Dermochelys coriacea (Southwest Atlantic Ocean subpopulation) 

Rationale

The Southwest Atlantic Leatherback subpopulation nests only in southern Brazil (Thomé et al. 2007), and this rookery is genetically distinct from all other sampled rookeries in the Atlantic (Dutton et al. 2013). The marine habitat for this subpopulation is thought to extend north across the equator and east to the coast of Atlantic Africa, southwest to Uruguay and Argentina, and southeast to South African waters (see Figure 1 in Supplementary Material). However, the geographic boundaries for this subpopulation lack resolution.

Based on analysis of long-term time series datasets of abundance—i.e. annual counts of nesting females and nests—this Southwest Atlantic Leatherback subpopulation has been increasing 232% during the past three generations and the population is considered not threatened according to IUCN Red List Criterion A2. Likewise, applying Criterion A4 reveals a population increase of 957% by the year 2040, or one generation from now, which indicates a not threatened category. Despite its extremely small population size (n=35 mature individuals) and range restriction (area of occupancy= ~320 km²), the population is not declining, so does not meet thresholds for threatened categories under Criterion B or C. Due to its very small abundance of mature individuals (n=35), this subpopulation meets the threshold for Critically Endangered under Criterion D.

We assessed this subpopulation under Regional Red List Guidelines (IUCN 2012) as well because the geographical distribution of the Southeast and the Southwest Atlantic subpopulations are identical (Wallace et al. 2010). The two subpopulations are genetically distinct and do not exchange in breeding individuals (Dutton et al. 2013), so the extinction risk for each of these subpopulations is not dependent on the other. Therefore, we assessed each subpopulation separately, and the category listing of one does not affect the listing of the other.

Justification

Application of Criterion A2 is appropriate, as population reduction may have been observed in the past where the causes of reduction may not have ceased OR may not be understood OR may not be reversible. This subpopulation is increasing by 232%, and will increase by 957% in another generation (i.e., by 2040), with an abundance of approximately 169 nests (~34 females) per year, or approximately 85 adult females total.

On applying Criteria B, the AOO (320 km²) initiates an Endangered categorization, but it is substantiated by only one subcriterion, i.e., the subpopulation is found at only a single location (it is not severely fragmented) and it is not declining; therefore, it does not meet thresholds for any threatened category. For Criterion C, no threatened category is triggered because the population is not declining. However, our assessment of available data under Criterion D triggers the Critically Endangered category because of the small and restricted population size at present (n=35 mature individuals).

From 1995/1996 to 2003/2004, 527 nests were laid on four beaches in the state of Espírito Santo, Brazil, representing between 1.2 and 18.4 females nesting annually (Thomé et al. 2007) Our analyses of long-term abundance trends in the Southwest Atlantic is corroborated by Thomé et al. (2007) who reported an average annual increase of 20.4% in nest numbers between 1995/1996 and 2003/2004. Egg collection has been minimized on the nesting beach and there is no subsistence hunting for these animals, however, fisheries bycatch is a major threat (Pinedo and Polachek 2004, Gallo et al. 2006, Thomé et al. 2007, Bugoni et al. 2008, Fiedler et al. 2012, Wallace et al. in press).

Assessment Procedure

We assessed the status of the Southwest Atlantic Leatherback subpopulation by Criteria A-D; as no population viability analysis has been performed, Criterion E could not be applied.

Criterion A: We compiled time series datasets of abundance of nesting activities from four beaches in the state of Espírito Santo, Brazil  (see Table 1 and Figure 2 in Supplementary Material). The time series data were nesting activities (e.g. tracks or nests) monitored for 28 years. For marine turtles, annual counts of nesting females and their nesting activities (more often the latter) are the most frequently recorded and reported abundance metric across index monitoring sites, species, and geographic regions (NRC 2010). We presented and analysed all abundance data in numbers of nests yr^-1 (Table 1 in Supplementary Material). We calculated annual and overall population trends for this subpopulation. The most recent year for available abundance data was 2010.

