5(b) What is the typical range of migration of the various species of sea turtles, in particular in relation to the territories (including overseas territories) of the countries concerned? What is the maximum range?

Dr. S. Eckert:

5.270. See my response to question 5(a).

Dr. J. Frazier:

5.271. It would be difficult to derive a value for a "typical range of migration" for a population of sea turtles, much less for a species. Firstly, precise information on migrations of sea turtles from the Indo-Pacific region is only recently becoming available. Secondly, much information is from tag returns, and this only reveals where the turtle was caught - not the route it traveled, nor where it was actually headed. Third, there is often tremendous variety in the final destinations and distances where turtles are recaptured, after being tagged and released.

5.272. A very brief review of some of the more remarkable data on migrations from the Indo-Pacific follows, all it is from nesting females. As more studies are carried out, especially using satellite telemetry, a much better information will come to light on the intricate relationships between nesting beaches, feeding ground, and migratory routes.

• Although at least 2,351 green turtles and 42 olive ridleys have been tagged at Hawksbay, Pakistan (Firdous, in press), there seems only one tag recovery from outside Pakistan. One green turtle tagged in Hawksbay was recaptured in the Gulf of Kutch, India (Firdous, 1991). The distance involved is relatively short, considering the distances that green turtles are known to have moved in other populations.

• Tens of thousands of olive turtles have been tagged at Gahirmatha, Orissa, India, but few if any have been reported recaptured away from India. There are observations that flotillas of these turtles may migrate from offshore Sri Lanka to Gahirmatha (Silas, 1984; Silas et. al., 1984).

• Long-range migration data are available for three species of sea turtle in Malaysia. Leatherbacks tagged at Terengganu have been captured at tremendous distances from their home beach, as far away as Taiwan, Japan and Hawai'i (Leong and Siow, 1980). Green turtles tagged in Sarawak have been recovered as far off as Philippines and California (Leh, 1989). During recent years, a wealth of information on migrations has been coming out of Malaysia. Green turtles nesting on Redang Island, off the coast of Terengganu, Peninsular Malaysia, have been tracked with satellite transmitters more than 1,600 km east to Sabah and Philippines, as well as some 1,000 km southeast into Indonesian waters (Liew et. al., 1995a; 1995b; Papi et. al., 1995). Once they have finished nesting in the Sabah turtle islands, green turtles disperse north and east to the Philippines and even to Palau Islands, as well as south into Indonesian waters; some of the distances between sites of marking and recapture are close to 2,000 km (Chan and Liew, 1996b). Hawksbills from the Sabah turtle islands also disperse east to Philippines (Chan and Liew, 1996b).

• There seems to be no information on tagging or tag returns or sea turtle migration from Thailand.

• From the United States there is a considerable amount of information on long-range tag returns, and more recently satellite telemetry. Eckert (1993) has reviewed findings from the North Pacific. Since then several studies of satellite telemetry have documented movements of green turtles from French Frigate Shoals to Hawaii and Johnson Atoll (Balazs, 1994; Balazs and Ellis, in press), as well as from Rose Island to Samoa (Balazs et. al., 1994). Hawksbill turtles have made shorter movements, within the Hawaiian Islands (Balazs et. al., 1997; in press). Pultz et. al. (in press) found that one of six green turtles tagged while nesting on Tinian Island, Commonwealth of Northern Mariana Islands, was recaptured in Philippines a year later. Dutton et. al. (in press) have found that one of two leatherbacks caught in Hawaii had a haplotype that has been found in Indonesia.

5.273. It is important to understand that in those regions with more active research activities, more scientific information is available. The absence of information is no proof of the absence of a phenomenon; until a systematic study has been carried out to objectively show that a specific phenomenon does not occur, one cannot draw defensible conclusions on the basis of the lack of information.

