117. In question number 3 relative to the leatherback population on St. Croix. The St. Croix project was indeed a nesting beach project whose goal was to try to enhance the leatherback population nesting at St. Croix, to the use of conservation measures on the beach, including egg relocation and active protection on the beach. This is a programme that my wife and I started back in 1981/1982 and one of the reasons that this programme was undertaken is that there was an identified source of mortality. There was an issue of possible beach development and then there was an issue of identified mortality of eggs due to poaching and, even more significantly, due to loss of the nests due to erosion. In the first season down there, we documented 65 per cent of the eggs were lost annually simply to natural erosive processes. So, our response as a conservation measure on that beach, was simply to relocate eggs to safer areas and thereby boost egg production substantially over what had been historical levels. Now, there is the current nesting population size, which is somewhere in the neighbourhood of a 100 females, and one of the questions says, would you say protecting nesting females on the nesting beaches and protecting eggs undergoing incubation had contributed towards the build-up of the nesting population in St. Croix? At this point, it is still too early to say. The population has been showing a slow but sure increase over the last few years and we are very encouraged that it may be providing support to the population. After all we have been doubling egg production on that beach since about 1982, so that has been a significant increase in the number of turtles being put into the sea.

118. Question 4 is actually a comment not a question so I will leave it at that. Do you want me to get into all the expert questions as well here? I will just race through them. In your expert opinion would trade prohibition on the import of shrimps to the US save sea turtles from shrimp trawlers and extinction? [Question 5 by Malaysia] Certainly. I think it would cause a net benefit to sea turtles. Any time you can reduce the number of turtles killed in shrimp trawlers you are going to benefit the populations, its relative to the trade prohibition. Let me just briefly describe a situation that I witnessed in Trinidad, a country in which I maintained an ongoing research programme for about the last five years. I am going to have a difficult time with the factual information here because, again, I am just pulling this off from my own experience and memory, I am sure this can be documented. A couple of years ago Trinidad was embargoed for failure to maintain shrimp trawler TEDs in their shrimp trawler fleet. The effect in Trinidad was extraordinary. I was down there just after this embargo had gone into place and the price of shrimp had dropped out of the market. The local market in which you could barely buy shrimp in previous years was flooded with shrimp and the value had gone way, way down. Clearly the impact to the shrimp fishery had been economically very significant. The response by the fisheries management agency in Trinidad was also extraordinary. In previous years, there was a bit of a jurisdictional split, turtles were jurisdictionally managed by their Wildlife Section under the Department of Agriculture and the Fisheries Department is a separate department and they had virtually ignored any of the requests from the Wildlife Section to support sea turtle conservation or minimise take in the fishery. Subsequent to the embargo the Fisheries Department immediately called the Wildlife Section and requested a biologist to come with them and advise them on what they should be doing. So Wildlife sent one of their sea turtles biologists over to work with both the shrimping fleet and the Fisheries Department and TEDs were put into place in a matter of months and it was an extraordinarily fast turnaround. So, relative to the effectiveness of the embargo in supporting the conservation of green turtles, my experience is that it was extremely effective in Trinidad and it was just a matter really of the Fisheries Department recognizing the need for this and taking steps to have the process occur and believe me, it did not take very long at all for them to get out there and start making sure that these devices were being utilized.

Chairman

119. Can I just confirm that your answer to that question is that a US trade prohibition would by itself, on its own, without any other measures, save sea turtles from shrimp trawlers and extinction? I just wondered whether that was the question you were really answering.

Dr. Eckert

120. Yes, I'm sorry I didn't see the "by itself" and in response to that, no. Certainly, what we have talked about extensively yesterday is that you need to take a multiple approach to the conservation of turtles addressing specific problems. I guess what I was interpreting this to mean was that, would it help to save sea turtles and the answer to that question would be, yes it would, as opposed to by itself, no it would not.

