Dr. S. Eckert:

5.232. Renaud et. al., (1991) noted that there were differences in catch rates between TED equipped nets and non-TED nets when comparing tests in the Atlantic Ocean and the Gulf of Mexico. However, he also noted that there were no statistical differences in catch rates between different areas within the Gulf of Mexico. Because no data was given to characterize the habitats used in this test, it is difficult to draw any conclusions from this data. Poiner et. al., (1990) compared catch rates between the North Australian prawn fishery and the US shrimp fishery and found comparable catch rates (between the US Gulf of Mexico and Northern Australia). To the best of my knowledge, there is no study that attempts to compare geographic differences in TED performance based on habitat or geographical area.

Dr. J. Frazier:

5.233. To work properly, TEDs must be adapted for the local conditions where they are to be used, taking into account: fishing gear, fishing technique, substrate type, bottom cover, and water depth, among other things. These sorts of adaptations are not unlike the modifications that fishermen have made to gear to be able to fish in different conditions. Sr. Randall Arauz, who has been working on TEDs in Costa Rica for the last four years, stated: "with proper modifications of the TED technology and fishing practices, together with scientific documentation, research to make TEDs work efficiently under virtually any fishing conditions, as we have proven in Costa Rica". (Arauz, 1997).

5.234. There is great variation in the fishing grounds of the Gulf of Mexico, Caribbean and East Pacific, where TEDs are being used. Fishing grounds of the Indo-Pacific are likely to be both similar and divergent from fishing grounds in the Americas. However, the principle of TED modification for local requirements is the same. Indeed, Thai gear specialists have carried out tests and devised two unique designs, the Thai Turtle Free Device (TTFD) and the Thai-Ku (Bundit et. al., 1997). Under the aegis of SEAFDEC, Thai fisheries officers have been disseminating this gear in other countries of the region (SEAFDEC, 1996; 1997a; 1997b; 1997c).

5.235. It must also be pointed out that the gear specialists of the Pascagoula Laboratory of the National Marine Fisheries Service have decades of experience in devising, modifying and testing TEDs. They have been actively training people as well as distributing gear and information in many different countries, in workshops both in the United States and abroad, since 1983 (see Appendix 2 "Transfer of TED Technology" contained in Annex II of the Report).

Mr. M. Guinea:

5.236. For TEDs to be effective in reducing the mortality of sea turtles, they have to be functional in the fishery. Part of their functionality is the willingness with which they are accepted by the fishery. This involves considerable modification and experimentation not only to provide the previously mentioned sense of ownership, but also to convince operators of the usefulness of new technology. Australian trawl fields are considerably different to the trawl fields of the Gulf of Mexico and the Caribbean Sea.⁴¹¹ Options such as bottom or top opening for the removal of sponge or sea turtles respectively, have to be explored. The set angle of the TED and the position in the net have to be modified for the nature of the benthic habitat and the species of sea turtles and their size as well as the nature of any other bycatch. There needs to be considerable modification and trials before TEDs or any other bycatch reduction device, e.g., fish eye etc., is accepted by the fishery.

5.237. From trials in Australia (Robins, 1995; Mounsey, 1995) and Thailand (Chokesanguan et. al., 1996), it is possible that the environmental conditions vary greatly between the localities. This is reflected in the performance and unacceptability of the unmodified TEDs.

Mr. H.-C. Liew:

5.238. Not able to comment.

Dr. I. Poiner:

5.239. Monitoring of TEDs and other bycatch reduction devices (BRDs) in tropical northern Australia (Brewer et. al., 1995, 1997; Robins-Troeger et. al., 1994) under commercial conditions demonstrate that TEDs performance depends on the nature of the sea bottom. Different areas require different types of TEDs. These results should be transferable to other parts of the Indo-Pacific waters. What these results show is that if TEDs are to be used they need to be selected and adapted to local fishing conditions and approaches to fishing. TEDs that are effective in the Gulf of Mexico and Caribbean Sea may not be appropriate for Indo-Pacific fisheries.