To apply Criterion A, three generations (or a minimum of ten years, whichever is longer) of abundance data are required (IUCN 2011). For A2, data from three generations ago (~100 yr) are necessary to estimate population declines beginning three generations ago through the present (i.e. assessment) year. The challenges of this requirement on long-lived species like marine turtles—with generation lengths of 30 yr or more—are obvious (see Seminoff and Shanker 2008 for review). Abundance data from ~100 yr ago are not available for leatherbacks anywhere in the world. We considered extrapolating backward using population trends based on current datasets inappropriate because estimates produced would be biologically unrealistic and unsubstantiated, given what is currently known about sea turtle nesting densities on beaches and other factors (Mrosovsky 2003). In the absence of better information, we assumed that population abundance three generations (~100 years, one generation estimated at 30 yr; see below) ago was similar to the first observed abundance than to assume that the population has always been in a decline (or increase) of the same magnitude as in the current generation (Table 1 in Supplementary Material). A similar approach was used in the Red List assessment of another long-lived, geographically widespread taxon, the African elephant (Blanc 2008). Thus, to apply Criterion A to this subpopulation, we assumed that the abundance at the beginning of an available time series dataset had not changed significantly in three generations, and therefore used the same abundance value in trend calculations.

We also applied A4 to the Southwest Atlantic subpopulation, using the same overall scheme as described above (Table 2 in Supplementary Material). Criterion A4 permits for analysis of population trend during a “moving window” of time, i.e. over three generations, but where the time window must include the past, present, and future. Therefore, we made the same assumption about earliest available historical abundance being equivalent to the subpopulation abundance for generations past, and estimated future population abundance in 2020, 2030, and 2040, i.e. within one generation. This future projection assumes that the derived population trend will continue without deviation during the next generation. Implicit in this assumption is that no changes to degree of threats impacting this subpopulation will occur during that time although the threat of fisheries in the future remains uncertain. Based on application of Criterion A4 to the available data, this subpopulation would have increased by 957% by the year 2040, or one generation from now. This would correspond to approximately 169 nests (~34 females) per year, or approximately 85 adult females total, and would represent a substantially increased population.

Criterion B: We defined extent of occurrence (EOO) as the total area included within the geo-referenced boundaries of the Southwest Atlantic Leatherback subpopulation (Figure 1 in Supplementary Material), which we calculated to be >48 million km². We defined area of occupancy (AOO) as the linear distribution of nesting sites within the EOO, which was estimated to be 160 km (Thomé et al. 2007), and then multiplied by 2 km to account for the IUCN Guidelines for calculating linear AOOs using minimum grid cell size of 2 km x 2 km. The AOO for this subpopulation was calculated to be 320 km², which initiates an Endangered categorization. Although the subpopulation is not severely fragmented, it is only known from a single location (i.e., biological rookery or genetic stock; Dutton et al. 2013),but  it does not exhibit a continuing decline nor extreme fluctuations of number of individuals. The IUCN defines extreme fluctuations as occurring when “population size or distribution area varies widely, rapidly, and frequently, typically with a variation greater than one order of magnitude.” Thus, inter-annual variations in nesting abundance do not qualify, because these oscillations do not reflect an extreme fluctuation in the number of mature adults in the population, only their annual nesting activities. Therefore, the Southwest Atlantic subpopulation did not trigger even the Vulnerable category under Criterion B, but it would qualify as Near Threatened.

Criterion C: To apply Criterion C, we first calculated the number of mature individuals in each subpopulation, i.e., the total number of adult females and males. First, we divided the current average annual number of nests (n=53, Table 1 in Supplementary Material) by the estimated clutch frequency (i.e. average number of clutches per female (n=5) used also by Thomé et al. 2007) to obtain an average annual number of nesting females. Next, we multiplied this value by the average re-migration interval (i.e. years between consecutive nesting seasons; n=2.5; Thomé et al. 2007 estimated it to be between 2 and 3 years) to obtain a total number of adult females that included nesting as well as non-nesting turtles. Finally, to account for adult males, we assumed that the overall sex ratio of hatchlings produced on nesting beaches in the Southwest Atlantic is approximately 75% female, or 3:1 female:male ratio suggesting largely female bias (Barata et al. 2004), and that this reflected the natural adult sex ratio. This calculation provided an estimated mature adult population of 35 individuals, initiating a Critically Endangered categorization (<250 mature individuals); however, because this subpopulation is not currently exhibiting a continuous decline, it does not satisfy subcriteria under C1 or C2. Consequently, no threatened category is triggered under Criterion C, but it would qualify as Near Threatened.

