; STF: shrimp trawl target fishery; ST-BYC: shrimp trawl bycatch; C-SSF: coastal small-scale fisheries; S-PEL: small pelagics.  Figure 2.5. Mean shifts in depth centroids (average depth weighted by the Habitat Suitability Index) and associated standard deviation for demersal species by 2041-2060 relative to 2001-2020 under RCP 8.5. Positive changes indicate species are shifting towards shallower waters.  22  2.3.3 Habitat suitability The spatial patterns of changes in habitat suitability by 2041-2060 were similar between RCP 2.6 and RCP 8.5, although in most cases the magnitude of change was greater for RCP 8.5 (Table 2.1). The habitat suitability of all species was projected to increase or remain the same by 2041-2060 relative to present in the northern and southern limits of the study area (northern Mexico, southern Ecuador and Peru), while it was projected to decrease in southern Mexico, Guatemala, El Salvador, Nicaragua and northern Costa Rica (Fig. 2.6). Consequently, the habitat suitability for all species combined is projected to decrease on average in all EEZs except Peru, with the largest declines from Guatemala to Costa Rica (Table 2.1).  The lowest species turnover rates across all species are also expected in Peru and Mexico (Table 2.2). Models projected high rates of local loss (as a proportion of species in the cell during baseline conditions) across all species throughout the study region, except along the northern and southern limits (Fig. 2.7), and high local invasion rates (as a proportion of species in the cell during baseline conditions) for Panama, Colombia and Ecuador (Fig. 2.8). Projections of habitat suitability for species caught by small-scale fisheries indicate declines will be strongest from Guatemala to Nicaragua (-16%). In contrast, species habitat suitability was projected to increase in the Galapagos and Peru (Table 2.1; Figure 2.6). Rates of local loss for species caught by small-scale fisheries was projected to be higher further from the coasts along Costa Rica, Panama, Colombia, Ecuador and Galapagos (Fig. 2.7). The highest local invasion rates were projected for northern Mexico and throughout Costa Rica, Panama, Colombia and Galapagos (Fig. 2.8). Projected turnover rates for species caught by small-scale fisheries were higher than for any other fishery, surpassing 38% between Costa Rica and Galapagos, and reaching almost 80% in Colombia (Table 2.2).  Projected decreases in habitat suitability for species caught by large pelagic fisheries were highest between Guatemala and Nicaragua (-20% to -26%) (Table 2.1) and projected turnover rates were highest between Guatemala and Ecuador (Table 2.2). Most invasions were projected to occur from Panama to Peru (Fig.2.8), whilst local losses were expected to remain high throughout the study region, except along the northern and southern limits (Fig. 2.7).  23   Table 2.1. Percent change in the habitat suitability projected by 2041-2060 relative to 2001-2020 for species groups caught in the four main fisheries (large pelagics, small pelagics, shrimp trawl (incl. of bycatch) and small scale) in the Pacific Exclusive Economic Zones of Mexico to Peru.  Pacific EEZ All fisheries Coastal small-scale fisheries Large-pelagics fisheries Small-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 Mexico -2.39 -4.41 -2.67 -4.98 -13.74 -14.42 -4.55 -10.00 -1.70 -4.00 9.21 8.51 Guatemala -10.18 -13.59 -12.23 -15.85 -18.25 -25.98 -31.24 -40.19 -9.86 -13.18 4.22 4.08 El Salvador -11.32 -13.45 -13.33 -15.50 -18.16 -24.61 -35.04 -42.13 -11.12 -13.10 0.21 0.94 Nicaragua -10.46 -13.94 -11.67 -15.97 -12.44 -20.19 -30.45 -45.69 -10.86 -13.86 -1.09 -2.02 Costa Rica -12.11 -9.04 -13.46 -10.37 -8.77 -8.80 -32.68 -28.20 -12.65 -9.32 0.68 1.31 Panama -4.62 -7.15 -5.58 -8.51 -2.99 -9.05 -17.64 -29.06 -4.81 -7.21 2.65 1.28 Colombia -2.42 -6.55 -3.14 -7.47 -1.62 -5.28 -8.03 -20.20 -2.48 -6.77 4.88 0.89 Ecuador -1.44 -0.89 -1.54 -1.13 1.57 -0.99 -3.52 -4.69 -1.53 -0.73 3.78 5.17 Galapagos -0.16 1.24 0.41 1.84 1.73 0.99 2.50 3.11 -0.46 1.25 4.35 5.54 Peru 6.01 4.74 5.91 4.48 3.37 3.44 17.42 13.90 6.63 5.15 13.10 13.08  24                                                 Figure 2.6. Projected change in the habitat suitability of species groups caught in the four main fisheries (large pelagics, small pelagics, shrimp trawl (incl. of bycatch) and small scale) by 2041-2060 relative to present (2001-2020) under RCP 8.5. Warm hues in the color ramp denote losses in habitat suitability while cool hues denote gains. 25   Table 2.2. Projected species turnover (%) by 2041-2060 relative to 2001-2020 for species groups caught in the four main fisheries (large pelagics, small pelagics, shrimp trawl (incl. of bycatch) and small scale) in the Pacific Exclusive Economic Zones from Mexico to Peru.  Pacific EEZ All fisheries Coastal small-scale fisheries Large-pelagics fisheries Small-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 RCP 2.6 RCP 8.5 Mexico 5.78 12.36 8.74 16.60 4.12 9.59 3.06 5.22 7.90 15.33 3.69 5.61 Guatemala 17.27 28.19 1.65 2.42 18.82 30.74 0.00 5.67 1.89 2.38 0.54 0.66 El Salvador 20.06 19.68 3.27 4.06 25.01 24.31 3.99 6.52 2.54 3.20 3.05 3.05 Nicaragua 10.56 11.68 6.22 7.21 14.39 18.56 24.24 30.30 5.87 6.77 3.18 5.45 Costa Rica 17.57 32.27 42.97 48.32 17.48 33.88 8.20 10.14 9.57 27.40 1.20 1.21 Panama 21.64 33.46 41.79 57.70 23.51 39.11 10.25 14.21 18.02 27.23 6.08 4.40 Colombia 22.34 41.12 57.42 79.47 22.23 46.74 7.19 7.55 15.85 32.78 2.05 3.37 Ecuador 10.71 28.89 12.76 38.20 9.48 28.98 5.81 21.00 16.56 45.88 3.53 3.84 Galapagos 5.19 14.88 10.39 38.70 1.82 7.04 5.27 14.83 12.04 38.44 2.79 2.82 Peru 4.37 8.69 4.17 9.38 6.87 10.62 3.71 7.04 3.91 11.15 1.54 1.65  26    Figure 2.7. Projected local extirpation rate of species groups caught in the four main fisheries (large pelagics, small pelagics, shrimp trawl (incl. of bycatch) and small scale) between for 2041-2060 relative to 2001-2020 under RCP 8.5. The extirpation rate represents the number of species lost from the cell as a proportion of species in that cell during baseline conditions. The larger the extirpation rate the warmer the color.   27    Figure 2.8. Projected local invasion rate of species groups caught in the four main fisheries (large pelagics, small pelagics, shrimp trawl (incl. of bycatch) and small scale) between for 2041-2060 relative to 2001-2020 under RCP 8.5. The invasion rate represents the number of species gained in a cell as the proportion of species in that cell during baseline conditions. The larger the invasion rate the warmer the color. 28  Models projected large declines in habitat suitability for species caught by small-pelagic fisheries from Guatemala to Colombia (Table 2.1), and a 17% increase for Peru (Table 2.1). Species turnover projections were below 30% (Table 2.2), with the highest local losses expected along the continental shelf, especially from Costa Rica to Ecuador (Fig. 2.7). Projected invasion rates were also highest for the continental shelf area (Fig. 2.8). The habitat suitability of species targeted by the shrimp trawl fishery was projected to increase in all EEZs except Nicaragua, and particularly in Mexico and Peru (Table 2.1, Fig.2. 6). Local loss rate projections were low throughout the study region apart from small areas in the Gulf of California (Fig.2. 7). Shrimp invasion rates were projected to follow a patchy distribution mainly along the coastline of northern Mexico and southward of Panama (Fig. 2.8). Target species in this fishery also showed the lowest turnover rates, with projections below 6% for all EEZs (Table 2.2).  In contrast to species targeted by the shrimp trawl fishery, the habitat suitability of species caught as bycatch was projected to decrease in all EEZs except the Galapagos and Peru, with the highest impacts estimated for the area between Guatemala and Costa Rica (9% - 13%) (Table 2.1). Patterns of decreasing habitat suitability and local loss mirrored that of small-scale fisheries, but with higher losses spanning further south (Fig. 2.6, 2.7). Invasion rates were projected to be much lower than for small-scale fisheries and were limited to the northern and southern limits of the study area (Fig. 2.8).  Most species in each EEZ were not projected to undergo declines in habitat suitability (Appendix A. 6). The changes in habitat suitability of species in the southern EEZs of Peru, Galapagos and Ecuador had narrower frequency distributions showing smaller changes in habitat suitability. In the remaining EEZs, changes in habitat suitability have skewed distributions, with a small percentage of species decreasing up to 100%. Habitat suitability was projected to increase for a small number of species.  2.4 Discussion Our results provide insights into the expected climate change impacts on regional biogeography and living marine resources in the Eastern Tropical Pacific between now and 2041-2060, a time frame deemed relevant for climate change to inform fisheries management actions. Results show limited 29  divergence in the projected oceanographic conditions and habitat suitability losses between greenhouse gas emissions scenarios from the present-day to 2041-2060 because of lagged responses of some oceanic variables from changes in atmospheric greenhouse gas concentrations.  Two general patterns emerge when examining the oceanographic processes responsible for the distribution of fisheries species: species are shifting towards cooler waters in the northern and southern margins of the Eastern Tropical Pacific, and towards more oxygenated, shallower waters. Temperature and oxygen were predicted by these models to be the most important variables shaping the environmental niche of the study species and therefore, warming and deoxygenation will likely drive the redistribution of species in the Eastern Tropical Pacific. These patterns may be representative of the broader response of marine biodiversity to climate change in the region. Such patterns also agree with expectations from proposed theory explaining the biological responses of marine fishes and invertebrates to changing temperature and oxygen levels. For example, the “oxygen and capacity limited thermal tolerance” theory (OCLTT) suggests that temperatures above an organism’s thermal tolerance threshold results in a smaller aerobic scope for physiological functions, like growth and reproduction (Pörtner et al., 2017). According to the Gill Oxygen Limitation Theory (GOLT), increase in oxygen uptake to meet higher metabolic demand for oxygen under ocean warming is limited by the constraints of available area for gaseous exchange of the gill (Pauly & Cheung, 2018). Thus, fish move to waters with temperatures that resemble those of their original habitats and satisfy organisms' oxygen needs.  2.4.1 Species shifts The Humboldt Current and California Current Eastern Boundary Upwelling Systems (produced by alongshore winds in Peru and northern Mexico) and the Equatorial Upwelling systems (near the equator produced by the Coriolis force) have a cooling effect along the northern and southern limits of the Eastern Tropical Pacific (Fiedler & Lavín, 2017). These upwelling systems are separated by the eastern Pacific warm pool along Central America, resulting in an inverse temperature gradient in the northern hemisphere (Fiedler & Lavín, 2017). Consequently, between 0⁰ and 15⁰ N species are moving towards the Equator instead of the poles (Fiedler & Lavín, 2017; Pörtner et al., 2017). As expected, species are 30  shifting at a faster rate in tropical areas with weaker latitudinal temperature gradients, where their preferred temperature has shifted further away. In contrast, the steeper temperature gradient along the Humboldt Current allows for species to find their preferred temperature within shorter distances (Robinson et al., 2015).  Species within the Eastern Tropical Pacific shift towards the Equator when waters warm during El Niño events (Sielfeld et al., 2010), further supporting the projected equator-wards shift of tropical species. Local losses, population recoveries and range extensions are common responses to El Niño–Southern Oscillation related inter-annual temperature variability in the Eastern Tropical Pacific (Mora & Robertson, 2005). Reports from as early as 1982 show that tropical shrimp shift southwards towards Peru during El Niño (Barber & Chavez, 1983). A more recent study identified 100 tropical species in Chilean waters (with sub-tropical and temperate climatic conditions) during El Niño years (Sielfeld et al., 2010).  In contrast to observed range shifts elsewhere (Dulvy et al., 2008; Poloczanska et al., 2013), these findings show that species are projected to shift towards shallower instead of deeper waters. This shoaling of species can be attributed to the expansion of Oxygen Minimum Zones, which drive most organisms into shallower and more oxygenated waters (Stramma et al., 2012). These shallower waters are also warmer, which increases the energy required to meet basic metabolic demands and may require organisms to compensate for temperatures outside their tolerance range (Craig, 2012; Gallo & Levin, 2016). Oxygen minimum zones are known to compress the habitats for both benthic and pelagic species (Gallo & Levin, 2016). For example, oxygen minimum zones have been shown to compress the habitat of billfish in the Eastern Pacific Warm Pool (Prince & Goodyear, 2006) and of small pelagics in Peru (Bertrand et al., 2011). The expansion of oxygen minimum zones was also found to force echinoderms in the California Current to shoal, while the contraction of oxygen minimum zone during El Niño temporarily expanded the habitat of demersal fish towards deeper waters in Peru (Arntz et al., 2006; Keller et al., 2015; Sato et al., 2017).  Decreases in the habitat suitability and local losses of the living marine resources focused on in this study, mainly occurred across the exclusive economic zones of Central America and Colombia, coinciding with the warming and expansion of the Eastern Pacific warm pool (Fiedler & Lavín, 2017). High 31  invasion rates along the northern limits of the study area (20-30˚N) and south of 10˚N could be caused by species shifting towards the cooler waters of the upwelling systems. Projections for upwelling systems under climate change are uncertain as Earth system models do not resolve upwelling processes well (Lluch-Cota et al., 2019). The potential increase in upwelling intensity in eastern boundary upwelling systems could affect species’ biogeography. For example, enhanced hypoxia and acidification associated with upwelling activity could limit the beneficial effects of cooling and higher primary productivities on habitat suitability (Bakun et al., 2015; Fiedler & Lavín, 2017).   2.4.2 Implications for fisheries and conservation Our findings highlight the importance of local-scale oceanographic and biological data to elucidate the multi-dimensional biogeographic shifts on fishery species and their potential impacts on fisheries in the region. Overall, changes in the habitat suitability and therefore composition of species caught by the four main fisheries are expected to be most severe along Central America, with substantial variations in the magnitude of impacts across fisheries.  The results suggest that shrimp-trawl fisheries may benefit from climate change in the Eastern Tropical Pacific because of the increase in habitat suitability of target species. Shrimp in general may be less vulnerable to climate change because of their fast population growth rates, high larval dispersal rates, and low ecological specificity (Hsieh et al., 2006). Yet, it is their fast population growth rates and short lifespans that make them vulnerable to the strong interannual climate variability characteristic of the ETP (Arreguín-Sánchez et al., 2015; López-Martı́nez et al., 2003; Sanz et al., 2017). Furthermore, shrimp throughout the region may be more vulnerable to climate change than models indicate because most stocks are overfished (Cisneros-Montemayor & Clarke, 2019). Overall, the shrimp-trawl fishery’s environmental impacts would increase, as it would continue to put additional pressure on shrimp bycatch species, which are projected to undergo strong declines in habitat suitability. In addition, shrimp are also targeted by small-scale fisheries, which can account for up to 80% of shrimp catches within a country. As climate change is projected to heavily impact species caught by small scale fisheries, they may increase 32  their shrimp catches. This may further impact the sustainability of shrimp stocks in the region, as the catches of small-scale fisheries are seldom recorded or managed. In contrast, small-scale fisheries are at high risk of impacts under climate change, with findings showing strong declines in habitat suitability and high local losses. However, such risk may be mitigated by fisheries catching species projected to move into their fishing grounds under climate change and therefore likely replacing a proportion of foregone catches. For example, despite projected declines in habitat suitability for species caught by small-scale fisheries in Costa Rica, Panama, Colombia, Ecuador and Galapagos, their waters will become more suitable for a large number of species currently not caught by the fisheries. If these species are able to colonize and establish populations in the newly available habitats, without causing local losses of traditional target species, 'invaders' may help compensate for the decrease in traditionally targeted species. The flexibility granted by the multi-gear, multi-species approach of small-scale fisheries may allow them to seize the opportunity of catching these new species on their fishing grounds (Lluch-Cota et al., 2019). On the other hand, small-scale fishing communities may be highly vulnerable to climate change impacts because they do not have the vessels to chase stocks on the move, many of their stocks are already overfished, they are highly dependent on short-term income (Lluch-Cota et al., 2019) and 'invading' species may not contribute to food security and livelihood opportunities in the same way. Any impact on the small-scale fishing sector is likely to affect food security and local economies to a much larger extent than official statistics would indicate, because their catches, employment and income are significantly under-reported (Pauly & Zeller, 2016). Findings from this study can help inform stewardship and sustainable fishing practices local communities need to adopt to support their needs into the future. Our estimates of climate impacts are based on all the species caught by each fleet, regardless of economic value. Future studies should provide detailed estimates of projected climate impacts of high and medium value commercial species. Abundant species where habitat suitability is expected to remain the same or increase should then be identified as new target species to offset catch losses. Appropriate processing mechanisms and marketing programs could then be planned in order to support fisheries adaptation climate change. 33              2.4.3 Model robustness and uncertainty While the regional geographic patterns of projected changes are considered reliable, the specific magnitude of projected changes is affected by model and scenario uncertainties (Cheung et al., 2016; Wabnitz et al., 2018). While the multi-model approach accounts for uncertainties associated with future emissions, species distribution models' and Earth system models' structures, several additional sources of uncertainty remain (Peterson & Soberón, 2012; Stock et al., 2011). First, the results depend on the accuracy of the bias-corrected ESM projections (Cheung et al., 2016) and the assumption that the correlation between the ESMs and the observed climatology will hold in the future. In addition, model outputs rely on ESMs, which exclude small-scale processes that would allow to better resolve regional upwelling dynamics and coastal processes. Second, the species distribution modelling approach applied here focuses on habitat suitability in a single species context. Therefore, the method does not account for ecosystem effects, such as inter-species dynamics, the ability of invading species to establish themselves in new habitats, or the possible biodiversity and habitat loss that invading species may cause (Pecl et al., 2017). Local losses and invasions have the potential to modify ecosystem structure and function (Marzloff et al., 2016; Pecl et al., 2017). Third, I may have underestimated the declines in habitat suitability by modeling the realized niches of species rather than populations, for which there is little data available. The average niche breadth of a species is much wider than that of a population, and therefore, will be less sensitive to projected changes in environmental conditions. Fourth, several tropical species are already close to their thermal limits, likely increasing their vulnerability to further warming (Sunday et al., 2011). Fifth, I do not consider other human stressors that have and will continue to drive impacts on living marine resources (Galbraith et al., 2017). For example, these models do not account for the environmental impacts of shrimp-trawl fisheries, nor for the effects of overfishing of large pelagics in the Eastern Tropical Pacific (Espinoza et al., 2018). Therefore billfish, tunas and other large pelagics are probably even more vulnerable to climate change than these results indicate. Sixth, the possibility of genotypical and phenotypical plasticity that determine potential acclimation and rapid evolution of marine species may reduce their sensitivity to climate change (Calosi et al., 2016). Evidence of evolutionary 34  responses of marine fishes and invertebrates to climate change is still limited and is an important area for further exploration in future studies. Overall, most of these sources of uncertainty (1 to 5) would likely produce stronger declines in habitat suitability while only one (6) may result in more optimistic projections. Our findings contribute to the understanding of species responses to the growing threat of climate change in a complex oceanographic region, in support of efforts to implement management and conservation actions. This study can help identify which species may need greater protection in the future and identify areas that would support greater resilience in the face of climate change, such as biodiversity refuges (Hoffmann et al., 2019; Kujala et al., 2013; McHenry et al., 2019). The identification of areas that may be particularly vulnerable to climate change may also be used to inform marine spatial planning for climate adaptation initiatives in the Eastern Tropical Pacific (Peñaherrera-Palma et al., 2018). Thus, species projections can inform policy decisions and conservation strategies that ensure the protection and sustainable use of living marine resources (Wilson et al., 2020).35  Chapter 3: A new metabolic index to understand the impacts of ocean warming and deoxygenation on global marine fisheries resources  3.1  Introduction Over the past 50 years, ocean warming and changes in water column ventilation have lowered the global ocean oxygen content by 1-3% and expanded the volume of oxygen-depleted waters (Bindoff et al., 2019; Breitburg et al., 2018; Schmidtko et al., 2017). Under continued global warming, Earth system models suggest that these trends will continue, with oxygen losses of 3-4% projected by the end of the 21st century relative to 2006-2015 (Keeling et al., 2009). At the individual level, ocean warming and deoxygenation affect the physiological performance of ectothermic, water-breathing marine organisms, such as fishes and invertebrates, by increasing their metabolic oxygen demand while reducing oxygen supply (Cheung et al., 2013a; Pauly & Cheung, 2018). Combined, these stressors compress an organism’s aerobic scope (the difference between maximum and standard metabolic rates), reducing the oxygen available to support processes beyond maintenance requirements, such as growth, movement and reproduction (Crear et al., 2020; Hoefnagel & Verberk, 2015; Pörtner et al., 2017; Sokolova, 2013).  The impacts of warming and deoxygenation on marine organisms can lead to profound population and ecosystem-level changes. As ocean warming increases oxygen demand beyond what can be provided by oxygen supplies, it may render currently occupied habitats unsuitable (Deutsch et al., 2015, 2020). For example, Deutsch et al. (2015) found that increasing temperatures will be the main driver of marine hypoxia, causing species to shift away from the warmer limits of their distributions. These shifting species will then modify the ecosystem composition, as well as its structure, function and services (Gallo & Levin, 2016). Deutsch et al. (2015) and Penn et al. (2018) developed a physiologically derived metabolic index to estimate the impacts of warming and deoxygenation on habitat viability. This metabolic index represents the relationship between oxygen supply and standard metabolic oxygen demand in ectothermic marine organisms. Application of the index requires information on experimentally-derived temperature-dependent oxygen thresholds, which are available only for a few, extremely well documented species (Penn et al., 2018), limiting the scope for application of this method. 36  Here, I present a new metabolic index, hereafter called the Biogeographically derived Metabolic Index (BDMI), which can be calculated for a broad range of marine fishes and invertebrates using data that are available from published databases (Table 3.1). The BDMI aims to facilitate research and assessment of climate-driven warming and deoxygenation impacts on the main global marine fisheries resources. The BDMI integrates both growth theory and metabolic theory (Cheung et al., 2013a; Pauly, 2010) to create a theoretical oxygen supply to demand ratio. I validate its use by comparing BDMI estimates with published values of the physiologically derived metabolic index developed by Deutsch et al. (2015) and Penn et al. (2018) and discuss model limitations.  3.2 Methods 3.2.1 Biogeographically derived Metabolic index (BDMI) The BDMI is essentially the ratio of oxygen supply to demand estimated for a species: 𝐵D𝑀𝐼 =  𝑝𝑂2,𝑠𝑢𝑝𝑝𝑙𝑦 𝑝𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 ⁄  (1) Where pO2 is oxygen partial pressure (in atm), pO2,supply  is the ambient oxygen pressure (in atm), and pO2, demand  is the species’ oxygen demand (in atm) at a maintenance metabolic rate (Cheung et al.,2011; 2013; Pauly, 2010). I define maintenance metabolic rate according to Pauly and Cheung (2018, page 16) as “the weight‐specific consumption of oxygen by fish that allows their survival under natural conditions (i.e., in habitats with prey and predators, and other stressors), or in simulated natural conditions, but not allowing for somatic growth.” Biomass growth can be expressed as the difference between anabolism and catabolism (Pauly, 2010). Therefore, the maintenance metabolic rate can also be expressed as the oxygen demand when anabolism equals catabolism: 𝑑𝑊𝑑𝑡= 𝐻𝑊𝑡𝑑 − 𝑘𝑊𝑡𝑏 (2) where the first term represents anabolism 𝐻 that scales exponentially with body weight W through a scaling coefficient d. This coefficient accounts for changes in respiratory efficiency as fish or invertebrate grow in mass. Mass-specific anabolism 𝐻 generally scales negatively with body size because factors such as oxygen supply, food availability and temperature limit the rate of anabolic biochemical reactions (Cheung et al., 2013a). On the other hand, I assume mass-specific catabolism 𝑘 is proportional to body 37  mass, with b close to 1. (Pauly, 2010) show that d can vary between 0.5 and 0.95. Here I use a value of 0.7, which results in a conservative estimate of oxygen demand at a maintenance metabolic rate (Pauly & Cheung, 2018). To examine the sensitivity of the BDMI to d, I also computed the index with a d of 0.9. The choice of d did not strongly affect BDMI output (Appendix B. 2, Appendix B. 6, Appendix B. 7, Appendix B. 8, Appendix B. 9). Based on the von Bertalanffy growth function (VBGF, von Bertalanffy, 1957) and calculated from equation (2), the coefficient k can be derived from the VBGF coefficient K (i.e., the instantaneous rate of growth towards the theoretical asymptotic weight (𝑊∞) ) at which anabolism equals catabolism, or dW/dt = 0. k=K/(1-d) (3) In addition, the anabolic coefficient H can be expressed as a function of 𝑊∞, assuming b =1:      𝑊∞ = (𝐻𝑘)1(1−𝑑)  (4a)      𝐻 = 𝑊∞1−𝑑 ∙ 𝑘  (4b) where 𝑊∞ is the asymptotic weight (in g), estimated from the asymptotic length 𝐿∞ (in cm) and the species´ length-weight relationship (Froese & Pauly, 2019). The VBGF’s coefficients 𝑊∞ and K have been calculated for thousands of fishes and invertebrates worldwide and the estimates are publicly available through online databases: FishBase (www.fishbase.org) and SeaLifeBase (www.sealifebase.org).  I account for the temperature dependence of anabolism and catabolism through the Arrhenius equation, a method that is commonly used to represent the role of temperature in accelerating the reaction rates of chemical and biological processes (Clarke & Johnston, 1999). The Arrhenius equation can be expressed as a rate of reaction 𝑒−(𝐸𝑎𝑘𝑏𝑇), where 𝐸𝑎 is the activation energy (in eV), 𝑘𝑏 is the Boltzmann constant (8.617333262×10−5 eV K−1 ) and T is temperature (in Kelvin). I also assume that the anabolic term 𝐻 in equations (4) is oxygen dependent, and that oxygen supply is at the threshold level when individual reaches 𝑊∞ yields:  𝐻 = 𝑔 ∙ [𝑝𝑂2] ∙ 𝑒−𝑗1/𝑇 (5) 𝑔 =  𝑊∞1−𝑑𝑘𝑝𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ 𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓 (6) 38  The anabolic coefficient can be re-expressed by combining equations (5) and (6),  𝐻 = 𝑊∞1−𝑑 𝐾/(1−𝑑)𝑝𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓[𝑝𝑂2]𝑒−𝑗1𝑇⁄  (7) The pre-exponential factor q of the growth equation´s catabolic term is not oxygen dependent: 𝑘 = 𝑞 ∙ 𝑒−𝑗2/𝑇 (8) 𝑞 = 𝐾/(1−𝑑)𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓 (9) The catabolic coefficient can be re-expressed by combining equations (8) and (9):  𝑘 =  𝐾/(1−𝑑)𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓 * 𝑒−𝑗2/𝑇 (10) Where 𝑗1 and j2 are coefficients calculated by dividing the activation energy by the Boltzmann constant (kb). I used the estimates of activation energy generated by Cheung et al. (2011) of 0.388 eV and 0.689 eV for anabolism and catabolism, resulting in j1 and j2 of 4500K and 8000K, respectively. Cheung et al. (2011) obtained these values through a parameterization of the growth equation generalized to model exploited marine fishes worldwide. Based on a median Q10 (temperature coefficient that represents the acceleration of the reaction rate with a 10 °C increase in temperature) of 2.4 for fishes across the studies assessed by Clarke and Johnston (1999), Cheung et al. (2011) estimated the activation energy Ea over a temperature range of 1-28 °C, which yields a value of ~ 8 for j2 . Next, I estimated j1 based on the ~0.7 slope of the regression between log (K) and log (𝑊∞) (Cheung et al., 2011; Pauly, 2010). Variations in the relationship between K and 𝑊∞ among species reflect the differences in the anabolic and catabolic coefficients and thus their respiratory and growth efficiency. To examine the sensitivity of the BDMI to this generalized temperature dependence across all species, I also computed the index using: a) species-specific j1 obtained from physiological parameters reported by Penn et al. (2018)  b) a species-specific j1 (Penn et al., 2018) and j2 (Cheung et al., 2011; Pauly, 2010).  I assume that each species has an ambient threshold oxygen level (pO2,thresh) below which oxygen becomes a limiting factor for survival (Pauly & Cheung, 2018). I incorporated the temperature dependence of this oxygen threshold into the Arrhenius equation (6). I estimated this threshold oxygen concentration pO2,thresh (in atm) as the 10th percentile of the oxygen concentration across a species´ distribution. I examined the sensitivity of BDMI model output to the selection of pO2,thresh by also 39  computing BDMI with pO2,thresh of 1st, 5th, 15th, 20th and 25th percentiles of oxygen concentration across species´ distributions. I expect demersal species that mainly live below the mixed layer (roughly below 100m) to be more influenced by environmental temperature and oxygen on the ocean bottom, while I would expect pelagic or shallow water species that mainly live above the mixed layer to be more influenced by environmental temperature and oxygen at the ocean surface. Therefore, I assumed that the median sea surface water temperature for pelagic and coastal species and sea bottom water temperature for demersal species across a species´ distribution reflects the species’ physiologically preferred temperature (𝑇𝑝𝑟𝑒𝑓 ) (Cheung et al., 2008). Furthermore, the latitudinal range of a species distribution may encompass temperature and oxygen breadth that replicate conditions across its depth distribution (Reygondeau, 2019).  