Question 4: Conservation measures on nesting grounds

4(a) What is your assessment of conservation programmes focusing on protection of eggs and hatchlings? Are there examples where these programmes have been proved effective in restoring a population of sea turtles, or in maintaining it at a sustainable level? Are their regional differences in this regard?

Dr. S. Eckert:

5.240. In my response to question 2(b), I have provided some assessment of sea turtle conservation methods. Of greatest importance to any sea turtle conservation programme is to address the problem that led to the "endangered" status of the stock or population as a first priority in conservation (Frazer, 1992). To the best of my knowledge, there has never been a case where enhancing reproductive output has been able to mitigate for juvenile and adult turtle mortality. Thus, while nesting beach programmes are important and useful in mitigating for historical over-harvest of eggs, I cannot advocate this technique as a mitigation for incidental mortality associated with fishing. The reason for this stance in sea turtle conservation is obvious when you consider what it means in terms of sea turtle population dynamics. Due to the low survival rate of sea turtle hatchlings and juveniles, one large juvenile or sub-adult turtle represents many hundreds (or thousands of eggs). Thus, for each turtle killed incidentally many eggs must be hatched and released over and above those that would survive naturally. With the highly depleted status of most nesting populations it is simply not feasible to increase hatch production at the levels required to mitigate for even small levels of incidental mortality.

5.241. An example of where protecting only nesting stocks as a conservation strategy has failed is for the loggerhead nesting stocks of North Carolina, South Carolina, and Georgia. This stock constitutes a unique nesting assemblage and is separated genetically from the larger Florida nesting population (Bowen et. al., 1993). The index nesting beach for this stock is on Little Cumberland Island. This is the best studied loggerhead nesting population in the world and thus much of our information on sea turtle population dynamics is based on this data from this beach (Frazer, 1983; Frazer, 1985; Richardson, 1978; Taylor, 1993, Bell and Richardson, 1978, Bowen et. al., 1993, Frazer and Richardson, 1985, Frazer and Richardson, 1986, Hillestad et. al., 1978, Frazer and Richardson, 1985b, Hillestad et. al., 1979, Kraemer and Richardson, 1979, Mrosovsky et. al., 1984. Stoneburner et. al., 1982, Richardson et. al., 1976a, Richardson et. al., 1979b, Richardson et. al., 1976, Richardson, 1978, Stoneburner and Richardson, 1981, Richardson, 1982, Richardson, 1992). Little Cumberland Island has provided an interesting test of nesting beach conservation, because prior to the initiation of nest protection in 1964, virtually 100 per cent of the nests were consumed by raccoons. After the initiation of protection by beach patrol and maintenance of an on-beach hatchery, almost 100 per cent of the eggs have been protected. Yet, between 1964 and 1991 the population declined approximately 65 per cent (NRC, 1990, Richardson, 1992). Accounting for a 20 -25 year delay in nesting population response due to maturity time in loggerhead (Frazer, 1983), nesting population numbers should have begun to rebound if egg protection was an appropriate conservation tool, and they have not. Similar trends in nesting have been seen in North and South Carolina. Such lack of recovery has been due to the mortality associated with shrimp fishing on the Atlantic coast (NRC, 1990).

Dr. J. Frazier:

5.242. As stated in earlier responses, the protection of eggs and hatchlings of sea turtles is essential for the long-term health of the population; without recruitment into the population from eggs and hatchlings, it will gradually "die of old age". However, "focusing" on protection of just eggs and hatchlings, and not reducing mortality in older animals will be doomed to failure (see my answers to questions 1(b), 1(c), 2(c) and 2(d)). It makes little sense to invest time, money, materials and effort protecting just eggs, only some of which will hatch, and fewer of which will grown into young turtles, if those turtles are under high risk, and their chances of survival are very low. Because egg protection produces rapid, tangible results (i.e., hundreds of scrambling baby turtles, just two months after eggs are laid) it provides quick and attractive rewards for conservation activities; furthermore, it is much simpler and less expensive than protection of animals in the sea or marine environments. Hence, as a rule egg protection attracts more attention than the more difficult, complex and time-consuming tasks of protecting turtles at sea. For decades, egg protection and head-starting (captive rearing) programmes have been carried out with the best of intentions, and the rapid, tangible results have consistently been activities that have been reported as evidences of success - routinely taken advantage of by politicians. However, over the last decade sea turtle conservationists have come to realize that concentrating on nesting beaches has routinely taken attention away from other, more needy activities (e.g., Mortimer, 1990; 1995; Suliansa et. al., 1996). As explained in several sea turtle conservation strategies (World Conference on Sea Turtle Conservation, 1982; IUCN, 1995; 1996; in press), the priority is integrated management and conservation.

