V. PANEL'S CONSULTATION WITH SCIENTIFIC EXPERTS

A. INTRODUCTION

5. 1. The Panel noted that none of the parties to the dispute had requested the Panel to consult experts. However, the Panel noted that parties had submitted a number of studies by experts and often quoted the same scientific documents to support opposite views. Under those circumstances, the Panel informed the parties that it had decided, acting on its own initiative, to seek scientific and technical advice pursuant to paragraph 1 and paragraph 2, first sentence, of Article 13 of the DSU. The Panel focussed its questions on two main areas: (i) approaches to sea turtle conservation in light of local conditions, and (ii) habitat and migratory patterns of sea turtles.

5.2. Regarding the criteria for selecting the experts, India, Malaysia, Pakistan and Thailand noted that the experts should be neutral, diverse in areas of expertise and geographically distributed as much as possible. The emphasis should be placed on experts who had knowledge and first-hand experience with respect to sea turtle populations in the areas of contention, namely Asia and South-East Asia. They should not come from the same university or the same team of research. Moreover, the experts should be asked to provide citations to all sources that they consulted for the purpose of providing information to the Panel and to attach copies of cited sources to any submissions to the Panel. India, Malaysia, Pakistan and Thailand further noted that the Panel had decided to seek expert opinion under the provisions of paragraph 1 and paragraph 2, first sentence, of Article 13 of the DSU, and had decided, therefore, not to establish an expert review group as foreseen in paragraph 2, second and third sentences, of Article 13 and Appendix 4 of the DSU. India, Malaysia, Pakistan and Thailand requested the Panel to conform as far as possible with the provisions of Appendix 4 of the DSU, and in particular with paragraph 3 of Appendix 4 which stated that, unless there was joint agreement of the parties to the dispute, citizens of parties to the dispute should not be called upon to render expert advice.

5.3. The United States fully supported the Panel having access to expert advice that it considered useful for the resolution of this dispute. The advice of qualified and impartial experts would support the scientific and technical information that the United States had presented to the Panel and would thus assist the Panel in resolving this dispute on the basis of the best available scientific data. According to the United States, the Panel's enquiry should be limited to resolving those factual issues necessary to determining whether the US measures met the criteria of Article XX(g) and (b); the Panel was not asked to address and decide general policy issues relating to shrimp trawling and sea turtles conservation. In order to determine whether the US measures related to the conservation of an exhaustible natural resource, or were necessary for the protection of animal life or health, the core scientific and technical issues were the following: (i) are sea turtles threatened or endangered worldwide?, (ii) does shrimp trawling without TEDs result in the death of large numbers of sea turtles?, (iii) do TEDs, when properly installed and used, significantly reduce the mortality of sea turtles caused by shrimp trawl nets?

5.4. According to the United States, the Panel should use two basic criteria in selecting the experts: (i) the persons selected should be "experts" with respect to those aspects of the dispute for which their opinions were sought; (ii) as stipulated in the Rules of Conduct for the Understanding on Rules and Procedures Governing the Settlement of Disputes they must be "independent and impartial, and shall avoid direct or indirect conflict of interest".³⁹⁶ In particular, no expert consulted by the Panel should be associated with the government of a party to the dispute. The disclosure requirements regarding the existence of any interest, and in particular employment interests, that could affect or raise doubts concerning a person's independence or impartiality also applied to the experts.³⁹⁷ Given the broad field covered by the questions, it was unlikely that many persons would have expertise with respect to each and every one of these questions. Each expert should therefore be instructed to answer only those questions in which they had expertise.

Panel Procedures with Regard to Scientific Expertise

5.5. The Panel asked the parties to the dispute to provide it with names of possible experts. The Secretariat, then sollicited brief curricula vitae from all proposed experts who were ready to assist the Panel. The parties were provided the opportunity to comment on these potential experts on the basis of the curricula vitae, and in particular to state any compelling objections they might have with regard to any individual.