Mr. M. Guinea:

5.274. The years that juvenile sea turtles spend in their pelagic existence after leaving the rookery enable them to drift around an ocean gyre. At any one time they may be thousands of kilometres away from their natal beach. (The Australian flatback is exceptional in not having a pelagic phase to its life cycle.) The coastal developmental habitats through which they pass as they mature, do not necessarily bring the sub-adult closer to its natal beach. The movement from the adult feeding ground to the nesting beach and return is considered a true migration. Tagging studies in Australia have indicated that loggerheads travel hundreds and even several thousand kilometres to nesting beaches and return during a reproductive migration. Green turtles have been recorded travelling up to 2600 km from rookery to feeding area, but most travel less than 1000 km. Hawksbills travel up to 2369 km in one instance but most travel a shorter distance. In Malaysia, green turtles travel over 1700 km after nesting (Liew, 1997). In India, olive ridleys travel within the country from Orissa to the Gulf of Mannar, over 1,000 km. Leatherbacks appear to retain their pelagic existence in adulthood and may, in their non reproductive state, be several thousand kilometres from their natal beach.

5.275. The concept of maximum range is attributable to sea turtles that migrate from a feeding area to a nesting area and return to the feeding area. The maximum ranges as reported above are in the order of 2,000 km. Breeding units from nesting beaches may be detected on feeding grounds. Comparisons between the genetic profiles of a sample of sea turtles at a nesting beach and a sample of adult mature females turtles on the feeding ground may indicate if they are of the same breeding unit. This may be supported further by tagging programmes. If a turtle has been tagged either at the feeding ground or the nesting beach, then its life history may be pieced together from subsequent recaptures on either feeding grounds and the nesting beach. It is the successful completion of the migration that separates a normal sea turtle from a "waif" which has been carried or drifted out of its "normal" range.

Mr. H.-C. Liew:

5.276. Turtles migrate from their feeding or "home" grounds to the nesting grounds when they are physiologically ready and are in the reproductive phase. This does not occur every year for the individual female but happens in cycles of between 2 to 7 or more years. This is because the females need to build up sufficient fat (or food) reserves to sustain them throughout the breeding season which may last up to 3-4 months before they are able to return back to their feeding grounds. What is known of the green turtles throughout this period, i.e. during migration and at the nesting grounds, they hardly feed. Hence migration ranges would be somewhat restricted. The migration ranges of most green turtles would be in the region of 500 to 2,500 km. Anything beyond that would put severe restrictions on their survival. Leatherbacks, however, being and ocean-pelagic species are capable of migrating over much longer distances.

Dr. I. Poiner:

5.277. See my response to 5 (a).

Question 6: Relation between sea turtles and shrimping grounds

6(a) Does sea turtle biology and in particular the spatial and temporal relation between sea turtles and shrimp differ between the Atlantic and the Indo-Pacific waters? To what extent do habitats and/or nesting grounds of the different species of sea turtles coincide with shrimp fishing grounds?

Dr. S. Eckert:

5.278. Due to the limited information available on the distribution of foraging turtles in Thailand, Malaysia, India and Pakistan I am not able to address the question of where shrimping and turtles might interact. Except for the few reports of where turtles have been killed by shrimping (Orissa, India, Terengannu, Malaysia, United States Atlantic coast and Gulf of Mexico), predicting where such interaction could occur is difficult.

Dr. J. Frazier:

5.279. It is important to understand that "sea turtle" refers to any one of five species of sea turtles, and "shrimp" refers to scores of species; in some countries, a dozen species of shrimp and prawn may be harvested. Each species will have its own life history, with different spatial and temporal characteristics. I am not versed in these details. The spatial and temporal relationship between sea turtles and shrimp trawling has been abundantly and systematically documented. The first global review on the subject was presented by Hillestad et. al. (1982), and since then much more information has become available. There have been specific studies in both northern and eastern Australia (Poiner and Harris, 1994; Robins, 1995 Guinea and Whiting, 1997); on the Pacific coasts of Guatemala, El Salvador, Nicaragua and Costa Rica (Arauz, 1990; 1996a; 1996b; Arauz et. al., 1997a; 1997b); Mexico (Olguin, 1996); along the southern Atlantic and Gulf of Mexico coasts of the United States (National Research Council, 1990; Crowder et. al., 1994; 1995; Weber et. al., 1995); and on the Caribbean coast of Venezuela (Marcano and Alio, 1994). There is also information in the scientific literature from many other countries, such as: Eritrea (Hillman and Gebremariam, 1996), India (e.g., Silas et. al., 1983a; 1983b; 1985; Pandav et. al., 1997), Kenya (Wamukoya et. al., 1996), Malaysia (Suliansa et. al., 1996; Ali et. al., 1997), Mauritius (Mangar and Chapman, 1996), Tanzania (Howell and Mbindo, 1996) and Turkey (Oruç et. al., 1997).