121. "What is the acceptable recognized method of determining population size of breeding units? [Question 6 by Malaysia] I am not sure that there is a complete consensus on this, but what I have stated in my response to the Panel was that the general consensus among biologists, at least among the biologists that I have spoken to and worked with, are that you need to monitor the population for a number of years. You need to monitor the number of nesting female for a number of years before you will be able to determine the trend. What I have suggested is that the time period needs to be approximately 3 migration cycles. The average re-migration cycle is defined as the number of years that the average female within a nesting aggregation takes between nestings. Now, in Australia, if I understand Colin Limpus' work, that is often 5 to 7 years, in the Caribbean where I have substantially more experience, it is 2 to 3 years. So it can vary depending on the individual region as to how long you need to monitor that population. It also depends somewhat on the species, in that olive ridleys and Kemp's ridleys often have annual nesting events and, thus, the annual average re-migration rate can be a single year thereby, (by the way I am using the calculations), 3 years might be adequate to indicate a trend for those nesting populations. Having said that, you have to realize there is a tremendous amount of variation in this depending on other environmental external factors, such as southern oscillations. Colin Limpus' work on green turtles has specifically pointed that out that the El Ni_o or the southern oscillation in Australia has a large impact on the re-migration rate of those animals and that their re-migration rate can change based on what's going on out there in their particular world, particularly relative to food availability. That is actually somewhat significant over the last few years, it is going to be interesting to see what has happened in the last 10 years in the green turtle populations of the Western Pacific because we have seen an increase in the rate of Southern oscillation events over the last 10 years. That may very well explain issues relative to a shifting, there may be a shifting in the re-migration intervals and in those populations. The best folks to speak to that are obviously the Australians who work on the species and have been able to document this very well.

122. "Please tell us your views of the concept of unit stocks or populations of breeding units and sea turtles". [Question 7 by Malaysia] We have worked this pretty heavily over the last day or so. There is no question, in my opinion, that the identification of unit stocks is an extraordinarily important management tool for marine turtles. It will be the tool by which we can apply proper management to sea turtle populations in the future. As I was the Chairman of the US Pacific Sea Turtle Recovery Team, that developed the recovery plans for the Pacific for the United States, and in those plans, we were all in uniform agreement that one of our top priorities is to see the stock ranges of all sea turtle nesting aggregations identified. However, what I would also like to say, is that currently we do not have enough information on those stocks and on those stock ranges to use it as a management tool. It is simply a goal that we are working very hard towards obtaining and it will be a number of years before I feel that it is adequately documented to become a useful management tool. This is why I have advocated an approach more of, when you see a problem with the sea turtle population due to, say, incidental take in this case, that you address that problem irrespective of individual stock, quantities of individual stock status. We simply do not have enough information on individual stock status to warrant the approach that says we do it stock by stock. So, if turtles are taken in shrimp trawling, I feel very strongly that you must address that issue immediately and not wait another 10 or 20 years until you can identify whether it's important to that particular population or not. There is enough evidence out there on the impact of incidental take on turtle populations, in populations that have been better studied, to indicate that the problem probably exists globally. "When studies on a particular sea turtle population are made, will the results applied to populations being studied?" [Question 8 by Malaysia] I think I have just covered that question as well. "Would you say that there are sea turtle populations in the world that are quite healthy?" [question 9 by Malaysia] I have a very difficult time saying that. Based on my experience it would be very difficult to say that there is any population of sea turtles that is quite healthy. You have to sort of look at the historical context of a question like this and the question sort of comes back to, do you want to back to 200 years, 500 years or 1 million years, as to how you answer a question like that. If you go back a million years obviously no, if you go back 10 years, there is some encouragement in a few populations that there may be something other than remnant populations. Otherwise I cannot be encouraged with our current status, even in the last 30 years that we see any population that could be considered healthy.

123. Would I agree that "loggerheads are the most vulnerable to shrimp trawling?" [Question 10 by Malaysia] The question "are they most vulnerable", all sea turtle populations are vulnerable to shrimp trawling. In the United States there has been some discussion about whether leatherbacks can be subject to shrimp trawling. In the United States and the coast of Georgia, North Carolina, there were large numbers of leatherbacks killed in the shrimp trawler a number of years ago and that number seems to fluctuate quite a bit probably based on what the migratory pathways the leatherbacks are on at the time. Basically, my feeling is that if you have a habitat in which any species of sea turtle and shrimp trawling occur you will have significant mortality of sea turtle populations. I think this is the way in which this thing has to be examined. If there are turtles there and shrimp trawling there, there will probably be mortality to the sea turtle populations. Thank you.