5.6. After careful consideration of the curricula vitae and of the comments made by the parties, the Panel selected the following five experts:

5.7. These experts were requested to serve, in their own personal capacities, as individual advisers under the authority of the Panel. The Panel noted that, in their disclosure forms, three of the proposed experts disclosed what might be considered to be potential conflicts of interest. However, the Panel decided to confirm their appointments being of the view that the disclosed information was not of such a nature as to prevent the individuals concerned from being impartial in providing the scientific information expected of them. The Panel has also taken into account the disclosed information when evaluating the answers provided. The Panel underlined that, in making its choice, it had been guided primarily by the need to gather expertise of the best quality and covering as wide a field as possible. In the small community of sea turtle specialists, it was difficult - if not impossible - to reconcile this need with an agreement by all the parties to the dispute on each and every individual concerned.

5.8. The Panel, in consultation with the parties, prepared specific questions which it submitted to each expert individually. The experts were requested to answer only those questions which fell within their field(s) of expertise. The parties agreed that their written submissions to the Panel, including the written versions of their oral statements, be provided to the selected experts. The written responses of the experts, as well as copies of the sources cited in support of their responses, were provided to the parties, which were afforded the opportunity to comment on them. The questions asked by the Panel and the answers provided by the experts are presented in Section V.B. The comments by the parties are reflected in Section V.C. The United States raised the fact that, in their comments, some parties had submitted new material, i.e. material which had not been submitted by the time of the second meeting of the Panel. The Panel specified that it did not intend to take this new material into account in evaluating the comments made by the parties; the Panel would take into account only those comments which were strictly related to the scientific issues under discussion with the experts.

5.9. On 21-22 January 1998, the experts were invited with the Panel and the parties to discuss their written responses to the questions and to provide further information. A transcript of this meeting is contained in Annex IV.

B. QUESTIONS BY THE PANEL AND VIEWS OF THE SCIENTIFIC EXPERTS

5.10. The Panel requested the experts to focus their answers on the situation prevailing in India, Malaysia, Pakistan, Thailand and the United States, and on the following species of sea turtles: loggerhead (Caretta caretta), Kemp's ridley (Lepidochelys kempi), olive ridley (Lepidochelys olivacea), green turtle (Chelonia mydas), leatherback (Dermochelys coriacea), and hawksbill (Eretmochelys imbricata). The experts were also asked to cite references where appropriate.

General Comments by the Experts

Dr. J. Frazier:

5.11. The questions cover a wide range of topics, and many of them are broadly phrased, so to provide complete anwers requires considering a large number of variables. In general, differences between species, time and place all bear on different biological interpretations. Several of the questions are phrased in such a way that it would appear that what was expected was not only a concise, simple answer, but also the reduction of a series of options to one single alternative. If biology and biological conservation were as simple as rocket science, it might have been possible to provide brief, clear-cut answers. But biology is the study of life, of variation and change. It would be both arrogant and deceitful to pretend that biology, and even worse, that I myself, could consistently produce simple answers to simple questions. Furthermore, biological conservation is an interactive, iterative process, during which there are endless events of learning and experimenting. Since biological conservation is an attempt to use the information that we have in order to steward the resources on which we depend, the challenge becomes all the greater, for the needs and desires of many people and societies become paramount.

5.12. Hence, in many cases it seemed that as much as an answer, what was warranted was an explanation, at least from my point of view; and my intention has been to not only respond to the questions presenting my point of view, but also to provide citations to information which bears on my opinion. There are several general principles which I espouse: (i) do not assume that a lack of information is negative information, nor a justification for denying or asserting a case; (ii) to paraphrase from the United Convention on the Law of the Sea: be more cautious when information is uncertain, unreliable or inadequate, the absence of adequate scientific information shall not be used as a reason for postponing or failing to take conservation and management measures; (iii) develop and implement integrated approaches - not "either-or" alternatives - for conservation biology and resource management; (iv) thus, in concerns of resource management and conservation, especially when they confront various threats, the Precautionary Approach, as explained in the FAO Code of Conduct for Responsible Fisheries is essential.

5.13. While it is fully understood that the case at hand deals with a dispute before the WTO in which five countries are directly involved, there are several aspects of this initial orienting statement that warrant comment, concerning the issue of endangered species of sea turtles and their conservation.