Mr. M. Guinea:

5.280. Just as there are a number of species of sea turtle there are even more species of shrimp. Generalizations about sea turtles and shrimp interactions should be avoided as different shrimp species of different market value have different preferred habitats. Particular shrimp species are targeted by the operators. In Australian trawl fields, some species e.g., the banana prawn (Penaeus merguiensis) form dense aggregations which discolour the shallow water and the schools of prawns form an image on depth sounders. Beam or otter trawls are used to target such aggregations. Tow durations rarely exceed 30 minutes. In such short tows on a targeted school, sea turtles are rarely captured. Other prawn species inhabit deep water (90m). Trawls may be longer, but turtles are seldom found at those depths and any negative impact is unlikely. Trawling for some tiger prawn species is conducted in shallower water, with 3 hours per tow being relatively common. They have the potential, if unchecked by restrictions, to interact with loggerhead, olive ridley and flatback sea turtles. Operators may target different shrimp species at times of the year. Or alternatively they may target different species within a single cruise.

5.281. Regions offshore from sea turtle rookeries, by the soft nature of the sea bed, may support a shrimp ground. These areas offshore from rookeries should have seasonal closures to fishing activities that have the potential to harm sea turtles. The extent of the closed area will depend on the species of sea turtles nesting. Some species may be protected by a 3 km wide refuge, but others e.g., leatherback, may require a width of 20 km for a successful refuge. This is a situation best left to the legislators of the respective countries. All the countries in the dispute have indicated that sanctuaries or seasonal refuges have been established offshore from nesting beaches.

Mr. H.-C. Liew:

5.282. In a broad general sense, they are similar but there will also be localised differences. In Asia, we have the wet and dry season brought about by the monsoon, which may be somewhat different in the Atlantic. Even within the same region, some sea turtle populations nest in the dry season, but others may nest in the wet season. There are also some locations, like in the Sabah turtle islands where nestings occur throughout the year. The season for shrimp trawling may also differ. Feeding habitats of different sea turtles would differ depending on their diet but these habitats may overlap. An area of seabed may have green turtles, hawksbills, loggerheads and ridleys occurring together as the area may have pockets of seagrass, sponges, crabs, shrimps, molluscs and fish there. On the other hand, over a seagrass area in an estuary, you may find only green turtles feeding there. Since loggerheads and ridleys feed on crustaceans and molluscs while green turtles and leatherbacks feed on seagrass/algae and jellyfish respectively, shrimping grounds would have a stronger association with loggerheads and ridleys than the other species. Not all sea turtle nesting grounds have good shrimping grounds in the vicinity. Sipadan Island, off Sabah, Malaysia, is a world renown nesting beach for green turtles but no shrimp trawlers can operate there as the waters off the island is 2,000 ft deep. Many such islands and atolls do occur throughout the Indo-Pacific.

Dr. I. Poiner:

5.283. Globally, tropical and sub-tropical shrimp fisheries are generally concentrated in relatively shallow coastal waters (< 80m). Sea-turtle nesting and foraging habitats also tend to occur in the shallow coastal waters. Hence there is and will continue to be significant interaction between shrimp fisheries and sea-turtles.

6(b) Are statistical comparisons of the interaction between shrimp trawling and sea turtle populations in the Atlantic and in the Indo-Pacific waters available? If so, what do they indicate?

Dr. S. Eckert:

5.284. To the best of my knowledge there are no statistical comparisons on shrimp fishery / sea turtle interaction between the waters around Thailand and Malaysia and the United States. However, there are some studies on the Australian prawn fishery (Dredge, and Trainor, 1994, Harris and Poiner, 1990, Poiner et. al., 1990), the latter comparing Northern Australian catch rates directly to Henwood and Stuntz (1987) report of US catch rates. While this study showed comparable catch rates between the US Gulf of Mexico and Northern Australia, the mortality rate for Australia was much lower. Unfortunately, for comparative purposes the Australian study was hampered in that the primary species caught (43 per cent) was the endemic Australian Flatback. This species has a very unique life history as compared to all other marine turtle species and it is not known if it has a higher resistance to drowning than other species. Thus, it is difficult to know if the different mortality rates are due to geographical or species composition differences between Australia and the United States.