Mr. Chairman

124. Do you wish to address the US questions as well?

Dr. Eckert

125. Let's look at question 1. ² I think a lot of this has been pointed out fairly clearly in some of our previous discussions and previous answers. Jack Frazier has brought up Indian Ocean turtle populations and the declines in some of those, I hope I am not speaking out of term for Jack. Certainly, the leatherback turtle population of Malaysia is not in recovery. The green turtle and hawksbill peninsular populations, I believe, are also in decline for the hawksbill of Malaysia. I am not sure what the status is of turtles in Thailand other than what we have written down. I am going to pass on with that question, it's probably more than I can try to answer at this time. I am going to let Mr. Guinea deal with Question 2 because it is really addressed more towards him. I have stated in my response that I think that the annual documented mortality of 5,000 turtles nesting at Gahirmatha is not minor. It probably represents quite a bit more than that. I think that the Murphy's study on what percentage of the turtles actually make it to shore after being killed in a drowning incident suggest the number is probably substantially higher than 5,000. But shy of getting into a numbers game on this thing, which I am not really equipped to do right here, I don't want to go any further than that. Question 3, "... does existence of all these threats to turtles make it more important or less important regarding turtle mortality in shrimp trawlers?" I guess, what you have to say about this is, I get back to this point. Mortality in shrimp trawling is a documented fact in a number of areas and in other areas it simply isn't documented. There are a few areas where it simply isn't documented. In those areas where it is documented, we obviously need to address the shrimp induced mortality on turtles. Areas where it is not documented, we need to use our best judgement to indicate rather there are turtles and trawlers working in the same habitats and if there are, one has to assume that mortality is going to be associated with that. That's my own opinion on that. "Do experts agree that TEDs that have been properly installed and used reduce the mortality of sea turtles in shrimp trawl nets?" [question 4 by the United States] Yes, I don't think there is any question there. "If all the world shrimp fleets used TEDs would it contribute to reduction, to the threat of sea turtles?" [question 5 by the United States] Yes, in my opinion it would. Any time you can introduce something that reduces the mortality of wild turtle populations, which are in trouble obviously, it is going to help.

126. Time and area closures: this is more of a management type question and it's a difficult one to answer, just simply from a science perspective, it is one that managers need to address but there have been problems with time and area closures in the US experience that I am familiar with. Primarily, I think that one of the most significant sources for the problem are that if you are using a time closure or a time and area closure; we are closing a particular area for a certain time. Very often it is difficult to respond to the turtles changing in their patterns, of either movements or migrations. We saw this in the US when, for example, leatherbacks started moving in closer to the coast than they had in previous years. During times of the year when turtles normally would not necessarily have been an issue in those waters, the management agencies had a difficult time responding by putting the regulations into place, by banning shrimping in the area. So, unless you close an area pretty much all the time, which is often very unpalatable to the fishing industry, using a time and area closure is often very cumbersome. I think in the case of the US, and I don't want to try to speak for the Fishery Services on this because this is their business, but it does seem to me that after a number of years of experience of trying to do time and area closures they realized that it was far simpler, from both an enforcement perspective and a management perspective, to simply require turtle excluder devices all the time so that they didn't have to try to deal with the vagrancies of the biology of the animals which we well pointed out that we don't understand very well. It also meant that they did not have to be confusing the fishing industry with closures on, closures off, closures on and continuous monitoring of what was going on out there relative to the turtle take. I think time and area closures can be a very very difficult from a management perspective and not necessarily well received by the fishing industry because it puts them in a source of confusion to know whether they will be able to fish or not. Most significantly when they are done they cannot be static they have to be plastic. All of a sudden you have got turtles in the water, you must call up the fleet and tell them to shut down. If you are not willing to do that, then time and area closures will not be effective for turtle populations, they simply move around too much.

