Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Exploring the Gulf of Mexico as a large marine ecosystem through a stratified spatial model Vidal Hernandez, Laura 2000

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

Item Metadata


831-ubc_2000-0297.pdf [ 13.98MB ]
JSON: 831-1.0074853.json
JSON-LD: 831-1.0074853-ld.json
RDF/XML (Pretty): 831-1.0074853-rdf.xml
RDF/JSON: 831-1.0074853-rdf.json
Turtle: 831-1.0074853-turtle.txt
N-Triples: 831-1.0074853-rdf-ntriples.txt
Original Record: 831-1.0074853-source.json
Full Text

Full Text

EXPLORING THE GULF OF MEXICO AS A LARGE MARINE ECOSYSTEM THROUGH A STRATIFIED SPATIAL MODEL by L A U R A V I D A L H E R N A N D E Z B. Sc., Universidad Nacional Autonoma de Mexico, 1991 A THESIS SUBMITED IN PARTIAL F U L F I L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OFN SCIENCE In THE F A C U L T Y OF G R A D U A T E STUDIES (Zoology department) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A M A R C H 2000 © Laura Vidal, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Z o o /o <\ The University of British Columbia Vancouver, Canada Date /P^ri / ZLoo O DE-6 (2/88) ABSTRACT The major fishing areas in the world consist of large shelf and adjacent areas, which may-be called Large Marine Ecosystems (LME). L M E usually consist of a maze of subsystems, and studies on their various features and resources are usually compartmentalised by institutional and political entities. As a consequence, we know little about the integrated behaviour of these systems and the large-scale impact of economic activities. This research about the Gulf of Mexico L M E integrates information from a variety of subsystems to construct an overall, spatially stratified mass-balance model using the Ecopath approach. Temporally and spatially explicit models were run using Ecosim and Ecospace to simulate diverse scenarios representing current and hypothetical changes in fishing effort. Maturity criteria suggest that the Gulf of Mexico was, in the 1980s and 1990s, a highly mature and relatively stable system, relatively resilient to impacts by the fisheries. However, the biomass of some groups were severely affected directly or indirectly (i.e., by-catch) and the landings of some groups are decreasing. Resources highly affected by the fisheries showed a potential for recovery, but tended towards new levels of abundance. Overall, the catches will not increase as greater fishing effort is applied. Spatial modelling highlighted the strong coupling nature between pelagic and demersal components of the ecosystem, and between the inshore and offshore subsystems. Estuaries are important to the integrity of the system. The design of Marine Protected Areas should be assessed on a per-species basis taking into account ecosystem information, because reduction of fishing mortality on some predators wil l affect the biomass of their prey species. Stratified spatial modeling was a useful tool to explore the dynamics of the Gulf of Mexico Large Marine Ecosystem. T A B L E O F C O N T E N T S Abstract i i Table of Contents i i i List of Tables vi i List of Figures x Acknowledgments ••• xii Chapter 1. The Gulf of Mexico as a Large Marine Ecosystem 1 1.1. Studying and 'Managing' Large Marine Ecosystems (LME) 1 1.2. Modelling of Large Marine Ecosystems 2 1.3. Modelling the effect of Fisheries on L M E 3 1.4. Research questions 4 1.5. Objectives 4 1.6. Research methods 4 1.6.1. The Ecopath approach 5 1.6.2. The Ecosim approach 6 a. Juvenile-adult split pools in Ecosim 8 1.6.3. The Ecospace approach 9 Chapter 2. Analysis of the Gulf of Mexico 12 2.1.1. Characterization of the Gulf of Mexico 12 2.1.2. Location 13 2.1.3. Meteorological aspects .' 13 2.1.4. Geology 13 2.1.5. Hydrology 15 2.1.6. Sub-ecosystems IV Coastal shallow regions /estuaries 18 Soft bottoms 19 Coral reefs 20 2.1.7. Primary production 21 iii 2.1.8. Secondary and tertiary production 23 2.1.9. Overview of the fisheries 24 Recreational fisheries 27 2.1.10. Bycatch and discards 28 2.1.11. Status of stocks 29 2.2. Models applied to the Gulf 30 2.2.1 Ecopath in the GoM 31 Chapter 3. Synthesis of a Mass-balance model for the G o M 37 3.1.1. Definition of the area modelled 37 3.1.2. Estimation of areas by depth and subsystems 37 3.1.3. Subsystems represented by previous models 40 3.1.4. Definition and description of functional groups 41 a. Non-fish groups • 42 Benthic producers 42 Phytoplankton 45 Meiobenthos 46 Macroinfauna 46 Macroepifauna or epibenthos 46 Zooplankton 47 Shrimps 48 Other decapods 49 Octopus 50 Seabirds 50 Marine mammals 51 Sea turtles 51 Detritus 52 Dead-discards 52 b. Fish groups 53 Oceanic fishes 54 Sharks 55 iv 3.1.5. Standardisation of models parameters 56 a. Weighting factor for biomass 56 b. Weighting factor for P/B and Q/B 58 3.1.6. Diet composition 61 3.1.7. Catches by unit of area 61 Commercial fisheries 61 Sport fisheries 64 3.1.8. Landings by fishing gear 65 3.1.9. Discards 66 3.1.10. Balance of the synthetic model 69 3.1.11. Uncertainty and Ecoranger 70 3.1.12. Adjusting mortality by fishing effort 72 3.1.13. Independent analysis of discard fluxes 73 3.2. Results and discussion 74 3.2.1. Structure of the Gulf of Mexico ecosystem 74 3.2.2. Summary statistics 76 3.2.3. Maturity analysis 80 3.2.4. Transfer efficiencies 82 3.2.5. Trophic impact assessment 82 3.2.6. Mean trophic level analysis of landings in the system 85 3.2.7. Comparison of discard fluxes with other fluxes in the system 87 Chapter 4. Temporal and spatial explicit modelling 89 4.1.Temporal dynamic multispecies modelling (Ecosim) 89 4.1.1. Making ontogenetic links (juvenile-adult split) explicit 89 4.1.2. Simulation of diverse fishing policies '. 90 4.1.3. Results and discussion 91 v 4.2. Spatial dynamic modelling (Ecospace) 98 4.2.1. Base map to predicted biomass distribution of all functional groups .99 4.2.2. Incorporation of spatial parameters 99 a) Ecological parameters 101 b) Fishing parameters 101 4.2.3. Adjustment of parameters 103 4.2.4. Simulations of spatial fishing policies (MPAs) 106 4.2.5. Results and discussion 109 Chapter 5. General discussion and conclusions 115 List of References 121 Appendices 148 Appendix 1. Estuaries of the Gulf of Mexico 148 Appendix 2. Original references used in previous models of G o M sub-ecosystems.... 149 Appendix 3. Seabirds species in the Gulf of Mexico 153 Appendix 4. Marine mammals in the Gulf of Mexico 154 Appendix 5. Sea turtles of the Gulf of Mexico 155 Appendix 6. Most common shark species in the Gulf of Mexico 156 Appendix 7. Estimated Ecopath mortality rates of all group-species in the Gulf of Mexico synthetic model 157 Appendix 8. Percentage of change in biomass of all living groups in the Gulf of Mexico synthetic model under Ecospace simulations 158 vi LIST OF T A B L E S Table 2.1 General oceanographic characteristics that define the Gulf of Mexico ecosystem 12 Table 2.2 Layers of water in the Gulf of Mexico 17 Table 2.3 Primary productivity in regions of the GoM ecosystem 21 Table 2.4 Mexican and US commercial fishing vessels operating in Gulf of Mexico waters 24 Table 2.5 Average reported annual landings (t) for the GoM, 1980-1997 25 Table 2.6 Average annual GoM landings from commercial fisheries from 1985 to 1997 by species 26 Table 2.7 Some GoM resources showing evidence of overexploitation 29 Table 2.8 Ecopath models applied to sub-ecosystem in the Gulf of Mexico 36 Table 3.1 Estimated areas (103 km2) of US, Mexican and Cuban territorial waters by depth intervals 39 Table 3.2 Estimated areas of subsystems in the Gulf of Mexico 40 Table 3.3 Total subsystems area (km2) represented by each mass-balance model....40 Table 3.4 Functional groups of the synthetic Gulf of Mexico model 43 Table 3.5 Feeding processes in zooplanktonic groups 48 Table 3.6 Parameters of oceanic fishes included in the synthetic model 55 Table 3.7 Some shark species with nursery areas in the Gulf of Mexico 56 Table 3.8 Conversion factors used to standardise biomass of functional groups 56 Table 3.9a Parameters of non-fish groups used in the synthetic model 59 Table 3.9b Parameters of -fish groups used in the synthetic model 60 Table 3.10a Diet matrix of non-fish groups included in the synthetic model 62 vii Table 3.10b Diet matrix of fish groups included in the synthetic model 63 Table 3.11 Annual landing values of commercial fisheries in the G o M 64 Table 3.12 Annual landing values of US recreational fisheries, and estimated landings by area used to represent the entire system 65 2 1 Table 3.13 Total commercial and sport landings (t- km • year") in the G o M area by fishing gear 66 Table 3.14 Absolute and relative percentage of discarded groups in offshore shrimp fisheries 68 Table 3.15 Annual estimated discards from trawlers operating in the Gulf of Mexico by strata 69 Table 3.16 Summary statistics for the original mass-balance model and the best model obtained using the Ecoranger pedigree routine 72 Table 3.17 Initial and adjusted biomass, fishing and predation mortality rates of highly catch resources in the Gulf. 73 Table 3.18 Final input and output parameters and some results from the balanced synthetic model 75 Table 3.19 Summary statistics computed by Ecopath for the Gulf of Mexico model in 1980s-1990s 78 Table 3.20 Transfer efficiency (%) between trophic level (TL) of the Gulf of Mexico model 82 Table 3.21 Value of discards and other fluxes compared 87 Table 3.22 Discards flux relative to production of particular discarded groups 87 Table 3.23 Primary Productivity required to sustain fluxes associated with fisheries 88 Table 4.1 Parameters defining the transition from juveniles to adults in the Ecosim split pools 90 Table 4.2 Scenarios to test policies related to changes in fishing effort in the Gulf of Mexico ecosystem 91 viii Table 4.3 Changes in the GoM ecosystem components as a result of Ecosim simulations varying effort in 25 years period 95 Table 4.4 Commercially important aggregated types by habitat 96 Table 4.5 Final biomass of aggregated types under scenarios tested in Ecosim....97 Table 4.6 Percentage of change between initial and final biomass of aggregated types under scenarios tested in Ecosim 97 Table 4.7 Final CPUE for exploited functional groups under different scenarios tested in Ecosim 97 Table 4.8 Percentage of change of CPUE (initial vs final) by fishing gear under different scenarios in Ecosim 97 Table 4.9 Preferred habitat of group species of the synthetic model used in the GoM base map 100 Table 4.10 Allocation of fishing gears by habitats in Ecospace 103 Table 4.11 Spatial ecological parameters used to simulate the current base map of the GoM 105 Table 4.12 Spatial scenarios tested in the Gulf of Mexico model with Ecospace 107 Table 4.13 Percentage of change in biomass of aggregated types and all alive species among M P A scenarios tested in Ecospace 112 Table 4.14 Percentage of change of catches by gear among M P A scenarios tested in Ecospace 114 ix LIST OF F I G U R E S Figure 1 Methodology used in the analysis of the Gulf of Mexico Large Marine Ecosystem 11 Figure 2 Gulf of Mexico 14 Figure 3 Antoine's (1972) geomorphologic provinces in the Gulf of Mexico 16 Figure 4 Largest Gulf of Mexico estuaries and reefs 22 Figure 5 Total commercial landings in the Gulf of Mexico, 1980 to 1997 26 Figure 6 Total commercial landings by fisheries in the Gulf of Mexico during 1985-1997 27 Figure 7 Mass-balance Ecopath models for the Gulf of Mexico sub-systems 35 Figure 8 Study area used in the synthetic model 38 Figure 9 Functional groups incorporated in the synthetic mass-balance model of the Gulf of Mexico 77 Figure 10 Matrix of mixed trophic impacts of the Gulf of Mexico synthetic mass-balance model 83 Figure 11 Mean trophic level of the landings in the Gulf of Mexico (by Mexican and US commercial fleets ) from 1985 to 1997 86 Figure 12 Mean trophic level of landings in the Gulf of Mexico (by Mexican and US commercial fleets) from 1985 (right end) to 1997 (left end) 87 Figure 13 Changes in the ratio of biomass in the Gulf of Mexico components with scenario 1 93 Figure 14 Changes in the ratio of biomass in the Gulf of Mexico components with scenario 2 93 Figure 15 Changes in the ratio of biomass in the Gulf of Mexico components with scenario 3 94 x Figure 16 Base map of the Gulf of Mexico Ecosystem showing the habitat types distribution 102 Figure 17 Relative primary productivity in the Gulf of Mexico 102 Figure 18 Ecospace scenarios tested. A) Scenario 1, B) Scenario 2, C) Scenario 3, and D) Scenario 4 108 Figure 19 Setting of an M P A scenario 109 Figure 20 Base map predicting biomass distribution of functional groups from the synthetic GoM model 110 Figure 21 Plot of relative biomass obtained in the base map predicting current patterns of distribution I l l Figure 22 Resultant distribution of fishing effort by gear in the Gulf of Mexico model 112 xi A C K N O W L E D G M E N T S First, I give my thanks to my supervisor, Dr. Daniel Pauly, for his guidance and support throughout my study program. Also, I thank to all my committee members, Drs. William Neill, Carl Walters and Tony Pitcher for their assistance, special thanks to Dr. Vi l ly Christensen for his invaluable support in this research project and for his inspiration. I also thank Drs. Francisco Arreguin-Sanchez, Joan Browder, Ussif Rashid Sumalia and Dennis Chitty, for their motivating attitudes in the scientific world. M y thanks go also to Kevern Cochrane and Sherry Manickchand-Heileman for their technical advice and help. M y warm thanks to many colleagues and friends, both in my past and present as a student and as a researcher, for their invaluable support and help, but also inspiring ideas and work. Particularly, I thank Mr. Leonardo Huato, Mrs. Martha Haro, Ms. Silvia Salas, Ms. Eny Buchary, Dr. Ratana Chuenpagdee, Mr. Tom Okey, Mr. Trevor Huton and Mr. Marcelo Vasconcellos. I thank the International Council for Canadian Studies and CONACyT for their financial support during my program. I also thank to Ms. Saskia Arnesen for proofreading this thesis. Personally, I want to express my deepest gratitude to my loving family specially to my husband, Gerardo Gomez and my parents, Jose M . Vidal and Eugenia Hernandez de V . who have always encouraged and supported me, who have giving me inspiration with their lives and values, but also who have always trusted me without reserve. Last but not least, I want to thank God who has filled my life with blessings (my family, my baby and my friends), but also has given me faith and the opportunity to accomplish and enjoy my goals and choices (my career). xii CHAPTER 1. THE GULF OF MEXICO AS A LARGE MARINE ECOSYSTEM The Gulf of Mexico (GoM) is a large marine ecosystem (LME) located between tropical and subtropical Atlantic latitudes, possessing a wide variety of marine habitats. It is one of the 49 LMEs defined by Sherman and Alexander (1991) that sustain 75% of the world marine nominal catch, and is considered one of the most economically productive bodies of water within Mexican and U.S. territorial waters. The Gulf is about 60% the size of the Caribbean Sea (w 1,942,500 km2), and has a fishery yield higher than 1 million tonnes a year. For the past two decades, both human expansion and commercial and recreational activities (fisheries, tourism, marine shipping, mineral extraction, oil and gas production) have increased on its coast; while several stocks show evident symptoms of overexploitation and several subsystems show severe damage. Even though the Gulf has been recognised as one ecological unit (Sherman and Alexander 1991) and strategy implementation for its sustainable development has been considered a priority by the sharing countries (Yanez-Arancibia et al. 1997), study and management are compartmentalised by institutional and political units. As a consequence, we know little about the integrated operations of the system and the large-scale impact of economic activities. 1.1. Studying and 'Managing' Large Marine Ecosystems ( L M E ) An understanding of how aquatic ecosystems work and how they are affected by increasing exploitation activities (e.g., fishing) will contribute to establishing a basis for proper and integral management. This will contribute to the sustainability of related economic and social activities, while helping to conserve biodiversity and health of the systems (Beddington 1984; Steele 1984; Sherman 1986). To pursue an integral management goal based on ecosystem knowledge is a great challenge, especially since major fishing areas in the world are concentrated in large oceanic areas shared by more than one political entity, either provinces, states or countries. Two classifications of large marine areas have been proposed so far. First, Sherman and Alexander (1986) defined Large Marine Ecosystem (LME) as large oceanic areas (larger than 200,000 km2) close to land with distinctive hydrography, topography and trophically-linked communities. Secondly, Longhurst (1995) suggested the term 'biochemical provinces' be applied 1 to different oceanographic units that generate distinctive patterns of nutrient fluxes found in marine ecology studies. Pauly (1998) emphasised the importance of standardising these criteria for studying ecological units based on a functional definition and mentioned the convenience of the latter classification for authors wishing to compare oceanological processes. Regardless of the classification, the intention to manage large areas as ecological units has shown much promise. Since the size and complexity of these ecosystems limit our ability to understand them, one way to approach them has been through modelling (Toft and Mangel 1991). The Gulf of Mexico is considered an independent entity in Sherman and Alexander's classification, whereas Longhurst classification system suggests the Gulf is integrated with the Caribbean Sea. For the current study and given that the GoM shows a unique combination of tropical and subtropical characteristics not shared with the Caribbean Sea, I considered it one 'independent' Large Marine Ecosystem for modelling purposes. 1.2. Modelling of Large Marine Ecosystems Two different approaches have been used in modelling LMEs: Simulation models, as developed to describe the ecosystems in the North and Bering Seas (Andersen and Ursin 1977; Laevastu and Favorite 1977) and, mass-balance models, used to describe e.g., the South China Sea and Alaska Gyre ecosystems (Pauly and Christensen 1993; Okey and Pauly 1998). Although dynamic simulation models allow us to test various hypotheses about community dynamics in the systems, their implementation has traditionally encountered methodological problems because of the complexity of the interactions among groups and the amount of data required (Pauly and Christensen 1993). In contrast, mass-balance modelling deals relatively simple processes during a defined period of time, under the assumption of steady-state over a given period, generally a year. Recently, these two approaches have been merged (Bundy 1997; Walters et al. 1997, 1998; Buchary 1999; Okey and Pauly 1998) with the mass-balance approach providing the basis for simulations in both time and space (see below). Traditionally, the models built to understand LMEs were constructed under the convenient assumption of homogeneity in time and space. However, these ecosystems are formed by the integration of a variety of subsystems such as mangrove, seagrasses, coral reefs, and oceanic areas, each with special characteristics and typical components. Until now, only the mass-balance model developed by Pauly and Christensen (1993) for the South China Sea was stratified 2 by the subsystems in the area. Pauly and Christensen (1993) represented that large marine ecosystem by a series of interlinked models each expressing a subsystem, with the purpose of maintaining biological reality in estimated biomasses and trophic interactions. From that study, good descriptions of subsystem dynamics and generalisations of the whole ecosystem were obtained. However, estimations such as primary production transfer efficiencies, and detritus flows only represent sums of independent components, rather than a true integration of the strata. 1.3. Modelling the effect of fisheries in L M E s Meaden and Do Chi (1996) recognised the spatially extensive nature of marine fisheries as an economic activity. Also, they point out that fisheries problems throughout the world are expanding from local to international scales. Evidently, temporal, technical arid some spatial regulations have not been enough to overcome the increasing overexploitation and destruction of marine resources. Since major fishing grounds in the world are located in Large Marine Ecosystems, and since these areas are a puzzle of subsystems, explicit definition of fishing grounds and ecosystem unit based allocation of fishery efforts may improve our understanding of the systems. They may also serve as a tool for spatially-based fisheries management (Caddy and Garcia 1986). Fortunately, our knowledge about resource distribution is rapidly increasing through the use of Information Technology (IT) and Geographic Information Systems (GIS). However, neither spatial allocation of fishing effort nor the spatial description of its impact has been well studied. This is of particular concern among fisheries scientists and managers because management policies are generally based on the unrealistic assumption that the fishing effort is evenly distributed in the areas. Our knowledge about the effects of the fisheries in particular areas would improve i f we knew not only where the resources are, but also where a particular type of fishing effort is applied. To enhance our understanding of these complex LMEs, I have integrated information from a variety of subsystems of a L M E using the Ecopath approach. This work on the Gulf of Mexico documents a procedure in which subsystems can be integrated to construct an overall spatially-stratified mass-balance model. This procedure was modified form that presented in Pauly et al. (1999b), where an approach was proposed which did not consider the Ecospace software of 3 Walters et al. (1999). Finally, using Ecosim and Ecospace, I show how the integrated model can be used to simulate diverse scenarios concerning changes in fishing effort. 1.4. Research questions 1. Can an integrated mass-balance trophic model, like Ecopath, be used to describe an LME? 2. Is it possible to predict the effect of current fishing trends in an L M E using a spatially integrated model? 3. Is it possible to predict spatial distribution of biomass of an L M E using a spatially integrated model? 1.5.0bjectives This study has the following objectives: • To integrate trophic interactions among functional groups of various subsystems of the Gulf of Mexico L M E in a form of a mass balance model; • To describe the structure and energetic fluxes of the entire Gulf of Mexico based on the integrated model; • To simulate temporal responses of the system to current fishing trends; • To simulate the spatial distribution of functional groups in the system under the status quo; • To simulate the spatial response of functional groups in the system under some protective spatial policies, such as Marine Protected Areas (MPAs). 1.6. Research methods Since the biotic components of ecosystems are strongly interrelated, especially through trophic links, fisheries scientists and managers of aquatic resources are currently beginning to consider that ecosystem approaches should replace single-species approaches (Walters et al. 1997). So far, four approaches have been used to account for trophic interactions within ecosystems: 1. Differential equation models of biomass dynamics (Larkin and Gazey 1982); 2. Multispecies virtual population analysis (MS VP A; Sparre 1991); 3. Bioenergetic modelling (Stewart et al. 1981; Kitchell et al. 1994) and 4. The Ecopath approach (Polovina 1984). The first three approaches require a large amount of data that complicate their usefulness to make reasonable estimates of trophic interaction parameters. These requirements make them inappropriate to 4 apply in the study of complex ecosystems especially in the tropics, where the economic resources supporting scientific studies are very limited. For these reasons, I integrated ecological and fisheries information from Mexico and U.S. Gulf waters in a spatially structured model and applied the Ecopath/Ecosim/Ecospace approach to modelling the Gulf of Mexico ecosystem to address the research questions (Fig. 1). 1.6.1. The Ecopath approach The Ecopath approach originated from a concept proposed by Polovina and Ow (1983) and Polovina (1984, 1993) to study marine ecosystems. Christensen and Pauly (1992, 1995) developed Ecopath further by incorporating network analysis and system maturity indices based on theories of R.E. Ulanowicz, H.T. Odum and E.P. Odum. Ecopath may be seen as a simplified equilibrium version of a multispecies virtual population analysis that defines an ecosystem as a unit with strong internal fluxes where each functional group (either single species or groups of species with similar feeding habits) is compartmentalised and related to each other mainly through predation. This approach allows the modeller to incorporate landings (catches) information as exports from the system Ecopath assumes stability in the conditions over the time represented in the model and assumes mass-balance among inputs (primary production and imports) and outputs (respiration and exports, e.g. catches) for each component or whole system. The result is a thermodynamically 'possible' system described by deterministic and linear equations (Eq.l) where all energy entering at any independent compartment should be accumulated, moved to other compartment through fluxes or leave the system. Production by (i)= all predation on (i) + non-predation losses of (i) + export of (i) where Pj is the production of i ; M2, is the predation mortality on i ; EEj is the ecotrophic efficiency of i , or the fraction of the production of i that is consumed within the system (EEj is Eq. 1 which can be re-expressed as: Eq. 2 5 usually unknown and calculated by Equation 2); 1-EEj is 'mortality of i by non-predation causes', or the fraction of the production of i that flows to detritus; and EXj is the export from the system of i (catches). Equation 2 can be re-expressed as n Bt \PIB\-EEi - Y^Bj .(Q/B) -Dcji-EXi = 0 Eq. 3 7=1 where Bj = biomass of the group (i); P/Bj = production/ biomes ratio of group (i) which is equal to total mortality rate (Zj) assuming equilibrium; Qj/Bj consumption/ biomass ratio of group (j); Dc jj =fraction of the prey (i) in the average diet of predator (j): the summation corresponds to total biomass of species (j) consumed by predators (j). This equation is repeated for all n functional groups in a set of linear equations (Eq. 3.4) that are solved using standard matrix algebra (Christensen and Pauly 1992) as outlined in the following equations Eq. 4: Bj • (P / B\ • EEj - Bj • (Q / B)1 • D C n - B 2 • (Q / B)2 • D C 2 1 - . . - B n • (Q / B)n • D C n l - EXj = 0 B 2 • (P / B)2 • EE 2 - Bj • (Q / B)1 • DC 1 2 - B 2 • (Q / B)2 • D C 2 2 - . . - B n • (Q / B)n • D C ^ - EX 2 = 0 Bn\PIB\ -EEn-Bx (Q/B\DCln - B2 • {QI B\ • DC2n-..-Bn .(QIB\-DCnX-EXn = 0 Eq. 4 Ecopath requires three of the following 4 input parameters: Bj, (P/Bj), (Q/Bj), EEj. If one input parameter is unknown, it can be calculated when Ecopath solves the equations. Diet composition information for each group is also required. Including estimates of exports and catches is also recommended. The solution of Equation 3 also will allow the calculation of the energy balance for each compartment, using Consumption by (i) = production by (i) + respiration by (i) + unassimilated food by (i) Eq. 5 The model is verified and the input parameters are evaluated when Equation 5 is rearranged and for each functional group the calculated respiration is positive and estimated EEj is between 0 and 1. 6 Ecopath 4 (Ecopath with Ecosim) allows the modeller to incorporate exports or specify catches from diverse fleet types and also to incorporate discards from these fleets and then decide the fate of these discards in the system. Also, the software includes a routine called Ecoranger that allows incorporating uncertainty in all the input data, where the level of confidence of the input parameters is specified through 'pedigree' information. Ecoranger runs through a Monte Carlo routine in a Bayesian context (Walters 1996). For the Gulf of Mexico ecosystem this uncertainty routine was used to select the 'best' model. Results of this routine are shown in Chapter 3. Ecopath allows users to estimate the maturity and stability of the ecosystem under study. For the Guf of Mexico, maturity and stability results from the flow analysis are described and discussed in Chapter 3 along with an overview of the trophic interactions and fluxes in the system. 1.6.2. The Ecosim approach Ecopath models provide a 'snap-shot' or static picture of an ecosystem's trophic structure. But since ecosystems are not static structures, the dynamic component was incorporated in the Ecosim routine developed by Walters et al. (1997. Ecosim (ver. 4) uses the mass-balance model as input to explore how the community might respond to changes in fishing mortality through simulation. This approach allows the modeller to answer questions like "what i f the fishing policies or the ecosystem balance changed. The linear equations that describe trophic fluxes in mass-balance (Eq. 1 to 4) are used to provide parameters for differential equations that define changes in the interactions when predator and prey biomasses and harvest regimes change. Perturbations in the ecosystem are introduced as changes in fishing mortality on various functional groups. Consequently, the equilibrium model is transformed into a dynamic simulation model that allows the modeller to assess the effect of the perturbation in each component in the system. The main Ecosim equation is Eq. 6 where: f[B) is a function of Bj i f (i) is a primary producer, or: 7 " / \ fiB) = SiYjCji • \Bi>Bj)tf 0)is a consumer; cjj{Bi,Bj) is the function used to predict consumption (Qy) from the biomass of the prey (Bj) and the predator (Bj) c is the food consumption rate per unit biomass; g is gross food conversion efficiency = P/Q; and M0 is other (non-predation) mortality. Fi is fishing mortality rate Ecosim also includes a representation of prey vulnerability, wherein each functional group is split into vulnerable and invulnerable biomass. The rate of transfer between these two components (v.f. or vulnerability factor) defines whether the prey is lightly vulnerable (v.f. <3) and in consequence the system is 'bottom-up' controlled, or i f the prey is highly vulnerable (v.f. >3) and in consequence the system is 'top-down' controlled. Changes in the fishing regimes can be simulated by changing the relative fishing mortality of a single functional group, of specific gear type, or of all the fishing gear combined in the system. Effort policies are set graphically through a 'sketchpad' interface using the computer mouse. For the Gulf of Mexico, the scenarios tested in Ecosim corresponded to current trends of change in fishing effort in the area. Historic information that support these simulations are described in Chapter 2, while the results of the dynamic multispecies modelling are shown in Chapter 4. 1.6.2a. Juvenile-adult split pools in Ecosim Very often, populations are formed by several age/size classes that play different ecological roles in ecosystems, not only because they are generally located in different habitats, but also because their feeding habits and predators are different. When these differences need to be emphasised in a modelled ecosystem, a routine introduced in Ecopath and Ecosim can be used. In Ecopath, different ontogenetic stages can represent different functional groups. Note that the ontogenetic linkages between juveniles and adults can not be accommodated in Ecopath, so the mixed trophic impact routine generated by the software must be interpreted carefully. 8 In Ecosim, the dynamic interactions between juvenile/ adult pools are explicitly defined. This dynamic software accommodates split pools by using a delay-difference model (Deriso 1980) that keeps track of absolute numbers and biomass of individuals in each pool to allow recruitment from juvenile to adult pools (Walters et al. 1997). To define the links among recruits and adults, some additional parameters are needed. These parameters are: 1. The age at transition to adults (tk, in years); 2. the curvature (respiration) parameter of the von Bertalanffy growth model (K,year -1). Other parameters that link split pools that were left at the default value are: ratio of average adult weight to weight at transition to adults (W a v g / Wk,) (default = 2), base proportion of food intake used for reproduction (default = 0.3), proportion of increase in food intake used for growth (default = 0.8), minimum time as juveniles (default =1 yr.) and maximum time as juveniles (default = 1.0001 yr.). In the Gulf of Mexico model, a very large number of split pools (juvenile-adult) could be described; however, for simplicity only five groups were explicitly split. These pools were peneaids shrimps, medium coastal invertebrates and fish feeders, coastal demersal carnivores, large shelf pelagic predators, and medium coastal fish feeders-large predators. Details of the split pools parameters incorporated in the model are given in Chapter 4. 1.6.3. The Ecospace approach. In section 1.3,1 mentioned the importance of incorporating spatially explicit information on the distribution of the organisms and fishing effort. This information could be incorporated for the Gulf of Mexico ecosystem to simulate the spatial distribution of functional groups using Ecospace. Ecospace is a routine developed by Walters et al. (1999) that makes the model spatially explicit based on user defined habitat preferences, movement rates, and vulnerabilities in different areas for each functional group. The Ecospace routine is incorporated in the Ecopath with Ecosim software and can be used to test spatial policy strategies regarding fishing effort (e.c. marine protected areas or MPA) . The program interface allows sketching of topographic features (shorelines, islands, areas of high primary productivity habitat types and preferences) and policy options such as location, size or shape of M P As with the computer mouse. Ecospace represents biomass dynamic patterns over two-dimensional space as well as a time grid using an Eulerian approach. This approach treats movement as 'flows' of organisms among fixed spatial reference cells, without retaining information about the past history of the organisms 9 present at any point. The instantaneous dispersal rates (nil) across each cell boundary are assumed to vary with: 1) pool type (active or passive/transport movement of organisms); 2) habitat type on the source cell side of the boundary (defined as preferred habitat) and, 3) response of organisms to predation risk and feeding conditions in source cell. For trophic interaction, harvesting and movement calculations, biomass densities are treated as homogeneous within each cell (Walters and Pauly 1998). Ecospace represents spatial distribution of fishing mortality using a relatively simple "gravity model" (Caddy 1975; Hilborn and Walters 1987), derived by assuming the fisheries (specific gear) distribute effort so as to maximise their catches (profits) depending on the attractiveness and the costs of the access to the cells (fishing grounds). Each biomass pool in each cell (i) is subject to a total fishing mortality rate equal to: F i c = 2 k F k c q K i Eq. 7 where q K i is catchability of type i organism by gear k, and F kc is the total mortality by gear k in cell c, by all gears together (Z k) The resultant biomass patterns predicted over time are shown both as color-coded density maps, and as relative biomass values to the original started biomass. Despite the fact that Ecospace does not account for seasonal changes or oriented migration, it provides one with information-rich graphs and tables (which contain decadal scale predictions of spatial biomass patterns) that could be give the modeler insights about the likely efficacy of alternative policies. I used Ecospace to simulate the spatial distribution of functional groups in the system under the status quo but also to evaluate some fishing policies (MPAs) that would protect a large portion of two of the most exploited subsystems in the Gulf (soft bottoms and estuaries). This last simulation will give us insights about the benefit of these hypothetical protected areas and the impact in the biomass and distribution of the resources in the entire ecosystem. This approach is described in the second section of Chapter 4. 10 60 s ^  t3 a O 1 3 g CN 1—1 M o Inclusion of fishing effort distribution and relative PP spatial variation Spatial simulation of various fishing policies in the system Construction of a basemap that recreate current patterns of biomass distribution of all functional groups (Ecospace) Adjustment of parameters IncorporatKm of spatial parameters (Ecospace) Temporal simulation of diverse fishing policies in the system (Ecosim) Incorporation of juvenile-adult split pool parameters and test of stability (Ecosim) • Inclusion of discards by trawlers Construction of a stratified mass-balance model of the GoM (Ecopath) Parametrization of integrated model (adjustment of parameters) Definition and incorporation of fishing fleets operating by strata by functional group Gaps in area filled by extrapolation using ecological and bathymetric-topographic characteristics. Addition of sensitive groups Estimation of areas through digital ization Estimation of parameters by weighted area and biomass Assignation of each submodel to represent coastal and deep continental shelf regions Standardardisation of unit of currency IdentificatK>n and integration of functional groups maintaining ontogenetic stages and subsystem trophic structures Identification of previous mass-balance models of subsystems in the study area Construction of a stratified schema by depth and key subsystems (habitats) Collection of ecological and spatial information (Literature review) Definition of the GoM ecosystem Fig 1. Methodology used in the analysis of the Gulf of Mexico Large Marine Ecosystem. 