Previous studies have demonstrated a strong correlation between biogeographically derived temperature preferences with those determined by physiological experiments (Pauly, 2010). Therefore, I assume this approach to estimate species’ temperature preferences provides reasonable approximations for species with no available experimentally determined temperature preferences. I acknowledge the uncertainties that variations in the vertical distribution of species across the water column may introduce. In the future, availability of temperature, oxygen and species distribution data across different depth levels may allow for the calculation of a three-dimensional BDMI.  Maintenance metabolic oxygen demand (𝑝𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑), or the oxygen demand when anabolism equals catabolism (i.e., dW/dt=0), can be obtained by solving the following equation, which integrates equations 2, 7 and 10: 𝑊 ∙𝐾 (1−𝑑)⁄𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓 ∙ 𝑒−𝑗2𝑇 = 𝑊𝑑 ∙𝑊∞1−𝑑 𝐾 (1−𝑑⁄ )𝑝𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓 ∙ 𝑝𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑𝑒−𝑗1/𝑇 (11) From it derives, the 𝑝𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 at a maintenance metabolic rate: 𝑝𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 = (𝑊1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗2/𝑇 𝑝𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ 𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓𝑊∞1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗1/𝑇 𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓) (12) The final biogeographically derived metabolic index then can be computed as follows: 𝐵D𝑀𝐼 =  𝑝𝑂2,𝑠𝑢𝑝𝑝𝑙𝑦 (𝑊1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗2/𝑇 𝑝𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ 𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓𝑊∞1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗1/𝑇 𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓)⁄   (13) 40  The input parameters of the BDMI are detailed in Table 3.1. Table 3.1. Data requirements for the Biogeographically derived Metabolic Index, with the values of the constants and fixed coefficients, as well as the units for each parameter.  Abbreviation Definition Units Sources Species distribution  At least 10 geographic locations randomly distributed throughout the species range ° GBIF  https://gbif.org/  OBIS  https://obis.org/  FishBase  https://fishbase.org  Aquamaps  https://www.aquamaps.org/  T, p02, cO2  Ambient temperature and oxygen  K, atm, mol/m3 World Ocean Atlas, https://nodc.noaa.gov/OC5/woa18/woa18data.html  CMIP5/6 Data Portal https://esgf-node.llnl.gov/projects/cmip5/  https://esgf-node.llnl.gov/search/cmip6/ W Average body weight or a third of the species asymptotic weight g FishBase  https://fishbase.org SeaLifeBase   https://sealifebase.org d Metabolic scaling coefficient  Pauly & Cheung (2018), Pauly (2010) b Catabolic scaling coefficient  Cheung et al. (2011) j1 Anabolism activation energy divided by the Boltzmann constant K Cheung et al. (2011) j2 Catabolism activation energy divided by the Boltzmann constant K Cheung et al. (2011) kb Boltzmann constant eV K−1   W∞ Asymptotic weight, converted from the L∞ and the length-weight relationship g FishBase  https://fishbase.org SeaLifeBase   https://sealifebase.org L∞ Asymptotic length cm FishBase  https://fishbase.org SeaLifeBase   https://sealifebase.org LWa a parameter from the length weight relationship  FishBase  https://fishbase.org SeaLifeBase   https://sealifebase.org LWb b parameter from the length weight relationship  FishBase  https://fishbase.org SeaLifeBase   https://sealifebase.org K  von Bertalanffy K  FishBase  https://fishbase.org SeaLifeBase   https://sealifebase.org  3.2.2 Illustrative examples I calculated average annual BDMI for 1971 – 2100 and three commercially-important species with substantially different ecology and phylogeny: Sharpsnout seabream (Diplodus puntazzo), Atlantic blue crab (Callinectes sapidus) and Atlantic cod (Gadus morhua). I selected these species because they fulfill 41  the data requirements for the metabolic index. Unfortunately, no commercial fish or invertebrates in the Eastern Tropical Pacific met the requirements of the metabolic index at the time of the analysis. The biological and ecological input parameters are described in Table 3.2. We used environmental data from the ocean surface for Sharpsnout seabream and Atlantic blue crab because they are demersal and bentho-pelagic species that are commonly distributed at depths shallower than 60 m. Given the coarse resolution of the environmental data, we assume surface data would better represent the variability within their habitat. We used environmental data from the ocean bottom for Atlantic cod because it is a demersal species with common maximum depth distributions between 150 m and 200 m. The asymptotic length (𝐿∞), von Bertalanffy K, and the length-weight relationship were obtained from Fishbase (Froese & Pauly, 2019) and SealifeBase (Palomares & Pauly, 2020). When more than one value for asymptotic length and von Bertalankky K were available, we used an average of all available values. While ocean oxygen observational data is not widely available, I can use World Ocean Atlas and Earth system model outputs to estimate the oxygen threshold (Table 3.1). For the illustrative examples I used oxygen and potential temperature data from a simulation of an Earth system model developed at the Geophysical Fluid Dynamic Laboratory (GFDL ESM2M; Dunne et al., 2013). The simulation has a monthly time step and spans the historical 1861-2005 period and the 2006-2100 period following the Representative Concentration Pathway 8.5 (RCP8.5) scenario (Burger et al., 2020). I averaged monthly values to produce annual oxygen and temperature time series. The RCP8.5 scenario represents a high emissions scenario with a radiative forcing of 8.5 W m-2 in year 2100 (Riahi et al., 2011). The ocean component of the GFDL ESM2M has a nominal 1o horizontal resolution and 50 vertical depth levels. The ocean biogeochemical and ecological component is version two of the Tracers of Ocean Productivity with Allometric Zooplankton module (TOPAZv2). In general, the model does well in simulating large-scale temperature and oxygen distributions in the ocean (e.g. Bopp et al., 2013; Dunne et al., 2013). However, the model tends to overestimate the volume extent of tropical oxygen minimum zones, a common bias in coarse-resolution global Earth system models (Cabré et al., 2015). Based on Henry’s Law, I converted simulated oxygen concentrations (c(O2); unit: mol/m3) into oxygen partial pressure (p(O2); unit: atm) by dividing the oxygen concentration by the pressure-corrected 42  solubility oxygen concentration (s(O2); Garcia & Gordon, 1992; Sarmiento & Gruber, 2006) and using the python package “seawater” (https://github.com/bjornaa/seawater): p(O2) = c(O2)/s(O2) (15) I divided c(O2) by the pressure-corrected solubility concentration to obtain the partial pressure of oxygen for any grid cell and depth (Sarmiento & Gruber, 2006). I then calculated the climatological averages of temperature and oxygen partial pressure for the baseline 1971-2000 period and for the future 2071-2100 period  Table 3.2. Biological and ecological input parameters necessary to estimate the Biogeographically derived metabolic index (BDMI) for Gadus morhua, Diplodus puntazzo and Callinectes sapidus. I show the parameter values for the standard application of the BDMI presented in the methods and results.  Parameter  Sharpsnout seabream (Diplodus puntazzo) Atlantic blue crab (Callinectes sapidus) Atlantic cod (Gadus morhua) Habitat Bentho-pelagic Normoxic Bentho-pelagic Normoxic Bentho-pelagic Normoxic Common depth range (m) 0 - 60 0 - 35 150 - 200 Environmental layer Surface Surface Bottom L∞ 62.200 20.000 200.000 K 0.360 0.900 0.180 LWa 0.013 0.128 0.007 LWb 3.030 2.700 3.070 d 0.700 0.700 0.700 j1 4.500 4.500 4.500 j2 8.000 8.000 8.000 Temperature preference  19.400 23.730 3.550 O2 threshold  0.199 0.200 0.164  under the RCP8.5 scenario. Temperature, oxygen and species occurrence data were re-gridded on a 0.5o latitude x 0.5o longitude global raster through linear interpolation for further analysis.  To determine the minimum distribution data requirements necessary to produce valid temperature preference and oxygen threshold estimates, I selected 10 000 random subsamples of temperature and oxygen data points that ranged between 3 and the maximum number of cells that comprise the species 43  distributions. The minimum sample size required to appropriately estimate these parameters was the breakpoint in the slope between the sample size and the range of parameter estimates, or the point in which the curve reached an asymptote in data information gain.  3.2.3 Index comparisons I compare BDMI-derived estimates with previously published physiologically derived metabolic index values (Penn et al., 2018) and expresses the ratio of oxygen supply to demand for marine fish or invertebrates as: 𝜙 = 𝐴𝑂𝑝𝑂2exp[𝐸𝑎𝑘𝐵∗(1𝑇−1𝑇𝑟𝑒𝑓)] (15) where 𝐴𝑂 is the inverse of the hypoxic threshold (minimum pO2 necessary to sustain the standard metabolic rate, in atm-1) and 𝐸𝑎, the activation energy representing temperature dependence of the oxygen threshold (in eV).  I used values of 𝐴𝑂 and 𝐸𝑎 provided by Penn et al. (2018). These parameters were generated through laboratory experiments that measured hypoxic thresholds at a minimum of three different temperatures. In these studies, the hypoxic threshold was defined as the oxygen level below which the resting metabolic rate cannot be maintained, causing an increase in anaerobic metabolism or mortality. 𝑝𝑂2 is the ambient oxygen partial pressure (in atm) for each cell in the global raster, 𝑘𝐵 is the Boltzmann constant, T is the potential temperature in Kelvin for each cell in the global raster and Tref is the reference temperature, set at 288.15 K (Penn et al., 2018). This method assumes oxygen supply increases with ambient oxygen partial pressure and respiratory efficiency.  Deutsch et al. (2015) defined the active metabolic demand (Φcrit) as the minimum oxygen supply necessary for the organism to sustain a population (e.g. growth, reproduction, defense, feeding, migration, etc.), which tends to be 2-5 times the physiologically derived metabolic index value (Deutsch et al., 2015). This method assumes that a viable habitat should have a physiologically derived metabolic index value above Φcrit. I estimated Φcrit as the 10th percentile of the physiologically derived metabolic index values calculated for a species throughout its distribution.  44  For each of the three selected species, I computed the BDMI and the physiologically derived metabolic index based on Penn et al. (2018) for baseline (1971-2000) and future (2071-2100) time periods. I compared absolute values of the metabolic indices, percent change in future metabolic index values relative to the baseline period, and future habitat loss estimated by each method. I used a linear model to explore the correlation between absolute values of the BDMI and the physiologically derived metabolic index for the baseline period. I also used a linear model to explore the correlation between the percent change in future metabolic index values relative to the baseline period for both methods. To minimize spatial autocorrelation within the data, I used a randomly selected subsample of 100 grid cells for each linear model. Finally, I estimated habitat loss as the area within a species' current distribution where future metabolic index values declined below Φcrit (the 10th percentile of the metabolic index values in the baseline species distribution range) and compared projections of lost habitat between the two methods. All analyses were carried out in the R programming environment (R Core Team, 2019).   3.3 Results and discussion  3.3.1 BDMI predictions The BDMI values for seabream, crab and cod for the baseline time period (1971 – 2000) were 0.79-2.04, 1.20-2.95 and 0.76-2.27 across their distribution range, with estimated Φcrit values of 1.04, 1.28, 1.25 respectively. Using these Φcrit as cutoff thresholds, I predicted that seabream, crab and cod would lose 16.7%, 46.8% and 13.0% of suitable habitat area, respectively, by the end of the 21st century relative to the baseline time period (Table 3.3).  3.3.2 Comparison between BDMI and physiologically derived metabolic index Our computed BDMI and the predicted loss of habitat areas relate closely with the previously published physiologically derived metabolic index. For all three species, correlations between the BDMI and physiologically derived metabolic index estimates had an R2 ≥ 0.98 for the baseline 1971-2000 period (Fig.3.1) and R2 ≥ 0.97 for the percent change by 2071-2100 relative to 1971-2000 (Fig. 3.2). The BDMI and physiologically derived metabolic index (hereafter referred to as the metabolic index) are so strongly 45   Table 3.3. Habitat loss by 2071-2100 relative to 1971-2000 according to the Biogeographically derived Metabolic Index (BDMI) and the Physiologically derived Metabolic Index. The baseline habitat area in km2, Φcrit, the 10th percentile of the metabolic index values within the species distribution for 1971-2000, is also indicated for both methods.  Species Baseline habitat area (km2) Habitat loss (%) Φcrit Biogeographically derived Metabolic Index  Physiologically derived Metabolic Index  Biogeographically derived Metabolic Index  Physiologically derived Metabolic Index  Gadus morhua 3687390 -13.01 -12.47 1.25 4.25 Callinectes sapidus 2852165 -46.83 -46.67 1.28 3.16 Diplodus puntazzo 1456658 -16.71 -17.04 1.04 7.54  46    Figure 3.1. Relationship between the Biogeographically derived metabolic index and the physiologically derived metabolic index (Penn et al., 2018) for baseline conditions (1971-2000) for three species: crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua). Dots (black) represent a random subsample of 100 cells selected to minimize spatial autocorrelation. Linear models (red line) are provided for each species. Note different y and x-axes.   47  correlated, because they both represent a ratio between oxygen supply and demand. However, differences in the indices' definitions of oxygen demand explain why both indices deviate slightly from a one-to-one relationship. For both indices, oxygen supply values represent ambient oxygen, but BDMI uses oxygen demand at the maintenance metabolic rate, while the metabolic index uses oxygen demand at the resting metabolic rate. In BDMI, maintenance metabolic oxygen demand supports survival, feeding and movement, but not growth (Pauly & Cheung, 2018), while the resting metabolic oxygen demand of the metabolic index occurs at the onset of mortality or anaerobic metabolism (Deutsch et al., 2015). Therefore, the BDMI theoretically represents a higher threshold for basic oxygen demand, resulting in an offset in the linear relationship between the two metabolic indices. In addition, the minimum metabolic index value across the current distributions of the three species, or Φcrit, represents the aerobic scope required to sustain a viable population. As expected, Φcrit is much lower for the BDMI (1.04 - 1.28) than the metabolic index (3.16 – 7.54; Table 3.3), because the maintenance metabolic oxygen demand used for the BDMI is much closer to the oxygen needed to sustain a viable population than the resting metabolic oxygen demand used by the metabolic index.  The relationship between the absolute values of BDMI and the metabolic index output for baseline conditions was slightly non-linear for crab, but linear for seabream and cod. The non-linearity for crab may be caused by the difference in growth types between fish and mollusks. Yet, there was a linear relationship between the change in both metabolic indexes by the end of the 21st century relative to baseline conditions (Fig. 3.2), and only a very small difference in habitat loss projections of 0.16% in the baseline habitat area (Table 3.3). Habitat loss projections as a percentage of baseline distribution areas were very similar between both indices (Table 3.3 and Fig. 3.3), with differences between them ranging from 0.16% for crab (Table 3.3) to 0.54% for cod (Fig. 3.3). There were small differences in the locations of habitat loss projected for the seabream and crab, but differences were much larger for cod. This larger difference observed for cod, may stem from the fact that the original 1° latitude x 1° longitude horizontal ocean resolution of the Earth system model may be too coarse to accurately capture the temperature and oxygen variations across the continental shelf regions, leading to error in the temperature preference and oxygen threshold estimates 48  of coastal demersal species. Yet, the total area of habitat loss projected for cod was very similar across both methods, with losses mainly projected along the southern limits of the species' distribution.     Figure 3.2. Linear regressions for the percent change by the end of the century (2071-2100) relative to the reference period (1971-2000) between the Biogeographically derived metabolic index and the physiologically derived metabolic index (Penn et al., 2018) for three species: crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua). Black dots represent a randomly selected subsample of 100 cells within the global raster.  49   Losses in habitat along the warmer limits of a species' distribution range, correspond to areas where the metabolic index and BDMI are closer to Φcrit. In these areas, warming causes oxygen demand to increase beyond oxygen supply, reducing the aerobic scope, with impacts on growth, reproduction and in some cases, survival (Pauly, 2010). Along the warmer edges of a species distribution, where aerobic scopes are smaller (Halsey et al., 2018), even small increases in temperature can have negative impacts. The BDMI and the metabolic index provide an understanding of the mechanisms underlying range shifts by uncovering the combined effects of temperature and oxygen on biogeography (Deutsch et al., 2015; Poloczanska et al, 2013).   Figure 3.3. Habitat loss by end of the century (2071-2100) relative to the baseline period (1971-2000) estimated by both indices (red), the Biogeographically derived metabolic index (light blue) and the physiologically derived metabolic index (green) for crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua).  3.3.3 Uncertainties The BDMI is subject to several sources of uncertainty that should be considered when interpreting the results. Deutsch et al. (2015) found that a warming-induced increase in oxygen demand is the main driver of higher metabolic index values and greater habitat loss. Yet, the BDMI assumes the same temperature dependence for all species, excluding an important source of variability. This generalization was necessary to lower data requirements for the BDMI and allow its application to a broad range of species (Cheung et al., 2011). Furthermore, the high correlation between the BDMI and the metabolic index indicates this generalization has little effect on the overall results.  50  To understand the effects of using a generalized temperature-dependence parameter on BDMI results, I calculated the index with species-specific temperature dependence values for the anabolic constant (Penn et al., 2018, Appendix B. 5) and the fixed temperature dependence for the catabolic constant (Cheung et al., 2011, Appendix B. 5), as well as with a species-specific temperature dependence parameter for both the anabolic and the catabolic constants (Pauly, 2010). Based on these sensitivity tests, I conclude that in the absence of species-specific temperature dependence parameters, the generalized version of the BDMI is well suited to quantify the effect of warming and deoxygenation on marine organisms (Appendix B. 1, Appendix B. 6, Appendix B. 7, Appendix B. 9). BDMI output for cod and seabream was similar across the two alternatives set of temperature-dependence parameter values (Appendix B. 1, Appendix B. 6, Appendix B. 7, Appendix B. 9). For crab, the BDMI with species-specific anabolic temperature dependence and a fixed catabolic temperature performed poorly; however, the predicted BDMI with species-specific anabolic and catabolic parameters, and generalized temperature dependence parameters relates closely with the metabolic index (Appendix B. 1, Appendix B. 6, Appendix B. 7, Appendix B. 9). The relationship between the absolute values of BDMI and metabolic index output for baseline conditions was non-linear for crab, but linear for seabream and cod. The non-linearity for crab was caused by the stronger species-specific temperature dependence used to compute the metabolic index (Appendix B. 1). Additional nuance can be added by understanding and modeling how temperature dependence changes across species or taxonomic groups.  Uncertainty in model output can be addressed by using a model ensemble approach (Frölicher et al., 2016). Yet, one should carefully consider the biases within the models and how they affect conditions within the study area. Given the potential impacts of deoxygenation on marine organisms and the recognized need to improve the global oxygen observation network (Keeling et al., 2009; Levin & Breitburg, 2015), there is hope that oxygen monitoring programs should begin filling this information gap soon (Moltmann et al., 2019; Pearlman et al., 2019). Access to better and more resolved data would result in better estimates of dissolved oxygen concentrations in marine habitats and thus improve the accuracy of the BDMI and its applications. Dissolved oxygen concentration data are more widely available than the partial pressure of oxygen, therefore, I also calculated the BDMI using dissolved oxygen concentrations. 51  Habitat loss projections from the BDMI using dissolved oxygen concentration differed from the metabolic index by less than 1.18% (Appendix B. 3, Appendix B. 6).  A key assumption in the calculation of BDMI is that the oxygen and temperature data points used in its derivation should be randomly distributed across the species' distribution. Therefore, it is important to compare the sample location to the species' distribution to ensure it adequately captures the species' environmental temperature and oxygen tolerance ranges. Results from the sensitivity analysis show that minimum sample sizes needed to represent a given species’ biogeography and temperature preferences vary across species: from 124 data points for oxygen and 75 for temperature for seabream, to 377 oxygen and 150 temperature data points for cod (Fig. 3.4). Application of this approach is currently underway for a larger number of species to define the absolute number of critical independent data needed (Clarke et al., in prep). The selection of the 10th percentile to determine species oxygen thresholds did not significantly affect habitat loss projections or correlations with the metabolic index (Appendix B. 8, Appendix B. 9, Appendix B. 10). Additional anthropogenic and climatic stressors not represented in BDMI can further limit an organism´s aerobic scope with impacts on habitat viability (Gallo & Levin, 2016). The aerobic scope can be limited directly by additional physiological stress (e.g., hyposalinity, lower primary productivity, lower food availability, ocean acidification), and indirectly by increased magnitudes of warming and deoxygenation (e.g., eutrophication and loss of coastal vegetation). Changes in primary productivity and resource availability under climate change may amplify the effect of temperature and oxygen on the growth of ectothermic marine organisms by causing additional physiological stress. Warming is projected to reduce primary productivity with impacts on food availability for higher trophic levels (Carozza et al., 2019). The declining food intake and increasing metabolic costs associated with warming will reduce growth rates and asymptotic sizes, lower optimal temperatures for growth, and restrict activity to shorter time periods (Huey & Kingsolver, 2019). Evidence of the impacts hyposalinity and acidification are less clear cut, while some studies report a synergistic interaction with warming and deoxygenation (Breitburg et al., 2019; Hassell et al., 2008; Wang et al., 2012), while others show no effect at all (Breitburg et al., 2019; Cheng et al., 2015; Serafin et al., 2019).  Overfishing may further amplify these effects (Free et al., 52  2019). Fisheries eliminate individuals with high performance aerobic scopes that invest more of their aerobic capacity in growth instead of the skittish behaviors that would allow them to avoid interactions with fishing gear (Duncan et al., 2019), selecting for smaller body sizes. Fisheries may also reduce genetic diversity thereby limiting the possible adaptive capacity of organisms to changes in temperature and oxygen. BDMI may not always be able to represent the impacts of warming and deoxygenation on species in the wild because responses can be modified by behavior and species interactions (Kordas et al., 2011; Spicer, 2014). The different sensitivities of species to warming and deoxygenation will change their distribution, abundance, physiological performance, activity levels and behaviour in different ways, modifying species interactions and ecosystem function (Kordas et al., 2011). While BDMI does not have the ability to model species interactions, it may be used as a driver in ecosystem models to study the effects of warming and deoxygenation on species interactions and ecosystem structure. . In summary, this study describes the derivation of the BDMI and its applications to estimate the impacts of warming and deoxygenation on water-breathing marine ectotherms that may not meet the data requirements of the metabolic index (Deutsch et al., 2015). The BDMI can be derived with all necessary input parameters and data available from open access sources (Table 3.1). Our study supports the broad applicability and use of BDMI to understand the impacts of ocean warming and deoxygenation on marine biodiversity and fishery resources.  53    Figure 3.4. Range of temperature preferences and oxygen thresholds estimated from 10 000 subsamples randomly selected across a species' distribution, where subsample sizes ranged from 3 to the maximum number of cells that comprise the species' entire distribution. The black lines indicate the sample size at which the ranges of temperature preference and oxygen threshold reach an asymptote. Results presented for crab (Callinectes sapidus), seabream (Diplodus puntazzo), and cod (Gadus morhua). 54  Chapter 4: Impact of warming and deoxygenation on pelagic fisheries of the Eastern Tropical Pacific  4.1 Introduction Ocean warming affects the physiological performance of marine fishes and invertebrates, with cascading effects on body size, abundance and survival (Cheung et al., 2013a; Deutsch et al., 2015). Metabolic rate (rate of conversion of food to ATP with oxygen as an input), generally scales positively with temperature, so more oxygen is needed to support aerobic metabolism in warmer waters (Clarke & Johnston, 1999). Under ocean warming, increases in oxygen demand may surpass the ability of some fishes to maintain sufficient oxygen supply, even in well-oxygenated pelagic environments (Deutsch et al., 2015). This may lead to a contraction of the aerobic scope (difference between the standard metabolic rate and the maximum metabolic rate), forcing trade-offs between vital oxygen demanding physiological processes that may affect an organism’s growth, reproduction and mortality rate (Baudron et al., 2014; Pörtner et al., 2017; Sokolova et al., 2013). Impacts at the organism level can have knock-on effects at the population, community and ecosystem levels (Baudron et al., 2014; Cheung et al., 2013a; Deutsch et al., 2015). Ocean warming is changing species composition of fisheries around the world (Cheung et al., 2013b; Gamito et al., 2015; Martínez-Ortiz et al., 2015). In temperate and sub-tropical regions, the average temperature preference of species in fisheries catches has increased since the 1970s. In the tropics, the average temperature preference of species in the catch has also increased but reached a plateau once only species with high temperature preferences remained (Cheung et al., 2013b). Such changes in average species temperature preferences have also been reported for non-exploited marine biological communities (Bates et al., 2014; Bianchi & Morri, 2003; Vergés et al., 2014). Oxygen limitation may be shaping the relationship between species´ temperature preferences and ocean warming (Cheung et al., 2013b; Pauly & Cheung, 2018). Acute or short-term warming can increase metabolic oxygen demand above supply, with impacts on body size and abundance that would lower the proportion such a species represents within the catch. Species living in warmer waters have 55  adapted to the long-term exposure to higher temperature by reducing their standard metabolic rates (Jutfelt, 2020; Sandblom et al., 2016). In other words, at a reference temperature, the oxygen demand of a temperate species will be higher than that of a tropical species. Therefore, as warming continues, catch composition is expected to shift towards a dominance of species with higher thermal tolerances and lower oxygen demands.  The impacts of warming and oxygen limitation on the physiological performance of marine fishes and invertebrates can be assessed with a Biogeographically derived metabolic index (BDMI), which is a ratio of oxygen supply to demand of a species or population (Chapter 3). A relatively high metabolic demand (low BDMI) indicates a higher sensitivity to warming and deoxygenation than a relatively low metabolic demand (high BDMI) (Deutsch et al., 2015). Ocean warming and deoxygenation lower BDMI by increasing oxygen demand and reducing oxygen supply, respectively (Deutsch et al., 2015). An indicator (Biogeographically derived Metabolic Index of the Catch, BDMC) combining BDMI with fisheries catch data can help elucidate the impacts of ambient temperature and oxygen on the physiological performance of species in the catch.  A second indicator (Mean Oxygen Demand of the Catch, MODC) combining species oxygen demand at a reference temperature with fisheries catch data can help elucidate the impacts of ambient temperature and oxygen on changes in catch composition. We hypothesize that ocean warming, initially, allows species that are warm adapted (i.e., with lower metabolic oxygen demand and higher BDMI) to thrive. Thus, we expect the average BDMI of the catch (BDMC) to increase as the Mean Oxygen Demand of the Catch (MODC) declines. As warming continues, eventually, only warm adapted species remain in the catch, and will begin to BDMC decline (Fig. 4.1).  In the Eastern Tropical Pacific Ocean, the latitudinal temperature gradient (average annual SST 23.1 °C - 29.3°C) and the large temperature and oxygen anomalies driven by El Niño Southern Oscillation provide a “natural experiment” to examine the relationship between temperature, oxygen, and the changes in physiological performance that shape community structure. Commercial pelagic fisheries in the Eastern Tropical Pacific Ocean may act as bellwethers of the impacts of warming and deoxygenation on pelagic ecosystems (Woodworth-Jefcoats et al., 2019), as they target many species with high oxygen demands, such as sardine, tuna and billfish (Bertrand et al., 2011; Mislan et al., 2017; Prince & 56  Goodyear, 2006). Furthermore, in the tropics, temperatures and oxygen concentrations are closer to species’ physiological tolerance limits, so even a small degree of warming could drive species to shift towards cooler, more oxygenated waters (Breitburg et al., 2018).  In this study, we aim to test the hypothesis that changing ocean temperatures and oxygen levels are playing an important role in shaping the composition of fisheries catches in the Eastern Tropical Pacific Ocean. Specifically, analyzing fisheries catch data from eight countries’ exclusive economic zones (Mexico to Ecuador) between 1970 and 2009, we expect that pelagic fisheries catch will be dominated by tropical species with low oxygen demands in warmer and/or less oxygenated waters and that temperature and oxygen thresholds may separate different species assemblages.   