127. Let's go to Question 7, "could Dr. Eckert please elaborate on his statement of 'seasonal migrations will not be expected in regions of warm waters'". I made that statement because in the discussion on migration in sea turtle populations. What we see, and it is based partly on my experience on the US West Coast and what I know about information from the US East Coast. One of the primary things that we look at, when we look at the presence or absence of turtles particularly on the West Coast, is water temperatures. There have been some nice studies done that indicate in the West Coast that turtles follow the 18 degree isotherm, at least the hard-shelled turtles, what we call the Thecate turtles, which are the greens, hawksbills, loggerheads. When the water temperature comes up to 18 degrees, turtles start moving in. Some satellite tracking studies I have done on an eastern specific green turtle male, which is actually quite unique (not a lot of work has been done with male turtles), showed that that animal very much preferred water temperatures at 22 degrees and he seemed to move to make sure that he was within 22 degrees. He was vertically in the water column, or horizontally, depending on his location. This was tracked up the US West Coast. My statement relative to that is that temperature can be an indicator of either biological factors that the animals are moving for, or simply physiologically what the animal needs. So that is the primary reason for turtles moving. My reason for saying that I did not expect to see that kind of behaviour in countries of the Western Pacific particularly is due to lack of knowledge about temperature structure there. I feel that I don't believe that they have those kind of temperatures they get down into areas in which the turtles will move out. My colleagues from there, particularly Liew from Malaysia, could probably answer best on that one. That is how that statement was generated, I guess I was suggesting that, what I had seen, was true for turtles in areas where you have temperature fluctuations. Leatherbacks, on the other hand, are not quite as temperature prone. Leatherbacks have been seen swimming around icebergs, they can deal with very very cold waters and are considered a north temperate species in distribution. Some of my satellite data is beginning to indicate that they are far more cosmopolitan than we ever realised before. However, on the coast of California and the West Coast of the United States, we do see a relationship between temperature and the distribution of leatherbacks probably based on water mass movements. When our temperatures get above about 16 degrees we see leatherbacks start moving into Californian waters, but they are coming from off shore and what they are usually following is often water masses that are coming in. So there is some suggestion that it may be not necessarily temperature related but water mass movement related. Based on what my satellite tracking data shows, what is probably driving things, that an overall desire to move for leatherbacks is based on food availability and where that food is going to occur. So, I guess what I would do is make a statement that temperature does seem to affect the seasonal migrations and movements of turtles in habitats in which they see those kind of temperatures and this is irrespective of nesting females who have their own thing going relative to the nesting season.

The representative of the United States

128. Mr. Chairman, I just want to clarify the question. The season migration we are talking about here is not the same thing as what Dr. Liew was talking about, when the turtles have one place to nest and one place to feed and they go back and forth. They are different things, is that right?

Dr. Eckert

129. Thank you. Yes, that is exactly what I am trying to point out. We need to separate the kind of migration that I was talking about here which is seasonal migrations, from reproductive migrations. This refers very much more to say, juveniles or non-breeding females or non-breeding males relative to migration. Now, on reproductive migrations yes, that has been much better defined. There are these reproductive migrations and you have to realize that most of our research has been very myopic when it comes to sea turtles. We have looked at reproductive females, it is sort of the equivalent to martians studying humans by studying them in the maternity ward. It has left big gaps in our understanding about the biology of these creatures and we are having to address those issues now, because what we are finding is that the other 99 per cent of their lives are significant to the overall biology of the species. So, yes, I am primarily addressing non-reproductive migrations. One final note about reproductive migrations is that males also seem to exhibit reproductive migrations though they are much more poorly defined. "The adoption by TEDs of shrimp trawling would take between 6 to 8 years, could experts comment on this point?" [question 8 by the United States] Yes, I have some experience in the history of how they were put in the United States, just having been in Georgia for ten years during some of the early years of the implementation of the use of TEDs. A lot of the problems were simply problems of introduction. I think a lot of those problems have been solved and I would say that 6 to 8 years may be a little longer than is absolutely necessary but, as we talked about before, a lot of this depends on the sociological factors associated with the particular industry. In Australia, they seem to have a very good cooperative relationship with a relatively small industry. In the United States, you had independent fisherman working in their own boats and their own areas and a very large fleet, certainly the number of vessels operating there were numbering in the thousands, in Australia it is numbering in the hundreds. So, there are a lot of those factors that have to be brought into play: of how easy it is to work with the fisherman, how easy it is to teach them how to use these and I would have to say that if Mr. Guinea feels it would take 6 to 8 years in Australia, it very well might. Although Mr. Poiner certainly has more expertise there than I would. But, in the US the long duration that it took to introduce was mainly because it was the first time that anybody tried to put something like that into place with a fleet of that size and there was simply a long learning process on how to introduce it. I think that in the last few years they found that the optimum way was simply to require it and go from there, it has been a lot more effective. So, I will turn it over to Jack. Thank you.