(a) All of the six listed species of marine turtles disperse and migrate over vast distances, with no respect to national boundaries. This has been amply proven in the scientific literature, with contributions by nationals of many of the five countries that are involved; the research includes tag and recapture, satellite telemetry, genetic analyses (notably of the D-loop of mitochondrial DNA), and geographic distribution/life history information. A few of the better known examples, and review papers that synthesize many citations are discussed below. Because of the biological realities, it would be artificial, incomplete, inadequate and deceptive to limit the responses to what is known for just the five nations involved in a dispute. The issue at hand involves many other nations neighbouring those five: the conservation and management of migratory marine animals - marine turtles in this case - can only be accomplished through full international cooperation. Furthermore, many basic aspects in the biology of marine turtles are poorly known, and information available for some of the countries involved is very limited. Hence, it is frequently necessary to draw from studies done elsewhere in order to provide a response.

(b) The issue at hand is far greater than sea turtle conservation. Human activities - in this case fishing, and in particular, bottom trawling - have major effects on marine organisms and environments, some of which are critical to sea turtle survival, and many of which are utilized for human consumption. The subject in dispute is a small, though highly visible part of a gargantuan dilemma before modern society: the destruction of bycatch as a major contributor to the declining status of the world's fisheries. The focus on marine turtle conservation is justified in and of itself. At the same time marine turtles are "flagship species", charismatic, highly visible, and easily identified; and they are employed as ambassadors of the seas in a strategy to facilitate the resolution of other resource conservation dilemmas, less visible and attractive to the general public. A brief description of this conservation strategy, in relation to a new regional convention, is presented in Frazier (1997a).

(c) Finally, no resource conservation or management can be effective without including humans and their societies in the equation. Limiting the discussion of sea turtle conservation to biological and technical questions, risks ignoring the basic social problems, on which the conservation problems rest. Problems of biological conservation and the human situation are tightly inter-linked, and can only be solved in concert. A discussion of this argument, as part of a critique of the magic of "sustainable development", is developed in Frazier (1997b).

Mr. M. Guinea:

5.14. The base unit for sea turtle conservation and management is the demographic unit (Chaloupka and Musick, 1997)³⁹⁸ or breeding unit (gene pool). A country may have a single, or several, breeding units within its territorial waters. Sea turtles feeding in the waters of that country may not belong to the breeding unit. This has been demonstrated by mixed populations of hawksbill turtles on a feeding ground in Northern Australia (Broderick et. al., 1994). The paradigm of breeding units is essential to assess the threats and status of the sea turtle resources of a geographic area (Limpus, 1997). The concept of sea turtles being a global resource is philosophically laudable, but cumbersome in terms of conservation strategies.

5.15. Generalizations regarding sea turtles "... [being] found in the same general habitats and [feeding] on the same types of food throughout the world. Their feeding habits and habitat put them in the direct path of shrimp trawls where they were captured"³⁹⁹ are incorrect and hamper management options of individual countries in managing their breeding units of sea turtles. Some species e.g., loggerhead, olive ridley, Kemp's ridley and flatback are generally at risk from shrimp trawling. But because of their preferred habitats most greens, and usually hawksbills and leatherbacks, are relatively unaffected by trawling. Sea turtles are very long lived with several decades required before hatchlings grow to sexual maturity. Any management strategy employed to increase the number of hatchlings will not be obvious on the nesting beaches, the accepted reference for the condition of the breeding unit, for some decades.

5.16. The embargo imposed by the United States on the affected countries has been ineffective in reducing any sea turtle mortality because the trawling effort remained unchanged in the affected countries, and alternative markets were found for shrimp banned from the US market. In Australian waters, the incidental catch of sea turtles is directly related to the fishing effort (Poiner et al., 1990). A similar relationship exists in the United States (US National Research Council, 1990). There is no indication that fishing effort decreased in any of the affected countries. The figures given relating to trade pre and post 1 May 1996 relate to exports of shrimp to the United States. India has indicated that other markets for their non-TED shrimp were found.⁴⁰⁰ This indicates that shrimp previously destined for the US market before 1 May 1996 could flood existing markets that do not require the use of TEDs for their imported shrimp. This was anticipated by Australia which exports considerable amounts of shrimp of which only a small proportion has been exported to the US market (Stanley, 1996). The embargo imposed by the United States has readjusted trade in shrimp without reducing the alleged mortality of sea turtles in the affected countries.