4.2 Methods 4.3 Catch data The geographical domain of this study included the Exclusive Economic Zones (EEZs) of countries bordering the Eastern Tropical Pacific Ocean from Mexico in the north, to Ecuador in the south. We present results for Ecuador and Galapagos separately, because of differences in catch composition and oceanographic conditions. Given the uncertainty in ESM projections for upwelling areas, I exclude Peru from the analysis. Our analysis focused on fisheries for small pelagics such as sardines (Sardinops spp., Opisthonema spp., Ethmidium spp.), anchovies (Engraulis spp., Cetengraulis spp.) and mackarels (Acanthocybium spp., Scomber spp., Scomberomorus spp., Euthynnus spp.), and large pelagics such as tuna (Thunnus spp.), mahi mahi (Coryphaena hippurus), billfish (Istiophorus sp., Makaira sp., Xiphias sp.) and sharks (Lluch-Cota et al.,2019). We used the reconstructed historical catch time series data from the Sea Around Us to calculate the catch-based indicators used in this study (Cisneros-Montemayor et al., 2015; Donadi et al., 2015; Haas et al., 2015; Harper et al., 2014; Lindop, 2015; Lindop et al., 2015; Pauly et al., 2020; Trujillo et al., 2015; Zeller et al. 2016). The Sea Around Us data is used because it has higher degree of taxonomic resolution and estimates of potential discards. To minimize the effects of fishing on the analysis, we only included catch time series from species that were caught for more than 30 years, as 57  fisheries commonly expand their number of target species as they deplete traditional target species. In total, this study included 25 species that contributed to 99% of the total pelagic catch from 1970 to 2009 (Appendix C. 8).    Figure 4.1. A diagram explaining hypothesized trends of mean oxygen demand of the catch (MODC) and biogeographic-derived metabolic index of the catch (BDMC) in subtropical and tropical pelagic catches. MODC is a catch-based indicator of average temperature independent oxygen demand that represent warm adaptation in fish populations. BDMC is a catch-based indicator of the ratio between oxygen supply and temperature-dependent oxygen demand and represents warm-adaptation in fish and warming increases in metabolic rate. I hypothesize that in subtropical areas, MODC declines with warming, as the abundance of species with high oxygen demands decreases and leads to an increase in BDMC. In tropical areas, MODC is stable because only tropical species with low oxygen demands remain in the catch, thus BDMC declines with warming-induced increases in oxygen demand and declining oxygen supply.  58  4.3.1 Environmental data We obtained sea surface temperature from the Centennial in-situ Observation Based Estimates (COBE) provided by NOAA/OAR/ESRL (https://www.esrl.noaa.gov/psd/) (Ishii et al., 2005). The temperature time series has a monthly time step from 1891 until present and is interpolated on a global grid of 1 ° latitude by 1° longitude.  In the absence of time series of oxygen concentration data for each of the studied exclusive economic zones, we used sea surface oxygen concentration (mol/m3) from ocean model hindcast simulations. Specifically, hindcast outputs were obtained from three Earth system models included in the Coupled Models Intercomparison Project Phase 6 (CMIP6): the second generation Earth system model developed by the Centre National de Recherches Météorologiques, CNRM-ESM2-1 (Séférian et al., 2019), the Community Earth system model Version 2, CESM2 (Danabasoglu et al., 2020) and the latest version of the Institut Pierre Simon Laplace climate model, IPSL-CM6A-LR (Boucher et al., 2020). The hindcast simulation is 62-year long (744 months), from 1948 to 2009 (Griffies et al., 2016; Orr et al., 2017). All data were re-gridded on a °1 latitude x °1 longitude map of the ocean.  For each EEZ, we estimated the annual average sea surface temperature and oxygen time series from 1970 to 2009. We produced global 1971-2000 climatologies of sea surface temperature and sea surface oxygen to estimate species temperature preferences and oxygen thresholds.   4.3.2 Mean Oxygen Demand of the Catch (MODC) and the Biogeographically derived metabolic index for the Catch (BDMC). To calculate BDMC and MODC, we first computed the theoretical oxygen level required by an individual to support its maintenance metabolic rate (Chapter 3): 𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 = (𝑊1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗2/𝑇 𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ 𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓𝑊∞1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗1/𝑇 𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓) (1) where W is average body weight (in g) and 𝑊∞ is the asymptotic weight (in g), estimated from the asymptotic length 𝐿∞ (in cm) and the species´ length-weight relationship (Froese & Pauly, 2019), while K is the von Bertalanffy K (Froese & Pauly, 2019). d is an anabolic scaling coefficient with a value of 0.7 (Pauly & Cheung, 2018). An abbreviated form of the Arrhenius equation (Clarke & Johnston, 1999) 59  𝑒−𝑗1/𝑇 and 𝑒−𝑗2/𝑇 is included in (1) to account for the temperature-dependence of anabolism and catabolism. Parameters j1 and j2 are the activation energies (0.388 eV and 0.689 eV; Cheung et al., 2011) divided by the Boltzmann constant, resulting in a j1 and j2 of 4500K and 8000K, respectively (Cheung et al., 2011). T is the temperature in Kelvin and 𝑂2,𝑠𝑢𝑝𝑝𝑙𝑦 is the ambient sea surface dissolved oxygen concentration in mol/m3. The species-specific oxygen threshold concentration (O2,thresh, expressed in mol/m3) is the 10th percentile of the sea surface oxygen concentrations across its distribution. 𝑇𝑝𝑟𝑒𝑓 is the species temperature preference (in Kelvin), expressed as the median sea surface temperature across the species’ distribution (Cheung et al., 2013a; Cheung et al., 2008).  The Biogeographically derived metabolic index (BDMI) is the ratio of oxygen supply to demand: 𝐵D𝑀𝐼 =  𝑂2,𝑠𝑢𝑝𝑝𝑙𝑦 𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 ⁄  (2) Where 𝑂2,𝑠𝑢𝑝𝑝𝑙𝑦  is the average annual dissolved oxygen concentration in each EEZ (in mol/m3) and 𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑   is the species’ oxygen demand (in mol/m3) computed in equation 1.  We then calculated a Biogeochemically-derived metabolic index of the Catch (BDMC) based on the community-averaged BDMI (from equation 2) weighted by the catch: 𝐵𝐷𝑀𝐶𝑦𝑟 =∑ 𝐵𝐷𝑀𝐼𝑖,𝑇𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟𝑛𝑖∑ 𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟𝑛𝑖   (3) where 𝐶𝑎𝑡𝑐ℎ𝑖,𝑠 is the weight of the catch of species i in each EEZ, 𝐵𝐷𝑀𝐼𝑖,𝑇 is the oxygen demand of species i at sea water temperature T and n is the total number of species. We also calculated a Mean Oxygen Demand of the Catch (MODC) that represented the average oxygen demand (O2 demand) from equation (1) weighted by the annual catch of species i:  𝑀𝑂𝐷𝐶𝑦𝑟 =∑ 𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 𝑖,𝑇𝑟𝑒𝑓∙𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟𝑛𝑖∑ 𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟𝑛𝑖   (4) where 𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟 is the catch of a species i in each EEZ, 𝑂2, 𝑑𝑒𝑚𝑎𝑛𝑑𝑖,𝑇 is the oxygen demand of species i at a reference sea water temperature, Tref as represented by the average sea surface temperature in the study region over 1970-2009.   Finally, the Mean Temperature of the Catch (MTC) was the community-averaged temperature preference weighted by the catch (Cheung et al., 2013a): 60  𝑀𝑇𝐶𝑦𝑟 =∑ 𝑇𝑝𝑟𝑒𝑓𝑖∙𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟𝑛𝑖∑ 𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟𝑛𝑖   (5) where 𝐶𝑎𝑡𝑐ℎ𝑖,𝑦𝑟 is the catch of a species i in each EEZ, 𝑇𝑝𝑟𝑒𝑓 is the temperature preference of species i.  We computed the annual BDMC, MODC and MTC for each exclusive economic zone from 1970 to 2009.  4.3.3 Spatial trends A visual inspection of the data revealed a non-linear relationship between temperature and both BDMC and MODC. Therefore, we applied segmented regressions to understand the relationship between the two environmental variables (temperature and oxygen) with BDMC and MODC, using the package ‘segmented’ in R (Muggeo, 2003, 2017). Segmented regressions may be useful to identify the threshold below which physiological performance begins to decline, and the threshold that separates temperate and subtropical communities from tropical communities. Given the uncertainties inherent in the limited amount of data analyzed in this study, future studies should examine different fisheries dependent and independent data from different location to further explore the generality of thresholds. 4.3.4 Temporal trends The annual rates of warming and deoxygenation in each EEZ were estimated as the slope of linear regressions between decade and temperature, as well as year and oxygen. We estimated the correlation between MODC, BDMC, MTC, temperature, oxygen level, and the Oceanic Niño Index (ONI, https://origin.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ONI_v5.php). ONI is the three-month running mean of sea surface temperature anomalies in the 3.4 Niño region (5N-5S, 170W-120W). Values above 0.5 indicate an EL Niño, while values below -0.5 indicate a La Niña.  We applied wavelet coherency analysis to detect any transient linear correlations between BDMC and MODC with El Niño Southern Oscillation (ENSO) time series (Gouhier et al., 2019). Firstly, we averaged the BDMC and MODC time series across ESM to obtain an ensemble model for each indicator, then standardized each time series by applying a Continuous Wavelet Transform function. Specifically, we applied the Morlett wavelet function, which is commonly applied to analyze ecological data. The Morlett function is a continuous wavelet from which we can extract time-dependent amplitude and frequency, 61  represented in Appendix C.6, Appendix C.7. We performed a cluster analysis based on the dissimilarities among the wavelet cross-coherency between BDMI and ONI, as well as between MODC and ONI with a maximum correlation analysis. All wavelet analyses were performed with the wavelet library developed by (Gouhier et al., 2019).  4.4 Results 4.4.1 Spatial trends Between 1970 and 2009, sea surface temperatures and sea surface oxygen were negatively correlated (R2 > 0.8). Average annual temperature and oxygen concentration ranged between 23.1 - 29.3°C and 0.20 - 0.22 mol/m3, respectively (Fig. 4.2). The northern and southern most exclusive economic zones of the Eastern Tropical Pacific Ocean (Mexico and Ecuador) had the lowest temperatures and highest dissolved oxygen concentrations. From Guatemala to Colombia (north to south), temperature declined, and oxygen  concentrations increased (Fig. 4.2).  MODC declined with temperature until a threshold of 25.18 °C - 25.21°C, after which it stabilized and remained low (Fig. 4.3, Table 4.1). On the other hand, BDMC increased with temperature until 25.18°C - 25.35°C, after which it declined (Fig. 4.3, Table 4.2). MODC increased with oxygen until it reached a threshold of 0.210-0.213 mol/m3, and then plateaued (Fig. 4.3, Table 4.1), while BDMC increased until 0.211-0.214 mol/m3, and then declined (Fig. 4.3, Table 4.2). Temperatures were much cooler than the threshold in Mexico, very close to the threshold in Ecuador and Galapagos, and much warmer in the remaining EEZ (Fig. 4.3). Oxygen concentrations were higher than the threshold in Mexico and Ecuador, very similar to the threshold in Galapagos, Colombia and Panama, and much lower in the remaining EEZ (Fig. 4.3).   4.4.2 Temporal trends The rate of warming and deoxygenation were negatively correlated (R2 = 0.68). The slope between temperature and year ranged from 0.008 to 0.014 year -1 (Appendix C. 2), while the slope between oxygen and year ranged from -5.17x10-5 to -2.11 x10-5 year -1 for CESM, -4.7 x10-5 to 2.43 x10-5 year -1 for CNRM,  62     Figure 4.2. Average sea surface temperature (°C) and sea surface oxygen concentration (mol/m3) (left panel) and their rate of change (right panel) in exclusive economic zones of the Eastern Tropical Pacific Ocean from 1970 to 2009 based on Centennial in-situ Observation Based Estimates sea surface temperature measurements and hindcast model outputs for oxygen concentration from three Earth system models: A) CESM2, B) CNRM-ESM2-1 , C) IPSL-CM6A-LR . The rate of change of sea surface temperature and sea surface oxygen from 1970 to 2009 is also shown for D) CESM2, E) CNRM-ESM2-1, F) IPSL-CM6A-LR. Temperature and oxygen data are categorized according to the values below the 33rd percentile, between the 33rd and the 67th percentile and above the 67th percentile. 63     Figure 4.3. Segmented regressions between A) sea surface temperature (°C) and the Mean Oxygen Demand of the Catch (MODC); B) sea surface temperature (°C) and the mean Biogeographically derived metabolic index of the catch (BDMC); C) sea surface oxygen (mol/m3) and MODC; D) sea surface oxygen (mol/m3) and BDMC. Results based on different oxygen outputs, CESM2 in red, CNRM-ESM2-1 in blue and IPSL-CM6A-LR in purple. Squares represent average MODC and BDMC values between 1970 and 2009 in each Exclusive Economic Zone: (ME) Mexico, (GU) Guatemala, (ES) El Salvador, (NI) Nicaragua, (CR) Costa Rica, (PA) Panama, (CO) Colombia, (EC), Ecuador, (GA) Galapagos.  and -4.95 x10-5 to 2.59x10-5 year -1 for IPSL (Appendix C. 3). Oxygen trends were similar for CNRM and IPSL outputs and were found to increase over the reference time period in Guatemala, El Salvador, Nicaragua and Ecuador, and to decline in the remaining EEZs. According to CESM model outputs, deoxygenation occurred in all EEZ, with the lowest rates of deoxygenation recorded for Galapagos and Ecuador, and the highest for Mexico, Panama and Colombia (Fig. 4.2).  64  Table 4.1. Segmented regression analyses between sea surface oxygen concentration (mol/m3) and Mean Oxygen Demand of the Catch (MODC) as well as between sea surface temperature and MODC of pelagic fisheries in the Eastern Tropical Pacific (from Mexico to Ecuador) from 1970 to 2009. Segmented regression analyses were conducted for each set of Earth system model (ESM) outputs used to compute MODC.   ESM R2 Breakpoint  Std. error Parameter Estimate Std.error t-value p-value  CESM2 0.6 0.213 mol/m3 0.000 Intercept 0.350 0.052 6.74 <0.001     Oxygen -0.944 0.248 -3.81 <0.001     Difference in slopes coefficient 6.637 0.457 14.52 <0.001 CNRM-ESM2-1  0.4 0.211 mol/m3 0.001 Intercept 0.202 0.081 2.50 0.013     Oxygen -0.238 0.386 -0.62 0.537     Difference in slopes coefficient 2.856 0.441 6.47 <0.001 IPSL-CM6A-LR  0.37 0.21 mol/m3 0.001 Intercept 0.212 0.067 3.19 0.002     Oxygen -0.293 0.321 -0.91 0.363     Difference in slopes coefficient 3.111 0.491 6.34 <0.001 CESM2 0.57 25.21 °C 0.099 Intercept 0.728 0.046 16.01 <0.001     Temperature -0.023 0.002 -12.29 <0.001     Difference in slopes coefficient 0.023 0.002 12.06 <0.001 CNRM-ESM2-1  0.58 25.18 °C 0.096 Intercept 0.727 0.045 16.22 <0.001     Temperature -0.023 0.002 -12.46 <0.001     Difference in slopes coefficient 0.023 0.002 12.15 <0.001 IPSL-CM6A-LR  0.57 25.21 °C  0.098 Intercept 0.716 0.044 16.20 <0.001     Temperature -0.023 0.002 -12.41 <0.001          Difference in slopes coefficient 0.023 0.002 12.15 <0.001 65  Table 4.2. Segmented regression analysis between sea surface oxygen (mol/m3) and mean Biogeographically derived Metabolic Index of the Catch (BDMC) as well as between sea surface temperature and BDMC of pelagic fisheries in the Eastern Tropical Pacific Ocean (from Mexico to Ecuador) from 1970 to 2009. Segmented regressions were conducted for each set of Earth system model (ESM) outputs used to compute BDMC.   ESM R2 Breakpoint   Std. error Parameter Estimate Std.error t-value p-value CESM2 0.6 0.214 mol/m3 0.000 Intercept -5.009 0.363 -13.81 <0.001     Oxygen 30.154 1.730 17.43 <0.001     Difference in slopes coefficient -55.253 4.560 -12.12 <0.001 CNRM-ESM2-1  0.3 0.211 mol/m3 0.001 Intercept -3.527 0.741 -4.76 <0.001     Oxygen 23.112 3.548 6.51 <0.001     Difference in slopes coefficient -15.310 3.903 -3.92 0.0170 IPSL-CM6A-LR  0.5 0.213 mol/m3 0.001 Intercept -3.005 0.411 -7.31 <0.001     Oxygen 20.751 1.985 10.45 <0.001     Difference in slopes coefficient -21.292 4.782 -4.45 <0.001 CESM2 0.6 25.35 °C 0.110 Intercept 0.845 0.334 -2.53 0.012     Temperature 0.092 0.014 6.74 <0.001     Difference in slopes coefficient -0.168 0.015 -11.47 <0.001 CNRM-ESM2-1  0.6 25.32 °C 0.120 Intercept -0.713 0.327 -2.18 0.030     Temperature 0.087 0.013 6.49 <0.001     Difference in slopes coefficient -0.159 0.014 -11.13 <0.001 IPSL-CM6A-LR  0.6 25.18 °C 0.000 Intercept -1.171 0.362 -3.23 <0.001     Temperature 0.106 0.015 7.13 <0.001     Difference in slopes coefficient -0.181 0.016 -11.62 <0.001  There was a high degree of interannual variability in all the BDMC and MODC time series. Average BDMC values were highest in Ecuador and Galapagos (Appendix C.4). BDMC followed an increasing trend through time in Ecuador and was consistently high in the Galapagos. Although BDMC was also high in Mexico during the 1970s, it declined until the 1980s, when it reached levels similar to Guatemala through Costa Rica, increased thereafter before declining again between 2005 and 2010. While BDMC in Guatemala increased throughout the time series, BDMC in El Salvador, Nicaragua and Costa Rica declined slightly. BDMC in Panama and Colombia was higher on average, through the time series, but with no long-term trend. MODC was highest in Mexico, where it increased until the 1980s, and declined thereafter (Appendix C.5). While MODC in Guatemala also followed a declining trend, MODC, on average, had much  66  lower values. MODC in El Salvador, Nicaragua, Costa Rica, Panama, Columbia and Galapagos remained consistently low through time, only with MODC in Panama, Colombia and Galapagos registering slightly lower values. Except for Mexico, MODC was highest in Ecuador, increasing slightly until 1990, after which it declined abruptly, increased sharply in 2000 and declining thereafter, although with considerable interannual variability. Year was not correlated with temperature nor oxygen, indicating long-term warming and deoxygenation trends were weak relative to the interannual variability produced by ENSO. BDMC and Oceanic Nino Index (ONI) were negatively correlated, as higher ONI values indicate warmer waters (Fig. 4.4). MODC and BDMC co-vary negatively with temperature and positively with oxygen (Fig. 4.4). MODC and BDMC were negatively correlated in all eight EEZs; however, the slope of the regression between these indicators was much steeper for some EEZ than for others (Table. 4.3). In Mexico, Guatemala, El Salvador, Nicaragua, Costa Rica and Ecuador, where the relationship was steeper, larger declines in BDMC were required to produce a decline in MODC. In other words, larger losses in physiological performance were required to produce a shift in species composition, making catch composition more stable. In Panama, Colombia, and Galápagos, a smaller decline in physiological performance was required to produce a shift in species composition. In these EEZ, the slope was weaker, as smaller declines in BDMC were required to change MODC (Table 4.3).   Country Slope R-squared p-value Mexico -1.012 0.806 <0.001 Guatemala -0.718 0.639 <0.001 El Salvador -0.5405 0.457 <0.001 Nicaragua -0.754 0.432 <0.001 Costa Rica -0.498 0.361 <0.001 Panama -0.227 0.036 0.125 Colombia -0.261 0.038 0.118 Ecuador -0.736 0.556 <0.001 Galapagos -0.402 0.099 0.027  Table 4.3. Linear regression analyses between the normalized Mean Oxygen Demand of the Catch (MODC) and the mean Biogeographically derived Metabolic Index of the Catch (BDMC) of pelagic fisheries in the Eastern Tropical Pacific Ocean from 1970 to 2009.  67   Year ONI Temperature Oxygen BDMC MODC MTC A) CESM2        MTC 0.1 0.0 0.6 -0.5 0.3 -1.0  MODC -0.1 0.0 -0.6 0.6 -0.3  -1.0 BDMC 0.0 -0.2 -0.6 0.6  -0.3 0.3 Oxygen -0.1 -0.2 -0.9  0.6 0.6 -0.5 Temperature 0.1 0.2  -0.9 -0.6 -0.6 0.6 ONI 0.0  0.2 -0.2 -0.2 0.0 0.0 Year  0.0 0.1 -0.1 0.0 -0.1 0.1 B) CNRM-ESM2-1      MTC 0.1 0.0 0.6 -0.5 0.3 -1.0  MODC -0.1 0.0 -0.6 0.6 -0.2  -1.0 BDMC 0.0 -0.3 -0.6 0.5  -0.2 0.3 Oxygen -0.1 -0.3 -0.8  0.5 0.6 -0.5 Temperature 0.1 0.2  -0.8 -0.6 -0.6 0.6 ONI 0.0  0.2 -0.3 -0.3 0.0 0.0 Year  0.0 0.1 0.0 0.0 -0.1 0.1 C) IPSL-CM6A-LR      MTC 0.1 0.0 0.6 -0.5 0.3 -1.0  MODC -0.1 0.0 -0.6 0.6 -0.3  -1.0 BDMC 0.0 -0.2 -0.6 0.6  -0.3 0.3 Oxygen -0.1 -0.2 -0.9  0.6 0.6 -0.5 Temperature 0.1 0.2  -0.9 -0.6 -0.6 0.6 ONI 0.0  0.2 -0.2 -0.2 0.0 0.0 Year  0.0 0.1 -0.1 0.0 -0.1 0.1    Figure 4.4. Correlation between the Mean Temperature of the Catch (MTC), the Mean Oxygen Demand of the Catch (MODC), the Biogeographically derived Metabolic Index (BDMC), sea surface oxygen (mol/m3), sea surface temperature (°C), Oceanic Niño Index (ONI) and year. Correlation analyses were conducted based on oxygen hindcast outputs from the Earth system models: A) CESM2, B) CNRM-ESM2-1, C) IPSL-CM6A-LR. Significant correlations were underlined (p<0.05). Correlation coefficients (R) are displayed inside every cell. Red indicates negative correlations and blue indicate positive correlations. The color scale represents the strength of the correlation, with dark tones representing R values closer to 1 or -1, and light tones representing values closer to 0.1 or -0.1.  68   Figure 4.5. Standardized time series of the Mean Oxygen Demand of the Catch (MODC) and Mean Biogeographically derived Metabolic Index of the Catch (BDMC) of pelagic fisheries in the exclusive economic zones in the Eastern Tropical Pacific Ocean. Standardized Oceanic Niño Index is also plotted. The dashed line is a reference point at 0.  69  The coherence between ENSO and BDMC time series was strong and consistent throughout the study period (Appendix C. 7), while the coherence between ENSO and MODC time series was episodic (Appendix C. 6). The warming and cooling produced by different ENSO phases always caused changes in BDMC but did not consistently cause changes in catch composition (MODC). In fact, the relationship between BDMC and MODC shifted between positive and negative, as MODC and BDMC responded differently to changes in temperature and oxygen (Figure 4.5). The coherence between BDMC and ENSO was very strong and similar across EEZs from El Salvador through Colombia and Galápagos (Appendix C. 7).  4.5 Discussion Our study sought to test the hypothesis that the sensitivity of species’ physiological performance to changing ocean temperatures and oxygen levels plays an important role in shaping the composition of fisheries catches in eight EEZs across the Eastern Tropical Pacific Ocean. To that end we applied two indices, Mean Oxygen Demand of the Catch (MODC) and Biogeographically derived metabolic index of the Catch (BDMC), and show that the constraints of warming on species´ physiological performance, through its relationship with oxygen demand and supply, does indeed shape pelagic catch composition across the region. While BDMC is an indicator of average species physiological performance in the catch, MODC is an indicator of average species oxygen demand at a reference temperature, which generally declines with temperature, allowing us to detect the effect of shifting species on average oxygen demands in the catch. In EEZs with on average cooler water, impacts of warming on physiological performance began excluding species with high oxygen demand from the catch (for example, Trachurus murphyi, Sarda chilensis, Thunnus alalonga), causing declines in MODC. In the warm waters of tropical EEZs, biological communities were already dominated by species with low oxygen demand (for example, Euthynnus linneatus, Thunnus orientalis, Acanthocybium solandri, Dosidicus gigas). These results support the hypothesis that warming-induced impacts on physiological performance have led to an increasing dominance of species with low oxygen demand in the catch (Cheung et al., 2013a). Importantly, we were able to determine clear temperature and oxygen thresholds that separate different species compositions and physiological responses to warming and deoxygenation. When temperature 70  was above the threshold (25.18-25.21 °C), species with low oxygen demand dominated the catch, and the average species physiological performance (BDMC) declined due to a narrowing gap between oxygen demand and supply. As warming continues, it is possible that more species will be excluded from the catch due to oxygen limitation.  Patterns in MODC and BDMC were mainly driven by the regional temperature gradient. Temperatures in Mexico and Ecuador are, on average, cooler than the threshold, as a result of the California Current and equatorial upwelling systems bringing colder nutrient rich waters to the surface, while temperatures in Central American EEZs were warmer than the threshold (Fiedler & Lavín, 2017). The oxygen and temperature thresholds we detected here may represent critical tipping points – where a small change could push a system into abrupt or irreversible change (Griffiths et al., 2013). These tipping points are indicative of profound shifts in ecosystem functions with important consequences for human use and the long-term delivery of ecosystem services. Such tipping points are typically hard to identify, yet if known can help shape monitoring and management programmes that support the sustainable management of marine resources. In Galapagos, Panama and Colombia, where oxygen concentrations were near the threshold, fisheries catch composition is highly variable. Consequently, pelagic fisheries in these EEZs may respond faster to short-term temperature variations such as marine heatwaves caused by El Niño. Catch compositions were much more stable in EEZs with oxygen level further above or below the threshold.  MODC and BDMC stabilized over time in EEZs characterized by warmer waters and where most species have low oxygen demand. Despite their low MODC, EEZs in the Eastern Pacific warm pool (Guatemala, El Salvador, Nicaragua, Costa Rica) had the lowest BDMCs. Therefore, they may be particularly vulnerable to additional stressors that cause physiological stress and limit organisms’ aerobic scope, such as ocean acidification, overfishing and declining (Horwitz et al., 2020; Laubenstein et al., 2018). Consequently, this region may be less resilient to El Niño that can drive temperatures above the physiological tolerance limits of many species (Smale et al., 2019).  The strong ENSO-driven interannual temperature variability was found to drive the temporal BDMC. ENSO continuously affects BDMC amongst EEZs in the Eastern Tropical Pacific Ocean in all EEZs except Mexico and Ecuador, where temperatures were cooler. The sharp temperature anomalies 71  increase species oxygen demand, thereby reducing the gap between oxygen demand and supply. This may have consequences on species growth and abundance. In some but not all ENSO events, changes in the ratio between oxygen demand and supply drive changes I catch composition. These results show the potential application of the BDMI as a tool that could be used to measure and eventually predict, the impacts of ENSO on fisheries resources.  Strong El Niño are known to affect fisheries catches and species distributions throughout the Eastern Tropical Pacific region (Fiedler, 2002; Leung et al., 2019; Mora & Robertson, 2005; Zapata Padilla, 2002). We found ENSO’s impact on BDMC to be lasting longer and becoming more immediate after stronger El Niño events (Appendix C.7). This could reflect the different ways in which El Niño affects a population, from the short-term effects of highly mobile pelagic species moving away from unfavorable conditions (Carlisle et al., 2017; Farchadi et al., 2019; Marín-Enríquez & Muhlia-Melo, 2018), to the longer-term effects of impaired population growth, recruitment and reproductive success (Watters et al., 2003). In addition to the direct climate effects, ENSO also causes shifts in ecological interactions (Ruiz-Cooley et al., 2017), which could create noise in the observed responses of BDMC and MODC. Deoxygenation is likely one of the key factors driving changes in the catch composition of pelagic fisheries in the Eastern Tropical Pacific Ocean. Oxygen itself may become limiting in pelagic environments as hypoxic waters are upwelled to the surface and as oxygen minimum zones expand (Duteil et al., 2018; Stramma et al., 2012), compressing the volume of suitable habitat for sardines, tuna and billfish (Mislan et al., 2017). This study was not able to detect the impact of deoxygenation on the catch composition of pelagic fisheries because it is based on the environmental conditions at the ocean surface, where oxygen concentration approximates oxygen solubility. However, habitats of pelagic species extend hundreds of meters below the surface, where oxygen levels decline rapidly and may limit species distributions. Future research should explore how the three-dimensional impacts of warming and deoxygenation on metabolically viable pelagic habitats may affect fisheries.  Fisheries may also confound trends in BDMC and MODC, as they can influence the relative composition of the catch. Specifically, fisheries may increase the proportion of small, fast growing species in the catch. These species may also have lower oxygen demands. Future studies should account for 72  fishing effort in models detecting factors causing trends in both indices. Alternatively, trends can be validated with fishery independent information. Future studies should also conduct separate analysis for oceanic large pelagic fisheries, coastal large pelagics fisheries and small pelagics fisheries, as they may be influenced by different oceanographic conditions. The key role that temperature was found to play in structuring catch composition of pelagic fisheries, directly and via its relationship with oxygen, suggests that catches may be vulnerable to warming under climate change. Here, we show how the application of BDMC and MODC can help improve our understanding of the impacts of warming and deoxygenation on marine fisheries. Together, these indicators reveal how the reductions physiological fitness caused by warming can lead to pelagic fisheries that are increasingly dominated by species with low oxygen demands. Importantly, identified thresholds and applied indices can help inform monitoring methods that support sustainable management measures for marine resources in the Eastern Tropical Pacific Ocean.     73  Chapter 5: Temperature and oxygen supply shape the demersal community in a tropical oxygen minimum zone  5.1 Introduction The open ocean lost 0.5-3.3% of its oxygen content between 1970 and 2010 (Schmidtko et al., 2017) and is projected to lose an additional 3.2-4.7% by the end of the 21st century under the high greenhouse gas emissions scenario (Keeling et al., 2009). However, oxygen levels are not homogenous throughout the ocean, oxygen minimum zones with concentrations below 0.05-0.08 mol/m3 develop in the Eastern Tropical Pacific and Atlantic oceans where high respiration rates meet sluggish ventilation (Hameau et al., 2020; Oschlies et al., 2018). Oxygen levels within oxygen minimum zones decreased ten times faster than the global average (4% per decade) between 1960 and 2010, quadrupling the volume of anoxic waters (Schmidtko et al., 2017). Oxygen minimum zones now occupy 8% of the ocean area (Limburg et al., 2017) and projections indicate they will continue to expand under ocean warming (Bopp et al., 2013; Stock et al., 2019).  The North and Equatorial Pacific Ocean are the main areas affected by deoxygenation, where 39.9% of global oxygen losses have occurred between 1960 and 2010 (Schmidtko et al., 2017). The North Eastern Pacific contains a large oxygen minimum zone that shoals up to 50 m below the surface and has large functionally anoxic cores (Gallo & Levin, 2016; Tiano et al., 2014). Few species are adapted to live in these extreme low oxygen environments (Childress & Seibel, 1998; Gallo & Levin, 2016). However, the high concentration of food along the oxygen minimum zone margins sustains large aggregations of some fish and crustacean species (Bianchi, 1991; Gallo & Levin, 2016) that support deep water shrimp and squat lobster fisheries along Central America (Wehrtmann et al., 2012). Gallo & Levin (2016) found that the biological implications of expanding oxygen minimum zones remain grossly understudied in this region, where hypoxic conditions are common in the shallow, warm waters of the continental shelf (Bianchi, 1991; Fiedler & Lavín, 2017). Organisms inhabiting waters with temperature and oxygen conditions closer to their physiological tolerance limits are likely to be more sensitive to further warming and deoxygenation  (Gallardo et al., 74  2019; Hofmann et al., 2011; Wishner et al., 2018). Warming exacerbates the impacts of expanding oxygen minimum zones on marine ectothermic organisms by accelerating metabolic rates. In this way, warming narrows the organism’s aerobic scope (Deutsch et al., 2015; Pauly & Cheung, 2018; Pörtner, et al., 2017) and leaves less oxygen available for basic physiological processes required to sustain a population, such as growth and reproduction. When ocean warming and deoxygenation surpass physiological tolerance limits, many species respond by shifting their distributions towards more oxygenated waters (Deutsch et al., 2020; Gagné et al., 2020). In many cases, oxygen minimum zone expansions drive organisms into shallower, warmer waters, where fishing effort is also higher (Craig, 2012; Prince & Goodyear, 2006).  