11 CHAPTER 2. ANALYSIS OF THE GULF OF MEXICO In this chapter I analyse the GoM from two perspectives. The first section (2.1) summarises important oceanographic and ecological characteristics in the area, and includes information about current trends in biological exploitation. The second section (2.2) provides information about the models applied to the area, while emphasising the use of mass-balance models to describe sub-ecosystems in the Gulf. 2.1.1. Characterization of the Gulf of Mexico The Gulf of Mexico is a large ecosystem that combines characteristics of tropical and warm temperate waters. It consists of a mix of oceanographic characteristics that gives rise to multiple subsystems with complex patterns of productivity. Currently, Cuba, Mexico and the USA, extract several renewable and non-renewable resources from the area, mainly from the shelf regions. General oceanographic characteristics of the ecosystem are described in several sections, which emphasise the complexity of this system (Table 2.1). Some of them are incorporated in further chapters as part of the integration of an overall model for the Gulf. Table 2.1 General oceanographic characteristics that define the Gulf of Mexico ecosystem. Characteristic Unit Location - 18°-30° Lat. N, 80°-98° Long. W Area Km 2 1,623,000 Maximum depth m 3,400 Climate - Tropical and warm-temperate Seasonal events 'Nortes' and hurricanes Annual runoff 106m3 29,100 to 192,800 Annual precipitation 101 0cm3 1,722 to 56,940 Superficial water temperature ° C 22-28 Superficial water salinity - 36.0 - 36.7 Primary production R C m V 25 - 350 12 2.1. 2. Location The GoM (Fig. 2) is bordered at the eastern, northern and northwestern regions by five USA states (Florida, Alabama, Mississippi, Louisiana and Texas), at the southwestern, south and part of eastern shores by five Mexican states (Tamaulipas, Veracruz, Tabasco, Campeche and Yucatan), and by Cuban waters at the southeastern region. The GoM connects to the Caribbean Sea through the 1,750-m deep Yucatan Channel (between Yucatan peninsula and Cuba) and with the Atlantic Ocean through the 800-m deep Florida Strait (between the Florida peninsula and Cuba) (Fig.2). 2.1.3. Meteorological aspects The Gulf of Mexico is warmer in the southern and eastern regions and warm-temperate towards northern regions. The climate changes from arid to subhumid from west to east. Annual precipitation on the US coast ranges from 142 cm year"1 in northern provinces, to 61-19 cm year" 1 in Lower Laguna Madre, Texas. In Mexican regions, annual precipitation is several times higher than on the US coast; it ranges from 1,722 x 10 1 0 to 56,940 x 10 1 0 cm 3. Annual runoff ranges from 192.8 to 29.1 x 109 m 3 , with the higher values corresponding to western and southern regions (Ibarra-Obando et al. 1997). Climatological events such as 'nortes' and hurricanes produce seasonal changes in precipitation and temperature. The 'nortes' are cold winds coming from the Northeast of the Gulf from December to March ('winter'). Hurricanes originate from equatorial areas, generally from July through November ('summer'). During these events, freshwater inflow considerably increases owing to intense precipitation. 2.1.4. Geology Of the total GoM area, approx. 40% is continental shelf. The shelf is broad off the West Coast of Florida (225 km), the Texas/ Louisiana border (201 km) and off Yucatan (280 km); while Alabama, Mississippi, Tamaulipas, Veracruz, Tabasco and Cuba shelves are narrower. The average depth of the Gulf is 1,524 m with the deepest regions greater than 3,400 m (Pica-Granados et al. 1991; Upton et al. 1992). 13 Antoine (1972) divided the GoM into 7 regions according to its geologic morphology. Regions 1-3 correspond to US waters from 0 to approximately 3,000 m min depth, regions 4-6 correspond to Mexican waters at the same depth, while region 7 corresponds to oceanic areas approximately 200 to 3,400 m in depth (Fig.3) (this study did not include Cuba). The sediments that form the GoM floor are variable in origin. The distribution of sediments and carbonates in the Gulf, closely correspond to Antoine's provinces. The tip of the Florida Peninsula to western Alabama, and the Yucatan Peninsula (Antoine Regions 1, part of regions 2 and 6, respectively) are formed mainly by carbonated sand of living origin. It includes patches of ancient and modern coral growth and large areas of mangrove swamps. The rest of the Gulf coast comprises alluvial sandy, silt, clay and muddy sediments derived from ancient and contemporary rivers and ancient glaciers. Major deltas responsible for sedimentary deposits in the area are the Mississippi, Atchafalaya, Bravo, Papaloapan, Grijalva and Coatzacoalcos deltas. These deltaic systems can form estuarine channels or bar-built estuaries along the coast. The latter are common in the western part of the Gulf. The open ocean sea floor is formed by blue sand and of Globigerina oozes (Lynch 1954). 2.1.5. Hydrology The most evident and important internal current in the Gulf of Mexico is 'La corriente del Lazo' ('string current') which connects the Yucatan strait and the Florida Strait. This current is a flux of water with high salinity («37) and surface temperatures that reach between 28° and 29 °C during summer, and 25° and 26 °C during winter. Anticyclonic currents, characterized by clock-wise and sinking movements to the western Gulf, occur when 'del Lazo' current loses strength in late summer and early fall. These currents commonly form in front of the Tamaulipas coast, western Florida, and on the Campeche Bank. When the anticyclonic currents are weaker and become separated from the original currents, they form cyclonic currents over the shelves adjacent to Florida, Texas-Louisiana shelf and Campeche Bank. These currents are directed to the western Gulf and are characterised by cold ascending waters and counter-clockwise movements. Austin (1955) defined two major zones with high geo-potential. One is the Yucatan current and the other is off of the Tamaulipas coast, where 2 main upwelling systems occur. 15 ( N O) a p m u B q Tides in the Gulf area are mainly diurnal (oscillating every 24 hour), but vary within regions. There are regions with mixed tides such as the Texas-Louisiana and Florida shelves, and semi-diurnal tides occur in the Campeche sound (oscillating every 12.5 h). Tidal range is small throughout the region with a maximum of 1 m in Florida and Veracruz and a minimum of 0.5 m along Louisiana and Texas coasts. Table 2.2 shows the well-defined layers of water in the Gulf of Mexico (Nowlin, 1971). Among them are 1. the superficial layer which originates from the Caribbean Sea and shows strong local variations by latitude and is due to seasonal events and, 2. the layer of minimum oxygen, which is considered extremely important in regards to the capture and distribution of nutrients. The location of this layer in the water column limits the productivity of the area (Morrison and Nowlin 1977). Table 2.2. Layers of water in the Gulf of Mexico. N-northern regions, S-southern regions a ) . Depth Temperature Salinity Layer (m) (°C) Mixed or superficial 0-150 19-20 N, 24-29 S 32 N, 36 S Subtropical 150-250 - 36-37 Minimum Oxygen 250-900 6-19 35-36 Intermediate Antarctica 900-1050 6.2 34-35 Deep North American 1050-1400 4 34.96 Bottom 1400-3400 4 34.96 a ) source: Morrison and Nowlin (1977) 2.1.6. Sub-ecosystems The Gulf of Mexico encompasses several distinct aquatic environments, each typically inhabited by different communities. Some examples of these sub-ecosystems are marshes, estuaries, lagoons, mangrove stands, coral islands, sea grass beds and coral reefs. The boundaries between these habitats are arbitrary. For the purposes of this study, I used what seemed to be the simplest divisions. I only considered three gross divisions: coastal shallow systems from 0 to 20 m depth contours, deeper continental shelf systems from 20 m to 200 m depth contours and oceanic deep water with depth greater than 200 m. 'Coastal shallow systems' could be subdivided into estuarine, non-estuarine/non-reef areas, and coral reef systems. The 'deeper continental shelf 17 systems' only considered soft bottoms. Since most studied and productive areas are estuaries, reefs and soft bottoms, these subsystems will be described as follows. Coastal shallow regions. Estuaries The coastline from the Dry Tortugas Islands off the southern tip of the Florida Peninsula to Cabo Catoche on the Northeast tip of the Yucatan Peninsula is about 4,645 km in length. This coastline is divided almost equally between Mexico and the U.S. (Upton et al. 1992; Ibarra-Obando et al. 1997, Rivera-Arriaga and Villalobos-Zapata 1998). Coastal shores are bordered by mangrove forest (mainly Rhizophora mangle and Avicennia germinans) in tropical Gulf regions (all west Florida and from Veracruz to Yucatan peninsula) and by salt and fresh marshes (Spartina spp. and Distichlis spicata mainly) in the northern portions of the Gulf. Tidal flats with sea lettuce and smooth cordbngrass (Ulva lactuca and Spartina alterniflora) are present on the floor of estuarine areas (Field 1991; To villa-Hernandez 1994). Estuarine waters account for approximately 25% of the total inshore surface area. In the US Gulf, there are 31 estuaries covering about 31,000 km 2 of water along the coast (NOAA 1990). Some important ones are: Florida Bay, Tampa Bay, Apalachicola Bay, Mobile Bay, Mississippi Sound, Breton/Chandeleur Sounds, Galveston Bay, Corpus Christi Bay and Lower Laguna Madre. The Mississippi and Atchafalaya rivers mainly influence these estuaries. In the Mexican region, 30 estuaries and lagoons accounting to 6,800 km 2 have been described (Lankford 1975; Ibarra-Obando et al. 1997). The most significant estuaries in ecological terms are the Laguna Madre, Laguna Tamiahua and Laguna Terminos. Five large rivers highly influence these coastal systems: Bravo river, Papaloapan river, Grijalva river, Coatzacoalcos river and Candelaria river (see Appendix 1). Although estuaries in the GoM have a wide water surface area, they are generally shallow (average 8 m depth), and receive a large freshwater inflow (average 110,950 • 106 m 3 • year"1). The Mississippi River dominates the inflow of fresh water into the area with 30,000 m3- s"1. These estuarine areas are considered of great importance since they support various important economic activities such as urban, fishery, recreational and agricultural activities (NOAA 1990). Ecologically, estuaries are important nursery habitats for a variety of marine organisms providing refuge from predation, habitat, and food. For example, at least 15 species of sharks are 18 known to use gulf coastal waters as nursery areas (Hueter et al. 1994; Bonfil 1997). Species comprising the four top fisheries in the US gulf (shrimp, menhaden, oyster and blue crab) use estuaries extensively. It is considered that the economic value of the commercially important estuarine dependant fisheries is about 650 million dollars a year (NOAA, 1990). Most relevant estuaries in the area are shown in Fig 4. Soft bottoms Soft bottom floors are formed by either carbonated sand and living sediments or by alluvial sandy and muddy sediments. Compared to the hard bottoms, soft bottoms are less stable sustrata. Several natural disturbances such as waves, storms, currents, alluvial influence and large amount of animals affect their surface (from few centimetres to a meter) through the suspension or removal of sediment particles. However, like the hard bottoms, soft bottoms increase their structural complexity with living organisms. Small organism such as foraminifers, coralline algae, corals, brachiopods, bryozoans, worms and molluscs utilise pebbles or other rocks to construct calcium carbonate structures; while bigger organisms such as sponges, anemones, gorgonians, sea pens, polychaete worms, crustaceans, sea urchins and crinoids, create their own tubular structures on the seabed (Watling et al. 1998). Shallow soft bottoms of the Gulf region are usually covered by patches of diverse species of smooth sea grass (e.g. Halodule beaudettei and Thalassia testudinum ) and sea lettuce (Ulva lactuca). The structural complexity of these habitats allows a large number of organisms to utilise soft bottoms as feeding and hiding places. Factors determining the composition of community assemblages in these areas appear to include seagrass biomass and blade density (Stoner 1983) nature of microhabitats, water movement (Adams 1976) and diverse physical/chemical parameters (Alevison et al. 1990). Major shrimp grounds occur in soft bottom areas; as a result, the main activity of trawlers is in those areas, with consequently constant deterioration of these natural communities. In the Mexican waters, shrimp grounds are mainly concentrated within the Campeche Bank and offshore Tamaulipas coast. In US waters, shrimping grounds are widely distributed on the entire Gulf shelf, however the fishing effort is distributed differently by province (Nance 1993). 19 Coral reefs. Several non coral reefs reefs, such as oyster and rocky reefs, occur in the Gulf; however, since only the coral reefs have been studied and identified properly, this study gives emphasis only to coral reefs. In the US Gulf, there are two main reef systems. The first and most extensive is the chain of the Florida Keys, Marquesas keys and Dry Tortugas, located at the Southwest tip of the Florida peninsula. The living coral reefs in this area represent the scattered fragments of what was once a much extensive living reef system. The second system contains the Flower Garden Banks located roughly 180 km south of the Texas and Louisiana border. Also, there are a few scattered heads of coral in the western shelf of the Florida shelf. In the Mexican gulf area, three main regions contain coral reefs: the Campeche Bank (Alacranes, Cayo Areas, Triangulos, Chinchorro and Isla Arenas), Yucatan east shelf, and Veracruz shelf (Lobos, Enmedio and Blanquilla). The majority of Campeche Bank reefs are dispersed at the northwestern region of the Yucatan Peninsula. The East Yucatan reefs are also dispersed systems bordering the northeastern Yucatan peninsula and North Yucatan shelf. Veracruz reefs are located very close to the land between the 20 and 50-m isobaths in the southeastern part of the Gulf. Cuban reefs are located very close to the coastline along the northwestern coast, which faces the Gulf of Mexico. These reef systems are shown in Fig 4. The characteristic zonation pattern of the coral reefs of the tropical western Atlantic region is much simpler than that found among Indo-Pacific reefs (Goreau 1959). However, species composition and diversity varies among Gulf reefs. For a southern reef, Arrecife de Lobos (Veracruz), Chavez et al. (1970) described two main regions: the algal ridge, where Thalassia testudinum and Halimeda opuntia are very abundant, and the reef rim, where common coral species such as Montastrea annularis, Diploria clivosa and Acropora palmata are located. They found six different biotopes with at least 80 animal species within more than five invertebrate taxa. In the Florida Reef Tract, four zones have been described from inshore to offshore: patch reefs, seagrass beds, bank reefs and intermediate/deep reefs. T. testudinum is the main component of seagrass beds; while Montastrea annularis, Acropora sp. and Millepora sp. are the common coral species in that area. Venier (1997) gives a detailed description of zonation patterns of these Florida reefs. 20 Fish communities are also a very important component in reef areas. In the Looe Key National Marine Sanctuary, 188 fish species have been observed (Bohnsack et al. 1987). This fore-reef of 18 km 2 is considered to be representative of the entire Florida reef tract since its species are typical of the rest of the region (U.S. Department of Commerce 1996). Fishing species such as groupers, snappers and porgies occupying coral reefs and rocky bottoms are optimum for exploitation. Reef systems in the Gulf undergo a wide range of use. The Flower Garden Banks are somewhat protected from exploitation, probably owing to their greater distance from the coast (Gittings 1999). In contrast, Florida Keys and most of the Mexican reefs have been used extensively, and overexploited. In Yucatan reefs, groupers are caught mainly in areas between 36 and 54-m depths, constituting one of the most important economical activities of that region (Arreguin-Sanchez et al. 1987). 2.1.7. Primary production Extensive primary production (PP) studies in the oligotrophic waters of the G o M were conducted in the 1970s and 1980s. Areal production rates obtained from these studies varied almost two orders of magnitude as a consequence of differences in the methodology, time of year, location and experimental conditions (Vargo et al. 1990). In general, I found reports of three well-defined zones: a) Highly productive zones with mean PP value of 300 g C • m 2 • year"1 mainly corresponding to lagoons, coastal areas, coral reefs and upwelling regions; b) zones of medium productivity where the PP is about 100 g C • m 2 • year"1 corresponding to the rest of the continental shelf and, c) zones of low productivity such as western Florida and the open Gulf where PP is about 25-30 g C • m 2 • year"1 (Table 2.3). Table 2.3. Primary productivity in regions of the GoM ecosystem. Level ofPP PP mean values (g O m2 • year'1) Regions Source a) High 200 and 350 Tamiahua and Terminos Lagoons, Walsh (1988); Merino or more Tamaulipas coast and Sonda de (1992); Day et al. Campeche (upwelling regions), Cabo (1982) Catoche, coral reefs and Mississippi delta b) Medium 100-160 the rest of the continental shelf except Kondratyeva and Sosa western Florida, Cuban gulf offshore (1966) waters, c) Low 25-30 western Florida and the open Gulf El-Sayed (1972); Flint and Rabelais (1981) 21 o U T3 ca o u CD a C3 O ' „ CD CTs C< ^ N e 2 £ 2 o u Si if? ro CD CD 03 CD 1 m 0 0 c CD 1 o O CD (-M S CN CD CD l-H oo" CQ o CD 00 <D cd •c o O CD i cd CN CN Phytoplankton biomass and productivity in the ocean are influenced by zooplankton grazing and nutrient excretion (leading to "regenerated productivity" Vargo et al. 1990). However, the higher primary production in coastal areas of the Gulf is evidently related to high runoff from rivers and contributions of production from mangrove defoliation. For example, in Terminos Lagoon an 2 1 annual mean leaf fall rate was calculated to be from 32 to 51 g C • m • year" (Day et al. 1996). On the other hand, in large open estuaries such as Tampa Bay, major sources of local productivity are phytoplankton, followed by mangroves (Lewis et al. 1988). The main sources of productivity in coral reefs are resident algal communities, symbiotic dinoflagellates associated with coral, and seagrass communities with associated epiphytes and macroalgae. 2.1.8. Secondary and tertiary production The shallow southern Gulf coast, less than 20 m deep, is considered to support the highest secondary and tertiary productivity in the entire ecosystem. In the US area, regions of secondary and tertiary productivity match with high primary productivity areas, mainly the Mississippi Delta region. One exception is off the Louisiana coast where excessive levels of organic enrichment have resulted in seasonally extensive hypoxic areas wherein the abundance of benthic communities is much decreased (Turner et al. 1986; Bierman et al. 1987). In Mexico, approximately 42% of the fishery resources are caught in shallow areas off the Campeche-Yucatan states, i.e., on Campeche Bank. This region is influenced by cyclic upwelling events that have been associated with an increase in catches (Gonzales 1975). One particularly productive area is Laguna de Terminos, where 64% of Campeche fisheries production was removed (about 19,160 t) in the late 1980s and early 1990s (Sanchez-Gil et al. 1993). 23 2.1.9. Overview of the Fisheries The fisheries in the GoM can be partitioned into subsistence, commercial, and recreational. As the total effort and catches of subsistence fisheries are not well documented, this study will refer only to commercial and recreational fisheries. In commercial fisheries, inshore and offshore fleets can be further distinguished. Inshore commercial fleets consist of boats between 7 and 15 m with outboard engines and performing trips of less than 5 days. They operate in waters of less than 20m depth. In Mexico these boats have less than 10 tonnes capacity (Ramos-Miranda et al. 1991). Boats in offshore commercial fleets are 15 to 26 m in length, and have inboard engines and larger storage capacities (>10 tonnes); they can stay at sea for more than 5 days, and tend to operate in waters of more than 20 m depth. Among commercial vessels, fishing effort in the GoM is more intense in inshore waters than in offshore waters. Recent estimates by governmental agencies (Table 2.4) show that 84% of all fishing vessels registered in the area (about 64,000 units) are inshore boats used in oyster collection, gill netting, and crabbing. The rest of the vessels operate in offshore waters and are mainly shrimp trawlers, longline vessels for reef and pelagic fishes, oyster dregers, purse seiners and finfish draggers. Offshore vessels often participate in more than one fishery (SEPESCA-INP 1994; Burrage etal. 1997). Although complete databases of fishing effort are not available, there is evidence that commercial fishing in the GoM has considerably increased in the last two decades. For example, the US shrimp fleet expanded 1.4 times from 1980 to 1992 (Nance 1993), and the number of boats in the Mexican fleet doubled between 1980 and 1997 (Miranda et al. 1991; S. Salas CINVESTAV, unpublished data). Table 2.4. Mexican and US commercial fishing vessels operating in Gulf of Mexico waters (mean for the 1990s) Country Inshore boats Offshore vessels Total Mexico 28,998 1,332 30,330 US 25,000 8,696 33,696 Total 53,998 10,028 64,026 source: NMFS (1997), SEMARNAP (1997) 24 Catches by the US fleet in the GoM represent 15% of the US national commercial catches and 19% of the value of US landings (NMFS 1997). For Mexico, the commercial fisheries in Gulf waters are remarkably important at local provincial scale and their production accounts for 25% in weight of the Mexican commercial catches. Like the majority of tropical fisheries, most of the fisheries in the Gulf have multispecies catches. There are very few fisheries that catch only one or two species (e.g. octopus, lobster); most catch a fraction of the target species or related species, and the remainder of the catch includes many species from different families. In total, reported Gulf landings from commercial fisheries are close to 1.2 millions tonnes • year"1 (NMFS 1997; S E M A R N A P 1997), of which 73% ( « equivalent to 895,000 tonnes) is taken by US fleets. The US state with the highest catch is Louisiana (73%). Of Mexican Gulf landings (close to 325,000 tonnes-year"1) about 72% corresponds to demersal organisms (Ramos-Miranda, 1991). The Mexican state with the highest catch is Veracruz (39%). Average landings in the GoM from 1980 to 1997 are shown in Table 2.5. Fig 5 shows that total G o M landings did not experience drastic changes during the period considered. However, there is a decrease of 13% from 1. 3 million tonnes during the 1980s to about 1.1 million tonnes in the 1990s. US landings showed major decrease, while Mexican landings tended to increase slightly. Table 2.5. Average reported annual landings (t) for the GoM, 1980-1997 a ) . a) State 1980-89 1990-97 1980-97 Alabama 12,164 11,045 11,667 Florida 59,297 49,993 55,162 Louisiana 728,477 568,852 657,533 Mississippi 155,037 88,742 125,573 Texas 46,223 44,172 45,312 US Subtotal 1,001,199 762,803 895,246 Tamaulipas 50,801 61,323 57,276 Veracruz 108,979 135,881 125,534 Tabasco 33,516 47,402 42,061 Campeche 67,021 69,762 68,708 Yucatan 36,067 47,827 43,304 Mex. Subtotal 296,384 362,196 325,124 GoM Total 1,294,682 1,124,999 1,220,369 source: NMFS (1997), SEMARNAP (1997) 25 •2 0.6 -a « 0.4 -h i 0.2 -0 i i i i i i i i i 1980 1982 1984 1986 1988 1990 1992 1994 1996 Years Fig. 5. Total commercial landings in the Gulf of Mexico, 1980 to 1997. Shrimp fisheries are the most profitable fishery in both countries because of the high market value of their catch, though they account only for 10-12% of total annual landings (^124,000 tonnes). Small coastal pelagic fisheries (e.g. menhaden and herring) are the largest in weight followed by shrimps, large decapods (e.g. crabs and lobsters) and molluscs. Among fish groups, demersal fish are the second highest in terms of weights caught. Cephalopods (e.g. octopus and squids) and miscellaneous other fishes are also caught (Table 2.6). A decreasing trend in landings per year is evident for small pelagic fish and shrimp. Landings for some resources, such as reef fish and cephalopods, have doubled in the last 12 years (Fig. 6). Table 2.6. Average annual GoM landings from commercial fisheries from 1985 to 1997 by species a-Species Average landings (t x 10 3) % Molluscs 40.0 4 Shrimps 123.8 12.5 Large decapods 44.1 4.4 Cephalopods 13.8 1.4 Small pelagic fishes 659.3 66.4 Medium and Large pelagic fishes 33.8 3.4 Demersal fishes 37.0 3.7 Reef fishes 27.8 2.8 Other fishes 16.9 1.3 0 source: NMFS (1997), SEMARNAP (1997) 26 Fig. 6. Total commercial landings by fisheries in the Gulf of Mexico during 1985-1997. Secondary y-axis is for small pelagic fish only. Recreational fisheries Recreational fisheries are centered at major tourist destinations in the US and Mexico. More than 3,700 public outdoor recreation sites exist in the US Gulf, with fishing grounds covering over 22,500 km 2 . Thirty percent of the national marine angling trips occur in Florida, Alabama, Mississippi and Louisiana waters (NMFS 1997). The majority of fishing grounds are located in the Mississippi delta region and on the west coast of Florida. In Florida, four types of fisheries boats that take paying passengers operate: inshore and offshore charter boats, guide boats, and head boats (Browder et al. 1981). On average, 3,491 boat fishing weeks are completed every year in this fishery and approximately 58 percent of effort takes place in waters deeper than 18 m (Brasher and Palko 1995). In these fisheries, the number of trips increased by 4 % from 1993 to 27 1997 and the number of participants in this fishery increased by 8 % in the same period (NMFS 1997). Several Mexican Gulf states attract recreational fisheries, including Yucatan, Veracruz, and Tamaulipas. However, there is no record of fishing effort, tonnage caught, or of species caught in those areas. Sixty six percent of recreational landings are recorded in Florida and 19 % in Louisiana. Florida recreational fisheries target small pelagic species such as dolphin, billfish, king mackerel, Spanish mackerel, little tunny, and reef species such as groupers (Browder et al. 1981; Brusher and Palko. 1995). Recent NMFS estimates of catches from recreational fisheries is 11,175 tonnes year"1; however, from 1980 to the present, landings by this fishery have tended to increase. Total landings have doubled from 1980 to 1997, in an average increment of 1,000 tonnes per decade. 2.1.10. Bycatch and discards As mentioned before, the Gulf of Mexico is located in the West Central Atlantic. This region has been recognized as one of the world's worst regions in terms of total bycatch (incidental catch) and discards (portion of the catch returned to the sea) by weight (« 1,601,000 tonnes discarded-year"1). Various gear used by commercial fisheries in the area are known to have bycatch and discards e.g. fish trawls, pelagic longline fisheries, bottom longline and commercial hook-and-line (Burrage et al. 1997). However, the most unselective fishery in the Gulf of Mexico is shrimp trawling, where the average ratio of fmfish bycatch/ shrimp is 10:1 and where 97% of this bycatch is discarded (Alverson et al. 1994). This lack of selectivity has been attributed to the small mesh size used, and the high abundance of non-target species in shrimp habitats (Upton etal. 1992). Information about bycatch and discards in the area is more readily available for US than for the Mexican fisheries. Nichols et al. (1987, 1990) estimated the annual bycatch from the shrimp fishery in offshore US waters to be between 180,000 and 450,000 t. This estimate is based on commercial vessel information from the mid-1970s and early 1980s. Most bycatch and discarded species are Atlantic croaker, spots, seatrout, catfish, diverse crustaceans, molluscs, dolphins, and sea turtles. The bulk of captured fishes are juveniles. Although the shrimp fishery is also conducted in inshore waters, there are no records of discard values or species discarded by this 28 activity. There is virtually no knowledge about how this flux returned to the sea from inshore and offshore shrimp fisheries is recycled into the system. 2.1.11. Status of stocks The current status of most of the stocks in the GoM is not well known, owing to lack of specific fishery-effort records, landings data, and knowledge of population dynamics needed for traditional stock assessment studies. However, it generally appears that fishing effort is intense and mortality levels are high. Several Gulf of Mexico stocks are overexploited. This is particularly serious in most reef and shallow coastal fish stocks, where the overexploitation includes a diversity of species. The declines in stocks of coastal shark in Yucatan and of reef fish in the Florida Keys are just two examples (Alvarez 1988; Bonfil 1990; Ault et al. 1998). Evidence of overexploitation of some species is shown in Table 2.7. Table 2.7. Some G o M resources showing evidence of overexploitation. Species Region Reference Ephinephelus morio Yucatan shelf Gonzalez-Cano et al. (1994); Arreguin-Sanchez et al. (1987b) Octopus maya Yucatan shelf Solis (1994) Strombus gigas All Caribbean region Alevizonetai. (1990) Orthopristis chrysopterus Yucatan shelf Arreguin-Sanchez (1987) Penaeus spp. Southern and northern gulf Arreguin-Sanchez et al. (1995); Gracia et al. (1995); Upton et al. (1992) Rangia cuneata South estuarine system Arreguin-Sanchez etai. (1994, 1995) Strombus gigas Yucatan shelf Arreguin-Sanchez et al. (1987b); Solis (1994) Scomberomorus cavalla South West Florida shelf GMSAFMC(1985) Scomberomorus maculatus South West Florida shelf GMSAFMC(1985) Xiphias gladius South West Florida shelf USDOC (1989) Thunnus thynnus South West Florida shelf Russell (1989) 29 2.2. Models applied to the Gulf In a previous section, I discussed some relevant patterns of increment in fishing pressure and evidence that several fishery resources are overexploited or approaching that level in the Gulf of Mexico. Since the combined effects of physical events and human activities influence long-term sustainability of the area, it is necessary to gain a deeper understanding of how the system as a whole works and how human activities might affect it. Several approaches have been applied to understand the complexity of the Gulf of Mexico ecosystem. Various models have addressed physical characteristics, such as circulation or river discharge of nutrients, in the ecosystem (Wiseman 1997; Alexander et al. 1997). Others coupled general biological and physical characteristics in local subsystems such the Louisiana-Texas shelf (Chen and Wiesenburg 1997) or in larger sections of the Gulf (Richards et al. 1989; Dagg et al. 1991). Richards et al. (1989) emphasised the importance of the interactions of oceanographic and meteorological characteristics (ocean climate) with the biological productivity and landings of the US Gulf region. Richard's model described the northern Gulf productivity affected by Mississippi River discharges, and described the most relevant zooplankton and fish groups interacting in food chains in that region. Finally, two models concerning fishery and ecosystem impacts have been built. Sheridan et al. (1984) built a model that emphasises ecological interactions between penaeid shrimp and bottom fish assemblages of the north-central nearshore GoM ecosystem. Analysis of this model indicated no long-term impact on shrimp stock when discards are reduced, i f the bottomfish do not prey selectively on shrimps. Brown et al. (1991) built a biologically detailed model for the Mexican and US Gulf continental shelf to recognise the impact of currently increasing fisheries on prey and predators. They concluded that predator species commonly caught in Gulf fisheries (e.g. snappers, groupers, sharks, drums and tuna) may be consuming approximately 78% of the estimated biomass of prey species (e.g. small mackerels, anchovies, menhaden, etc.), while the fisheries catch about 15% of the same resources. This indicates that the production rates (P/B) of these prey species would have to be very high to maintain stability. 30 Since then, several studies have addressed the interactions of the biological components in particular Gulf subsystems. Those studies often assume mass-balance to simplify the analysis of such complex systems. Relevant models are those built with the Ecopath approach. 2.2.1. Ecopath in the GoM The Ecopath approach and software have been widely used for different regions of the GoM. So far, 11 mass-balance models have been built for the Gulf of Mexico continental shelf sub-ecosystems. Table 2.8 summarizes key features of these models. The models representing those areas could be divided into 1. Estuarine, 2.other coastal areas (from 0 to 20 m depth), and 3. soft-bottom areas (from 20 to 200 m depth). In total, there are five models for estuarine areas (TAM-4, MAN-4 , CEL-6, TER-4 and TAP-4), one coastal model for a non-estuarine/non reef area (SWG-4), one model for coral reefs (FLO-1) and four models for soft bottom systems (Fig 7). The total area represented by all models is approximately 391,000 km , which corresponds to almost 66 % of the total continental shelf and close to 24% of the total gulf area. Each model emphasised different aspects of the systems. For modelling purposes sub-ecosystems were defined with respect to fishing grounds e.g. Celestun coast (CEL-6), South-western Gulf (SWC-4), or using well defined natural 'limits' as in the case of estuaries e.g. Tamiahua (TAM-4), Mandinga (MAN-4), or some shelf sections of the systems (SWS-4). Most models were published in the early 1990s (Christensen and Pauly 1993) and more recent years (Venier and Pauly 1997; Rosado-Solorsano et al. 1998; Manickchand et al. 1998); however, the data used here originated mainly from the late 1980s and early 1990s. 78% of input data were obtained from local or regional (Gulf) studies. Biomass often was calculated using the swept area method or by fixing the value of ecotrophic efficiency and leaving the software to estimate it. Other input parameters, such as Q/B, were generally calculated through the use of empirical formulas (Pauly 1986; Palomares and Pauly 1989). Finally, P/B and diet composition were commonly taken from local assessments or from other similar Gulf or tropical ecosystems for which these values had been calculated (Appendix 2). Models from systems with accurate landing information, usually reported by local fishery agencies, included catches. For some small estuaries represented (TAM-4, MAN-4) catch information was unavailable. The reef model (FLO-1) did not incorporate catches since the Looe Key reef is a National Marine Park, with very controlled fishing activities. Discards of fisheries were only included in the model of 31 Browder (1993) for the continental shelf (NOR-1); however, this information was incorporated within the catch data. Functional groups for shelf and coastal GoM models were selected using diverse criteria. Often, fish groups were selected as the most abundant families captured during surveys or abundance studies (TAP-4, SWS-4, FLO-1). In other cases, only fish susceptible to being caught by specific fishing gear such as demersal trawls or purse seines were considered (SWC-4, CAM-6). In the Tamiahua lagoon (TAM-4) model, only fish species considered residents in the area were included. In all cases the commercial importance of those abundant fish species was always emphazised. Functional groups of invertebrates included species consumed by fish, those abundant in survey studies, or commercially important species. As a result, a large variety of functional groups were represented in the models. Groups from neritic systems are well-represented, especially primary producers (phytoplankton and benthic producers), miscellaneous invertebrates (molluscs and annelids), crustaceans (crabs and shrimps), small prey demersal and pelagic fish, and miscellaneus demersal and pelagic predators. A l l models included only one detritus group. The under-represented groups in the models are large zoobenthos feeders (rays), large predators (sharks), large pelagics (tuna, billfish), marine mammals and sea turtles. A l l the models used average annual information for input parameters; although some authors recognised that the systems represented show strong seasonality (e.g., YUC-6 , FLO-1). Energy units used in models vary from grams of Carbon (MAN-4) to grams dry or wet weight. In the process of mass-balancing each model, some adjustments were made to the input data. In cases when one basic parameter was absent or was found to be unrealistic (often biomass estimates), it was left unknown so that the software could calculate it using equation 3 (section 1.6.1.) The diet matrix was slightly modified when diet information from other systems was used. Gaps in local knowledge were evident for some ecosystems causing some challenges when building the model. A summary of these gaps is as follows: 32 • Coastal models only; Absence of knowledge of detritus imports from other systems; therefore it was necessary to include a best approximation of detritus imports to reach a balance in the system (YUC-6, SWC, CAM-6, CEL-6, FLO-1 and MAN-4); In some systems, catches are unknown and therefore not considered in models (TAM-4 and MAN-4). • Soft bottom models only: A paucity of local knowledge of pelagic components of the systems; those groups are represented in a rough form (almost all models). • Both types of models: - Unknown exchange flux of migratory species from other shelf systems; migration was not included (all models); Unknown sources of production from other local sources and adjacent systems (estuaries, mangroves or upwelling systems); therefore, these were not incorporated (e.g. MAN-4, SWC-4, CEL-6 and YUC-6 ); Few local studies of fish diet composition; several diet references used correspond to similar, but not local systems; Some other information gaps caused no problems when the models were built, but limited interpretations owing to the importance of the omitted groups. They were as follows: - Higher vertebrate groups such as marine mammals, sea birds and sea turtles are poorly represented or not represented at all; - Discards from fisheries are poorly included. Although several models were considered preliminary, a large amount of information was obtained from them and their analysis has lead to a better understanding of the areas modelled. Important insights were found such as high coupling of benthic and pelagic components in the shelf system (SWS-4) and the impact of reopening prohibited fishing methods or the consequences of intensifying them in reef systems (FLO-1) were indicated. The analyses also indicated the importance of detritus and detritivores during balancing of coastal systems models 33 (TER-4, SWGc-4). Also, estimates of the quantitative contribution of primary producers in an overall coastal system were obtained from one model (SWC-4). In spite of challenges in construction of the integrated GoM model, the incorporation of information provided by these models and the filling of important gaps in them must serve to increase the understanding of how the entire Gulf of Mexico ecosystem works. The synthesis of this information is presented in the following chapter. 