CR

Critically Endangered

A2bd

3.1

Rationale

Analysis of historic and recent published and unpublished accounts indicate extensive subpopulation declines in all major ocean basins over the last three Hawksbill generations as a result of over-exploitation of adult females and eggs at nesting beaches, degradation of nesting habitats, take of juveniles and adults in foraging areas, incidental mortality relating to marine fisheries, and degradation of marine habitats. Analyses of subpopulation changes at 25 Index Sites distributed globally (see W-Figure 1 in attached PDF) show an 84 to 87% decline in number of mature females nesting annually over the last 3 Hawksbill generations (see W-Table 1 in attached PDF). Numerous populations, especially some of the larger ones, have continued to decline since the last assessment of the species (Meylan and Donnelly 1999). Today, some protected populations are stable or increasing, but the overall decline of the species, when considered within the context of three generations, has been in excess of 80%.

Assessment Procedure:

In accord with the IUCN Red List Categories and Criteria, the Hawksbill is listed as Critically Endangered (CR A2bd) because it meets the following criteria:

A. Reduction in population size based on:

2. An observed, estimated, inferred or suspected population size reduction of >80% over the last 10 years or three generations, whichever is the longer, where the reduction or its causes may not have ceased OR may not be understood OR may not be reversible, based on (and specifying):

(b) an index of abundance appropriate to the taxon; and

(d) actual or potential levels of exploitation.

This assessment measures changes in populations based on the number of mature individuals (IUCN 2001a), specifically changes in the annual number of nesting females.

Index Sites:

Choice of Index Sites. Reliable historic data are not available for all subpopulations, so the present report quantifies population trends by examining data from 25 Index Sites (see W-Figure 1, IND-Table 1, PAC-Table 1 and ATL-Table 1 in attached PDF). Index Sites were chosen to represent broad regional subpopulation trends over time and include representative major nesting areas as well as many of the lesser nesting areas for which quantitative data are available. An estimated 41% of the current global population of nesting females is represented by index sites.

The most reliable method of monitoring trends in sea turtle populations are long-term population assessments conducted at the nesting beach (Meylan 1982) and these are used as an appropriate index of abundance for the taxon (IUCN 2001a, 2001b). But, estimating the total number of adult females in a nesting population is complicated by the fact that an individual female typically nests several times within a breeding season, and follows a non-annual breeding schedule, with intervals of two to seven years separating consecutive nesting seasons. Individuals also may be reproductively active for decades (Carr et al. 1978, FitzSimmons et al. 1995, Mortimer and Bresson 1999). Long-term monitoring is thus essential to document true population change. Few long-term studies of nesting Hawksbills exist, in part because sea turtle research did not become popular until the 1970s, and by then many populations had already been reduced to low levels (Meylan 1999).

Interpretation of long-term data can be complicated. Because Hawksbills mature slowly, an over-exploited nesting population may already be in decline for decades before the damage manifests itself as a decrease in numbers of nesting turtles on the nesting beach. Meanwhile, documented increases in numbers of nesting females must be interpreted cautiously, as they do not always reflect an absolute increase in the size of the population. In situations where protection is afforded a breeding population that previously had been subject to intense exploitation, numbers of egg clutches laid are likely to rise precipitously at the newly protected rookery. This is because, with protection, individual females survive not only to lay their full complement of three to five egg clutches within a single nesting season, but also return to breed in subsequent seasons.

Because of the extended and complicated life cycle of the Hawksbill, to quantify only a single stage in the life cycle will not always adequately portray the true status of the entire population. For example, where over-exploitation of nesting females or eggs has impeded reproduction during long periods of time, estimates of population decline based only on numbers of nests may significantly underestimate the overall population decline at those sites because they will not reflect the absence of juvenile foraging turtles in the wider population (Mortimer 1995). Although studies on foraging grounds are useful, reliable quantitative data on the size of foraging populations, and especially historical data describing foraging populations, are generally not available. Interpretation of foraging data is further confounded by the mixing of animals from various nesting populations at the foraging grounds (Broderick et al. 1994, Encalada et al. 1996). Similarly, recent increases on some Caribbean nesting beaches demonstrate the difficulty in predicting increasing numbers of sea turtles. Although reduced effort in the Cuban Hawksbill fishery has spared more than 55,000 large animals on its foraging grounds since the early 1990s (Mortimer et al. 2007), to date regional nesting increases are still relatively small.