Dr. Frazier

130. I will start with Malaysia then. Just passing very quickly over the first questions directed to Scott. The question of jurisdiction, although this is not directed to me, I feel that this concept is very important. When turtles are on the high seas outside the EEZs of any country, they are on the high seas in the commons, in the world commons, and I think that is one of the reasons why sea turtle biologists often speak of this as a global resource because at that point any nation can have an impact on it. Clearly, they are not under the impact of trawlers out there, trawlers do not work in the high seas, but they are a global commons when they are in that part of their life stage. As a concept in conservation biology, when you have a charismatic flagship species like this; my hope is that it will draw nations together to collaborate in the conservation of these shared resources, whether it is in shrimp trawling or high sea drift netting, long lining or whatever. Just that point to me is very important as a strategy in conservation.

131. The first question [by Malaysia] to all experts. Using TEDs won't grow more hair on my head, I won't have more girlfriends, it won't solve the problems of poverty and suffering on the planet. But using TEDs will contribute, if they are correctly done, to reducing sea turtle mortality and will be part of an integrated approach to conserving a very complex resource. I don't think it is useful for us to go to extremes and say what is the greatest, what is the smallest, by itself will it do this. I don't believe we can effectively conserve a complex resource by limiting our vision to something like this. It would be the same as human health, you would not raise a child by just giving it rice, you have to give it other sources of food, you have to give it love, education and so forth. So, the rice is necessary but it is not sufficient alone. By itself, it is not enough but I do not understand why we should consider that by itself.

132. The question of population size, breeding units or management stocks [question 6 by Malaysia], I tried to address that in my comments. It is a source of discussion because we are just beginning to understand this with sea turtles. If they were white rats or squirrels or lizards on a wall it would be much easier for us to manage them because we could understand these animals with much simpler investigation. With animals which take decades to reach maturity, which live for long periods, to maintain their populations must breed for long periods, which move about the world's oceans, some of them more than others, it is very, very difficult to understand just what a breeding unit is. To determine the population size in normal population biology means everything that is part of a breeding unit, a management stock from the very smallest animals to the very largest animals; I tried to address this in my comments, and I hope I have been clear. The trouble with sea turtle biology is that most, all of our work is limited to the beaches because it is easier to get to beaches. Biologists will be constantly complaining that we do not have enough funds to do what we need to do. If you have to work on the high seas it is much more expensive, it's much more complex but that's where the turtles spend 99 per cent of their lives. Most of what we know about sea turtles is from what they do on beaches. Therefore most of what we can do in terms of estimating population size is what happens on beaches. This is why we see graphs of egg production or graphs of numbers of nests or graphs of number of females. That's where we have the best idea of what's happening. This is not a complete idea; it is an index of what is happening. In that respect we must be very careful, because the number of eggs produced on a beach represents a number of phenomena which are subject to variation. The number of eggs in a single clutch will vary from female to female. Each female will lay a number of clutches in a season, that number of clutches will vary between females, between species, between populations and indeed between years ... [tape turnover] ... This is then an indicator of the number of females that nested in that year, but it has to be looked at as a very sloppy indicator. You cannot just take one number and divide it by the number of eggs to get to the exact number of females that nested in that year.