Mr. M. Guinea:

5.243. Conservation measures devoted to eggs and hatchlings have been successful for some breeding units of some species e.g., olive ridleys in Orissa. Mortimer (1995) elegantly distils the evidence for protecting eggs and adults. Each strategy has its individual strengths and possible scenarios for delaying such conservation measures. Conservation involving coastal communities will gain popular support and have a greater chance of being maintained, than a piece of legislation which affects only a small proportion of the population i.e., fishers, or companies and which is out of sight of the community. Like fishing, conservation can become an industry, if properly structured.

5.244. The conservation measures employed by Malaysia (Liew, 1997) and Thailand (Monansunsap, 1997) appear to be successful. The measures have community support and sponsorship from a number or organizations.

5.245. There will be regional differences regarding the effectiveness of conservation programmes focussing on protection of eggs and hatchlings. These will be based on the culture of the area and the socio-economic climate that prevails as well as depending on the breeding unit to which the sea turtles belong. The sea turtles may display plasticity in life history strategies which may be confounded by differing pivotal temperatures, sex ratios and stable age structure. Each breeding unit will respond in a similar manner but at a differing rate to identical conservation measures. Conservation measures that protect nests or eggs will make a significant contribution to the continued survival of the breeding unit.

Mr. H.-C. Liew:

5.246. Protection of eggs and hatchlings are important to ensure the continued sustainability of sea turtle populations. However, they must be conducted properly and in tandem with other conservation strategies determined for each locality. Where possible, eggs should be incubated in natural nests in situ and hatchlings immediately released on hatching and not retained for long periods as still practised in some countries. There are a few examples where turtle populations have shown apparent recovery or sustained where conservation efforts focus on protection given to turtle nesting beaches, their eggs and hatchlings. However, such recoveries were only apparent after many years of strict conservation measures due to the long periods turtles need before they mature after emergence as hatchlings. Some of these include the green and hawksbill populations in the Turtle Islands of Sabah, Malaysia; the leatherback populations of South Africa, the leatherback population in St. Croix and Surinam and the green turtles of the French Frigate Shoals, Hawaii.

Dr. I. Poiner:

5.247. All sea turtles populations in the Indo-Pacific region including southeast Asia are severely depleted and/or are subjected to over-harvest (legal and illegal) and/or excessive incidental mortality. Some countries (e.g., Malaysia and Thailand) have instigated management measures to prohibit or control egg harvests as a conservation measure but there is no evidence of recovery of any of these populations (Limpus, 1997).

4(b) Considering the long timeframe some species of sea turtles need to reach reproductive age, is it still difficult for biologists to anticipate the effects of the more recent programmes on the populations concerned or is it now possible to assess whether egg protection methods are capable of ultimately preventing marine turtle extinction and, if properly implemented, will in fact do so?

Dr. S. Eckert:

5.248. In previous answers I have touched on the disadvantages of using nesting counts for determining population trends. The same is pretty much true for understanding the effect of conservation actions or nesting beach perturbations. Due to the long time it takes for turtles to reach reproductive maturity, it will often take a generation time (25-50 years) to see the fruits of such efforts revealed on the beaches. However, as noted in the example provided by the Little Cumberland Island loggerhead study, we are reaching a point in some projects that enough time has elapsed for the effects of nesting beach conservation actions to be determined. This, combined with improvements in our sea turtle population models (for a review see Chaloupka and Musick, 1996) is indicating the need for a balanced conservation approach and illustrating the fallacy of focusing only on nest beach conservation as a means to restore depleted turtle populations. Finally, consider this illustration. If it takes 1000 eggs to produce 2 adult turtles (Frazer, 1983) (this is probably a minimum estimate) and only slightly less for 2 sub-adult (stage 3) turtles, then for every turtle we want to replace we must hatch just under 500 eggs. If there is a relatively minor incidental shrimping mortality of stage 3 turtles, 100 as an example, then just under 50,000 eggs will need to be protected to mitigate for the fishery mortality. Further, this 50,000 has to be in excess of what is already being produced on the beach, since the current beach production is likely not enough to maintain the population (based on the assumption that most population are already in decline). From this example it can be seen why it is so difficult to use nesting beach conservation as a mitigation for fishery mortality, and why such an approach simply will not work, as was demonstrated at Little Cumberland Island.