At the community level, structural shifts occur when the average organism transitions from being an oxy-regulator (oxygen demand is independent on oxygen supply) to an oxy-conformer (oxygen demand depends on supply) (Claireaux & Chabot, 2016). Once oxygen demand depends on the supply, losses in oxygen supply directly impact physiological performance. At the population level, organisms with impaired physiological performance will also have lower abundances until a threshold at which the habitat is no longer metabolically viable (Pörtner et al., 2017). The sensitivity of species assemblages to ocean warming and further deoxygenation in oxygen minimum zones can be assessed using a Biogeographically derived Metabolic Index (BDMI) (Chapter 3). BDMI is essentially a ratio of a species´ oxygen supply to demand that varies according to changes in temperature and oxygen. Oxygen is essential for the long-term survival of a species, so a habitat is considered metabolically viable when environmental oxygen supply is higher than the species’ demand (Chapter 3). A reduction in oxygen supply caused by ocean deoxygenation and/or an increase in oxygen demand caused by warming would lower BDMI values and thus reduce the habitat viability. Similar indices have been developed and applied to study the effects of warming and deoxygenation on a limited number of marine ectothermic species (Deustch et al., 2015). However, the BDMI is applicable to a wider range of species, as input parameter values are available for over 1000 marine ectothermic water breathers of commercial importance. Consequently, the BDMI can be readily applied to study oxygen minimum zone fauna in the Eastern Tropical Pacific for which limited biological data is available.  75  In this study, I aim to examine the relationship between ocean warming, deoxygenation and species composition in the Eastern Tropical Pacific oxygen minimum zone. Specifically, I develop and apply two community-based indices to understand the role of temperature and oxygen in structuring demersal fish and invertebrate communities along the dynamic oxygen minimum zone margins in the Costa Rican Pacific. In the absence of long-term oceanographic and ecological time series to elucidate the impacts of expanding oxygen minimum zones on biological communities, I use episodic oceanographic and biological records collected during different phases of El Nino Southern Oscillation (ENSO). The wide ranges of temperature and oxygen levels that ENSO produces along oxygen minimum zone margins create a ‘natural experiment’ to explore the effects of warming and expanding oxygen minimum zone margins on marine fauna communities (Leung et al., 2019). Based on the results from this ‘natural experiment’, I discuss the implications of projected ocean warming and deoxygenation for marine biodiversity and fisheries associated with oxygen minimum zones.   5.2 Methods 5.2.1 Demersal species community survey  Scientific surveys of the demersal fish and invertebrate community along the Costa Rican Pacific were conducted onboard commercial shrimp trawlers (22.5 m long, with 270 hp engine and two standard epibenthic nets 20.5 m long, mouth opening of 5.35 m x 0.85 m, mesh size, 4.45 cm, cod-end mesh size 3.0 cm). Surveys were conducted in August 2008, May 2009, March 2010, and February 2011. During each survey, 15-minute trawls were deployed to collect samples at a speed of 2.1 - 5.7 km/h. These surveys followed a stratified design, in which at least one tow was conducted at 150 m, 250 m and 350 m every half a degree latitude. For each trawl, I collected the following information: geographic coordinates, depth, duration, species level catch weight and bottom temperature (Fig. 5.1). Organisms caught during each trawl were identified and weighed (total wet weight – TW). Abundance and biomass data were standardized to catch per unit of effort (CPUE) in kilograms per hour (kg × trawl hr-1). Sea bottom temperature (°C) was recorded at each sampling station with an SBE-25 CTD (Sea-Bird Electronics, Inc., Bellevue, WA). I grouped the survey data by 1° latitude, 1° longitude and 10 m depth groups. 76   Figure 5.1. Trawl sample locations along the Pacific coast of Costa Rica, 2008-2011. Sampling was carried out at 145-350 m depths.  In the absence of in situ oxygen measurements, I used Earth system model hindcast from the Centre National de Recherches Météoroglogiques Earth system model version 2,CNRM-ESM2-1 (Séférian et al.,2019), Community Earth system model version 2, CESM2 (Danabasoglu et al., 2020), Version 6 of the Institut Pierre‐Simon Laplace (IPSL) climate model, IPSL-CM6A-LR (Boucher et al., 2020) available from the Coupled Models Intercomparison Project Phase 6 (CMIP6) database . The hindcast simulation covers the time period from January 1948 to December 2009 across 50 depth levels (Griffies et al., 2016; Orr et al., 2017). I transformed the dissolved oxygen concentration (mol/m3) data to a 1° latitude x 1° longitude grid for each depth level between 140 and 350 m. I created monthly oxygen climatologies for normal, EL Niño and La Niña between 1948 and 2009. ENSO state was determined based on the Oceanic Niño Index: La Niña (<-0.5), El Niño (>0.5), Normal (-0.5–0.5), which is available from the NOAA Climate Prediction Center (http://www.cpc.ncep.noaa.gov/products/monitoring_data/). I associated an oxygen climatology value to each trawl based on ENSO phase, month, latitude, longitude and depth level. The trawl 77  surveys in 2008 and 2009 were conducted under a neutral ENSO phase, while during 2010 El Niño conditions were present and in 2011 La Niña conditions were present.  5.2.2 Community-level indicators (Biogeographically derived Metabolic Index of the Catch, Mean Oxygen Demand of the Catch) I derived two community-level indicators from the BDMI, hereafter called Biogeographically derived Metabolic Index of the Catch (BDMC), and Mean Oxygen Demand of the Community (MODC). BDMC  is an indicator of the physiological performance of organisms within the community, while MODC is an indicator of their metabolic traits (see Chapter 3). I used the two indices to identify the oxygen threshold that separates the different responses of the community to changes in oxygen and temperatures.  To calculate the BDMC, I firstly estimated the oxygen demand for each species in each trawl according to the methods presented in Chapter 3. This algorithm estimates oxygen demand of marine fishes and invertebrates based on the von Bertalanffy growth equation and metabolic theory: 𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 = (𝑊1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗2/𝑇 𝑂2,𝑡ℎ𝑟𝑒𝑠ℎ 𝑒−𝑗1/𝑇𝑝𝑟𝑒𝑓𝑊∞1−𝑑 (𝐾 (1−𝑑))⁄  𝑒−𝑗1/𝑇 𝑒−𝑗2/𝑇𝑝𝑟𝑒𝑓) (1) where W is average body weight (g) and d is a scaling coefficient with a value of 0.7 (Pauly & Cheung, 2018). 𝑊∞ is the asymptotic weight (g), estimated from the asymptotic length 𝐿∞ (cm) and the species´ length-weight relationship (Froese and Pauly 2019), while K is the von Bertalanffy K (Froese & Pauly, 2019). An abbreviated form of the Arrhenius equation (Clarke and Johnston 1999), 𝑒−𝑗1/𝑇 and 𝑒−𝑗2/𝑇 accounts for the temperature dependence of anabolism and catabolism, where j1 and j2 represent the activation energy divided by the Boltzmann constant. The coefficients j1 and j2 have values of 4500K and 8000K, respectively (Cheung et al., 2011). T is the temperature in Kelvin measured in situ. The species oxygen threshold O2,thresh (mol/m3) is the 10th percentile of the dissolved oxygen concentrations across a species´ distribution, while the species temperature preference 𝑇𝑝𝑟𝑒𝑓(Kelvin) is the median sea water temperature across a species´ distribution (Cheung et al.,2013a; Cheung et al., 2013b). I estimated species oxygen thresholds and temperature preferences based on the dissolved oxygen concentrations and temperatures for trawls in which species were present. Due to data limitations, I included 95.6% of the total catch biomass in the analysis. 78  I then estimated the Biogeographically derived Metabolic Index (BDMI) as the ratio of oxygen supply to demand (1). Oxygen demand (mol/m3) for each species was computed using equation 1 while oxygen supply is the ambient dissolved oxygen concentration (mol/m3). If oxygen supply exceeds demand, BDMI takes a value of one or above and the habitat is ‘metabolically viable’ (Chapter 3).  I computed BDMC and MODC for each trawl and Earth system model combination. BDMC represents the community averaged BDMI weighted by the catch and was used here as an indicator of the physiological performance of species within the community. I calculated BDMC as:  𝐵𝐷𝑀𝐶̅̅ ̅̅ ̅̅ ̅̅ ?̅? =∑ 𝐵𝐷𝑀𝐼𝑖,𝑇𝐶𝑎𝑡𝑐ℎ𝑖,𝑠𝑛𝑖∑ 𝐶𝑎𝑡𝑐ℎ𝑖,𝑠𝑛𝑖   (2) where 𝐶𝑎𝑡𝑐ℎ𝑖,𝑠 is the weight of the catch of species i in each trawl s, 𝐵𝐷𝑀𝐼𝑖,𝑇 is the oxygen demand of species i at sea water temperature T and n is the total number of species. MODC was calculated as the average oxygen demand (O2, demand from eq. 1) weighted by the catch:  𝑀𝑂𝐷𝐶𝑠 =∑ 𝑂2,𝑑𝑒𝑚𝑎𝑛𝑑 𝑖,𝑇𝑟𝑒𝑓𝐶𝑎𝑡𝑐ℎ𝑖,𝑠𝑛𝑖∑ 𝐶𝑎𝑡𝑐ℎ𝑖,𝑠𝑛𝑖   (3) where 𝐶𝑎𝑡𝑐ℎ𝑖,𝑠 is the catch of species i in each trawl s, 𝑂2, 𝑑𝑒𝑚𝑎𝑛𝑑𝑖,𝑇 is the oxygen demand of species i at a reference sea water temperature T ref and n is the total number of species. T ref was the average sea bottom temperature in all trawls, which was 12.87 °C.  In combination, BDMC and MODC were used to detect the effects of warming and deoxygenation on the demersal community in the Costa Rican oxygen minimum zone. If the community composition remains constant, BDMC will increase with ambient oxygen and decrease with temperature because of increasing metabolic oxygen demand. However, if oxygen becomes limiting, MODC will decline as species with high oxygen demand are excluded from the community and species with lower oxygen demands become more dominant. At this stage, BDMC trends will depend on the balance between declines in community-mean oxygen demand and oxygen supply.   A visual inspection of the data revealed a non-linear relationship between oxygen, temperature and both BDMC and MODC. I therefore conducted segmented regressions to examine the effect of oxygen, temperature and their interaction on these indicators (Muggeo, 2003; Muggeo, 2017). I aggregate tows across years because of the small sample size. The latitudinal and longitudinal range of these tows is very 79  narrow, and the composition of elasmobranchs and bony fish do not change across latitude or longitude within the 150-250 m depth range (Clarke et al., 2015; Clarke et al., in prep). In this chapter, I assume that the changes in temperature and oxygen across different depths and ENSO phases drive the changes in catch composition. The oxygen breaking point then was used to separate the trawl samples into two different communities. I assessed the difference in species richness and biomass across these communities, then used a SIMPER to detect which species caused the largest differences (Clarke et al., 2014). I examined the differences in temperature, oxygen, MODC, BDMC and depth centroids across sampling years with an ANOVA and a post-hoc Tukey test. When necessary, I normalized these variables with a Tukey transformation. Depth centroids for each species i per year (Cheung et al., 2009) were calculated as:  𝐷𝑒𝑝𝑡ℎ 𝑐𝑒𝑛𝑡𝑟𝑜𝑖𝑑 =  𝐷𝑒𝑝𝑡ℎ𝑖∗𝐶𝑎𝑡𝑐ℎ𝑖𝐶𝑎𝑡𝑐ℎ𝑖 (3) Where 𝐷𝑒𝑝𝑡ℎ𝑖  is the depth of the tow i and 𝐶𝑎𝑡𝑐ℎ𝑖 was the catch in weight of species i.  Finally, I evaluated the ability of the Biogeographically derived Metabolic Index (BDMI) to predict the presence and absence of each species in the Costa Rican Pacific region. Firstly, I identified the species that were most abundant in the survey i.e., species that were present in at least 23 tows (25% of the total 91 tows). Then, I calculated the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) by comparing the species’ presence predicted by the BDMI with the presence records from the trawl survey data. This analysis was performed with the pROC package in R (Robin et al., 2011).  5.3 Results Dissolved oxygen concentrations throughout the study area ranged between 0.0023 - 0.0588 mol/m3, while temperatures ranged between 9.6 °C and 15.5 ° C. Spatial variability of the oxygen outputs from each ESM was high, which resulted in a high variability in BDMC and MODC values (Fig. 5.2, Fig. 5.4, Fig. 5.5,  Table 5.1). A wide range of temperatures and dissolved oxygen concentrations were found across the sampling years (Table 5.1, 5.2). Temperatures and oxygen were consistently higher in 2008 through 2010 (neutral and El Niño) than during 2011 (La Niña) (Fig. 5.2). Specifically, temperature was significantly 80  cooler in 2011 than in 2009, while oxygen (CESM2) was lower in 2011 than in any other year (Table 5.2). This corresponded with lower BDMC in 2008 and 2011, and with the decline of MODC during this period. Simultaneously, average depth centroids became significantly shallower in 2011, especially in comparison to 2009 (Fig. 5.2, Table 5.2). BDMC (CESM2) was higher in 2009 and 2010 than 2008 and 2011, while BDMC (IPSL-CM6A-LR) was lower in 2008 than in any other year. On the other hand, MODC (CESM2, IPSL-CM6A-LR) was higher in 2008 than in any other year, and MODC (CNRM-ESM2-1) was higher in 2008 and 2009 than in 2011 (Table 5.2).    Figure 5.2. Temperature (A), dissolved oxygen concentration (B), Biogeographically derived Metabolic Index of the Catch (BDMC) (C), and Mean Oxygen Demand of the Catch (MODC) (D) across sampling years along the Pacific coast of Costa Rica (150 – 350 m). Outputs based on CESM2 (blue), CNRM-ESM2-1 (yellow) and IPSL-CM6A-LR (red). The boxplots represent the median and the lower and upper hinges represent the 25th and 75th percentiles. 81   Figure 5.3. Depth centroids of demersal species along the Pacific coast of Costa Rica (150 – 350 m) from 2008 to 2011.  Table 5.1. Analysis of variances of the differences in temperature, dissolved oxygen concentration (mol/m3), BDMC, MODC (mol/m3) and depth centroids (m) across sampling years (2008-2011). Results for the changes in dissolved oxygen concentration, BDMI and MODC are based on output from three different Earth system models (ESM; CESM2, CNRM-ESM2-1, IPSL-CM6A-LR). ESM Dependent variable  Independent variable d.f. Sum of squares Mean square F value p-value CESM2 Oxygen Year 3 3.5700000 1.1900000 7.959 <0.01   Residuals 87 13.0100000 0.1495000    BDMC Year 3 2.5630000 0.8543000 8.797 <0.01   Residuals 87 8.4490000 0.0971000    MODC Year 3 0.0000100 0.0000033 7.649 <0.01   Residuals 87 0.0000378 0.0000004   CNRM-ESM2-1  Oxygen Year 3 78.3000000 26.1100000 0.987 0.40   Residuals 87 2301.3000000 26.4500000    BDMC Year 3 0.0359000 0.0119650 1.801 0.15   Residuals 87 0.5779000 0.0066420    MODC Year 3 0.0000037 0.0000012 6.968 <0.01   Residuals 87 0.0000153 0.0000002   IPSL-CM6A-LR  Oxygen Year 3 0.0202000 0.0067380 1.420 0.24   Residuals 87 0.4127000 0.0047440    BDMC Year 3 11.4000000 3.8010000 6.008 <0.01   Residuals 87 55.0400000 0.6330000    MODC Year 3 0.0000467 0.0000156 8.263 <0.01   Residuals 87 0.0001638 0.0000019    Temperature Year 3 11.1800000 3.7280000 3.122 0.03   Residuals 87 103.8900000 1.1940000    Depth centroids Year  3 0.0000067 0.0000022 3.047 0.03     Residuals 108 0.0000787 0.0000007     82  Table 5. 2.Tukey tests to identify the years with differences in temperature, dissolved oxygen concentration (mol/m3), BDMC, MODC (mol/m3) and depth centroids (m) of demersal species along the Pacific coast of Costa Rica. I show results for the three different Earth system models (ESM; CESM2, CNRM-ESM2-1, IPSL-CM6A-LR).  ESM Dependent variables Years Difference p-value CESM2 BDMC 2009-2008 0.2871 0.01  BDMC 2010-2008 0.2549 0.10  BDMC 2011-2009 -0.3909 <0.01  BDMC 2011-2010 -0.3587 0.01  Oxygen 2011-2008 -0.3349 0.01  Oxygen 2011-2009 -0.5020 <0.01  Oxygen 2011-2010 -0.4108 0.02  MODC 2009-2008 -0.0006 0.01  MODC 2010-2008 -0.0007 0.02  MODC 2011-2008 -0.0008 <0.01 CNRM-ESM2-1  MODC 2011-2008 -0.0005 <0.01  MODC 2011-2009 -0.0004 0.01 IPSL-CM6A-LR  BDMC 2009-2008 0.7957 <0.01  BDMC 2010-2008 0.9288 0.01  BDMC 2011-2008 0.6243 0.03  MODC 2009-2008 -0.0012 0.01  MODC 2010-2008 -0.0014 0.02  MODC 2011-2008 -0.0019 <0.01  Depth centroid 2011-2009 0.0006 0.05   Temperature 2011-2009 -0.8686 0.02  I identified a threshold oxygen concentration below which BDMC and MODC responded differently to changes in oxygen (p < 0.01). For all ESMs, BDMC increased with oxygen while MODC values remained low and stable when ambient oxygen levels were below 0.013-0.039 mol/m3 (Fig. 5.4, Fig. 5.5). However, when oxygen level was above these values, MODC increased with ambient oxygen, with BDMC either plateauing for CESM2, increasing at a slower rate for CNRM-ESM2-1, or declining for IPSL-CM6A-LR (Fig. 5.4, 5.5). The relationship between temperature and both indicators mirrored their relationship with oxygen as temperature and oxygen co-variated positively in the study area (Fig. 5.4, Fig. 5.5).  83  I subsequently used the 0.013-0.039 mol/m3 oxygen level as a threshold to separate two different faunal communities according to ambient oxygen levels and hypoxia-tolerance of species within the communities (Table 5.3, 5.4). Overall, the average number of species and weight caught per tow was lower in the low oxygen community (Appendix D. 1). Ten out of 51 species accounted for over 50% of the differences in species composition between these communities, with Squilla biformis, Pleuroncodes monodon, Heterocarpus vicarious, Merluccius angustimannus and Cherublemma emmelas being most abundant in the oxygen minimum zone community, while Pontinus sierra, Solenocera agassizi, Physiculus rastrelliger, Baldwinella eos, Engyophrys sanctilaurentii and Peprilus snyderi being more abundant in the higher oxygen community (Table 5.5).  Table 5.3. Break-points and adjusted R squared of the segmented regressions between BDMC and oxygen and temperature, as well as MODC, oxygen and temperature. I present results based on the outputs of each Earth system model (ESM; CESM2, CNRM-ESM2-1, IPSL-CM6A-LR).  Dependent variable ESM Adjusted R2 Break-point St. error  Independent variables p-value BDMC CESM2 0.85 0.013 0.001 Oxygen 0.01    10.413 4.685 Temperature 0.15  CNRM-ESM2-1  0.98 0.039 0.001 Oxygen <0.01    11.924 0.392 Temperature 0.90  IPSL-CM6A-LR  0.67 0.027 0.009 Oxygen <0.01       12.948 0.147 Temperature 0.12 MODC CESM2 0.33 0.011 0.003 Oxygen 0.07    11.877 2.554 Temperature 0.93  CNRM-ESM2-1  0.39 0.038 0.006 Oxygen 0.37    11.881 0.555 Temperature 0.69  IPSL-CM6A-LR  0.38 0.032 0.005 Oxygen 0.02       11.679 0.724 Temperature 0.78    84  Table 5.4. Results of the segmented regression between each indicator (BDMC, MODC) and, temperature, oxygen and their interaction. I conducted segmented regressions based on the outputs of three different Earth system models (ESM; CESM2, CNRM-ESM2-1, IPSL-CM6A-LR). Dependent variables ESM Independent variables Estimate Std. Error t value p-value BDMC CESM2 Intercept -0.5966 3.8401 -0.155 0.88   Temperature – segment 1 0.0402 0.3909 0.103 0.92   Oxygen – segment 1 319.7797 63.7567 5.016 <0.01   Temperature – segment 2 0.0486 0.3932 0.124 0.15   Oxygen – segment 2 -43.9693 14.7808 -2.975 0.01   Temperature:Oxygen -16.9229 5.1732 -3.271 <0.01  CNRM-ESM2-1  Intercept 0.6637 0.1085 6.117 <0.01   Temperature – segment 1 0.0372 0.0087 4.299 <0.01   Oxygen – segment 1 -10.5044 3.9059 -2.689 <0.01   Temperature – segment 2 0.0125 0.0059 2.135 0.9   Oxygen – segment 2 10.4393 0.7804 13.377 <0.01   Temperature:Oxygen -0.5660 0.3036 -1.864 0.07  IPSL-CM6A-LR  Intercept -18.1950 18.2100 -0.999 0.33   Temperature – segment 1 1.6340 1.4840 1.101 0.29   Oxygen – segment 1 2323.1440 1623.1430 1.431 0.17   Temperature – segment 2 10.9780 6.3760 1.722 0.12   Oxygen – segment 2 444.2080 1350.4910 0.329 <0.01     Temperature:Oxygen -153.1310 127.4650 -1.201 0.25 MODC CESM2 Intercept 0.0069 0.0033 2.085 0.04   Temperature – segment 1 -0.0003 0.0003 -1.082 0.28   Oxygen – segment 1 -0.6945 0.3563 -1.949 0.05   Temperature – segment 2 -0.0001 0.0004 -0.330 0.93   Oxygen – segment 2 -0.1151 0.1151 -1.000 0.07   Temperature:Oxygen 0.0613 0.0301 2.035 0.04  CNRM-ESM2-1  Intercept 0.0161 0.0035 4.549 <0.01   Temperature – segment 1 0.0001 0.0003 0.472 0.64   Oxygen – segment 1 0.1150 0.1273 0.903 0.37   Temperature – segment 2 0.0003 0.0002 1.534 0.69   Oxygen – segment 2 0.0318 0.0270 1.180 0.31   Temperature:Oxygen -0.0079 0.0100 -0.796 0.43  IPSL-CM6A-LR  Intercept 0.0006 0.0063 0.100 0.92   Temperature – segment 1 0.0001 0.0006 0.162 0.87 85  Dependent variables ESM Independent variables Estimate Std. Error t value p-value   Oxygen – segment 1 0.2937 0.2256 1.302 0.20   Temperature – segment 2 0.0008 0.0007 1.092 0.78   Oxygen – segment 2 0.1704 0.0600 2.841 0.02     Temperature:Oxygen -0.0251 0.0178 -1.405 0.16     Figure 5.4. Segmented regression between Biogeographically derived Metabolic Index of the Catch (BDMC) and temperature, oxygen and its interaction. Model predictions (line), prediction confidence intervals (ribbon) and data are shown (circles). The model based on the output of three Earth system models, CESM2 is represented in blue (A, B), CNRM-ESM2-1 in yellow (C,D) and IPSL-CM6A-LR in red (E,F).   86      Figure 5.5. The segmented regressions between Mean Oxygen Demand of the Catch (MODC) and temperature, oxygen and its interaction. Model predictions, prediction confidence intervals and data points are included. Models were conducted based on the output of three Earth system models, CESM2 in blue (A, B), CNRM-ESM2-1 in yellow (C, D) and IPSL-CM6A-LR in red (E, F).     87  Table 5.5. Mean abundance of species in the communities in low oxygen and higher oxygen environments along the Pacific coast of Costa Rica. Dissimilarities between communities were measured with similarity percentage (SIMPER) analysis. I show results from the ten species that contribute to most dissimilarities, based on outputs of three different Earth system models. Higher mean abundances are highlighted in bold.  ESM Species Mean Abundance per trawl Cumulative dissimilarity (% contribution) Low oxygen Higher oxygen CESM2 Squilla biformis 0.45 0.20 0.11  Pleuroncodes monodon 0.27 0.19 0.21  Pontinus sierra 0.19 0.29 0.28  Heterocarpus vicarius 0.15 0.08 0.32  Solenocera agassizi 0.07 0.17 0.37  Physiculus rastrelliger 0.12 0.16 0.42  Baldwinella eos 0.06 0.17 0.46  Merluccius angustimanus 0.17 0.11 0.50  Cherublemma emmelas 0.13 0.01 0.54  Peprilus snyderi 0.05 0.11 0.57 CNRM-ESM2-1  Squilla biformis 0.46 0.13 0.11  Pleuroncodes monodon 0.27 0.19 0.21  Pontinus sierra 0.21 0.24 0.26  Baldwinella eos 0.07 0.19 0.30  Physiculus rastrelliger 0.13 0.14 0.35  Solenocera agassizi 0.08 0.14 0.39  Merluccius angustimanus 0.17 0.10 0.43  Heterocarpus vicarius 0.16 0.01 0.47  Engyophrys sanctilaurentii 0.02 0.15 0.51  Cherublemma emmelas 0.13 0.00 0.54 IPSL-CM6A-LR  Squilla biformis 0.47 0.13 0.11  Pleuroncodes monodon 0.27 0.19 0.21  Pontinus sierra 0.21 0.23 0.26  Baldwinella eos 0.06 0.18 0.30  Engyophrys sanctilaurentii 0.01 0.17 0.34  Physiculus rastrelliger 0.13 0.13 0.39  Peprilus snyderi 0.04 0.16 0.43  Heterocarpus vicarius 0.17 0.01 0.47  Merluccius angustimanus 0.17 0.10 0.51  Solenocera agassizii 0.08 0.14 0.55 88   When BDMI was applied to predict the occurrences of species across the sampling locations, the predictions from BDMI were better than random for all species (Appendix D.1). Occurrences predicted with temperature and oxygen outputs from the CESM2 model agreed most with observed occurrences (AUC above 0.7 in 25% of the species) amongst predictions using different ESMs (Appendix D.1).  5.4 Discussion Our results support the hypothesis that changing ambient oxygen and temperature contributed to ENSO-driven shifts in the demersal community of marine fauna along the oxygen minimum zone margins in the Costa Rican Pacific. The main findings supporting this hypothesis are: (1) the faunal community structure, as indicated by BDMC and MODC, are positively correlated with environmental oxygen concentration; (2) species´ depth distributions track the changes in temperature and oxygen closely; (3) the fauna community at the lowest oxygen levels was comprised of species with very low oxygen demands.  The different responses of MODC across an environmental oxygen threshold level (0.013-0.039 mol/m3) suggest the existence of two different ecological communities that are adapted to different environmental conditions. Species inhabiting the oxygen minimum zone in the Costa Rican Pacific are known to have physiological and behavioral adaptations that allow them to maintain very low oxygen demands (0.0066 - 0.0259 mol/m3), granting them access to the high food availability under reduced predation and reduced competition characteristic of these hypoxic areas (Gallo & Levin, 2016). The steep oxygen and temperature gradients along the oxygen minimum zone margins have been known to separate different ecological communities in relatively small geographic areas (Bianchi, 1991; Gallo & Levin, 2016; Wishner et al., 2018). Here, I identified the threshold value of ambient oxygen level of 0.013-0.039 mol/m3 that separated distinct communities. This threshold is below the 0.05 – 0.08 mol/m3 (Hameau et al.,2020) limit used to define oxygen minimum zone; however, ESMs overestimate the spatial extent of oxygen minimum zones in the Eastern Tropical Pacific, and therefore, it is possible the threshold identified in this study is closer to 0.05 – 0.08 mol/m3 (Hameau et al., 2020).  Particularly, the community associated with lower oxygen levels identified in this study is comparable to the one detected by Bianchi (1991) in Nicaragua (300-350m), which also had a low species 89  diversity and was dominated by Pleuroncodes monodon, Merluccius angustimanus and Cherublemma emmelas. While both communities identified here had a much lower species richness than the highly oxygenated coastal waters that most commercial species inhabit (Gallo & Levin, 2016), the lower oxygen community had amongst the lowest species richness in Costa Rican waters. Biomass was higher in the higher oxygen environment along the oxygen minimum zone margin, where the high concentration of food attracts large aggregations of Pleuroncodes monodon and Squilla biformis (Bianchi, 1991; Gallardo et al., 2019).  Species inhabiting the oxygen minimum zone responded quickly to the relatively small ENSO-driven changes in temperature and oxygen by shifting their depth distributions, which may indicate that species are living close to their tolerance limits (Wishner et al., 2018). Species responded to the low temperature and oxygen supply during La Niña by shoaling, while they inhabited deeper waters when temperature and oxygen supplies were higher during normal and El Niño years. These shifts are consistent with the documented responses of fish and invertebrates to short and long term shoaling of oxygen minimum zone, which are shifting species distributions (Gilly et al., 2013; Grantham et al., 2004; Stewart et al., 2014); changing in community compositions (Bertrand et al., 2011; Chu & Gale, 2017; Papiol et al., 2017), declining biodiversity and biomass (Gallo & Levin, 2016). When applying the BDMC and MODC to test a hypothesis, it is important to consider several sources of uncertainty. Particularly, oxygen hindcasts from CMIP6 ESMs cannot fully replicate the spatial and temporal oxygen trends in the tropics, including the strong interannual variation in oxygen across normal, EL Niño and La Niña phases (Cabré et al., 2015). Despite the wide variability in absolute oxygen values, the effect of oxygen on BDMC and MODC was strong and produced similar general trends across models. This effect is strong despite the dampening effect that temperature has on the relationship between oxygen and both indicators. Temperature and oxygen are positively correlated in the study area, so an increase in oxygen that leads to higher BDMC and MODC occurs at the same time as an increase in temperature that leads to lower BDMC and MODC. Given the uncertainties in the environmental data and the limited geographic and temporal range of the data, it is possible that the existence of thresholds or breaks in the data is artificial. Future studies should explore the universality of tipping points.  Yet, the 90  relatively high prediction ability increases our confidence in the results even further. All AUC values were above 0.5, which is high considering the fact that ecological interactions and fishing impacts tend to be more important than the environment in shaping species distributions at such fine spatial resolutions (Peterson et al., 2011).  I can draw on the community response to the inter-annual variability in temperature and oxygen supply along the oxygen minimum zone margins during different ENSO phases to infer possible consequences of the projected long term warming and oxygen minimum zone expansion. I expect species to continue shifting as they track environmental oxygen and temperature envelopes. Oxygen minimum zone expansions may compress the habitat of hypoxia tolerant species at the deep end of their depth distribution range, while warming may compress their habitat at the shallower end (Deutsch et al., 2015). Expanding oxygen minimum zones can also be compressing the habitat of hypoxia sensitive species towards the coast (Rose et al., 2019). Upper boundaries of the oxygen minimum zone along Central America already reach 50 m below the surface in some places, so any additional shoaling is bound to affect shallow water species.  The impacts of oxygen minimum zone expansions on demersal faunal abundance and community structure are likely to reduce fishery production. Fisheries inside oxygen minimum zones are generally not very productive, as hypoxia reduces trophic efficiency (Bertrand et al., 2011; Rose et al., 2019). The Costa Rican shrimp-trawl fishery operating within hypoxic waters was not sustainable, the landings of both target species (Heterocarpus vicarious and Solenocera agassizi) declined substantially between the fishery’s beginning in the 1980s and the mid 2000´s (Wehrtmann & Nielsen Munoz, 2009). The prospect of developing a new fishery based on another target species in the oxygen minimum zone is low because the more palatable texture created by muscle has high oxygen demands that cannot be met in hypoxic environments (Gallo & Levin, 2016). The key role that oxygen and temperature have played in structuring the demersal faunal community in the Costa Rican Pacific region suggests that such faunal community ocean is vulnerable to deoxygenation and warming under climate change. Using BDMC and MODC, I was able to show how oxygen supply limits the physiological performance of the average organism within the demersal community along the oxygen minimum zone margins in Costa Rica. Oxygen minimum zone expansions have the 91  potential to affect marine biodiversity and marine food production along the Pacific of Costa Rica and potentially the region. Although climate-mitigation is essential to minimize oxygen minimum zone expansion and the scale of potential impacts, the scope for adaptation will depend on a better understanding of the rate and magnitude at which it is occurring, as well as its ecological impacts.   92  Chapter 6: General discussion and conclusions  6.1 Synthesis of the main findings In this thesis, I aimed to better understand the impacts of climate change on species impacted by fisheries in the Eastern Tropical Pacific Ocean, a region that extends from Baja California, Mexico to northern Peru. I modeled the impacts of climate change on the biogeography and ecophysiology of exploited species, in order to assess the risk of climate change on fisheries. The results support the need for regional to local-scale modeling to inform climate change adaptation strategies. In Chapter 2, I projected the impacts of climate change on species caught (targeted, discarded or as bycatch) by the four main fisheries (small-scale, large pelagics, small pelagics and shrimp trawl) in the Eastern Tropical Pacific Ocean using an ensemble of species distribution models. Results showed that by the mid 21st century, environmental conditions will become less suitable for 31-63% of species in each Exclusive Economic Zone. Species are projected to shift towards the equator, seeking the more favorable, cooler habitats that are driven by upwelling. Simultaneously, habitat suitability was projected to increase or remain the same in the northern and southern margins of the Eastern Tropical Pacific but decreased by up to 14% along Central America. Such regional changes in habitat suitability reflect the direction of projected range shifts to areas where environmental conditions would become more suitable for the species under climate change. Findings also showed that the expansion of oxygen minimum zones will force species into shallower, inshore waters. Small pelagic fisheries and small-scale fisheries were the biggest losers. The habitat suitability of key small-pelagic species declines up to -45.7%, and species turnover of key species for small scale fisheries decline up to 80%. Although shrimp habitat suitability is projected to increase or remain the same, species caught as bycatch in shrimp trawl fisheries are projected to undergo large losses in habitat suitability of up to ~ 14%. Chapter 3 revealed the key role of warming and deoxygenation in shaping future marine fisheries resource availability in the Eastern Tropical Pacific Ocean. These environmental drivers are among the main factors affecting the future sustainability of fish stocks and fisheries (Koenigstein et al., 2016), yet, available projections of climate impacts seldom incorporate the mechanism linking them together. In 93  Chapter 3, I aim to present a new method to further our understanding of how warming and deoxygenation may affect marine organisms. I developed a Biogeographically derived Metabolic Index (BDMI) to determine if a habitat will be viable or not by comparing ambient oxygen supply to demand. In contrast to previous metabolic indices that are dependent on experimentally-derived estimates of temperature-dependent oxygen demand, the BDMI is based on oxygen demand estimates computed from growth parameters and species’ biogeography that are available from open access databases for thousands of marine fishes and invertebrates. I assessed the performance of the BDMI by comparing it with a metabolic index developed by Deutsch et al. (2015) and Penn et al. (2018) that require parameters estimated from physiological experiments. The comparison for three fish and invertebrate species revealed that both metabolic indices were significantly and positively correlated and projected similar magnitudes of habitat loss. Our results support the potential of the BDMI to better understand the combined effects of ocean warming and deoxygenation on a wide range of marine species in the Eastern Tropical Pacific Ocean and elsewhere.  