5.17. Affected countries may still export shrimp to a third country(s) for either processing or transshipment to the US market.⁴⁰¹ A number of countries in their third party submissions indicated that they did not have trawl fleets and did not allow trawling in their waters, but were involved with trade in shrimp.⁴⁰²

5.18. The report "Decline of Sea Turtles" (US National Research Council, 1990) was a fine body of work by a highly respected group of scientists, but it focused on mainland Unites States of America with some references to its Caribbean Territories but scarcely mentioned the Pacific Ocean States and the Pacific Ocean Territories. Its outlook is therefore ethnocentric and relates to the decline of sea turtles in the Gulf of Mexico, Western Atlantic Ocean and Caribbean Sea by essentially the US shrimp trawling fleet. I have difficulty extrapolating its conclusions to the global scale. The Australian Endangered Species Scientific Subcommittee has been evaluating a nomination for otter trawls as a key threatening process. After nearly two years of deliberation, it is unprepared to so make such a recommendation because of the equivocal reports of the relative effects of trawling on Australian sea turtles and other causes of decline e.g., egg predation. It will seek further advice before making another statement in approximately one year's time.⁴⁰³

Question 1: Status of sea turtle populations - Past and current threats

1(a) Biologists consider that sea turtle populations around the world are affected by various factors, mainly anthropogenic. Are sea turtles threatened or endangered worldwide? Have the causes of any decline of sea turtle populations been the same for all species of sea turtles? Have these causes been similar in different parts of the world? Have these causes been similar over time?

Dr. S. Eckert:

5.19. There can be no question that global sea turtle populations have declined significantly to the point where all species are in danger of extinction. Leatherbacks, green turtles, hawksbills, olive ridleys, and the Kemp's ridley are classified as Endangered in International Union for Conservation of Nature and Natural Resources (IUCN) Red Data Book and the loggerhead is classified as Vulnerable. Such listing reflects on the global status for each species. Further, all of these species are included on Appendix I of the Convention on International Trade In Endangered Species of Wild Flora and Fauna (CITES). While such listing is designed to regulate cross border trade in listed species and not control within country utilization, listing by CITES does reflect on the global status of the species.

5.20. Relative to commenting on the status of the species within the regions of the disputing parties, some discussion of how population status is determined is necessary. It is possible to evaluate the status of stocks within different regions, but these values cannot be applied as though regional populations were independent management units. The reason for such limitation is that we do not know the full geographic distribution of each stock, and that stock status is assessed by nesting beach census. Our current rudimentary understanding of sea turtle life history and cohort movements or migrations does not yet allow us to define individual stock boundaries or home ranges (Musick and Limpus, 1996). For example, all sea turtle species except one (Australian Flatback) have a pelagic phase in their development, whose duration is not yet well defined but is apparently in excess of 5 years (Musick and Limpus, 1996). Loggerhead sea turtle hatchlings dispersing from Japanese nesting beaches move across the North Pacific Ocean and reside off the US west coast and the Baja Peninsula of Mexico, before returning to Japan to continue development to maturity (Bowen et. al., 1995). A similar developmental migration occurs in the Atlantic with loggerheads hatched on the US east coast, migrating to Eastern Atlantic developmental habitats. Unfortunately, these are the only sea turtle stocks that we understand where the pelagic developmental phase of theirs lives is spent. All other species are unknown, but similar cycles are likely. Without a clear understanding of the distribution of individual stocks, it is not feasible to determine their population status. Thus, consideration of population status must still be based on the global species status.

5.21. Research into individual stock boundaries is still in its infancy. Improvements in stock identification techniques using mitochondrial and nuclear DNA, as well as improved satellite telemetry are rapidly changing what we know about stock ranges of turtle populations. Unfortunately, both are relatively new methods and sample sizes are still very small. Often information gathered using these new methods cause us to extend what we previously considered the home range of an individual stock. In 1996 I discovered that leatherback turtles distribute across ocean basins covering far greater ranges than had been expected from tag return data (S. Eckert, 1997). By satellite tracking 3 leatherbacks from the nesting beaches on the Caribbean island of Trinidad it is apparent that leatherback females circumnavigate the north Atlantic ocean annually. In the Pacific my ongoing satellite tracking studies of leatherbacks indicate that these turtles migrate from Mexican and Central American nesting beaches to Chile and Peru and probably also circumnavigate the entire Pacific Ocean. DNA analysis of leatherback caught in the north Pacific and stranded on the west coast of the US indicate that leatherback nesting stocks from Malaysia (and probably Thailand as well), Indonesia, the Solomon Islands, Mexico, and Costa Rica distribute throughout the ocean basin (Peter Dutton, NMFS pers. com.).