133. Indeed, if you could note a number of females that nested on a beach, all the females that nested on a beach in a single year, that only tells you the number of females that nested in that year. That doesn't tell you the number of females in the population, nor is it simple to take that number and estimate the total number of females in the population. The numbers of females every year can vary, there has been some very dramatic changes in numbers of females nesting in a single year from work done in Australia. There is an island called Raine Island, my colleagues from Australia can, I am sure, explain this much better than I can, but in Raine Island they cannot count the number of turtles in a season for a number of reasons. The first reason is that it is a remote area and it is difficult to get to; the second is that the density of turtles is so great that it is almost physically impossible to count them. It is like Gahirmata in India, or some of the nesting beaches in Central America. It is physically impossible to count the turtles, there are so many. As I recall in some years they have had in a single night in excess of 10,000 turtles in a single night, and then the next year the greatest number of turtles nesting in a single night, several hundred, I believe. So, what has happened? Why are the turtles doing this? Does it mean they have all gone extinct? No, these are tremendous fluctuations. In Australia, they have good information which shows this is relevant to the southern oscillation and there seems to be a relationship to food availability. In other areas we are much more ignorant, we do not know what is happening. We still have these fluctuations. I have tried to address this also, I have tried to include scientific references which deal with this. Green turtles are famous for this but other species also show this. So it is very very difficult with our information to be able to estimate what is the size of a population. With long-term data, Scott Eckert prefers to use an indication of generation times, or maturation times, or re-migration times to give an index of what length of time we need to monitor. There are other people who would simply say we need 10 or 20 years. The point is we need many years of information with these long-lived animals, and this in fact is not something unique to sea turtles, it is something that in biology we have begun to realize that problems of scale are tremendously important. If we are going to understand biological phenomena, we are going to need to understand the tremendous fluctuations that happen in the living world. So I am not sure whether I have done anything more than confuse you with this, I'm sorry if I have done that because I myself am confused, this is very complex. We need to have better information here, clearly we need it, but at the moment at least I consider myself to be terribly ignorant with this topic.

134. "Please tell us your views about the concept of unit stocks or populations or breeding units of sea turtles". [question 7 by Malaysia] Well, this touches back on what I have just been mentioning and what I would add here is something which, especially in fisheries biology, I believe now is an international accord. Certainly, I hope it becomes much much more talked about, and this is the precautionary approach. This is well described in the FAO Code of Conduct for Responsible Fisheries. In simple terms, the precautionary approach is, if we don't know enough we have to be careful. Especially if we are managing resources which are shared by many different people and on which generations of people will depend. The history of the world shows us that, we, as a species, independent of which nationality, have not been careful. The status of fishery stock shows us that we are in a desperate situation. We have to be more careful with the way we manage fishery stocks. Sea turtles are impacted by the way we manage fishery stocks. That is my concept on it. As I mentioned in one of the letters I wrote to the Secretary, if this was as simple as rocket science we could give you straight answers. It's not as simple as rocket science, it's much more complex. "When studies of any particular sea turtle population are made with the results applied to the population study or whether you as a scientist generalize your data, for all sea turtles irrespective of the species...". [question 8 by Malaysia] No, I would not generalise to all species nor to all populations. Indeed, I would say the data I have found are relevant to this population for the time when I was studying it. That is careful science. However, if I don't know enough from another population or from another period or from another species and my level of ignorance forces me to admit that I don't know and I'm forced to take a decision to protect that species, I then have to start grabbing the best information I can find, and I then grab at the closest I can find which is comparable. If that means that I don't know enough about migrations and I have to go to another ocean basin to say, well, I know that this species in another ocean basin does this, I will therefore assume it is not that different in this ocean basin. I will use that as my best approach for management until I have better information.

135. "Notwithstanding the status listings, would you not say that there are sea turtle populations in the world which are quite healthy and which have benefitted from long-term conservation programmes started some 30 years ago?" [question 9 by Malaysia] Definitely, any conservation programmes, especially if they started 30 years ago, would have benefitted sea turtles. What worries me is "quite healthy". I am not sure what "quite healthy" means. As a global phenomenon we have seen that there have been declines of sea turtles populations around the world, that is why they are listed in IUCN and CITES as endangered species. This is not a trivial observation, it is a global observation. If a few years ago someone had asked me where are the healthiest populations of sea turtles, I would have turned to my colleagues in Australia and said, they have them in Raine Island. But now I am terribly depressed to know that even those populations are declining. Colin Limpus, whom you have heard cited numerous times during these proceedings, has shown that populations in Australia which were enormous, are now in decline. The reason they are in decline evidently is that when they migrate out of Australia into other waters of neighbouring countries, they are subjected to very high mortality, in some cases a direct mortality. So I would prefer to put, instead of "quite healthy", "at a lesser degree of risk". I am not comfortable with "quite healthy".