Previous research has detected a signal of warming on global marine fisheries (Cheung et al., 2013b), yet, the role that deoxygenation plays in shaping fishery catches has not been explored. I bridge this gap in Chapters 4 and 5, by examining the sensitivity of species to warming and deoxygenation with two catch-based indicators based on the BDMI - Biogeographically derived Metabolic Index of the Catch (BDMC) and Mean Oxygen Demand of the Catch (MODC). In Chapter 4, I applied these indices to time series data of pelagic fisheries in the Eastern Tropical Pacific Ocean between 1970 and 2009. Temperature was the main factor driving oxygen limitation in pelagic catches. Findings also show that El Niño Southern Oscillation (ENSO)-driven temperature variability caused the observed temporal BDMI and MODC patterns during the studied period. I identified temperature and oxygen thresholds of 25.20 °C and 0.21 oxygen mol/m3, respectively, that separated catch composition according to species’ sensitivity to warming and deoxygenation. Below the oxygen threshold, catch composition was stable and dominated by species with low oxygen demand, while the metabolic index declined as warming increased species’ oxygen demand. Above the oxygen threshold, warming caused pelagic catches to become increasingly dominated by species with low oxygen demand. The oxygen threshold may represent a tipping point 94  where catch composition transitions from an assemblage that includes temperate and subtropical species with relatively high oxygen demands, to an assemblage that is dominated by tropical species with relatively low oxygen demands.  In Chapter 5, I used the two indicators applied in Chapter 4 (BDMC, MODC) to examine the effect of temperature and oxygen on species composition in a demersal community in the oxygen minimum zone off the Costa Rican Pacific coast. I used data obtained from a scientific sampling campaign using shrimp trawl vessels during 2008-2011. The wide ENSO-driven temperature and oxygen fluctuations produced along the margins of the oxygen minimum zone provided a ‘natural experiment’ to explore the effects of warming and hypoxia on demersal fish and invertebrate communities. My results show that oxygen was the main factor constraining organisms’ physiological performance and shaping catch composition in this community, although species distributions were sensitive to changes in both temperature and oxygen. A 0.01-0.04 mol/m3 oxygen threshold separated the marine fauna into two distinct communities, according to their hypoxia tolerance. During El Niño years when the Eastern Tropical Pacific Ocean was warmer and more oxygenated, average oxygen demand by the faunal community was higher and average species depth centroids were deeper relative to the cooler, less oxygenated conditions during La Niña.   6.2 Key uncertainties The BDMI is based on the Oxygen and Capacity-limited Thermal Tolerance (OCLTT; Pörtner et al., 2017) and the Gill Oxygen Limitation Theory (GOLT; Pauly & Cheung, 2018) that have contributed to a debate in the scientific literature that has at times been polarized. A particular source of contention around the GOLT and OCLTT are the mechanisms driving the decline in physiological performance when temperatures increase above tolerance levels (Jutfelt et al., 2018; Pörtner et al., 2017). The role of oxygen supply in limiting fish growth explained in the GOLT has also been debated (Lefevre et al., 2017; Marshall & White, 2019). However, these critiques are based on individual physiological experiments that fail to account for evolutionary and ecological considerations in both the experimental design and interpretation of the results (Pauly & Cheung, 2018; Pörtner & Giomi, 2013). For example, oxygen 95  limitation occurs in fish that are closer to their maximum sizes, yet much smaller fish are generally used in laboratory experiments (Leiva et al., 2019). Future approaches that integrate knowledge gleaned from the fields of evolution, ecology, physiology and climate change may provide a unifying theory (Jutfelt et al., 2020; Leiva et al., 2019) that would advance the assessment of the effects of ocean warming and deoxygenation on marine species and fisheries. All modeling work conducted in this thesis depended on the output of Earth system models, which have three main sources of uncertainty relevant to this thesis: 1) Earth system models represent open ocean conditions well, but are not able to replicate the variability and extremes that often occur in coastal areas. In addition, they do not account for the hypoxic events that occur in coastal areas due to eutrophication and warming (Breitburg et al., 2018). Therefore, I may not be able to fully capture the total impacts of climate change on species in coastal environments. 2) Earth system models underestimate the rate of deoxygenation in tropical oceans but overestimate the spatial extent of oxygen minimum zones (Cabre et al., 2015; Oschlies et al., 2018). Uncertainty in oxygen projection for the tropics results from two opposing trends: warming lowers oxygen solubility and therefore, dissolved oxygen, at the same time that it reduces biological oxygen consumption (Bopp et al., 2017). 3) Furthermore, upwelling is one of the most important processes driving temperature and oxygen trends in the Eastern Tropical Pacific Ocean, and yet, there is a high degree of uncertainty in how they might change in the future.  I do not consider the impacts of additional climate drivers that have the potential to limit organisms’ aerobic scope, such as ocean acidification and declining primary productivity. The Eastern Tropical Pacific has the lowest surface pH levels in the world, as the hypoxic waters below the thermocline have very low pH and are transported to the surface by the multiple upwelling systems (Fiedler & Lavín, 2017). Food availability may also decline throughout the region, as the warming-induced stratification of surface waters limit nutrient transport to the mixed layer and supresses primary productivity (Fiedler & Lavín, 2017). Additional anthropogenic factors such as overfishing, habitat degradation and pollution also have the potential to cause physiological stress and limit an organisms’ aerobic scope. 96  Another important source of uncertainty derives from the species-specific application of the BDMI that does not consider the impacts of deoxygenation on trophic interactions. Yet, the physiological stress caused by deoxygenation affects the behaviour, feeding and movement patterns that structure many of the current ecological interactions among species (Breitburg et al., 2018). Fishing can also shape community composition by increasing the proportion of small, fast growing species in the catch. Future studies should explore the role of fishing effort in driving trends in BDMC and MODC, by including it as a factor in modeling and analyzing fishery independent information.  6.3 Future research areas The BDMI has many potential applications to further our understanding of warming and deoxygenation on marine biodiversity and fisheries. Current studies on the impacts of warming and deoxygenation on marine biodiversity and fisheries largely focus on the horizontal dimension of the ocean. Analyses of two-dimensional (horizontal) habitat loss require data that are more readily available and provide useful approximations of changes in the geographic distributions of fisheries resources. Yet, the BDMI can be extended to estimate three-dimensional habitat loss. Although the assessment of three-dimensional habitat loss requires three-dimensional (longitude, latitude, depth) oxygen and temperature data currently not widely available, the approach can estimate habitat compression and vertically resolved changes in catchability (Schirripa et al., 2017). By accounting for the increasing catchability of commercial species, fisheries management may avoid resource depletion (Craig, 2012; Farchadi et al., 2019; Gallo & Levin, 2016; Prince et al., 2010). A three-dimensional application of the BDMI could also improve modeling climate impacts on meso-pelagic fish (Brito-Morales et al., 2020), which are key diet items of current fishery resources and may soon become key fishery species due to their high biomass (Hidalgo & Browman, 2019).  The BDMI provides an opportunity to explore the impacts of historic and projected warming deoxygenation on global fisheries and living marine resources. By including the mechanisms driving the historic changes of warming on fisheries as detected by Cheung et al. (2013b), the BDMI can predict future impacts. For example, Cheung et al. (2013b) showed that MTC will reach a plateau in tropical 97  EEZs as only species with high temperature preferences remain in the catch. The BDMI can estimate the increase in temperature and decline in oxygen required for fisheries to reach this plateau. In addition to uncovering the impacts of long-term mean conditions, the BDMI could improve predictions of the impacts of El Niño, by providing the mechanism driving warming impacts in both well oxygenated pelagic environments, and demersal environments. The BDMI may also be used to identify tipping points that separate assemblages with different responses to warming and deoxygenation. In Chapter 4 and 5, I detected a potential threshold that separates the more stable species compositions at the cool and warm extremes of the pelagic environment in the Eastern Tropical Pacific Ocean. Pelagic catches in EEZs close to a tipping point had a highly variable trait composition after El Niño, presumably as the ecosystem state fluctuated around equilibrium (Dakos et al., 2019; Guerin et al., 2013). Warming and deoxygenation will lead to a dominance of species/populations with low oxygen demands and low metabolic indices. These populations have smaller sizes and lower reproductive output, which increase their sensitivity to additional environmental stress and the probability of reaching the tipping point (Dakos et al., 2019). A global application of the BDMC and MODC could strengthen our understanding of this method´s potential to detect tipping points (van Nes et al., 2016) and whether they change across different oceanographic biogeographies.  Identifying potential tipping points is important because crossing them may be irreversible, increase the cost of inaction and redistribute ecosystem benefits and burdens (Selkoe et al., 2015). Options to avoid some of the negative consequences of reaching a tipping point include : a) a reduction in local pressures (Halpern, 2017; Rilov et al., 2019; Selkoe et al., 2015), b), implementation of precautionary management measures that tie management targets to ecosystem thresholds, c) monitoring of tipping point indicators such as the BDMI and MODC and d) an increase in monitoring and interventions as the risk of tipping points increase (Selkoe et al., 2015).  Future impacts of warming and deoxygenation on habitat loss, fisheries and ecosystem state may be buffered or exacerbated through ecological interactions. Warming and deoxygenation triggers behavioral adaptations that can modify species interactions, such as avoidance, reduced activity or lower 98  feeding rates (Breitburg et al., 2018). This will intensify with the local extirpation of key species that could trigger trophic cascades. At the same time local invasions could have destabilizing effects on ecosystem structure and function (DeRoy et al., 2020). Therefore, it is important to explore the role of oxygen limitation at the ecosystem level, using BDMI as a driver.  6.4 Recommendations for climate-smart fisheries in the Eastern Tropical Pacific Ocean Warming and deoxygenation limit the aerobic scope of marine ectothermic organisms in the Eastern Tropical Pacific, with direct impacts on physiological performance and catch composition. Additional anthropogenic and climatic stressors can limit an organism´s aerobic scope both directly, by causing additional physiological stress (e.g. lack of food and pollution) and indirectly, by increasing the magnitude of warming and deoxygenation (e.g. eutrophication and loss of coastal vegetation). Therefore, reducing these human stressors on fisheries resources may compensate for some of the catch lost due to climate change (Cheung et al., 2018; Gaines et al., 2018; Galbraith et al., 2017; Sumaila & Tai, 2020). Common stressors throughout the Eastern Tropical Pacific include fisheries habitat degradation, pollution, run-off from urban and agricultural lands (Lluch-Cota et al., 2019). Overfishing is the main stressor affecting fisheries resources in the Eastern Tropical Pacific (Lluch-Cota et al., 2019). Overfished or depleted populations have been shown to be more vulnerable to warming (Free et al., 2019). Fisheries eliminate individuals with high performance aerobic scopes that invest more of their aerobic capacity in growth instead of the skittish behaviors that would allow them to avoid interactions with fishing gear (Duncan et al., 2019). Oxygen limitation also selects for smaller sizes due to geometric constraints in the gill’s ability to supply growing fish bodies with enough oxygen (Daniel Pauly & Froese, 2020). Thus, fishing and oxygen limitation both select for smaller body sizes and faster growth rates. To counteract this selection process, fisheries should increase the size at first capture, preserve the larger individuals (Pauly & Froese, 2020) and reduce fishing effort in proportion to the climate-induced decreases in carrying capacity (Arreguín-Sánchez et al., 2015; Cheung et al., 2018). In cases where estimates of carrying capacity are not available, decreases in habitat suitability such as those presented in Chapter 2 may be used as proxies (Wabnitz et al., 2018). This would increase body 99  size, allow populations to recover and increase their resilience to climate change (Diaz Pauli et al., 2017; Pauly & Froese, 2020). Reducing the stress of fisheries could increase the metabolic index and size of species, increasing reproductive output and staving off oncoming tipping points brought on by warming and deoxygenation (Dakos et al., 2019). Yet, the weak fisheries governance, monitoring and lack of enforcement may hinder climate adaptation efforts in the region (Defeo & Castilla, 2012). Initiatives to improve fisheries management in the region can use ENSO as a tool to better understand the impacts of warming and deoxygenation on fisheries. Through its impacts on temperature and oxygen, ENSO variability affects respiratory demand in the Eastern Tropical Pacific (Bindoff et al., 2019; Ito et al., 2020), affecting fisheries for small pelagics, shrimp (Bertrand et al., 2011; Castro-Ortiz & Lluch-Belda, 2008), large pelagics such as tuna and billfish as well as demersal coastal resources (Arntz et al., 2006; Leung et al., 2019). Forecasting tools are essential to prepare and adapt fisheries for the increasingly strong El Niño events (Yang et al., 2018; Yu & Kao, 2007), yet, to build these tools, a better understanding of the mechanisms driving the impacts is required. The BDMI can help uncover these mechanisms and facilitate the transition into the realm of short-term predictions if the spatio-temporal coverage of oceanographic, biodiversity and fisheries monitoring programs is improved (Lluch-Cota et al., 2019; Webb et al., 2013). Management frameworks, in turn, should have the flexibility and regional scope to respond to these predictions (Mills et al., 2013).  Projections of the shifts in species distributions and resulting impacts on catches can inform regional initiatives to manage fisheries to be climate-smart, support resilience as well as be proactive in developing adaptive opportunities and reducing potential conflicts that may arise when commercial species begin to shift across borders (Gullestad et al., 2020; Oremus et al., 2020; Pinsky et al., 2018; Pinsky & Fogarty, 2012). They may also support the development of incentives to cooperate among Regional Fisheries Management Organisations, despite shifting opportunities that arise from climate-driven changes (Miller, 2007).  The results of this thesis suggest that fisheries have already been impacted by warming and deoxygenation. The severity of projected impacts and levels of committed warming show that adaptation and mitigation are essential to preserve the benefits that fisheries provide to society in terms of 100  livelihoods, food and nutrition, as well as income (Bindoff et al., 2019). Although many EEZ in the region are taking an active role in mitigating climate change (e.g. Costa Rica’s plan to become carbon neutral by 2050), they only account for a small percentage of global emissions, so action by the main historic and current CO2 emitters is necessary to reach internationally agreed to targets (Janssens-Maenhout et al., 2019). 101  Bibliography Allison, E. H., Perry, A. L., Badjeck, M.-C., Adger, W. N., Brown, K., Conway, D., Halls, A. S., Pilling, G. M., Reynolds, J. D., Andrew, N. L., & Dulvy, N. K. (2009). Vulnerability of national economies to the impacts of climate change on fisheries. Fish and Fisheries, 10(2), 173–196. https://doi.org/10.1111/j.1467-2979.2008.00310.x Angulo, A., López, M., Bussing, W., Rosa, A., 2016. Colección ictiológica del Museo de Zoología de la Universidad de Costa Rica, in: Del Moral-Flores, L.F., Ramírez-Villalobos, Á.J., Martínez-Pérez, J.A., González-Acosta, A.F., Franco-López, J. (Eds.), Colecciones ictiológicas de Latinoamérica. Facultad de Estudios Superiores Iztacala, Universidad Nacional Autónoma de México & Sociedad Mexicana de Ictiología, Mexico, pp. 56–65. Arafeh-Dalmau, N., Brito-Morales, I., Schoeman, D. S., Possingham, H. P., Klein, C. J., & Richardson, A. J. (2020). Incorporating climate velocity into the design of climate-smart networks of protected areas [Preprint]. Ecology. https://doi.org/10.1101/2020.06.08.139519 Arntz, W. E., Gallardo, V. A., Gutiérrez, D., Isla, E., Levin, L. A., Mendo, J., Neira, C., Rowe, G. T., Tarazona, J., & Wolff, M. (2006). El Niño and similar perturbation effects on the benthos of the Humboldt, California, and Benguela Current upwelling ecosystems. Advances in Geosciences, 6, 243–265. Arreguín-Sánchez, F., Monte-Luna, P. del, & Zetina-Rejón, M. J. (2015). Climate Change Effects on Aquatic Ecosystems and the Challenge for Fishery Management: Pink Shrimp of the Southern Gulf of Mexico. Fisheries, 40(1), 15–19. https://doi.org/10.1080/03632415.2015.988075 Bakun, A., Black, B. A., Bograd, S. J., García-Reyes, M., Miller, A. J., Rykaczewski, R. R., & Sydeman, W. J. (2015). Anticipated Effects of Climate Change on Coastal Upwelling Ecosystems. Current Climate Change Reports, 1(2), 85–93. https://doi.org/10.1007/s40641-015-0008-4 Bakun, Andrew. (1990). Global Climate Change and Intensification of Coastal Ocean Upwelling. Science, 247(4939), 198–201. https://doi.org/10.1126/science.247.4939.198 Barber, R. T., & Chavez, F. P. (1983). Biological Consequences of El Nino. Science, 222(4629), 1203–1210. https://doi.org/10.1126/science.222.4629.1203 102  Basille, M., Calenge, C., Marboutin, É., Andersen, R., & Gaillard, J.-M. (2008). Assessing habitat selection using multivariate statistics: Some refinements of the ecological-niche factor analysis. Ecological Modelling, 211(1–2), 233–240. https://doi.org/10.1016/j.ecolmodel.2007.09.006 Bates, A. E., Barrett, N. S., Stuart-Smith, R. D., Holbrook, N. J., Thompson, P. A., & Edgar, G. J. (2014). Resilience and signatures of tropicalization in protected reef fish communities. Nature Climate Change, 4(1), 62–67. https://doi.org/10.1038/nclimate2062 Baudron, A. R., Needle, C. L., Rijnsdorp, A. D., & Marshall, C. T. (2014). Warming temperatures and smaller body sizes: Synchronous changes in growth of North Sea fishes. Global Change Biology, 20(4), 1023–1031. https://doi.org/10.1111/gcb.12514 Bertrand, A., Chaigneau, A., Peraltilla, S., Ledesma, J., Graco, M., Monetti, F., & Chavez, F. P. (2011). Oxygen: A Fundamental Property Regulating Pelagic Ecosystem Structure in the Coastal Southeastern Tropical Pacific. PLOS ONE, 6(12), e29558. https://doi.org/10.1371/journal.pone.0029558 Bianchi, C. N., & Morri, C. (2003). Global sea warming and “tropicalization” of the Mediterranean Sea: Biogeographic and ecological aspects. Biogeographia – The Journal of Integrative Biogeography, 24(1). https://doi.org/10.21426/B6110129 Bianchi, G. (1991). Demersal assemblages of the continental shelf and slope edge between the Gulf of Tehuantepec (Mexico) and the Gulf of Papagayo (Costa Rica). Marine Ecology Progress Series, 73, 121–140. https://doi.org/10.3354/meps073121 Bindoff, N. L., Cheung, W. W. L., Kairo, J. G., Arístegui, J., Guinder, V. A., Hallberg, R., Hilmi, N., Jiao, N., O’Donoghue, S., Suga, T., Acar, S., Alava, J. J., Allison, E., Arbic, B., Bambridge, T., Philip W Boyd, Bruggeman, J., Butenschön, M., Chávez, F. P., … Whalen, C. (2019). Changing Ocean, Marine Ecosystems, and Dependent Communities. In H.-O. Pörtner, D. C. Roberts, V. Masson-Delmotte, P. Zhai, M. Tignor, E. Poloczanska, K. Mintenbeck, A. Alegria, M. Nicolai, A. Okem, J. Petzold, B. Rama, & N. M. Weyer (Eds.), IPCC Special Report on the Ocean and Cryosphere in a Changing Climate (p. 142). 103  Blasiak, R., Spijkers, J., Tokunaga, K., Pittman, J., Yagi, N., & Österblom, H. (2017). Climate change and marine fisheries: Least developed countries top global index of vulnerability. PLOS ONE, 12(6), e0179632. https://doi.org/10.1371/journal.pone.0179632 Bopp, L., Resplandy, L., Orr, J. C., Doney, S. C., Dunne, J. P., Gehlen, M., Halloran, P., Heinze, C., Ilyina, T., Séférian, R., Tjiputra, J., & Vichi, M. (2013). Multiple stressors of ocean ecosystems in the 21st century: Projections with CMIP5 models. Biogeosciences, 10, 6225–6245. https://doi.org/10.5194/bg-10-6225-2013 Bopp, L., Resplandy, L., Untersee, A., Le Mezo, P., & Kageyama, M. (2017). Ocean (de)oxygenation from the Last Glacial Maximum to the twenty-first century: Insights from Earth System models. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 375(2102), 20160323. https://doi.org/10.1098/rsta.2016.0323 Boucher, O., Servonnat, J., Albright, A. L., Aumont, O., Balkanski, Y., Bastrikov, V., Bekki, S., Bonnet, R., Bony, S., Bopp, L., Braconnot, P., Brockmann, P., Cadule, P., Caubel, A., Cheruy, F., Codron, F., Cozic, A., Cugnet, D., D’Andrea, F., … Vuichard, N. (2020). Presentation and evaluation of the IPSL-CM6A-LR climate model. Journal of Advances in Modeling Earth Systems, 12(7), e2019MS002010. https://doi.org/10.1029/2019MS002010 Breitburg, D. L., Baumann, H., Sokolova, I. M., & Frieder, C. A. (2019). Chapter 6. Multiple stressors – forces that combine to worsen deoxygenation. In D. Laffoley & J. M. Baxter (Eds.), Deoxygenation: Everyone’s problem—Causes, impacts, consequences and solutions (pp. 225–246). IUCN. https://doi.org/10.2305/IUCN.CH.2019.13.en Breitburg, D., Levin, L. A., Oschlies, A., Grégoire, M., Chavez, F. P., Conley, D. J., Garçon, V., Gilbert, D., Gutiérrez, D., Isensee, K., Jacinto, G. S., Limburg, K. E., Montes, I., Naqvi, S. W. A., Pitcher, G. C., Rabalais, N. N., Roman, M. R., Rose, K. A., Seibel, B. A., … Zhang, J. (2018). Declining oxygen in the global ocean and coastal waters. Science, 359(6371), eaam7240. https://doi.org/10.1126/science.aam7240 Brito-Morales, I., Schoeman, D. S., Molinos, J. G., Burrows, M. T., Klein, C. J., Arafeh-Dalmau, N., Kaschner, K., Garilao, C., Kesner-Reyes, K., & Richardson, A. J. (2020). Climate velocity reveals 104  increasing exposure of deep-ocean biodiversity to future warming. Nature Climate Change, 10(6), 576–581. https://doi.org/10.1038/s41558-020-0773-5 Burger, F. A., John, J. G., & Frölicher, T. L. (2020). Increase in ocean acidity variability and extremes under increasing atmospheric CO2. Biogeosciences, 17(18), 4633–4662. https://doi.org/10.5194/bg-17-4633-2020 Cabré, A., Marinov, I., Bernardello, R., & Bianchi, D. (2015). Oxygen minimum zones in the tropical Pacific across CMIP5 models: Mean state differences and climate change trends. Biogeosciences, 12(18), 5429–5454. https://doi.org/10.5194/bg-12-5429-2015 Calenge, C. (2006). The package “adehabitat” for the R software: A tool for the analysis of space and habitat use by animals. Ecological Modelling, 197(3), 516–519. https://doi.org/10.1016/j.ecolmodel.2006.03.017 Calosi, P., De Wit, P., Thor, P., & Dupont, S. (2016). Will life find a way? Evolution of marine species under global change. Evolutionary Applications, 9(9), 1035–1042. https://doi.org/10.1111/eva.12418 Carlisle, A. B., Kochevar, R. E., Arostegui, M. C., Ganong, J. E., Castleton, M., Schratwieser, J., & Block, B. A. (2017). Influence of temperature and oxygen on the distribution of blue marlin (Makaira nigricans) in the Central Pacific. Fisheries Oceanography, 26(1), 34–48. https://doi.org/10.1111/fog.12183 Carozza, D. A., Bianchi, D., & Galbraith, E. D. (2019). Metabolic impacts of climate change on marine ecosystems: Implications for fish communities and fisheries. Global Ecology and Biogeography, 28(2), 158–169. https://doi.org/10.1111/geb.12832 Castro-Ortiz, J. L., & Lluch-Belda, D. (2008). Impacts of interannual environmental variation on the shrimp fishery off the Gulf of California (CalCOFI Rep No. 49; pp. 183–190). Cheng, B. S., Bible, J. M., Chang, A. L., Ferner, M. C., Wasson, K., Zabin, C. J., Latta, M., Deck, A., Todgham, A. E., & Grosholz, E. D. (2015). Testing local and global stressor impacts on a coastal foundation species using an ecologically realistic framework. Global Change Biology, 21(7), 2488–2499. https://doi.org/10.1111/gcb.12895 105  Cheung, W. W. L. (2018). The future of fishes and fisheries in the changing oceans. Journal of Fish Biology, 92(3), 790–803. https://doi.org/10.1111/jfb.13558 Cheung, W. W. L., Bruggeman, J., & Butenschön, M. (2018). Chapter 4: Projected changes in global and national potential marine fisheries catch under climate change scenarios in the twenty-first century. Impacts of Climate Change on Fisheries and Aquaculture: Synthesis of Current Knowledge, Adaptation and Mitigation Options. Ed. by M. Barange, T. Bahri, MCM Beveridge et al. FAO Fisheries and Aquaculture Technical Paper, 627. Cheung, W. W. L., Dunne, J., Sarmiento, J. L., & Pauly, D. (2011). Integrating ecophysiology and plankton dynamics into projected maximum fisheries catch potential under climate change in the Northeast Atlantic. ICES Journal of Marine Science, 68(6), 1008–1018. https://doi.org/10.1093/icesjms/fsr012 Cheung, W. W. L., Jones, M. C., Reygondeau, G., & Frölicher, T. L. (2018). Opportunities for climate-risk reduction through effective fisheries management. Global Change Biology, 24(11), 5149–5163. https://doi.org/10.1111/gcb.14390 Cheung, W. W. L., Jones, M. C., Reygondeau, G., Stock, C. A., Lam, V. W. Y., & Frölicher, T. L. (2016). Structural uncertainty in projecting global fisheries catches under climate change. Ecological Modelling, 325, 57–66. https://doi.org/10.1016/j.ecolmodel.2015.12.018 Cheung, W. W. L., Lam, V. W. Y., & Pauly, D. (2008). Dynamic Bioclimate Envelope Models to predict climate-induced changes in distribution of marine fishes and invertebrate (Fisheries Center Research Reports No. 16; pp. 5–50). http://legacy.seaaroundus.s3.amazonaws.com/doc/Researcher+Publications/dpauly/PDF/2008/Books%26Chapters/DynamicBioclimateEnvelopeModelToPredictClimateInducedChangesInDistributionsofMarineFishes%26Invetebrates.pdf Cheung, W. W. L., Lam, V. W. Y., Sarmiento, J. L., Kearney, K., Watson, R., & Pauly, D. (2009). Projecting global marine biodiversity impacts under climate change scenarios. Fish and Fisheries, 10(3), 235–251. https://doi.org/10.1111/j.1467-2979.2008.00315.x 106  Cheung, W. W. L., Lam, V. W. Y., Sarmiento, J. L., Kearney, K., Watson, R., Zeller, D., & Pauly, D. (2010). Large-scale redistribution of maximum fisheries catch potential in the global ocean under climate change. Global Change Biology, 16(1), 24–35. https://doi.org/10.1111/j.1365-2486.2009.01995.x Cheung, W. W. L., & Pauly, D. (2016). Impacts and effects of ocean warming on marine fishes. In D. Laffoley & J. M. Baxter (Eds.), Explaining Ocean Warming: Causes, scale, effects and consequences. IUCN, International Union for Conservation of Nature. https://doi.org/10.2305/IUCN.CH.2016.08.en Cheung, W. W. L., Sarmiento, J. L., Dunne, J., Frölicher, T. L., Lam, V. W. Y., Palomares, M. L. D., Watson, R., & Pauly, D. (2013). Shrinking of fishes exacerbates impacts of global ocean changes on marine ecosystems. Nature Climate Change, 3(3), 254–258. https://doi.org/10.1038/nclimate1691 Cheung, W. W. L., Watson, R., & Pauly, D. (2013). Signature of ocean warming in global fisheries catch. Nature, 497(7449), 365–368. https://doi.org/10.1038/nature12156 Childress, J. J., & Seibel, B. A. (1998). Life at stable low oxygen levels: Adaptations of animals to oceanic oxygen minimum layers. Journal of Experimental Biology, 201(8), 1223–1232. Chu, J. W. F., & Gale, K. S. P. (2017). Ecophysiological limits to aerobic metabolism in hypoxia determine epibenthic distributions and energy sequestration in the northeast Pacific ocean. Limnology and Oceanography, 62(1), 59–74. https://doi.org/10.1002/lno.10370 Cisneros-Montemayor, A. M., & Clarke, T. M. (2019). Exploración de los Posibles Impactos de las Reglas de la OMC Sobre Subsidios a la Pesca: El Caso del Camarón en la Costa Oeste de América Latina (p. 85). International Institute for Sustainable Development. https://www.iisd.org/library/subsidios-pesca-camaron-america-latina Claireaux, G., & Chabot, D. (2016). Responses by fishes to environmental hypoxia: Integration through Fry’s concept of aerobic metabolic scope. Journal of Fish Biology, 88(1), 232–251. https://doi.org/10.1111/jfb.12833 107  Clarke, A., & Johnston, N. M. (1999). Scaling of metabolic rate with body mass and temperature in teleost fish. Journal of Animal Ecology, 68(5), 893–905. https://doi.org/10.1046/j.1365-2656.1999.00337.x Clarke, K. R., Gorley, R. N., Somerfield, P. J., & Warwick, R. M. (2014). Change in marine communities: An approach to statistical analysis and interpretation, 3rd edn. Primer-E Ltd. http://plymsea.ac.uk/id/eprint/7656/ Clarke, T. M., Espinoza, M., Ahrens, R., & Wehrtmann, I. S. (2016). Elasmobranch bycatch associated with the shrimp trawl fishery off the Pacific coast of Costa Rica, Central America. Fishery Bulletin, 114(1). Craig, J. K. (2012). Aggregation on the edge: Effects of hypoxia avoidance on the spatial distribution of brown shrimp and demersal fishes in the Northern Gulf of Mexico. Marine Ecology Progress Series, 445, 75–95. https://doi.org/10.3354/meps09437 Crear, D. P., Brill, R. W., Bushnell, P. G., Latour, R. J., Schwieterman, G. D., Steffen, R. M., & Weng, K. C. (2020). The impacts of warming and hypoxia on the performance of an obligate ram ventilator. Conservation Physiology, 7(1). https://doi.org/10.1093/conphys/coz026 Dakos, V., Matthews, B., Hendry, A. P., Levine, J., Loeuille, N., Norberg, J., Nosil, P., Scheffer, M., & De Meester, L. (2019). Ecosystem tipping points in an evolving world. Nature Ecology & Evolution, 3(3), 355–362. https://doi.org/10.1038/s41559-019-0797-2 Danabasoglu, G., Lamarque, J.