Criterion D: The Southwest Atlantic subpopulation has 35 mature individuals which triggers a Critically Endangered categorization (<50 individuals). Therefore this subpopulation is listed as Critically Endangered under Criterion D.

Estimating Generation Length:

Leatherback age at maturity is uncertain, and estimates range widely (see Jones et al. 2011 for review). Reported estimates fall between 9-15 yr, based on skeletochronology (Zug and Parham 1996), and inferences from mark-recapture studies (Dutton et al. 2005). Furthermore, updated skeletochronological analyses estimated Leatherback age at maturity to be between 26-32 yr (mean 29 yr) (Avens et al. 2009). Extrapolations of captive growth curves under controlled thermal and trophic conditions suggested that size at maturity could be reached in 7-16 yr (Jones et al. 2011). Thus, a high degree of variance and uncertainty remains about Leatherback age at maturity in the wild. Likewise, Leatherback lifespan is unknown. Long-term monitoring studies of Leatherback nesting populations have tracked individual adult females over multiple decades (e.g. Santidrián Tomillo et al. unpublished data, Nel and Hughes unpublished data), but precise estimates of reproductive lifespan and longevity for Leatherbacks are currently unavailable.

The IUCN Red List Criteria define generation length to be the average age of parents in a population; older than the age at maturity and younger than the oldest mature individual (IUCN 2011). Thus, for the purposes of this assessment, we estimated generation length to be 30 yr, or equal to the age at maturity (estimated to be 20 yr on average), plus a conservative estimate of reproductive half-life of 10 yr, as assumed by Spotila et al. (1996).

Sources of Uncertainty

Although monitoring of nesting activities by adult female sea turtles is the most common metric recorded and reported across sites and species, globally, there are several disadvantages to using it as a proxy for overall population dynamics, some methodological, some interpretive (NRC 2010). First, because nesting females are a very small proportion of a sea turtle population, using abundance of nesting females and their activities as proxies for overall population abundance and trends requires knowledge of other key demographic parameters (several mentioned below) to allow proper interpretation of cryptic trends in nesting abundance (NRC 2010). However, there remains great uncertainty about most of these fundamental demographic parameters for Leatherbacks, including age at maturity (see Jones et al. 2011 for review), generation length, survivorship across life stages, adult and hatchling sex ratios, and conversion factors among reproductive parameters (e.g., clutch frequency, nesting success, re-migration intervals, etc.). These values can vary among subpopulations, further complicating the process of combining subpopulation abundance and trend estimates to obtain global population abundance and trend estimates, and contributing to the uncertainty in these estimates. For example, in the present case of the Southwest Atlantic Leatherback subpopulation, our assumption of 3:1 female:male sex ratio has a significant impact on the estimate of mature individuals in the subpopulation; i.e. under the assumption of a 1:1 sex ratio, the number of mature individuals (n=54) would have exceeded the threshold for Critically Endangered, thus qualifying this subpopulation for the Endangered category. Our assumption was based on published data (Barata et al. 2004), but more information is needed to increase confidence in this assumption of adult sex ratios, as well as assumptions included in estimates of other demographic parameters.

Second, despite the prevalence of nesting abundance data for marine turtles, monitoring effort and methodologies can vary widely within and across study sites, complicating comparison of nesting count data across years within sites and across different sites as well as robust estimation of population size and trends (SWOT Scientific Advisory Board 2011). For example, monitoring effort on Matura beach, Trinidad, has changed multiple times since the early 1990s, which necessitated a modelling exercise to estimate a complete time series for years with reliable monitoring levels (Table 2 in Supplementary Material). Furthermore, there was a general lack of measures of variance around annual counts provided for the assessment, which could be erroneously interpreted as equally high confidence in all estimates. Measures of variance around annual counts would provide information about relative levels of monitoring effort within and among rookeries, and thus reliability of resulting estimates. For all of these reasons, results of this assessment of global population decline should be considered with caution. For further reading on sources of uncertainty in marine turtle Red List assessments, see Seminoff and Shanker (2008).