34 o 3 a a, D. ca CU -a o E .c CS a o o W x> a H >- S Q.I •9 '-s S a S M l Z « W cu 3 O c <u o Q bO Q ca 3 OS OS XI < < C o o bo ca 1 SC o CN o e cu O Q U 00 fr 3 CU 3 < I N S u o o bO ca J <c o z 3 z < s 3 C CO U z E «• CJ JS IS •£ J3 O 1_ cu cn cu cn CU J3 o E o u -3 "9 •g Co I - „ °o — c <*> CU ~ ^ O O M Q & W "So o CN cu 60 3 -S CJ X ) cu _ CO O CU JZ c/> X> D. •J= 3 O. O cu l— o SB c/l 3 w w CU cu -ti « 2 o k •£ | t/J <u £ • J= •SI'S I* § s « I cu rti OS Os Q 3 oo OS Os U SC i T3 C Pi W H c o o bO 00 o z fr 3 Os N cu > ca Xi O W u c o o be cs! H oo w w o CN CS b 00 .£ 3 T3 cu I ca a. Q cu w C ca a o 3 <*H *-* CU s a (3 C o o c c ca rr, Os Os ca N CU J3 o C ca oo 3 bo cu t < ca o o W H w SC H O o c o bo o ca rs J= a o cd C o CU ^3 w 1 S c ca E 5 w u Os Os O -J CJ CJ o o —1 < Q 2 o on c o o ts o J2 o OS Os T3 o m oi o z —I a <C H z CP Z H z o u >s T3 J3 C cj cu 3 > tS 'C C •a -c § o o & E o n. ce E . S3 B -w cfl to •S.J5 i E M C CJ cu W cb h "5b o ti o x> o OS Os cu C ca E Ji 'S S C I § J3 i w H oo W o oo c o ctl I 8 bo Q cu J3 co E o ti o J3 <S o Os Os •a c cu U • ca oo cj > < z < CQ w sc u M OH s < u cu c o ~bh E o ts o JO o XI m Os OS <u N CU x ; o c 3 00 CU t; < u D ffl SC 00 Z < < U CHAPTER 3. SYNTHESIS OF A MASS- BALANCE MODEL FOR THE GOM This chapter describes the methodology I used to build a mass-balanced model representing the entire Gulf of Mexico ecosystem in the late-1980s and early-1990s. The chapter is divided in two major sections. The first section (3.1) shows the process of building a spatially structured model that incorporates some information mentioned in Chapter 2, as well as information proceeding from previous models of subsystems. The second section (3.2) describes the results from the integrated model and the characteristics obtained for the Gulf of Mexico ecosystem. 3.1.1 Definition of the area modelled In the previous chapter, the spatial complexity of the Gulf of Mexico was described. To build a representative model, it was required to define and estimate the total area in the Gulf, and to estimate sub-areas by depth and subsystems. Fig.8 shows the total area defined with the following limits: from Cabo Catoche in Yucatan (21° 53' Lat. N . and 86° 57' Long. W) to Punt Perpetua on the western extreme of Cuba (25° 12' Lat. N and 84° 57' Long. W) to close the Yucatan Channel and, from Bahia Santa Clara in North Western Cuba (22° 59' Lat N.and 80° 28' Long. W), to Long Sound in Florida Bay (25° 12' Lat. N . and 80° 28' Long. W) to close the Florida Strait. These limits were chosen for simplicity reasons since I assumed that main internal fluxes of the ecosystem are expressed in that area. 3.1.2 Estimation of areas by depth and subsystems Total Gulf area and total Continental shelf area (from 0 to 200 m depth) were estimated from a digitized US map with a scale of 1:2,750,000 (Dept. Interior U.S. Geological Survey, 1956). Calculated areas were compared with independent estimations using GEBCO- Digital Map (Carocci, unpublished data). Areas for US and Mexico territorial waters located at less than 200-m depth (intervals of 10, 50 and 100 m) were determined from digitalized topographic maps with bathymetric contours of US and Mexican coasts (Dept. Interior U.S. Geological Survey, 1956; Mexican Military Dept, 1958). For Cuba, this information was not available but relative 37 (N O) apmutn proportions from the other two countries were used. Table 3.1 shows those estimated areas by depth by country. Table 3.1. Estimated areas (103 km 2) of US, Mexican and Cuba territorial waters by depth intervals. 0-10 10-20 20-50 Depth range( 50-100 m) 100-200 0-200 >200 USA 51 45 91 67 55 310 Mexico 25 28 89 78 53 274 -Cuba 1 1 4 4 2 12 -Subtotal 78 74 184 150 110 596 1,026 Total . . . . . . 1,623 To estimate areas by subsystems, I subdivided them into five categories, using depth and ecological characteristics as described in section 2.5: estuaries, shallow non-estuaries/non reef areas, coral reefs, soft bottoms, and deep ocean region. The first three subsystems are in coastal regions bellow 20-m depth. Soft bottoms cover mainly all the continental shelf area from 20 to 200 m depth. The largest subsystem in the Gulf is the oceanic region deeper than 200 m. Table 3.2. shows the estimated areas of the five subsystems considered. The estimation was based on some reported data but also based on some assumptions. First, estuarine areas were obtained using independent reports from several lagoons and gross estimations from US estuarine water surface and Mexican estuarine areas (NOAA 1990; Ibarra-Obando et al. 1997). From this, total estuarine area is estimated to be about one quarter of the total area below 20-m depth. Secondly, since complete estimations about coral reef extensions in the area were not available, some reports and an arbitrary assumption (not more that 1% of the total shelf area are reefs) was used to estimate 2,500 km 2 for these systems. Finally, shallow non-estuaries/non reef areas were calculated as the difference of total coastal area minus estuarine and reef areas, while, soft bottom areas were estimated to be the rest of the continental shelf. The oceanic area was the estimated area greater than 200m depth. In absence of detailed information about Cuban subsystems, relative proportions from the other two countries subsystems were extrapolated. 39 Table 3.2. Estimated areas of subsystems in the Gulf of Mexico. Subsystem Depth Area (m) (10 3 km2) Estuaries 0-20 34 Shallow non-estuarine 0-20 119 Coral reefs 10-50 2 Soft bottoms 20-200 441 Open ocean >200 1,026 Total Gulf 0 - 3,400 1,623 3.1.3 Subsystems represented by previous models In section 2.2.1, I referred to the models so far built to represent subsystems in the Gulf of Mexico. For simplicity reasons, I considered only mass-balance models. Since not all the Gulf area is explicitly covered by those models, it was necessary to expand each model to represent sub-areas for the total area. For this purpose, I used a modified classification of GoM provinces made by Antoine (1972) using geologic characteristics (section 2.1.4). The modification included the Cuban region as province 7 and the oceanic region was then defined as province 8. After this classification, I assigned a code to each model. The letters correspond to the name of the system and the number corresponds to the modified Antoine's province in which the system is located (e.g. Tamiahua model in province 4 has code TAM-4). Since only 9 models were available at the beginning of this study, these were used to represent the total Gulf area. The last two models (TAP-4 and SWS-4) were incorporated later. Most of the models did not provide estimations of the area modelled; to obtain them, oceanographic characteristics found in the literature and those mentioned in Chapter 2 were used (Table 3.3). Table 3.3 Total subsystem area (km2) represented by each model. Model Subsystem represented * Subsystem area Represented area TAM-4 Estuary 2-4 800 11,823 MAN-4 Estuary 2-4 5 11,823 CEL-6 Estuary 1,6-7 28 8,318 TER-4 Estuary 5 2,500 2,014 SWC-4 Non-estuary 1-7 22,203 118,620 FLO-1 Coral reef 1,3,4,6,7 18 2,500 NOR-1 Soft-bottom 1-4 212,402 253,816 CAM-6 Soft-bottom 5-7 102,186 93,738 YUC-6 Soft-bottom 5-7 51,178 93,738 Total Continental shelf 391,320 596,391 *numbers correspond to the provinces modified from Antoine (1972) 40 Since the G o M deep oceanic subsystem has not been modelled with the Ecopath mass-balance approach, a modified version of a generic oceanic model, earlier used to represent the central South China Sea (Pauly and Christensen, 1993), was used to represent this ecosystem. Once the models were assigned to represent all areas in the Gulf, it was necessary to define representative functional groups, standardise the units and integrate the information provided. 3.1.4. Definition and description of functional groups Some important functional groups well represented in the models were taken directly from the subsystems models and were kept independent (e.g. octopus, other decapods, etc). Others were considered independent stocks, although it is known that they can move to adjacent waters (e.g. some reef fishes). Finally, others were incorporated to represent groups not considered by previous models (e.g. oceanic fish groups, sea birds and planktivorous marine mammals). As well, given the large amount of discards in the area reported by various authors (section 2.10), another functional group called 'dead discards' was included. Some criteria used in the aggregation of functional groups were common for all community members. These are: 1) ISSCAAP codes and Taxonomic categories as by Pauly and Christensen (1993); 2) habitat type (specified for each group); 3) size (specified for each group); 4) food habits: planktivores, herbivores, detritivores, invertebrate feeders, piscivores or carnivores and omnivores and, 5) ontogenetic stages: juvenile and adults. Other criteria were specific for non-fish or fish functional groups and they will be mentioned when these groups are described. Criterion number five, recently incorporated in the Ecopath approach, was included in this study to be able to distinguish different ontogenetic states of some functional groups (Walters et al. 1997). This criterion was applied in 4 fish groups separated out into juvenile and adults (i.e. grunts, snappers, tuna and sharks groups). Shrimps were also dis-aggregated, to represent the differential impact of fishing fleets. In all those cases, it was assumed that the juveniles occur near the coast and the adults occur deeper onto the shelf. Initially, it was necessary to identify and aggregate synonyms by taxa (specific names, generic name or common name) used by different authors. Aggregation of species within generic groupings also had to be identified. 41 Of the forty functional groups used in the integrated model, half were non-fish groups. Three primary producers (inshore and offshore phytoplankton and benthic plants) were identified as well as four secondary producers (zooplankton groups), 29 consumers groups with different intermediate trophic levels, two top predators (oceanic sharks and piscivorous marine mammals) and two detritus groups (detritus and bycatch). A general description of functional groups is given below, but specific components of each can be seen in Table 3.4. 3.1.4 a. Non- fish groups Twenty non-fish groups were incorporated in the model. Specific criteria to aggregate these groups were: 1) habitat type: inshore and offshore in planktonic groups where the spatial distribution could reflect different stocks and sustain different sub-trophic webs; 2) size: benthic organisms < 0.5 mm and benthic organism > 0.5 mm; and 3) food habits: herbivores and carnivores in planktonic groups, and planktivores and piscivorus among marine mammals groups. Each group will be described in the following sections (the number in parentheses categorises each functional group). Benthic producers (1) A l l cyanobacteria, benthic diatoms and macroalgae encrusted, attached to or lying on benthic substrate are considered benthic primary producers. Most common types of macroalgae are: green algae (Chlorophyta), brown algae (Phaeophyta) and red algae (Rhodophyta). When brown and green macroalgae are alive, they are grazed, used as nursery areas, or used by a variety of organisms for protection. They also reduce wave action along some coastal areas. When these algae are fragmented they are consumed by detritivores and macrofauna. Red algae are the principal component of coral reefs in Caribbean regions. Moore and Wetzel (1988) estimated the biomass of Thalassia testudinum standing crop in the Terminos Lagoon to be between 1,000 to 1,500 g W W m"2. Also, they calculated the primary productivity of the Thalassia in 1,300 tonnes year"1 in the same area. 42 Table 3.4. Functional groups o f the synthetic G u l f of Mexico model Group Name Members Common species 1 Benthic producers Chlorophyta, Phaeophyta and Rhodophyta Thalassia testudinum 2-3 Phytoplankton (inshore and offshore) Diatoms, dinoflagellates and Cianophyceae (algae) Chaetoceros spp., Pendulus spp., Phyrodinium spp. 4 Meiobenthos Nematodes, turbelaria, copepods, several minor phyla 5 Macroinfauna Smal l molluscs and invertebrates, polychaetes 6 Macroepi fauna Large molluscs and echinoderms, sponges,sea cucumbers, seastars Busycon spp. 7-8 Herbiv. Zooplankton. (inshore and offshore) Copepods ostracodes, mysids, sergestids, euphausiids Acantia tonsa, Centropages fiircalus, Eucalanus spp. 9-10 Carnv. Zooplankton. (inshore and offshore) Ichthyoplankton (fish and invertebrate larvae), chaetognaths , annelids Sagitta spp. 11-12 Shrimp (juvenile and adult) Penaeids Penaeid aztecus, P. seliferus, P duorarum 13 Other decapods Lobster and crabs Callinectes sp., Menippe mercenaria, Panulirus argus 14 Octopus Octopus spp. Octopus maya, 0. vulgaris 15 Coastal small planktonic feeders Clupeidae (Herring, menhaden) Engraulidae (anchovies) Juv Carangids and Lutjanids, small planktivores (squid) Opistonema spp., Harengula spp., Sardinella spp. Anchoa spp. 16 Coastal small demersal invertebrate feeders Ari idae(catfish) Gerreidae (mojarras) Mu l l idae (goatfish) Scianidae (croakers) Tetraodontidae Ostraciidae (cowfish) Arius felis Diapterus spp. Mulloidichthys martinicus Bardiella spp. Lagucephalus spp. Acanthostraceon spp. 17-18 Coastal medium invertebrates and small fish feeders (juveniles and adults) Haemulidae (grunts) Paralichthydae (flounders) Sparidae (pinfish) Scianidae (drums and seatrouts) Bal ist idae Centropomidae (snooks) Orthopristis spp. Paralichlhys lethostigma Lagodon spp. Sciaenops ocellatus leatherjacket Centropomus undecimalis 19-20 Coastal large demersal carnivores, (juvenile and adults) Serranidae (groupers) Lutjanidae (snappers) Dasyatidae and Myl iobat idae (rays) Ephinephelus spp. L. campechanus, L. griseus, Rhomboplites spp. 21 Smal l coastal detritus and plant feeeders Gobiidae(gobies) Mugi l idae (mullets) Sparidae (porgies) Mugil spp. Archosargus spp. 22 Med ium coastal fish feeders Synodontidae (lizardfish) Coastal sharks and juveniles o f large species Saurida spp., Synodus spp. Rhizoprionodon lerraenovae 23 Large predators Oceanic sharks Alopias sp., Isurus oxyrinchus 43 Table 3.4. continuation Group Name Members Common species 24-25 Large shelf pelagic predators, (juvenile and adults) Istiophoridae (billfish) Scombridae (tuna, mackerel) Sphyraenidae (barracuda) Carangidae Coryphaenidae (dolphins) Makaira nigricans, Tetrapturus albidus Thunnus sp., Scomberomorus cavalla Sphyraena barracuda Caranx spp., Trachinotus spp. Coryphaena hippurus 26 Large reef planktivores Carangidae Holocentridae (soldierfishes) Innermiidae Pomacentridae Serranidae (creole-fish) Decapterus macarellus Myripristis jacobus Inermia vittata Chromis cyaneus, C multilineatus Paranthias furcifer 27 Other fishes Batrachoididae Cyprionodontidae Hemirhamphidae Odontaspidae Perciformes Poecilidae 28 large reef carnivores Aulostomidae (trumpetfish) Balistidae (filefish, triggerfish) Belonidae (needlefish) Diodontidae (porcupine, balloonfish) Holocentridae (squirrelfish) Labridae (hogfish) Muraenidae (moray) Pomacanthidae (angelfish) Serranidae (coney, red hind) Sparidae (porgies) Aulostomus maculatus Aluterus scriptus Strongylura notata Diodon hystrix Holocentyrus ascensionis Bodianus rufus Echelycore nigricans Pomacanthus arcuatus Epinephelus fulvus Calamus ba/onado 29 large reef herbivores Balistidae (filefishesh) Kyphosidae (bermuda chub) Scaridae (parrotfishes) Aluterus schoepfi Kyphosus sectatrix Scarus coelestinus 30 small reef carnivores Callionymidae (lancer dragonet) Chaetodontidae (butterflyfish) Cirrhitidae (redspotted hawkfish) Clinidae (blennies) Labridae (wrasses, razorfishes) Ostaciidae (trunkfishes) Serranidae (perches, hamlet, bass) Tetradontidae (puffers) Malacoctenus gilli Halichoeres bivattatus Lactophrys bicaudalis Diplectrum formosum Sphoeroides spengleri 31 small reef herbivores Acanthuridae (surgeonfishes) Blenniidae Gobidae Pomacentridae (damselfishes) Scaridae (darrotfishes) Acanthurus bahianus Scartella cristatus Coryphoplerus dicrus Pomacentrus diencaeus Sparisoma alomarium 32 small pelagic predators Scombridae (Spanish mackerel, bonito, little tunny) Pomatomidae (bluefish) Carangids Belonidae (needlefish) Elopidae (ladyfish) Sarda sarda, Euthynnus alletteratus Pomatomus saltatrix Seriola fasciata Strongylura spp. Elops saurus 33 Mesopelagics Myctophids, gonostomatids and stemoptychids 34 Bathypelagics Anglerfish and Cyclotone 35 Seabirds Pelicans, gulls, storm petrels, terns and jaegers Oceanodroma spp., Chilidonias spp. 36 Pisciv. m. mammals Odontocetes (dolphins) Tursiops tructatus, Stenella spp. 37 Plankv. m. mammals Mistycetes (whales) Eubalaena glacialis, Balaenoptera spp. 38 Sea turtles Chelonia mydas, Caretta caretta, Lepidochelys spp Eretmochelys imbricata, Dermochelys coriacea 39 Detritus P O M and D O M 40 Deaddiscards discards from shrimp trawls 44 Phytoplankton (2-3) Steidinger (1973) delimited four broad phytoplankton assemblages for the eastern Gulf of Mexico based on depth strata: 1. Estuarine; 2. Estuarine-coastal; 3. Coastal-open ocean and, 4. Open ocean. Although the first strata show strong overlap among their species, some generalisations could be made. In estuarine and coastal waters the numerically dominant components are microflagellates (<15 pm) and nanoplankton; while, in coastal and open gulf waters eurythermal and euryhaline dinoflagellates and diatom species are common. In the open Gulf, tropical and subtropical oceanic species of dinoflagellates (such as Amphisolenia, Heterodinium, and Pyrocystis), diatoms (such as Ethmodiscus, Gossleriella and Planktonielld) and cyanobacteria (Trichodesmium) are present. In deeper layers of the Gulf coccolithophorids are also very abundant. Diatoms (Class Bacilloriophyceae), dinoflagellates (Dinophyceae) and Cyanophyceae algae are principal components of phytoplankton along the Mexican Gulf coast. In lagoons, the most common and abundant diatom genus is Chaetoceros, with 29 species. Among them, 75% have neritic origin and the rest have oceanic origin. Most common species are: Chaetoceros diversus, C compressus, C. affinis, C. didymus, C. pendulus, C laciniosus and C. danicus (Torres 1987). Common dinoflagellate species are: Pyrodinium, Ceratium fusus and C. pentagonum (Gomez-Aguirre 1989; Ochoa and Ramirez 1987). Cyanophyceae algae are very abundant in shallow calm waters where there are high nutrient imports. Numerical abundance of phytoplanktonic species in the coastal open Gulf is often one to two orders of magnitude lower than in the estuarine regions (Vargo et al. 1990). In Mexican coastal lagoons, densities show large variation, ranging from 8 to 5.5 cells/ 1 x 106 in southern states (Tabasco and Campeche) to 463 cells/ 1 x 106 in the western region of Veracruz, Tamiahua lagoon (Licea et al. 1987; Santoyo 1994). Net phytoplankton production calculated in the Terminos lagoon was about 690 g WW m"2 y _ 1 (Day et al. 1988). In US waters, the highest surface phytoplankton stocks have been found in the discharge plumes of the Mississippi and Atchafalaya Rivers and within the coastal boundary. Concentrations of shelf phytoplankton are higher in Louisiana waters than in Texas waters (Dagg et al. 1991). 45 For this study, two phytoplanktonic groups were defined to maintain some degree of differentiation between inshore and offshore waters. Again, inshore species correspond to shallow strata less than 20-m depth; and offshore species correspond to waters deeper that 20-m. Meiobenthos (4) I separated benthic functional groups into three: meiobenthos, macroinfauna and macroepifauna (Day et al. 1989). The Meiobenthos group includes microfaunal and meiofauna organisms and consists of very small organisms (1 to 1000 um) such as nematodes, harpacticoids, copepods, turbelarians and several other minor phyla that live in spaces between sand grains. This group plays an important role in the decomposition of detritus on shallow estuary bottoms (Day et al. 1989). Burnett (1981) found a biomass higher than 100 g/m2 of microfaunal organisms in shallow waters and between 2 to 5 g/m2 in deep waters. Meiobenthos biomass from other west Atlantic coasts (South North Carolina, USA) was estimated to fluctuate from 1.6 g/m2 between 250-750 m to 0.1 g/m2 at depths greater that 2800-m (Coull et al. 1977). Macroinfauna (5) Organisms that live underneath the sediment surface in tubes or burrows, and larger than 0.5 mm constitute this group. Most of them are decomposers. Representative components are some bivalves, many polychaetes, and nematodes, among others. Biomass estimates helped identify two regions in the GoM: highly productive areas close to the shore with macroinfauna biomass 2 2 of 10-300 g/m , and more extensive shelf regions with biomass values between 0.1 to 10 g/m (Zenkevitch al. 1971). Macroepifauna or epibenthos (6) Macroepifauna are those animals living on the surface of the bottom or on elevated surfaces, bigger than 0.5 mm in size. They are mainly suspension feeders. Some examples are oysters, barnacles, sponges, hydroids, some polychates, tunicates, echinoderms, and several molluscs. Perez-Rodriguez (1980) reported 307 species of gastropods and pelecipods in the Gulf of Mexico and Caribbean continental shelf. Vazquez-Bader et al. (1994) reported 28 species (13 families) of molluscs and 22 species (11 families) of echinoderms in soft bottoms of the Mexican continental shelf. In that study, scallops (Pelecipods) were the most numerically abundant of the 46 mollusc groups, and Gastropods were the most diverse groups regarding species. Among echinoderms, the most abundant and diverse groups were sea stars. Zooplankton (7-10) Zooplankton is a large assemblage of microscopic organisms living in the water column either during only part of their life cycle (meroplankton) such as fish and invertebrate larvae; as permanent residents (holoplankton) or, accidental/temporary (tychoplankton) including benthic forms of ostracods, nematodes and, more rarely platyhelmints. Vargo et al. (1990) gives a detailed description of the most abundant zooplanktonic components in the Gulf. Main components of the oceanic zooplankton communities are holoplanktonic species such as calanoid copepods (Clausocalanus, Euchaeta and Eucalanus), euphausiids (Euphausia and Stylocheiron), Chaetognaths (Sagitta enflata) and planktonic decapods such as Lucifer faxoni. Holoplankton and meroplanktonic species are abundant in shelf waters. Among the holoplankton, epipelagic oceanic copepods (Arcocalanus, Temora, etc), euphausiids (Euphausia), calanoid copepods (Centropages, Eucalanus, Labidocera etc) and chaetognaths are common. Copepod species composition changes across the shelves (Hopkins 1990). Among the meroplankton, zooplanktonic larvae of shrimps, lobsters, crabs, scallops and squids are abundant. In nearshore areas, copepods (especially Acartia tonsa) are the most abundant (87%) holoplanktonic group (Dagg et al. 1981). Species commonly found in estuaries were: Acartia tonsa, Temora turbinata, T. stylifera, Labidocera nerii, Centropages furcatus, Paracalanus quasimodo and Farranula carinat. The meroplankton group is represented by shrimp larvae and benthic larvae from cirripeds, echinoderms, gastropods, pelecypods and polychaetes. The species composition of theses planktonic communities are often a function of salinity (Vargo et al. 1990) The biomass of zooplankton copepods in the Gulf is higher in low salinity surface waters (e.g. lagoons and close to river deltas) than in oceanic areas. In shallow waters the biomass varies between 0.01 and 0.02 g/m2, and in oceanic waters biomass is about 0.0006 g/m2 (Ortner et al. 1984; Suarez 1994; Ortufio 1998). 47 Another important component of the zooplanktonic community is the ichtyoplankton or planktonic fish larvae, which are typically carnivorous. These organisms are extremely important to determine the recruitment of commercially and non-commercially important species. They are also distributed differentially in oceanic, shelf and coastal waters. However, for the purpose of this study detail in this aspect of the planktonic community is not relevant. This group is considered as part of the carnivorous component of the zooplankton. There are two principal methods of feeding among zooplankton organisms: filter feeding and raptorial feeding. Commonly, animals with the former method are herbivorous and those with the latter mechanism are carnivorous. However, the two process may be found in the same species, especially in crustaceans (Table 3. 5). Table. 3.5. Feeding processes in zooplanktonic groups a ) . Group Feeding processes Protozoa, Ctenophores, Coelenterates and Annelids Raptorial Urochordates Filter feeding Molluscs and Crustaceans Filter and raptorial feeding source: Parsons etal. (1984) Using the relevant information about zooplankton distribution and food habits expressed above, four zooplankton groups were incorporated in the model: herbivorous and carnivorous groups, both split into inshore and offshore components. Shrimp (11-12) Ten shrimp species are abundant in the Gulf area (Upton et al. 1992; Arreguin-Sanchez and Chavez 1985). They are distributed irregularly in shallow (< 40 m) and middle depth waters (41-60 m) associated with different types of bottoms (GMFMC, 1981b; Rulifson, 1981). Among them, only three penaeid species (P. aztecus, P. setiferus and P. duorarum) sustain valuable fisheries in both countries. During different life stages, from postlarvae/ juveniles to young adults/ adults, these organisms sequentially inhabit two basic environments. In early stages, they are bottom-dwelling in estuaries and lagoons and feed on microorganisms (amphipods, tanaids, 48 polychaetes, miysids and copepods) algae and detritus (Minello et al. 1989). As young adults, they migrate to sea with more omnivorous predatory habits. Cannibalism is common among juvenile and adult shrimp (Christmas and Etzold 1977). This resource is a classical example of sequential fishery where the inshore fishery catches early stages while adults are caught by the offshore fishery. Rulifson (1981) estimated the biomass of juvenile penaeids shrimps between 1.05 and 3.07 g m (average 2.06 g/ m2) in different estuarine areas in the western Atlantic region. Minello et al. (1989) suggested that predation is the major direct cause of brown shrimp mortality in estuaries of the northern GoM. Knudsen et al. (1996) estimated monthly instantaneous mortality of the same species in a Louisiana coastal marsh, which converted to annual values range from 6.6 to 15.2/ year. Given these ecological characteristics, I distinguished juveniles and adults shrimp as two different functional groups Other decapods (13) This group includes crabs and lobsters. Brachyuran crabs are the most diverse and abundant decapods in the GoM with 352 species recognized, within 22 families. Crabs are most abundant in the limestone-based shelf than in the sand or mud sustrates (Powers 1977). Vazquez-Bader et al. (1994) reported 86 species of these crustaceans within the soft bottoms of the Mexican continental shelf. They found that families with the most species were Majidae (spider crabs), Portunidae (swimming crabs), Diogenidae and Penaeidae (shrimps). The most numerically abundant species among portunids were Callinectes similis, C. sapidus and Portunus spinicarpus. Callinectes species are very abundant in estuarine systems where they feed on infauna and epifaunal organisms and sustain important local fisheries (Rosas 1994). In US waters, common crabs species are Menippe mercenaria and M adina. The former is commercially exploited in southwest coast of Florida and Texas; the latter is mainly caught as bycach in other fisheries (Perry et al. 1995). Lobsters are the other component of this group. Three species of spiny lobster are recognised in the area: Panulirus argus, P. guttatus and P. laevicauda (Gracia and Kensler, 1980). The first species sustain a very important fishery in the Yucatan shelf. Adults are located in coral reef 49 areas where they reproduce, and juveniles live in shallow vegetated bottoms (Witham et al. 1968). Octopus (14) There are seven octopus species reported for the Mexican Gulf area. Among them, two are exploited commercially in Veracruz and Yucatan: Octopus maya and O. vulgaris. These species inhabit in Thalassia testudinum prairies, caves, or cavities in the sea floor and coral reefs, generally down to 50-m depth (Solis 1994). These animals are carnivores preying on bivalve molluscs, snails, and crabs. This resource sustains one of the most important fisheries along the northern and west coast of the Yucatan Peninsula (Solis 1962, 1994). In US waters, catches of octopus seem not to be very important numerically. Total mortality for these organisms in Mexican waters has increased drastically in the last decade (Solis et al. 1986; Arreguin-Sanchez 1992) due to an increase of directed fishing pressure. Seabirds (35) Harrison (1983) catagorized seabirds as coastal, offshore, and pelagic birds which have their usual habits and food sources in the sea. 46 seabird species, mainly represented in nine families, have been reported in the Northern Gulf of Mexico (Duncan et al. 1980; Peake 1996) Appendix 3. The most abundant species are: Storm petrels (Oceanodroma spp.), Terns (Chilidonias spp.), Gulls (Larus spp.) and Jaegers (Stercorarius spp) based on number of sights. The Family Laridae has the highest variety of species (26). Not all seabirds species are permanent residents in the area; many of them occupy the area during nesting or winter seasons only. Specific numbers of resident or migratory species were not available. Primary food items consumed by most seabirds in the world are small schooling fishes (clupeids), crustaceans, and cephalopods occurring in the upper to mid-water portion of the column. Other common items consumed are demersal fishes in juvenile stages, inshore benthic fishes, carcasses offish, and discards (Barret et al. 1993) 50 Marine mammals (36-37) Thirty-one cetacean species have been reported in the Gulf area (Solis 1995; Waring et al. 1996). This cetacean community has been divided into shelf and oceanic components. Odontocetes mainly occupy shelf regions from depth of <18 m to more than 200 m (Mullin et al. in press). Some Odontocetes and all Mysticetes live in the oceanic region. A l l reports point out that the shelf community is numerically larger than oceanic community (Waring et al. 1997; Mullin et al. in press). For this model, I divided these organisms into two groups: Piscivores or Odontocetes (24 species) and Planktivores or Mysticetes (6 species) See Appendix 4. The most abundant piscivorous mammals are bottlenose and spotted dolphins (Tursiops sp. and Stenella sp.). Among planktivores, there are no precise numerical estimates by species, but blue whales (Balaenoptera musculus) could represent the highest biomass of this group owing mainly to their large average weight. Also, there is a lack of detailed distribution data for Mysticetes group, and for that reason I assumed that they are more abundant in areas deeper than 50 m. Biomasses of these groups were obtained from stock assessment estimations of the population of all marine mammals in the area reported by Waring et al. (1996). Other input parameters were obtained using a weighting method described below for piscivorous mammals, and from the baleen whales in the model by Okey and Pauly (1998) of Alaska's Prince William Sound model. Sea turtles (38) Six sea turtle species have been described in the region (Appendix 5). Although biomass estimations of these animals are unavailable, it is known that numbers of nesting females and landing values in the Gulf have drastically decreased since 1950s due to high exploitation of adults and eggs (Rebel 1974). Since 1978, these animals are protected and considered endangered in the US and Mexico Gulf waters by the Endangered Species Act. As a consequence of their protection, recent studies on the west coast of Florida show that some Kemp's ridley stocks may be recovering: the number of animals has tripled from the late 1980s to early 1990s (Schmid 1998). This author estimated that 159 Kemp's turtles constituted the annual ridleys population at the Corrigan reef, Florida (3.2 tonnes • km ). These animals live primarily in shallow coastal waters with vegetated bottoms (e.g. seagrass beds), and over oyster 51 beds, and coral reefs; some species are known to be more pelagic in deep waters (e.g. leatherback turtle). They commonly consume seagrasses, benthic invertebrates (crabs, molluscs, sponges, and urchins), jellyfish and occasionally fish (mullets) (Mortimer 1981). I therefore incorporated this group into the model not for reasons of abundance, but to emphasise its importance in the ecology of the system. Biomass was left to be calculated by the software and other Input parameters were taken from Polovina and Ow (1983). Detritus (39) Detritus in sea water could be classified by the ratio of its components in dissolved (DOC) and particulate carbon (POC=ca. 1 pm diameter is particulate), and by the nature of the its material in non-living and living components. DOC (or DOM) originates from dissolved primary products from phytoplankton and benthic algae, and it is a major food supply for filter-feeding organisms. POC (or POM) originates from organic matter from excretory processes (excrements and feces) and carcasses of organisms. When POC is deposited at the bottom, it will be available to carrion consumers (fish and invertebrates) and/or decomposed by bacteria. In my model, I assumed that all detritus consumers feed on the living components in the detritus, which according to Gordon (1970) accounts for 14 to 79% of the particulate material in depths above 100 m. A rough estimate of detritus biomass (D, ton/m2) was obtained using a derived empirical equation from Pauly et al. (1993b) as a function of primary production (PP, t C • km" year"1) and euphotic depth (E, in m). However, the input value was modified to allow equilibrium in the model (section below). Dead-discards (40) Discards were included in the detritus group due to the amount and size of the material discarded (chunks of flesh instead of small aggregates) and to emphasise the ecological importance of this flux. This study included only discards from the shrimp fishery, since it is the major contributor to discards in the area (Section 2.9 Chapter 2). A l l discards were assumed to be dead. Seabirds, dolphins, crabs, and a variety offish, including sharks (Hill et al. 1990; Wassenberg et al. 1990) consume discards. Thus, this biomass returned to the sea could be considered some 52 proportion of potential food recycled into the system. Comparisons of discards fluxes with other fluxes in the system are presented later in this chapter. 3.1.4 b. Fish groups There is a large diversity of fish groups in the Gulf of Mexico. Among cartilaginous fish, 36 shark species (within 11 Families) and 15 rajiformes (within 6 Families) have been identified (Branstetter 1981; Bonfil 1997; Castillo-Geniz et al. 1998). Among teleosteans, 598 species (101 Families) are described for the coastal area in the GoM (Robins et al. 1986). Only in the south Gulf (approx. shelf area from Tamiahua lagoon to Cabo Catoche), 250 demersal species (102 genera and 55 families) were described by Yanez-Arancibia et al. (1985). Among those species, 17% (43 spp.) were considered dominant due to their abundance, frequency and distribution. One hundred twenty two species were recorded from an estuarine system in the same region (Laguna de Terminos). Among them, 6% were defined as permanent residents, 16 % use the lagoon as a feeding or nursery area, and 79 % were occasional visitors to the estuary (Yannez-Arancibia et al. 1985b). The incorporation of such species diversity was not the aim of this study. Only commercially important, abundant species or very unique groups (reef groups) were incorporated in the synthetic model. Specific criteria (added to those mentioned before) used to aggregate fish in the model were: 1) habitat type: coastal, shelf, reef or oceanic; 2) habit type: demersals or pelagics; and 3) size of general fish: small (< 30 cm TL a), medium (30-60 cm TL) or large (>60 cm TL) and, size of reef fish: small (< 20 cm) and large (> 20 cm). I aggregated fish for this study in 20 groups (the number before the name defines the group and one representative example is shown in parentheses b ): 15. Coastal small planktonic feeders (clupeids); 16. Coastal small demersal invertebrate feeders (catfish); 17. Coastal medium invertebrate and small fish feeders (grunts) juveniles; 18. Coastal medium invertebrate and small fish feeders (grunts) adults; 19. Coastal large demersal carnivores (groupers) juvenile; a TL is Total length measured from the tip of the jaw to the extreme end of caudal fin. b These representative names will be used throughout the rest of this document to designate species groups. 53 20. Coastal large demersal carnivores (groupers) adults; 21. Small coastal detritus and plant feeders (mullets); 22. Medium coastal fish feeders (coastal sharks); 23 Large predators (oceanic sharks); 24. Large coastal fish and invertebrate pelagic feeders (tuna) juvenile; 25. Large coastal fish and invertebrate pelagic feeders (tuna) adults; 26. Large reef plaktivores (soldierfishes); 27. Other fishes. 28. Large reef carnivorous (moray); 29. Large reef herbivores (parrotfishes); 30. Small reef carnivores (blennies); 31. Small reef herbivores (damselfish); 32. Small pelagic predators (bluefish); 33. Mesopelagics; 34. Bathypelagics. Since names of groups tend to be self-explanatory and the specific components of each functional group are already shown in Table 3.4, only I will describe some of the fish groups that require more detail. Oceanic fishes Available information of oceanic fishes of the Gulf is scarce. Lantern fishes seem to be the typically dominant deepwater fish fauna in the region (Devaney 1969). In general, oceanic fishes are divided in mesopelagics and bathypelagics. The former are fishes that spend daytime in depths between 200 and 1,000 m; and during the night feed on zooplankton in the epipelagic zone. Some examples are: myctophids (lantern-fishes), gonostomatids and stenoptychids. Bathypelagics are fishes that live below depths of 1,000 m. Some examples are anglerfish and Cyclotone. Parameters of these groups were modified from Pauly and Christensen's (1993) South China Sea generic oceanic model and are shown in Table 3.6. 