-F., Bacmeister, J., Bailey, D. A., DuVivier, A. K., Edwards, J., Emmons, L. K., Fasullo, J., Garcia, R., Gettelman, A., Hannay, C., Holland, M. M., Large, W. G., Lauritzen, P. H., Lawrence, D. M., Lenaerts, J. T. M., Lindsay, K., Lipscomb, W. H., Mills, M. J., … Strand, W. G. (2020). The Community Earth System Model Version 2 (CESM2). Journal of Advances in Modeling Earth Systems, 12(2), e2019MS001916. https://doi.org/10.1029/2019MS001916 Defeo, O., & Castilla, J. C. (2012). Governance and governability of coastal shellfisheries in Latin America and the Caribbean: multi-scale emerging models and effects of globalization and climate change. Current Opinion in Environmental Sustainability, 4(3), 344-350. 108  DeRoy, E. M., Scott, R., Hussey, N. E., & MacIsaac, H. J. (2020). High predatory efficiency and abundance drive expected ecological impacts of a marine invasive fish. Marine Ecology Progress Series, 637, 195–208. https://doi.org/10.3354/meps13251 Deutsch, C., Ferrel, A., Seibel, B., Pörtner, H.-O., & Huey, R. B. (2015). Climate change tightens a metabolic constraint on marine habitats. Science, 348(6239), 1132–1135. https://doi.org/10.1126/science.aaa1605 Deutsch, C., Penn, J. L., & Seibel, B. (2020). Metabolic trait diversity shapes marine biogeography. Nature, 585(7826), 557–562. https://doi.org/10.1038/s41586-020-2721-y Diaz Pauli, B., Kolding, J., Jeyakanth, G., & Heino, M. (2017). Effects of ambient oxygen and size-selective mortality on growth and maturation in guppies. Conservation Physiology, 5(1). https://doi.org/10.1093/conphys/cox010 Dufresne, J.-L., Foujols, M.-A., Denvil, S., Caubel, A., Marti, O., Aumont, O., Balkanski, Y., Bekki, S., Bellenger, H., Benshila, R., Bony, S., Bopp, L., Braconnot, P., Brockmann, P., Cadule, P., Cheruy, F., Codron, F., Cozic, A., Cugnet, D., … Vuichard, N. (2013). Climate change projections using the IPSL-CM5 Earth System Model: From CMIP3 to CMIP5. Climate Dynamics, 40(9), 2123–2165. https://doi.org/10.1007/s00382-012-1636-1 Dulvy, N. K., Rogers, S. I., Jennings, S., Stelzenmüller, V., Dye, S. R., & Skjoldal, H. R. (2008). Climate change and deepening of the North Sea fish assemblage: A biotic indicator of warming seas. Journal of Applied Ecology, 45(4), 1029–1039. https://doi.org/10.1111/j.1365-2664.2008.01488.x Duncan, M. I., Bates, A. E., James, N. C., & Potts, W. M. (2019). Exploitation may influence the climate resilience of fish populations through removing high performance metabolic phenotypes. Scientific Reports, 9(1), 11437. https://doi.org/10.1038/s41598-019-47395-y Dunne, J. P., John, J. G., Shevliakova, E., Stouffer, R. J., Krasting, J. P., Malyshev, S. L., Milly, P. C. D., Sentman, L. T., Adcroft, A. J., Cooke, W., Dunne, K. A., Griffies, S. M., Hallberg, R. W., Harrison, M. J., Levy, H., Wittenberg, A. T., Phillips, P. J., & Zadeh, N. (2013). GFDL’s ESM2 Global Coupled Climate–Carbon Earth System Models. Part II: Carbon System Formulation and 109  Baseline Simulation Characteristics. Journal of Climate, 26(7), 2247–2267. https://doi.org/10.1175/JCLI-D-12-00150.1 Duteil, O., Oschlies, A., & Böning, C. W. (2018). Pacific Decadal Oscillation and recent oxygen decline in the eastern tropical Pacific Ocean. Biogeosciences; Katlenburg-Lindau, 15(23), 7111–7126. http://dx.doi.org.ezproxy.library.ubc.ca/10.5194/bg-15-7111-2018 Eddy, T. D. (2019). Chapter 7—Building confidence in projections of future ocean capacity. In A. M. Cisneros-Montemayor, W. W. L. Cheung, & Y. Ota (Eds.), Predicting Future Oceans (pp. 69–76). Elsevier. https://doi.org/10.1016/B978-0-12-817945-1.00007-1 Elith, J., Leathwick, J. R., & Hastie, T. (2008). A working guide to boosted regression trees. Journal of Animal Ecology, 77(4), 802–813. https://doi.org/10.1111/j.1365-2656.2008.01390.x Elith, Jane, Phillips, S. J., Hastie, T., Dudík, M., Chee, Y. E., & Yates, C. J. (2011). A statistical explanation of MaxEnt for ecologists. Diversity and Distributions, 17(1), 43–57. https://doi.org/10.1111/j.1472-4642.2010.00725.x Espinoza, M., Díaz, E., Angulo, A., Hernández, S., & Clarke, T. M. (2018). Chondrichthyan Diversity, Conservation Status, and Management Challenges in Costa Rica. Frontiers in Marine Science, 5. https://doi.org/10.3389/fmars.2018.00085 Evans, T. G., Diamond, S. E., & Kelly, M. W. (2015). Mechanistic species distribution modelling as a link between physiology and conservation. Conservation Physiology, 3(1). https://doi.org/10.1093/conphys/cov056 Facultad de Ciencias Biológicas, 2001. Colección de Crustáceos Decápodos Marinos de las Costas Mexicanas (UANL) (Base de datos REMIB-CONABIO). Universidad Autónoma de Nuevo León, San Nicolás de los Garza, Nuevo León, México. FAO. (2020). The State of World Fisheries and Aquaculture 2020: Sustainability in action. FAO. https://doi.org/10.4060/ca9229en Farchadi, N., Hinton, M. G., Thompson, A. R., & Yin, Z.-Y. (2019). Modeling the dynamic habitats of mobile pelagic predators (Makaira nigricans and Istiompax indica) in the eastern Pacific Ocean. Marine Ecology Progress Series, 622, 157–176. https://doi.org/10.3354/meps12996 110  Fiedler, P. C. (2002). Environmental change in the eastern tropical Pacific Ocean: Review of ENSO and decadal variability. Marine Ecology Progress Series, 244, 265–283. https://doi.org/10.3354/meps244265 Fiedler, P. C., & Lavín, M. F. (2017). Oceanographic Conditions of the Eastern Tropical Pacific. In P. W. Glynn, D. P. Manzello, & I. C. Enochs (Eds.), Coral Reefs of the Eastern Tropical Pacific: Persistence and Loss in a Dynamic Environment (pp. 59–83). Springer Netherlands. https://doi.org/10.1007/978-94-017-7499-4_3 Forster, J., Hirst, A. G., & Atkinson, D. (2012). Warming-induced reductions in body size are greater in aquatic than terrestrial species. Proceedings of the National Academy of Sciences, 109(47), 19310–19314. https://doi.org/10.1073/pnas.1210460109 Frazão Santos, C., Agardy, T., Andrade, F., Calado, H., Crowder, L. B., Ehler, C. N., García-Morales, S., Gissi, E., Halpern, B. S., Orbach, M. K., Pörtner, H.-O., & Rosa, R. (2020). Integrating climate change in ocean planning. Nature Sustainability, 3(7), 505–516. https://doi.org/10.1038/s41893-020-0513-x Free, C. M., Thorson, J. T., Pinsky, M. L., Oken, K. L., Wiedenmann, J., & Jensen, O. P. (2019). Impacts of historical warming on marine fisheries production. Science, 363(6430), 979–983. https://doi.org/10.1126/science.aau1758 Froese, R., & Pauly, D. (2019). FishBase. https://www.fishbase.org/ Frölicher, T. L., Rodgers, K. B., Stock, C. A., & Cheung, W. W. L. (2016). Sources of uncertainties in 21st century projections of potential ocean ecosystem stressors. Global Biogeochemical Cycles, 30(8), 1224–1243. https://doi.org/10.1002/2015GB005338 Funes‐Rodríguez, R., Zárate‐Villafranco, A., Hinojosa‐Medina, A., González‐Armas, R., & Hernández‐Trujillo, S. (2011). Mesopelagic fish larval assemblages during El Niño-southern oscillation (1997–2001) in the southern part of the California Current. Fisheries Oceanography, 20(4), 329–346. https://doi.org/10.1111/j.1365-2419.2011.00587.x 111  Gagné, T. O., Reygondeau, G., Jenkins, C. N., Sexton, J. O., Bograd, S. J., Hazen, E. L., & Houtan, K. S. V. (2020). Towards a global understanding of the drivers of marine and terrestrial biodiversity. PLOS ONE, 15(2), e0228065. https://doi.org/10.1371/journal.pone.0228065 Gaines, S. D., Costello, C., Owashi, B., Mangin, T., Bone, J., Molinos, J. G., Burden, M., Dennis, H., Halpern, B. S., Kappel, C. V., Kleisner, K. M., & Ovando, D. (2018). Improved fisheries management could offset many negative effects of climate change. Science Advances, 4(8), eaao1378. https://doi.org/10.1126/sciadv.aao1378 Galbraith, E. D., Carozza, D. A., & Bianchi, D. (2017). A coupled human-Earth model perspective on long-term trends in the global marine fishery. Nature Communications, 8(1), 14884. https://doi.org/10.1038/ncomms14884 Gallardo, M., Rojas, I., Brokordt, K., Lovrich, G., Nuñez, V., Paschke, K., Thiel, M., & Yannicelli, B. (2019). Life on the edge: Incubation behaviour and physiological performance of squat lobsters in oxygen-minimum conditions. Marine Ecology Progress Series, 623, 51–70. https://doi.org/10.3354/meps12984 Gallo, N. D., & Levin, L. A. (2016). Fish Ecology and Evolution in the World’s Oxygen Minimum Zones and Implications of Ocean Deoxygenation. In B. E. Curry (Ed.), Advances in Marine Biology (Vol. 74, pp. 117–198). Academic Press. https://doi.org/10.1016/bs.amb.2016.04.001 Gamito, R., Costa, M. J., & Cabral, H. N. (2015). Fisheries in a warming ocean: Trends in fish catches in the large marine ecosystems of the world. Regional Environmental Change, 15, 57–65. https://doi.org/DOI 10.1007/s10113-014-0615-y Garcia, H. E., & Gordon, L. I. (1992). Oxygen solubility in seawater: Better fitting equations. Limnology and Oceanography, 37(6), 1307–1312. https://doi.org/10.4319/lo.1992.37.6.1307 Gattuso, J.-P., Magnan, A., Billé, R., Cheung, W. W. L., Howes, E. L., Joos, F., Allemand, D., Bopp, L., Cooley, S. R., Eakin, C. M., Hoegh-Guldberg, O., Kelly, R. P., Pörtner, H.-O., Rogers, A. D., Baxter, J. M., Laffoley, D., Osborn, D., Rankovic, A., Rochette, J., … Turley, C. (2015). Contrasting futures for ocean and society from different anthropogenic CO2 emissions scenarios. Science, 349(6243), aac4722. https://doi.org/10.1126/science.aac4722 112  GBIF, 2017. GBIF Home Page [WWW Document]. URL https://www.gbif.org Gilly, W. F., Beman, J. M., Litvin, S. Y., & Robison, B. H. (2013). Oceanographic and Biological Effects of Shoaling of the Oxygen Minimum Zone. Annual Review of Marine Science, 5(1), 393–420. https://doi.org/10.1146/annurev-marine-120710-100849 Giorgetta, M. A., Jungclaus, J., Reick, C. H., Legutke, S., Bader, J., Böttinger, M., Brovkin, V., Crueger, T., Esch, M., Fieg, K., Glushak, K., Gayler, V., Haak, H., Hollweg, H.-D., Ilyina, T., Kinne, S., Kornblueh, L., Matei, D., Mauritsen, T., … Stevens, B. (2013). Climate and carbon cycle changes from 1850 to 2100 in MPI-ESM simulations for the Coupled Model Intercomparison Project phase 5. Journal of Advances in Modeling Earth Systems, 5(3), 572–597. https://doi.org/10.1002/jame.20038 Gouhier, T. C., Grinsted, A., & Simko, V. (2019). biwavelet: Conduct Univariate and Bivariate Wavelet Analyses (0.20.19) [Computer software]. https://CRAN.R-project.org/package=biwavelet Grantham, B. A., Chan, F., Nielsen, K. J., Fox, D. S., Barth, J. A., Huyer, A., Lubchenco, J., & Menge, B. A. (2004). Upwelling-driven nearshore hypoxia signals ecosystem and oceanographic changes in the northeast Pacific. Nature, 429(6993), 749–754. https://doi.org/10.1038/nature02605 Griffies, S. M., Danabasoglu, G., Durack, P. J., Adcroft, A. J., Balaji, V., Böning, C. W., Chassignet, E. P., Curchitser, E., Deshayes, J., Drange, H., Fox-Kemper, B., Gleckler, P. J., Gregory, J. M., Haak, H., Hallberg, R. W., Heimbach, P., Hewitt, H. T., Holland, D. M., Ilyina, T., … Yeager, S. G. (2016). OMIP contribution to CMIP6: Experimental and diagnostic protocol for the physical component of the Ocean Model Intercomparison Project. Geoscientific Model Development, 9(9), 3231–3296. https://doi.org/10.5194/gmd-9-3231-2016 Griffiths, S. P., Olson, R. J., & Watters, G. M. (2013). Complex wasp-waist regulation of pelagic ecosystems in the Pacific Ocean. Reviews in Fish Biology and Fisheries, 4(23), 459–475. https://doi.org/10.1007/s11160-012-9301-7 Guerin, G. R., Biffin, E., & Lowe, A. J. (2013). Spatial modelling of species turnover identifies climate ecotones, climate change tipping points and vulnerable taxonomic groups. Ecography, 36(10), 1086–1096. https://doi.org/10.1111/j.1600-0587.2013.00215.x 113  Guisan, A., & Thuiller, W. (2005). Predicting species distribution: Offering more than simple habitat models. Ecology Letters, 8(9), 993–1009. https://doi.org/10.1111/j.1461-0248.2005.00792.x Guisan, A., & Zimmermann, N. E. (2000). Predictive habitat distribution models in ecology. Ecological Modelling, 135(2), 147–186. https://doi.org/10.1016/S0304-3800(00)00354-9 Gullestad, P., Sundby, S., & Kjesbu, O. S. (2020). Management of transboundary and straddling fish stocks in the Northeast Atlantic in view of climate-induced shifts in spatial distribution. Fish and Fisheries, 21(5), 1008-1026. https://doi.org/10.1111/faf.12485 Guo, C., Lek, S., Ye, S., Li, W., Liu, J., & Li, Z. (2015). Uncertainty in ensemble modelling of large-scale species distribution: Effects from species characteristics and model techniques. Ecological Modelling, 306, 67–75. https://doi.org/10.1016/j.ecolmodel.2014.08.002 Gutiérrez García, R., 2006. Evaluación del estado de explotación del camarón costero (Litopenaeus y Farfantepenaeus) del Pacífico de Nicaragua. Período 2000-2005. Centro de Investigaciones Pesqueras y Acuicolas, Managua, Nicaragua. Gutiérrez García, R., 2004. Crucero de pesca comercial de camarón de profundidad Heterocarpus affinis, en el Pacífico nicaragüense Enero - febrero 2004. Centro de Investigaciones Pesqueras y Acuicolas, Managua, Nicaragua. Gutiérrez García, R., 2003. Evaluación del langostino Pleuroncodes planipes en el Pacífico nicaragüense, por el método de área barrida. Junio 2003. Centro de Investigaciones Pesqueras y Acuicolas, Managua, Nicaragua. Halpern, B. S. (2017). Chapter 13—Addressing Socioecological Tipping Points and Safe Operating Spaces in the Anthropocene. In P. S. Levin & M. R. Poe (Eds.), Conservation for the Anthropocene Ocean (pp. 271–286). Academic Press. https://doi.org/10.1016/B978-0-12-805375-1.00013-1 Halsey, L. G., Killen, S. S., Clark, T. D., & Norin, T. (2018). Exploring key issues of aerobic scope interpretation in ectotherms: Absolute versus factorial. Reviews in Fish Biology and Fisheries, 28(2), 405–415. https://doi.org/10.1007/s11160-018-9516-3 114  Hameau, A., Frölicher, T. L., Mignot, J., & Joos, F. (2020). Is deoxygenation detectable before warming in the thermocline? Biogeosciences, 17(7), 1877–1895. https://doi.org/10.5194/bg-17-1877-2020 Hameau, A., Mignot, J., & Joos, F. (2019). Assessment of time of emergence of anthropogenic deoxygenation and warming: Insights from a CESM simulation from 850 to 2100 CE. Biogeosciences; Katlenburg-Lindau, 16(8), 1755–1780. http://dx.doi.org.ezproxy.library.ubc.ca/10.5194/bg-16-1755-2019 Hassell, K. L., Coutin, P. C., & Nugegoda, D. (2008). Hypoxia, low salinity and lowered temperature reduce embryo survival and hatch rates in black bream Acanthopagrus butcheri (Munro, 1949). Journal of Fish Biology, 72(7), 1623–1636. https://doi.org/10.1111/j.1095-8649.2008.01829.x Hernandez, P. A., Graham, C. H., Master, L. L., & Albert, D. L. (2006). The effect of sample size and species characteristics on performance of different species distribution modeling methods. Ecography, 29(5), 773–785. https://doi.org/10.1111/j.0906-7590.2006.04700.x Hidalgo, M., & Browman, H. I. (2019). Developing the knowledge base needed to sustainably manage mesopelagic resources. ICES Journal of Marine Science, 76(3), 609–615. https://doi.org/10.1093/icesjms/fsz067 Hoefnagel, K. N., & Verberk, W. C. E. P. (2015). Is the temperature-size rule mediated by oxygen in aquatic ectotherms? Journal of Thermal Biology, 54, 56–65. https://doi.org/10.1016/j.jtherbio.2014.12.003 Hoffmann, S., Irl, S. D. H., & Beierkuhnlein, C. (2019). Predicted climate shifts within terrestrial protected areas worldwide. Nature Communications, 10(1), 4787. https://doi.org/10.1038/s41467-019-12603-w Hofmann, A. F., Peltzer, E. T., Walz, P. M., & Brewer, P. G. (2011). Hypoxia by degrees: Establishing definitions for a changing ocean. Deep Sea Research Part I: Oceanographic Research Papers, 58(12), 1212–1226. https://doi.org/10.1016/j.dsr.2011.09.004 Horwitz, R., Norin, T., Watson, S.-A., Pistevos, J. C. A., Beldade, R., Hacquart, S., Gattuso, J.-P., Rodolfo-Metalpa, R., Vidal-Dupiol, J., Killen, S. S., & Mills, S. C. (2020). Near-future ocean 115  warming and acidification alter foraging behaviour, locomotion, and metabolic rate in a keystone marine mollusc. Scientific Reports, 10(1), 1–11. https://doi.org/10.1038/s41598-020-62304-4 Hsieh, C., Reiss, C. S., Hunter, J. R., Beddington, J. R., May, R. M., & Sugihara, G. (2006). Fishing elevates variability in the abundance of exploited species. Nature, 443(7113), 859–862. https://doi.org/10.1038/nature05232 Huey, R. B., & Kingsolver, J. G. (2019). Climate Warming, Resource Availability, and the Metabolic Meltdown of Ectotherms. The American Naturalist, 194(6), E140–E150. https://doi.org/10.1086/705679 Hutchinson, G. E. (1957). The multivariate niche. Cold Spring Harbor Symposia on Quantitative Biology, 22, 415–421. Instituto de Biología, 2003. Colección Nacional de Peces del Instituto de Biología (IBUNAM) (Base de datos REMIB-CONABIO). Universidad Nacional Autónoma de México, México, D.F., México. Instituto de Ciencias del Mar y Limnología, 2001a. Colección de Referencia de Crustáceos del Pacífico Mexicano (ICMyL-MAZ, UNAM) (Base de datos REMIB-CONABIO.). Universidad Nacional Autónoma de México, Mazatlán, Sinaloa, México. Instituto de Ciencias del Mar y Limnología, 2001b. Colección de Referencia de Peces del Pacífico Mexicano (ICMyL-MAZ, UNAM) (Base de datos REMIB-CONABIO.). Universidad Nacional Autónoma de México, Mazatlán, Sinaloa, México. INVEMAR, 2017. Sistema de Información Ambiental Marina [en línea]: [WWW Document]. URL http://siam.invemar.org.co/sibm-busqueda-avanzada IUCN, 2018. The IUCN Red List of Threatened Species. Version 2017-1. [WWW Document]. URL http://www.iucnredlist.org Ixquiac, M., 1998. Análisis de la composición y distribución de la fauna de acompañamiento del camarón (FAC) en el Océano Pacífico guatemalteco dentro de las isóbatas de 10 a 100 m durante los cruceros de investigación enero 1996 a febrero 1998 (Tesis de Licenciatura). Centro de Estudios del Mar y Acuicultura, Universidad de San Carlos de Guatemala, Ciudad de Guatemala, Guatemala. 116  Janssens-Maenhout, G., Crippa, M., Guizzardi, D., Muntean, M., Schaaf, E., Dentener, F., Bergamaschi, P., Pagliari, V., Olivier, J. G. J., Peters, J. A. H. W., van Aardenne, J. A., Monni, S., Doering, U., Petrescu, A. M. R., Solazzo, E., & Oreggioni, G. D. (2019). EDGAR v4.3.2 Global Atlas of the three major greenhouse gas emissions for the period 1970–2012. Earth System Science Data, 11(3), 959–1002. https://doi.org/10.5194/essd-11-959-2019 Jones, M. C., & Cheung, W. W. L. (2015). Multi-model ensemble projections of climate change effects on global marine biodiversity. ICES Journal of Marine Science, 72(3), 741–752. https://doi.org/10.1093/icesjms/fsu172 Jutfelt, F. (2020). Metabolic adaptation to warm water in fish. Functional Ecology, 34(6), 1138–1141. https://doi.org/10.1111/1365-2435.13558 Jutfelt, F., Norin, T., Åsheim, E. R., Rowsey, L. E., Andreassen, A. H., Morgan, R., Clark, T. D., & Speers-Roesch, B. (2020). The aerobic scope protection hypothesis: A mechanism explaining reduced growth of ectotherms in warming environments? [Preprint]. EcoEvoRxiv. https://doi.org/10.32942/osf.io/zc3bm Jutfelt, F., Norin, T., Ern, R., Overgaard, J., Wang, T., McKenzie, D. J., Lefevre, S., Nilsson, G. E., Metcalfe, N. B., Hickey, A. J. R., Brijs, J., Speers-Roesch, B., Roche, D. G., Gamperl, A. K., Raby, G. D., Morgan, R., Esbaugh, A. J., Gräns, A., Axelsson, M., … Clark, T. D. (2018). Oxygen- and capacity-limited thermal tolerance: Blurring ecology and physiology. Journal of Experimental Biology, 221(1). https://doi.org/10.1242/jeb.169615 Keeling, R. F., Körtzinger, A., & Gruber, N. (2009). Ocean Deoxygenation in a Warming World. Annual Review of Marine Science, 2(1), 199–229. https://doi.org/10.1146/annurev.marine.010908.163855 Keller, A. A., Ciannelli, L., Wakefield, W. W., Simon, V., Barth, J. A., & Pierce, S. D. (2015). Occurrence of demersal fishes in relation to near-bottom oxygen levels within the California Current large marine ecosystem. Fisheries Oceanography, 24(2), 162–176. https://doi.org/10.1111/fog.12100 Kilsby, C. G., Cowpertwait, P. S. P., O’Connell, P. E., & Jones, P. D. (1998). Predicting rainfall statistics in England and Wales using atmospheric circulation variables. International Journal of Climatology, 117  18(5), 523–539. https://doi.org/10.1002/(SICI)1097-0088(199804)18:5<523::AID-JOC268>3.0.CO;2-X Koenigstein, S., Mark, F. C., Gößling‐Reisemann, S., Reuter, H., & Poertner, H.-O. (2016). Modelling climate change impacts on marine fish populations: Process-based integration of ocean warming, acidification and other environmental drivers. Fish and Fisheries, 17(4), 972–1004. https://doi.org/10.1111/faf.12155 Kordas, R. L., Harley, C. D. G., & O’Connor, M. I. (2011). Community ecology in a warming world: The influence of temperature on interspecific interactions in marine systems. Journal of Experimental Marine Biology and Ecology, 400(1), 218–226. https://doi.org/10.1016/j.jembe.2011.02.029 Kujala, H., Moilanen, A., Araújo, M. B., & Cabeza, M. (2013). Conservation Planning with Uncertain Climate Change Projections. PLoS ONE, 8(2). https://doi.org/10.1371/journal.pone.0053315 Kwiatkowski, L., Bopp, L., Aumont, O., Ciais, P., Cox, P. M., Laufkötter, C., Li, Y., & Séférian, R. (2017). Emergent constraints on projections of declining primary production in the tropical oceans. Nature Climate Change, 7(5), 355–358. https://doi.org/10.1038/nclimate3265 Lam, V. W. Y., Allison, E. H., Bell, J. D., Blythe, J., Cheung, W. W. L., Frölicher, T. L., Gasalla, M. A., & Sumaila, U. R. (2020). Climate change, tropical fisheries and prospects for sustainable development. Nature Reviews Earth & Environment, 1(9), 440–454. https://doi.org/10.1038/s43017-020-0071-9 Lam, V. W. Y., Cheung, W. W. L., Reygondeau, G., & Sumaila, U. R. (2016). Projected change in global fisheries revenues under climate change. Scientific Reports, 6(1). https://doi.org/10.1038/srep32607 Laubenstein, T. D., Rummer, J. L., Nicol, S., Parsons, D. M., Pether, S. M. J., Pope, S., Smith, N., & Munday, P. L. (2018). Correlated Effects of Ocean Acidification and Warming on Behavioral and Metabolic Traits of a Large Pelagic Fish. Diversity, 10(2), 35. https://doi.org/10.3390/d10020035 Laufkötter, C., Vogt, M., Gruber, N., Aita-Noguchi, M., Aumont, O., Bopp, L., Buitenhuis, E., Doney, S. C., Dunne, J., Hashioka, T., Hauck, J., Hirata, T., John, J., Le Quéré, C., Lima, I. D., Nakano, H., Seferian, R., Totterdell, I., Vichi, M., & Völker, C. (2015). Drivers and uncertainties of future global 118  marine primary production in marine ecosystem models. Biogeosciences, 12(23), 6955–6984. https://doi.org/10.5194/bg-12-6955-2015 Lefevre, S., McKenzie, D. J., & Nilsson, G. E. (2017). Models projecting the fate of fish populations under climate change need to be based on valid physiological mechanisms. Global Change Biology, 23(9), 3449–3459. https://doi.org/10.1111/gcb.13652 Leiva, F. P., Calosi, P., & Verberk, W. C. E. P. (2019). Scaling of thermal tolerance with body mass and genome size in ectotherms: A comparison between water- and air-breathers. Philosophical Transactions of the Royal Society B: Biological Sciences, 374(1778), 20190035. https://doi.org/10.1098/rstb.2019.0035 Lek, S., & Guégan, J.-F. (1999). Artificial neural networks as a tool in ecological modelling: An introduction. Ecological Modelling, 120(2–3), 65–73. https://doi.org/10.1016/S0304-3800(99)00092-7 Lenoir, J., & Svenning, J.-C. (2015). Climate-related range shifts – a global multidimensional synthesis and new research directions. Ecography, 38(1), 15–28. https://doi.org/10.1111/ecog.00967 Leung, S., Thompson, L., McPhaden, M. J., & Mislan, K. A. S. (2019). ENSO drives near-surface oxygen and vertical habitat variability in the tropical Pacific. Environmental Research Letters, 14(6), 064020. https://doi.org/10.1088/1748-9326/ab1c13 Levin, L. A., Ekau, W., Gooday, A. J., Jorissen, F., Middelburg, J. J., Naqvi, S. W. A., Neira, C., Rabalais, N. N., & Zhang, J. (2009). Effects of natural and human-induced hypoxia on coastal benthos. Biogeosciences, 6(10), 2063–2098. https://doi.org/10.5194/bg-6-2063-2009 Levin, Lisa A., & Breitburg, D. L. (2015). Linking coasts and seas to address ocean deoxygenation. Nature Climate Change, 5, 401–403. https://doi.org/10.1038/nclimate2595 Limburg, K. E., Breitburg, D., & Levin, L. A. (2017). Ocean deoxygenation – a climate-related problem. Frontiers in Ecology and the Environment, 15(9), 479–479. https://doi.org/10.1002/fee.1728 Lluch-Cota, S. E., Arreguín-Sánchez, F., Salvadeo, C. J., & del Monte Luna. (2019). Climate change impacts, vulnerabilities and adaptations: Northeast Tropical Pacific marine fisheries. In Impacts of climate change on fisheries and aquaculture: Synthesis of current knowledge, adaptation and 119  mitigation options. FAO Fisheries and Aquaculture Technical Paper. 627 (pp. 207–219). Food & Agriculture Org. López-Martı́nez, J., Arreguı́n-Sánchez, F., Hernández-Vázquez, S., Garcı́a-Juárez, A. R., & Valenzuela-Quiñonez, W. (2003). Interannual variation of growth of the brown shrimp Farfantepenaeus californiensis and its relation to temperature. Fisheries Research, 61(1–3), 95–105. https://doi.org/10.1016/S0165-7836(02)00239-4 Lotze, H. K., Tittensor, D. P., Bryndum-Buchholz, A., Eddy, T. D., Cheung, W. W. L., Galbraith, E. D., Barange, M., Barrier, N., Bianchi, D., Blanchard, J. L., Bopp, L., Büchner, M., Bulman, C. M., Carozza, D. A., Christensen, V., Coll, M., Dunne, J. P., Fulton, E. A., Jennings, S., … Worm, B. (2019). Global ensemble projections reveal trophic amplification of ocean biomass declines with climate change. Proceedings of the National Academy of Sciences, 116(26), 12907–12912. https://doi.org/10.1073/pnas.1900194116 Maestri, R., & Duarte, L. (2020). Evolutionary Imprints on Species Distribution Patterns Across the Neotropics. In V. Rull & A. C. Carnaval (Eds.), Neotropical Diversification: Patterns and Processes (pp. 103–119). Springer International Publishing. https://doi.org/10.1007/978-3-030-31167-4_6 Marín-Enríquez, E., & Muhlia-Melo, A. (2018). Incidental catch of the rare shortbill spearfish (Tetrapturus angustirostris) by the tuna purse seine fleet in the eastern tropical Pacific Ocean. Ciencias Marinas, 44(1), 21–32. https://doi.org/10.7773/cm.v44i1.2727 Marshall, D. J., & White, C. R. (2019). Aquatic Life History Trajectories Are Shaped by Selection, Not Oxygen Limitation. Trends in Ecology & Evolution, 34(3), 182–184. https://doi.org/10.1016/j.tree.2018.12.015 Martínez-Ortiz, J., Aires-da-Silva, A. M., Lennert-Cody, C. E., & Maunder, M. N. (2015). The Ecuadorian Artisanal Fishery for Large Pelagics: Species Composition and Spatio-Temporal Dynamics. PLOS ONE, 10(8), e0135136. https://doi.org/10.1371/journal.pone.0135136 Marzloff, M. P., Melbourne‐Thomas, J., Hamon, K. G., Hoshino, E., Jennings, S., Putten, I. E. van, & Pecl, G. T. (2016). Modelling marine community responses to climate-driven species redistribution to 120  guide monitoring and adaptive ecosystem-based management. Global Change Biology, 22(7), 2462–2474. https://doi.org/10.1111/gcb.13285 McClatchie, S., Gao, J., Drenkard, E. J., Thompson, A. R., Watson, W., Ciannelli, L., Bograd, S. J., & Thorson, J. T. (2018). Interannual and Secular Variability of Larvae of Mesopelagic and Forage Fishes in the Southern California Current System. Journal of Geophysical Research: Oceans, 123(9), 6277–6295. https://doi.org/10.1029/2018JC014011 McHenry, J., Welch, H., Lester, S. E., & Saba, V. (2019). Projecting marine species range shifts from only temperature can mask climate vulnerability. Global Change Biology, 25(12), 4208–4221. https://doi.org/10.1111/gcb.14828 Miller, K. A. (2007). Climate variability and tropical tuna: Management challenges for highly migratory fish stocks. Marine Policy, 31(1), 56–70. https://doi.org/10.1016/j.marpol.2006.05.006 Mills, K. E., Pershing, A., Brown, C. J., Chen, Y., Chiang, F.-S., Holland, D. A., Lehuta, S., Nye, J. A., Sun, J. C., Thomas, A. C., & Wahle, R. (2013). Fisheries Management in a Changing Climate: Lessons from the 2012 Ocean Heat Wave in the Northwest Atlantic. Oceanography, 26(2), 191–195. Mislan, K. a. S., Deutsch, C. A., Brill, R. W., Dunne, J. P., & Sarmiento, J. L. (2017). Projections of climate-driven changes in tuna vertical habitat based on species-specific differences in blood oxygen affinity. Global Change Biology, 23(10), 4019–4028. https://doi.org/10.1111/gcb.13799 Moltmann, T., Turton, J., Zhang, H.-M., Nolan, G., Gouldman, C., Griesbauer, L., Willis, Z., Piniella, Á. M., Barrell, S., Andersson, E., Gallage, C., Charpentier, E., Belbeoch, M., Poli, P., Rea, A., Burger, E. F., Legler, D. M., Lumpkin, R., Meinig, C., … Zhang, Y. (2019). A Global Ocean Observing System (GOOS), Delivered Through Enhanced Collaboration Across Regions, Communities, and New Technologies. Frontiers in Marine Science, 6. https://doi.org/10.3389/fmars.2019.00291 Mora, C., & Robertson, D. R. (2005). Causes of Latitudinal Gradients in Species Richness: A Test with Fishes of the Tropical Eastern Pacific. Ecology, 86(7), 1771–1782. https://doi.org/10.1890/04-0883 121  Morley, J. W., Selden, R. L., Latour, R. J., Frölicher, T. L., Seagraves, R. J., & Pinsky, M. L. (2018). Projecting shifts in thermal habitat for 686 species on the North American continental shelf. PLOS ONE, 13(5), e0196127. https://doi.org/10.1371/journal.pone.0196127 Muggeo, V. M. R. (2003). Estimating regression models with unknown break-points. Statistics in Medicine, 22(19), 3055–3071. https://doi.org/10.1002/sim.1545 Muggeo, V. M. R. (2017). Interval estimation for the breakpoint in segmented regression: A smoothed score-based approach. Australian & New Zealand Journal of Statistics, 59(3), 311–322. https://doi.org/10.1111/anzs.12200 Nguyen, K. D. T., Morley, S. A., Lai, C.-H., Clark, M. S., Tan, K. S., Bates, A. E., & Peck, L. S. (2011). Upper Temperature Limits of Tropical Marine Ectotherms: Global Warming Implications. PLOS ONE, 6(12), e29340. https://doi.org/10.1371/journal.pone.0029340 OBIS, 2017. Ocean Biogeographic Information System [WWW Document]. URL www.obis.org Oremus, K. L., Bone, J., Costello, C., García Molinos, J., Lee, A., Mangin, T., & Salzman, J. (2020). Governance challenges for tropical nations losing fish species due to climate change. Nature Sustainability, 3(4), 277–280. https://doi.org/10.1038/s41893-020-0476-y Orr, J. C., Najjar, R. G., Aumont, O., Bopp, L., Bullister, J. L., Danabasoglu, G., Doney, S. C., Dunne, J. P., Dutay, J.-C., Graven, H., Griffies, S. M., John, J. G., Joos, F., Levin, I., Lindsay, K., Matear, R. J., McKinley, G. A., Mouchet, A., Oschlies, A., … Yool, A. (2017). Biogeochemical protocols and diagnostics for the CMIP6 Ocean Model Intercomparison Project (OMIP). Geoscientific Model Development, 10(6), 2169–2199. https://doi.org/10.5194/gmd-10-2169-2017 Oschlies, A., Brandt, P., Stramma, L., & Schmidtko, S. (2018). Drivers and mechanisms of ocean deoxygenation. Nature Geoscience, 11(7), 467–473. https://doi.org/10.1038/s41561-018-0152-2 Páez-Osuna, F., Sanchez-Cabeza, J. A., Ruiz-Fernández, A. C., Alonso-Rodríguez, R., Piñón-Gimate, A., Cardoso-Mohedano, J. G., Flores-Verdugo, F. J., Carballo, J. L., Cisneros-Mata, M. A., & Álvarez-Borrego, S. (2016). Environmental status of the Gulf of California: A review of responses to climate change and climate variability. Earth-Science Reviews, 162, 253–268. https://doi.org/10.1016/j.earscirev.2016.09.015 122  Palomares, M.L.D. & D. Pauly. Editors. (2020). SeaLifeBase. World Wide Web electronic publication. www.sealifebase.org, version (07/2020). Papiol, V., Hendrickx, M. E., & Serrano, D. (2017). Effects of latitudinal changes in the oxygen minimum zone of the northeast Pacific on the distribution of bathyal benthic decapod crustaceans. Deep Sea Research Part II: Topical Studies in Oceanography, 137, 113–130. https://doi.org/10.1016/j.dsr2.2016.04.023 Pauly, D. (2010). Gasping fish and panting squids: Oxygen, temperature and the growth of water-breathing animals. Oldendorf/Luhe (Germany) International Ecology Inst. Pauly, D., & Cheung, W. W. L. (2018). On confusing cause and effect in the oxygen limitation of fish. Global Change Biology, 24(11), e743–e744. https://doi.org/10.1111/gcb.14383 Pauly, D., & Froese, R. (2020). Are smaller fish pre-adapted to warmer oceans? In Daniel Pauly & V. Ruiz-Leotaud (Eds.), Marine and Freshwater Miscellanea II (Vol. 28, pp. 131–141). http://eprints.uni-kiel.de/49280/ Pauly, D., & Zeller, D. (2016). Catch reconstructions reveal that global marine fisheries catches are higher than reported and declining. Nature Communications, 7(1), 1–9. https://doi.org/10.1038/ncomms10244 Pauly D., Zeller D., Palomares M.L.D. (Editors), 2020. Sea Around Us Concepts, Design and Data (seaaroundus.org). Payne, N. L., & Smith, J. A. (2017). An alternative explanation for global trends in thermal tolerance. Ecology Letters, 20(1), 70–77. https://doi.org/10.1111/ele.12707 Pearlman, J., Bushnell, M., Coppola, L., Karstensen, J., Buttigieg, P. L., Pearlman, F., Simpson, P., Barbier, M., Muller-Karger, F. E., Munoz-Mas, C., Pissierssens, P., Chandler, C., Hermes, J., Heslop, E., Jenkyns, R., Achterberg, E. P., Bensi, M., Bittig, H. C., Blandin, J., … Whoriskey, F. (2019). Evolving and Sustaining Ocean Best Practices and Standards for the Next Decade. Frontiers in Marine Science, 6. https://doi.org/10.3389/fmars.2019.00277 Pecl, G. T., Araújo, M. B., Bell, J. D., Blanchard, J., Bonebrake, T. C., Chen, I.