Dr. J. Frazier:

5.249. As yet, no species of sea turtle is known to reach reproductive age in less than 10 years. Green and loggerhead turtles, for which the best information on growth rates is available, are generally thought to require about 30 years to reach maturity. The long time needed to reach maturity means that only long-term data will permit a true understanding of trends in the population. As was explained earlier, turtle populations are evaluated by counting nesting females, nests, or eggs. These counts represent only a small segment of the total population and there is tremendous variation in clutch size, number of clutches per female, inter-nesting intervals, and nesting activity from year to year. Hence, estimating population size from what is seen during a nesting season on a beach has clear limitations.

5.250. It does not matter whether the conservation measure is egg protection or use of TEDs; it takes years of systematic information to be able to decipher the trends in size of a sea turtle population. Because the animals have complex life cycles and need a long time to maturity, they are subjected to many different sources of mortality over long periods of time. It is most prudent to carry out integrated conservation, providing a variety of measures for habitat protection and reducing mortality. This strategy of integrated conservation for sea turtles has been adopted in numerous international fora, for well over a decade (e.g., World Conference on Sea Turtle Conservation, 1982; IUCN, 1995; 1996; in press).

Mr. M. Guinea:

5.251. Egg protection strategies have been employed for less than one sea turtle generation. The nesting beaches are the only points of reference to measure the success of such conservation measures. Ideally if the developmental habitats were known, then an increase in relative abundance of sea turtles may be demonstrated, but developmental habitats may, in fact, be defined more by carrying capacity than the absolute abundance of sub-adult sea turtles. Egg protection measures for olive ridleys in India and green turtles in Malaysia appear to be successful. The relative significance of egg protection is difficult to determine without knowing the other threatening processes impacting on the breeding unit.

Mr. H.-C. Liew:

5.252. Egg protection methods alone is not sufficient especially if other threats are still present and have significant impacts on the population. For populations, if any, where egg exploitation is high while the threats from the other factors are negligible, then egg protection methods would suffice.

Dr. I. Poiner:

5.253. Sea turtles are very long lived animals that mature at a relatively late age (ca 30 to 50 years). The interval between breeding events is also very extended (ca 5 to 15 years, depending on the species). While many eggs are produced, and egg predation is high, natural mortality of sub-adults and adults is probably relatively low. Because recruitment to the adult population is low, recovery from low population number (if non-natural sources of mortality have been removed) will be slow, and there is no clear documented cases of recovery in the world. Our only estimates of recovery times come from modelling studies.

5.254. Crouse et. al., (1987) and Crowder et. al., (1994) used a stage-based model for United States loggerhead sea turtles to conclude from sensitivity analysis that reducing annual mortality of large juveniles, sub-adults and adults was most important to ensure long-term viability of the stock and suggested egg protection programmes are ineffective. Modelling studies of loggerhead turtles in the United States following the introduction of TEDs which should have reduced mortalities suggest recovery will be slow e.g., 70 years or more would be required for the simulated population to increase by an order of magnitude (Crowder et. al., 1994). This was because of the high relative reproductive value individuals at these stages/ages in the model. However, other models by Somers (1994), and Chaloupka and Limpus (MS) concluded that protection of eggs/hatchlings would also have a major impact on long-term stock viability but give no estimation of recovery times.

Question 5: Migratory patterns

5(a) What are the migratory patterns of the various species of sea turtles mentioned above? Are the migratory patterns similar in different regions of the world? In particular, do sea turtles migrate seasonally -and if so, are those seasons clearly defined- or do they migrate all year round?