54 Table 3.6. Parameters of oceanic fishes included in the model Oceanic fishes Biomass P/B Q/B (t ww km"2) (year1) (year1) Mesopelagics 2.6 0.6 2.9 Bathypelagics 0.02 0.1 0.4 source: Pauly and Christensen (1993). Sharks Sharks are a large group of animals that live in many oceanic subsystems and play a very important role as top predators. Habitat and ecological roles can be used to divide them in shallow water sharks and oceanic sharks. Juveniles of several species and adults of size smaller than 150-cm (TL) form the former group, called 'cazones' in Mexico. Among these coastal animals, adults are top predators moving between adjacent shelf areas; while early stages occupy lagoons nursery areas (Hueter et al. 1994; Castillo et al. 1998) (Table 3.7). Juvenile sharks are preyed upon by other fish species including other sharks. In the Mexican Gulf, these small sharks sustain important coastal fisheries (Bonfil 1997; Castillo-Geniz et al. 1998). In the model, these sharks are aggregated with medium coastal fish feeders. Sharks larger than 150 cm (TL), living in deep shelf and oceanic waters represent the highest proportion of predators in the area. Appendix 6 shows a list of shark species considered here. Studies in Atlantic lagoons show that juveniles of some shark species, such as bull sharks (Carcharhinus leucas), eat mainly small fish such as stingrays and marine catfish (Snelson et al. 1984), while oceanic shark species, such as the shortfin mako (Isurus oxyrinchus), have a diet consisting to 90% of oceanic teleost species (e.g. bluefishes, swordfishes, searobins). The rest of the diet of these large animals includes crustaceans, molluscs, other elasmobranchs, marine mammals and other animal remains. Consumption rate (Q/B) tends to be higher for some oceanic sharks species (8.01 to 11.0 year"1) than for coastal species (4.0 to 7.3 year"1) (Stillwell et al. 1982; Wetherbee et al. 1990). The production rate (P/B) for some coastal species was reported to change between 0.57 to 0.82 year"1 (Wetherbee et al. 1990). 55 Table 3.7. Some shark species with nursery areas in the Gulf of Mexico a ) . Species Nursery area Carcharhinus falciformis South Gulf lagoons C. leucas L. Terminos, Tampa Bay/Charlotte Harbor C. limbatus L. Yalahau, Tampa Bay/Charlotte Harbor C. signatus Off Veracruz area C. acronotus Tampa Bay/Charlotte Harbor Rhizoprionodon terraenovae Tamaulipas coast, Tampa Bay/Charlotte Harbor Sphyrna tiburo L. Terminos, Charlotte Harbor ' sources: Hueter et al. (1994); Castillo et al. (1998) 3.1.5. Standardisation of model parameters After the selection of functional groups, input and output parameters from models were standardised to selected units (wet weight tonnes per km 2 per year) assuming that the models represented possible balanced systems. Conversion factors for biomass are shown in Table. 3.8. Next, an average biomass of each functional group was estimated as a weighted mean over areas as will be shown in the following section. Table 3.8. Conversion factors used to standardise biomass of functional groups. Groups CI D W D W / W W Microalgae 0.40 0.18" Macrophytes 0.32 0.18a Shrimps 0.15 0.27 b ' c Zooplankton 0.32 0.16a Fishes 0.48 0.20a Other vertebrates 0.45 0.20a Polychaetes 0.30 0.20^ Molluscs 0.40 0.20M Benthos - 0.09b Other crustaceans 0.45 0.20bc Microcrustaceans 0.45 0.25b'c Detritus 0.10 0.25e C= Carbon, DW= Dry weight, WW= Wet weight; a. Jorgensen (1979); b. Lie (1968); c. Steel (1974). 3.1.5 a. Weighting factor for Biomass In this section, I describe how I obtained the mean (standardised) biomass to represent a functional group in the integrated model. First, I obtained a weighted mean biomass by functional group representing each subsystem. Secondly, I calculated a total subsystem biomass 56 by functional group and, finally I standardised the mean biomass by functional group for the entire GoM ecosystem. In the first step it was necessary to remember that each functional group is an aggregation of species. In this section the functional group will be defined as (i) and each component species will be defined as (j). The area of the subsystem represented by the model in which the functional group is located will be called (Ax) (Table 3.3 in section 3.1.3). So, for each functional group in a given subsystem the relative biomass of the functional group (i) in the subsystem (x) (B*ix) was calculated through a weighting factor by area as follows. B'jx = Bj*Ax Eq. 8 At = I Ax B*ix = (Z B' jxVAt Eq. 9 where B'jx is the absolute biomass for species (j) in subsystem represented by the model, Bj is relative biomass for species j taken from the models and Ax is the area represented by the model that reports that group. Finally, SB'jx is the absolute biomass of all species in (i) and, At is the total area of the subsystem. This calculation was done for each functional group by subecosystem. As an example, the weighted mean biomass of the functional group 'other decapods' formed by two species and represented in two soft bottom models was obtained as follows: 'Other decapods' (i) is formed by crabs (jl) and lobsters (j2). These groups are components of the model 1 and model 2 both representing soft bottoms subsystem. Model lfarea = 253 xlO 3 km 2) Model 2 (area = 187 x lO 3 km 2) Bj 1 = 1 Bj2 = not reported so, Bj 1 = 15.7 Bj2 = 8.0 assumed equal to model 2 so, total soft bottom area = 441 xlO 3 km 2 , and the mean biomass of other decapods in soft bottoms will be: B*ix = [(1+8 )•( 253 »10 3) + (15.7+8)' (187 xlO 3 )] / 441 -103 Biomass of other decapods in soft bottoms = 13.59 t/ km 57 Secondly, following the same logic as before (Eq. 8 and 9), this weighted mean biomass of each i was used to obtain a total subsystem biomass when multiplied by the total subsystem area. Finally, all these absolute biomasses of i by subsystem were added up and divided by the total GoM area (1.6 millions km2) to obtain a standardised mean biomass to be incorporated in the integrated model. These mean biomass values then, were available for all functional groups except for those not represented in the models: seabirds, piscivores and planktivorous marine mammals and sea turtles. Biomasses for these groups were obtained using reported numbers of animals in close or similar areas, along with average adult weights. A l l biomass values for non-fish and fish groups are shown in Tables 3.9a and 3.9b. 3.1.5 b. Weighting factor for P/B and Q/B Fluxes and rates were expressed on a annual basis. To obtain the mean P/B parameter per functional groups (i), first absolute Production and Consumption were calculated for each species (j) in that functional group per subsystem as follows: Pj =P /B jx*B ' jx Eq. 10 where Pj is the total production of species (j), P/B jx is the Production per Biomass ratio reported for j in the area x and, B'jx is the absolute biomass for species (j) computed above (Eq. 8). Then, a mean P/Bix per functional group was obtained using the weighted factor by absolute biomass of all species through: P/Bix = E P j / S B ' j x > Eq. 11 Finally, a standardised mean P/B for each functional group was obtained when all the absolute productions by subsystems were added up and divided by the total G o M biomass for any (i). Following the same logic, a standardised mean Q/B for each functional group was obtained through a total consumption Q/Bix. Thus using this weighting method, the parameters P/B and Q/B of almost all non-fish and fish groups were obtained. For non-represented groups, these parameters were obtained using literature available for similar areas. Tables 3.9a and 3.9b show the mean standardised parameters and mean values for non-fish and fish groups by subsystem. 58 CD -a o T3 CO 3 O & 43 W «C i B O B CD CD I = OH d Os rn CD ¥ 2 Z E tS E Z E CA B Z E „ oo ^ ^ 00 o oo oo m o oo o r--CN ^ o o o 12 o o o o OS "1 in 1 O m SD VO — S S o S 8 rs CNS 2 2 '2 m °° ~ ZZ o r— CN OS OS O O O O p-O — s 2 ~ S 9 s 2 S gg OS sb _ ^ - in C N - io „ »n V"i ~ o — O r< H os os o — x; o IL! *n CN — V"> O CN O rn o H^ O OS CN 2 8 as in so O — a O CN SO . p , £ CN 2 o - ^ 5 ^ ° C J M r - W M C O s f c N . ^ c n v o . ^ ; c o ^ ^ i o - ; c n M c N ^ ^ c s i o 6 c n j ^ o ^ c n d \ b ^ r i o 6 c * i SO cs — — o o o O O O O 00 Os CN Os -3-— CN S 2 °° ~ 4 0 N ^ so oo Os — i ~ OS o o o o CN o o ^ 8 — o — o M © o CO r-- in oo r--m o o m tN m so cu 43 _t — ZZ "~> I ; SO — I I in ^ Cl ^ o o o o BO c •c o D. 3 O bo J3 C C O O N ° ° in ^ as H 2 » 1^; c— r~ J . O N V" ) \S C N ¥ 2 Z E 5 E o o »r> N O » -ro CT\ ro I fN — rn "3- 00 ^ o 2 o C O o S o ro ^ oo r i 2 ^ § ^ 5 2 ^ H 2 O un m © DO O oo ro 2 oo' £ <=> © C O un — — o — o o ( N ^ O \ N ^ O O \ ^ UO r » u n r o u n u n " 3 ' r o o \ un O o vo I un ~- OO CM ro (N ' 'o i r— —- — ~ © © m i — i o o —* d o — —i *0 (N CO <N —• © O O O —' O un ui un o O O T J - m O N un a\ o o d — ~ ^ ^ O ^j- r-- m — d d t/^  _ _ ' f N O O f N O O O O © 2 \D un © — -^1 m — <N — un ~ d d d © © — —< • O " ro d d — — t ro J ° oo 0 0 <=> " <*! £ N 2 r~ X. f i 2 2 — — <N O N co «n oo 3 2 p p . S3 S - j . M rs 60 i2 < 5 < 2 cj o "3 I I I < on O 2 in *o r-X „ .2 a> « C/5 Cg ™ " • S .2 Q -2 c c oj S C cu 3 <u (2 5 Q 5 S ^ O - (N m fN ro ro rO ro CO I 3.1.6 Diet composition Diet composition data for non-fish and almost all fish groups were obtained through reported volume proportions of all items consumed by those groups in all submodels. For the rest of the groups, diet compositions representing functional groups were obtained through a mean weight of components of most abundant species using the consumption of each species (Q; = B; • Q/Bj). For groups not previously represented, original references of diet composition were used. Tables 3.10a and 3.1 Ob show diet matrix for non-fish and fish groups incorporated in the model. Specific references used in the models for subsystems for all parameters by functional group can be seen in Appendix 2. 3.1.7 Catches per unit area Commercial fisheries Although catches were estimated for some of the models used in this study, I considered that global catches reported by governmental agencies better represent exports from the entire system. Based on the information provided in section 2.1.8, I considered three main fisheries relevant to this model: inshore commercial, offshore commercial, and sport fishery. I assumed that all retained bycatch was included in landings records. Published data of commercial fishing landings from 1985 to 1997 were taken from fisheries management agencies in Mexico and U S A (SEMARNAP 1997; NMFS 1997). A n average landing per year by species was obtained for those years. Since total annual landings by species and total annual accumulated landings are different by approx. 170,000 tonnes; the first was corrected using the relative proportion by species. Then, to obtain the relative catch by area per functional group, the corrected landing value was divided by the total Gulf area (1,623,000 km z). Table 3.11 shows the average landings of functional groups for the period 1985-97 in the Gulf. The proportion of inshore and offshore catches was based on the knowledge provided by the subsystem models and by additional literature. When these estimates of landings by species were different from other values reported by fisheries scientists, I corrected the landing values. 61 o SO o d o O o o o >n o CN tN o o © © © d d o o o o o ro in CN 00 o o d d d d ro OS o Os Os o . I ro r- o ro ro ro CN *— o o O o ro CN O O o o d d d d d d d d d d o o o o o o o o oo in OO o CN CN ro CN o o in O O O O d © d d d d d d o o o © o o o r~ r- ro ro o CN o o O o O © © d d in CN ro m o ro o O SO CN ro —< o o o o O O d d d d o d o o d o o o CN — d d o o m o CN </"> d d o o CN t-CN CN d d o o CN © o o d o © d o d o d — oo Os d o o OS O o o uo ~- oo Os ro o d so o o i n i n r o © O s © © © — - i n o o o o o o r o » n — T f r o O O — o o o o o o o o o ro in in o o oo in o in o o o © d © © d d © d _ _ o © © © — o © © © © © © © o © d d p ' c 2 J3 .c/> o 43 CJ 9 <£ e 3 i 2 3 ,o c -S c2 -« C c ft §-£ 1 s f. ° § § £ 2 S 2 a. o o N -a ' o p. •e -e E E CD CJ CS C3 X X U CJ 3 < 3 S ^ Q, u m O CL I S O O U CJ p. a § = — E LA CJ a. Sb g C3 2 60 •S 1 / 1 CJ CO c 3 _cj 3 _ <J 3 is o d §1 Os — © © ro © — © © © r-ro O — © d © CN SO © o o CN CN SO © q tS .!£ CO >s CO CJ J3 co O C ^ - c N r O T f i n s o r ^ o o o s © ' - H c N r o ^ r i n s o c ±; (3*- <u»; 5 c - .S i- SP n c cots >T3 £ - o 3 O cj £ T J ^ 3= ,5 ca — C343 C ^ c » c > o ^ f S f ^ ^ i n ^ l ^ c o i ^ O ' - - c N f n , t i r i — — i - r i N N N N r i s N N r N i f N ^ r ^ ^ f O r o r n o o in o o CN d d o o o r> SO CN d d Os o © m in o os Tt o Os o o o o o o © d o © d o o d o © o d o o <N O o — © © IS — — OO o o o o o o CN o o © VD CN CN d 15 o o o o in — oo O CN O O © o © © in oo O oo o o o d o oo o t m M os — — (NO o o o o o o o o o o d o CN o © Os © Is o  — m o o © CN Tt © © o o o — o o o CN CN — ro o o o o o d d © so o o o o r-o ro in m so — o o © © © p © o o © © ©' m c N T t T t © r o i n © i n © i n o o r o o s — o— in i n T t s o i n O s © o o r o o o c N © i n — in © © ro o — Io o o o o c s o o o o o o o o o p o o o p o o ' o ' o ' o ' o o o o o ' o o ' o o ' o o o o o o ro CN CN r-O Tt Tt o o o o © — in o o o o o © 00 o oo o o ( N ro o o Tt OS o oo oo CN o o ro © Tt o o o o © o © d o o o o © in o o o o ( N o <N o oo o o oo o in o o o o in o o CN o o Tt o in o o o o o ro CN o o © ©• © © o o o d © o ©• oo o o CN O ( N o CO o ro o o ( N o ro SO o CN CN o o o o © O in in ro o o CN in o CN so o o in o in o in o © o o o © o o © d ©' © o o d © © ©' o ©' ©' o in o o o o r-o o Tt o CN o o r-o o Os o o o Tt o o o CN o o o o © o in o © CN o d © © d o © ©• d o d o © o SO CN © — ro © — © o — o o o p © © d d d © © © ' — — ro I — O I so © o 2 ISO ro Tf o o in O ro © in in — o o CN CN o ©' © © © CN Tt ©' Tt o o so oo in CN in ro CN — O ro CN — ©' O © © © © SO in o © so ro Tt CN so — — Os © O in o d d © d so CN CN SO Tt o o so oo in r— ro r~ © — CN © d o © © in o oo o CN — O CN O d o ' © ' Os in Tt © o © ©' d oo — — © © o m o CN © so in ro in © — — o © d o © (N O © OO © o o Tf — SO OO O © CN O d d © d © CN O d © Tf in © — o d d o o o o © © o o o © t-~ © in ro ro CN — — — r--— T t c N p p p p p o p d d d d d d d d d d |o © o o o s — © o o © in m r ^ m — cNinin — in o — — o o c N p p p p dddcidddd>d>ci o oo i n s o r - o o o s © — rNroTtinso r - - o o o s © — c N r o T t i n s o r ^ o o a s © — cNroTtinso — — — CNCNCNCNCNCNCNCNCNCNrororororororo r~ oo os © I ro ro ro Tt I Table 3.11. Annual landing values of commercial fisheries in the G o M a ) . Group -species Total Landings (t- 103-year"') Landings by area (t • km 2 , year"') Inshore Offshore 6 Macroepifauna 92.9 0.057 -11 Juvenile shrimps 57.6 0.036 -12 Adult shrimps 86.5 - 0.053 13 Other decapods 50.6 0.019 0.012 14 Octopus 15.3 0.009 -15 Clupeids 744.1 0.183 0.275 16 Catfishes 39.6 0.015 0.010 18 Grunts 2.3 0.001 0.001 20 Groupers 30.3 0.011 0.007 21 Mullets 19.5 0.012 -22 Coastal sharks 7.9 0.003 0.002 23 Oceanic sharks 8.2 0.005 -25 Tunas 21.9 0.008 0.005 27 Other fishes 8.9 0.003 0.003 32 Bluefishes 4.4 0.001 0.001 Total 1,190 0.360 0.380 source: SEMARNAP and NMFS 1987 data records. Inshore area < 20 m depth, offshore area 20-200 m depth Sport fisheries Only US records of sport fishery landings from 1981 to 1997 were used in the study (NMFS 1998). Landing weights by species were aggregated by functional group and then divided by 2, half the Gulf area (about 811 x 103 km 2) because it represented only the US sport fisheries. However, I assumed that these recreational landings could possibly represent the entire system. For sharks, I corrected the landing value using the estimation of Casey et al. (1985) since the reported value of catches by sport fishers was very small when compared to independent data. These landing values by area are shown in Table 3.12. 64 Table 3.12. Annual landing values of US recreational fisheries, and estimated landings by area used to represent the entire system Group-species Landings Landings by area tonnes (t • km2- year"1) 15 Clupeids 9 0.00001 16 Catfishes 849 0.001 18 Grunts 867 0.001 20 Groupers 3198 0.004 21 Mullets 7 0.00001 23 Oceanic sharks 1973 0.0024 25 Adult tunas 2665 0.0033 27 Other fishes 532 0.0013 32 Bluefishes 1077 0.014 Total 11,177 0.014 a) source: NMFS (1998) 3.1.8. Landings by fishing gear Inshore and offshore landings were subsequently divided by fishing gear to assess the possible effect of particular gear in the ecosystem. For this purpose, and since sport fisheries account for a very small amount of the total Gulf landings (« 2 %), all landings from commercial and recreational fisheries were aggregated by gear. Eight fishing gear were considered: inshore and offshore trawlers, inshore and offshore longline, inshore and offshore gillnet, purse seiners and miscellaneous gear. The exact amount caught by each gear was not available. However, the landings by area were assigned to each gear depending on the habits (pelagics, demersals, etc) of the resources and using local and international information of what resources are caught by each gear (INP 1994; Sainsbury 1971; v. Brandt 1972; Nedelec et al. 1990) Table 3.13 shows the commercial and sport landings aggregated by fishing gear. 65 Table 3.13 Total commercial and sport landings (t- km 2- year"1) in the GoM area by fishing gear. Group- species Fishing gears Itw Ilg Igt Otw Olg Ogt PS Misc. Total 6 Macroepifauna - - - - 0.057 0.057 11-12 Shrimps 0.036 - 0.053 - - 0.089 13 Other decapods - - - - 0.031 0.031 14 Octopus - - - - 0.009 0.009 15 Clupeids - - - - 0.458 0.458 16 Catfish - 0.015 - 0.0001 0.01 0.025 17-18 Grunts 0.001 - - 0.0011 - 0.0021 19-20 Groupers 0.006 0.005 - 0.006 0.003 0.02 21 Mullets - 0.006 - - 0.006 0.012 22 Coastal sharks 0.002 0.001 - - - 0.003 23 Oceanic sharks - - 0.008 - 0.008 24-25 Tunas 0.007 - - 0.009 - 0.016 27 Other fishes 0.001 0.001 0.001 0.001 0.001 0.001 0.006 32 Bluefish 0.001 - - 0.0015 - 0.003 Total 0.036 0.018 0.028 0.054 0.026 0.014 0.458 0.104 0.74 G-S = group species, Itw= inshore trawl, Ilg= inshore Ion g line, Igt= inshore gillnet, Otw= offshore trawl, 01g= offshore long line, Ogt= offshore gill net, PS= purse seiners, Misc.= miscellaneous gears. 3.1.9 Discards In section 2.1.10,1 mentioned that discards from shrimp trawls operating in inshore and offshore waters consist of a huge volume of flesh returned to the sea. Also I mentioned that this flesh is a potential food source for several consumers in the system. To incorporate this flux into the integrated model it was necessary to calculate the total amount discarded in the area and the discards by species. The first step in this process was to define the amount of bycatch and discards in the area. Shrimp trawls incidentally catch diverse finfish species, varieties of invertebrates and some high vertebrates groups. However, there are only good estimates of finfish bycatch in offshore waters of the northern GoM (Nichols et al. 1987, 1990; Burrage at al. 1997). Therefore, these were the main sources of information about bycatch and further calculations were based on those 66 estimates. An average value was obtained and modified to represent only discards (3% extracted) for that region (Alverson et al. 1994). To obtain an estimate of Mexican discards, I calculated a possible bycatch using proportions of shrimp landed between Mexico and the US. The annual Mexican shrimp catch is about one fifth of the US catch (22,237 tonnes vs 101,941 tonnes); this proportion was maintained to assume that Mexican bycatch should be approximately one fifth of the US shrimp bycatch. Then, once again, amounts discarded were obtained assuming that 97% of bycatch is discarded. The second step was to obtain relative proportions of discard by functional group. Initially, the relative percent of incidental catch of finfish and other invertebrates in relation with the shrimp caught was estimated using values in Mexican and US literature (Hagenkotter 1993; University of Georgia, unpublished; Upton et al. 1992; Branstetter 1997). I then obtained specific total finfish (bony and cartilaginous fishes) and invertebrate discards (non-commercial shrimps, crabs, echinoderms and molluscs) using detailed reports from the area (Bryan et al. 1982; Nichols 1987, 1990; Palomino 1996). Proportions of these functional groups in discards are shown in Table 3.14. Finally, I combined these proportions with total discarded values obtained before to obtain discarded amounts by functional groups in the area. For higher vertebrates, I used independent reports of turtles and dolphins killed in shrimp trawls and finfish trawls (Upton et al. 1992; Quiroga et al. LNP, unpublished). I assumed that all higher vertebrates are discarded. Then, I calculated the average amount discarded using average weight of sea turtles and piscivorus marine mammals present in the area. 67 Table 3.14. Absolute and relative percentage of discarded groups in offshore shrimp fisheries. Other fish division includes catfish, snappers, lizardfish and sharksa). Groups Absolute % Relative % Absolute % Fishes 47 Croakers 51 24 Spots 11 5 Seatrouts 9 4 Drums 8 4 Porgies 3 1 Mackerels 5 2 Searobins 11 5 Other fishes 2 1 Invertebrates 35 Crustaceans 63 22 Molluscs & echinodems 37 13 Higher vertebrates 18 Dolphins 2.5 0.5 Sea turtles 97.5 18 Total - - - 100 sources: Hagenkotter (1993); University of Georgia (unpublished); Upton et al. (1992); Branstetter (1997); Bryan et al. (1982); Nichols (1987, 1990); Palomino (1996); Upton et al. (1992); Quiroga et al. INP (unpublished). For the inshore shrimp fisheries, I estimated bycatch using information from an artisanal shrimp trawl fishery in the Gulf of Paria, Trinidad (Maharaj et al. 1991). These authors reported that 42% of total shrimp landings is discarded dead, and that from this volume, finfish account for 61%> and crabs for 39%. To differentiate among fishfish species in incidentally caught in inshore waters, I used the same proportions obtained in the offshore fisheries. No higher vertebrates discarded were reported in this fishery. Finally, total shrimp discards by functional group relative to area were estimated by dividing the discards by the total Gulf area by strata (Table 3.15). Not all discards are dead when they are returned to the sea (Perret et al. 1996), but in general, survival should be negligible for most species (Wassenberg et al. 1990; Goodyear 1992). Some factors affecting the survival of discards are the number of biological, environmental and fishing operational patterns (Alverson et al. 1994). Here, I assumed that all discards were dead. 68 The specific estimated values for discards by area were incorporated in the model using the discards input interface of Ecopath 4, so the program will make them available for potential consumers in the system. Table 3.15. Annual estimated discards from trawlers operating in the Gulf of Mexico by strata Group-species Discards by area (t • km2- year"') Inshore Offshore Total discards (t- 103-year"') 6 Macroepifauna - 0.076 123 13 Other decapods 0.01 0.123 215 16 Catfishes 0.21 0.317 850 17 Juvenile grunts 0.01 0.012 32 24 Juvenile tunas - 0.014 23 27 Other fishes 0.03 0.044 127 36 Dolphins - 0.003 4.0 38 Sea turtles - 0.098 159 Total 0.26 0.69 1,534 a) sources: Nichols et al. (1987, 1990); Burrage at al. (1997) 3.1.10. Balance of the synthetic model Once the set of parameters were available, the model was parameterised and balanced. To balance the model the assumption of thermodynamic equilibrium was evaluated for each functional group in the system. There are several physiological criteria to assess this equilibrium: ecotrophic efficiency (EE) of each functional group must be between 0 and 1; gross efficiency (production / consumption rate) should be between 0.1 and 0.3 (except for very small and fast-growing organisms), a positive flux of matter to detritus should exist, and respiration / assimilation rate should be less than but close to 1 in top predators. I constructed the initial GoM model using the mean values provided in tables 3.9 to 3.15. Also, for all groups consuming detritus, the unassimilated /consumption rate was set to 0.4 instead of the 0.2 assigned to non-detritivores groups. This was done assuming that the detritivores assimilate less food than non-detritivores because they consume the bacteria aggregates that use the detritus as a substrate, more than the detritus itself. This live material in the detritus accounts 69 for 14 to 79 % of the total weight of detritus (in areas above 100 m) and to less than 3% in deep waters (Gordon 1970). However, after one parametrization attempt, the system proved to be unbalanced in some group parameters. The groups that were highly unbalanced (with EE values from 18 to 70) were sea turtles, carnivores zooplankton inshore and juvenile coastal demersals. Other unbalanced groups (with EE values lower than 9) were mainly coastal and juvenile groups. The reason for the unbalance in the majority of the cases was due to high predation mortality; only for sea turtles and dolphins was the cause mainly related to catches (specifically discards). Thus, I modified some parameters for those functional groups in descending order starting with the group of highest EE. First, slight modifications were made to the least known parameters, such as diet composition. In the case of the sea turtles, it was also necessary to dramatically decrease the predation on them. Then, when necessary, B, Q/B and P/B parameter values were modified within the range of variation provided by the underlying models. In very few cases, when these changes still did not allow the model to balance, alternative parameters were used based on published literature. For some groups where the input parameters did not allow the balance (groups 1, 2, 4, 8 and 38), biomasses were left to be calculated by the software while fixing EE to values around 0.9. 2 1 Finally, for balance in the detritus groups it was necessary to import 1000 tonnes/km /year" of detritus matter into the system (1.5 times the primary production reported by E l Sayed et al. 1972). Once all groups were balanced, the Ecoranger routine was used to introduce uncertainty to the model constructed. 3.1.11. Uncertainty and Ecoranger The models representing subsystems in the Gulf were created using single sets of mean annual input parameters. However, since the synthetic model is from several sources, it was necessary to take into account the uncertainties incorporated in the data. This was accomplished through the use of the Ecoranger routine in Ecopath with Ecosim (Christensen and Pauly 1995). Ecoranger allows the modeller to include a mean or mode value for each input parameter (B, P/B, Q/B, EE Ex, and Diet composition) and a range of variation and distribution. The distribution chosen could be normal, triangular or regular depending on previous knowledge. The routine runs in a Monte-Carlo fashion, evaluating model alternatives with combinations of input variables 70 selected randomly from the specified distributions. Only the models meeting the mass-balance constraint are accepted, and the modeller can set the selection criteria to choose the best model. In this analysis, I chose to work with the pedigree information for each parameter of each functional group using the main source of information or the source of the parameter with value closest to the mean average obtained by the weighting method. The pedigree of an Ecopath input (parameter) is a coded statement categorizing the origin of a given input and specifying the likely uncertainty associated with that input (Help, Ecopath ver.4). The input is considered with the best 'quality' when it is estimated from local data from the region modelled. For example, for the biomass parameter, high confidence value is assigned when the biomass was estimated by sampling with high precision, while low confidence value is assigned when the biomass was estimated by the Ecopath software using the Equation 3 (Section 1.6.1). Pedigree of the biomass parameter Uncertainty (+/- %) Missing parameter (estimated by Ecopath) n.a. From other model 80 Guesstimate 80 Approximate or indirect method 50-80 Sample based, low precision 40 Sample based, high precision 10 I used the minimum residuals criteria with random seed process and a minimum of 10,000 runs to evaluate the best resultant model. The best model obtained was unbalanced in a few groups with EE values not larger than 3. I proceeded to balance this 'best unbalanced model' with few additional changes in biomass and diet composition. Finally I ran the balanced model and obtained some important summary statistics. I compared these results with the statistics of my original base model (Table 3.16) and selected and saved the best model for further analysis. 71 Table 3.16. Summary statistics for the original mass-balance model and the best model obtained using the Ecoranger pedigree routine. Statistics Base model Model from Ecoranger Total throughput 11,167 a 11,304 a Throughput cycled (excluding detritus) 127 a 58 a Finn's cycling index 14.8 10.6 Finn's mean path length 3.6 3.2 Net system production 379 a 1198 a Total biomass 304 b 239 b Total catches 1.6 a 1.6a Prop, flow originated from detritus 0.62 0.57 t/W/year bt/km2 3.1.12. Adjusting mortality by fishing effort Since one of my objectives was to compare the effect of incrementing fishing effort for some resources, I compared the fishing mortality rates estimated by the software after parameterisation with reported values from the literature (Brown et al. 1991; Castro et al. 1997). As I mentioned in section 2.11, several resources in the area have been reported as overexploited, However, given the landings, discards and biomass estimated for most of the groups in the area, the resultant fishing mortality rates were very small compared to the estimated mortality by predation. It is assumed that an overexploited resource shows fishing mortality rates (F) equal to double of the predation mortality rate (F > 2 (M2)) (Table 2 in Pauly et al. 1986b). In my model for the exploited resources (groups 11 to 25) the fishing mortality rate was smaller than half of the predation mortality rate. Evidently these results are a consequence of the high biomass of some resources, i f we assume that F= catch/biomass and that the reported landings are trustworthy. Thus, I adjusted these estimates by decreasing the biomass input in the model until F was closer to the reported values and EE was not higher than 1. Note that some resources have a very patchy habitat distribution into the potential habitat on the shelf (e.g. shrimps and groupers); therefore, their estimated mean densities and their fishing mortalities are unrealistically low. Table 3.17 shows the initial and adjusted biomass, fishing and predation mortality rates for the integrated model. 72 Table 3.17. Initial and adjusted biomass, fishing and predation mortality rates of highly exploited resources in the Gulf. Groups- species Initial Adjusted B F M2 B F M2 11 Juvenile shrimp 1.1 0.030 2.27 0.30 0.12 7.39 12 Adult shrimp 3.4 0.020 2.07 0.20 0.24 6.32 20 Groupers 0.2 0.050 0.22 0.10 0.08 0.28 21 Mullets 2.0 0.006 0.20 0.90 0.01 0.63 22 Coastal sharks 0.1 0.030 0.33 0.04 0.07 0.39 23 Oceanic sharks 0.1 0.130 0.16 0.04 0.20 0.19 B= biomass (t/km2) F= fishing mortality rate (year"1) M2= Predation mortality rate (year"') 3.1.13. Independent analysis of discard fluxes Section 3.1.8 describes how total discards of shrimp fisheries were estimated and incorporated into the synthetic model. However, with the information obtained it is also possible to obtain some insights on how much those discard fluxes represent in terms of other fluxes in the ecosystem. Thus, I compared the discards flux obtained with 1. Total commercial landings; 2. Total production of discarded groups (also: production of particular groups); 3. Total consumption of potential feeders; 4. Primary production. The first comparison was straightforward using the total fluxes already calculated. The second and third comparisons were done using the absolute production and consumption values calculated with equation (3) shown in section 3.15b. For the third comparison, potential consumers of discards were considered to be all carnivorous marine biota reported to eat discards. Some examples of these consumers are mobile invertebrates (crabs), carnivorous fishes (sharks), sea birds and dolphins (Wassenberg et al. 1990; Hi l l et al. 1990), and those groups that are assumed to be discard eaters due to their food habits and size, such as porgies and groupers. At the end, I considered 12 functional groups as potential discard eaters: juvenile and adult shrimps (11-12), other decapods (13), juvenile and adult coastal medium invertebrates and small 73 fish feeders (17-18), juvenile and adult coastal large demersal carnivores (19-20), medium coastal fish feeders (22), large predators (23), other fishes (27), sea birds (35) and piscivores marine mammals (36). Finally, I calculated the Primary Production Required (PPR) to sustain annual discards using the approach of Pauly and Christensen (1995), and compared it with PPR to sustain total commercial landings. 3.2. Results and discussion The synthetic Ecopath model constructed for the Gulf of Mexico provided us with basic information about structure, interactions among its components and energetic fluxes within the system. This information will allow us to have insights into the response and resilience of the system to externally imposed perturbations (mainly related to fishing impact) and its maturity. Also it allow us to compare the system with other ecosystems. In the following sections a range of statistics will be analysed for the Gulf of Mexico ecosystem using the model that I considered the best in maintaining a net system production (budget) closest to reported values (the original base model). 3.2.1. Structure of the Gulf of Mexico ecosystem Table 3.18 presents the final input parameters and some important results from the balanced model. Figure 9 graphically shows the position of the components of the synthetic model of the Gulf of Mexico ecosystem. 74 CD C3 o >V T 3 CD * J "3 o "3 o CD CS 4 3 o o CD T3 O a CD 43 fl >v c/l T3 CD O C ca "3 42 CD 6 o C+H CD 5 CD i ^ o a M c T J 3 fl 00 3 ° PHCS o ^ T3 » fl w CO * J C 3 3 .S cd ctj o •9 S 1—H ro CD O 1 « CM 0 0 tN > 0 0 vo CN 00 Uo CN tN CO tN "o CO d CO o\ CO uo 00 0-1 VO CO tN 00 Uo uo —. CO 00 CN rv. Ov Ov O N 00 Ov 00 t \ ON 00 M - ON tN IN "O Ov OO o\ V O Ov vo IN tN Ov Ov "o °o M - 00 •<»• uo uo Ci Ci Ci cd ci ci ci ci 0 0 ci 0 C> <=> Ci Ci Ci CO C O ci ci ci ci CO C> C O 0 ci ci ci CO CO <oi CO ci ci Ov Ov <N Ov CO CO CO uo <N Ov CO 0 uo uo vo tN I N u-i vo Uo CO Ov Ov vo <N Ov CO VO - H uo CN Uo CN CN CO CN ro tN CO <N CO CN u-i CN Uo vo CO vo vo I N vo IN tN vo uo 06 CO uo CN CO co CN K vd °6 CO ci uS vd vci vd vd 06 06 vd <o 06 CN OQ CN CN UO CN UO CO uo CO CN vo tN 00 co uo CN co uo uo CO CO uo ?> uo vo r-1 Ov vo co VO Ov co co vo <N Ov tN vo IN Ov CN '—, vo uo vo tN tN VO °o UO vo 00 vo co d C i uo co M-CO co CN vo co vo tN co vo co CO UO co CO <o> <N CO CO 0 C i ci co Ci cd CO CO cd cd cd cd cd cd cd cd cd cd cd cd cd cd CO cd cd cd cd cd cd cd cd cd cd cd cd ci ci CO C i CO co M- Ov CN P^  Ov >» ^ ^o c . C i c> co cd cd c> c i c i Uo Uo vo C l co ci cd ci ci uo <N uo "0 uo Cv. Ov 0 Ov vo <N Ov t \ CO Ov 00 Ov uo CO t \ CN vo CN Uo UO CO °o Ov CN co »^ Ov ,* co vo l \ Ov r\ CO Ci CO C i CN 00 Ov CS <N Co uo c C i CO CO c > > ci Ci ci ci ci ci ci c i c i ci ci Cj ci ci ci cd ci Ci ci ci ci ci ci cd Ci CO CN vo — O O O N CO CN 00 uo CN Ov Ov CO Ov C i 00 CN vo co CO CN <N c vo uo co Ov Ov CO CN Ov c. 00 co CO Ov CN <N CN Ov 00 vo c <N Ov C i Ov M-C i Ov uo 0 00 00 c. CO VO >>• CO CO ?° vo C i UO uo C i R d CS <N Ov CO ?° vo c CN C i C i UO 0 C CO C i ci C> ci ci Ci ci ci Ci ci ci ci ci ci cd ci ci ci ci ci ci ci ci ci Ci ci ci ci cd cd C) cd ci ci ci ci UO uo 00 r~ 00 0 0 0 0 Ov uo 0 CN 00 VO uo CN Ov p vo 00 vo OV uo 00 00 00 0 0 0 0 uo 00 CN 00 0 0 O O 0 ro 0 00 0 00 uo -* VO ro O Ov CO 0 0 r~ 0 uo Ov 0 uo CN UO CN 00 vd 00 r- d CN CO Ov d co 06 CN 0 d vd 00 06 uo d ro CN CN CN CO Ov CN d d 00 0 CO O O Ov 0 0 CN 0 p O p 0 r- O 0 p CN 0 CN O •<* ro VO vo uo CO •* CN uo uo Ov CO CN uo uo 0 Ov ro uo Ov uo uo 0 vq 0 as r~ vo VO uo 0 0 uo 0 uo d VO CN CN CN CN uo u-i r-i r~ u-i d d 0 d d d d d d d d d u-i 0 d d CN CN VO CN VO uo 0 0 uo uo uo 0 CO CN CN 00 0 VO CN CN uo 0 CN O 0 CN CN 0 Ov uo 0 uo 0 CN 00 O VO CO wo O CN 00 ro vq 0 0 0 uo 0 CN 0 0 > <N vd CO 0 vd vd d vd d d 00 CN u-i CN CN d d d d d d d d d d d d 0 d d d d 0 d o o o o v£i h OO ON O m ro m m In Figure 9, each box represents a functional group situated vertically from the primary producers and detritus at the bottom, to the top predators at the top. Horizontally, the diagram shows diverse levels of depth moving from inshore waters at the left to the oceanic region at the far right. The trophic flows are not displayed in this scheme because there are too many connections among the groups to be shown and understood in this format. However, standing stock biomass (B) and production (P) are represented. The size of the boxes show that the detritus biomass is very large compared with the rest of the groups. It can be noticed that the Gulf of Mexico ecosystem spans over more than four trophic levels. Top predators, with trophic level of approximately 4, are represented by 9 groups: large demersal carnivores (groupers in shelf areas), large coastal predators (sharks), marine mammals and, juveniles and adults of large pelagic fishes (tunas and mackerels). Secondary and tertiary consumers, with trophic levels from 2.5 to 3.5, are very abundant (18 groups), while first grade consumers with trophic level smaller than 2.5, are represented by 8 functional groups in the system. Although the system seems to show very strong competition for resources due to high numbers of groups located at similar trophic levels, it is necessary to remember that the model was built to integrate functional groups that represent all trophic levels in different habitats. For this reason, groups at the same trophic level do not necessarily compete since they occur in different habitats. 3.2.2. Summary statistics The summary parameters generated by Ecopath can be used in model comparisons. Until now, the most extensive source of summary parameters for comparison purposes of diverse ecosystems in the world is the document by Christensen and Pauly (1993b). For this reason, on some occasions, I will compare the results of my model with the results of this document. Several of these parameters originated from a study by Odum (1969) which defined ecosystem attributes that can be used to assess the development and maturity of a given ecosystem. The last four statistics define some attributes of the fisheries in the area. Table 3.19 shows a list of the summary statistics that define the mass-balance model from the Gulf of Mexico ecosystem. 