-C., Clark, T. D., Colwell, R. K., Danielsen, F., Evengård, B., Falconi, L., Ferrier, S., Frusher, S., Garcia, R. A., Griffis, R. B., Hobday, A. J., Janion-Scheepers, C., Jarzyna, M. A., Jennings, S., … Williams, S. E. (2017). 123  Biodiversity redistribution under climate change: Impacts on ecosystems and human well-being. Science, 355(6332). https://doi.org/10.1126/science.aai9214 Peñaherrera-Palma, C., Arauz, R., Bessudo, S., Bravo-Ormaza, E., Chassot, O., Chinacalle-Martínez, N., Espinoza, E., Forsberg, K., García-Rada, E., & Guzmán, H. (2018). Justificación biológica para la creación de la MigraVía Coco-Galápagos. MigraMar y Pontificia Universidad Católica Del Ecuador Sede Manabí. Portoviejo, Manabí, Ecuador. Penn, J. L., Deutsch, C., Payne, J. L., & Sperling, E. A. (2018). Temperature-dependent hypoxia explains biogeography and severity of end-Permian marine mass extinction. Science, 362(6419). https://doi.org/10.1126/science.aat1327 Peterson, A. T., & Soberón, J. (2012). Integrating fundamental concepts of ecology, biogeography, and sampling into effective ecological niche modeling and species distribution modeling. Plant Biosystems - An International Journal Dealing with All Aspects of Plant Biology, 146(4), 789–796. https://doi.org/10.1080/11263504.2012.740083 Peterson, A. Townsend, Soberón, J., Pearson, R. G., Anderson, R. P., Martínez-Meyer, E., Nakamura, M., & Araújo, M. B. (2011). Ecological Niches and Geographic Distributions (MPB-49). Princeton University Press. Phillips, S. J., Dudík, M., Elith, J., Graham, C. H., Lehmann, A., Leathwick, J., & Ferrier, S. (2009). Sample selection bias and presence-only distribution models: Implications for background and pseudo-absence data. Ecological Applications, 19(1), 181–197. https://doi.org/10.1890/07-2153.1 Pinsky, M. L., & Fogarty, M. (2012). Lagged social-ecological responses to climate and range shifts in fisheries. Climatic Change, 115(3–4), 883–891. https://doi.org/10.1007/s10584-012-0599-x Pinsky, M. L., Morley, J. W., & Frölicher, T. L. (2018). Can we adapt to species on the move? The Effects of Climate Change on the World’s Oceans Book of Abstracts, S10-Invited-12829. Poloczanska, E. S., Brown, C. J., Sydeman, W. J., Kiessling, W., Schoeman, D. S., Moore, P. J., Brander, K., Bruno, J. F., Buckley, L. B., Burrows, M. T., Duarte, C. M., Halpern, B. S., Holding, J., Kappel, C. V., O’Connor, M. I., Pandolfi, J. M., Parmesan, C., Schwing, F., Thompson, S. A., & 124  Richardson, A. J. (2013). Global imprint of climate change on marine life. Nature Climate Change, 3(10), 919–925. https://doi.org/10.1038/nclimate1958 Poloczanska, E. S., Burrows, M. T., Brown, C. J., García Molinos, J., Halpern, B. S., Hoegh-Guldberg, O., Kappel, C. V., Moore, P. J., Richardson, A. J., Schoeman, D. S., & Sydeman, W. J. (2016). Responses of Marine Organisms to Climate Change across Oceans. Frontiers in Marine Science, 3. https://doi.org/10.3389/fmars.2016.00062 Pörtner, Hans-O., Bock, C., & Mark, F. C. (2017). Oxygen- and capacity-limited thermal tolerance: Bridging ecology and physiology. Journal of Experimental Biology, 220(15), 2685–2696. https://doi.org/10.1242/jeb.134585 Pörtner, Hans-Otto, & Giomi, F. (2013). Nothing in experimental biology makes sense except in the light of ecology and evolution – correspondence on J. Exp. Biol. 216, 2771-2782. Journal of Experimental Biology, 216(23), 4494–4495. https://doi.org/10.1242/jeb.095232 Prince, E. D., & Goodyear, C. P. (2006). Hypoxia-based habitat compression of tropical pelagic fishes. Fisheries Oceanography, 15(6), 451–464. https://doi.org/10.1111/j.1365-2419.2005.00393.x Prince, E. D., Luo, J., Goodyear, C. P., Hoolihan, J. P., Snodgrass, D., Orbesen, E. S., Serafy, J. E., Ortiz, M., & Schirripa, M. J. (2010). Ocean scale hypoxia-based habitat compression of Atlantic istiophorid billfishes. Fisheries Oceanography, 19(6), 448–462. https://doi.org/10.1111/j.1365-2419.2010.00556.x Reygondeau, G. (2019). Chapter 9—Current and future biogeography of exploited marine groups under climate change. In A. M. Cisneros-Montemayor, W. W. L. Cheung, & Y. Ota (Eds.), Predicting Future Oceans (pp. 87–101). Elsevier. https://doi.org/10.1016/B978-0-12-817945-1.00009-5 Riahi, K., Rao, S., Krey, V., Cho, C., Chirkov, V., Fischer, G., Kindermann, G., Nakicenovic, N., & Rafaj, P. (2011). RCP 8.5—A scenario of comparatively high greenhouse gas emissions. Climatic Change, 109(1), 33. https://doi.org/10.1007/s10584-011-0149-y Rilov, G., Mazaris, A. D., Stelzenmüller, V., Helmuth, B., Wahl, M., Guy-Haim, T., Mieszkowska, N., Ledoux, J.-B., & Katsanevakis, S. (2019). Adaptive marine conservation planning in the face of 125  climate change: What can we learn from physiological, ecological and genetic studies? Global Ecology and Conservation, 17, e00566. https://doi.org/10.1016/j.gecco.2019.e00566 Rio, I. J. L., Franco, N. I. J. (2011). Informe de resultados de la campaña de investigación pesquera: Centroamérica-Pacífico 2010, 10 de noviembre al 16 de diciembre de 2010. B/O Miguel Oliver. Organization of Fishing and Aquaculture in Central America., Spain., Spain., & Proyecto Apoyo al Proceso de Integración de la Pesca y la Acuicultura Centroamericana. Robertson, D. R., & Allen, G. R. (2015). Shorefishes of the tropical Eastern Pacific: Online information system. Version 2.0. Balboa: Smithsonian Tropical Research Institute. https://biogeodb.stri.si.edu/sftep/en/pages Robin, X., Turck, N., Hainard, A., Tiberti, N., Lisacek, F., Sanchez, J.-C., & Müller, M. (2011). pROC: An open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics, 12(1), 77. https://doi.org/10.1186/1471-2105-12-77 Robinson, L. M., Hobday, A. J., Possingham, H. P., & Richardson, A. J. (2015). Trailing edges projected to move faster than leading edges for large pelagic fish habitats under climate change. Deep Sea Research Part II: Topical Studies in Oceanography, 113, 225–234. https://doi.org/10.1016/j.dsr2.2014.04.007 Rodgers, K. B., Lin, J., & Frölicher, T. L. (2015). Emergence of multiple ocean ecosystem drivers in a large ensemble suite with an Earth system model. Biogeosciences; Katlenburg-Lindau, 12(11), 3301. http://dx.doi.org.ezproxy.library.ubc.ca/10.5194/bg-12-3301-2015 Rollinson, N., & Rowe, L. (2018). Temperature-dependent oxygen limitation and the rise of Bergmann’s rule in species with aquatic respiration. Evolution, 72(4), 977–988. https://doi.org/10.1111/evo.13458 Rose, K. A., Gutiérrez, D., Breitburg, D., Conley, D., Craig, K. J., Froehlich, H. E., Jeyabaskaran, R., Kripa, V., Mbaye, B. C., Mohamed, K. S., Padua, S., & Prema, D. (2019). Impacts of ocean deoxygenation on fisheries. In Ocean deoxygenation: Everyone’s problem Causes, impacts, consequences and solutions (pp. 519–544). International Union for Conservation of Nature and Natural Resources. https://portals.iucn.org/library/sites/library/files/documents/2019-048-En.pdf 126  Ruiz-Cooley, R. I., Gerrodette, T., Fiedler, P. C., Chivers, S. J., Danil, K., & Ballance, L. T. (2017). Temporal variation in pelagic food chain length in response to environmental change. Science Advances, 3(10), e1701140. https://doi.org/10.1126/sciadv.1701140 Sandblom, E., Clark, T. D., Gräns, A., Ekström, A., Brijs, J., Sundström, L. F., Odelström, A., Adill, A., Aho, T., & Jutfelt, F. (2016). Physiological constraints to climate warming in fish follow principles of plastic floors and concrete ceilings. Nature Communications, 7(1), 11447. https://doi.org/10.1038/ncomms11447 Sanz, N., Diop, B., Blanchard, F., & Lampert, L. (2017). On the influence of environmental factors on harvest: The French Guiana shrimp fishery paradox. Environmental Economics and Policy Studies, 19(2), 233–247. https://doi.org/10.1007/s10018-016-0153-6 Sarmiento, J. L., & Gruber, N. (2006). Ocean Biogeochemical Dynamics. Princeton University Press. Sato, K. N., Levin, L. A., & Schiff, K. (2017). Habitat compression and expansion of sea urchins in response to changing climate conditions on the California continental shelf and slope (1994–2013). Deep Sea Research Part II: Topical Studies in Oceanography, 137, 377–389. https://doi.org/10.1016/j.dsr2.2016.08.012 Schirripa, M. J., Goodyear, C. P., & Forrestal, F. (2017). Longline data simulation: Integrating #-D species habitat with oceanographic data and depth distributions of pelagic longline hooks. Collect. Vol. Sci. Pap. ICCAT, 73(9), 3025–3034. Schmidtko, S., Stramma, L., & Visbeck, M. (2017). Decline in global oceanic oxygen content during the past five decades. Nature, 542(7641), 335–339. https://doi.org/10.1038/nature21399 Schwing, F. B., Palacios, D. M., Bograd, S. J., Grove, P., & Schwing, F. (2005). El Niño Impacts on the California Current Ecosystem. U.S. CLIVAR Newsletter, 3(2), 5–8. Séférian, R., Nabat, P., Michou, M., Saint‐Martin, D., Voldoire, A., Colin, J., Decharme, B., Delire, C., Berthet, S., Chevallier, M., Sénési, S., Franchisteguy, L., Vial, J., Mallet, M., Joetzjer, E., Geoffroy, O., Guérémy, J.-F., Moine, M.-P., Msadek, R., … Madec, G. (2019). Evaluation of CNRM Earth System Model, CNRM-ESM2-1: Role of Earth System Processes in Present-Day 127  and Future Climate. Journal of Advances in Modeling Earth Systems, 11(12), 4182–4227. https://doi.org/10.1029/2019MS001791 Selden, R., & Pinsky, M. (2019). Chapter 20—Climate change adaptation and spatial fisheries management. In A. M. Cisneros-Montemayor, W. W. L. Cheung, & Y. Ota (Eds.), Predicting Future Oceans (pp. 207–214). Elsevier. https://doi.org/10.1016/B978-0-12-817945-1.00023-X Selkoe, K. A., Blenckner, T., Caldwell, M. R., Crowder, L. B., Erickson, A. L., Essington, T. E., Estes, J. A., Fujita, R. M., Halpern, B. S., Hunsicker, M. E., Kappel, C. V., Kelly, R. P., Kittinger, J. N., Levin, P. S., Lynham, J. M., Mach, M. E., Martone, R. G., Mease, L. A., Salomon, A. K., … Zedler, J. (2015). Principles for managing marine ecosystems prone to tipping points. Ecosystem Health and Sustainability, 1(5), 1–18. https://doi.org/10.1890/EHS14-0024.1 Serafin, J., Guffey, S. C., Bosker, T., Griffitt, R. J., De Guise, S., Perkins, C., Szuter, M., & Sepúlveda, M. S. (2019). Combined effects of salinity, temperature, hypoxia, and Deepwater Horizon oil on Fundulus grandis larvae. Ecotoxicology and Environmental Safety, 181, 106–113. https://doi.org/10.1016/j.ecoenv.2019.05.059 Sielfeld, W., Laudien, J., Vargas, M., & Villegas, M. (2010). El Niño induced changes of the coastal fish fauna off northern Chile and implications for ichthyogeography (Cambios de la fauna íctica del norte de Chile inducidos por El Niño y sus implicancias en la ictiogeografía). Revista de Biología Marina y Oceanografía, 45(S1), 705–722. Smale, D. A., Wernberg, T., Oliver, E. C. J., Thomsen, M., Harvey, B. P., Straub, S. C., Burrows, M. T., Alexander, L. V., Benthuysen, J. A., Donat, M. G., Feng, M., Hobday, A. J., Holbrook, N. J., Perkins-Kirkpatrick, S. E., Scannell, H. A., Sen Gupta, A., Payne, B. L., & Moore, P. J. (2019). Marine heatwaves threaten global biodiversity and the provision of ecosystem services. Nature Climate Change, 9(4), 306–312. https://doi.org/10.1038/s41558-019-0412-1 Sokolova, I. M. (2013). Energy-Limited Tolerance to Stress as a Conceptual Framework to Integrate the Effects of Multiple Stressors. Integrative and Comparative Biology, 53(4), 597–608. https://doi.org/10.1093/icb/ict028 128  Spicer, J. I. (2014). What can an ecophysiological approach tell us about the physiological responses of marine invertebrates to hypoxia? Journal of Experimental Biology, 217(1), 46–56. https://doi.org/10.1242/jeb.090365 Stewart, J. S., Hazen, E. L., Bograd, S. J., Byrnes, J. E. K., Foley, D. G., Gilly, W. F., Robison, B. H., & Field, J. C. (2014). Combined climate- and prey-mediated range expansion of Humboldt squid (Dosidicus gigas), a large marine predator in the California Current System. Global Change Biology, 20(6), 1832–1843. https://doi.org/10.1111/gcb.12502 Stock, C. A., Alexander, M. A., Bond, N. A., Brander, K. M., Cheung, W. W. L., Curchitser, E. N., Delworth, T. L., Dunne, J. P., Griffies, S. M., Haltuch, M. A., Hare, J. A., Hollowed, A. B., Lehodey, P., Levin, S. A., Link, J. S., Rose, K. A., Rykaczewski, R. R., Sarmiento, J. L., Stouffer, R. J., … Werner, F. E. (2011). On the use of IPCC-class models to assess the impact of climate on Living Marine Resources. Progress in Oceanography, 88(1), 1–27. https://doi.org/10.1016/j.pocean.2010.09.001 Stock, C. A., Cheung, W. W. L., Sarmiento, J. L., & Sunderland, E. M. (2019). Chapter 2 - Changing ocean systems: A short synthesis. In A. M. Cisneros-Montemayor, W. W. L. Cheung, & Y. Ota (Eds.), Predicting Future Oceans (pp. 19–34). Elsevier. https://doi.org/10.1016/B978-0-12-817945-1.00002-2 Stramma, L., Prince, E. D., Schmidtko, S., Luo, J., Hoolihan, J. P., Visbeck, M., Wallace, D. W. R., Brandt, P., & Körtzinger, A. (2012). Expansion of oxygen minimum zones may reduce available habitat for tropical pelagic fishes. Nature Climate Change, 2(1), 33–37. https://doi.org/10.1038/nclimate1304 Sumaila, U. R., Cheung, W. W. L., Lam, V. W. Y., Pauly, D., & Herrick, S. (2011). Climate change impacts on the biophysics and economics of world fisheries. Nature Climate Change, 1(9), 449–456. https://doi.org/10.1038/nclimate1301 Sumaila, U. R., & Tai, T. C. (2020). End Overfishing and Increase the Resilience of the Ocean to Climate Change. Frontiers in Marine Science, 7. https://doi.org/10.3389/fmars.2020.00523 129  Sunday, J. M., Bates, A. E., & Dulvy, N. K. (2011). Global analysis of thermal tolerance and latitude in ectotherms. Proceedings of the Royal Society B: Biological Sciences, 278(1713), 1823–1830. https://doi.org/10.1098/rspb.2010.1295 Tapia García, M., 1997. Diversidad dinámica y patrones reproductivos en la comunidad de peces demersales del Golfo de Tehuantepec (Bases de datos SNIB-CONABIO). Universidad Autónoma Metropolitana, Unidad Iztapalapa, México, D.F. Thuiller, W., Georges, D., Engler, R., & Breiner, F. (2016). biomod2: Ensemble Platform for Species Distribution Modeling (3.3-7) [Computer software]. https://cran.r-project.org/web/packages/biomod2/biomod2.pdf Tiano, L., Garcia-Robledo, E., Dalsgaard, T., Devol, A. H., Ward, B. B., Ulloa, O., Canfield, D. E., & Peter Revsbech, N. (2014). Oxygen distribution and aerobic respiration in the north and south eastern tropical Pacific oxygen minimum zones. Deep Sea Research Part I: Oceanographic Research Papers, 94, 173–183. https://doi.org/10.1016/j.dsr.2014.10.001 Tittensor, D. P., Eddy, T. D., Lotze, H. K., Galbraith, E. D., Cheung, W., Barange, M., Blanchard, J. L., Bopp, L., Bryndum-Buchholz, A., Büchner, M., Bulman, C., Carozza, D. A., Christensen, V., Coll, M., Dunne, J. P., Fernandes, J. A., Fulton, E. A., Hobday, A. J., Huber, V., … Walker, N. D. (2018). A protocol for the intercomparison of marine fishery and ecosystem models: Fish-MIP v1.0. Geoscientific Model Development, 11(4), 1421–1442. https://doi.org/10.5194/gmd-11-1421-2018 Urban, M. C. (2019). Projecting biological impacts from climate change like a climate scientist. WIREs Climate Change, 10(4), e585. https://doi.org/10.1002/wcc.585 van Nes, E. H., Arani, B. M. S., Staal, A., van der Bolt, B., Flores, B. M., Bathiany, S., & Scheffer, M. (2016). What Do You Mean, ‘Tipping Point’? Trends in Ecology & Evolution, 31(12), 902–904. https://doi.org/10.1016/j.tree.2016.09.011 van Vuuren, D. P., Edmonds, J., Kainuma, M., Riahi, K., Thomson, A., Hibbard, K., Hurtt, G. C., Kram, T., Krey, V., Lamarque, J.-F., Masui, T., Meinshausen, M., Nakicenovic, N., Smith, S. J., & Rose, S. 130  K. (2011). The representative concentration pathways: An overview. Climatic Change, 109(1), 5. https://doi.org/10.1007/s10584-011-0148-z Vergés, A., Steinberg, P. D., Hay, M. E., Poore, A. G. B., Campbell, A. H., Ballesteros, E., Heck, K. L., Booth, D. J., Coleman, M. A., Feary, D. A., Figueira, W., Langlois, T., Marzinelli, E. M., Mizerek, T., Mumby, P. J., Nakamura, Y., Roughan, M., van Sebille, E., Gupta, A. S., … Wilson, S. K. (2014). The tropicalization of temperate marine ecosystems: Climate-mediated changes in herbivory and community phase shifts. Proceedings of the Royal Society B: Biological Sciences, 281(1789), 20140846. https://doi.org/10.1098/rspb.2014.0846 Wabnitz, C. C. C., Lam, V. W. Y., Reygondeau, G., Teh, L. C. L., Al-Abdulrazzak, D., Khalfallah, M., Pauly, D., Palomares, M. L. D., Zeller, D., & Cheung, W. W. L. (2018). Climate change impacts on marine biodiversity, fisheries and society in the Arabian Gulf. PLOS ONE, 13(5), e0194537. https://doi.org/10.1371/journal.pone.0194537 Wang, Y., Hu, M., Cheung, S. G., Shin, P. K. S., Lu, W., & Li, J. (2012). Immune parameter changes of hemocytes in green-lipped mussel Perna viridis exposure to hypoxia and hyposalinity. Aquaculture, 356–357, 22–29. https://doi.org/10.1016/j.aquaculture.2012.06.001 Watters, G. M., Olson, R. J., Francis, R. C., Fiedler, P. C., Polovina, J. J., Reilly, S. B., Aydin, K. Y., Boggs, C. H., Essington, T. E., Walters, C. J., & Kitchell, J. F. (2003). Physical forcing and the dynamics of the pelagic ecosystem in the eastern tropical Pacific: Simulations with ENSO-scale and global-warming climate drivers. Canadian Journal of Fisheries and Aquatic Sciences, 60(9), 1161–1175. https://doi.org/10.1139/f03-100 Webb, E. L., Friess, D. A., Krauss, K. W., Cahoon, D. R., Guntenspergen, G. R., & Phelps, J. (2013). A global standard for monitoring coastal wetland vulnerability to accelerated sea-level rise. Nature Climate Change, 3(5), 458–465. https://doi.org/10.1038/nclimate1756 Wehrtmann, I. S., Arana, P. M., Barriga, E., Gracia, A., & Pezzuto, P. R. (2012). Pesquerías de camarones de aguas profundas en América Latina: Una revisión. Latin American Journal of Aquatic Research, 40(SPECISSUE), 497–535. https://doi.org/103856/vol40-issue3-fulltext-2 131  Wehrtmann, I. S., & Nielsen Munoz, V. (2009). The deep water fishery along the Pacific coast of Costa Rica, Central America. Latin American Journal of Aquatic Research, 37(3), 543–554. https://doi.org/10.3856/vol37-issue3-fulltext-19 Wilson, K. L., Tittensor, D. P., Worm, B., & Lotze, H. K. (2020). Incorporating climate change adaptation into marine protected area planning. Global Change Biology, 26(6), 3251–3267. https://doi.org/10.1111/gcb.15094 Wishner, K. F., Seibel, B. A., Roman, C., Deutsch, C., Outram, D., Shaw, C. T., Birk, M. A., Mislan, K. a. S., Adams, T. J., Moore, D., & Riley, S. (2018). Ocean deoxygenation and zooplankton: Very small oxygen differences matter. Science Advances, 4(12), eaau5180. https://doi.org/10.1126/sciadv.aau5180 Wisz, M. S., Hijmans, R. J., Li, J., Peterson, A. T., Graham, C. H., & Guisan, A. (2008). Effects of sample size on the performance of species distribution models. Diversity and Distributions, 14(5), 763–773. https://doi.org/10.1111/j.1472-4642.2008.00482.x Woodworth-Jefcoats, P. A., Blanchard, J. L., & Drazen, J. C. (2019). Relative Impacts of Simultaneous Stressors on a Pelagic Marine Ecosystem. Frontiers in Marine Science, 6. https://doi.org/10.3389/fmars.2019.00383 Yang, S., Li, Z., Yu, J.-Y., Hu, X., Dong, W., & He, S. (2018). El Niño-Southern Oscillation and its impact in the changing climate. National Science Review, 5(6), 840–857. https://doi.org/10.1093/nsr/nwy046 Yu, J.-Y., & Kao, H.-Y. (2007). Decadal changes of ENSO persistence barrier in SST and ocean heat content indices: 1958–2001. Journal of Geophysical Research: Atmospheres, 112(D13). https://doi.org/10.1029/2006JD007654 Zapata Padilla, L. A. (2002). Effects of El Niño and La Niña on the Fisheries of the Colombian Pacific. Investigaciones Marinas, 30(1), 205–206. https://doi.org/10.4067/S0717-71782002030100090    132  Appendices Appendix A  Chapter 2: Climate change impacts on living marine resources in the Eastern Tropical Pacific A.1 Species caught by the coastal small-scale, large pelagics, small pelagics, shrimp-trawl bycatch and shrimp-trawl fisheries in the tropical Eastern Pacific.  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Acanthocarpus alexandri   X   Acanthurus xanthopterus X  X   Achirus klunzingeri X  X   Achirus mazatlanus X  X   Achirus scutum X  X   Aetobatus narinari X  X   Albula esuncula X  X   Albula nemoptera X  X   Albula pacifica   X   Albula vulpes X  X   Alectis ciliaris X  X   Alopias pelagicus X X    Alopias superciliosus X X    Alopias vulpinus X X    Alphestes immaculatus X  X   Alphestes multiguttatus X  X   Aluterus monoceros X  X   Aluterus scriptus X  X   Anchoa argentivittata X  X  X Anchoa eigenmannia X  X  X Anchoa ischana X  X  X Anchoa lucida X  X  X Anchoa mundeola X  X  X Anchoa nasus X  X  X Anchoa panamensis X  X  X Anchoa spinifer X  X  X 133  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Anchoa starksi X  X  X Anchoa walkeri X  X  X Anchovia macrolepidota X  X  X Ancylopsetta dendritica X  X   Anisotremus davidsonii X  X   Anisotremus interruptus X  X   Aphos porosus   X   Apogon pacificus   X   Apogon retrosella   X   Argentina aliceae X  X   Argentina sialis   X   Argopecten ventricosus   X   Ariopsis guatemalensis X  X   Ariopsis seemanni X  X   Ariosoma gilberti   X   Ariosoma prorigerum   X   Arothron hispidus X  X   Astroscopus zephyreus   X   Atherinella argentea   X   Atractoscion nobilis X  X   Auxis rochei X     Bagre panamensis X  X   Bagre pinnimaculatus X  X   Bairdiella armata X  X   Bairdiella ensifera X  X   Bairdiella icistia X  X   Balistes polylepis X  X   Bathycongrus macrurus   X   Bathygobius andrei   X   Batrachoides boulengeri   X   Batrachoides waltersi X  X   Bellator gymnostethus X  X   Bellator loxias   X   134  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Bellator xenisma   X   Bodianus diplotaenia X     Bollmannia chlamydes   X   Bollmannia ocellata   X   Bollmannia stigmatura   X   Bothus leopardinus   X   Bregmaceros bathymaster   X   Brotula clarkae X  X   Brotula ordwayi X  X   Calamus brachysomus X  X   Callinectes arcuatus X  X   Callinectes bellicosus X  X   Canthidermis maculata X  X   Carangoides otrynter X X X   Caranx caballus X  X   Caranx lugubris X X X   Caranx melampygus X X X   Caranx sexfasciatus X  X   Caranx vinctus X  X   Carcharhinus falciformis X X    Carcharhinus galapagensis X X    Carcharhinus leucas X  X   Carcharhinus limbatus X X    Carcharhinus longimanus X X    Carcharhinus obscurus X X    Carcharhinus porosus X  X   Cathorops dasycephalus X  X   Cathorops fuerthii X  X   Cathorops multiradiatus X  X   Caulolatilus affinis X  X   Caulolatilus princeps X  X   Centropomus armatus X  X   Centropomus medius X  X   135  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Centropomus nigrescens X  X   Centropomus robalito X  X   Centropomus undecimalis X  X   Centropomus unionensis X  X   Centropomus viridis X  X   Cephalopholis panamensis X  X   Cetengraulis mysticetus X  X   Cetorhinus maximus  X    Chaetodipterus zonatus X  X   Chaetodon humeralis   X   Chanos chanos X  X   Cheilotrema saturnum X  X   Cherublemma emmelas   X   Chilara taylori   X   Chilomycterus reticulatus   X   Chione undatella X  X   Chlorophthalmus mento   X  X Chloroscombrus orqueta X  X  X Citharichthys fragilis X  X   Citharichthys gilberti X  X   Citharichthys gordae   X   Citharichthys platophrys X  X   Citharichthys sordidus X  X   Citharichthys stigmaeus X  X   Citharichthys xanthostigma X  X   Coelorinchus canus   X   Coelorinchus scaphopsis X  X   Cookeolus japonicus X  X   Coryphaena equiselis  X    Coryphaena hippurus  X    Cyclopsetta panamensis X  X   Cyclopsetta querna X  X   Cynoponticus coniceps X  X   136  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Cynoscion albus X  X   Cynoscion nannus X  X   Cynoscion parvipinnis X  X   Cynoscion phoxocephalus X  X   Cynoscion reticulatus X  X   Cynoscion squamipinnis X  X   Cynoscion stolzmanni X  X   Cynoscion xanthulus X  X   Daector dowi   X   Dasyatis dipterura X  X   Dasyatis longa X  X   Decapterus macarellus X X X   Decapterus macrosoma X  X   Decapterus muroadsi X  X   Decodon melasma X  X   Dermatolepis dermatolepis X     Diapterus aureolus X  X   Diapterus brevirostris X  X   Diapterus peruvianus X  X   Diodon eydouxii X  X   Diodon holocanthus X  X   Diodon hystrix X  X   Diplectrum eumelum X  X   Diplectrum euryplectrum X  X   Diplectrum labarum X  X   Diplectrum macropoma X  X   Diplectrum maximum   X   Diplectrum pacificum X  X   Diplectrum rostrum   X   Diplobatis ommata   X   Echidna nocturna   X   Elattarchus archidium   X   Elops affinis X  X   137  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Engyophrys sanctilaurentii X  X   Epinephelus analogus X  X   Epinephelus cifuentesi X  X   Epinephelus labriformis X  X   Epinephelus quinquefasciatus X  X   Ethusa ciliatifrons   X   Etropus crossotus X  X   Etropus peruvianus X  X   Eucinostomus argenteus X  X   Eucinostomus currani X  X   Eucinostomus dowii X  X   Eucinostomus entomelas X  X   Eucinostomus gracilis X  X   Eugerres lineatus   X   Euphylax robustus   X   Euthynnus affinis X X    Euthynnus lineatus X X    Farfantepenaeus brevirostris X  X X  Farfantepenaeus californiensis X  X X  Fistularia commersonii X  X   Fistularia corneta X  X   Fowlerichthys avalonis   X   Galeocerdo cuvier X     Galeorhinus galeus X  X   Genyatremus dovii X  X   Genyatremus pacifici X  X   Gerres cinereus X  X   Ginglymostoma cirratum X  X   Gnathanodon speciosus X  X   Gnathophis cinctus   X   Gobiomorus maculatus X  X   Gobiosoma paradoxum   X   Gymnothorax dovii   X   138  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Gymnothorax equatorialis   X   Gymnothorax panamensis   X   Gymnothorax phalarus   X   Gymnura marmorata X  X   Haemulon californiensis X  X   Haemulon flaviguttatum X  X   Haemulon maculicauda X  X   Haemulon scudderii X  X   Haemulon sexfasciatum X  X   Haemulon steindachneri X  X   Haemulopsis axillaris X  X   Haemulopsis elongatus X  X   Haemulopsis leuciscus X  X   Halichoeres chierchiae X  X   Halichoeres nicholsi X     Halichoeres notospilus X     Halichoeres semicinctus X  X   Haliporoides diomedeae X  X X  Harengula thrissina X  X  X Hemanthias peruanus X  X   Hemanthias signifer X  X   Hemicaranx leucurus X  X  X Hemicaranx zelotes X  X  X Hemiramphus saltator X  X   Hepatus kossmanni   X   Heterocarpus affinis    X  Heterocarpus hostilis    X  Heterocarpus vicarius    X  Heterodontus mexicanus X  X   Hippoglossina bollmani X  X   Hippoglossina stomata X  X   Hippoglossina tetrophthalma X  X   Holacanthus passer X     139  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Holothuria impatiens X  X   Hoplopagrus guentherii X  X   Hyporhamphus naos X     Hyporhamphus snyderi X     Hyporthodus acanthistius X  X   Ilisha fuerthii X  X  X Isopisthus remifer X  X   Istiompax indica  X    Istiophorus platypterus  X    Isurus oxyrinchus  X    Isurus paucus  X    Kajikia audax  X    Katsuwonus pelamis X X X   Kyphosus analogus X  X   Kyphosus elegans X  X   Kyphosus ocyurus X  X   Kyphosus vaigiensis X     Lagocephalus lagocephalus X  X   Larimus acclivis X  X   Larimus argenteus X  X   Larimus effulgens X  X   Larimus pacificus X  X   Lepidopus fitchi X  X   Lepophidium microlepis   X   Lepophidium negropinna X  X   Lepophidium pardale   X   Lile gracilis X  X  X Lile stolifera X  X  X Litopenaeus occidentalis X   X  Litopenaeus stylirostris X   X  Lobotes pacificus X  X   Lobotes surinamensis X  X   Lonchopisthus sinuscalifornicus   X   140  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Lophiodes spilurus   X   Lutjanus aratus X  X   Lutjanus argentiventris X  X   Lutjanus colorado X  X   Lutjanus guttatus X  X   Lutjanus inermis X  X   Lutjanus jordani X  X   Lutjanus novemfasciatus X  X   Lutjanus viridis X  X   Makaira nigricans  X    Manta birostris  X    Megalops atlanticus X     Megapitaria aurantiaca X  X   Menticirrhus elongatus X  X   Menticirrhus nasus X  X   Menticirrhus panamensis X  X   Menticirrhus undulatus X  X   Merluccius angustimanus X  X   Merluccius gayi X  X   Merluccius productus X  X   Microgobius erectus   X   Microlepidotus brevipinnis X  X   Microlepidotus inornatus X  X   Micropogonias altipinnis X  X   Micropogonias ectenes X  X   Modiolus americanus X  X   Modiolus capax X  X   Monolene asaedae   X   Monolene maculipinna X  X   Mugil cephalus X  X   Mugil curema X  X   Mulloidichthys dentatus X  X   Mustelus californicus X  X   141  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Mustelus dorsalis X  X   Mustelus henlei X  X   Mustelus lunulatus X  X   Mycteroperca xenarcha X  X   Myliobatis californica X  X   Myrophis vafer   X   Narcine entemedor X  X   Narcine vermiculatus X  X   Nasolamia velox X     Nebris occidentalis X  X   Negaprion brevirostris X  X   Nematistius pectoralis X  X   Nematocarcinus agassizii    X  Nemichthys scolopaceus X  X   Neobythites stelliferoides   X   Neoopisthopterus tropicus X  X   Nezumia stelgidolepis X  X   Nicholsina denticulata X     Nodipecten subnodosus X  X   Notarius osculus X  X   Notarius troschelii X  X   Occidentarius platypogon X  X   Octopus vulgaris X  X   Odontoscion xanthops   X   Ogilbia ventralis   X   Oligoplites altus X  X  X Oligoplites refulgens X  X  X Oligoplites saurus X  X  X Ophichthus frontalis X  X   Ophichthus remiger   X   Ophichthus triserialis   X   Ophichthus zophochir   X   Ophidion fulvum   X   142  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Ophioscion scierus X  X   Ophioscion strabo X  X   Ophioscion typicus   X   Opisthonema bulleri X  X  X Opisthonema libertate X  X  X Opisthonema medirastre X  X  X Opisthopterus equatorialis X  X  X Opistognathus punctatus   X   Orthopristis cantharinus X  X   Orthopristis chalceus X  X   Orthopristis reddingi X  X   Otophidium indefatigabile   X   Panulirus gracilis    X  Paralabrax auroguttatus X  X   Paralabrax humeralis X  X   Paralabrax loro X  X   Paralabrax maculatofasciatus X  X   Paralabrax nebulifer X  X   Paralichthys adspersus X  X   Paralichthys aestuarius X  X   Paralichthys woolmani X  X   Paralonchurus dumerilii X  X   Paralonchurus goodei X  X   Paralonchurus petersii X  X   Paranthias colonus X  X   Pareques viola X  X   Pasiphaea americana   X   Peprilus snyderi X  X   Perissias taeniopterus   X   Peristedion barbiger   X   Physiculus nematopus   X   Physiculus rastrelliger   X   Pinctada mazatlanica X  X   143  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Platymera gaudichaudii   X   Plesionika trispinus   X X  Pleuroncodes monodon   X X  Pleuroncodes planipes   X X  Pleuronichthys ocellatus X  X   Pleuronichthys ritteri   X   Pleuronichthys verticalis   X   Pliosteostoma lutipinnis X  X   Polydactylus approximans X  X   Polydactylus opercularis X  X   Polylepion cruentum X  X   Pomadasys bayanus X  X   Pomadasys branickii X  X   Pomadasys macracanthus X  X   Pomadasys panamensis X  X   Pontinus furcirhinus   X   Pontinus vaughani X  X   Porichthys analis   X   Porichthys greenei   X   Porichthys margaritatus X  X   Porichthys myriaster   X   Portunus asper   X   Portunus xanthusii   X   Prionace glauca  X    Prionotus albirostris   X   Prionotus birostratus   X   Prionotus ruscarius X  X   Prionotus stephanophrys X  X   Pristigenys serrula X  X   Pronotogrammus eos X  X   Protrachypene precipua X   X  Psenes sio X  X   Pseudobalistes naufragium X  X   144  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Pseudupeneus grandisquamis X  X   Pteria sterna X  X   Pteroplatytrygon violacea  X    Raja equatorialis   X   Raja inornata X  X   Raja velezi X  X   Rhinobatos glaucostigma X  X   Rhinobatos leucorhynchus X  X   Rhinobatos productus X  X   Rhinoptera steindachneri X  X   Rhizoprionodon longurio X  X   Rimapenaeus byrdi X   X  Rimapenaeus faoe X  X   Rimapenaeus fuscina X  X   Rimapenaeus pacificus X  X   Roncador stearnsii X  X   Rypticus bicolor   X   Rypticus nigripinnis   X   Sarda chiliensis X  X  X Sarda orientalis X  X  X Sardinops sagax X  X  X Scarus compressus X  X   Scarus ghobban X  X   Scarus rubroviolaceus X  X   Scomber japonicus X  X  X Scomberomorus concolor X    X Scomberomorus sierra X  X  X Scorpaena guttata X  X   Scorpaena histrio X  X   Scorpaena mystes X  X   Scorpaena russula X  X   Scorpaena sonorae   X   Sebastes caurinus X     145  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Sebastes chlorostictus X     Sebastes ensifer X     Sebastes macdonaldi X     Sebastes melanostomus X     Sebastes miniatus X     Selene brevoortii X  X  X Selene orstedii X  X  X Selene peruviana X  X  X Seriola peruana X  X  X Serranus psittacinus X  X   Serrivomer sector X     Sicyonia aliaffinis   X   Sicyonia disdorsalis X  X   Sicyonia penicillata X  X   Sicyonia picta   X   Solenocera agassizii X   X  Solenocera florea X   X  Solenocera mutator   X X  Sphoeroides angusticeps X     Sphoeroides kendalli   X   Sphoeroides lispus   X   Sphoeroides sechurae X  X   Sphoeroides trichocephalus   X   Sphyraena argentea X  X  X Sphyraena ensis X  X  X Sphyraena idiastes X    X Sphyrna corona X  X   Sphyrna lewini X X X   Sphyrna mokarran X X    Sphyrna tiburo X  X   Sphyrna zygaena X X X   Spondylus crassisquama X  X   Squilla biformis   X   146  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Squilla mantoidea   X   Stellifer chrysoleuca X  X   Stellifer ephelis X  X   Stellifer ericymba X  X   Stellifer illecebrosus X  X   Stellifer oscitans X  X   Stellifer zestocarus X  X   Syacium ovale   X   Symphurus atramentatus X  X   Symphurus atricaudus X  X   Symphurus callopterus X  X   Symphurus chabanaudi X  X   Symphurus elongatus   X   Symphurus fasciolaris   X   Symphurus leei   X   Symphurus melanurus   X   Symphurus melasmatotheca   X   Symphurus oligomerus   X   Synchiropus atrilabiatus   X   Synodus evermanni   X   Synodus lacertinus   X   Synodus lucioceps   X   Synodus scituliceps X  X   Synodus sechurae   X   Tetrapturus angustirostris  X    Thunnus alalunga  X    Thunnus albacares  X    Thunnus obesus  X    Thunnus orientalis  X    Torpedo peruana   X   Trachinotus kennedyi X  X   Trachinotus paitensis X  X   Trachinotus rhodopus X  X   147  Species Coastal small-scale fisheries Large-pelagics fisheries Shrimp-trawl bycatch Shrimp-trawl fisheries Small-pelagics fisheries Trichiurus lepturus X  X   Trinectes fonsecensis X  X   Triplofusus princeps X  X   Tylosurus pacificus X X X   Umbrina roncador X  X   Umbrina xanti X  X   Urobatis halleri X  X   Urolophus maculatus X  X   Urotrygon aspidura   X   Urotrygon chilensis   X   Urotrygon munda   X   Urotrygon nana   X   Urotrygon rogersi X  X   Xiphias gladius  X    Xiphopenaeus kroyeri X   X  Zalieutes elater   X   Zapteryx exasperata X   X      Sources:  SAGARPA. (2010). Carta Nacional Pesquera. Secretaría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Acuacultura. Diario Oficial De La Federación, México, 1–688; SAGARPA. (2012). Carta Nacional Pesquera. Secretaría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Acuacultura. Diario Oficial De La Federación, México 1–236.  