Dr. S. Eckert:

5.255. Despite many years of sea turtle flipper tagging, and an increasing number of satellite telemetry studies, our understanding of the migratory movements for sea turtle populations are still very limited. In particular, we know almost nothing of the migratory movements of juvenile turtles during early development or even after they have settled in coastal habitats. As noted earlier, we have only one clear pattern of migration resolved for the loggerhead during this part of its life phase, yet even for that species our sample sizes are small and we know nothing of the timing of the migration. Further, virtually all other migration information is associated with mature female turtles.

5.256. However, some hints at what sea turtles are capable of can be gleaned from recent studies. Early in this document, I described something of what my own satellite telemetry studies are telling us of the migratory capabilities of the leatherback. They have demonstrated a capacity to travel in excess of 11,000 km in a single year, and all indications are that they make north/south migrations annually. In the Pacific it is likely that mature female leatherbacks circumnavigate the Pacific Ocean during the 2 or 3 years between nesting seasons. My current hypothesis for the movement of leatherbacks in the Pacific is that females from the 2 major colonies (Mexico/ Central America and Irian Jaya/Solomon Islands) as well as the minor colonies (e.g. Malaysia) distribute into a clockwise migration of the Pacific Ocean with turtles stopping to feed in areas of high productivity. What I have shown for the Atlantic Ocean is that leatherbacks are very adept at knowing where to anticipate areas of high food availability and will readily migrate great distances to access those resources. Supporting data for the theory of the migration cycle of Pacific leatherbacks is currently being gathered by satellite telemetry and DNA stock assessment, and thus far the hypothesis is supported. Significantly, this make the leatherback a species that shares many government jurisdictions. It is highly probable that Malaysia, Thailand and the United States all share responsibility Pacific leatherbacks during a single nesting migration.

5.257. Green turtle females have well-documented long distance post-nesting migrations. Most of the data is from tag returns, which are somewhat problematic when trying to understand migratory cycles. Such data usually only represents a one-way trip or a stop along a possibly longer journey, because invariably the turtle is killed and thus the tag is recovered. Most green turtle post-nesting migrations are between 1,500 and 3,000 km (Kolinski, 1991, 1992, Meylan, 1982, Mortimer and Carr, 1987, Pritchard, 1973, Balazs, 1976). Even more valuable has been a recent plethora of satellite tracking studies of female green turtle post-nesting migrations, though in most cases the duration of tracking has been too short for the determination of annual movement patterns (Balazs, G.H. 1994, Balazs et. al., 1994, Liew et. al., 1995, Luschi et. al., 1996).

5.258. Migrations or movements of juvenile or foraging green turtles are not as well investigated. It is likely that the species exhibits the same planktonic existence of other species for the first years after hatching. Balazs (1976) proposed for the Hawaiian green turtle nesting population at the French Frigate Shoals that hatchling probably disperse to the west, though how far and how long is unknown. Generally loggerhead females also make long post nesting migrations in excess of 1000 km; they are generally shorter than what is documented from green turtles (Bell and Richardson, 1978, Hughes 1974, Meylan, 1982, Margaritoulis, 1988). Developmental migrations of juvenile loggerheads is probably better understood than any other species. In both the Pacific and Atlantic, hatchling loggerheads circle their respective ocean basins during their first years of life (Carr, 1987, Bowen et. al., 1995) and return to the coast they were hatched on to settle. From those foraging area they will make migrations to their natal beaches to nest. Early literature on the migration behaviour of hawksbill suggested that they were relatively sedentary and did not make long migrations (Bustard, 1979). Meylan et. al., (1997) summarizes hawksbill migrations and concludes that they migrate comparable distances to green or loggerhead turtles. The longest migration was 2,925 km, with a large number in excess of 1,000 km. Meylan et. al., (1997) also summarizes studies on hawksbill juveniles both in the Caribbean and Pacific that suggests that juveniles probably remain in the same habitat or area for many years and may only move to other developmental habitats as they grow.