5.22. Disagreements on population status often revolve around confusion on what constitutes a population. To compound this confusion, what is usually referred to as a population by the scientific community actually refers to a nesting population or nesting stock. A nesting population describes only the mature females utilizing a particular beach or area for nesting. Traditional monitoring methods for a sea turtle "nesting population" is to count the number of females annually nesting at particular beach, and utilize these counts to calculate nesting population status. The primary reason for this approach is that nesting activities are obvious and can last for many days or weeks after nesting. However, it should be realized that such methods have limitations that must be accounted for when conducting trend analysis.

5.23. There are often stochastic fluctuations in the annual numbers of females nesting in any given year that may be brought about by environmental conditions such as the southern oscillation or El Niño events (Limpus and Nicholls, 1988). Such fluctuations can be quite large. There are also regional differences in what is known as the remigration interval, or the time between nesting seasons for an individual turtle. Generally for most species and most regions this interval is 2-3 years, but in some areas may extend to 5-7 years (Van Buskirk and Crowder, 1994, Limpus et. al., 1992, Dodd, 1988, Witzell, 1983, Hughes, 1974). The reasons for this difference is yet unclear, but may be reflective of local foraging habitat quality. Thus it is recommended that when defining a population trend, census coverage be maintained for 3 times the average remigration cycle, which for most species and most populations requires a nesting beach be monitored for a 6-9 years. One exception to this monitoring duration is for the ridley species which tend to nest annually. Confusion as to population status is often due to trend analysis being carried out on census durations that are too short and thus overly influenced by stochastic fluctuations.

5.24. Determining population or stock status based on the numbers of nesting females can also sometimes mask population status because female sea turtles generally require between 20 and 35 years to reach maturity. Thus, conservation actions or perturbations to the nesting beach population can take many years to be reflected on the number of females nesting annually. This is likely why the leatherback nesting population at Terengannu, Malaysia took so long to collapse. It took at least 40 years of almost 100 per cent egg harvest for this population to be reduced to an effectively extinct nesting population (Chua, 1988a, 1988b, Chan and Liew, 1996). When examining population status, it is critical to remember that these long lag times can confound trend analysis.

5.25. Some analysis of each species current status can be summarized as follows:

5.26. Global population outlook for the leatherback sea turtle is extremely poor. Since 1980 most data indicates that the global population has declined substantially. Of the 28 nesting areas reviewed by Spotila et. al., 1996, 10 may be in decline, 5 may be increasing and 13 may be stable. Even more importantly the largest nesting populations (Mexico, French Guiana/Suriname, Irian Jaya, Gabon, Malaysia), only one may be stable (French Guiana/Suriname). Most of the decline has been in the Pacific Ocean with the nesting populations of Malaysia virtually gone, the nesting populations of Irian Jaya in doubt (Bhaskar 1985, Stark, 1993 ), but likely reduced, and the once largest nesting population in the world in Mexico almost gone (Spotila et. al., 1996, Sarti et. al., 1996).

5.27. Of all the species, leatherbacks have the most regular long distance migration through the waters of a large number of countries. In the Atlantic, leatherbacks tracked by satellite travelled to the North Atlantic and then south to Africa in a single year (Eckert, 1997). During this single year migration, the turtles passed through the jurisdiction of as many as 7 countries. In the Pacific it appears that females nesting in Mexico and Central America reside for some time in the coastal waters of Chile, but based on DNA data, will also migrate to the northeast Pacific and then down the coast of the Western United States to Mexico. Thus, it can be predicted that the home range for all nesting populations of the leatherback in the Pacific extends to virtually every government region of the Pacific.

5.28. Green turtle populations for the region(s) are also in decline. According to Groombridge and Luxmoore (1989 ) "around half of the extant nesting populations are either known or suspected to be depleted or in decline, ...". The draft Recovery Plan for US Pacific Populations of the green turtle (NMFS and USFWS, 1996b), which describes all US Pacific ocean populations as well as those of the Republic of Palau, the Federated States of Micronesia and the Republic of the Marshall Islands, states that "green turtle throughout the insular Pacific region has likely continued to decline due to directed harvest (both illegal and legal) and negative impacts to essential habitats".