76 S 0) u 01 o o 0 1" o r-0- tl £ to a. | D. o o CD o <+H o o T3 *u o s CD o i s • — ii xi ii xi •s o & o o S3 • o Jjb o ON r-Several of these statistics related to maturity will be discussed in further sections; however, some general aspects are discussed here. The analysis of major flows in the system (table 3.19) show that consumption (predation) is the highest flow (41%) followed by flows to the detritus (22%). The high consumption should be interpreted with caution as predation is the product of consumption rate multiplied by the biomass of the consumers, and the biomass in some of the models used in the integration were calculated assuming fixed values of ecotrophic efficiencies (EE). I considered that among the input parameters the biomass estimates are the least trustworthy since the biomass, when estimated, is calculated based on very localised surveys. The high flow to detritus is very possible since 62 % of the flows originate from detritus. The importance of detritus in the system is mentioned further on several occasions. Table 3.19. Summary statistics computed by Ecopath for the Gulf of Mexico model in 1980s-1990s. Dashes represent no dimension in those parameters. Parameter Value Units Sum of all consumption 4547 t km "2 • year Sum of all exports 1380 t km ~2 • year Sum of all respiratory flows 1786 t km "2 • year Sum of all flows to detritus 2455 t km 2 • year Sum of imports (detritus) 1000 t km "2 • year Total system throughput 11,167 t km "2 • year Total biomass (without detritus) 304 t-km"2 Net system production 379 t km "2 • year Total Pp /total respiration * 1.2 Year Total Pp/total biomass * 7 Year Connectance index 0.23 -Overhead 75.3 % Total biomass/total throughput * 0.027 -System Omnivory index * 0.23 -Fraction of total flow originating from detritus * 0.62 -Finn's cycling index * 14.8 -Finn's mean path length * 3.6 -Total catches 1.6 t km "2 • year Mean trophic level of the catch 2.9 -PPR (% of total Pp) 4.6 % Gross efficiency (catch/net p.p) 0.0007 -Pp = Primary production; PPR = Primary production required to sustain fisheries defined by Pauly and Christensen (1995) * = Maturity attributes defined by Odum (1969) 78 Comparing the mean trophic level of the catch (TL 2.9) to the trophic level of the majority of the components in the system (Table 3.18), it can be noticed that the fishery is not competing directly for prey with top predators in this system. As mentioned before, the most important fisheries are concentrated on small pelagic fishes (TL 2.6), macroepifauna (TL 2.2), shrimps (TL 2.6-2.9) and several groups of fishes with TL smaller than 3.9. Only small amounts of large pelagics are taken from the system. The gross efficiency of the fishery is very low (GE= 0.0007) which indicates either the fishery is concentrated on apex predators (which is not the case in the GoM system) or the system as a whole is underexploited. Since most of the resources sustaining fishing are in shelf areas (small areas compared with the oceanic region), it is very possible that the synthetic model is not sensitive to the impact of the fisheries in more coastal regions. Other possible comparisons among fluxes in the Gulf of Mexico ecosystem are related to the consumption and production of predators and prey in the system, and the relationship between the total production of these groups and the landings reported in the area every year. I mentioned one study regarding this issue (Brown et al. 1991) in Section 2.2. Even when the models are not comparable due to the extension of the areas modelled, I consider it interesting to assess the relation between some fluxes in the way Brown et al. (1991) mentioned. In my model, I considered it more accurate to compare the total absolute consumption of all predator groups (usually groups with T L higher than 2.9) to the absolute production of prey groups (groups with TL smaller than 2.9). The result was that the total consumption of predators only accounts for approximately 17% of the prey production. Clearly, there is a large contrast in calculations obtained with both models. Since my model represents an area of more than twice the area in Brown et al., it is possible that the large area dilutes not only the total effect of catches in the system (as mentioned before) but also the effect of predation. Moreover, the majority of the groups included in the synthetic model correspond to shelf habitats and the reported landings and discards come from the same smaller region. Thus, I consider that my estimations could be valid in terms of general Gulf fluxes, but that for the management of fisheries operating only in shelf regions, Brown's et al. values could be more accurate. 79 3.2.3. Maturity analysis Odum (1969) described a system for measuring the maturity of an ecosystem. Ulanowicz (1986), who integrated concepts from'thermodynamics and information theory, further developed this idea and gave new interpretation of ecosystem growth and development. Ulanowicz proposed that when ecosystems mature, their growth (thermodynamic component) and development (information component) increases. Increase in size is shown as the sum of all flows in the system (total system throughput of all components) while development is expressed by an increase in the organisation and information content of the flows which is quantified as the average mutual information. Some of these maturity concepts are incorporated as Ecopath routines and have been used to compare maturity and stability (resilience) levels of 41 aquatic ecosystems from all over the world (Christensen and Pauly 1993b). Although none of these 41 models represent Large Marine Ecosystems, I used four criteria of maturity resulting from the results of these routines (Table 3.19) to assess the maturity and possible resilience of the Gulf of Mexico ecosystem compared to other aquatic systems. Odum (1969) stated that as ecosystem mature they become more dependent on detrital flows than on flows from primary producers. With this criterion, the Gulf of Mexico is mature, since it shows higher dependency on detritus (62%) than on direct grazing of primary producers. As mentioned in section 3.1.10 (balance or parameterisation of the synthetic model), a supplemental detritus flow was shown to be very important in balancing the model. This requirement for the model is highly consistent with the amount of detritus (and nutrients) that the Mississippi River and variety of Mexican rivers introduce annually into the system (section 2.1.7). Odum (1971) described another index of maturity as the ratio between total primary production and total system respiration (Pp/R). This ratio would approach 1.0 as systems mature. Pp would grossly exceed total respiration for upwelling systems, while respiration would exceed production for systems with high organic pollution. The Pp/R ration of the Gulf of Mexico was 1.2, which seems consistent with the fact that there are several upwelling systems in the Gulf (section 2.1.7), but also several large sources of organic pollution in the area (almost all the river discharges, especially the Mississippi River). This criterion, as well suggests that the Gulf of Mexico ecosystem is mature. 80 Margalef s (1968) criterion of maturity as a function of the ratio between total system production and total system biomass (P/B) also supports the idea that the Gulf is a mature system. In developing systems the P/B ratio is high (close to 100) and in mature systems it is low (smaller than 50). Pauly and Christensen (1993) derived a ranking of maturity from 1 to 41 using several aquatic systems in the world. The Gulf of Mexico P/B ratio is 1.25. Cycling is another measure to assess maturity in the systems (Odum, 1969). Finn (1976) expressed cycling as a percentage of the total throughput that is actually recycled in the system. This percentage is assumed to increase as the systems mature. Christensen (1992) showed that maturity is related to stability sensu Rutledge et al (1976) which is related to system overhead. As a consequence, Christensen and Pauly (1993b) assumed that the Finn's Cycle Index (FIC) should also be related to system overhead. They showed that when FCI is plotted against system overhead for a large number of ecosystems (Figure 6 in Christensen and Pauly 1993), the relationship is expressed as a parabola whose maximum point is the optimal stable point. Values away from the maximum are less stable. The Gulf of Mexico is located close to the maximum point, with FCI equal to 14.8 % and system overhead equal to 75.3 % (close to the Gulf of Thailand system), supporting the idea of a system that is highly mature and as a consequence, stable. Christensen and Pauly (1993b) considered path length as another descriptor of an ecosystem. This descriptor is defined as the average number of groups that a flow passes through. It is calculated as the total throughput divided by the sum of the exports and the respiration (Finn 1980). When the FCI is correlated with the Finn's mean path length, another insight of the maturity of the systems could be found. The FCI of 14.8 plotted against a mean path length of 3.6, locate the Gulf of Mexico system within intermediate range among shelves and upwelling/ocean systems. This position is consistent with the characteristics of the GoM and with the way the model was structured. In Fig 9 of Christensen and Pauly (1993b), the Gulf of Mexico will be located very close to the Gulf of Thailand. 81 3.2.4. Transfer efficiencies (TE) A routine in Ecopath can aggregate the entire system in discrete trophic levels sensu Lindeman (1942) through a reverse calculation of fractional trophic levels. From the information generated, it is possible to distribute the flows originated by producers or detritus by trophic levels, and the estimate the transfer efficiency between them. For the Gulf of Mexico model the transfer efficiencies are shown in table 3.20. Table 3.20. Transfer efficiency (%) between trophic level (TL) of the Gulf of Mexico model. Source/TL II III IV V VI VII VIII IX From producers 15.5 14.8 10.2 7.9 6.3 4.6 - -From detritus 14.7 12.2 8.9 6.9 5.3 4.3 - -All flows 15.1 13.5 9.6 7.5 5.9 4.5 3.0 2.5 The mean transfer efficiency in the system (calculated as geometric mean for all flow from II to IV) is equal to 11.4 %. This value means that approximately eleven percent of the energy that enters a trophic level is transferred to the next consumer. The trophic efficiencies are higher at lower trophic levels and decrease at higher levels due to increased respiration (Lindeman 1942). The TE value for the G o M of 11 % is located between the one reported for the deeper tropical shelf of the Gulf of Thailand (TE = 12%) and the one reported for the Ocean South China Sea deeper ocean (TE = 10%), both systems described by Christensen and Pauly (1993). 3.2.5. Trophic impact assessment Fig. 10 shows the relative impact (direct or indirect) that changes in the biomass of one group will have on the biomass of other groups in the ecosystem based on the synthetic model. Those bars pointing upwards indicate positive impacts, while the bars pointing downwards show negative impacts. The majority of the groups, even the non-detritivorous ones, respond positively to an increase in the biomass of detritus. On the other hand, increases in dead-discards (as a detritus group) cause negative effects in all groups that show response to it, except sharks, which apparently benefit. 82 s 3UJ3S ssjry 10 sjoyswo MX 3J0l|SJJ.O 10 SJOLJSUI 6"1 3J0L)SU| MX 3J0L|SU| spjeosippesa snjiJlsa ssHJn} 33s ujaieui'w osy spjiq ess S3!Ge|8dAL)}Bq S3|6e|3dOS3LU P3jd|3dLUS qj3l|J33JUJS| JB0)38JIDS qjsqiss J6J| | jeojssjGj] ssgsij jsgio >|B|dj33j6j| 6B|3d6"inpv 6e|3d6iAnr p3jdS800Bj| HsijsBoopsw wudsBoows jBOujspseoonpv jBoaispseooAnp SBoapsw |npv SB03P3LU Anp wspssooius >l|dsBciows sndopo spodsospjsgio duiuLjs iinpy dujuqs Anp WO dooz UJBO oysui ^doozjeo J J O 0O2A|qj3H Usui )|doo2 qj3|-| B u n e j j d s o j o B u i eunsiuiojOBH S0L|lU3q0|3H SyO )|UB|dOlAL|d qsu| >|UB|doiAqd X S S c £ 9 - 2 p 1 ? - ° £ I f ! o c 7 -SB " T 3 0 O O _ _ * * W W 3 = 1' I1 i ' 'i' 555" o O s o o ,£3 I EC cd 0) .is o o ca C M O c+-3 a C M O C O • M u -a £ o H M -a X o • M 2 P H There are some groups whose increase in biomass has a negligible effect on other groups (not apparent due to the scale used in the graph). These are small coastal detritus and plant consumers (mullets and gobies), large reef planktonics, small reef carnivores and bathypelagics. The negligible effect of these resources in the entire ecosystem seems related to their very particular role. For example, reef organisms represent a very small biomass in the system and their role is only related to other reef groups. On the other hand, some groups positively or negatively affect a large number of groups. Some of these important groups are: macroinfauna, macroepifauna, other decapods, small coastal planktonics (clupeids, engraulids), adult of medium coastal invertebrates and fish feeders (grunts, croakers, etc.), large coastal predators (large sharks), adults of large pelagics (mackerels and tuna) and other fishes. Most of these resources are basic prey in the system (macroepifauna or clupeids) while others are top predators with varied diet. The importance of sea turtles in the system is apparently negligible, and need not be considered here. However, it is necessary to remember that the landings of these groups were not incorporated into the model since the productivity of the turtles proved to be insufficient to sustain the group. The estimated discards of these animals and the predation by other groups were reduced for the same reason. Unfortunately, the result was a negligible interrelation of these animals with the rest of the elements in the model, and as a consequence, any analysis that includes this group should be taken very carefully. Evidently each fishery is affected positively by biomass increments of the resources caught. However, biomass increment of some resources causes large negative impacts on several fisheries. These resources are adults of medium coastal invertebrates and fish feeders (grunts and croakers), other fishes and small pelagic predators (small scombrids, bluefish and ladyfish). These effects should be related to competition interaction for prey species, since most of these impacting groups have trophic levels higher (TL=3.2 to 3.6) than the average for the fishery (TL=2.9). Resources that cause positive response to several fisheries are microinfauna, macroepifauna, adult shrimps, other decapods and octopus. These effects seem more related to the indirect benefit of the increment of prey food for the resources caught. Currently, support for the idea that marine mammals compete with fisheries for the resources in the sea is increasing. Trites et al. (1997) suggests that this competition is mainly due to indirect competition for primary production (food web competition) rather than direct competition. In the 84 Gulf of Mexico model, only the increment of piscivorous marine mammals biomass (principally dolphins) show a slight negative effect on two inshore fisheries, the inshore trawlers and longline fisheries. This effect could be related to the fact that most of the dolphins (main component of piscivores marine mammals group) consume inshore resources and some of them are those groups discarded by trawlers. The negative effect on inshore trawlers due to dolphins is because all catches (incidental or not) from that fishery are considered for the impact assessment. In this case, those shared resources are bycatch (discards) and not target species for inshore trawlers. The fisheries that affect the most resources in the system are offshore trawlers and the offshore longline fishery. In the first case, the response in the system is mainly due to the large amount of discards. In the second case, the variety and amount of organisms caught directly by the fishery cause the negative effect. 3.2.6. Mean trophic level analysis of landings in the system One analysis of changes in the trophic level of landings from commercial fleets in a time series of 13 years (1985-97) was performed using the trophic level calculated by Ecopath and reported landings mentioned in section 2.19. The results are shown in Fig 11 and 12. Fig 11 shows that, contrary to the expected by the theory of fishing down marine food web (Pauly et al. 1998, the mean trophic level of landings seem to be relatively stable fluctuating between 2.50 to 2.53 from 1985 to 1995. But, since 1995 the TL of landings has increase from 2.50 to 2.57. It is true that the time considered may not be long enough to clearly notice such extensive changes in a TL of the landings in a system; but for this case one element should be considered. Landings of several higher trophic level resources, such as demersal fish, reef fish and medium and large pelagics (TL = 3.0-3.9), have almost doubled in the last 3 years, while landings of low TL («2.7) organism such as small pelagics and shrimps have been declining almost constantly in all the time considered (see Fig 6). As an example, in the Yucatan peninsula, the Mexican government has encouraged the exploitation of offshore resources since 1995 trying to reduce the extensive fishing pressure on inshore resources that show symptoms of overexploitation. As a consequence more medium/large pelagic and reef fish (groupers) were caught in those years (S. Salas, 1998 pers. com). It is also known that the increasing export market of cephalopods (octopus) in the late 1990s, has promoted the exploitation of those resources in the Yucatan shelf (S. Salas, 1998 pers. com). 85 Fig 12 shows that landings in the area from 1985 to 1995 have decreased in total amount while the mean trophic level of the catch has fluctuated. Only three recoveries in the amount of landings are shown in the trend, but these fluctuate only between 900 to 1,400 x 10 tonnes a year. After 1995, when the landings showed a jump in their trophic level, the last recovery of the landings are shown, but again no more than 100 x 103 tonnes a year. These fisheries show the typical curve where the highest landings are associated with low trophic level, pattern predicted by the fishing down the food web theory (Pauly et al. 1999). The current increase in trophic level of landings and amount of landings may be due to the exploitation of pristine fishing grounds rather than any recovery of the system. a o s 2.58 2.57 2.56 2.55 2.54 2.53 2.52 2.51 2.50 2.49 84 86 90 92 Y e a r s 94 96 98 Fig. 11. Mean trophic level of the landings in the Gulf of Mexico (by Mexican and US commercial fleets) from 1985 to 1997. 86 1997 1992 X 1985 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 Landings (t x 106) Fig 12. Mean trophic level of landings in the Gulf of Mexico (by Mexican and US commercial fleets) from 1985 (right end) to 1997 (left end). 3.2.7. Comparison of discards fluxes with other fluxes in the system Results of comparisons between discard flux and total commercial landings, total consumption by potential discards eaters, total production of discarded groups, and finally, total primary production are shown in Tables 3.21, 3.22 and 3.23. Table 3.21. Value of discards and other fluxes compared. Comparison Variable flux value (t -IO6 -year"1) Discards as % of other variable Discards 1.50 -1. Discards vs. Commercial landings 1.20 130.0 2. Discards vs. Q of discards eaters 452.0 0.30 3. Discards vs. P of discarded group 350.0 0.40 Table 3.22. Discards flux relative to production of particular discarded groups. Discarded groups Discards % of P 6 Macroepifauna 0.04 13 Other decapods 0.57 16 Catfishes 33.00 18 Adults grunts 2.91 25 Adults tunas 9.00 27 Other fishes 0.52 36 Dolphins 51.00 38 Sea turtles > 1000.00 2.57 J 2.56 TJ 2.55 I 2- 5 4 | 2.53 1 2.52 cu g 2.51 2.50 87 Table. 3.23. Primary Productivity required to sustain fluxes associated with fisheries. Fluxes % PPR (t/year) Commercial fishery 12.5 Discards 12.0 Total 24.5 * The amount of discards estimated from trawlers operating in the systems is approximately the same (possibly higher) as the total landings in the area (Table 3.21). This means that of almost 3 x 106 tonnes caught annually in the area, one half is discarded. This 'waste' is mainly (73 %) a result of the operation of offshore trawlers. This huge amount of discards represents no more than 1% of the total production of discarded groups and no more than 1 % of the consumption by discard eaters, given the parameters used in this model (Table 3.21). This means that the discarded groups are not losing a lot of production when they are caught and discarded, and that the food available to discard eaters from this source represents only a small snack. However, it was necessary to asses the particular impact of discards by groups discarded. These results are shown in table 3.22. For most of the groups (invertebrates and highly productive fish), the impact of discards is again no more that 2 % of their production. For others that are not very productive or abundant (croakers, mackerels and higher vertebrates), the discards represent 10 to over 1000 % of their productivity. Table 3.23 shows that the PPR to sustain fluxes associated with fisheries is 25 % of the total PP in the Gulf of Mexico. This percentage predominantly sustains the landings of phytoplanktivorous groups. Discards also comprise high trophic level organisms (including higher vertebrates) whose productivity may be reduced by trawler discards. In summary, several of the maturity criteria analysed here suggest that the Gulf of Mexico is a highly mature and relatively stable system. These criteria support the idea of a system not strongly impacted by fisheries. However, the production of some groups is severely affected by discards, and the landings are decreasing in time. Therefore, the Gulf of Mexico may be susceptible to perturbations caused by fisheries. This hypothesis will be further tested using the dynamic multispecies modelling approach described in the next chapter. 88 CHAPTER 4. TEMPORAL AND SPATIALLY EXPLICIT MODELLING This chapter analyses temporal and spatial dynamics of the Gulf of Mexico based on the synthetic model. The chapter is divided into two sections: (4.1) simulates temporal dynamics in the system (Ecosim) when diverse fishing policies related to fishing effort are applied; (4.2) applies spatial parameters (Ecospace) to predict current spatial patterns of biomass distribution in the system, which serve as basis to test some scenarios related to the use of marine protected areas. 4.1. Temporal dynamic multispecies modelling (Ecosim) Since no natural ecosystem shows constant structure and flows over time, a dynamic routine was developed for the analysis of the mass-balance models (Walters et al. 1997). Herein, the Ecosim routine replaces the linear equations that describe the trophic fluxes in mass-balance equilibrium (Ecopath), with differential equations needed to predict changes in biomass through time. I used this routine to explore the stability of the GoM model, especially in response to drastic changes in fishing pressure, and to test various fishing policies. The model was considered stable when five extreme scenarios failed to induce collapse. The vulnerability factor was set at 0.3 (default value corresponding to intermediate state between top and bottom control) for all functional groups. 4.1.1. Making ontogenetic links (juvenile-adult split) explicit For this study, I defined five pairs of functional groups linked ontogenetically. These groups are penaeid shrimps (juvenile and adults), coastal medium invertebrates and small fish feeders (juvenile and adult grunts), coastal large demersal carnivores (juvenile and adult groupers), medium coastal fish feeders-large predators (juvenile and adult sharks), and large shelf pelagic predators (juvenile and adult tunas). Additional Ecosim parameters that explicitly linked the ontogenetic pairs and were defined for this analysis were the curvature parameter of von Bertalanffy growth model (K, in year"1) and the age at which juveniles become adults (Tk, in years). Other parameters were set as default values 89 (section 1.6.2a). Table 4.1 presents these additional parameters defined for each of the pairs included in this model. Table 4.1. Parameters defining the transition from juveniles to adults in the Ecosim split pools. Functional groups K(year1) T k (year _1) 11-12 Penaeid shrimps 2.4 a 0.5 b 17-18 Grunts 0.33 c 1.0 d 19-20 Groupers 0.13 c 3.0° 22-23 Sharks 0.17 c 4.0 c 24-25 Tunas 0.2 c 2.32° a Fernandez et al. (1991); b Castro & Arreguin-Sanchez (1997); c Chavez (1994); d FishBase (1998). 4.1.2. Simulation of diverse fishing policies I used Ecosim to test the effect of currently increasing trends of fishing exploitation in US and Mexican waters and some other scenarios that do not occur but which I considered interesting. The simulations correspond to a 25-year period with in most of the cases, a gradual increase in fishing pressure (3-6 % increase per year). The scenarios tested are shown in Table 4.2. 90 Table 4.2. Scenarios to test policies related to changes in fishing effort in the Gulf of Mexico ecosystem. Scenario Fishing effort (f) Fleet 1. Comb3 Trebling All fleets 2. Ten2 Trebling during 10 years and after All fleets doubling of effort 3. Otw2 * Doubling Offshore trawlers 4. Ps4 * Quadrupling Purse seine 5. Off2 * Trebling All offshore fleets 6. Testl Reduction by half for inshore, doubling All fleets for offshore, PS and Misc. current effort 7. Test2 Doubling for inshore, PS and Misc, and All fleets tripling for offshore * Current trends of fishing effort 4.1.3. Results and discussion Simulated responses of the Gulf of Mexico ecosystem to scenarios one to four are shown in Figures 13-15. These results represent responses due only to biotic interactions (predation and competition) among ecosystem components and fishing. In these figures, each line represents one functional group. Negative changes in biomass are indicated by lines declining below the initial Ecopath biomass at the beginning of the simulation. Lines increasing above this initial biomass represent an increase of the original biomass of those groups. Since the Gulf of Mexico synthetic model incorporates a large number of functional groups, only those changes larger than ten percent of the original biomass are shown in the figures. Table 4.3 shows the relative changes in ecosystem components as a result of all Ecosim simulations and provides various possible explanations for those responses. It is noticeable that in cases where trawler effort was simulated to increase, the decrease of biomass of some groups (such as catfishes and piscivorous marine mammals) is mainly caused by the negative effect of the bycatch of those groups. This is consistent with the results shown in section 3.2.7. For other resources with trophic level ranging from 3 to 4 (e.g., oceanic sharks, adult and juvenile mackerel and tunas and coastal sharks) the biomass decrease was caused either by the strong indirect competition for available prey (organisms with TL = 2-3) between the predators and the 91 fisheries or by the strong reduction of their prey through the fisheries (Scenarios 3 and 4). Benefits obtained by some medium trophic level groups, such as small coastal predators or bluefishes, juvenile and adult grunts and sea birds, were due to their generalist diet, mainly consisting of low trophic level organisms (TL=l-2) whose biomass increases when other predators are caught by the fisheries or when the availability of bycatch increases as a consequence of the simulation. Some scenarios show particularly interesting results. Scenario 2 (Fig. 14) attempts to mimic a situation where in view of greatly reduced biomasses, fisheries managers would reduce the fishing effort. Results from this scenario support the idea of the GoM as a stable system since the model shows the possibility of a gradual recovery in all affected groups. However, this implies that the affected groups may settle at a new level of abundance. The time required for a new stable level to establish itself is surprisingly short. From the results of Scenario 3, two conclusions emerge. First, the strongest effect of an increment of effort by offshore trawlers on affected resources is due to their bycatch and resulting discards. Second, the increase of effort does not have strong consequences on the stability of the system. The first assumption is supported by the results of the independent analysis of bycatch and discards in the synthetic mass-balance model (section 3.2.6). However, the second conclusion requires much caution, because this model does not explicitly account for the effect of the destruction of the soft bottom biota by trawler operations. Buchary (1999) found a very dramatic impact of trawling on living bottom structure and demersal species in the Java Sea ecosystem using a modelling approach similar to that used here. Results from Scenario 6, shown in Table 4.3, support the idea that inshore and offshore strata are coupled, and that the biomass of inshore resources does not increase substantially under low fishing pressure i f the predation in offshore waters remains high. Apparently, for some species this is related to a failure in recruitment when offshore fisheries decrease the biomass of reproductive adults. 92 Fig. 13. Changes in the ratio of biomass in the Gulf of Mexico components with scenario 1. Bottom panel shows the simulation of gradual increase of effort by all fleets until tripled at year 25. Upper panel shows the response of ecosystem components to the scenario. See text for details. Fig. 14. Changes in the ratio of biomass in the Gulf of Mexico components with scenario 2. Bottom panel shows the simulation of rapid increase of effort by all fleets, sustained triple effort for ten years and then decreased and sustained effort at double. Upper panel shows the response of ecosystem components to the scenario. See text for details. 93 Fig . 15. Changes in the ratio o f biomass in the G u l f o f Mexico components with scenario 3. Bottom panel shows the simulation o f gradual increase of effort by offshore trawlers until it is doubled at year 25. Upper panel shows the response o f ecosystem components to the scenario. See text for details. 94 CO & CD >> I/O. CN o CO O •c A 00 fl > CO fl o 'M a CO O o W o -4—» CO CD crj t/3 crj fl CD fl o I o o B <D *^ CO O O <U o tu <D - » -t—» E3 =0 • fl cu co o CD Ci oo: c s ft <u cu 1*2 r cd >-u CU •a 1 -S ftl cu. c> a CD ^ *1 cu CU g w o JS *- <S2 5 S3 iS » cu >-> u CU — 43 cs C/5 o u o oo o U O C L , < ; U U O C H U o rn <-i n i ro o ro O 8 43 TJ 3 C3 43 cu CQ 00 ro <-T3 3 3 -3 cu cu CM t; s o 42 o A o 5 s 2 to -C cu « fl 43 cu 43 >, 6 0 3 -O ^ t/3 o a ca ^ p. w « ° § <fl -TJ *j O 2 "2 o 3 43 13 60 fl O 43 .3 "o CO 33 - .3 3 Q 43 o •4-> CO O >. 42 § .2 i j cu JJ fl 42 cu CU ft C o _ •fl o O CU ft cu X > CU -J3 CU ™ O fl o ro A | o o cu CU "0 £ .3 J=o 42 « cu fl cu 42 C/5 o C3 cu cu cu .. o co T3 o 43 cu •3 cu c£ <s cu o 52 co 43 o fl ,22 M o ^ • f l • f l cu T3 T3 -° cu ft 60 O ro O <N o o w ft4= g CL 43 3 <S S c/) o -a o CN IH CU •2 .> •O 43 00 CU 60 cu ft43 Z • fl-§ 1 8 fl S 42 cd cl ? 5 u B " 3 o ro V V cu •a A co *J C/3 CU _Q C/D CO V V o O <4_ 3 o 00 _ cu co 3 cu 43 00 « c*3 ft 3 o CO Ifl CU 60 42 « ^ CO 35 cu CT t>j o t>3 ro ro —> a CO o oo o o ro ro IT) Results from Ecosim can provide further insights about the direction and the relative magnitude of changes in response to simulated disturbance or management actions. To compare the effects among the scenarios tested, I aggregated the commercially important species (aggregated types) by habitat (Table 4.4). Tables 4.5 and 4.6 show the biomass and the percentage of change in biomass of these aggregated types following various Ecosim runs. A negative percentage indicated that the type biomass decreased as a result of the simulation, compared to the initial biomass. Zero implies no change, and a positive value means an increase in the biomass of that group. Table 4.4. Commercially important aggregated types by habitat. Aggregated types by habitat Groups-Estuarine prey 11 Juvenile shrimps 21 Mullets 16 Catfish 17 Juvenile grunts Shelf prey 15 Clupeids 13 Other decapods 12 Adult shrimps 18 Adult grunts Shelf predators 14 Octopus 19 Juvenile groupers 20 Adult groupers 22 Coastal sharks 24 Juvenile tuna 32 Bluefishes Oceanic predators 23 Oceanic sharks 25 Adult tuna Table 4.5 shows that, for all scenarios, the highest biomass is archived by prey shelf species followed by estuarine prey, shelf predators and oceanic predators. Estuarine prey show higher biomass in scenario 3, while shelf prey are evenly abundant in scenarios 2, 3 and 5. Shelf and oceanic predators do not show strong differences among the scenarios tested. A more informative effect of the scenarios is observed when comparing the percentage change of biomass related to its initial value (arithmetical mean of the groups considered). Table 4.6 shows that oceanic predators are impacted the most in all scenarios, scenario 1 being the most damaging for this group, followed by scenario 7. Scenario 4 slightly benefits estuarine prey while negatively affecting shelf prey. A l l scenarios, except for scenario 3, negatively affect shelf predators. Without taking the effect of bycatch into account, it seems that the smaller damage to 96 Table 4.5. Final biomass of aggregated types under scenarios tested in Ecosim (t/km ). Types\ Scenario 1 2 3 4 5 6 7 Estuarine prey 3.6 3.8 4.4 4.8 4.4 4.6 3.8 Shelf prey 16.1 16.4 16.4 15.6 16.4 16.3 16.3 Shelf predators 3.5 3.3 3.4 3.3 3.4 3.4 3.4 Oceanic predators 0.1 0.2 0.2 0.2 0.2 0.2 0.1 Table 4.6. Percentage of change between initial and final biomass of aggregated types under scenarios tested in Ecosim. TypesVScenario 1 2 3 4 5 6 7 Estuarine prey -18 -15 -6 1 -6 -2 -15 Shelf prey 4 4 2 -3 2 1 3 Shelf predators -14 -6 2 -7 -3 -2 -10 Oceanic predators -61 -41 -4 -5 -29 -22 -49 Table 4.7. Final CPUE for exploited functional groups under differerent scenarios tested in Ecosim. Group-species 1 Scenarios 3 4 5 12 Adult shrimp 0.050 0.058 0.012 0.061 13 Other decapods 0.144 0.156 0.038 0.161 15 Clupeids 0.355 0.253 0.358 0.249 16 Catfish 0.004 0.334 0.131 0.323 20 Adult groupers 0.008 0.005 0.002 0.010 21 Mullets 0.011 0.007 0.003 0.007 25 Adult tunas 0.009 0.017 0.005 0.020 Table 4.8. Percentage of change of CPUE (initial vs final) by fishing gear under different scenarios in Ecosim. Gear Scenarios 1 3 4 5 Inshore trawl -73 -57 -76 -59 Inshore long line -31 -45 -78 -48 Inshore gill net -59 -53 -77 -55 Offshore trawl -62 -10 -77 -9 Offshore long line -47 -45 -78 -8 Offshore gill net -75 -57 -77 -18 Purse seiners -22 -45 -22 -46 Miscellaneous -12 -45 -77 -45 total -50 -34 -59 -33 97 fishing resources is caused by the increment of effort of gear with more specific targets such as trawlers and purse seiners. An analysis of catch per unit effort (CPUE) for commercial important resources was made, based on the catch from those scenarios in which fishing effort increased gradually (scenarios 1,3,4 and 5). Table 4.7 shows that final CPUE of clupeids is the highest (> 0.25) in all scenarios, followed by catfish and other decapods. Adult shrimp show an intermediate («0.05) final CPUE almost evenly among scenarios 1, 3 and 5 1, while the smallest final CPUE (<0.01) occurs in adult groupers, mullets and adult tunas. Table 4.8 shows the percentages of change between initial CPUE and final CPUE for all fishing gear analysed by scenarios with gradual increment of effort. This shows that all scenarios negatively affect the CPUE of all fishing gear. This result is interesting since, as mentioned in section 3.1.12, the ECOPATH estimates of fishing mortalities were smaller than reported in the literature. A l l percentages of change range between 10% and 78%. The overall minimum percentage is obtained in scenario 5 (-33.4%), however this value is not extreme compared to other scenarios. Scenarios 1 and 3 have less of an effect on CPUE for inshore gear where the larger effort is by offshore trawlers and all gear combined. The CPUE of purse seiners is less affected by scenario 4 and by the combined effort of all gear. Miscellaneous gear are highly affected by the increment of purse seiners. 4.2. Spatial dynamic modelling (Ecospace) Organisms in aquatic systems are not evenly distributed. They 'select' preferred habitats depending on the availability of resources (space, food, etc.) and exposure to predation. Fishing effort is also allocated differentially in the ecosystem depending on the richness of the grounds and the cost to operate in them. Integrating this information into aquatic ecosystem studies improves both our knowledge and the performance of the management. Since the Gulf of Mexico integrates a wide diversity of aquatic environments (section 2.1.6) and fisheries, I considered it very useful to predict spatial patterns of biomass of the functional groups and to simulate dynamics using the Ecospace approach (Walters et al. 1999) 1 Note that shrimp fishing mortality is biased downward for reason mention on page 72 section 3.1.12. 98 4.2.1. Base map to predicted biomass distribution of functional groups. The first step was to construct a base map over which the simulation would plot current patterns of biomass distribution for all functional groups in the system. To accomplish this, I defined the same five habitats that I used to construct the Ecopath model, i.e., estuaries, non-estuaries, reefs, soft bottoms and oceanic regions. These habitats were mapped into a grid of 28 by 37 cells. Each cell has a length of 55 km and is solid black when representing land, or contains a number (from 1 to 5) when representing an aquatic area with a defined subsystem or habitat type. The number of cells representing each habitat type is based on the estimates of area by subsystems presented in section 3.1.2. (Table 3.2). Since the estuarine and reef areas are distributed at smaller scale than provided by the base map, the allocation of cells representing those two subsystems was based on the distribution of the larger estuaries and reefs in the GoM. Fig. 16 shows the base map of the Gulf of Mexico, and the habitat type distribution. 4.2.2. Incorporation of spatial parameters Additional spatial parameters by functional group were included to define how the organisms 'select' their distribution within the system. These parameters are preferred habitat, instantaneous dispersal rate, dispersal rate in non-preferred habitat, vulnerability to predators in non-preferred habitat, and relative feeding rate in non-preferred habitat. The last three parameters are defined as multiples of the preferred habitat parameters. Preferred habitats were taken from general references and particular studies about the organisms in the system. They are shown in Table 4.9. 