Data Sources for Index Sites. To assess long term changes in the nesting populations at each of the 25 Index Sites, we used several types of data sources, often in combination with each other. For sites for which data on annual numbers of nesting females are not available we used other indices of nesting abundance, including numbers of nests recorded, numbers of nesting females killed, numbers of nesting females recorded per unit of patrol effort, and numbers of egg clutches collected for human consumption or for incubation in hatcheries. At some sites, different measures of Hawksbill abundance were used, including tortoiseshell export statistics, and total numbers of slaughtered animals (including both nesting and foraging turtles). The data were derived from a multitude of sources, including published scientific and historical literature and unpublished reports. We are grateful to the numerous researchers, especially the members of the MTSG Hawksbill Task Force, who generously provided their unpublished data and the benefit of their personal experience to ensure that the most up-to-date information be included in this assessment (see Acknowledgements in the attached PDF). As noted in the text and accompanying tables, such information is recorded as in litt. citations.

Unfortunately, for sea turtles and other long-lived species, decades of long-term quantitative data are seldom available. Few Hawksbill nest-monitoring projects were carried out in the 20^th Century on populations that are now depleted or remnants of their former size (Meylan 1999). Nevertheless, to estimate changes in populations over time, the contributions of historically large, but now depleted, populations needs to be considered. Where quantitative data are lacking, old naturalist’s records, historic egg collection data, and tortoise shell trade statistics are often the best source of information about populations, and can be used to estimate former abundance and subsequent declines. Unfortunately, while some excellent information about the enormous trade in tortoiseshell is available, in many areas of the world researchers will never know the full extent of the Hawksbill declines that have taken place before and during the 20^th Century. For example, Hawksbills were likely found in some numbers along the eastern coasts of the Pacific and Atlantic although now they have become scarce.

Extrapolated Data For Index Sites. In the present assessment, where quantitative data are available, population abundance estimates are based on raw data, and linear and exponential extrapolation functions (IUCN 2001a). In some subpopulations, more than one trajectory was exhibited over the 3-generation interval; changes in subpopulation size are thus often based on a combination of raw data and extrapolations. If no change is believed to have occurred outside the time interval for which published abundance data are available, we use the raw data to determine the change in population size. However, when it appeared that change in subpopulation abundance occurred outside the interval for which raw data were available, we used extrapolation techniques to determine the overall change. Linear extrapolations were used when it was believed that the same amount of change occurred each year, irrespective of total subpopulation size. Exponential extrapolations were used when it was believed that change was proportional to the subpopulation size. In cases where there is a lack of information on the specific rate of change, we used both linear and exponential extrapolations to derive a population estimate. However, when either the linear or exponential function produced an obviously unrealistic number, we included the unrealistic figures in the tables summarizing estimated population change over three generations (and noted them as being unrealistic), but we did not use those unrealistic figures to estimate population changes for the ocean basin under consideration (see IND-Table 3, PAC-Table 3, and ATL-Table 3 in attached PDF).

Backward Extrapolations of Increasing Populations. Significant increases in nesting populations during the past two decades have been recorded at a number of nesting localities, particularly in the Atlantic Ocean at the following Index Sites: Antigua (JumbyBay), Barbados, Cuba (Doce Leguas Cays), Mexico (YucatanPeninsula), Puerto Rico (MonaIsland), and US Virgin Islands (Buck Island Reef National Monument). The observed population increases correlate with implementation of protective measures at these nesting sites in combination with decreased exploitation at neighbouring foraging grounds (especially in Cuba). However, most of these now-increasing populations were not monitored prior to implementation of protective measures (the presence of researchers on the beach is often a significant element of the actual protection afforded such sites).