For further information about this species, see 46967838_Dermochelys_coriacea_Southwest_Atlantic_Ocean_subpopulation.pdf.

A PDF viewer such as Adobe Reader is required.

Leatherbacks are distributed circumglobally, with nesting sites on subtropical and tropical sandy beaches and foraging ranges that extend into temperate and sub-polar latitudes. See Eckert et al. (2012) for review of Leatherback geographic range. The Southwest Atlantic Leatherback subpopulation nests only in Brazil (Thomé et al. 2007) and its marine habitat is thought to extend north across the equator in Brazil and east to the coast of Atlantic Africa, southwest to southern Brazil, Uruguay, and Argentina, and southeast to South African waters (see Figure 1 in Supplementary Material); however, the geographic boundaries for this subpopulation lack resolution. 

Leatherbacks are a single species globally comprised of biologically described regional management units (RMUs; Wallace et al. 2010), which describe biologically and geographically explicit population segments by integrating information from nesting sites, mitochondrial and nuclear DNA studies, movements and habitat use by all life stages. RMUs are functionally equivalent to IUCN subpopulations, thus providing the appropriate demographic unit for Red List assessments. There are seven Leatherback subpopulations, including the Southwest Atlantic Ocean, Southeast Atlantic Ocean, Northwest Atlantic Ocean, Northeast Indian Ocean, Southwest Indian Ocean, East Pacific Ocean, and West Pacific Ocean. Multiple genetic stocks have been defined according to geographically disparate nesting areas around the world, and in the Atlantic Ocean in particular (Dutton et al. 1999, 2013), and are included within RMU delineations (Wallace et al. 2010; shapefiles can be viewed and downloaded at: http://seamap.env.duke.edu/swot).

See the species account for a summary of the details. For a thorough review of Leatherback biology, please see Eckert et al. (2012).

Threats to Leatherbacks (and other marine turtle species), vary in time and space, and in relative impact to populations. Threat categories were defined by Wallace et al. (2011) as the following:

1) Fisheries bycatch: incidental capture of marine turtles in fishing gear targeting other species;

2) Take: direct utilization of turtles or eggs for human use (i.e. consumption, commercial products);

3) Coastal Development: human-induced alteration of coastal environments due to construction, dredging, beach modification, etc.;

4) Pollution and Pathogens: marine pollution and debris that affect marine turtles (i.e. through ingestion or entanglement, disorientation caused by artificial lights), as well as impacts of pervasive pathogens (e.g. fibropapilloma virus) on turtle health;

5) Climate change: current and future impacts from climate change on marine turtles and their habitats (e.g. increasing sand temperatures on nesting beaches affecting hatchling sex ratios, sea level rise, storm frequency and intensity affecting nesting habitats, etc.).

The relative impacts of individual threats to all Leatherback subpopulations were assessed by Wallace et al. (2011). At a global scale, fisheries bycatch was classified as the highest threat to Leatherbacks globally, followed by human consumption of Leatherback eggs, meat, or other products and coastal development. Due to lack of information, pollution and pathogens was only scored in three subpopulations and climate change was only scored in two subpopulations. Enhanced efforts to assess the impacts of these threats on Leatherbacks—and other marine turtle species—should be a high priority for future research monitoring efforts.

Leatherbacks are protected under various national and international laws, treaties, agreements, and memoranda of understanding. A partial list of international conservation instruments that provide legislative protection for leatherbacks are: Annex II of the SPAW Protocol to the Cartagena Convention (a protocol concerning specially protected areas and wildlife); Appendix I of CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora); and Appendices I and II of the Convention on Migratory Species (CMS); the Inter-American Convention for the Protection and Conservation of Sea Turtles (IAC), the Memorandum of Understanding on the Conservation and Management of Marine Turtles and their Habitats of the Indian Ocean and South-East Asia (IOSEA), the Memorandum of Understanding on ASEAN Sea Turtle Conservation and Protection, and the Memorandum of Understanding Concerning Conservation Measures for Marine Turtles of the Atlantic Coast of Africa.