136. "Which species are most vulnerable to trawling?" [Question 10 by Malaysia] Certainly, the information which is available from the US shows very clearly that loggerhead and Kemp's ridley are highly susceptible to trawling. Those are the species which are most abundant in those waters. Where the behaviour of a turtle means that it will share time in the water in areas where shrimp occur or shrimp trawlers occur, it makes that turtle obviously subject to being caught in trawlers. My information is that in Surinam (where I have never worked and I must be clear here), there seems to be some confusion at some point. I am not responding to this information with my own personal experience. I am responding as a scientist reviewing information that I can gather. I have not worked in Surinam, the information I have from reading this scientific information is that indeed green turtles and olive ridley's are subject to capture, and at some degree of importance in Surinam. We do not have good data from Gahirmatha. Colleagues who work on that beach, and this number has been bantered around quite a bit, have shown that there is a large number of turtles which are stranded in Gahirmatha. There is some question as to whether that is in gill nets or trawlers or some other type of fishing activity. No one has done a study, those data don't exist. However, the simplest explanation that I see and my colleagues, for example, colleagues from the Wildlife Institute of India, see that a significant number of those animals are being drowned in shrimp trawlers. We have no data, no study has been done and that takes me back to the precautionary approach. To say that I don't know cannot mean the problem does not exist. We have to be very careful, not to do that because that will simply make it more difficult to solve the problem. I would say that our lack of information makes it very difficult to know whether outside of areas where there have been long-term studies, whether other species are highly susceptible. I would certainly say that in Surinam, other species other than loggerhead and Kemp's ridley are susceptible from the data I have read, and the indications that have come from Orissa indicate that olive ridley is heavily affected there. Along the Pacific Coast of Central America, there are good data that olive ridleys are captured in very large numbers; the estimates for just the country of Costa Rica were 20,000 turtles a year captured in shrimp trawlers. In other countries north of Costa Rica there is less information and less study, but the numbers appear to be very significant. I am confused by a lack of information, but the snippets, the pieces of information I see from the people who work in Sabah make me very worried that there seems to be some kind of interaction between green turtles and trawling in the south. The reason I say this is indeed in this submission by Malaysia, the last page, which is a study done by Mohammed Suliansa. He ends by saying that there is a need for study of accidental capture of turtles in trawl nets and that there is a need to educate the trawl operators. My guess, I would need to speak to him, my guess is that he is concerned, and the reason that he is concerned, that he has manifested here, is that there is a problem. I remember in one of the submissions, that there are data which show that there are strandings of turtles when trawling begins in Sabah. The same phenomenon has been well documented in the US, in Texas and Louisiana especially, that when trawling begins in Sabah, strandings begin. The number of turtles which are documented is very small, and a large number of the turtles which were documented, there was a large number of those turtles for which the cause of drowning was not known. I don't have the report in front of me but, as I recall, a very significant percentage of the drownings, of the sources of mortality that could be determined, were from trawling and those were green turtles. So I am very concerned that we don't know, but certainly the people who are working in Sabah seem to be indicating that there is a very high potential for a problem. So, again my long elaborated comment which hasn't provided you with a dogmatic yes or no, again I am manifesting my ignorance. We don't have enough information but I would say, yes, other species are subject and we have to be careful. The lack of information cannot be used to prove that there is a lack of a phenomenon. If we can do studies, if someone can do a study and show that so much effort of trawling has produced so many turtles, then we can make a comparison. Without that information we are blind. I think I'll finish here.