Dr. J. Frazier:

5.285. Systematic studies of the interactions between shrimp trawling and sea turtles have been carried out in both Northern and Eastern Australia (Poiner and Harris, 1994; Robins, 1995); on the Pacific coasts of Guatemala, El Salvador, Nicaragua and Costa Rica (Arauz, 1996a; 1996b; Arauz et. al., 1997a; 1997b); along the southern Atlantic and Gulf of Mexico coasts of the US (National Research Council, 1990; Crowder et. al., 1994; 1995; Weber et. al., 1995); and on the Caribbean coast of Venezuela (Marcano and Alio, 1994).

Mr. M. Guinea:

5.286. Available data indicate that sea turtle mortality rates are higher in the Gulf of Mexico (29 per cent) and Atlantic Ocean Shrimp Fishery (21 per cent) than has been found in the Northern Prawn Fishery (6-10 per cent) and the East Coast Trawl Fishery (1-6 per cent) of Australia (Robins, 1995). The catch rates of sea turtles per unit of effort was greater in the American shrimp fisheries (0.0031-0.0487 per net h) than in the above mentioned Australian fisheries (0.0057-0.01 per net h). The species of turtles impacted also differed, with loggerheads, Kemp's ridleys and greens being present in the US fisheries and loggerheads, flatbacks, olive ridleys, greens and hawksbills present in the Australian fishery.

Mr. H.-C. Liew:

5.287. Unable to source such information.

Dr. I. Poiner:

5.288. Poiner and Harris (1996) compared the incidental catch of sea turtles in Northern Australia with the Gulf of Mexico and southern North Atlantic. The catch rate of turtles in the Australian Northern Prawn Fishery (mean 0= 0.0113, 95 per cent CI 0.0012 turtles) is higher than the rate Henwood and Stuntz (1987) reported for the Gulf of Mexico (mean = 0.0031, 95 per cent CI 0.0008 turtles) but lower than the rate they reported for the southern North Atlantic (mean = 0.0487, 95 per cent CI 0.0041 turtles). Most prawn trawling in the southern North Atlantic fishery occurs in water depths less than 18 m and, as in the Northern Prawn Fishery, catch rates vary with water depth, with the highest catch rates in water around 14 m deep. In the Gulf of Mexico, prawn trawling occurs in water depths up to 80 m, but unlike the other two fisheries, the turtle catch rate appears to be fairly constant over all depths up to 30 m.

5.289. The turtle mortality rates for the Gulf of Mexico and southern North Atlantic prawn fisheries were estimated as 29 per cent and 21 per cent of captures (Henwood and Stuntz 1987), which is higher than the 14.1 per cent estimated for the Northern Prawn Fishery. The difference may be due to different species having different mortality rates. The loggerhead dominates the American catches: 94 per cent of the southern North Atlantic and 86 per cent of the Gulf of Mexico catches. Its mortality rates were estimated as 29 per cent and 30 per cent respectively (Henwood and Stuntz 1987). The same species is a small component of the Northern Australian catch (10 per cent), but its estimated mortality rate is similar to the American rates (22 per cent). The loggerhead therefore appears to be particularly susceptible to drowning. In contrast, the dominant turtle in the Northern Prawn Fishery catch, the flatback (59 per cent), has a low mortality rate: 11 per cent. This species is endemic to Northern Australia and tends to be found inshore in relatively shallow (<40 m) muddy waters and possibly has a higher resistance to drowning in trawls (11 per cent mortality) compared to the other species. The difference in the overall mortality rates of turtles in the American and Australian fisheries may, therefore, be due to the dominant species being more or less susceptible to drowning.

6(c) Are all species of sea turtles significantly affected by shrimp trawling in different regions of the world? Or are some species likely to be more or less affected due to their nesting/feeding habits and migratory patterns and such divergences as might occur in those habits and patterns in different parts of the world ?

Dr. J. Frazier:

5.290. Any population of sea turtles that suffers mortality of breeders or near breeders in shrimp trawls will be significantly affected, independent of the species or locality. Some sea turtle populations may be more vulnerable to shrimp trawling than others because of spatial and temporal differences in occurrence of turtles and shrimp. By the same token, some human populations may be more vulnerable to cocaine addiction than others, but in all human populations this drug represents a risk to society.