99 Table 4.9. Preferred habitat of group species of the synthetic model used in the GoM base map. Group-species Preferred habitat A l l Oceanic Soft bottoms Reefs Non-Estuaries Estuaries Benthic producers Phytoplk. Inshore Phytoplk. offshore X Meiobenthos X Macroinfauna Macreoepifauna Herb.zooplk. inshore Herb.zooplk. offshore X Carn.zooplk. inshore Carn.zooplk. offshore X Juvenile shrimp Adult shrimp Other decapods Octopus Clupeids Catfishes Juvenile grunts Adults grunts Juvenile groupers Adult groupers Mullets Coastal sharks Oceanic sharks X Juvenile tuna Adult tuna X Soldierfishes Other fishes Morays Parrotfishes Blennies Damselfishes Bluefishes Mesopelagics X Bathypelagics X Seabirds Pisciv.m. mammals X Plank, m. mammals X Sea turtles Detritus X Dead discards X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X 100 4.2.2a. Ecological parameters In the version of Ecospace that I used (January 1999), default instantaneous dispersal rates were provided which were proportional to the trophic level of the group, with values ranging from 1 to 100 (km-year"1). Since this calculation evidently assigned a very small dispersal rate to phytoplanktonic groups and planktivorous marine mammals, the dispersal rates of these groups were adjusted to higher values in order to represent the large influence of currents on the dispersion of planktonic groups and the large mobility of baleen whales in the oceans. Dispersal rates, vulnerability to predation and feeding rates in non-preferred habitats were initially left at the default values (5, 2 and 0.5 respectively) Another crucial element in the distribution of organisms in aquatic systems is the availability of food. The distribution of primary production matches with the distribution of secondary and tertiary production in these systems. In section 2.1.7,1 explained the differences in the magnitude of primary productivity within the GoM ecosystem as a result of the upwelling regions, the influences of runoff from rivers and the presence of reefs. I considered it important to incorporate this information in my spatial model. Thus, I included a map of relative primary production in the area using the option provided in Ecospace for this purpose. Figure 17 shows the map of relative PP that emphasises the differences between inshore, offshore, reef and oceanic sub-systems. 4.2.2b. Fishing parameters A l l fishing gears do not operate in all habitats. The place of operation is restricted to the distribution of target species, bottom type, column water depth and currents, among other characteristics. Given the fact that Ecospace is based in Ecopath and Ecosim for the simulation, I was required to allocate the operation of the fisheries defined in Ecopath to the habitats defined in Ecospace. The default Ecospace allocation of the fisheries is to all defined habitats, but in the Gulf of Mexico model it was possible to give some specificity to where the fleets operated. None of the fisheries was assigned to operate in oceanic waters. Table 4.10 gives the allocation of the fishing gear operations. 101 Fig. 16 Base map of the Gulf of Mexico ecosystem showing the habitat types distribution (numbers). 1. Ocean; 2. Soft bottom; 3. Reefs; 4. Non-estuarine coastal areas and 5. Estuarine areas , 11 11.5 1 •1.5 1 | l .4 1 I 1.5 1.4 1 1.5 1.4 1 1.4 1. 1 1.3 1. .9 1.3 1. .9 1.3 1. .9 14 1.4 .9 |1.5 1. Il.4 1. _ 1 . 5 1.5 1.4 1.4 1 5 • i 1 3 | • 5 | i 14 1.4 1 3 1 3 1.3 1 4 1 4 Tif^gn3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 1.3 4 1.3 1.3 1.3 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 12 1.2 4 1.3 1. 4 .9 .9 4 .9 3 .9 |1.3 1. H.3 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 1. 1.3 1. 1 1.3 1 11.4 1. 1.4 1.4 1.4 1.4 1.4 11.5 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 1. 1.2 1 1.2 1 1.2 1 1.2 1 1.2 1.3 1 1.4 1.4 1.4 1.4 1. 1.4 1.6 1.4 1. 1.4 1.4 1.4 1.3 1.4 I Fig. 17. Relative primary productivity in the Gulf of Mexico. Numbers indicate a gradient of production from very productive areas (1.6 in reefs) to low productive areas (0.9 oceanic regions). 102 Table 4.10. Allocation of fishing gears by habitats in Ecospace Fishing gear Habitat of operation Soft bottoms Reefs Non-Estuaries Estuaries Inshore trawlers X Inshore long line X X X Inshore gill net X X Offshore trawlers X Offshore long line X Offshore gill net X Purse seiners X X Miscellaneous X X X X To allocate the fisheries to some particular regions within the habitat of operation, I incorporated an extra cost to the fishing operation in some regions. For the offshore shrimp trawlers, enough information about allocation of fishing effort within the exploited habitats is available. Upton et al. (1993) gave a detailed description of the fishing effort applied to nearshore and offshore US Gulf waters by the shrimp trawlers. From this study it is known that the greatest number of vessel trips in nearshore (<20 m depth) and offshore (>20 m depth) areas is in Mississippi, Alabama and Texas waters. In Mexican Gulf waters, two well recognized areas with high fishing effort by offshore trawlers are the coasts of Tamualipas-north Veracruz, and the Campeche Bank (coast of Campeche state) (TNP, 1999). This information was incorporated in the base map by assigning an operational cost 50 times normal base to offshore shrimp fishery in the less productive areas (those areas excluding the rich areas mentioned before). For the rest of the fishing fleets, no extra cost of operation was included due to the absence of detailed information on the spatial distribution of fishing effort. 4.2.3. Adjustment of parameters Once the spatial conditions were defined, Ecospace was run to simulate 25 years of changes, at time steps of 0.25 years. The first run used the parameters already described for Ecopath and Ecosim. However, the result was that some resources were predicted to occur in areas from which they are known to be absent, notably the central gyre. To restrict the distribution of these groups, some adjustments were made. For some reef organisms, it was necessary to decrease the 103 dispersal rate to prevent their movement to adjacent non-reef areas. For other organisms (phytoplankton inshore, herbivorous zooplankton inshore, several reef groups and oceanic fishes) it was necessary to increase by several times their vulnerability to predation in non preferred habitats and/or to decrease their possibility to find food in non-preferred habitats. For meso- and bathypelagics, it was necessary to adjust the relative primary production of oceanic regions in the base map to a higher value than the one set at the beginning of the simulation (from 0.7 to 0.9). This change made those areas more productive for oceanic fishes. The final spatial ecological parameters after adjustments are shown in Table 4.11. 104 Table 4.11. Spatial ecological parameters used to simulate the current base map of the GoM. Group species Dispersal rate in Dispersal in non- Vulnerability in Feeding rate in preferred habitat preferred habitat non-preferred non-preferred (km/yr) (relative) habitat (relative) habitat (relative) Benthic producers 1 5 2 0.01 Phytoplk. Inshore 100 5 10 0.01 Phytoplk. Offshore 100 5 2 0.50 Meiobenthos 35 5 2 0.50 Macroinfauna 37 5 2 0.10 Macreoepifauna 40 5 2 0.50 Herb.zooplk. inshore 33 5 10 0.10 Herb.zooplk. offshore 31 5 2 0.50 Carn.zooplk. inshore 61 5 10 0.10 Carn.zooplk. offshore 65 5 2 0.50 Juvenile shrimp 49 10 2 0.10 Adult shrimp 59 10 2 0.10 Other decapods 56 5 2 0.10 Octopus 73 5 2 0.10 Clupeids 59 5 2 0.10 Catfishes 71 5 2 0.10 Juvenile grunts 70 5 2 0.50 Adults grunts 75 5 2 0.50 Juvenile groupers 91 5 2 0.10 Adult groupers 94 5 2 0.50 Mullets 38 5 2 0.10 Coastal sharks 95 5 2 0.50 Oceanic sharks 96 10 1 0.80 Juvenile tuna 93 5 2 0.30 Adult tuna 100 5 2 0.50 Soldierfishes 26 5 2 0.001 Other fishes 75 5 2 0.30 Morays 10 5 6 0.10 Parrotfishes 34 5 20 0.001 Blennies 5 5 6 0.001 Damselfishes 31 5 6 0.001 Bluefishes 88 5 2 0.50 Mesopelagics 76 10 10 0.10 Bathypelagics 52 5 10 0.10 Seabirds 85 5 2 0.50 Pisciv.m. mammals 97 5 1 0.40 Plank, m. mammals 92 5 2 0.40 Sea turtles 56 5 2 0.50 Detritus 10 5 2 0.50 Dead discards 1 5 2 0.50 A n additional element to consider in the adjustment of Ecospace parameters was the final value of biomass of the organisms relative to their original biomass value. This information is shown 105 as a trim plot as Ecospace simulations progress. In this plot, after the first run in Ecospace some organisms showed a biomass smaller than 1/8 their original value. In some cases, the groups even collapsed after a few years. Evidently, this was a result of the conditions of high mortality by natural or by fishery causes, or by scarce food availability in the system for those organisms. Shrimps (juvenile and adults) decline was due to a movement parameter that affected the juveniles while moving to adult areas. Other groups whose biomass decreased more than expected (<l/3 of their original value) were offshore carnivorous zooplankton, juvenile and adult of groupers, and large and small reef herbivores. To maintain these organisms without severely decreasing their biomass, their original vulnerability default values to their predators were changed in Ecosim. The final vulnerability value for zooplankton was 0.25, 0.2 for reef herbivores, 0.1 for groupers and 0.05 for shrimps. For shrimps it also was necessary to increase the biomass in Ecopath from 0.3 to 1.12 for juveniles and from 0.2 to 3.38 for adults (the original values originated from the weighting method described in section 3.1.5a). These changes of vulnerability factors do not represent an artefact to adjust the model. As pointed out by Walters et al. (1997) natural ecosystems are likely to have a mixture of high and low vulnerabilities in their components. In complex systems, it is difficult to predict functional responses to changing fishing patterns. With these adjustments, the final base map with predicted biomass distribution of all functional group was obtained. 4.2.4. Simulations of spatial fishing policies (MPA) Once the base map with reasonable predictions of biomass was obtained, it was possible to simulate some fishing policies related to marine protected areas in the region. A marine protected area (MPA) as defined by IUCN 1 has the objective to protect the biodiversity and environmental quality of a marine area for sustainable resource use. There are few protected areas in Gulf waters. In the Mexican Gulf region only three coastal regions in Yucatan are currently considered protected: Dzilam de Bravo, Ria Celestun and Ria Lagartos. These regions account for a total area of approximately 1,400 km 2 (WCMC 1998). In US waters, there are 24 protected areas in the Florida Keys National Marine Sanctuary (Ault et al. 1997) and the Flower Garden National 1 IUCN definition of marine protected area is any area of intertidal or subtidal terrain, together with its overlying water and associated flora, fauna, historical and cultural features, which has been reserved by law or other effective means to protect part or all the enclosed environment (resolution GA17.38, 17th General Assembly, IUCN) 106 Marine Sanctuary in Texas. In this case, the areas protected consist mainly of coral reefs. The extension of the largest sections of these regions (Key Largo and Looe Key reef) is not larger 2 2 than 300 km (Causey 1995). Recently, the protection of a reef region of 1,096 km was accepted by the Gulf of Mexico Fishery Management Council to avoid extensive fishing mortality of the reproductive aggregations of gag grouper in the southeast coast of Florida (AOC-Fishlink 1999). The addition of all these protected areas would add up to no more than 1,400 km 2 which correspond to less than 1% of the total Gulf waters. DeMartini (1993) found that the efficacy of a marine reserve is species-dependent, but is also related to the natural and fishing mortality and to the rate of transfer from the reserve to the fishery ground. Yield models show that the reserve size and transfer rates are very important to increase fishing yield at high fishing effort (Russ et al. 1992; Attwood et al. 1995). For short-lived, fast-growing species, a reserve may be small (MR=10%) while for long-lived, slow-growing species it should be larger than 30% (DeMartini 1993). Unfortunately, in addition to the small amount of protected areas in the Gulf, no study has been done so far to assess the possible result of a policy that would protect another and more extensive subsystem such as soft bottoms. These regions sustain very important shrimp fisheries and are strongly exploited and damaged by trawling1. For this reason, I simulated M P As of several sizes in soft bottoms and one M P A in estuarine regions to assess the effect on the entire ecosystem, on some commercially important species aggregated by habitat (as in the Ecosim simulations, Table 4.3), and on the total catches by all fishing gear. Among the scenarios of MPA's in 60% of soft bottoms, two options were tested to answer the question: What will be the difference between having a number of small MP As versus having a single large M P A with the same combined surface area. These scenarios are shown in table 4.12 and figures 18a to 18d. Table 4.12. Spatial scenarios tested in the Gulf of Mexico model with Ecospace. Scenarios Setting of MPA's 1 40% of soft bottoms (fragmented) 2 60% of soft bottoms (fragmented) 3 60% of soft bottoms (single area) 4 20% of estuaries Note that fishing mortality for some resources in the model is biased downward for reason mention on page 72. 107 The way to create the simulations when comparing responses of several M P A scenarios is shown in Fig. 19. It is necessary to let the system stabilise, then set a base time (starting biomass and catch), let the program run for some times steps (5), then pause the simulation, create the M P A in the base map and continue the simulation. The final time (final biomass and catch) should be located where the result of the simulation is again stable. Fig. 19. Setting of an M P A scenario. Red lines indicate start and end of the simulation. 4.2.5. Results and discussion The simulated base map with predicted biomass distribution of all functional groups is shown in Fig. 20. Only three groups have a narrower distribution than in reality: adult shrimps (not on all soft bottoms), adult tunas (not in all oceanic regions) and piscivorous marine mammals (not in all oceanic regions). For shrimps, an important cause of mortality is predation (see Appendix 7) and since all their predators are located over soft bottoms, it is not surprising that shrimps have restricted their distribution in those areas. For adult tunas and dolphins, the narrower patterns are caused by higher availability of food in soft bottoms and other non-oceanic systems than in the oceanic habitat. Since these organisms have high dispersion rates and wide range habitats, it is not surprising that the model predicted they would show higher biomass in non-oceanic regions than in food-poor oceanic regions. Other interesting distribution patterns are those shown by dead-discards and detritus. Dead-discards are distributed just as expected, more concentrated in those high effort fishing areas of trawlers than in the rest of the soft bottom areas. 109 o _> m g <u c -S >> > <* o <U _c •C CO a - a I S 3 O S U (90 V CO O 2 ed C O 2 <a S a 5 .2 g ° a O - ° *3 U cu 3 G -C "2 ° » C TJ T ! U I) ^ CO *S co <U C3 co £j O fi a o .a x x> o l i u OH T3 S <H CO • CO s i ! CQ O x> O CN 1. v> o IM i \ Herb, */ QJ R IS is i 1 V S "WV I m l * <r ro Q. > B F u 1 2? • IS& E D .£ c UJ Q. o CL G o r\) fj hi J l L J b <D Si 3 > '-4—> •M g 1 -S o p 13 2 ° cb w a | P O 8 u S a. e 8 .2 sb 3 c^ O C3 R O o o - — .o o O ^ HO •JJ CD CD 3 P 2j M a ™ ' C M S » fi T3 U CO * J co CD cn co M o B * co X o M*2 -3 « <D _ >H CD OH T3 cS ^ > co fi <H co P • (ti pq O J3 d CN • — P O On the other hand, detritus is extensively distributed in the entire area following a natural decreasing gradient of concentration from inshore to oceanic areas. The plot of relative biomass in the system after the 25 years simulation (Fig 21) shows that there are only three groups whose final biomass decreased below 1/3 of their original value. These are shrimps (juvenile and adults) and small reef herbivores. As explained before regarding shrimps, the predation mortality for the reef group is also high and, in this case, very concentrated in the small-preferred habitats (reef areas). The rest of the groups showed a final biomass between 0.4 and 2.3 times the original value. Seabirds are the group with highest relative biomass and smallest mortality. On the other hand, the resultant distribution of fishing effort is shown in Fig. 22. The fishing effort for all gear is located in the areas where they operate. For the offshore trawlers the major effort is concentrated in those rich areas where fishing effort is known to be high. For other offshore gear, fishing effort is distributed evenly since no spatial information of fishing cost was included for them. Fishing effort of inshore gear as well as for purse seiners and miscellaneous gear is concentrated in coastal areas adjacent to the land. s e a b i r d s J u u . c . d e m . c a r n i v a d . c . d e m . c a r n i v s m . r e e f h e r b i u . j u v e n i l e s h r i m p • . :. • • . " -25 y e a x Fig. 21. Plot of relative biomass obtained in the base map predicting current patterns of distribution. Ill Fig. 22. Resultant distribution of fishing effort by gear in the Gulf of Mexico model. Concentration of effort is shown in scale at the bottom. The results of the M P A simulations comparing the percentage of change between initial and final biomass are shown in Table 4.13. The meaning of the results (negative, zero and positive values) follow the same logic as in Table 4.6. Table 4.13. Percentage of change in biomass of aggregated types and all alive species among M P A scenarios tested in Ecospace. Aggregated types MPA Scenarios 1 2 3 4 Estuarine prey 0.4 0.5 0.5 1.0 Shelf prey -0.7 0 -0.05 0.02 Shelf predators -2.5 -2.5 0.3 -0.4 Oceanic predators -0.4 -0.7 7.8 0 All alive groups 0.12 „ n , „ i.. J:_ 0.04 0.09 0.04 •all alive groups equal to all groups excluding detritus and dead-discards. 112 Offshore Lg Offshore Gt J J K . J . " W -Low Fig. 22. Resultant distribution of fishing effort by gear in the Gulf of Mexico model. Concentration of effort is shown in scale at the bottom. The results of the M P A simulations comparing the percentage of change between initial and final biomass are shown in Table 4.13. The meaning of the results (negative, zero and positive values) follow the same logic as in Table 4.6. Table 4.13. Percentage of change in biomass of aggregated types and all alive species among M P A scenarios tested in Ecospace. Aggregated types MPA Scenarios 1 2 3 4 Estuarine prey 0.4 0.5 0.5 1.0 Shelf prey -0.7 0 -0.05 0.02 Shelf predators -2.5 -2.5 0.3 -0.4 Oceanic predators -0.4 -0.7 7.8 0 All alive groups 0.12 0.04 0.09 0.04 "all alive groups equal to all groups excluding detritus and dead-discards. 112 These results show that the joint biomass of all live groups experienced a very slight increase (< 0.1 %) after the M P A scenarios1. Overall, it seems that to protect the biomass in the system, the best option is to protect soft bottoms in 40 % of the total area. Interesting is the fact that equal overall benefit is obtained when a large fragmented M P A (60%) is set up and when a small percentage of estuaries are protected (20%). This supports the notion that estuarine systems are important in the GoM ecosystem. However, looking at the results by aggregated types, only the biomass of estuary prey is conserved in all M P A cases. Evidently the highest benefit (1%) is obtained when the estuaries are protected, presumably because the inshore fisheries decrease the biomass of prey in estuarine areas. Shelf prey species biomass is conserved when the M P A correspond to 60% fragmented soft bottom area and when estuaries are protected (scenarios 2 and 4), but slightly decreased in other scenarios. In my opinion, this is a consequence of greater exposure of these animals to predation by all species that otherwise would be caught by the fisheries, especially when the protected shelf area is small (40%) or when it is large but concentrated. Shelf predator biomass is only conserved when the MPA's soft bottom is large (60%) and condensed (scenario 3) but, in other cases, it decreases. This is a consequence of the increase of competition for resources when the protected shelf area is small or very exposed to predation (very fragmented). Scenario 3 substantially benefits the biomass of ocean predators (« 8%) in the area. These organisms (oceanic sharks and adult tunas) actually obtain most of their food from shelf regions. Having high mobility and few competitors, evidently they benefit when their prey biomass is higher in shelf regions. Results from the comparison of the percentage of change in catches by all gear are shown in Table 4.14. In these cases, greater effects of the M P As could be noticed by looking at the biomass changes. The only scenario that does not cause negative effect on the catches of all gear is the protection of a small area of estuaries (20%) \ On the other hand, the overall fisheries are affected the most when a large single portion of soft bottoms (60%) is protected (scenario 3). Since the trawl fishery is the most intensive in offshore areas, it is not surprising that their total catch is affected when M P As are located over soft bottoms. This fishery benefits slightly when 1 However note that fishing mortality is biased downward for some resources as mentioned in page 72. 113 the estuaries are protected, probably due to the protection of juvenile shrimp. The offshore long line fishery is severely affected by scenario 3, a fact that is not surprising since the biomass of shelf and oceanic predators (target species of this fishery) is protected the most by that scenario. Scenario 3 has a positive effect on purse seine and miscellaneous fisheries. In the first case, the reason is not obvious. In the second case, the main target species of miscellaneous fisheries are macroepifauna, other decapods and octopus (Table 3.13) and the biomass of these organisms increases when large, contiguous areas of soft bottoms are protected (Appendix 8). The negative effect of scenario 1 on inshore trawlers is, in my opinion, related to the indirect effect of high mortality by predation on adult shrimps in protected areas. Since the soft bottom protected area is located close to most estuarine areas (Fig 21a) and the fishing mortality of predators is reduced in that scenario, more adult shrimps die, affecting the recruitment of that group. Table 4.14. Percentage change in catches by gear among M P A scenarios tested in Ecospace. Fishing gears MPA Scenarios 1 2 3 4 Inshore trawlers -10.0 0 0 0 Inshore long line 0 0 0 0 Inshore gill net 0 0 0 0 Offshore trawlers -4.0 -0.3 -9.2 0.4 Offshore long line 0 0 -20.0 0 Offshore gill net 0 0 0 0 Purse seiners 0 0.1 4.1 -0.6 Miscellaneous 2 0 4.0 0 Total -2.2 -0.1 -3.4 0.2 In summary, the temporal and spatial modelling described in this chapter gave some insights about how the Gulf of Mexico ecosystem may respond to increments of fishing pressure or to the protection of some important productive areas. Although the Gulf system never exhibited dramatic changes in any of the simulations, it was evident that intensive fisheries affect some of its components directly or indirectly. Also, the coupling among some pelagic and demersal components, and between inshore and offshore systems, was evident. Overall, the fisheries will not increase their catch per effort as greater fishing effort is applied. The establishment of Marine Protected Areas should be assessed on a per-species basis taking into account ecosystem information, as reducing the fishing mortality of some predators will affect the biomass of some prey species. 114 CHAPTER 5. GENERAL DISCUSION AND CONCLUSIONS The importance of studying Large Marine Ecosystems has been stressed by K . Sherman and co-workers in a series of edited volumes (Sherman et al. 1986, 1989, 1990, 1991 and 1993). Ecosystems consist of a mosaic of spatial elements with distinct biological, chemical and physical characteristics linked through biological and physical transport mechanisms. Variation in time and variation in space are constituent elements of LMEs and must be taken in account to ensure the long-term sustainability of biomass yields from these systems (Sherman et al. 1991) and the health of the systems (Sherman, 1986). The Gulf of Mexico proved to be an appropriate example in the study of Large Marine Ecosystems for at least two reasons: 1) its different components have been the subject of intense study which provides enough biological, temporal and spatial information for preliminary broader analyses; but also, 2) it is a system for which synthesis of information is required to quantify the impact of human activities, such as fisheries. From the research questions mentioned in Chapter l a ) , some objectives were formulated to direct this research. To this point, these objectives were accomplished gradually, since I have proceeded in a series of steps to add more information into the analysis of the Gulf of Mexico ecosystem. I started with the integration of the subsystems information into a synthetic mass-balance model to describe the structure and energetic fluxes of the entire Gulf (Chapters 2 and 3). I proceeded with temporal simulation of the system under different policies of fishing effort, to examine temporal responses of the system to current fishing trends (Chapter 4, first section). Finally, I introduced spatial heterogeneity into the ecosystem analysis to simulate the spatial distribution of functional groups in the system under the status quo and to simulate the spatial response of functional groups under various M P As policies (Chapter 4, second section). a ) Can an integrated mass-balance trophic model such as Ecopath, be used to describe a L M E , such as the G o M ? Is it possible to predict the effect of current fishing trends in a L M E using a spatially integrated model? Is it possible to predict spatial distribution of biomass of a L M E using a spatially integrated model? 115 Even though the Ecopath-Ecosim-Ecospace approach is often perceived as representing 'simple' approach to ecosystem modelling, it represents, up to now, the most appropriate suite of methods to study this oceanic region (and others) given the amount of information required and that which is available. This study represents the first attempt to explicitly incorporate spatial heterogeneity in the study of one Large Marine Ecosystem using a trophic model. Spatial heterogeneity usually is not incorporated into ecosystems simulations because of the logistic difficulties this creates (Holling 1973). In Chapter 2, I reviewed the key oceanographic and ecological characteristics of the Gulf of Mexico and information about the fisheries, to show the major driving forces operating in the system. The Gulf system is a mosaic of spatial elements linked through biological and physical transport mechanisms. Current fisheries operate mainly in shallow continental shelf areas with a trend toward increasing fishing effort. Also there is a tendency for the fisheries to move to offshore areas. US landings account for close to 3/4 of the total landing in the area, with a decreasing trend from the 1980s to the 1990s. In Mexican waters there is a trend towards increasing landings. However, this seems more related to the exploitation of previously unfished areas and populations in the last decade, than to the recovery of overexploited resources. Traditionally, the main exploited resources in the area are small pelagic fishes, macroepifauna and shrimps. However, from the middle of the 1990s, the fisheries seem to have shifted to higher trophic level organisms. I mentioned the different mass-balance models previously constructed to describe Gulf subsystems, and emphasised the gaps of knowledge that restricted analysis of those models. Thus, I constructed the synthetic model trying to incorporate all available information and filling those knowledge gaps to obtain a description of the entire system using the method initially proposed in Pauly et al. (1999b). This included the incorporation of total catches reported from the area, inclusion of discards by trawlers, incorporation of sensitive functional groups (such as marine mammals, turtles and birds) and stratification using homogeneous strata, largely defined by depth. Gaps related to the precise quantitative values of primary production and detritus imports were addressed by incorporating distribution of primary production and by assuming a large import of detritus into the system. Moreover, several spatial components defining the systems were included (distribution of several subsystems and distribution of fishing effort - at least for one of the most extensive fisheries in the region, shrimp trawling). Gaps related to 116 migratory patterns and about illegal and unreported catches could not be filled due to the absence of sufficient information. Some of the important findings resulting from the integration of submodels were the high degree of coupling between benthic and pelagic components and the importance of detritus to the balance of the system. Groups that showed larger impact in several components in the systems were macroinfauna, macroepifauna, other decapods, small coastal planktonics and, medium coastal invertebrate and fish feeders (grunts and croakers) as prey; and, as predators large predators (oceanic sharks) and large pelagic fishes (tunas and mackerels). These groups either represent medium trophic levels linking primary producers and top predators, or are top predators themselves. The most abundant groups were secondary and tertiary consumers with TL from 2.5 to 3.5. But the highest biomass was in the detritus group. Indices of maturity described by Odum (1969, 1971), Margalef (1968) and Finn (1976) support the idea of the Gulf of Mexico as a highly mature and relative stable system. The Gulf is sustained by high contribution of detrital matter, the ratio between its total primary production and total system respiration approaches one, the ratio between total system productivity and total system biomass is low and the relation between Finn's cycling index and system overhead locate the system close to the maximum (stability) point. In several of these criteria the GoM is similar to the Gulf of Thailand. The main flux in the system and the major cause of mortality in the Gulf components is related to consumption or predation. However, the integration of responses by the model to fishing pressure gives us an idea of the importance of fishing in the functioning of the system. Overall, Gulf fisheries operate on medium trophic levels (average 2.9) with very low gross efficiencies, which may be related to the scale of the model. The information from the mass-balance model tells us that the fisheries have very small direct impact on the productivity, biomass or mortality of the groups caught. However, the analysis of mean trophic levels shows us that landings are declining while the trophic level of the catch is increasing. This increase in trophic level of the catch could be related to the exploitation to pristine fishing grounds in the area. Moreover, the indirect impacts of fisheries through bycatch-discards are strong. This is not only due to the total amount discarded and the PP required to sustain it (both nearly equal to the fishery flux), but also because discards represent a very important percentage (>30%) of the productivity of some 117 groups, such as coastal demersal invertebrate feeders (catfishes) and higher vertebrates (dolphins and sea turtles). The use of Ecosim to assess the effect of the fisheries shows the importance of benthic-pelagic linkages through predation and competition. In all Ecosim scenarios tested, the most affected groups were those with low productivity, directly caught by the fisheries, or those (usually pelagic groups) indirectly affected through the extraction of their prey (usually demersal or inshore groups). Those that benefited from the increase in fishing pressure were groups with generalist diets (such as birds) whose food was rendered more abundant by the discards, or those whose competition for prey was reduced. Among the scenarios tested, 3 and 4 (increase of fishing effort by offshore trawlers and purse seiners) were the least damaging in terms of total biomass of live groups. This supports the idea that using selective gear to control the mortality and impact on the ecosystems is beneficial. However, a very important aspect to consider (and to explicitly include in further analysis) is the intensive impact that trawlers have on the productivity of some groups, and on benthic biota. A l l temporal scenarios show a substantial decrease in the final CPUE by almost all fleets. This once again supports the idea that doubling or tripling fishing effort, and reducing the biomass of commercial species, would reduce the profitability of the fisheries. The spatial modelling analysis using Ecospace proved capable of predicting realistic biomass distribution patterns of the functional groups in the area. However, it was necessary to incorporate 'bottom up' control in the vulnerability of some groups, proved to be instrumental to avoid the collapse of some groups (shrimps, and some reef fishes). This support Mangel's (1991) suggestion concerning the relevance of comparing 'bottom up' control with 'top down' control when seeking causes for dynamics of some systems. Evidently, more information about vulnerability factors among the components of aquatic systems wil l enhance our predictions in their distribution and abundance. In general terms it could be said that, except for the evident benefit on ocean predators biomass by scenario 3 (MPA of 60% soft bottoms - condensed areas), none of the other Ecospace M P A scenarios cause notable benefit or damage to the biomass of the commercially important and other species. Without doubt, an M P A of 60% of the soft bottoms area seems very unrealistic for socio-economical reasons, although this large area represents only 16% of the total Gulf area. Of 118 interest is the fact that in terms of biomass of commercially important species the protection of such an extensive region is not the best option. Protecting 40% of the soft bottom areas appears to be a better option overall, although some species seemed to be affected when the fisheries did not keep their predators down. When the effect on the final biomass of all resources is assessed, the overall result of the protection of 20% estuarine areas is similar to the protection of an extensive, dispersed soft bottom area (Scenario 2). This means that the protection of approximately 7 km of estuaries benefits the total biomass of groups the same way as the protection of 265 km of soft bottoms. The contrast seems extreme; however, the importance of estuaries in the Gulf ecosystem was previously known and the protection of some estuarine areas merits further study. Additionally, this scenario (# 4) is the only one that does not affect the total catches of all fishing gear (except purse seiners whose catch is slightly reduced). The use of stratified spatial modelling and its analysis was a useful tool to explore the dynamics of the Gulf of Mexico L M E . However, to enhance the quality of these assessments there are some elements that could be improved. A n important problem in the integration of this synthetic model through the use of information from subsystem models was the high non-independence of diet composition data and some parameters among adjacent subsytems. This problem was very common in models of Mexican subsystems, where insufficient information was available. For the same models, biomass estimates of some resources were difficult to obtain and for that reason were usually calculated by the software or taken from models of similar or adjacent systems. For the entire area, biomass estimates of sea turtles, sea birds, and different categories of plankton groups were unavailable and needed also to be calculated by the software or taken from models of similar systems. As mentioned in section 3.1.12 (page 72), estimated Ecopath fishing mortalities for some resources appear to be excessively low, specially for resources with very patchy distribution such as shrimps. This issue will have to be addressed in a subsequent version of this model, jointly with the question on how, within Ecospace, movements between juveniles and adults (shrimps, section 4.2.3, page 106) generates 'hidden' mortalities not accounted in the version of Ecopath with Ecosim (4.0) used here. A key assumption in the model is the assumption of equilibrium in the dynamics of the system, while we know that a system is affected by diverse oceanological and biologic interactions (as 119 well as by external human activities) that cause shifts in species dominance (Richards and McGowan 1989). I believe that this model and its analysis can serve as a basic building block to increase our knowledge about the GoM ecosystem. Further analyses of the carrying capacity of the system, the primary production required to sustain all catches from the system (including illegal and unreported catches), and detailed analysis by trophic levels and comparison with other systems could be performed. However these analyses have their limits. The synthetic model represents the area in its entire scale, while the fisheries are more concentrated in only 40 % of the total area, thus 'diluting' the fisheries into a larger area. For this reason, management of particular stocks should be kept at the scale that is appropriate to those stocks (Beddington 1986). However, management should also consider the effect of interacting components. Management of an entire region as a L M E is far from being understood and accomplished. Gaps of knowledge of exploited functional groups should be filled regionally (basic ecological parameters, biomass estimates, diet composition and migratory patterns). Spatial information of landings by specific gear types, discards by gear types and fishing effort should be gathered. Major driving forces concerning the variability in biomass yield need to be addressed. Finally, factors that are known to affect the system and its components, but are not included in this model, such as seasonal patterns, pollution, habitat destruction, among others, should be dealt with i f the entire ecosystem is to be treated and managed as a unit. 120 LIST OF REFERENCES Abarca-Arenas, L . G. 1987. Aspectos morfologicos y relaciones ecologicas de las especies de la Familia Gerreidae en la Laguna costera de Santecomapan, Veracruz, Mexico. B . Sc. thesis, ENEP-Iztacala, U N A M , Mexico. Abarca-Arenas, L . G., and Valero-Pacheco, E., 1993. Towards a trophic model of Tamiahua, a coastal lagoon in Mexico. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26, pp. 181-185. Aguirre-Leon, A . , and Yanez-Arancibia, A. , 1986. Las mojarras de la Laguna de Terminos: Taxonomia, biologia, ecologia y dinamica trofica (Pisces: Gerreidae). An. Inst. Cienc. del Mary LimnoL, UNAM, 13:(1) 369-444. Alevison, W. S., and Bannerot, S. P., 1990. Fish communities and fisheries biology. In: Continental Shelf Associates Inc., ed. Synthesis of available biological, geological, chemical, socioeconomic, and cultural resource information for the South Florida area, M M S U.S. Department of Interior Minerals Management Service Atlantic OCS Region. Alexander, R. B. , Smith, R. A. , and Schwarz, G. E., 1997. Estimates of the sources of nitrogen delivered to the Gulf of Mexico by the Mississippi River. In: Q. R. Dokken, G. Meier, and H . Kumpf, eds., Models and mass-balance calculations for the Gulf of Mexico, Report of workshop, April 2-4, Metairie, Louisiana, pp. 68. Alvarez, H. , 1988. Analisis de la pesqueria de cazon de la Peninsula de Yucatan. M . Sc. thesis, Centro de Investigation y Estudios Avanzados del Instituto Politecnico Nacional, CLNVESTAV, Merida, Yucata, Mexico. 