5.29. Green turtles nesting populations throughout Malaysia are also in decline (de Silva, 1982, 1987, Eckert, 1993, Chan and Liew, 1996). On peninsular Malaysia, green turtle nesting populations declined 43 per cent between 1956 and 1982 (Eckert, 1993). Given the large and continuing (illegal) egg take in Sabah and Sarawak these nesting populations will continue to decline. Between 1965 and 1973 more than 6 million eggs were harvested from the Turtle Islands (de Silva, 1982 in Eckert, 1993) and turtle egg poaching continues (Francis Liew, in Eckert, 1993) despite the areas classification as a marine turtle refuge. In neighbouring Sarawak 1-3 million eggs were collected per year between 1927 - 1960, 500,000 per year in the 1960's and <300,000 eggs collected until 1986 (Banks, 1986 in Eckert, 1993). In 1989 and 1990, 185,461 and 117,701 eggs respectively were collected (Eckert, 1993). Further, recent information suggest that development pressures from Malaysian business interests at the Turtle Islands may also threaten nesting populations there (Romeo Trono, pers. com.).

5.30. As tropical coral reef residents, hawksbill sea turtles are faced with very much the same suite of threats faced by green turtles. However, the global populations are generally considered to be in far worse condition than green turtles. The Recovery Plan for the US Pacific Populations of the hawksbill turtle (NMFS and USFWS, 1996e) describes that status of the species very well:

This latter issue is the primary reason that hawksbill population are in so much worse shape than the green turtle.

5.31. While the olive ridley is considered the most numerous species of sea turtle, its populations have also been reduced. In Pacific Mexico, overexploitation of the nesting females and their eggs caused the collapse of 3 of the 4 arribada beaches (Eckert, 1993). Harvest of nesting females was so extensive (for the leather trade) that between 75,000 and 100,000 females were killed each year (despite a legal limit of 20,000). In May of 1990, the harvest of turtles was banned in Mexico. Population status for stocks nesting in India are far less clear. Based on my review of data presented by Dash and Kar (1990) there is no clear trend in nesting population status at Gahirmatha.

5.32. Globally loggerhead populations are considered in less danger of immediate extinction than most other sea turtle species. However regionally there have been serious population declines, particularly in the south-east United States (NRC, 1990). The primary cause for these local population collapses have been shrimp fishing (NRC, 1990). However, some re-consideration of the global status of loggerhead may be warranted in light of the rapid growth of longline fishing methods. Beside shrimp trawling loggerheads are the most frequently caught sea turtle species in longline type fisheries. (Aguilar et. al., 1992, 1993, Balazs and Pooley, 1994).

5.33. It is extremely difficult to credit any one particular cause with decline for all sea turtle species. In the Draft Recovery Plans for US Pacific populations of sea turtles (NMFS and USFWS, 1996 drafts a-f) we identified 29 different general categories of threats to marine turtles, 26 of those were anthropogenic. Those anthropogenic sources fall under 3 headings: (a) direct intentional take of turtles for food or commercial product; (b) incidental take by fisheries; and (c) destruction of habitat. Historically the most significant threats fall within headings (a) and (b).

5.34. The tremendous decline in leatherback sea turtle populations can probably be attributed to over-harvest of eggs, and incidental take in fisheries. The best example of the destruction of any nesting population of sea turtles by over-harvest of eggs was the leatherback nesting population at Terengannu, Malaysia. Mortality of adult turtles was limited at this nesting colony due to religious constraints but commercial egg take was in excess of 90 per cent for over 50 years and caused a slow decline to less that 100 females (Chua, 1988a, 1988b). Some mortality to this population can likely be attributed to the high-seas driftnet fishery which operated through the 1980's and early 1990's, and to trawl fisheries operating off the coast in the early 1980's (Wetherall et. al., 1993, Chan and Liew, 1996). The once large Mexican/Costa Rican populations of nesting leatherback is likely a good example of the impact gillnet and longline fisheries can have to a sea turtle population. Throughout the 1980's the high seas driftnet fleet caught approximately 1000 leatherbacks per year (Wetherall et. al., 1993, Eckert and Sarti, 1997). While this take was likely from all of the nesting stocks in the Pacific, the exceptionally large numbers of leatherbacks nesting in Mexico and Central America, has probably meant that the majority of those killed in the north Pacific were from those stocks. In the mid-1980's, Chile and Peru initiated large scale gillnet and longline fisheries for swordfish and it is estimated that they kill in excess of 2000 leatherbacks per year in this fishery (Eckert and Sarti, 1997). In only 10 years, the population of leatherbacks nesting in Mexico alone has declined over 95 per cent (Sarti et. al., 1996). This decline occurred despite extensive efforts by Mexico to protect their nesting stocks of sea turtles on the beaches.