Mr. M. Guinea:

5.291. All species of sea turtles are not adversely affected by shrimp trawls. Some species have preferred habitats which do not always coincide with shrimp trawl fields. These habitats can be identified and, if need be, seasonal closures to turtle threatening activities may be imposed. Even on relatively uniform substrates sea turtle distribution is clumped rather than random. This gives rise to "hot spots" where sea turtles abound while in seemingly similar areas nearby, they are scarce. After almost one year of trials in the Northern Prawn Fishery, the Australian Fisheries Management Authority (Sachse and Wallner, in press) are looking towards a log-book programme of all sea turtle captures, resuscitation procedures for comatose sea turtles, and closure of some areas such as seagrass beds to protect juvenile tiger prawns and green sea turtles as well as the implementation of TEDs on a voluntary basis. This example from Australia demonstrates the complexity of introducing new technology into a fishery as well as adopting an ethos of responsible fishing. Any legislation requiring the use of TEDs on shrimp trawls would require the allocation of additional resources for enforcement of any such legislation. Australia is encouraging voluntary compliance by stressing the advantages of using Trawl Efficiency Devices (TEDs) in the fishery. This will take considerable time.

Mr. H.-C Liew:

5.292. All species of sea turtles have the potential of being caught in a shrimp trawl as much as any other sea creature large enough to be retained by the cod-end of the trawl net. The only difference is the probability of encounters. Some of the factors that dictate this probability are:

• Number of trawlers operating in the area, their size, power, efficiency, size of nets, trawl time, etc.

• How much the trawling grounds overlap with turtle feeding grounds.

• Species of turtles which will dictate their feeding habits, resting habits, migration routes, how long they remain on the seabed as opposed to midwater or surface, do they feed in the same area as the operating trawlers.

• Offshore internesting habitats, depth, frequency of shrimp trawlers operating there.

• Whether shrimp trawling seasons coincide with nesting seasons.

• Laws and regulations protecting the turtles.

• Enforcement of regulations.

• Awareness and education of the fishermen to turtle conservation.

All these factors do vary from region to region, hence the probability of encounters cannot be the same. There is no doubt that in some regions of the world, sea turtles are significantly affected by shrimp trawling where the probability of encounters is high but the same cannot apply for all regions. Moreover, threats due to other causes may impact the turtles more significantly than shrimp trawling for some regions.

Dr. I. Poiner:

5.293. See my answer to 6(b).

C. COMMENTS BY THE PARTIES

1. Comments by India

5.294. A review of experts' opinions shows that causes of decline of sea turtles should not be broadly categorised as being due to anthropogenic and natural causes. In the Draft Recovery plans for US Pacific populations of Sea Turtles⁴¹², Eckert et al. have identified 26 different types of anthropogenic threats. The degree and magnitude of these different types of threats are not the same for all species of sea turtles. Even for the same species differences exist in different geographical regions of the world. All experts have indicated that the causes of decline of sea turtle populations have changed over time for each region and for each species. Again all the experts have emphasized the lack of information in this regard, which makes it difficult for categorization of different threats. For sea turtle populations in different parts of the world, the general consensus is that in the past populations have vanished due to exploitation of eggs, habitats and adults for commercial purposes. There was a great demand for eggs and byproducts all over the world which led to the flourishing trade. Now such large scale exploitation for commercial purposes has completely stopped in many countries including in India. Although eggs have a special value for certain other qualities in addition to nutrition in some countries, in India, there is no such tradition. Again, from the experts' opinions, it emerges that the concept of sea turtles being a global resource, while being philosophically laudable, is cumbersome in terms of conservation strategies.

5.295. The 1997 Limpus study⁴¹³ has given an overview of the status of marine turtles of South East Asia and the Western Pacific region. His report does not include the status of turtles in Indian waters. Mr. Guinea has given most extensive answer to the question quoting data from several sources. We generally endorse his views.