Using only the raw data available for these now-increasing sites, it would be impossible to estimate the overall rate of population change during the past three turtle generations, since in most cases data for the protected sites are only available from the mid-1980s onward. There is no reason to doubt that these increasing populations had suffered the same sort of declines as other nesting populations in the region for which earlier data exist. Rather than eliminate these populations from the summary calculations for the ocean basin (and over-estimate the rate of decline), we incorporated these data by extrapolating backwards from 1985, using the average population trajectory calculated for all the other Index Sites in the region for which there are data prior to 1985. The results of these calculations are presented in ATL-Table 6 (in attached PDF).

Qualitative Information

Numerical historic rates of change in the sizes of nesting populations at the Index Sites describe only one aspect of the global conservation status of the Hawksbill turtle, and tend to be somewhat biased towards those subpopulations for which long-term quantitative data exist. A wealth of information also exists about the current status of many of the world's Hawksbill nesting populations, as well as the various modern-day factors, both positive and negative, affecting them. These include: a) the residual impacts from long-term tortoiseshell trade; b) current levels of purposeful slaughter and egg collection; c) incidental capture in fishing gear; d) destruction of nesting beaches caused by unregulated coastal development, oil pollution, sea level rise and accompanying erosional processes, and elevated incubation temperatures; e) damage to foraging habitat caused by sea water warming, and pollution; and f) efforts to raise awareness, and to coordinate and legislate protection. Such information is critical to a complete understanding of the current status of Hawksbill populations around the world.

For 58 countries around the world we have compiled information on current estimated population sizes and qualitative information about current trends in nesting and foraging populations, and the factors influencing them either positively or negatively (see IND-Table 5, PAC-Table 5 and ATL-Table 7 in attached PDF) The inclusion of such relatively qualitative information ensures that even those countries with the fewest resources for monitoring and enforcement can be represented in this assessment; and these areas are often the ones where greatest exploitation and declines have occurred (IUCN 2001b).

Uncertainties in the Assessment Process

As with any assessment based on historic data or small data sets, there is uncertainty relating to the final results of this report. The sources of uncertainty are rooted in the procedure itself as well as in the stochastic nature of Hawksbill biology. Both sources of uncertainty are ultimately related to a lack of information, and when dealing with an animal as long-lived as a Hawksbill, this can be a particularly acute problem.

Since the last Hawksbill assessment (Meylan and Donnelly 1999), the IUCN Standarads and Petitions Working Group have developed a system of regression equations to address population changes over time and produce estimates of previous population sizes. With care to filter out overly regressed populations, this system appears to be adequate. Scale of population change needs to be cautiously addressed: on the one hand, declines can go no lower than 100%, but potential population increases are limitless. Small population declines that may be difficult to observe annually can be devastating over several generations. For example, a hypothetical Hawksbill population numbering 1,000 females declining at a steady rate of 1% annually would have declined by 50% in only 68 years and by 75% in 135 years.

Another issue of concern is the fact that most of the increasing nesting populations in the Caribbean were included as Index Sites in this assessment, while many declining populations were not included due to lack of data. At many sites, the simple process of monitoring a population offers significant protection. Meanwhile, adjacent unprotected and unmonitored nesting sites may be suffering significant decline due to poaching and destruction of nesting habitat that are unrecorded. A case in point is that of Antigua/Barbuda, where the relatively small Jumby Bay nesting population, which has been intensely monitored since 1987, has increased by 79% (+23 turtles) during the past two decades. Meanwhile, the other 35 known Hawksbill nesting beaches of Antigua/ Barbuda neither have been afforded protection, nor has the status of their nesting populations been monitored (see ATL-Table 7 in attached PDF). We are concerned that in the Atlantic Ocean, protected populations are over-represented in our assessment, thus causing the assessment to under-estimate the true rate of regional population decline.

Seychelles, in the Indian Ocean, is one of the few places in the world where records of long-term monitoring of both protected and unprotected beaches are available (see IND-Table 4 in attached PDF). For the inner islands of Seychelles, monitoring was conducted at all 22 islands during both the early 1980s and the early 2000s. Nesting populations at the two islands that had been well-protected since the 1970s increased by 389% during a period of two decades; meanwhile, the nesting populations at 13 islands that had received no protection prior to 1994 declined by 59% during the same period. When all 22 of the inner islands are considered together, there was an overall decline of 24% in the total nesting population between the early 1980s and the early 2000s.