5.259. Annual migrations for most species are only poorly documented or understood. I have noted where it appears that Atlantic leatherbacks make annual north-south migrations. There is also a seasonal presence of leatherbacks at various areas along the US East and West Coasts (Shoop and Kenney, 1992, Stinson, 1984). Stinson (1984) also documented the seasonal abundance of loggerheads, olive ridley and East-Pacific green turtles on the US West coast, and concluded that these species follow the 18° C isotherm. Morreale (1990) has also indicated that there is a strong correlation between temperature and presence of Kemp's ridleys and loggerhead sea turtles in Long Island sound and the coastal waters of New York. With the exception of reproductive migrations and leatherback, migratory movement of most species seem to be temperature driven. Given the relatively warm waters of Malaysia, Thailand, India and Pakistan it would not be expected that resident turtle population would exhibit annual or seasonal migrations in those countries.

Dr. J. Frazier:

5.260. The individuals of a population of sea turtles, that nests on a nesting beach, are likely to have migrated to a variety of feeding grounds. Leatherbacks make the largest movements, while in general hawksbills migrate the shortest distances. Olive ridleys take up a pelagic existence, at least in the Eastern Tropical Pacific (Plotkin et al., 1995; 1997). In any event, information on "migratory patterns" is very incomplete, and we are only beginning fully to appreciate the degree to which sea turtles move around the oceans. It has been known for decades - even centuries - that sea turtles migrate over vast distances; Brongersma (1972) compiled hundreds of records from the Atlantic coast of Europe (where sea turtles do not breed), the first of which was from the 1300s. Today, with the exception of the Australian Flatback, there are records of every species of sea turtle crossing ocean basins: viz loggerhead (e.g., Brongersma, 1972; Dodd, 1988; Bowen, 1995; Bowen and Karl, 1997); green (e.g., Brongersma, 1972; Bowen, 1995; Hirth, 1997); leatherback (e.g., Brongersma, 1972; Pritchard and Trebbau, 1984; Eckert and Sarti, 1997); hawksbill (e.g., Marcovaldi and Filippini, 1991; Meylan et. al., in press); Kemp's ridley (e.g., Brongersma, 1972; Pritchard and Marquèz, 1973); and olive ridley (e.g., Pitman, 1990; Plotkin et. al., 1995). The absence of information simply is not evidence with which to conclude that turtles do not migrate. New scientific tools such as genetic analyses (Bowen, 1995; Bowen and Karl, 1997) and satellite transmitters, are providing valuable new insights into the question of sea turtle migrations.

5.261. Generally nesting is seasonal, although in some populations nesting may occur through the year, or much of the year, with a peak in activity at a certain time of year. The migrations for which sea turtles are famous occur between nesting grounds and feeding grounds, and between feeding grounds and nesting grounds. When nesting is seasonal, these migrations will also be seasonal However, some turtles may move over large areas between nesting seasons, as seems to be the case with the leatherback. In addition to the migrations of breeding adults to and from nesting grounds, the immature turtles disperse over vast areas of the oceans, apparently taking up temporary, sequential residence in various "developmental habitats" as they mature. These movements are often referred to as migrations also, although they generally are thought not to involve return trips. Little is known about these "immature migrations".

Mr. M. Guinea:

5.262. All sea turtle species except the Australian flatback undergo extensive ocean migrations during their life. Hatchlings, after they leave the nesting beach, spend a long period, possibly a decade, at sea. In response to an unknown trigger they take up residence in an inshore feeding area. Several of these inshore feeding areas may be used as the turtle grows to maturity. Adult sea turtles are thought to migrate to nesting beaches and back to their feeding areas using the magnetic field of the earth (Lohman et. al., 1997). They are capable of crossing deep water (>2,000 m) on these migrations. The migration may be independent of the coastline of alternatively may be along the coast. The return path appears to be essentially the same route. This is done individually without any social facilitation of others or herding within the breeding unit.

5.263. The migrations are similar but at the same time they are uncoordinated. Reproductive migrations are in response to conducive nesting conditions developing in the coming months at a rookery, possibly over 1,000 km from the feeding area. In mixed feeding grounds, turtles from one breeding unit may leave at a different time and in a different direction to those of another breeding units. Some turtles may not breed that year and will remain resident on the feeding area.