5.296. Dr. Eckert's statements regarding "regional populations were independent management units"⁴¹⁴ and that "consideration of population status must still be based on the global species status"⁴¹⁵ are contradictory. Even the DNA analysis and satellite telemetry data by Dr. Eckert show that the Atlantic leatherback turtles migrate within the Atlantic Ocean. Similarly, the Pacific population remains restricted to the Pacific Ocean. The sweeping generalisation regarding leatherback nesting stocks from Malaysia/Thailand/Indonesia being distributed throughout the Ocean basin is based on personal communication (Peter Dutton, NMFS)⁴¹⁶ and needs to be reinforced by more objective data. Though the methods of monitoring nesting population status have limitations, methods remaining the same over years can be used for conducting trend analysis. Dr. Eckert further lists a few causes for population decline for different species. The following are our comments on Dr. Eckert's views.

5.297. Dr. Eckert's views do not include the recent data (MTN, 1996) regarding the recovery of Mexican population of olive ridleys. For population status of ridleys in India he has incorporated old data and not the recent publications which shows recovery in populations (Mohanty-Hejmadi, 1994).⁴¹⁷ Further, Dr. Eckert has quoted several sources regarding the death of some five thousand turtles attributed to incidental catch in trawlers. As pointed out in the presentation by Indian experts to the WTO panel, the paper mostly lists the number and type of fishing vessels in Orissa in which the number of the shrimp trawlers is much less than the other fishing vessels. The conclusion that all the dead turtles are due to shrimp trawler activities is not true. Although 5,000 may seem a large number from other population point of view, India would like to draw attention to Mr. Guinea's comment that "the annual mortality of 5000 from fishing trawls and set nets from the nesting population of 600,000 with a recruitment of 85,000 appears relatively minor".⁴¹⁸ Further, Eckert has mentioned the attempts of State of Orissa to build fishing harbours besides the sanctuary.⁴¹⁹ On this point India repeats that the jetty, specially the Tachua jetty which would have affected the Gahirmatha population, was never commissioned by the Government. Further the whole area has now been declared as a marine sanctuary with the zone up to 20 km from the coast line having been declared as a "No fishing zone". At present the coast guards and the Indian Navy are patrolling the area to enforce the Government's conservation programmes. Regarding loggerhead sea turtles, Dr. Eckert has not provided data on Indo-Pacific populations.

5.298. Dr. Eckert has shown that causes of mortality for sea turtles are different in different parts of the world. For example, decline in the loggerhead populations in North Carolina is due to driftnet fishery in the high seas. The South American swordfish driftnet and longline fisheries are the causes of mortality of Pacific leatherbacks. Dr. Eckert has also shown that gill nesting in South America is a major problem. The facts presented by Dr. Eckert show that no generalisation can be made about ranking sources of mortality in different geographic areas of the world. Dr. Eckert has given data only about leatherback sea turtles and no data about India to rank sources of mortality. Regarding anthropogenic threats, Dr. Eckert has not included the recent developments in sea turtle protection in India. He has given the status of harvest of eggs and adults prior to 1985. Indian experts have already provided factual data to the Panel to demonstrate the highly successful steps taken by India to conserve and protect its sea turtle populations.

5.299. Dr. Eckert, Mr. Guinea and Mr. Liew have given only qualitative information to differentiate between shrimp trawl and other fishing gear in terms of the mortality they represent to marine turtles. No qualitative information has been provided on the effect on sea turtle mortality due to trawl fisheries, gillnet fisheries, longline fisheries, purse seine fisheries, fish traps, fish bombing. No information has been provided about the degree of coincidence of the above type of fishing activities in different regions of the world where significant number of turtles occur e.g. feeding, breeding, migrating etc.

5.300. Scott Eckert's view that in India, the direct harvest of eggs and meat is apparently still a problem⁴²⁰ is not factually correct, as Indian experts have provided the Panel already with factual data to demonstrate that direct harvest of endangered sea turtle eggs and meat is not a problem in India. Regarding the influence of socio-economic factors on the choice and enforcement of conservation programmes, Dr. Eckert has mainly covered the conditions in America relating to application of TEDs. He has not given any answer relating to socio-economic factors in the five countries involved in this dispute. We agree with Mr. Guinea's assessment that the so-called by-catch in US terms is a commodity with either a subsistence or retail value. Mr. Guinea has given a more realistic account on the subject. However, his comment that sea turtle eggs are also eaten is not applicable to India. Large scale exploitation of eggs has been effectively banned since mid 1970s. Mr. Liew's views that turtles are slaughtered in Asia is not true for India.⁴²¹ The same applies for eggs.