137. I am going now to the questions provided last night. "Are other sea turtles found in each of the complainants waters that the members of population are not yet showing signs of recovery?" [question 1 by the United States] The issue of recovery has been deliberated. Again we must be very careful, we are dealing with complex animals that have these long periods of time to mature, which have long life cycles. To understand them we need long-term data. A short-term study will not give us the information we need. There are two populations which in voluminous proceedings have been discussed as having recovered and I am told to leave them aside, well at least the Sabah population aside. Under the instructions here that would leave me with Gahirmatha. I don't know what's happening at Gahirmatha. Obviously large numbers of turtles are nesting at Gahirmatha, at Devi and Rushikula. There are three mass nesting areas in Orissa that have been discovered so far. One has to be very careful, I address this in my comments, with numbers, especially from mass nesting areas. I mentioned previously the difficulty that the Australian biologists have in counting green turtles on Raine Island. With olive ridleys in mass nesting situations, this is even worse. There can be tens of thousands of animals on a beach in a single night. It is physically impossible to have a good count. I really sincerely wish that everyone here could see this. It is a phenomenon which makes you understand why sea turtle biologists are as crazy as we are. It is something that renews your faith in the world to see this, it is phenomenal.

138. In order to estimate what's happening with the population, you need long-term data. If you are going to use numbers those numbers have to be something that you have to have confidence in. If you are estimating numbers, you need to have what are called confidence intervals around those numbers. I can count everyone in this room right now and produce what is called an exact estimate. But if there were so many people in the room that I could not count them and I had to do a statistical procedure of counting, a sample of them, I would have to be careful to be scientifically rigorous. I would have to have this done in such a way that there were confidence intervals around that number. This has been a major source of difficulty with the mass nesting populations. Statistics for these numbers are very sloppy and there have been attempts now on a beach in Costa Rica, called Nancite, to try different ways of estimating mass nesting populations. One is to use a transect down the beach and to count over a period of days at a fixed interval, the numbers of turtles that are in the transect. Another is to use what is called quadrat (a square area on the beach) and count the number of turtles in that quadrat. Then you have enough transects and enough quadrats to produce an average and then a variation around that average which gives you a confidence interval. Unfortunately, these two methods don't always give the same estimates and we are now a little bit confused about what we should do, because on some beaches people have been using transects and on other beaches people have been using quadrats and we are not sure how to compare these data. My colleagues at the Wildlife Institute of India are very concerned about getting counts from Gahirmatha which have confidence intervals. I spoke with some length with Bivash Pandav who is the graduate student who is doing this study. We have discussed some of the pros and cons about the widths and lengths of transects and I am confident that he is a remarkably dedicated man and I am sure he will be getting very good data from Gahirmatha which will provide confidence intervals. We need long-term data by the same institution on beaches. Gahirmatha was monitored by the Forest Department of Orissa, the Central Marine Fisheries Institute for some years, by other institutions, and every institution will have its own way of counting. So it is very difficult to know just what's happened at Gahirmatha. I would be pessimistic to say that I don't want that population to be in recovery, or to be growing or to be at least stable, but I don't know that, I can't say that because I don't understand enough about the way the numbers have been derived and I don't see confidence intervals around those estimates. So I would be careful about saying anything about Gahirmatha at the moment. Sorry for such a long talk, but they are such complex animals. Yes, "if they suffer mortality ..." [question 1 by the United States] clearly, that is what we have been trying to explain. Any source of mortality, especially directed at the animals which have been able to survive this long period of maturation, is very very costly for the population. I'm not saying that eggs are not important, I'm simply saying that those animals that have managed to have the good luck or good sense to do the right things for 10 or 20 years, those are now very valuable to the population. If we loose them, it is a very costly loss for maintaining the population.

139. This magic number of 5,000 turtles stranding in Gahirmatha [question 2 by the United States]. Firstly, let's be very clear that 5,000 is a number of strandings, it's not mortality. We don't know how many died but certainly many, many more than 5,000. How many more I can't say, 10,000, 15,000, I can't say. But certainly if 5,000 were counted dead on the beach, the number that died would be many many more. We need to know how long they float, we need to know what the currents were, we need to know what the winds were. The only people that, as far as I know, have looked at this in a systematic way are researchers from South Carolina and I have provided a synopsis of that in my comments to show some of the problems of trying to interpret stranding data in terms of mortality. So, let's be clear that this is not mortality, this is stranding. I am worried by that number. I realize that large numbers of turtles nest at Gahirmatha, but a continual mortality of in excess of 5,000 animals (some number we don't know) worries me and I would not be comfortable to call it minor or relatively minor.

To Continue With Chapter 140

² See Appendix 2 to this Annex.