133 pp. Alvarez, J. H . , and Arreguin-Sanchez, F., 1990. Evaluation de la pesqueria de cazon (Rhizoprionodon terraenovae) de peninsula de Yucatan, Mexico. In: 2nd Marine Science Congress, La Habana, Cuba. Alverson, D. L. , Freeberg, M . S., Murawski, A. , and Pope, J. G., 1994. A global assessment of fisheries bycatch and discards. FAO Fisheries Technical Paper, 339: 223. Amaratunga, T., 1983. The role of cephalopods in the marine ecosystem. In: J. F. Caddy, ed. Advances in Assessment of World Cephalopod Resources, F A O Fisheries Technical Paper, 231, pp. 379-412. Andersen, K. , and Ursin, E., 1977. A multispecies extension to the Beverton and Holt theory of fishing, with accounts of phosphorus circulation and primary production. Medd.Danm. Fisk=og. Havunders.N.S, 7: 319-435. 121 Antoine, J., 1972. Structure of the Gulf of Mexico. In: R. Henry, ed. Contribution to the Geological and Geophysical Oceanography of the Gulf of Mexico, Houston, pp. 1-34. AOC-Fishlink, 1999. Close fishing area will protect gag grouper spawning. A O K -Fishlink communication network . The Marine Fish Conservation Network (MFCN). Arreguin-Sanchez, F., 1976. Notas preliminares sobre las jaibas (Portunidae, Callinectes spp.) de las lagunas de Mandinga, Veracruz. Memorias de la reunion de Pesca Costera en Mexico, Instituto Nacional de la Pesca, Mexico, pp. 159-171. Arreguin-Sanchez, F., and Chavez, E. A . , 1985. Estado del conocimiento de las pesquerias de camaron en el Golfo de Mexico. Investigaciones Marinas, 2:(2) 23-44. Arreguin-Sanchez, F., Cabrera, M . A. , and Mexicano-Cintora, G., 1987a. Dinamica de la poblacion del mero (Ephinephelus morio) del Banco de Campeche. In: Simp. Biol. Ocean. Pesq., Mexico, La Paz, B.C.S. Arreguin-Sanchez, F., Seijo- Gutierrez, J. C , Fuentes-Castellanos, D., and Solis-Ramirez, M . J., 1987b. Estado del conocimiento de los recursos pesqueros de la plataforma continental de Yucatan y region adyacente. Documento interno CRIP Yucalpeten, CINVESTAV Merida, Delegation de Pesca Yucatan, Merida,. Arreguin-Sanchez, F., 1992. Growth and Seasonal Recruitment of Octopus maya on Campeche Bank, Mexico. NAGA, The ICLARM Quarterly, 1: 31-34. Arreguin-Sanchez, F., Valero-Pacheco, E., and Chavez, E. A. , 1993a. A trophic box model of the coastal fish communities of the Southwestern Gulf of Mexico. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26, pp. 197-205. Arreguin-Sanchez, F., Seijo, J. C , and Valero-Pacheco, E., 1993b. A n application of Ecopath II to the north continental shelf ecosystem of Yucatan, Mexico. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26 pp. Arreguin-Sanchez, F., Ramos-Miranda, J., Sanchez, J. A . , Flores-Hernandez, D., Sanchez-Gil, P., and Yanez-Arancibia, A. , 1994. A diminishing blackish water clam (Rangia cuneata) fishery and perspectives of recovery in Lagoon of Pom, Campeche, Mexico. Jaina, 5:(2) 7. Arreguin-Sanchez, F., Flores-Hernandez, D., Sanchez-Gil, P., Sanchez, J. A. , Ramos-Miranda, J., and Yanez-Arancibia, A. , 1995. Stock-recruitment relationships (SRRs) : A scientific challenge of support fisheries management in the Mexican Gulf of Mexico. Jaina, 6:(1) 9. 122 Attwood, C. G., and Bennett, B. A. , 1995. Modelling the effect of marine reserves on the recreational shore-fishery of the south-western Cape, South Africa. S. Afr. J. Mar. Sci, 16: 227-240. Ault, J. S., Bohnsack, J. A . , and Meester, G. A . , 1997. Florida Keys National Marine Sanctuary: Retrospective (1979-1995) reef fish assessment and a case for protected marine areas. Proceedings of the 2nd Worl Fisheries Congress. Developing and Sustaining World Fisheries Resources. Brisbane, Australia, pp. 415-425. Ault, J. S., Bonhsack, J. A. , and Meester, G. A. , 1998. A retrospective (1979-1996) multispecies assessment of coral reef stocks in the Florida Keys. Fish. Bull. 96:(3) 395-414. Austin, G. B. , 1955. Some recent oceanographic surveys of the Gulf of Mexico. Trans. American Geophysics Union, 36:(5) 885-892. Ayala-Perez, L . A. , Rojas-Galaviz, J. L. , and Aviles- Alatriste, O. A. , 1984. Crecimiento, reproduction y dinamica poblacional de Anchoa mitchilli (Pisces: Engraulidae) en el sistema Palizada- del Este, Campeche, Mexico. Rev. Biol. Trop., 44 (3)/45(l): 507-518. Barba-Torres, J. F., and Gaspar-Dillones, M . T., 1987. Notas acerca de la biologia de Eugerres plumeri (Cuvier, 1830) y de Cynoscion arrenarius (Ginsburg, 1929) en el sistema Lagunar Carmen-Machona, Tabasco, Mexico. In: 9th National Zoology Congress, Villahermosa, Tabasco, Mexico. Barradas, C. C , 1991. Aspectos biologicos-pesqueros preliminares de la mojarra rayada Eugerres plumieri (Cuvier) en la laguna de Tampamachoco, Veracruz, Mexico. B. Sc. thesis, Universidad Veracruzana, Xalapa, Veracruz. 41 pp. Barrett, R. T., Becker, P. FL, Furness, R. W., Hunt, G. L. , Latrouite, D., Montevecchi, W. A. , Olsen, B. , Skov, H. , Tasker, M . L. , and Wright, P. J., 1993. Report of the study group on seabirds/fish interactions. International Council for the Exploration of the Sea, Copenhagen. Batllori, E., Chavez, E. A. , Diaz de Leon, A. , Herrera, A . , Garduno, J., Gonzalez, A. , Gonzalez, M . , and Torruco, D., 1987. Caracterizacion estructural de la Laguna de Celestun, Yucatan, Mexico. In: 1st Marine Science Congress, La Habana, Cuba. Beddington, J. R., 1984. The response of multispecies systems to perturbations. In: R. M . May, ed. Exploitation of marine communities, Springer-Velag, Berlin, pp. 209-255. Beddington, J. R., 1986. Shifts in resource populations in Large Marine Ecosystems. In: K. Sherman and L . M . Alexander, eds., Variability and Management of Large Marine Ecosystems, Westview Press Inc; A A A S Selected Symposium, Colorado, 99, pp. 9-31. 123 Berrien, P., and Finan, D., 1977. Biological and fisheries data on king mackerel Scomberomorus cavalla (Cuvier). National Marine Fisheries Service (Sandy Hook),Technical Service Report, 8, 40 pp. Bierman, V . J. J., Hinz, S. C , Zhu, D., Wiseman, W. J. J., Rabalais, N . N . , and Turner, R. E., 1997. A water quality model for the Gulf of Mexico program. In: R. D. Quenton, G. Meier, and H . Kumpf, eds., Workshop: Models and mass balance calculations for the Gulf of Mexico, Gulf of Mexico Fundation, EPA and N O A A NMFS, Metairie, Louisiana, pp. 73. Bohnsack, J. A. , Harper, D. E., McClellan, D. B. , Sutherland, D. L. , and White, M . W., 1997. Resource survey of fishes within Looe Key National Marine Sanctuary. N O A A Tech Mem NOS M E M D 5, 108 pp. Bonfil, R., 1990. Contribution to the fisheries biology of the silky shark Carcharhinus falsciformis (Bibron, 1839) from Yucatan, Mexico. M.Sc. thesis, University of Wales, Bangor. 77 pp. Bonfil, R., 1997. Status of shark resources in the Southern Gulf of Mexico and Caribbean: implications for management. Fisheries Research, 29: 101-117. Bozada, L . M . , Corpi, R., Lila, H . R., Manzo, N . L. , and Chavez, Z., 1982. Dinamica poblacional de Callinectes danae (Smith) ( Decapoda: Portunidae) del estero de Casitas, Veracruz. In: 4th National Zoology Congress, Mazatlan, Mexico. Branstetter, S., 1981. Biological notes on the sharks of the north central Gulf of Mexico. Contrib. Mar. Sci., 24: 13-34. Branstetter, S. G., 1997. Bycatch and its reduction in the Gulf of Mexico and South Atlantic Shrimp Fisheries. Report of the Gulf and South Atlantic Fisheries Development Foundation, Inc., NOAA. 27 pp. Browder, J. A. , Connor, J., and Sullivan, E., 1991. Paying-passeger recreational fisheries of the Florida Gulf coast and Keys. Marine Fisheries Review, 43:(8) 12-20. Browder, J. A. , 1993. A pilot model of the Gulf of Mexico continental shelf. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26, pp. 279-284. Brown, B . E., Browder, J. A. , Powers, J., and Goodyear, C , 1991. Biomass, Yield models, and Management Strategies for the Gulf of Mexico Ecosystem. In: K . Sherman, L. M . Alexander, and B . D. Gold, eds., Food chains, yields, models and management of Large Marine Ecosystems, Westview Press, Colorado, pp. 125-163. Brasher, H . A. , and Palko, B. J., 1985. Charterboat catch and effort from southeastern U.S. waters,1983. Marine Fisheries Review, 47:(3) 54-66. 124 Bryan, C. E., Cody, T. J., and Matlock, G. C , 1982. Organisms captured by the commercial shrimp fleet on the Texas brown shrimp (Penaeus aztecus) grounds. Texas Parks and WUdlife Dept., Tech. Ser., 31: 26. Buchary, E. A. , 1999. Evaluating the effect of the 1980 trawl ban in the Java Sea, Indonesia: An ecosystem-based approach. M . Sc. Thesis. The University of British Columbia, Vancouver. 134 pp. Bundy, A. , 1997. Assessment and management of multispecies, multispecies, multigear fisheries: a case study from San Miguel Bay, the Philippines. Ph.D thesis, The University of British Columbia, Vancouver,B.C. Canada. 396 pp. Burnett, B . R., 1981. Quantitative sampling of nannobiota (microbiota) of the deep-sea benthos. I l l The Bathyl San Diego Trough. Deep-Sea Res., 28: 649-663. Burrage, D. D., Branstetter, S. G., Graham, G., and Wallace, R. K. , 1997. Development and implementation of fisheries bycatch monitoring programs in the Gulf of Mexico. M A F E S , Information Bulletin 324. Cabrera, M . A. , 1986. Contribution al conocimiento de la pesqueria de carito (Scomberomorus cavalla) en la Peninsula de Yucatan. B . Sc. thesis, Universidad National Autonoma de Mexico, U N A M . Cabrera, J. L . , Salas, S., and Arreguin-Sanchez, F., 1990. A preliminary approach to estimated migration: the case of the spiny lobster (Panulirus argus) from Yucatan, Mexico. In: 43rd Annual Session of the Gulf of Mexico and Caribbean Fisheries Institute, 4-9 November, Miami, Florida. Cabrera-Vazquez, M . A. , and Arreguin-Sanchez, F., 1987. Dinamica de la poblacion de carito Scomberomorus cavalla del Banco de Campeche, Mexico. In: 1st marine Science Congress, La Habana, Cuba. Caddy, J. F., and Garcia, S. M . , 1986. Fisheries thematic mapping. A prerequisite for intelligent management and development of fisheries. Ocednographie Tropicale, 21 :(1) 31-52. Caillouet, C , Jr., Shaver, D., Teas, W., Nance, J., Revera, D., and Cannon, A. , 1995. Relationship between sea turtle stranding rates and shrimp fishing intensities in the northwestern Gulf of Mexico: 1986-1989 versus 1990-1993. Fishery Bulletin, 94:(2) 237-249. Casey, J. G., and Hoey, J. J., 1985. Estimated catches of Large Sharks by U.S. Recreational fishermen in the Atlantic and Gulf of Mexico. NOAA Technical Report NMFS (31) 15-20. 125 Castillo, R. M . , Perez, M . A. , and Kobelkowsky, A. , 1988. Estudio de la biologia de Bairdiella ronchus (Pisces: Scianidae) de la laguna de Tampamachoco, Veracruz, Mexico. Univ. y Ciencia. U.J.A.T., 3:(10) 69-84. Castillo-Geniz, J. L. , Marquez-Farias, J. F., Rodriguez de la Cruz, M . C , Cortes, E., and Cid del Prado, A. , 1998. The Mexican artisanal shark fishery in the Gulf of Mexico: towards a regulated fishery. Mar. Freshwater Res., 49: 611-620. Castro, J. L, 1983. Sharks of North American waters. Texas A & M University College Station, USA. Castro, R. G., and Arreguin-Sanchez, F., 1991. Evaluation de la pesqueria de camaron cafe Penaeus aztecus del Litoral Mexicano del noroeste del Golfo de Mexico. Ciencias Marinas, 17:(4) 147-159. Castro, R. G., and Arreguin-Sanchez, F., 1997. Mortality and stock assessment of the brown shrimp Penaeus aztecus (crustacea: Penaeidae), in the northwest Gulf of Mexico. Rev. Biol. Tropical, 45:(3) 1075-1083. Causey, B. , 1995. A n ecosystem approach to managing marine protected areas for sustainable use. In: A . J. Hooten and M . E. Hatziolos, eds., Proceedings of a workshop. Sustainable financing mechanisms for coral reef conservation., Washington, DC. U S A World Bank, Washington, D.C, 9, pp. 48-53. Chavance, P., Flores-Hernandez, D., Yanez-Arancibia, A. , and Amezcua Linares, F., 1984. Ecologia, biologia y dinamica de poblaciones de Bairdiella chrysoura en la Laguna de Terminos Sur del Golfo de Mexico (Pisces: Sciaenidae). An. Inst. Cienc. del Mar y Limnol, UNAM., 11:(1) 123-162. Chavez, E. A . , 1963. Contribution al conocimiento de la biologia de los robalos, chucumite y Constantino (Centropomus spp.) del Estado de Veracruz (Pisces: Centropomidae). Ciencia XXXII, 5: 141-161. Chavez, E., Hidalgo, E., and Sevilla, M . L. , 1970. Datos acerca de las comunidades bentonicas del Arrecife de Lobos, Veracruz. Rev. Soc. Mex. Hist. Nal, 31: 211-280. Chavez, E. A. , and Fernandez, M . S., 1976. Contribution al conocimiento de la biologia de la jaiba prieta (Callinectes rathbunae, Decapoda, Portunidae) del Estado de Veracruz. 41p. (MS). Chavez, E. A. , 1981. Towards a rational management of western Gulf of Mexico shore fisheries. In: G. E. Lasker, ed. Applied systems and cybernetics, Pergamon Press, 4, pp. 2018-2023. 126 Chavez, L . R., Franco, J., and Abarca-Arenas, G., 1987. Distribution y abundancia de las especies marinas de la cuenca central de la Laguna de Tamiahua, Veracruz, durante el ciclo 1985-1986. In: Mem. IX Congreso de Zoologia, Villahermosa, Tabasco, pp. 18-26. Chavez, E. A. , Garduno, M . , and Arreguin-Sanchez, F., 1993. Trophic dynamic structure of Celestun Lagoon, Southern Gulf of Mexico. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26, pp. 186-192. Chavez, E., 1994. Los recursos marinos de la Peninsula de Yucatan. In: A . Yanez-Arancibia, ed. Recursos faunisticos del litoral de la Plataforma de Yucatan, E P O M E X , Campeche, Mexico, 2, pp. 1-12. Chen, C , and Wiesenburg, D. A. , 1997. Influences of river discharge on biological production in the inner shelf: A coupled biological and physical model of the Louisiana-Texas shelf. In: Q. R. Dokken, G. Meier, and H . Kumpf, eds., Models and Mass-balance calculations for the Gulf of Mexico, Report of workshop, April 2-4, Metairie, Louisiana, pp. 69. Christensen, V . , and Pauly, D., 1992. Ecopath II- A software for balancing steady -state models and calculating network characteristics. Ecol. Modelling, 61: 169-185. Christensen, V. , 1992. Network analysis of trophic interactions in aquatic ecosystems. Ph.D. thesis, Royal Danish School of Pharmacy, Copenhagen. 55 pp. Christensen, V . , and Pauly, D., 1993. Trophic models of Aquatic Ecosystems. I C L A R M Conf. Proc, 390 pp. Christensen, V . , and Pauly, D., 1993b. Flow Characteristics of Aquatic Ecosystems. In: V. Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc, 26, pp. 338-352. Christensen, V . , and Pauly, D., 1995. Fish production, Catches and the Carring Capacity of the World Oceans. Naga, The ICLARM Quarterly, 18:(3) 34-40. Christmas, J. Y . , and Etzold, D. J., 1977. The shrimp fishery of the Gulf of Mexico United States: a regional management plan. Gulf Coast Research Laboratory, Ocean Springs, MS.,Tech. Report Ser., 2:128. Claro, R., 1981. Ecologia y ciclo de vida de la biajaiba, Lutjanus synagris (Linnaeus), en la plataforma cubana. III. Nutrition. Ciencias Biologicas, 6: 93-109. Claro, R., 1983. Ecologia y cilco de vida de la rabirrubia Ocyurus chrysurus (Bloch) en la plataforma cubana. I. Identidad, distribution, habitat reproduction y alimentation. Rev. Invest. Inst. Ocean. Acad. Cienc. Cuba, 15: 34 . 127 Claro, R., and Colas, I., 1987. Algunos elementos sobre el balance energetico del pargo criollo Lutjanus analis (Cuvier 1828) en condiciones experimentales. Reporte de Investigation del Instituto de Oceanologia, Academia de Ciencias de Cuba, 58: 19. Contreras, M . , Arreguin-Sanchez, F., Sanchez, J. A. , Moreno, V . , and Cabrera, M . A. , 1990. Mortality and polulation size of the red grouper (Epinephelus morid) fishery from the Campeche Bank. In: 43rd Annual Session, Gulf and Caribbean Fisheries Institute Florida. Coull, B. C , Ellison, R. L. , Fleeger, W. E., Higgins, R. P., Hope, W. D., Hummon, W. D., Reiger, R. M . , Sterrer, W. E., Thiel, H . , and Tietjen, J. H. , 1977. Quantitative estimates of the meiofauna from the deep sea off North Carolina, U.S.A. Mar. Biol, 39:233-240. Crosby, M . P., Newell, R. J. E., and Langdon, L . J., 1990. Bacterial mediation in the utilization of carbon and nitrogen from detrital complexes by Crassostrea virginica. Limnol Oceanogr, 35:(3) 625-639. Dagg, M . , Grimes, C , Lohrenz, S., McKee, B. , Twilley, R., and Wiseman, W. J., 1991. Continental shelf food chains of the Northern Gulf of Mexico. In: K . Sherman, L . M . Alexander, and B . D. Gold, eds., Food chains, yields, models and management of Large Marine Ecosystems, Westview Press, Inc. Colorado, pp. 67-106. Davis, R. W., Fargion, G. S., and Peake, D. E., 1996. Bird Surveys. Distribution and abundance of cetaceans in the north-central and western Gulf of Mexico: Final Report. Texas Institute of Oceanography and the National Marine Fisheries Service, New Orleans, L A . , Final report, Vollume II. Day, J. M . , Barreiro, M . F., Lei-Lou, F., and Madden, C , 1982. Primary Production in the Laguna de Terminos, a tropical estuary in the southern Gulf of Mexico. Oceanol. Acta, 5:(4) 269-276. Day, J. W. J., Madden, C. J., Ley-Lou, F., Wetzel, R. L. , and Machado, A. , 1988. Aquatic primary productivity in Terminos lagoon, chap. 13. In: A. Yafiez-Arancibia and J. W. Day, eds., Ecology of Coastal Ecosystems in the Southern Gulf of Mexico: The Terminos Lagoon Region, Instituto de Ciencias del Mar y Limnologia, Universidad Nacional Autonoma de Mexico. Coastal Ecology. Instituto L S U , Mexico, D.F., pp. 221-235. Day, J., Hall, C , Kemp, W., and A, Yafiez-Arancibia, A . , 1989. Estuarine Ecology. Wiley- Intercience publication, John Wilwy & sons. New York. 558 pp. Day, J. W. J., Coronado-Molina, C , Vera-Herrera, F. R., and Twilley, R., 1996. A seven year record of aboveground net primary production in a mangrove forest in Laguna de Terminos, Mexico. Jaina, 7:(2) 12. 128 De la Cruz, G., and Franco, J., 1981. Relaciones tro fleas de la ictiofauna de la laguna de Santecomapan, Veracruz, Mexico. In: Memoirs of the VII Latinamerican Simposium on Biological Oceanography, Acapulco, Mexico, pp. 535-546. De la Cruz-Aguero, G., 1993. A preliminary model of Mandinga Lagoon, Veracruz, Mexico. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26, pp. 193-196. DeMartini, E. E., 1993. Modeling the potential of fishery reserves for managing Pacific coral reef fishes. Fish. Bull, 91: 414-427. Department, M . M . , 1958. Estados Unidos Mexicanos, Matamoros 14R-VII (all maps from the Gulf of Mexico coast). Departamento Cartografico Militar (D.C.M.) para la Comision Intersecretarial Coordinadora del levantamiento de la Carta Geografica de la Republica Mexicana, Mexico, D.F. Department Interior U.S. Geological Survey, 1956. Corpus Christi, Texas (all maps from Gulf of Mexico coast). U.S. Geological Survey and National Ocean Service, Denver, Colorado. Desiro, R. B. , 1980. Harvesting strategies and parameter estimation for an age-structured model. Can. J. Fish. Aquat. Sci., 37: 368-382. DeVane, J. C , 1978. Food of king mackerel, Scomberomorus cavalla in Onslow Bay-North Carolina. Transactions of the American Fisheries Society, 107:(4) 583-586. Devaney, T., 1969. Ecological interpretation of distribution of the lanternfishes (Myctophidae) in the Straits of Florida. Ph. D. thesis. University of Miami, Miami. Duncan, C. D., and Havard, R. W., 1980. Pelagic birds of the northern Gulf of Mexico. A preliminary summary distribution and abundance with comments on field identification. American Birds, 34:(2) 122-132. El-Sayed, S. Z., Sackett, W. M . , Jeffrey, L . M . , Fredericks, A. , Saunders, R. P., Conger, P. S., Fryxell, G., Steindinger, K . A. , and Earle, S. A. , 1972. Chemistry, primary productivity, and benthic algae of the Gulf of Mexico. In: V . C. Bushnell, ed. Serial Atlas of the Marine Environment, Am. Geogr. Soc, 22, pp. 1-22. Fernandez, M . I., Castro, M . R., Gonzalez, C. A . , Schultz, R. L. , Nunez, M . G., Wakida, K. A. , Uribe-Martinez, J., Sandoval, Q. M . , Toledo, D. P., Ramirez, S. G., and Aguilar, I. A. , 1998. Fundamento tecnico para el establecimiento de vedas en el Golfo de Mexico en 1998. INP, S E M A R N A P , Mexico, D.F. Field, D. W., Reyer, A . J., Genovese, P. V . , and Shearer, B. D., 1991. Coastal wetlands of the United states. A n Accounting of a valuable national resource. NOAA and U.S. Fish & Wildlife Services. 58. 129 Finn, J. T., 1976. Measures of ecosystem structure and function derived from analysis of flows. J. Theor. Biol, 56: 363-380. Finn, J. T., 1980. Flow analysis of models of the Hubbard Brook ecosystem. Ecology (6):562-571. Finucane, J. H. , and Vaught, R. N . , 1986. Species profile of Atlantic threadherring, Opistonema oglinum (Lesueur, 1818). NMFS-SEFC,NOAA Tech. Mem., 182. FishBase, 1998. FishBase 98 CD-ROM. I C L A R M , Manila. Flint, R. W., and Rabalais, N . N . , 1981. Environmental studies of a marine ecosystem, South Texas outer Continental Shelf. University of Texas Press, Austin, Texas, 240 pp. Galtsoff, P. S., 1964. The American oyster Crassostrea virginica Gmelin. Fish. Bull. (64):231-235. Garcia, A. , 1976. Contribution al conocimiento sobre las relaciones troficas de la ictiofauna marina del area de Laguna Verde, Veracruz. B . Sc thesis, Escuela National de Ciencias Biologicas , INP, Mexico. 72 pp. Garcia, G. J. D., 1985. Estudio del contenido estomacal y tramas troficas de algunas especies icticas (Arius melanopus, Gunther; Bairdiella ronchus, Cuvier y Valenciennes; Centropomus undecimalis, Bloch; Lutjanus griseus, Linnaeus) de la laguna de Tampamachoco, Tuxpan, Veracruz (1981-1982). , Escuela National de Estudios Superiores -Zaragoza, U N A M , Mexico. 70 pp. Garcia, G. A . , 1989. Ecologia y pesqueria de camaron bianco Penaeus settiferus (Linnaeus, 1767) en la Laguna de Terminos, Sonda de Campeche. Ph.D. thesis, Universidad National Autonoma de Mexico , U N A M . Gittings, S., 1999. Florida Gardens National Marine Sanctuary. http:/, 2:( 1) G M F M C , G. o. M . F. M . C , 1991b. Fishery management plan for the shrimp fishery of the Gulf of Mexico, United States waters (revised November 1981). G M F M C , Tampa, Florida. G M S A F M C , G. o. M . a. S. A . F. M . C , 1985. Amendment 1 to the fishery management plan, environmental impact statement for the coastal migratory pelagic resources (mackerels). Tampa and Charleston. Gomez-Aguirre, S., 1989. Florecimiento de Pyrodinium (Dinoflagellata) en la Laguna de Tamiahua y su persistencia en el periodo 1984-1986. In: Coloquio sobre Invertebrados, Mexico, D.F., Hidrobiologia Tampamachoco, Depto. E l Hombre y su Ambiente, U A M -Xochimilco. 130 Gonzalez, O. C , 1975. Dependencia afloramiento-capturas y pronostico a corto plazo en la region oriental del Banco de Campeche. Rev. Invest., INP, 1 :(1) 3-25. Gonzalez, G. C , 1986. Mortalidad y crecimiento de Mugil curema Valenciennes en la laguna de Tampamachoco, Tuxpan, Veracruz, Mexico. In: II Reunion Alejandro Vilalobos. Biologia de laguna de Tamiahua. 22-24 octubre. U N A M , Mexico, 70 pp. Gonzalez, M . E., Sanchez, J. A. , and Arreguin-Sanchez, F., 1990. Popultion dynamics of the red snapper (Lutjanus campechanus) fishery from the Campeche Bank, Mexico. In: 43rd Annual Session, Gulf and Caribbean Fisheries Institute Florida. Gonzalez-Cano, J., Arreguin-Sanchez, F., and Contreras, M . , 1994. Diagnostico del estado de la pesqueria del mero (Ephinephelus morio) en el Banco de Campeche. SEPESCA, Mexico. Gonzalez-Sanson, G., and Rodriguez-Vinas, L. , 1983. Alimentation natural de Eugerres brasilianus (Cuvier) y Gerres einereus (Walbaum) (Pisces :Gerreidae) en las lagunas costeras de Tunas de Zaza, Cuba. Revista de Investigaciones Marinas, 48:(1) 91-133. Goodyear, C. P., 1992. Red snappers in U.S. waters of the Gulf of Mexico. NMFS Southeast Fisheries Center, Miami lab, 75 Virginia Beach Drive, Miami F l 33149, Miami, F l , Contribution M I A 91/92-70. Gordon, D. C , 1970. Some studies on the distribution and composition of particulate organic carbon in the North Atlantic Ocean. Deep-Sea Res., 17: 233-243. Goreau, C. P., 1959. The ecology of Jamaican coral reefs. Ecology, 40: 67-90. Gratia, A. , and Kensler, C. B. , 1980. Las langostas de Mexico: su biologia y pesqueria. An. Inst. Cienc.del Mary Limnol., Univ. Nal. Auton. Mexico, 7:(2) 111-127. Gratia, A. , and Vazquez-Bader, A. R., 1995. Pesquerias de camaron del Sur del Golfo de Mexico. Estado actual y alternativas de manejo. Jaina, 6:(1) 9-10. Guevara Carrio, E., Bosch Mendez, A. , Suarez Mulkay, R., and Lalana Rueda, R., 1994. Alimentation natural de tres especie de pargos (Pisces: Lutjanidae) en el archipielago de los Canarreos, Cuba. Revista de Investigaciones Marinas, 15:(1) 63-72. Guzman, H . V . , 1987. Dinamica poblacional del camaron rosado (Penaeus duorarum duorarum Burkenroad 1939) del Banco de Campeche durante tres temporadas de pesca 1975-1977. B. Sc.thesis, Universidad Veracruzana, Veracruz, Mexico. 40 pp. Hagenkotter, C. I., 1993. Final Report on the Hagenkotter Beam Trawl Project. Gulf & South Atlantic Fisheries Development Foundation, GSAFDFI No. 49-08-13564. 131 Harrison, P., 1983. Seabirds: An identification guide. Houghton Mifflin Co., Boston, 448 pp. Heald, E. J., 1969. The production of organic detritus in a south Florida estuary. University of Miami, Coral Gables. Hildebrand, H. , 1963. Hallazgo del area de anidacion de la tortuga marina lora, Lepidochelys kempi (Garman) en la costa occidental del Golfo de Mexico. Ciencia, 22:(1) 105-112. Hildebrand, H . H. , 1979. A Historical review of the status of sea turles populations in the Western Gulf of Mexico. In: K . Bjorndal, ed. Worl Conference on Sea Turtle Conservation, Washington, D.C., Smithsonian Institution Press, pp. 447-453. Hi l l , B. J., and Wassenberg, T. J., 1990. Fate of discards from prawn trawlers in Torres Strait. Aust. J. Mar. Freshwater Res., 41: 53-64. Holling, C. S., 1973. Resilience and stability of ecological syatems. Ann. Rev. Ecol. Syst. (4) 1-23. Hopkins, T. L. , 1990. Zooplankton. In: I. Continental Shelf Associates, M M S U.S. Department of the Interior Minerals Management Service Atlantic OCS Region, ed. Synthesis of available Biological, Geological, Chemical, Socieconomic, and Cultural Resource Information for the South Florida Area, pp. 195-230. Hueter, R. E., and Manire, C. A. , 1994. By-catch and catch release mortality of small sharks in the Gulf coast nursery grounds of Tampa bay and Charlotte Harbor. Mote Marine Laboratory, N O A A / N M F S , Technical report, 368. Hunter, J., and Feller, R. J., 1987. Immunological dietary anlysis of two peneid shrimp species from a South Carolina tidal creek. J. Exp. Mar. Biol. Ecol. (107) 61-70. Ibarra-Obando, S., Smith, S. V. , and Contreras-Espinosa, F., 1997. Mexican coastal lagoons overview. In: S. V . Smith, S. Ibarra-Obando, P. R. Boudreau, and V . F. Camacho- Ibar, eds., Comparison of Carbon, Nitrogen and Phosphorus fluxes in Mexican Coastal Lagoons., LOICZ reports & studies, The Netherlands, 10, pp. 84. LNP, 1994. Atlas Pesquero de Mexico. Secretaria de Pesca, LNP, 234 pp. LNP, 1999. Camaron del Golfo de Mexico y Mar Caribe. In: INP, ed. Diagnostico de las pesquerias relevantes en Mexico, Documento interno INP, S E M A R N A P , Mexico, D.F. Kanciruck, P., 1980. Ecology of juvenile and adult of Palinuridae (spiny lobster). In: J. S. Cobb and B . C . Philips, eds., The biology and management of lobsters, Academic Press, New York. 132 Kitchell, J. F., Eby, E. A. , He, X . , Schindler, D. E., and Wright, R. M . , 1994. Predatory-prey dynamics in an ecosystem context. J. Fish. Biol, 45: 209-226. Knudsen, E. E., Rogers, B . D., Paille, R. F., Herke, W. H. , and Geaghan, J. P., 1996. Juvenile white shrimp growth, mortality, and emigration in weired and unweired Louisiana marsh ponds. North. Amer. J. Fish. Manag., 16:(3) 640-652. Kondratyeva, T. M . , and Sosa, E., 1966. Primary productivity of Cuban coastal waters. Issledovaniya Tsentral'no Amerikanskikh morey (Studies on the Central American seas), Kiev: Noukova Dumka Publishing House, 1, pp. 68-80. Laevastu, T., and Favorite, F., 1977. Preliminary report on dynamical numerical marine ecosystem model (DYNTJMES II) for Eastern Bering Sea. U.S. Natl Mar. Fish. Serv. Northwest and Alaska Fish. Center, Seatle. (mimeo) Lankford, R. R., 1976. Coastal laggons of Mexico, their origin and classification. In: M . Wiley, ed. Estuarine processes. Circulation of sediments and transfer of material in the estuary, Academic Press, pp. 182-215. Larkin, P. A . , and Gazey, W., 1982. Applications of ecological simulation models to management of tropical multispecies fisheries. In: D. Pauly and G. I. Murphy, eds., Theory and management of tropical fisheries, I C L A R M Conference Proceedings, 9, pp. 123-140. Lewis, R. R., and Estevez, E. D., 1988. Ecology of Tampa Bay Florida: A n estuarine profile. Mangrove Systems, Inc, Tampa, Flo.,Biological Report, 85(7.18). Licea, S., Luna, R., and Delgado, D., 1987. Fitoplankton y PrOductividad. In: S. Licea, ed. Informe final del estudio de evaluation del plancton y la productividad primaria de la laguna de Tamiahua, Veracruz. Informe Tecnico, Instituto de Ciencias del mar y Limnologia, Universidad National Autonoma de Mexico, Mexico, pp. 58-122. Lindeman, R. L. , 1942. The trophic-dynamic aspect of ecology. Ecology (23) 399-418. Longhurst, A . R., 1995. Seasonal cycles of pelagic production and consumption. Prog. Oceanogr., 36:(2) 77-167. Lynch, S. A . , 1954. Geology of the Gulf of Mexico. U.S. Fish and Wildlife Service Fisheries Bulletin, 55:(89) 67-83. Maharaj, V. , and Recksiek, C , 1991. The by-catch from the artisanal shrimp trawl fishery, Gulf of Paria, Trinidad. Mar. Fish. Rev.(53) 10-15. 133 Mallard Colmenero, L. , Yafiez-Arancibia, A. , and Amezcua Linares, F., 1982. Taxonomia y aspectos biologicos y ecologicos de los Tetraodontidos de la Laguna de Terminos, sur del Golfo de Mexico (Pisces: Tetraodontidae). An. Inst. Cienc. del Mar y Limnol. UNAM, 9:(1) 161-212. Mangel, M . , 1991. Empirical and theoretical aspects of fisheries yield models for Large Marine Ecosystems. In: K . Sherman, L . M . Alexander, and B. D. Gold, eds., Food chains, yields, models and management of Large Marine Ecosystems, Westview Press, Oxford, pp. 243-261. Manickchand-Heileman, S., Soto, L . A. , and Escobar, E., 1998. A preliminary trophic model of the Continental shelf, south-western Gulf of Mexico. Estuarine, Coastal and Shelf Science, 46: 885-899. Manickchand-Heileman, S., Arreguin-Sanchez, F., and Lara-Dominguez, A . , 1998. A trophic model and network analysis of Terminos lagoon, southern Gulf of Mexico. Ann. Inst. Ciencias del Mar y Limnologia, UNAM. Mann, K . H. , 1982. Eclogy of coastal waters. A systems approach. Studies in Ecology., University of California Press, Blackwell, Oxford, 8, pp. 188-190. Manooch, C. S., 1976. Age, growth and mortality of the white grunt, Haemulon plumieri Lacepede (Pisces: Pomadasyidae) from North Carolina. In: Southeast Association of Game and Fish Commissions 30th Annual Conference. Manooch, C. S., and Borans, C. A. , 1982. Distrubution, abundance, age and growth of the tomtate, Haemulon aurolineatum, along the southeastern United States coast. Fish. Bull, 80:(1) 1-18. Margalef, R., 1968. Perspectives in ecological theory. University Press, Chicago, 111 pp. Marquez, R. M . , Villanueva, A. , and Sanchez, M . P., 1979. The Population of the Kemp's Ridley Sea Turtle in the Gulf of Mexico- Lepidochelys kempii. In: K . Bjorndal, ed. Worl Conference on Sea Turtle Conservation, Washington, D.C., Smithsonian Institution Press, pp. 159-164. Martinez, O. J., 1987. Aspectos poblacionales y pesqueros de Diapterus auratus (Ranzani 1842) en la Laguna de Tampamachoco, Veracruz. B . Sc. thesis Universidad Veracruzana, Tuxpan, Veracruz. 52 pp. Meaden, G. J., and Do Chi, T., 1996. Geographical information systems. Aplications to Marine Fisheries. FAO, Rome, Fisheries technical paper, 356. 335.pp. Merino, M . L, 1992. Afloramiento en la Plataforma Continental de Yucatan: Estructura y fertilization. Ph.D. thesis Especialidad en Oceanografia Quimica, Universidad National Autonoma de Mexico, Mexico, D.F. 134 Mexicano-Cintora, G., and Arreguin-Sanchez, F., 1987. Dinamica de las poblaciones de rubia (Lutjanus synagris) y canane (Ocyurus chrysurus) de las costas de Yucatan, Mexico. In: 1st Congreso de de Ciencias del Mar, 9-12 June, La Habana, Cuba. Mexicano-Cintora, G., Salas, S., and Cabrera, M . A. , 1996. Analysis of the fishery and population dynamics of the Atlantic thread herring (Opisthonema oglinum) in Yucatan, Mexico. Ciencias Marinas, 22:(1) 9-22. Minello, T. J., Zimmerman, R. J., and Martinez, E. X . , 1989. Mortality of young brown shrimp Penaeus aztecus in estuarine nurseries. Transactions of the American Fisheries Society, 118: 693-708. Moe, M . A. , 1969. Biology of the red grouper Epinephelus morio (Valenciennes) from the Eastern Gulf of Mexico. Florida Department of Natural resources Professional Paper Series, 10. Moore, K . A. , and Wetzel, R. L. , 1988. The distribution and productivity of seagrasses in the Terminos Laggon. In: A . Yanez-Arancibia and J. W. J. Day, eds., Ecology of coastal ecosystems in the southern Gulf of Mexico: the Terminos Lagoon Region, ICMyL, U N A M ; Coast. Ecol. Inst. L S U . Editorial Universitaria, Mexico, D.F. Morrison, M . J., and Nowlin, W. D., 1977. Repeated nutrient, oxygen and density sections through the Loop Current. J. Marine Research, 35:(1) 105-127. Mortimer, J. A. , 1981. Feeding Ecology of Sea turtles. In: K . Bjorndal, ed. Worl Conference on Sea Turtle Conservation, Washington, D.C., Smithsonian Institution Press, pp. 103-109. Mullin, K . D., and Hansen, L. J., in press. Marine mammals in the northern Gulf of Mexico. In: T. G. Mcllwain and H . E. Kumph, eds., Gulf of Mexico: A Large marine ecosystem, Simposium Proceedings. Nance, J. M . , 1993. Effort trends for the Gulf of Mexico shrimp fishery. N O A A , NMFS-SEFSC, Technical Memorandum, 337, 37 pp. Naughton, S. P., and Saloman, C. H. , 1981. Stomach contents of juveniles of king mackerel (Scomberomorus cavalla) and Spanish mackerel (S. maculatus). Northeast Gulf Sci., 5:(1) 71-74. Nedelec, C , 1990. Definition and classification of fishing gear categories. Food and Agriculture Organization of the United Nations FAO, Rome,FAO fisheries technical paper, 222, 92 pp. 135 Nichols, S., Shaw, A. , Pellerin, G. J., and Mullen, K. , 1987. Estimates of shrimp fleet bycatch for thirteen species in the offshore waters of the U.S. Gulf of Mexico. 1972-1989. NMFS, Pascagoula, MS, 39568. Nichols, S., Shaw, A. , Pellerin, G. J., and Mullen, K. , 1990. Updated estimates of shrimp fleet bycatch in the offshore waters of the U.S. Gulf of Mexico 1972-1989. NMFS, Pascagoula, MS, 39568. NMFS, 1997. Annual landings from fisheries of United States. http:/www/ NMFS, N . M . F. S., 1998. Fisheries Statistics and Economic Division. Recreational Fisheries. http:/ N O A A , 1990. Estuaries of the United States. Vital statistics of a national resource base. N O A A , U.S, Department of Commerce, Special 20th anniversary report, 20. Nowlin, W. D., 1971. Water masses and general circulation of the Gulf of Mexico. Oceanology, 452: 173-178. Nunez, A. , 1987. Contribution al conocimiento biologico pesquero del charal Anchoa hepsetus (Linnaeus, 1975) de las costas de Campeche. B. Sc. thesis, Universidad Autonoma de Nuevo Leon, Mexico. 39 pp. Ochoa, R., and Ramirez, H . R., 1987. Variation estacional de los dinoflagelados en la Laguna de Tamiahua, Veracruz, Mexico. Memorias del VI Simposio Latino Americano sobre Oceanografia Biologica, 15-19 nov. 1981, Acapulco, Guerrero, Mexico. Odum, E. P., 1969. The strategy of ecosystem development. Science, 164: 262-270. Odum, E. P., 1971. Fundamentals of Ecology. Saunders, Philadelphia, 574 pp. Okey, T. A . , and Pauly, D., 1998. Trophic mass-balance model of Alaska's Prince William Sound ecosystem, for the post-spill period 1994-1996. The Fisheries Centre, University of British Columbia, Vancouver, B.C. Canada,Research reports, ISSN 1198-6727. Opitz, S., 1991. Quantitative Models of Trophic Interactions in Caribbean Coral Reefs. Ph. D. thesis, Kiel University, Germany. Opitz, S., 1993. A quantitative model of the trophic interactions in a Caribbean coral reef ecosystem. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M , Conf. Proc. 26, 259-267. 136 Ortner, P. B. , Ferguson, R. L. , Piotrowicz, S. R., Chesal, L. , Berberian, G., and Palumbo, A. V. , 1984. Biological consequences of hydrographic and atmospheric advection within the Gulf Loop Intrusion. Deep-Sea Res, 31: 1101-1120. Ortuno, M . G., 1998. Biomasa zooplanctonica en el Golfo de Mexico durante los veranos de 1986 a 1990. B . Sc. thesis. (Biologo), Universidad Nacional Autonoma de Mexico, Mexico, D.F. Palomares, M . L . D., and Pauly, D., 1989. A multiple regression model for predicting the food consumption of marine fish populations. Aust. J. Mar. Freshwat. Res., 40: 259-273. Palomino, O. I., Castro, R. G., Fernandez, R., and Rangel, A . , 1996. Composition estacional de la fauna de acompanamiento del camaron en el noroeste del Golfo de Mexico. Centra regional de Investigation pesquera Tampico, Tampico, Ver.Jnforme Tecnico del INP (Documento Interno no Publicado). Parsons, T. R., Takahashi, M . , and Hargrave, B. , 1984. Biological oceanographic processes. Butterworth-Heinemann Ltd, London, England. Paul, R. K . G., 1981. Natural diet, feeding and predatory activity of the crabs Callinectes arcuatus and C. toxotes (Decapoda, Brachyura, Portunidae). Marine Ecology Progress Series, 6: 91-99. Pauly, D., 1985. Population dynamics of short-lived species with emphasis on squids. North Atlantic Fisheries Organization Science Council Studies, 9: 143-154. Pauly, D., 1986. A simple method for estimating the food consumption of fish populations from growth data of food conversion experiments. U.S. Fish Bull., 84:(4) 827-840. Pauly, D., and Soriano, M . L. , 1986b. Some practical extensions to Beverton and Holt's Relative Yield-Per-Recruit model. In: J. L. Maclean, L . B . Dizon, and L . V . Hosillos, eds., The First Asian Fisheries Forum, Manila, Philippines, pp. 491-495. Pauly, D., and Christensen, V . , 1993. Stratified models of Large Marine Ecosystems: A general approach and an application to the South China Sea. In: K . Sherman, L. Alexander, and B . Gold, eds., Large Marine Ecosystems: Stress Mitigation and Sustainability, A A A S , Washington, D.C., pp. 148-174. Pauly, D., Soriano-Bartz, M . L. , and Palomares, M . L . D., 1993b. Improved Construction, Parametrization and Interpreatation of Steady-State Ecosystem Models. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc, 26, pp. 1-13. Pauly, D., and Christensen, V. , 1995. Primary production required to sustain global fisheries. Nature, 374: 255-257. 