5.301. In general we agree with the views of Mr. Liew that all measures that prevent sea turtles from being killed are important.⁴²² We further endorse the views of Mr. Guinea that nesting habitats should be preserved as should the offshore refuse habitats for nesting females.⁴²³ Dr. Eckert's views that very little is being done by most of the countries in this dispute to protect juvenile or resident adult sea turtles⁴²⁴ is not true. For most of the populations of Southeast Asia, the feeding areas remain unknown. Therefore, more emphasis has been placed on protection to adults and eggs. The offshore turtle sensitive areas declared as Marine Wildlife Sanctuaries have given adequate protection to mating/breeding/feeding/developmental habitats of sea turtles in India. India does not have "headstarting" programme for any species of sea turtle populations at present. We endorse Mr. Guinea's opinion that the Gahirmatha population of sea turtles is increasing or at least stable through protection of the nesting area.⁴²⁵

5.302. Dr. Eckert's view is that TEDs reduce sea turtle mortality only when installed and operated properly.⁴²⁶ Even in the United States, with years of education and conservation programmes, improper use of TED has resulted in continuing mortality of turtles. Dr. Eckert has admitted that he does not have any direct experience working with trawler fisherman from other countries involved in this dispute. In reality, improper use of TEDS, inefficiency in implementation and an ineffective monitoring mechanism can significantly affect the efficiency of TEDS, both in terms of loss of catch and protection of various species of sea turtles. The socio-economic conditions prevailing in the South Asian region require a different approach to that chosen in the United States in so far as the proper use, successful implementation and foolproof monitoring mechanism is concerned. No answer has been given by Dr. Eckert to the question "can alternative measures such as seasonal and time closures, area closures or low-time limitations achieved equivalent or better results". In areas of low congregation TEDs will reduce, but not eliminate deaths of sea turtles caused by shrimp trawl net. So far, there is no study on the efficiency of TEDs in areas of high turtle congregation zones and whether in such areas TEDs can significantly reduce the mortality of sea turtles has to be studied.

5.303. India agrees with the interpretation of Mr. Liew on the points that after many years of experiments, publicity campaigns, TED trials, the United States mandated use of TEDs in 1980. As recently as 1994, NMFS reported poor compliance with US requirements. resulting in a record number of dead sea turtles. India also agrees with his views that considering the socio-economic conditions, educational level, language and cultural differences, it will take some time to convince and introduce the use of TEDs in different countries. These are all time consuming processes. Mr. Guinea has also mentioned that for TEDs to be accepted, the technology has to be adopted for the local area. There is no data on TED efficiency in Indian coastal waters except for demonstration of a few hours. No data has been collected on the efficiency of TEDs or their effect on bycatch. In so far as the "data on rate of turtle stranding in areas where TEDs are currently required or on the relationship between turtle stranding and shrimping activities in areas where TEDs are required", India endorses the views of Mr. Guinea. Data on TEDs efficiency during commercial shrimping seems to be most extensive for the United States and data from other countries and geographical locations would be necessary for comments. On this last question, India further endorses the views of Mr. Liew.

To Continue With Chapter 5.304

⁴¹² NMFS and USFWS, (1996) drafts a-f, p. 5.

⁴¹³ Limpus, C. (1997, Marine Turtle Population of South East Asia and the Western Pacific Region: Distribution and Status, Proceedings of the Workshop on Marine Turtle Research and Management in Indonesia, Jember, East Java, Indonesia, November 1996.

⁴¹⁴ Eckert para. 5.20.

⁴¹⁵ Ibid.

⁴¹⁶ Eckert para. 5.21.

⁴¹⁷ Eckert para. 5.31.

⁴¹⁸ Guinea para. 5.199.

⁴¹⁹ Eckert para. 5.37.

⁴²⁰ Eckert para. 5.129.

⁴²¹ Liew para. 5.89.

⁴²² Liew para. 5.183.

⁴²³ Guinea para. 5.181.

⁴²⁴ Eckert para. 5.172.

⁴²⁵ Guinea, para. 5.189.

⁴²⁶ Eckert para. 5.202.