5.35. Green turtle population declines can generally be attributed to intense harvest for meat, eggs and turtle products, and secondarily to incidental take in fisheries. This species has been highly sought as a source of food, both commercially and by indigenous peoples (Groombridge and Luxmoore, 1989). While most countries have laws to limit such take, those laws have generally been ineffective, such that large scale harvest still continues (Eckert, 1993). However in some areas such as the Pacific coast of Mexico and NE South America and Thailand shrimp trawling has also been a significant source of mortality for these species Hill, 1991, Eckert, 1993, Chantrapornsyl, 1997).

5.36. There are 2 primary causes of population decline for the loggerhead sea turtle. In the southeastern United States, it was estimated that shrimp trawling accounted for the mortality of 50,000 loggerheads per year (NRC, 1990). However, this threat in the United States has been largely eliminated with the application of TEDs in shrimping trawls (Henwood and Stuntz, 1987). In the Pacific Ocean, the high seas driftnet fleets also caught large numbers of loggerheads during the 1980's and early 1990's, but this threat has been largely removed by the outlawing of that fishery (Wetherall et. al., 1993). A particularly serious and growing source of mortality for this species is pelagic longline fisheries in the Pacific and the Mediterranean. Loggerhead turtles will feed on bait used in longline fisheries and become hooked. Large numbers of mostly juvenile loggerheads are killed or injured by these fisheries (Aguilar, 1992, Aguilar, 1993, Argano, 1983, Balazs and Pooley, 1994).

5.37. Olive ridleys have long been harvested in Central America for eggs, meat and skin. This harvest was so intense in Mexico that 3 of the 4 arribada beaches were extirpated by the 1980's (Eckert, 1993). Such harvest was banned in Mexico by 1990 and there is some evidence that the remaining arribada population may be recovering (Marquez, 1996b). Olive ridleys are also heavily impacted by shrimp fishing in Central America, India, Suriname (Hoekert and Schouten, 1996) and to a lesser extent in Mexico. The incidental take of olive ridleys in India is exceptionally severe which supports the largest nesting aggregation of this species in the world. Annually 5,000 - 8,000 dead turtles wash up on the beaches of Orissa which are attributed to incidental take in shrimp trawls. Despite laws banning such fishing, large scale shrimp fishing is occurring within the Bhitara Kinika Sanctuary (the primary nesting area for olive ridleys in India) and more than 4,000 olive ridleys stranded dead on the nesting beach during 1996/97 (Das, 1998). Finally there is evidence that the incidental mortality of olive ridleys due to shrimp fisheries is not limited to reproductive adults, but also to what are likely resident juveniles (Pandav and Choudhury, 1995). Two things are clear relative to the incidental take of olive ridleys in India. The first is that there are severe problems with enforcement of regulations protecting these important olive ridley nesting beaches and, secondly, there seems to be conflicts between the State and Federal government as well as the fisheries resource management agencies in India over the need to protect sea turtles. This latter problem is well defined by incidents in which the State of Orissa attempted to build fishing harbours within and alongside the sanctuary to support increased shrimp fishing, despite the protected status of the area (Andrews, 1993, Mohanty-Hejmadi, 1994, Das, 1998 in press).

To Continue With Chapter 5.38

³⁹⁶ See WT/DSB/RC/1, Article II.1 (hereinafter the "Rules of Conduct").

³⁹⁷ Article III.1 and VI.2 of the Rules of Conduct.

³⁹⁸ The complete references of the literature and other sources cited by the experts can be found in Annex III.

³⁹⁹ See above paragraph 3.61.

⁴⁰⁰ See above paragraph 3.125.

⁴⁰¹ Verbal presentation by India at FAO Responsible Fishing Workshop Darwin, NT, Australia, 24-26 July 1997.

⁴⁰² See above paragraphs 4.49 and 4.61-62.

⁴⁰³ Interim Advice to the Minister of the Environment from the Endangered Species Scientific Subcommittee of a Public Nomination to Schedule 3 of the Endangered Species Protection Act 1992.