By Bryan P. Wallace

We’ve all heard many words to describe Dermochelys coriacea. Superlative. Ancient. Enormous. Beautiful. I’ll add another one that I think describes them well: weird. They have weird physiology: They can stay warm in frigid waters and avoid overheating in tropical waters, and they can dive more than a kilometer (3,280 feet) deep. We don’t really know what they do down there. They lay weird eggs that are not even eggs—shelled albumen gobs containing no yolk that serve no apparent purpose (and we’ve looked for one). They have weird semi-bony shells. They grow to huge sizes, and they thermoregulate and migrate across ocean basins while eating only jellyfish, for goodness’ sake! What if we have underestimated their weirdness when trying to assess and understand their population status?

Lots of Data, but Little Clarity

If we don’t know where we are and where we’ve been, it’s hard to know where we might be going. This is true for many things, including the status of sea turtle populations, and especially for the status of leatherback populations. Decades of effort by hundreds of researchers and volunteers on nesting beaches, in the water, and even via satellite-relayed movements and oceanographic conditions, have yielded mountains of data about sea turtles and how they live.

Yet despite all the knowledge we have gained, we’re still unable to understand the leatherback population trends we see or understand why we haven’t seen recovery after decades of conservation effort. In part, this paradox is rooted in a history of assessing leatherback status in ways that are fraught with inconsistencies. Flawed assumptions about how to go from a snapshot in time and space to a full global picture, which beaches to include, and more have confounded efforts to get a complete picture about how leatherbacks are doing. On multiple occasions, researchers have sounded the alarm that leatherbacks are in decline, but each time uncertainties have remained about the underlying dynamics of the situation and about what those dynamics might mean for the future.

The Challenge of Understanding the Past

Nearly 30 years ago, in 1996, Jim Spotila and coauthors asked the question “Are leatherbacks going extinct?” After they compiled data, rough estimates, and personal communications from researchers around the world, their analysis indicated that several populations were indeed declining, especially those in the Pacific Ocean, and that global leatherback abundance had declined from 115,000 to 34,500 adult females. A few years later, in 2000, the IUCN Marine Turtle Specialist Group concluded in a Red List assessment that the global leatherback population deserved critically endangered status, a listing that was again largely driven by the rapidly declining East Pacific and peninsular Malaysia populations. In fact, by the early 2000s, the once large leatherback population in Terengganu, Malaysia, was extinct. Seems pretty bad, right?

As with any trend, however, its accuracy depends on the source information. Both the Spotila et al. and Red List assessments used a baseline population estimate of 115,000 adult female leatherbacks worldwide, published in 1982 by Peter Pritchard—more than 80,000 of which were in the eastern Pacific alone. This 1982 estimate was itself an enormous increase over a much more modest population estimate by the same author in 1971 of 29,000–40,000 females globally, with approximately 8,000 in the eastern Pacific. This massive change was based on a two-day aerial survey that was of leatherback nesting in three Mexican states and was undertaken in 1980, which stimulated a vast extrapolation of existing (largely anecdotal) nesting abundance values, many of which had been conveyed to authors personally and without information about monitoring effort.

Until recently, this type of reporting—personal communications, second-hand information, and a lack of consistent monitoring or data analysis underpinning back-of-envelope calculations—was the norm. The Spotila et al. paper and the Red List assessment had simply adopted the 115,000 number because it was the best available science at the time. However, several researchers have noted that Pritchard’s numbers are almost certainly overestimates for the reasons mentioned previously. Further, no estimates before or since have landed global leatherback numbers in the same order of magnitude as Pritchard’s 1982 tally, even as more and more beaches have been included with improved monitoring, reporting, and collaboration over time. So does this mean that leatherbacks were not actually declining toward extinction globally?

Status Improves—or Does It?