Baile, D. S., Lucas, A. R., Ostaiza, A. Z., and Gracia, J. Á. (2014). Monitoreo del desembarco de pesca artesanal en el Estuario del río Chone, Ecuador, entre octubre 2013 y enero 2014. La Técnica, (12), 26-37.  Beatriz Naranjo unpublished data  Box, S. J., and Bonilla, R. S. (2009). Evaluación de las Prácticas Pesqueras en Pesquerías de Pequeña Escala del Golfo de Fonseca, Honduras. Recomendaciones para el Manejo.  148  Campos, J. A. (1986). Fauna de acompañamiento del camarón en el Pacífico de Costa Rica. Revista de Biología Tropical, 34(2), 185-197.  Chicas-Batres, F.A., J.A. González-Leiva and W.E. Ramírez-Vásquez. (2012). Ecología básica de los peces del Golfo de Fonseca: Bases para el manejo de la pesca artesanal. Escuela deBiología, Facultad de Ciencias Naturales y Matemática, Universidad de El Salvador.35 pp  Tayler Clarke unpublished data  De la Rosa Meza, K. (2005). Fauna de acompañamiento de camarón en Bahía Magdalena, BCS México (Doctoral dissertation, Instituto Politécnico Nacional. Centro Interdisciplinario de Ciencias Marinas).  Fuentes Rivera, C. I., and Hernández Rodríguez, N. R. (2004). Distribución y abundancia de la íctiofauna con importancia comercial asociada a la pesca de arrastre de camarones peneidos (Penaeus stylirrostris, P. vannamei, P. occidentalis, P. californiensis y P. brevirrostris) en la costa salvadoreña (Doctoral dissertation, Universidad de El Salvador).  CIPA-INPESCA (2007). Guía Indicativa. Nicaragua y el Sector Pesquero. Actualización al año 2007. http://www.inpesca.gob.ni/  Gutierrez, R. and Eslaquit, B. (2009). Resultados del primer monitoreo realizado al camarón costero del Pacífico de Nicaragua. Noviembre - Diciembre 2008. Centro de Investigaciones Pesqueras y Acuícola (CIPA), INPESCA Nicaragua.  Hernández-Covarrubias, V., Chávez-Herrera. D., Muñoz-Rubí, Melchor Aragón J. M. Villegas Hernández F. (2012). Fauna de acompañamiento de camarón en la ribera adyacente a la boca de Macapule, Navachiste, Sinaloa, 2011. Informe Técnico. INAPESCA  Hernandez-Noguera, L. (2011). Análisis pesquero y socioeconómico del camaron carabali Trachypenaeus byrdi (Burkenroad, 1934) en la parte interna del Golfo de Nicoya, Costa Rica. Maestría en Ciencias Marinas y Biológicas de la Escuela de Ciencias Biológicas de la Universidad Nacional (UNA), Puntarenas, Costa Rica.  Jolon-Morales M.R., Sanchez-Castañeda R., Villagrán-Colón J.C. Mechel C. Kinh H. A. (2005). Estudiosobre los Recursos Pesqueros (de escama) en el Litoral Pacifico y Mar Caribe de Guatemala.Guatemala: UNIPESCA-AECI. 128 p 149   López-Martínez, J., Herrera-Valdivia, E., Rodríguez-Romero, J., and Hernández-Vázquez, S. (2010). Peces de la fauna de acompañamiento en la pesca industrial de camarón en el Golfo de California, México. Revista de Biología Tropical, 58(3), 925-942.  Martínez-Muñoz, M. A. (2012). Estructura y distribución de la comunidad íctica acompañante en la pesca del camarón (Golfo de Tehuantepec. Pacífico Oriental, México).  Ministerio de Agricultura y Ganadería, CENTRO DE DESARROLLO DE LA PESCA Y LA ACUICULTURA, Unidad de Estadística, El Salvador, Estadísticas Pesqueras y Acuícolas, Año 2006, Vol. 33.  Monitoreo del desembarco de pesca artesanal en el Estuario del río Chone, Ecuador, entre octubre 2013 y enero 2014  Nieto Navarro, J. T. (2010). Estructura y organización de la ictiofauna de fondos blandos del sur de Sinaloa: Análisis ecológico y topología de taxa (Doctoral dissertation, Instituto Politécnico Nacional. Centro Interdisciplinario de Ciencias Marinas).  Puentes, V., Madrid, N., and Zapata, L. A. (2007). Composición de la captura en la pesquería del camarón de aguas profundas (Solenocera agassizi Faxon, 1893: Farfantepenaeus californiensis holmes, 1900, y Farfantepenaeus brevirostris Kingsley, 1878) del oceano pacifico colombiano. Gayana (Concepción), 71(1), 84-95.  Ross-Salazar, E., Posada, J. M., Melo, G., Suárez, C., and Ventura-Pozuelo, A. E. (2014). Guía de identificación: peces de importancia comercial en la costa Pacífica de Costa Rica. Fundación MarViva. San José, Costa Rica; Posada, J.M., E. Ross Salazar, G.Melo, N. Sánchez and A.E.   Ventura Pozuelo (2017), Guía de identificación: Peces de importancia comercial en la costa Pacífica de Panamá. Fundación MarViva. Ciudad de Panamá. Panamá, 255 p.; Ross Salazar, E., J.M.Posada, G.Melo, A.Díaz, L.Jaramillo and A.E. Ventura Pozuelo (2017). Guía de identificación: Peces de impotancia comercial en la costa Pacífica de Colombia. Fundación MarViva, Bogotá, Colombia 251.pp   Sánchez Castañeda, R. (2000). Caracterización de la pesca artesanal en el área del humedal Manchón Guamuchal, Ocós, San Marcos y sus opciones de desarrollo local. Tesis Lic. Acuicultura. Guatemala, USAC. 150   Wehrtmann, I. S., and Echeverría-Sáenz, S. (2007). Crustacean fauna (Stomatopoda: Decapoda) associated with the deepwater fishery of Heterocarpus vicarius (Decapoda: Pandalidae) along the Pacific coast of Costa Rica. Revista de Biología Tropical, 55(Supp1). Zapata, L.A., G. Rodríguez, B. Beltrán-León, G. Gómez, A. Cediel, R. Avila and C. Hernández. (1999). Evaluación de recursos demersales por el método de área barrida en el Pacífico colombiano. Inst. Nac. Pes. y Acuic. -INPA-. Boletín Científico 6: 177-226. Tente links no: Google. Similares em: Rede SciELO.  http://www.seaaroundus.org/   151    A.2 Geographic occurrence records for all the 505 modeled marine and invertebrate species. Color scale represents the number of species with presence record in each 0.5-degree longitude by 0.5-degree latitude grid. Areas in grey represent cells with no presence records.    152  A.3 Results of Generalized Linear Models utilized to downscale environmental parameters in the tropical Eastern Pacific. For all models, n was 12229 and p was < 2.2e-16. ESM: Earth system model.   Habitat ESM Dependent variables  Independent variables R2 Demersal GFDL-ESM-2G Sediment Sediment+Latitude*Longitude+Depth 0.99 Demersal GFDL-ESM-2G Temperature Temperature+Latitude*Longitude+Depth 0.74 Demersal GFDL-ESM-2G Salinity Salinity+Latitude*Longitude+Depth 0.69 Demersal GFDL-ESM-2G Oxygen Oxygen+Latitude*Longitude+Depth 0.92 Demersal GFDL-ESM-2G Primary productivity Primary productivity+Latitude*Longitude+Depth 0.56 Pelagic GFDL-ESM-2G Mixed Layer Depth Mixed Layer Depth+Latitude*Longitude 0.89 Pelagic GFDL-ESM-2G Temperature Temperature+Latitude*Longitude 0.84 Pelagic GFDL-ESM-2G Salinity Salinity+Latitude*Longitude 0.70 Pelagic GFDL-ESM-2G Oxygen Oxygen+Latitude*Longitude 0.82 Pelagic GFDL-ESM-2G Primary productivity Primary productivity+Latitude*Longitude 0.55 Demersal IPSL-CM5-MR Sediment Sediment+Latitude*Longitude+Depth 0.99 Demersal IPSL-CM5-MR Temperature Temperature+Latitude*Longitude+Depth 0.79 Demersal IPSL-CM5-MR Salinity Salinity+Latitude*Longitude+Depth 0.70 Demersal IPSL-CM5-MR Oxygen Oxygen+Latitude*Longitude+Depth 0.85 Demersal IPSL-CM5-MR Primary productivity Primary productivity+Latitude*Longitude+Depth 0.65 Pelagic IPSL-CM5-MR Mixed Layer Depth Mixed Layer Depth+Latitude*Longitude 0.89 Pelagic IPSL-CM5-MR Temperature Temperature+Latitude*Longitude 0.86 Pelagic IPSL-CM5-MR Salinity Salinity+Latitude*Longitude 0.77 Pelagic IPSL-CM5-MR Oxygen Oxygen+Latitude*Longitude 0.90 Pelagic IPSL-CM5-MR Primary productivity Primary productivity+Latitude*Longitude 0.65 Demersal MPI-ESM-MR Sediment Sediment+Latitude*Longitude+Depth 0.99 Demersal MPI-ESM-MR Temperature Temperature+Latitude*Longitude+Depth 0.89 Demersal MPI-ESM-MR Salinity Salinity+Latitude*Longitude+Depth 0.78 Demersal MPI-ESM-MR Oxygen Oxygen+Latitude*Longitude+Depth 0.87 Demersal MPI-ESM-MR Primary productivity Primary productivity+Latitude*Longitude+Depth 0.56 Pelagic MPI-ESM-MR Mixed Layer Depth Mixed Layer Depth+Latitude*Longitude 0.87 Pelagic MPI-ESM-MR Temperature Temperature+Latitude*Longitude 0.84 Pelagic MPI-ESM-MR Salinity Salinity+Latitude*Longitude 0.81 Pelagic MPI-ESM-MR Oxygen Oxygen+Latitude*Longitude 0.55 Pelagic MPI-ESM-MR Primary productivity Primary productivity+Latitude*Longitude 0.38 153  A.4 Surface and bottom environmental parameters selected by the Ecological Niche Factor Analysis (ENFA) as important in determining the species environmental niche.  Selected environmental parameters Number of demersal species Number of pelagic species  Temperature 299 206 Salinity 279 79 Oxygen 284 189 Primary Productivity 120 170 pH 270 86 Sediment 38 NA Mixed Layer Depth NA 118    154  A.5 Average Area Under the Curve for each species distribution model (SDM) and Earth system model (ESM) for 505 marine fish and invertebrate species in the Eastern Tropical Pacific.     155  A.6 Percentage of the 505 study species according to the proportional change of habitat suitability in the Pacific Exclusive Economic Zones between 2001-2020 and 2041-2060.  156  Appendix B  Chapter 3: A new metabolic index to understand the impacts of ocean warming and deoxygenation on global marine fisheries resources  B.1 Relationship between the BDMI estimated with oxygen in atm using different temperature dependence parameters (j1,j2) and the physiologically derived metabolic index (Penn et al., 2018). In red I compare the BDMI estimated with the temperature dependence parameters detailed in the methods, in blue the species-specific temperature dependence of the anabolic term compiled by Penn et al., (2018), and in yellow in the temperature dependence of the anabolic and catabolic term based on Penn et al., (2018) and Pauly (2010). The first row shows the relationship between the BDMI and the physiologically derived metabolic index (Penn et al., 2018) for the baseline period (1971-2000). The second row illustrates the changes in BDMI and physiologically derived metabolic index (Penn et al., 2018) by the end of the century (2071-2100) relative to the baseline period (1971-2000).    157  B.2 Relationship between the BDMI estimated with oxygen in atm with metabolic scaling parameter of 0.7 in red and of 0.9 in yellow based on Pauly and Cheung (2018) and the physiologically derived metabolic index (Penn et al., 2018). The first row shows the relationship between the BDMI and the physiologically derived metabolic index (Penn et al., 2018) for the baseline period (1971-2000). The second row illustrates the changes in BDMI and physiologically derived metabolic index (Penn et al., 2018) by the end of the century (2071-2100) relative to the baseline period (1971-2000).   158  B.3 Relationship between the BDMI estimated with oxygen in atm (red) and oxygen in mol/m3 (blue) and the physiologically derived metabolic index (Penn et al., 2018). The first row shows the relationship between the BDMI and the physiologically derived metabolic index (Penn et al., 2018) for the baseline period (1971-2000). The second row illustrates the changes in BDMI and physiologically derived metabolic index (Penn et al., 2018) by the end of the century (2071-2100) relative to the baseline period (1971-2000).   159  B.4 Biological and ecological input parameters necessary to estimate the Biogeographically derived metabolic index (BDMI) for Gadus morhua, Diplodus puntazzo and Callinectes sapidus. I show the parameter values for the standard application of the BDMI presented in the methods and results, as well as for the values of the parameter values to conduct sensitivity analysis that explore the sensitivity of BDMI output to changes in the values of different parameters, including the effect of different metabolic scaling (d), anabolic temperature dependence (j1), temperature dependence (j1,j2), oxygen units (oxygen threshold in mol/m3 and atm) and the percentile used to estimate species oxygen thresholds (p01, p05, p10, p15, p20, p25).   Sensitivity analysis Species  L∞ K LWa LWb d j1 j2 Temperature preference  oxygen threshold  BDMI Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.199242273 BDMI Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.199953679 BDMI Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.163605535 d Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.9 4500 8000 19.4092405 0.199242273 d Callinectes sapidus 20 0.9 0.1284 2.7 0.9 4500 8000 23.7334512 0.199953679 d Gadus morhua 200 0.177 0.00708 3.07 0.9 4500 8000 3.5479608 0.163605535 j1 Gadus morhua 200 0.177 0.00708 3.07 0.7 4874 8000 3.5479608 0.163605535 j1  Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 2669 8000 19.4092405 0.199242273 j1  Callinectes sapidus 20 0.9 0.1284 2.7 0.7 2669 8000 23.7334512 0.199953679 j1,j12 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 2669 4065 19.4092405 0.199242273 j1,j12 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 2669 13759 23.7334512 0.199953679 j1,j12 Gadus morhua 200 0.177 0.00708 3.07 0.7 4874 9609 3.5479608 0.163605535 O2 (mol/m3) Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.205625642 O2 (mol/m3) Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.24596 0.2036782 O2 (mol/m3) Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.257750646            160  Sensitivity analysis Species  L∞ K LWa LWb d j1 j2 Temperature preference  oxygen threshold  p01 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.190370877 p01 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.194543131 p01 Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.15135856 p05 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.196564881 p05 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.198321781 p05 Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.156465892 p10 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.199242273 p10 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.199953679 p10 Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.163605535 p15 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.201165185 p15 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.201696896 p15 Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.170285571 p20 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.201956206 p20 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.202755 p20 Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.174899595 p25 Diplodus puntazzo 62.2 0.36 0.01259 3.03 0.7 4500 8000 19.4092405 0.202641667 p25 Callinectes sapidus 20 0.9 0.1284 2.7 0.7 4500 8000 23.7334512 0.20346057 p25 Gadus morhua 200 0.177 0.00708 3.07 0.7 4500 8000 3.5479608 0.177724862 161  B.5 Comparing the habitat loss (%) and Φcrit projected by the physiologically derived metabolic index (MI, Penn et al., 2018) and different parametrizations of the BDMI: using oxygen partial pressure (atm) and dissolved oxygen concentration (mol/m3); j1, species-specific temperature dependence parameters for the anabolic term compiled by Penn et al., (2018); j1, j2: the same species-specific temperature dependence parameters for the anabolic term and scale it with the catabolic temperature dependence as in Pauly (2010); d: I used a value 0f 0.9 and for standard I use the parameter values described in the methods.   Species Parameters Oxygen BDMI habitat loss % Φcrit MI habitat loss % Φcrit Difference in lost habitat D. puntazzo d atm -16.71 0.93 -17.04 7.54 -0.33   mol/m3 -18.10 0.90 -17.04 7.54 1.06  j1 atm -16.50 0.88 -17.04 7.54 -0.54   mol/m3 -17.50 0.85 -17.04 7.54 0.46  j1, j2 atm -18.15 1.27 -17.04 7.54 1.11   mol/m3 -19.80 1.22 -17.04 7.54 2.76  Standard atm -16.71 1.04 -17.04 7.54 -0.33   mol/m3 -18.10 1.00 -17.04 7.54 1.06 C. sapidus d atm -46.83 1.03 -46.67 3.16 0.16   mol/m3 -47.70 0.99 -46.67 3.16 1.03  j1 atm -47.31 1.35 -46.67 3.16 0.64   mol/m3 -48.21 1.31 -46.67 3.16 1.54  j1, j2 atm -46.69 1.13 -46.67 3.16 0.02   mol/m3 -47.30 1.06 -46.67 3.16 0.63  Standard atm -46.83 1.28 -46.67 3.16 0.16   mol/m3 -47.70 1.23 -46.67 3.16 1.03 G. morhua d atm -13.01 1.00 -12.47 4.25 0.54   mol/m3 -13.65 0.93 -12.47 4.25 1.18  j1 atm -12.33 1.27 -12.47 4.25 -0.14   mol/m3 -12.27 1.19 -12.47 4.25 -0.20  j1, j2 atm -12.12 1.16 -12.47 4.25 -0.35   mol/m3 -15.68 1.08 -12.47 4.25 3.21  Standard atm -13.01 1.25 -12.47 4.25 0.54     mol/m3 -13.65 1.16 -12.47 4.25 1.18  162  B.6 Linear regressions between the BDMI and the physiologically derived metabolic index (Penn et al., 2018) for the baseline period (1971-2000). For j1, I use species-specific temperature dependence for the anabolic term (Penn et al., 2018). For j1, j2, I use the same species-specific temperature dependence for the anabolic term and scale it with the catabolic temperature dependence as in Pauly (2010). For d, I use 0.9 as in Pauly and Cheung (2018) and for standard I use BDMI computations detailed in the methods.  Species Parameters R2 Equation p-value D. puntazzo d 1.00 BDMI=-0.33+0.15* Physiologically derived metabolic index <0.001 D. puntazzo j1 0.99 BDMI=-1.26+0.28* Physiologically derived metabolic index <0.001 D. puntazzo j1, j2 0.98 BDMI=0.67+0.08* Physiologically derived metabolic index <0.001 D. puntazzo Standard 1.00 BDMI=-0.43+0.19* Physiologically derived metabolic index <0.001 C. sapidus d 0.99 BDMI=0.65+0.13* Physiologically derived metabolic index <0.001 C. sapidus j1 0.97 BDMI=1.15+0.07* Physiologically derived metabolic index <0.001 C. sapidus j1, j2 1.00 BDMI=-0.39+0.46* Physiologically derived metabolic index <0.001 C. sapidus Standard 0.99 BDMI=0.77+0.17* Physiologically derived metabolic index <0.001 G. morhua d 0.99 BDMI=0.34+0.16* Physiologically derived metabolic index <0.001 G. morhua j1 0.98 BDMI=0.53+0.18 Physiologically derived metabolic index <0.001 G. morhua j1, j2 1.00 BDMI=0.04+0.26* Physiologically derived metabolic index <0.001 G. morhua Standard 0.99 BDMI=0.42+0.20* Physiologically derived metabolic index <0.001 163  B.7 Linear regressions between the change in BDMI and the physiologically derived metabolic index (Penn et al., 2018) by end of century (2071-2100) relative to the baseline period (1971-2000). For j1, I use species-specific temperature dependence for the anabolic term (Penn et al., 2018), for j1, j2, I use the same species-specific temperature dependence for the anabolic term and scale it with the catabolic temperature dependence as in Pauly (2010), for d, I use 0.9 by Pauly and Cheung (2018) and for standard I use BDMI computations detailed in the methods.  Species Combination R Equation p-value Diplodus puntazzo d 0.99 BDMI=-0.64+1.21* Physiologically derived metabolic index <0.001  j1 0.93 BDMI=-1.75+1.69* Physiologically derived metabolic index <0.001  j1, j2 0.91 BDMI=0.84+0.65* Physiologically derived metabolic index <0.001  Standard 0.99 BDMI=-0.66+1.20* Physiologically derived metabolic index <0.001 Callinectes sapidus d 0.99 BDMI=0.59+0.62* Physiologically derived metabolic index <0.001  j1 0.88 BDMI=1.32+0.37* Physiologically derived metabolic index <0.001  j1, j2 1.00 BDMI=-0.64+1.14* Physiologically derived metabolic index <0.001  Standard 0.99 BDMI=0.52+0.61* Physiologically derived metabolic index <0.001 Gadus morhua d 0.96 BDMI=-0.23+0.70* Physiologically derived metabolic index <0.001  j1 0.93 BDMI=-0.30+0.62* Physiologically derived metabolic index <0.001  j1, j2 1.00 BDMI=-0.01+0.97* Physiologically derived metabolic index <0.001   Standard 0.97 BDMI=-0.04+0.71* Physiologically derived metabolic index <0.001   164  B.8 Linear regressions between the physiologically derived metabolic index and the Biogeographically derived metabolic index (BDMI) for the baseline period (1971-2000). I tested the sensitivity of BDMI output to the selection of the percentile that defines species oxygen thresholds by comparing the relationship between both metabolic indices across 6 different percentiles (p01 = 1st percentile, p05= 5th percentile, p10 = 10th percentile, p15=15th percentile, p20= 20th percentile, p25=25th percentile).  Species Parameters R2 Equation p-value D. puntazzo p01 1 BDMI=-0.44+0.01* Physiologically derived metabolic index <0.001 D. puntazzo p05 1 BDMI=-0.41+0.19* Physiologically derived metabolic index <0.001 D. puntazzo p10 1 BDMI=-0.42+0.19* Physiologically derived metabolic index <0.001 D. puntazzo p15 1 BDMI=-0.41+0.19* Physiologically derived metabolic index <0.001 D. puntazzo p20 1 BDMI=-0.40+0.19* Physiologically derived metabolic index <0.001 D. puntazzo p25 1 BDMI=-0.42+0.19* Physiologically derived metabolic index <0.001 C. sapidus p01 0.99 BDMI=0.83+0.17* Physiologically derived metabolic index <0.001 C. sapidus p05 0.99 BDMI=0.78+0.17* Physiologically derived metabolic index <0.001 C. sapidus p10 0.99 BDMI=-0.77+0.17* Physiologically derived metabolic index <0.001 C. sapidus p15 1 BDMI=0.73+0.18* Physiologically derived metabolic index <0.001 C. sapidus p20 0.99 BDMI=0.74+0.17* Physiologically derived metabolic index <0.001 C. sapidus p25 0.99 BDMI=0.75+0.17* Physiologically derived metabolic index <0.001 G. morhua p01 0.99 BDMI=0.47+0.21* Physiologically derived metabolic index <0.001 G. morhua p05 0.99 BDMI=0.44+0.20* Physiologically derived metabolic index <0.001 G. morhua p10 0.99 BDMI=0.41+0.20* Physiologically derived metabolic index <0.001 G. morhua p15 0.99 BDMI=0.40+0.19 Physiologically derived metabolic index <0.001 G. morhua p20 0.99 BDMI=0.39+0.18* Physiologically derived metabolic index <0.001 G. morhua p25 0.99 BDMI=0.39+0.18* Physiologically derived metabolic index <0.001        165  B.9 Linear regressions between the percent change in BDMI and the physiologically derived metabolic index (Penn et al., 2018) by end of century (2071-2100) relative to the baseline period (1971-2000). I tested the sensitivity of BDMI output to the selection of the percentile that defines species oxygen thresholds by comparing the relationship between both metabolic indices across 6 different percentiles (p01 = 1st percentile, p05= 5th percentile, p10 = 10th percentile, p15=15th percentile, p20= 20th percentile, p25=25th percentile).  Species Parameters R2 Equation p-value D. puntazzo p01 0.99 BDMI=-0.56+1.22* Physiologically derived metabolic index <0.001 D. puntazzo p05 0.99 BDMI=-0.71+1.2* Physiologically derived metabolic index <0.001 D. puntazzo p10 0.99 BDMI=-0.51+1.22* Physiologically derived metabolic index <0.001 D. puntazzo p15 0.99 BDMI=-0.54+1.22* Physiologically derived metabolic index <0.001 D. puntazzo p20 0.99 BDMI=-0.57+1.22* Physiologically derived metabolic index <0.001 D. puntazzo p25 0.99 BDMI=-0.58+1.21* Physiologically derived metabolic index <0.001 C. sapidus p01 0.99 BDMI=0.77+0.63* Physiologically derived metabolic index <0.001 C. sapidus p05 0.99 BDMI=0.70+0.63* Physiologically derived metabolic index <0.001 C. sapidus p10 0.99 BDMI=-0.68+0.62* Physiologically derived metabolic index <0.001 C. sapidus p15 1 BDMI=0.60+0.62* Physiologically derived metabolic index <0.001 C. sapidus p20 0.99 BDMI=1.04+0.64* Physiologically derived metabolic index <0.001 C. sapidus p25 1 BDMI=0.57+0.62* Physiologically derived metabolic index <0.001 G. morhua p01 0.97 BDMI=-0.07+0.71* Physiologically derived metabolic index <0.001 G. morhua p05 0.96 BDMI=-0.46+0.68* Physiologically derived metabolic index <0.001 G. morhua p10 0.96 BDMI=-0.31+0.68* Physiologically derived metabolic index <0.001 G. morhua p15 0.97 BDMI=-0.13+0.71 Physiologically derived metabolic index <0.001 G. morhua p20 0.97 BDMI=-0.13+0.70* Physiologically derived metabolic index <0.001 G. morhua p25 0.96 BDMI=-0.04+0.71* Physiologically derived metabolic index <0.001  166  B.10 Percent habitat loss by end of century (2071-2100) relative to baseline conditions (1971-2000) and Φcrit values estimated by the Biogeographically derived metabolic index computed using six different percentiles to define the species oxygen threshold (p01 = 1st percentile, p05= 5th percentile, p10 = 10th percentile, p15=15th percentile, p20= 20th percentile, p25=25th percentile).       Species Parameters Habitat loss (%) Φcrit Callinectes sapidus p01 -46.8333 1.316823 Callinectes sapidus p05 -46.8333 1.291733 Callinectes sapidus p10 -46.8333 1.281191 Callinectes sapidus p15 -46.8333 1.270118 Callinectes sapidus p20 -46.8333 1.26349 Callinectes sapidus p25 -46.8333 1.259108 Diplodus puntazzo p01 -16.7094 1.086202 Diplodus puntazzo p05 -16.7094 1.051974 Diplodus puntazzo p10 -16.7094 1.037838 Diplodus puntazzo p15 -16.7094 1.027917 Diplodus puntazzo p20 -16.7094 1.023891 Diplodus puntazzo p25 -16.7094 1.020428 Gadus morhua p01 -13.0092 1.349181 Gadus morhua p05 -13.0092 1.305141 Gadus morhua p10 -13.0092 1.248185 Gadus morhua p15 -13.0092 1.199221 Gadus morhua p20 -13.0092 1.167584 Gadus morhua p25 -13.0092 1.149023       167  Appendix C  Chapter 4: Impact of warming and deoxygenation on pelagic fisheries of the Eastern tropical Pacific  C.1 Temperature preferences and oxygen thresholds of all the species included in the computation of BDMC and MODC for the Eastern Tropical Pacific Ocean. Temperature preferences were estimated as the median temperature values throughout the species distribution and were computed based on COBE sea surface temperature averages for 1970-2009. Oxygen thresholds are the 10th percentile of sea surface oxygen values throughout the species distributions and were computed based on hindcast model outputs from CESM2, CNRM-ESM2-1 and IPSL-CM6A-LR.    168  C.2 Average sea surface temperature (°C) in Exclusive Economic Zones of the Eastern Tropical Pacific Ocean (1970-2009).           169  C.3 Average sea surface oxygen (mol/m3) in the Exclusive Economic Zones of the Eastern Tropical Pacific Ocean (1970-2009). Each color represents the hindcast output of a different Earth system model.            170  C.4 Biogeographically-Derived Metabolic Index of the Catch (BDMC) time series for Exclusive Economic Zones of the Eastern Tropical Pacific Ocean (1970-2009). Each color represents the BDMC computed with different oxygen hindcast output.            171  C.5 Mean Oxygen Demand of the Catch (MODC, mol/m3) time series for EEZ in the Eastern Tropical Pacific Ocean (1970-2009) computed with oxygen outputs from three different Earth system models.             172   C.6 Coherence between Mean Oxygen Demand of the Catch and the Oceanic Niño Index for pelagic fisheries in the Eastern Tropical Pacific. The horizontal axis represents time, while the vertical axis represents frequency. Red colors represent areas where the two time series co-vary while blue colors represent areas where there is a low dependence between both time series. The absence of arrows means both time series co-vary with no lag or lead. Arrows to the right indicate the time series in phase (they move in the same direction), arrows to the left mean the time series are in anti-phase (they move in the opposite direction). Right-down or left-up arrows: first variable leads, right-up, left-down: second variable leads.         173     174  C.7 Coherence between the Biogeographically derived Metabolic Index of the Catch of pelagic fisheries in the Eastern Tropical Pacific Ocean and the Oceanic Niño Index. The horizontal axis represents time, while the vertical axis represents frequency. Red colors represent areas where the two time series co-vary while blue colors represent areas where there is a low dependence between both time series. The absence of arrows means both time series co-vary with no lag or lead. Arrows to the right indicate the time series in phase (they move in the same direction), arrows to the left mean the time series are in anti-phase (they move in the opposite direction). Right-down or left-up arrows: first variable leads, right-up, left-down: second variable leads.          175  176   C.8 Catches of pelagic species in Pacific Exclusive Economic Zones between 1970 and 2009, represented as a percentage of its total pelagic catches. Percentages higher than 10% are highlighted in red.  Species Colombia  Costa Rica  Ecuador  El Salvador Galapagos Guatemala  Mexico  Nicaragua  Panama  Acanthocybium solandri 0.02 1.09 0.00 0.00 0.10 0.89 0.00 0.00 0.00 Aetobatus narinari 0.00 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00 Cetengraulis mysticetus 43.09 0.00 7.12 0.00 0.00 0.00 0.00 0.00 64.96 Coryphaena hippurus 0.64 6.05 0.02 3.04 0.52 14.23 1.72 28.18 0.38 Dosidicus gigas 0.00 0.00 0.43 0.00 0.00 3.01 5.65 0.00 0.00 Engraulis mordax 0.00 0.00 0.00 0.00 0.00 0.00 12.36 0.00 0.00 Engraulis ringens 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ethmidium maculatum 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Euthynnus lineatus 0.80 0.71 0.03 0.00 0.05 1.11 0.13 0.00 0.14 Istiophorus platypterus 0.13 1.64 0.00 0.00 0.04 0.49 0.02 0.00 0.12 Katsuwonus pelamis 32.65 26.87 2.69 25.75 58.47 14.25 3.51 10.32 3.70 Makaira mazara 0.00 0.04 0.00 0.00 0.00 0.00 0.01 0.00 0.00 Odontesthes regia 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Opisthonema libertate 0.00 0.00 17.65 0.00 0.00 0.00 0.00 0.00 23.10 Sarda chiliensis 0.00 0.00 0.00 0.00 0.00 0.00 0.69 0.00 0.00 Sardinops sagax 0.00 0.00 31.29 0.00 0.00 0.00 61.63 0.00 0.00 Scomber japonicus 0.00 0.00 38.07 0.00 0.00 0.00 2.62 0.00 0.00 Scomberomorus sierra 1.62 0.00 0.28 0.00 0.03 0.93 0.00 0.00 1.37 Thunnus alalunga 0.00 0.00 0.00 0.00 0.00 0.00 1.06 0.00 0.00 Thunnus albacares 16.91 61.12 1.93 68.06 25.75 58.03 9.66 59.71 5.67 177  Thunnus obesus 2.70 1.54 0.45 2.35 14.16 2.44 0.19 0.75 0.20 Thunnus orientalis 0.00 0.00 0.00 0.00 0.01 0.00 0.66 0.00 0.00 Trachurus murphyi 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Tylosurus crocodilus 0.00 0.00 0.00 0.00 0.00 3.01 0.00 0.00 0.00 Xiphias gladius 1.43 0.94 0.04 0.80 0.87 1.56 0.09 1.03 0.35  178  Appendix D  Chapter 5: Temperature and oxygen supply shape the demersal community in a tropical Oxygen Minimum Zone  D.1 Area Under the Curve comparing presence with predicted presence based on a Φcrit threshold of the local minimum.  Species Area Under the Curve IPSL-CM6A-LR  CNRM-ESM2-1  CESM2 Argentina aliceae 0.58 0.53 0.70 Ariosoma prorigerum 0.66 0.59 0.75 Bathycongrus macrurus 0.58 0.65 0.60 Bollmannia umbrosa 0.64 0.56 0.61 Brotula clarkae 0.70 0.66 0.63 Cherublemma emmelas 0.50 0.51 0.50 Citharichthys platophrys 0.60 0.61 0.58 Coelorinchus canus 0.56 0.55 0.51 Coelorinchus scaphopsis 0.56 0.53 0.69 Coryphaenoides capito 0.50 0.51 0.51 Cynoscion nannus 0.56 0.52 0.52 Decodon melasma 0.62 0.53 0.68 Diplectrum euryplectrum 0.65 0.53 0.75 Engyophrys sanctilaurentii 0.58 0.51 0.61 Gymnothorax equatorialis 0.63 0.55 0.69 Hemanthias peruanus 0.65 0.54 0.73 Heterocarpus vicarius 0.50 0.51 0.50 Hippoglossina bollmani 0.63 0.56 0.72 Hippoglossina tetrophthalma 0.65 0.55 0.70 Kathetostoma averruncus 0.57 0.52 0.52 Lepophidium microlepis 0.60 0.67 0.60 Lophiodes caulinaris 0.50 0.51 0.54 Lophiodes spilurus 0.55 0.51 0.51 Merluccius angustimanus 0.57 0.52 0.52 Monolene asaedae 0.50 0.53 0.53 Mustelus henlei 0.64 0.53 0.71 Ophichthus remiger 0.54 0.51 0.56 179  Species Area Under the Curve  IPSL-CM6A-LR  CNRM-ESM2-1  CESM2 Peprilus medius 0.93 0.93 0.87 Peprilus snyderi 0.56 0.51 0.55 Peristedion barbiger 0.60 0.53 0.53 Peristedion crustosum 0.50 0.51 0.53 Physiculus nematopus 0.54 0.56 0.55 Physiculus rastrelliger 0.56 0.51 0.51 Pleuroncodes monodon 0.50 0.50 0.50 Pontinus furcirhinus 0.62 0.53 0.69 Pontinus sierra 0.55 0.50 0.52 Porichthys margaritatus 0.76 0.79 0.79 Prionotus albirostris 0.62 0.55 0.74 Prionotus stephanophrys 0.65 0.59 0.75 Pronotogrammus eos 0.52 0.53 0.53 Raja velezi 0.63 0.53 0.69 Rhynchoconger nitens 0.59 0.76 0.61 Serranus aequidens 0.56 0.52 0.55 Solenocera agassizii 0.53 0.53 0.50 Squatina californica 0.64 0.59 0.76 Squilla biformis 0.50 0.50 0.50 Symphurus callopterus 0.63 0.55 0.68 Symphurus melanurus 0.54 0.57 0.54 Synodus scituliceps 0.57 0.68 0.61 Torpedo peruana 0.60 0.54 0.56 Zalieutes elater 0.52 0.53 0.56         

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0395020/manifest

Comment

Related Items