137 Pauly, D., 1998. Large Marine Ecosystems: Analysis and Management. Benguela Dynamics. S. Afr. J. mar. Sci., 19: 487-499. Pauly, D., Trites, A . W., Capuli, E., and Christensen, V . , 1998. Diet composition and trophic levels of marine mammals. ICES Journal of merine Science, 55: 467-481. Pauly, D., Christensen, V. , and Dalsgaard, J., 1999. Fishing Down Marine Food Webs. Science, 279: 860-863. Pauly, D., Arreguin-Sanchez, F., Browder, J., Christensen, V . , Manickchand-Heilemann, S., Martinez, E. and Vidal. L . 1999b. Toward a stratified mass-balance model of trophic fluxes in the Gulf of Mexico Large Marine Ecosystem. In: Ff. Kumpf, K . Steidinger and K. Sherman, ed. Blackwell Science (publishers). The Gulf of Mexico Large Marine Ecosystem- Assessment, Sustainability, and management, pp. 278-293. Peacock, N . A . , 1974. A study of the spiny lobster fishery of Antigua and Bermuda. In: Gulf. Caribb. Fish. Inst. Proc, pp. 117-130. Peake, D. E., 1996. Bird Surveys. In: R. W. Davis and G. S. Fargion, eds., Distribution and Abundance of Cetaceans in the North-central and Western Gulf of Mexico, U.S. Department of the Interior, Minerals Management Service, Gulf of Mexico OCS Region, Galveston, Texas, II, pp. 271-304. Perret, W. S., Bowman, P. E., and Savoie, L. B. , 1996. Bycatch in the shrimp fishery of Louisiana. In: B. Baxter and S. Keller, eds., Bycatch: considerations for today and tomorrow, Alaska Sea Grant College Program Report, University of Alaska, Fairbanks, 96-03, pp.137-142. Perry, H. , Brehm, W., Trigg, C , and Stuck, K. , 1995. Fishery-related morphometric characteristics of Menippe adina from the North-Central Gulf of Mexico. North Am. J. Fish. Manag. (15) 639-646. Perez, H . M . A. , and Kobelkowsky, A. , 1991. Estudio de los peces que habitan areas someras con vegetation sumergida de la laguna de Tampamacho, Veracruz. ,Informe Tecnico Interno 1990-1991, Consejo Divisional de Ciencias Biologicas y de la Salud, Universidad Autonoma Metropolitana-Iztapalapa. Perez-Rodriguez, R., 1980. Moluscos de la plataforma continental del Golfo de Mexico y Caribe mexicano. Ph D. thesis, Universidad National Autonoma de Mexico, Mexico, D.F. 338 pp. Pica-Granados, Y . , Ponce-Velez, M . G., and Barron-Echaury, M . E., 1991. Oceanografia Geologica (Golfo de Mexico y Mar Caribe Mexicano). In: G. De la Lanza Espino, ed. Oceanografia de Mares Mexicanos, A.G.T. Editor, S.A., Mexico, D.F., 569 pp. 138 Polovina, J. J., and Ow, M . D., 1983. ECOPATH: a user's manual and program listings. National Marine Fisheries Service, N O A A A , Honolulu, Adm. Rep., H-83-23, 46 pp. Polovina, J. J., 1984. Model of a coral reef ecosystem, the E C O P A T H model and its application to French Frigate Schoals. Coral reefs, 3: 1-11. Polovina, J. J., and Ralston, S., 1987. Tropical snapper and groupers: biology and fisheries management. Westview Press, Boulder and London. Polovina, J. J., 1993. The first ECOPATH. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc, 26, pp. vii-viii . Powers, L . W., 1977. Crabs (Brachyura) of the Gulf of Mexico. Contributions in Marine Science, 20:(215) 7-. Quiroga, B. C , Ulloa, R. P., Guzman, A . P., and Valdez, G. A . , unpublished. Informe preliminar sobre el programa de observadores en la flota arrastrera de escama que opera en el Banco de Campeche. INP Instituto Nacional de la Pesca, Mexico, D.F.. Documento interno (no publicado). Ramos-Miranda, J., Flores-Hernandez, D., and Sanchez-Gil, P., 1991. Pesca artesanal , Panorama Actual en el Estado de Campeche. Jaina, 2:(2) 20-21. Randall, J. E., 1967. Food habits of the reef fishes on the West Indies, pp. 654-847. Rebel, T. P., 1974. Sea Turtles and the turtle industry of the West Indies, Florida and the Gulf of Mexico. University Miami Press, Coral Gables, Florida, pp. 35-165. Ribic, C. A. , Davis, R., Hess, N . , and Peake, D., 1997. Distribution of seabirds in the northern Gulf of Mexico in relation to mesoscale features: initial observations. ICES Journal of Marine Science, 54: 545-551. Richards, W. J., and McGowan, M . F., 1989. Biological productivity in the Gulf of Mexico: Identifying the causes of variability in fisheries. In: K . Sherman and L . M . Alexander, eds., Biomass yields and geography of Large Marine Ecosystems, Westview Press, Inc., A A A S Selected Simposia series, Colorado, pp. 493. Rivera-Arriaga, E., and Villalobos-Zapata, G., 1998. The Mexico perspective. The Gulf of Mexico: Environmental issues. Gulf of Mexico Foundation, 1 Robins, C. R., Carleton, R. G., and Douglass, J., 1986. A field guide to Atlantic Coast Fishes. North America. Houghton Mifflin Company, Boston, New York, 354 pp. Rosado-Solorzano, R., and Guzman del Proo, S. A. , 1998. Preliminary trophic structure model for Tampamachoco lagoon, Veracruz, Mexico. Ecological Modelling, 109: 141-154. 139 Rosas, C , Lazaro-Chavez, E., and Buckle-Ramirez, F., 1994. Feeding habits and food niche segregation of Callinectes sapidus, C. rathbunae, and C. simUis in a subtropical coastal lagoon of the Gulf of Mexico. J. Crustacean Biol, 14:(2) 371-382. Rulifson, R. A. , 1981. Substrate preferences of juvenile penaeid shrimps in estuarine habitats. Contributions to Marine Science, 24: 35-52. Russ, G. R., Alcala, A . C , and Cabanban, A . S., 1992. Marine reserves and fisheries management on coral reefs with preliminary modelling of the effects on yield per recruit. Proceedings of hte Seveth International Coral Reef Symposium, Guam, pp. 978-985. Russel, R., 1989. Tuna management. Sport fishing, 4:(6) 40-44. Rutledge, R. W., Bacore, B. L. , and Mulholland, R. J., 1976. Ecological stability: an information theory viewpoint. J. Theor. Biol (57) 355-371. Sainsburg, J. C , 1971. Commercial fishing methods, an Introduction to vessels and gears. Whitefriars Press Ltd, London, 119 pp. Salas, S., 1984. Contribution al conocimiento biologico pesquero de la mojarra (Calamus pennatula) de la pesca de arrastre de Yucatan. B. Sc. thesis, ENEP-Iztacala, U N A M , Mexico. Salas, S., 1986. Analisis de la pesqueria de de la mojarra (Calamus pennatula) en las costas del Estado de Yucatan. M . Sc. thesis, CINVESTAV-LNP, Unidad Merida, Mexico. Saloman, C. H. , and Naughton, S. P., 1984. Food of crevalle jack (Caranx hippos) from Florida, Louisiana and Texas. NOAA-NMFS-SEFC,Tech. Mem., 134, 37 pp. Sanchez-Gil, P., Flores-Hernandez, D., Ramos-Miranda, J., Arreguin-Sanchez, F., and Sanchez-Chavez, J., 1993. Habitat, recursos pesqueros y estrategias de explotacion: hacia un diagnostico pesquero del Edo. de Campeche. Jaina, 4:(2) 2. Santoyo, H. , 1994. Fitoplancton y Productividad. In: G. De la Lanza Espino and C. M . Caceres, eds., Lagunas Costeras y el Litoral Mexicano, Universidad Autonoma de Baja California Sur, Mexico, D.F., pp. 221-245. Schmid, J. R., 1998. Marine turtle populations on the west-central coast of Florida: results of tagging studies at the Cedar Keys, Florida, 1986-1995. Fisheries Bulletin, 96:(1) 589-602. S E M A R N A P , 1997. Anuario estadistico de pesca. Secretaria National de la Pesca, Mexico, D.F.,http:/ SEPESCA-INP, 1994. Atlas pesquero de Mexico. Mexico, D.F., 234 pp. 140 Sheridan, P. F., J.A., B. , and Powers, J. E., 1984. Ecological interactions between penaeid shrimp and bottomfish assemblages. In: J. A . Gulland and B . J. Rothschild, eds., Penaeid shrimps- their biology and management, Fishing News Books Limited, Miami, Florida, pp. 235-258. Sherman, K. , and Alexander, L. M . , 1986. Variability and Management of Large Marine Ecosystems. A . S. S. 99, ed. Westview Press Inc. Washington, 319 pp. Sherman, K. , and Alexander, L. M . , 1989. Biomass Yields and Geography of Large Marine Ecosystems. A . S. S. 111, ed. Westview Press Inc.,, Washington, 493 pp. Sherman, K. , Alexander, L . M . , and Gold, B. D., 1990. Large Marine Ecosystems: Patterns, processes, and yields. A A A S , ed. American Association for the Advancement of Science, Washington, 242 pp. Sherman, K. , Alexander, L . M . , and Gold, B. D., 1991. Food Chains, Yields, Models, and Management of Large Marine Ecosystems. Library of Congress Cataloging in Publication Data, Colorado, 320 pp. Sherman, K. , Alexander, L . M . , and Gold, B . D., 1993. Large Marine Ecosystems: Stress, mitigation, and sustainability. A A A S , ed. American Association for the Advancement of Science, Washington, 376 pp. Sierra, L . M . , and Popova, O. A. , 1982. Particularidades de la alimentation del cibi (Caranx ruber Bloch) en la region suroccidental de la plataforma Cubana. Reporte de Investigation del Instituto Oceanologia, Academia de Ciencias de Cuba, 3, 1-19. Sierra, L. M . , 1983. Caracteristicas de la alimentation del jallao, Haemulon album, en la plataforma suroccidental de Cuba. Rev. Invest. Inst. Ocean. Acad. Cienc. Cuba, 11: 17 pp. Sierra, L. M . , and Diaz-Zaballa, J., 1984. Alimentation de dos especies de sardina, Harengula humeralis (Cuvier, 1829) y Harengula clupeola (Cuvier, 1829) en la costa N de la Ciudad de la Habana (rada del Instituto de Oceanologia). Reporte de Investigation del Instituto de Oceanologia, Academia de Ciencias de Cuba, 25: 1-18. Sierra, L . M . , Garcia-Cagide, A. , and Hernandez, A. , 1986. Aspectos de la biologia del cibi amarillo {Caranx bartholomaei, Cuvier, 1833) , en la region oriental del Golfo de Batabano, Cuba. Rev. Invest. Inst. Ocean. Acad. Cienc. Cuba, 55: 36. Sierra, L. M . , 1989. Dinamica del ritmo de la alimentation de varias especies de peces neriticos del Golfo de Batabano en diferentes anos. Rev. Invest. Inst. Ocean. Acad. Cienc. Cuba, 2, 1-14. 141 Snelson, F. F. J., Mulligan, T. J., and Williams, E., 1984. Food habits, occurrence, and population structure of the bull sharks, Carcharhinus leucas in Florida coastal lagoons. Bull.Mar.Sci., 34: 71-80. Solis, M . J., 1962. Contribution al estudio del pulpo (Octopus vulgaris, Lamarck) de la Sondade Campeche. Trab. Divulg., 3:(24) 1-36. Solis, M . J., 1967. Aspectos biologicos del pulpo (Octopus vulgaris, Lamarck) de la Sonda de Campeche. Trabajos de Divulgation, Inst. Nal. de la Pesca, Mexico, 3:(24) 1-30. Solis, M . J., and Chavez, E. A. , 1986. Evaluation y regimen optimo de pesca del pulpo en la Peninsula de Yucatan, Mexico. An. Inst. Cienc. del Mar y Limnol. UNAM, 13:(3) 1-18. Solis, M . J., 1988. E l recurso pulpo del Golfo de Mexico y el Caribe. In: LNP, ed. Los Recursos Pesqueros del Pais, INP, pp. 463-478. Solis, M . J., 1994. La Pesqueria del pulpo del Golfo de Mexico y Caribe Mexicano. In: S. Garcia del Real, ed. Atlas Pesquero y Pesquerias Relevantes de Mexico, LNP. CENEDIC, Colima. Solis, M . J., 1994b. Mollusca de la Peninsula de Yucatan, Mexico. In: A . Yafiez-Arancibia, ed. Recursos faunisticos del litoral de la Peninsula de Yucatan, Universidad Autonoma de Campeche, E P O M E X Serie Cientifica, Campeche, Mexico, 2, pp. 13-32. Solis, R. M . , 1995. Un caso de varamiento de Mesoplodon europaeus Gervais , 1855 (Mammalia: Cetacea) en Playas de Chelem, Y u c , Mexico. Jaina, 6:(3) 12-13. Soto, L . A. , and Escobar, E., 1995. Coupling mechanisms related to benthic production in the SW Gulf of Mexico. In: Proceedings of the European marine Biological Society Meeting, pp. 233-242. Sparre, P., 1991. Introduction to multispecies virtual population analysis. ICES Mar. Sci. Symp., 193: 12-21. Steele, J. H. , 1984. Kinds of variability and uncertainty affecting fisheries. In: M . R .M. , ed. Exploitation of marine communities, Springer-Verlag, Berlin, pp. 245-262. Steidinger, K . A . , 1973. Phytoplankton ecology: A conceptual review based on eastern Gulf of Mexico research. CRC Crit. Rev. Mirobiol, 3: 49-68. Stevenson, D. K. , 1981. A review of the marine resources of the Western Central Atlantic Fisheries Commission (WECAFC) region. Contrib. Prop. Fish. Biol. FAO, Firm.203 pp. 142 Stewart, D. J., Kitchell, J. F., and Crowder, L . B. , 1981. Forage fishes and their salmonid predators in Lake Michigan. Trans. Am. Fish. Soc, 110: 751-763. Stillwell, C. E., and Kohler, N . E., 1982. Food, Feeding Habits, and Estimates of Daily Ration of the Shortfin mako (Isurus oxyrinchus) in the Northwest Atlantic. Can. J. Fish. Aquat. Sci., 39: 407-414. Stoner, A . W., 1983. Distribution ecology of amphipods and tanaidaceans associated with three seagrass species. J. Crust. Biol, 3: 305-518. Suarez, E., 1994. Comunidades zooplanctonicas de las lagunas costeras. In: G. De la Lanza Espino and C. M . Caceres, eds., Lagunas Costeras y el Litoral Mexicano, Universidad Autonoma de Baja California Sur, Mexico, D.F., pp. 248-268. Toft, C. A. , and Mangel, M . , 1991. Discussion: From individuals to ecosystems; the papers of Skellam, Lindeman and Hutchinson. Bull. Mathematical Biol, 53:(1) 121-134. Torres, C , 1987. Sistematica y distribution del Genero Chaetoceros, Ehreberg (Bacillariophycea), en la Laguna de Terminos, Campeche, Mexico. B. Sc. thesis (Biologo), Universidad Veracruzana, Veracruz. Torres, R., and Chavez, E., 1987. Evaluation y diagnostico de la pesqueria de rubia (Lutjanus synagris L.) en el estado de Yucatan. Cienc. Mar. Mex., 13:(1) 7-29. Torres Lara, R., and Salas Marquez, S., 1990. Crecimiento y mortalidad de la rubia Lutjanus synagris de las costas de Yucatan durante las temporadas de pesca 1983-1985. Ann. Inst. Ciencias del Mar y Limnol. UN AM, 17:(2) 205-214. Tovilla-Hernandez, C , 1994. Manglares. In: G. De la Lanza Espino and C. Caceres-Martinez, eds., Lagunas costeras y el litoral mexicano, U A B C S , Mexico, pp. 372-423. Trites, A . , Christensen, V. , and Pauly, D., 1997. Competition between fisheries and marine mammals for prey and primary production in the Pacific Ocean. J. Northw. Atl. Fish. Sci., 22: 173-187. Turner, R. E., Kaswadji, R., Rabalais, N . N . , and Boesch, D. F., 1986. Long-term changes in the Mississippi River quality and its relationship to hypoxic continental shelf waters. In: M . P. Lynch and K . L. McDonald, eds., Proceedings of the 10th National Conference on Estuarine and Coastal Management, Tools of the Trade, New Orleans, L A , pp. 261-266. Ulanowicz, R. E., 1986. Growth and Development: Ecosystem Phenomenology. Springer-Verlag, New York, 203 pp. 143 University of Georgia, unpublished. R/V Georgia bulldog. Various B R D Ddesign Testing conducted off Cape Canaveral, FI. The University of Georgia, Marine Extension Service. 715 Bay Street, Brunswick, GA. 31523; Gulf & Soth Atlantic Fisheries Development Foundation, Brunswick, G A , N O A A Award No. NA17FD-0103-01. Upton, H . F., Hoar, P., and Upton, M . , 1992. The Gulf of Mexico shrimp fishery: Profile of a Valuable National Resource. Center for Marine Conservation, Washington, 113 pp. US Department of Commerce, Administration, N . O. a. A. , Service, N . O., Management, O. o. O. a. C. R., and Division, S. a. R., 1996. Florida National Marine Sanctuary. Final Management Plan/Environmental Impact Statement. U.S. Department of Commerce, N O A A . U S D O C , 1989. Report of the NMFS swordfish stock assessment workshop March 20-24, 1989. U.S. Department of Commerce, NO A A A , NMFS, Miami, Florida. v. Brandt, A. , 1972. Revised and enlarged fish catching methods of the world. Eyre & Spottiswoode Ltd, Thanet Press, London, England, 240 pp. Valero-Pacheco, E., 1989. Contribution al conocimiento del camaron rosado Penaeus duorarum en la Sonda de Campeche durante 1975-1976. B . Sc. thesis, ENEP-Iztacala, U N A M , Mexico. 31 pp. Vargo, G. A . , and Hopkins, T. L. , 1990. Plankton. In: Continental Shelf Associates Inc., ed. Synthesis of available biological, geological ,chemical, socieconomic, and cultural resource information for the South Florida Area, M M S U.S. Department of the Interior. Minerals Management Service Atlantic OCS Region, pp. 195-230. Vazquez-Bader, A . R., and Gratia, A. , 1994. Macroinvertebrados benticos de la Plataforma Continental del suroeste del Golfo de Mexico. Publicaciones Especiales 12, Mexico, D.F., 113 pp. Vega, M . E., and Hernandez, M . , 1987. Aspectos trofodinamicos de los peces dominantes de la Ria de Celestun , Yucatan, Mexico. In: I Congreso de Ciencias del Mar, 9-12 junio, La Habana, Cuba. Vega, M . E., 1990. Interaction trofodinamica entre los bagres Arius melanopus (Agassiz 1829) y Arius felis (Linneaus 1766), en las costas de Celestun Yucatan, Mexico. Ann. Cienc. del mar. y Limnol, UNAM, 17:(2) 271-275. Vega-Cendejas, M . E., Arreguin-Sanchez, F., and Hernandez, M . , 1993. Trophic fluxes on the Campeche Bank, Mexico. In: V . Christensen and D. Pauly, eds., Trophic models of aquatic ecosystems, I C L A R M Conf. Proc. 26, pp. 206-213. 144 Venier, J. M . , 1997. Seasonal ecosystem models of the Looe Key National Marine Sanctuary, Florida. M.Sc. thesis, The University of British Columbia, Vancouver, B.C. Canada. Venier, J. M . , and Pauly, D., 1997. Trophic dynamics of a Florida Key coral reef ecosystem. Proc 8th Int Coral Reef Sym, 1: 915-920. Walsh, J. J., 1988. On the nature of Continental Shelves. Academic Press Inc. U.S.A. Walters, C. J., 1996. Suggested improvements for ecopath modelling. In: D. Pauly and V . Christensen, eds., Mass-balance models of North-Eastern Pacific ecosystems, Fisheries Centre Research Report, 4 (1). Walters, C. J., Christensen, V. , and Pauly, D., 1997. Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Rev. Fish. Biol. Fish., 7: 139-172. Walters, C. J., Pauly, D., and Christensen, V . , 1999. Ecospace: prediction of mesoscale spatial patterns in trophic relationships of exploited ecosystems, with emphasis on the impacts of marine protected areas. Ecosystems 2:539-554. Waring, G. T., Palka, D. L. , Mullin, K . D., hain, J. FL, Hansen, L . J., and Bisack, K . D., 1997. U.S. Atlantic and Gulf of Mexico Marine Mammal stock Assessments-1996. U.S. Department of commerce, N O A A , NMFS, Massachusetts, N O A A Technical memorandum, NMFS-NE-114. Wassenberg, T. J., and Hi l l , B . J., 1990. Partitioning of material discarded from prawn trawlers in Moreton Bay. Aust. J. Mar. Freshwater Res., 41: 27-36. Watling, L. , and Norse, E. A. , 1998. Disturbance of the seabed by Mobile Fishing Gear: A comparison to forest clearcutting. Conservation Biology, 12:(6) 1180-1197. W C M C , 1998. Marine protected areas of Mexico. Wetherbee, B. M . , Gruber, S. H. , and Cortes, E., 1990. Diet, Feeding Habits, Digestion, and Consumption in sharks, with special reference to the lemon shark, Negaprion brevirostris. Elasmobranchs as living resources: Advances in the Biology, Ecology, Systematics and Status of the fisheries. 90: 29-47. Wiseman, W. J. J., 1997. Modelling the physics of the Gulf of Mexico: A large marine ecosystem. In: Q. R. Dokken, G. Meier, and H . Kumpf, eds., Models and mass-balance calculations for the Gulf fo Mexico, Report of workshop, April 2-4, Metairie, Louisiana, pp. 67. 145 Witham, R., Ingle, R. M . , and Joyce, A. , 1968. Physiological and ecological studies of Panulirus argus from the St. Lucie estuary. Fla. Bd. Conserv. Mar. Lab Tech Ser., 53. 31pp. Yafiez-Arancibia, A. , Sanchez-Gil, P., Villalobos, Z. G., and Rodriguez, C. R., 1985. Distribution y abundancia de peces demersales de la plataforma continental mexicana del Golfo de Mexico. In: A . Yanez-Arancibia, ed. Recursos Pesqueros Potenciales de Mexico: La pesca acompanante del camaron, Instituto de Ciencias del Mar y Limnologia, U N A M , Instituto National de la Pesca, Mexico, D.F., pp. 315-398. Yanez-Arancibia, A . , Lara-Dominguez, A. , Sanchez-Gil, P., Vargas, M . I., Garcia, A . M . , Alvarez-Guillen, H. , Tapia, G. M . , Flores-Hernandez, D., and Amezcua, L . F., 1985b. Ecology and Evaluation of fish community in coastal systems: Estuary-shelf interrelationships in the southern Gulf of Mexico. In: A . Yanez-Arancibia, ed. Fish Community Ecology in estuaries and coastal lagoons: Towards an Ecosystem Integration, U N A M Press, Mexico, pp. 475-498. Yanez-Arancibia, A . , and Day, J. W. J., 1988. Ecologia de los ecosistemas costeros en el sur del Golfo de Mexico. La region de la Laguna de Terminos. U N A M , Ed. Universitaria, Mexico, 525 pp. Yanez-Arancibia, A . , Rojas, J. G., Sanchez-Gil, P., and Lara-Dominguez, A. , 1997. The Gulf of Mexico. Strategic Priority for Sustainable Development. Rev. Mex. Caribe (4) 130-147. Yanez-Arancibia, A. , and Sanchez-Gil, P., 1986. The demersal fishes of the southern Gulf of Mexico shelf. 1. Environmental characterization, ecology and evaluation of species, population and communities. Rev. Inst. Cienc. del Mar y Limnol. UNAM, Special publication 9 Yanez-Arancibia, A. , and Lara-Dominguez, A . L. , 1988. Ecology of three sea catfishes (Ariidae) in a tropical coastal ecosystem- southern Gulf of Mexico. Marine Ecology Prog. Ser., 49: 215-230. Yanez-Arancibia, A. , Aguirre Leon, A. , and Soberon Chavez, G., 1988b. Estuarine-related fisheries in Terminos Lagoon and adjacent continental shelf (Southern Gulf of Mexico). In: E. Maltby, P. J. Dugan, and J. C. Lefeuvre, eds., Conservation and Development the sustainable use of wetland resources. Proceeding of the Third International Wetlands Conference. The IUCN Wetlands Programme, pp. 145-154. Zenkevitch, L . A. , Filatova, A. , Belyaev, G. M . , Lukyanova, T. S., and Suetova, I. A. , 1971. Quantitative distribution of zoobenthos in the world ocean. Bull, der Moskauer Gen. derNaturforscher,Abt. Biol., 16: 27-33. 146 Zetina Rejon, M . J., 1997. Patrones troficos de utilization espacio-temporal de los habitats criticos de pastos marinos y raices de manglar por tres especies de peces dominantes (Harengula jaguana, Poey 1985; Eucinostomus gula, Cuvier in C & V 1830; y Archosargus rhomboidalis, Linnaeus 1758) en la Laguna de Terminos, Campeche, Mexico. Univ. Auton. Campeche, Campeche. 64 pp. 147 Appendix 1. Estuaries of the Gulf of Mexico (NOAA, 1990; Ibarra-Obando etai. 1997) Estuaries on the US (31,080 km2) Estuaries in Mexico (6,800 km2) Florida Bay Laguna Madre South Ten Thousand Islands Laguna Morales North Ten Thousand Islands Laguna San Andres Rookery Bay Laguna Chijol Charlotte Harbor Laguna Puerto Viejo Caloosahatchee River Laguna de Tamiahua Sarasota Bay Laguna Tampamachoco Tampa Bay Laguna Grande Suwanee River Laguna Verde Apalachee Bay Laguna de Mandinga Apalachicola Bay Laguna Camaronero St. Andrew Bay Laguna Tlalixcoyan Choctawhatchee Bay Laguna de Alvarado Pensacola Bay Laguna de Santecomapan Perdido Bay Laguna del Ostion Mobile Bay Laguna del Carmen Mississippi Sound Laguna Machona Lake Borgne Laguna Tupilco Lake Pontchartrain Laguna Mecoapan Breton/Chandeleur Sounds Estero de Chiltepec Mississippi River Laguna Atasta Barataria Bay Laguna de Terminos Terrebonne/Timbalier Bays Laguna Sabancuy Atchafalaya/Vermilion Bays Estero Yalton Calcasieu Lake Laguna Celestun Sabine Lake Estero Progreso Galveston Bay Estero de Telchac Brazos River Estero de Punta Arenas Matagorda Bay Estero El Islote San Antonio Bay Estero Lagartos Aransas Bay Laguna Yalahua Corpus Christi Bay Upper Laguna Madre Lower Laguna Madre Total (km2) 37,800 148 5 r o ON ON <U N CD > C3 43 u £3 r o ON Os <u N CD > 6 r o ON o CU CU N N I Si 00 • i -00 r o r o CD ON ON C ON ON ro ON ON N > CS 43 U 5 ON ON cn O CD x> o fl CD OH ii °N ^ ON fl cd 00 ro "3 JU ON X fc ON " o I K ? oo 00 ON CU ~ fl ca a cd NO CJ ON r o < « ON O ON « a fl o CO «J 03 on 5 3 & CD > ~J i l « r o a 2? ~ ON CU w ca is "S •s N G< o O CQ N _ CD CL, 43 ^  o s 00 N * CD i -? ON 33 CS 00 r o u CD ON o C ON -o 2 .9 5 fl oo 73 oo ON > e ^ < N CL> > 0O ON n ON OO ^ ON ' fl C3 fl O N C3 CJ CD S CD 43 K.I 43 o H o c fl cd cd O w N CD 3 2J oo ON .3 w 3 - ~ OO C3 NO NO ON ON N CJ > C3 43 fl fl - . N ID 43 o fl ca 00 N CD « 2 cd ON e 2 CD - -fl c ca CU N <D .3 So g C 43 ca -< O U NO OO ON N CD > cd 43 u fl fl ca o oo C3 & © IM 00 00 ON cu CN ON UO ON ' ON « 3 cu o W fl fl ca o "S W oo fl CD >o ca oo oo ON . oo ^ ON cs r-1 cu a ca in 42 O o a w ca W CQ uo ON ON t— OO "—' ON H »—i ca w 42 . O O 2 w o ta H ° ca o CQ oo oo ON «5 C3 CQ uo ON ON 3 oo 42 ON O O ' co A •»« fl ca - c o o M ffl oo ON C3 •s fl o cd | ca N =a y 00 I ^ . 3 42 3 o\ 00 OO <D ON t O < .2 ® a ON NO O ON ON ON N CD ca 43 u fl C ca 43 o u N CD fl C ca CD 3^ ca 3 . oo a e l ^ 2 NO oo N CD > ca 43 u o 00 ON o u CU I 43 3 O cu fl o S o CU a cu s I ca 42 O o crt W 3« fcl oo ON cu •c ta CQ ON ON CU N CD > cd 43 u N CD 43 CJ fl cd 00 i .3 A oo CD r o ON ON NO ON N CD fl ca CD ON cd cj CJ ON o O cd CD 0_ CD r o ON 00 ON ON £> id O CL, '—^ CD rO r£ ON ^ ON fl fl w ed « C fl W cu N CD c^ 43 u cd r o ON ON N J4; 43 2 u o N CD > ca 43 u o oo CN ON 00 ON w ^ o **^» . El a o 3 cu cd Cd ^ fl X s >^ CQ 2 NO ON o oo cu u e CU CJ CU fl C cu a. a < •a o u a o CU ca e ca cu la O e o 4. fl ° 2 •2 1> 'S CD 2 C3 •s § s s "S. cu e CD fl o .3 „ "S '3j w w e o c Cd Q. «" 1 5 N ^ o fl c ca . cu 1 « u c 33 o oo a e o fl a ca cu o ON s •a ca fl c ca '3 CU > 3 n. 43 &0 fl o D. ca u cu fl s & o o 5 ca > T3 3 °s 2 ON N CJ O fl ea 00 ca o .a ^ 3 ca a ON - 3 1 t l ^ 3 < • - H C3 OH 60 o cj C > .g ro ON •c o C3 « ro ON ON CU ca oo a\ oo ca ro Os Os T3 fl CJ u I ca 60 CJ > N CJ • f l o C ca oo i .3 A ^ 60 ro CJ as tl ON « CU ca ca N N N CJ CJ CJ • f l • f l • f l CJ o CJ fl C fl C3 ca ca 00 i oo • 00 .3 ^ .3 ^ v 5 ^-fl x> A ca ' f l -O 60 CO 60 ro 60 ro CJ OS CJ Os JU ON t l ON C Os tl OS <; c < < C cj > 2 T3 fl ca N C J 1 ca fl ro cj Os I- ON < _ o ca o < PH NO ON ON CU C3 (H O U • o fl ca o cu M _ C C3 <D 60 H g< . ~ i ^ C3 ro o os a co CU ON ON N i—H CJ w > C3 a •+-* o r~ oo ON CQ w N CJ fl x> - f l oo t i ON C3 w ro ON ON CU en ca T3 2? CJ d-a a cj ca > o N CJ • f l u CO oo ON N -H CJ w cj ca C ' o ca (-j 3 ^ ^ CO CO rH u ON oo H ON ON NO o ON N V JS o c ca oo • .3 A N CJ • f l o fl fl '3 x> 60 co 60 ro CJ ON cj ON t| ON tl ON oo ON CU N CJ > ca XI U M > ° oo oo ON ON ^ r*. r*. ° J cu C o CJ hJ I .5 A 60 < C cj S '3 •o ca < > HO i n oo ON CU H 00 • i ^ fl £ j ! 'fl SS ca os « 3 Hfl o o CN oo ON ca ON 0 o ON ON C J PH C3 < c ca cj C ca o 1 PH oo ON "ca •8 N N ca ro oo ON NO n » « . •a* gi ca ^ B _. j f l ca S f i do > < oo oo oo ON ON *• C cu o o a 3 Q _ca ca U PH ^ o oo ^ Os _ H 00 5 °^  ca a o H 60 HO > -i3 > ea "t; ca nfl ca J 3 o m u NO ON ro oo ON c CJ bo ca ro ON ON cu N CJ • f l o c ca 00 3 00 CJ c < CJ Q oo ON CJ a o ea >-H PH 00 en ON ca o ^ T3 C J fl C ea ea c 3 _ CQ -2 T 3 o S fl 8 T3 B cu a I <» e o B "a H ^ ^" o u > o •fl cj C < t l cj §1 (A 1-<u E cu •a x u ii _ -a « ,** S OJ w JS ca u _ H O U .5 ca s ca T3 fl -fl -a .3 c « "2 3 fl CJ ca > « .2, es i. _ J3 ca cu s« .—. 1- 4> ea «J -O 3 U .5 <Si ea oo i—i i r-CJ ea X( 1 CJ ea C J ea T J >. cS • f l o ea PH ea o tH u N N CJ O 43 43 o o a fl ca cd OO 00 .9 ^ .3 — 3 40 fl ca 60 ro 60 ro <U ON <U ON t ON ti ON <: o < O fl fl 42 •fl -fl ro 3 fl ON 60 60 ON ca <u i2 ti t -a S fl S 3 55 C3 ON OS 3^ ~j M N N <U S 43 .S | ca o o > > fl c o <J oo oo CQ ro ON ON CJ - f l fl CJ U i ca 60 > ^ ^ ^ fl >^  uo o fi oo oo ON w 0O ON ON ON ON N u 43 o c ca 00 fl o Pi fl oo ON fl 4 2 60 ro ca ON fcl <=N < C 8 g S •s "3 1 , 3 3 A -fi 3 ^ '> 3 -a eo B o 3 K e t 13 O « fc O (2 o 2 < H o o O N ON ^ ON ON oo W W ON « "3 J c/1 "3 od cu cu 2. fl 2 ^a -g O O o u a CL, V 3 ca E o fi NO S 0 0 42 ON O w C CO ca ca t-i i—< ^ 00 ca ON fl ^ >^  fl o o 1 3 ON ON ON ON ca —< —1 oo ON N ' ca 8 55 « '3 JS 42 ca CO N 42 » ca /K oo O ca N o .g •3 o •> £ 3 -9 .a M « fc 2 <: <u N ca 43 o 00 g 'fl 1? oo r-ca oo fc ON < w r-ON ON fl o Di ca ca N i-H O 00 w 55 « fl 1 § ca .9 > t> ~3 O0 ON ON co ON ca * O ^ fl fl ca ca ^ ON u o c o r-OO ON fl o 3 Pi fl ca .9 > o PH ON ro NO 0O O S ON CJ 8 o ca fl cu s a C o u o fl cu a a. < u on CJ ca fl 'C ca CX 00 « fl i . ca cu S en cu — fl •-cu > cu 3 2 > ^ 2 S a U u ca e ca ca fl '3 ca 3 ca ca fl '3 ca fc ca 00 ca u ca V cu cu ca c M E « a C/J a. 2 CU fl cu ca o u fl cu fN fN ca 43 00 E5 Or -9 PH ca E o | 5 1:1 C U 3 .51 e o u •3 c cu Q. D. < o a cc 8? 3 1 s « S 2 2 cu c L U ^ « & s IT) «s TT CM N C J 43 o fl cd 00 I .3 _ 3 ca 60 r o C J O N tl O N •< o fl N •3 cj fl > 60 c5 C J 43 N O 00 . : as r~ H N OO ^ 2 O N C 3 fl, r H IH cy - - u N b C3 -9 c3 rH U / - ^ C J 43 r H * g 2 oo C -•a ca .3 3 60 cj t < _ N N 2 S 2 ~ .a s •8 O > oo cj JA a ca ca T3 §« cj U ON ON • i—I r o O N O N O 2 |S o fl ca N r " O §0 ca ON ca r o O N O N cu N C J 43 o G ca oo 3 60 C J < JS r-C N &"a ca r o O N O N T3 c C J u I ca 60 C J > r o O N O N C J T3 O (D O I ca 60 cj > fl ca N on fl oo a ON C N O N O N C J fl C J M o U ca cj ca T 3 C O T3 O ca J3 t/1 N cj 43 cj 00 .5 3 60 C J < OO O N > ca 43 u C N C3 ON 3^ N C J > ca 43 u ca > o o CH c cj 00 M f - as SO SO OS 2 2 ^ rS C C fl t! Pi oo oo ON ^ r\, NO oo C ON 2 r f l ^ 60 • 2 £ ca - H ca fl I-c «» ca S P O oo 73 2 oo C o ca TS cu v. O. w 'SD "3 a r o r o O N O N O N O N r H V ' C3 C3 cu cu >s >, 3 3 C3 C3 PH PH O N N O O N T3 o 13 o £ oo W cu Q ON to C N Appendix 3. Seabirds species in the GoM (Duncan et al. 1980; Peake 1996) Species Common name Family Habitat Calonectris diomedea Cory's Shearwater 1 cs,o Pterodroma spp. Unidentified gadfly-petrel 1 o Puffinus Iherminieri Audubon's Shearwater 1 cs,o Oceanites oceanicus Wilson's Storm-petrel 2 0 Oceanodroma castro Band-rumped Storm Petrel 2 0 Oceanodroma leucorhoa Leach's Storm Petrel 2 0 Phaeton aethereus Red-billed Tropicbird 3 0 Phaeton lepturus White-tailed Tropicbird 3 0 Pelecanus erythrorhynchos W ite Pelican 4 c,cs Pelecanus occidentalis Brown Pelican 4 c,cs Sula bassanus Northern Gannet 5 CS,0 Sula dactylatra Masked Booby 5 CS,0 Sula leucogaster Brown Booby 5 CS,0 Sula sula Red-footed booby 5 Phalacrocorax auritus Double-crested cormorant 6 Phalacrocorax olivaceus Olivaceous cormorant 6 Phalacrocorax sp. Unidentified Cormorant 6 o Fregata magnificens Magnificent Frigatebird 7 c,cs,o Phaloropus fulicaria Red Phalarope 8 0 Phaloropus tricolor Wilson's Phalarope 8 0 Anous stolidus Brown Noddy 9 0 Anous tenuirostris White Capped Tern 9 CS,0 Chilidonias niger Black Tern 9 CS,0 Larus argentatus Herring Gull 9 c,cs,o Larus atricilla Laughing Gull 9 cs,o Larus delawerensis Ring-billed Gull 9 CS,0 Larus dominicanus Dominican Gull 9 cs,o Larus marinus Great black-beacked Gull 9 cs,o Larus Philadelphia Bonaparte Gull 9 o Larus pipixcan Franklin's Gull 9 o Larus sabini Sabine's gull 9 Rynchops niger Black Skimmer 9 0 Stercorarius parasiticus Parasitic Jaeger 9 0 Stercorarius pomarinus Pomaric Jaeger 9 cs,o Sterna albifrons Little Tern 9 CS,0 Sterna anaethetus Bridled Tern 9 cs,o Sterna antillarum Least Tern 9 CS,0 Sterna caspia Caspian Tern 9 CS,0 Sterna dougallii Roseate Tern 9 CS,0 Sterna forsteri Forster's Tern 9 0 Sterna fuscata Sooty Tern 9 CS,0 Sterna hirundo Common Tern 9 CS,0 Sterna maxima Royal Tern 9 C,CS,0 Sterna nilotica Gull-billed Tern 9 0 Sterna sandvicensis Sandwich Tern 9 CS,0 Xema sabini Sabine Gull 9 CS,0 0= oceanic, CS= Continental shelf, C=coastal distribuition Families 1 Procellariidae 6 Phalacrocoracidae 2 Hydrobatidae 7 Fragatidae 3 Phaethontidae 8 Scolopacidae 4 Pelecanidae 9 Laridae 5 Sulidae N O O N O N NU M s u 05 o u cu CU JS OS £ £ en s "C CS s a a < 5-c ° a J * 1 a w eg .a c« C/J £3 a. Q. Q IO CN ON ^H CN O O ON ro © CN ro IO ro NO t i—H ON ON NO ro NO WN C~- CN -^ ON ON r - o 1—1 NO ro oo o CN ro ro ro ON i n o CN ON T f NO 00 0\ r - ro ro ON ON T f CN T t ro CN CN CN ON ro CN 00 rs NO NO ON ro ro NO 00 m ^ H ro NO IT) —' ON r— ro CN r - m N O ~ o CN o C N ON l O — T t — jo 5 CN _ ON J - CN <t n m oo O N O N N O ro ~ O N _ ^ i l ' Tt — — OO N O N O 22 O N TJ-^ 2 K £ 2 ON — ON ON CO m CN T f ON OO 00 T t ro O N 2 C T \ ° 2 - 2 ' * " * ' " C 3 N O O ^ ! ro r-ro O CO OO oo ON i—i T f o ro CN m oo T f ro m NO NO o CN 22 in ro ro OO OO ro CO NO ON in o <3N ON r - NO in 00 o o o I O o ON NO I O o\ I O in in in r-~ in ro ro NO ro OO ON in CN CO ro ON ro oo oo NO ro NO CN O ro ro ro ro ro ro — ro ON ON I O ON CN CN T t CN CN oo NO >n in T t T t CO CN ro 0O ro T t CN CN CN T t NO CN T t T t ro ' S 2 CJ S s s s s o o o o •o c J= B K to £ 1 , 1 E E -22 22 I o o o ° o © o O "CJ "CJ " O "O E E E 6 .5 E E E E o o o o o o o o CN CN CN CN A A A A ttt VI VI 1/1 fl> »- I- »- (-_2 u u u u w td cd cd & & & & 8- & & 8-QJ <U <D <D 4-J V V V « (N N (N CN O T3 T3 " d "O * * * * # # E E E E E E E E E E E E E E E E E E o © CN o © CN o o CN o © CN o © CN o © CN o © CN o © CN o o CN © o CN © o CN © o CN © © CN © © © © CN CN 200 200 30 m © © CN A A A A A A A A A A A A A A A A A CN A O O O O O O O O O O o O O s s S 2 S s ,g .g .g .g .g J-J a. J 3 a. J -« _a JS £ • JS o. o •o o •o o T3 o -a o •a CJ CJ CJ CJ CJ o o o o o CJ CJ CH CJ CJ 5 tt tt tt m CQ CQ CQ CQ 43 CJ E ra JS JS "3 J 3 e o CJ "3 43 CJ I* J2 e CQ E •S - a cj C O CQ "3 43 > ed 3 -B o c g u a 5 B B e e e | ft, a . •S a •S ft, "11 cs es "a S a S» =5 c3 .a g. « Si w £ P s s ^ 1 1 O 55 a, M is cj •a 3 a * E C C3 w "3 ^ § ~ CO a a N ? •« a ** 5r § ^S, O S, a 2 * ^ I I tt, o, Co £ 3 § 8 -8 -3 -S -S § .S "c3 "3 "3 § *S a c c B So JS" d) CD CO <U ^ CO £3 C O C O Oo Q §• _ .a "3 "3 -sT o 2 S co co CO - > B C3 <U Co ^ C3 v^ j ^ )^ Q -S a o 3 Cj O E B Co Co -S JS a a 09 OQ B C L CO 3" .5 so a NS «5 • a CJ • a JO E o •tt cn ed r - l CJ w X C CJ O r" ed E o E § % » § 3>5 j - S T3 0 « 2 | | | •S « a jo CJ -a £ -a -5 H ~ s CJ r- +H 1 ^ I 8 rt 2 o * J ? CO a n o 43 CJ J<« C © © . . CN CJ cd E o E IS o S 1 o S =5 C J cj C3 - ~ " g o fl W C " ca ca 0 >» a Os 43 00 1 rv l—i O o © r*, fl ~ 2 <o f3 CN co £ 2 ^ -fl oo sp SS ca - 9 cu oo fl bo fl o OH CJ O •fl to S 2 42 5 . bO co C U 43 A3 CJ 3 t3 fl ca a ^ co I 33 CO C3 I % J g s t ; 8 .g , 5 •w ca § „ co" 42 to ca -C! cj C3 13L co 3 •§ ^ p-O •S- ao W CJ CU CO -5 -2 = S3 O fl _ I f fl OS ca co O •fl « -s 3 § 2 c O c .fl N O 4= p-s ca fl o CJ 3 • ° o ca O cj ^ CO CO CJ ta CO CO Cf l fl -a [3 * ' fl 11 & o o ca fl- fl CJ co fl co 1 3 1 3 cfl /—s CO ^ cj CJ 33 rri -t-* CJ 3 fl O 4 3 p. ca CJ 2 3 4 3 C co CJ tfl "co 43 P ] ca 143 I-3 1 CU 3 O O 4 3 > 3 4 2 "2 H. ca S fl 4 - cj 4 3 3 O »-* o > 3 O o > 3 O O > CO CO co —3 CO • a ca CJ fl fl s s ca CO E3 * ° o co Z 2 « 2. =a 00 00 i2 "3 I P ca ca — ca fl fl 4 3 fl fl * -a w S fl ca O co 00 <*3 00 o 42 2. to iS1 1- CJ i - fc ca 3 £ u 00 c P. - C cj ca cj cj fl 42 3 co ca co "2 ca ca C cj P. 0 1 0 - co _ ! 8 ca =a '-fl tS Si & S E 42 00 42 o. fl o V co fl , . fl .2 4 ? O « O. M O » co ca M fl £ > £ ca - 0 0 3* c+2 j - . CJ 3 3 CJ so ° S P O 42 fl CJ 42 CJ CJ -0 p. = 1.2 l l 1-,CJ ,CJ . J C CJ CJ 2 CJ CJ CJ CJ ja fc fc g < < < P > o „ < co _ ca o CO CJ ca fl. CJ 43 CJ 3 & ca p E a ca u >< >s ca cj ca —-H U <C OJ « fl 3 ca 00 1- c y a CJ 43 o flj 3 a fl o CJ CJ 3 O CO CO cj ca ca ca cj ca . CJ H H > o CJ co ca 4 2 ca H ^ cj ca 43 w fl 3 5 O C3 hJ u CJ "S § <L> O B-C ca ca c U J? c '3 N -2 S S 3 .2- fti 3 o 00 < g ca ca x y cj H ^ a H z S o u z u 00 3 c CJ CJ co -§ >s s cu 5 CJ C3 'c o cj 00 00 o 1 3 0 ca T3 CJ 43 u T3 'C I cu -Ss cu -St cj •§ I" CJ ca rs 'c o U Ji fl 'C CO '!• CJ C3 cu cj s cu -SC CJ •§ I-^1 CJ ca fl c _o 4 3 O 4>d I 43 cu •« cj o 5 ca fl 4 3 U ca •e CJ 43 C3 CU cj .a o cj iS cu -s: CJ C2 0 0 ON ON u ON ON c» CU -*-> CU a K o o "S cu s o cu en "3 cu C A St a o S a o w -fc" t/3 O ON ON a o ca c cu On a X X X X X X X X X X X X X X X X X X X X X X X X X X X •a Cd 43 co _, ao ta <D ed X M co w «-» 00 a u > CO -J3 tD ed O a & i 1 3 fl cd 43 co 3"-a cd a o 43 CO O a> cd to •a cd 43 CO •3 Tl 4 H 0 co co 40 o 4  14 N ^ w aj 43 S i3 " to 00 « "2 a S ^ 2 5 s a s a ca -5 s CJ R &3 &3 5r ."a ^ s I cu 5 -o a a ° C3 J •§ "3 a s s &, t> 3 s i? C-3 o a C3 R R S_ « . " 3 cs «a to to a a to a 5 a1 <S,»S> a 3 § a oo w C3 <3 CD cd fl cd >5 -2 O ^ ^ -R »R g ^ ^ r 5 3 r Q r * 3 o u o •5 -St CJ a a •S -5 -3 -3 g g a a a •S -3 -5 -5 -3 -3 l . iv Sv « CJ C3 a g g r Q r * 3 r " 3 U U U 3 cu cu S cu -3 a, n , t-i t-i a a s R C3 C3 »R >R CJ CJ 3 to g o a , c-] 3 R C3 <3 -R -R CJ CJ CU C3 O R CU e C3 R >» sb -X 5 a 3 u l | cu CJ o cu to s R •§ 55 R O a a a a ts o o o o o R C5 ^ I S C5 g 3 ^ ?S Cl, O, C L fl cl • i-H • i—I i—-I +-» cd cd 00 00 b :y 2 5 S >» O J-t o oo c •S I O 2 cd fl taspii idae lidae idae don Alopi cd riak O Alopi h-1 H CO cd fl is •a cd 43 cj I-I cd U u cd fl 43 00 Appendix 7. Estimated Ecopath mortality rates of all group species in the Gulf of Mexico synthetic model Group species Mortality rates (year1) z . ) F b ) M2C) M0 d ) 1 Benthic producers 10.24 0 8.192 2.048 2 Phytoplk. Inshore 104.64 0 94.18 10.46 3 Phytoplk. offshore 88.13 0 68.42 19.71 4 Meiobenthos 10.00 0 9.00 1.00 5 Macroinfauna 6.49 0 5.23 1.26 6 Macreoepifauna 1.70 0.001 0.68 1.02 7 Herb.zooplk. inshore 22.10 0 17.86 4.24 8 Herb.zooplk. offshore 22.12 0 19.91 2.21 9 Carn.zooplk. inshore 15.00 0 13.49 1.52 10 Carn.zooplk. offshore 5.00 0 3.78 1.22 11 Juvenile shrimp 4,00 0.032 2.23 1.74 12 Adult shrimp 3.50 0.016 0.59 2.89 13 Other decapods 5.00 0.019 2.33 2.65 14 Octopus 1.12 0.003 1.08 0.03 15 Clupeids 1.70 0.088 1.15 0.46 16 Catfishes 1.02 0.359 0.42 0.24 17 Juvenile grunts 0.34 0.005 0.29 0.05 18 Adults grunts 1.61 0.006 0.84 0.76 19 Juvenile groupers 0.56 0.055 0.45 0.05 20 Adult groupers 0.43 0.075 0.28 0.07 21 Mullets 0.72 0.013 0.63 0.08 22 Coastal sharks 0.55 0.067 0.39 0.09 23 Oceanic sharks 0.39 0.178 0.19 0.03 24 Juvenile tuna 0.52 0.033 0.30 0.18 25 Adult tuna 0.51 0.128 0.33 0.05 26 Soldierfishes 1.40 0 0.25 1.15 27 Other fishes 1.39 0.007 0.72 0.66 28 Morays 0.95 0 0.63 0.32 29 Parrotfishes 1.55 0 1.13 0.42 30 Blennies 1.60 0 1.24 0.36 31 Damselfishes 1.90 0 1.87 0.03 32 Bluefishes 0.67 0.005 0.63 0.03 33 Mesopelagics 0.61 0 0.34 0.27 34 Bathypelagics 0.15 0 0.12 0.03 35 Seabirds 5.40 0 2.33 3.07 36 Pisciv.m. mammals 0.10 0.04 0.05 0.01 37 Plank, m. mammals 0.05 0 0.01 0.04 38 Sea turtles 0.15 0 0.06 0.09 Z = Total mortality rate F = Fishing mortality rate M2 = Predation mortality rate M0 = Mortality for other causes 157 Appeodix 8. Percentage of change in biomass of all living groups in the Gulf of Mexico synthetic model under Ecospace simulations (20 years). Group species 1 Scenarios 2 3 4 1 Benthic producers 0.239 0.021 -0.048 0.038 2 Phytoplk. Inshore 0.207 0.103 0.103 0.000 3 Phytoplk. offshore -0.364 0.030 0.030 0.030 4 Meiobenthos 0.263 0.061 0.053 0.053 5 Macroinfauna 0.059 -0.029 -0.077 -0.047 6 Macreoepifauna -0.236 0.185 0.312 0.142 7 Herb.zooplk. inshore -0.347 0.233 0.233 0.233 8 Herb.zooplk. offshore -0.057 -0.539 -0.539 -0.539 9 Carn.zooplk. inshore 0.000 0.000 0.000 0.000 10 Carn.zooplk. offshore 1.714 -0.312 -0.320 -0.297 11 Juvenile shrimp -2.105 -1.064 1.064 0.000 12 Adult shrimp -2.799 -1.289 0.258 -1.031 13 Other decapods 0.146 0.084 0.689 0.104 14 Octopus -1.434 -0.721 1.287 -0.772 15 Clupeids -1.990 -0.738 -1.033 -0.394 16 Catfishes -0.630 -0.422 -2.426 0.316 17 Juvenile grunts 3.205 2.698 3.492 3.016 18 Adults grunts 1.950 1.944 -0.108 1.404 19 Juvenile groupers -14.286 -14.286 0.000 0.000 20 Adult groupers 0.000 0.000 0.000 0.000 21 Mullets 1.168 0.930 0.000 0.698 22 Coastal sharks 0.000 0.000 0.000 0.000 23 Oceanic sharks 0.000 0.000 4.762 0.000 24 Juvenile tuna -1.724 0.000 -1.754 0.000 25 Adult tuna -0.719 -1.439 10.791 0.000 26 Soldierfishes 0.000 0.000 0.000 0.000 27 Other fishes -0.998 -0.152 0.399 -0.076 28 Morays 0.000 0.000 0.000 0.000 29 Parrotfishes 0.000 0.000 0.000 0.000 30 Blennies 0.000 0.000 0.000 0.000 31 Damselfishes 0.000 0.000 0.000 0.000 32 Bluefishes 2.400 0.000 2.326 -1.550 33 Mesopelagics 1.750 1.714 1.749 1.750 34 Bathypelagics 0.000 0.000 0.000 0.000 35 Seabirds 0.000 0.000 0.000 0.000 36 Pisciv.m. mammals 0.000 0.000 0.000 0.000 37 Plank, m. mammals 1.205 1.205 2.410 2.410 38 Sea turtles -14.286 -14.286 -14.286 -14.286 158 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items