5.264. The migration of a breeding unit will be seasonally to the rookery at the beginning of the breeding season and away from the rookery at the end of the nesting season. This largely goes unnoticed, except where the sea turtles pass through straits, cross shallow water or around geographic projections. This seasonality of green sea turtle migration through the waters of the Torres Straits, north of Australia, have been exploited for centuries by the indigenous islanders (Johannes and MacFarlane, 1991).

5.265. The timing and intensity of the migrations through the straits varies with the number of sea turtles nesting that season and the number of males that migrate to the breeding areas. Males leave the breeding area early in the nesting season and return to their feeding ground. Within the nesting area, movements by the female will be relatively short, 2-20 km, and coincide with movements to the nesting beach to lay the clutch and return to the offshore refuge while awaiting the maturation of the next clutch. After her final clutch the female returns to her distant feeding grounds.

Mr. H.-C. Liew:

5.266. Much has yet to be learned about sea turtle migration. From various evidences gathered, sea turtle hatchlings do not seem to migrate but head offshore on entering the sea to drift and be carried by oceanic currents for about 5-7 years. The oceanic currents may carry some of these hatchlings thousands of kilometres along oceanic gyres and may be transported across the Pacific or Atlantic Ocean. On becoming juveniles, only the leatherback will continue this ocean-pelagic existence while the other species would begin to work their way towards shallower waters. When they find suitable feeding areas, they would establish these areas as their foraging or feeding grounds, where they may remain for many years. The range of these feeding grounds may vary between species and between turtles. As to whether they have multiple distant feeding grounds and migrate amongst them is not known. The most significant migration that sea turtles perform is their migration between feeding grounds and nesting grounds (see answer below).

Dr. I. Poiner:

5.267. Sea turtle breeding stocks usually comprise multiple rookeries within a region while foraging areas and developmental habitats comprise a mix of turtles from several genetically distinct stocks (Bowen et. al., 1995; Broderick et al., 1994). The breeding adults usually migrate relatively long distances from the foraging areas to the traditional breeding rookeries. I will illustrate this life history pattern using Australian loggerhead (Caretta caretta) and green (Chelonia mydas) turtle populations (Limpus 1997).

5.268. The Australian nesting populations of loggerhead sea turtles are genetically distinct from those in other countries and within Australia there are two genetically independent breeding populations. Breeding occurs in the summer months for both populations. Breeding females migrate up to 2,600 km from feeding areas to aggregate at traditional nesting beaches (breeding males have not been studied). In eastern Australia, females migrate from northern and eastern Australia, Indonesia, Papua New Guinea, Solomon Islands and New Caledonia. In Western Australia, recorded migrants come from Northern and Western Australia and Indonesia. Mean remigration period is 3.8yr. At the completion of the breeding season the female returns to the same feeding site as she occupied before the breeding migration.

5.269. The green turtle has a global distribution in all oceans with nesting occurring mostly in tropical areas. The Australian nesting populations are genetically distinct from those in neighbouring countries. Within Australia there are at least 5 genetically independent stocks. In addition, there are green turtles that feed in Australia that are part of stocks that breed in other countries: Indonesia (Java), northeastern PNG, New Caledonia and Pacific Mexico. Breeding occurs in the summer months for the east coast and west coast populations and during the winter for the northern populations. Breeding females and males migrate up to 3000 km from feeding areas to aggregate at traditional nesting beaches. In Eastern Australia, females migrate from Northern and Eastern Australia, Indonesia, Papua New Guinea, Vanuatu, Fiji and New Caledonia. In Western Australia, recorded migrants come from Northern and Western Australia and Indonesia. Mean remigration period = 5.8 years for females and 2.1 years for males. At the completion of the breeding season the adult returns to the same feeding sites it occupied before the breeding migration.

To Continue With Chapter 5.270

⁴¹¹ US Embargo on the Import of Wild-Caught Shrimp, Submission by Australia to the US Secretary of State in support of its request for certification under Section 609(b), April 1996. See above paragraph 4.2.