Since those days, we’ve learned much more about leatherbacks and gained data from significant beaches in parts of the world that had been overlooked or unmonitored when the early assessments were done. In 2013, the leatherback status on the Red List was updated by evaluating nesting turtle abundance data through 2010 from many of the same places included in Spotila et al.’s paper, plus many more locations. This updated Red List assessment confirmed the critically endangered status of the West and East Pacific leatherback subpopulations (also called regional management units, or RMUs) while highlighting that the Northwest Atlantic RMU was abundant and stable, if not increasing. Further, the Southeast Atlantic RMU appeared to be at least as abundant—though with an undetermined trend—as the Northwest Atlantic RMU. But with the apparently robust Northwest Atlantic RMU included, the global status looked better than it had two decades prior. The global Red List assessment improved from critically endangered in 2000 to vulnerable in 2013. Finally, some good news!

Some apparent improvements in status were due to differences in which rookeries were included in the various analyses. For example, Pritchard’s aerial survey–based estimates of global population size did not include the large rookeries of Trinidad and Tobago in the Wider Caribbean Region nor Gabon on the West African coast, and Spotila et al.’s estimates of those rookeries were far lower than those used in the updated Red List assessment. In fact, in his first global assessment in 1971, Pritchard mentioned that only one or two leatherbacks nested nightly in northeastern Trinidad, and Spotila reported about 200 females per year in 1996, whereas now it’s normal to see hundreds per night during the peak nesting season. Similarly, while Pritchard asserted in 1971 that “a moderate amount of nesting” probably occurred in West Africa, by 1996, Spotila reported fewer than 5,000 females per year there. But by 2010, estimates of leatherback abundance were an order of magnitude higher.

Maybe leatherbacks really had increased in several places—and globally. Even after one accounted for discrepancies among different assessments, it seemed that things might be looking up for leatherbacks in some places and that there were more leatherbacks in the world than we had previously known. Those findings also gave much-needed hope for a brighter future in places where leatherbacks had declined and not yet recovered, such as the eastern Pacific.

More Populations in Decline

Just a few years after the updated 2013 Red List assessment was published, nervous whispers that leatherbacks actually might be in decline in parts of the Wider Caribbean rose to a chorus of concern, prompting a regional analysis of nest abundance trends, with data through 2017 amassed by the Wider Caribbean Sea Turtle Conservation Network. This analysis showed that annual numbers of leatherback nests had dropped on almost every nesting beach examined, producing a regionwide negative trend that accelerated in the most recent decade analyzed. For example, leatherback abundance in French Guiana—considered for decades to be robust and stable—had declined from tens of thousands of nests per year in the 1990s to a few hundred per year by 2017. The Red List status for the Northwest Atlantic RMU was updated with those new data in 2019, changing the status from least concern to endangered.

To make matters worse, soon after this Northwest Atlantic status update, a new global assessment of leatherback status delivered more bad news. In a comprehensive evaluation of abundance and trends from all leatherback RMUs through the year 2020, the U.S. National Oceanic and Atmospheric Administration biological review team confirmed the dire status of the Pacific RMUs and the Northwest Atlantic RMU. The team also revealed that the Southeast Atlantic RMU—thought to be the most abundant on the planet just 10 years earlier—was actually in decline. Other RMUs (Southwest Indian Ocean, Southwest Atlantic Ocean) are relatively small and geographically restricted, making them susceptible to declines as well, if the right threats were to come along.

Leatherbacks Are Weirdly Unique and Uniquely at Risk

Today, looking back on all the status assessments since the 1970s and accounting for their associated caveats, the global leatherback population trend does indeed appear to be downward. So maybe it’s time to ask the question posed by earlier researchers: Are leatherbacks going extinct? Like, for real this time?

To be clear, I consider myself an optimistic realist about sea turtle status. I tend to think that the extinction of any species—particularly at a global scale—is highly unlikely. After all, sea turtles survived the asteroid that took out most of life on Earth (including their non-avian dinosaur cousins), not to mention shifting continents and climate ups and downs over millennia. And despite everything we’ve thrown at them in the past few centuries—and it has been a lot—sea turtle populations seem to be hanging on everywhere and even bouncing back in some places. The drumbeat of good news for sea turtles appears to be getting louder, a testament to the incredible conservation efforts performed by so many people in so many places. To me, the trends underscore turtles’ resilience in the face of adversity, especially with sustained help from humans. Slow and steady just might win the race after all.

But leatherbacks do not seem to show the same resilience as other sea turtle species, at least not in the past five decades. Why are they doing so poorly? When are they going to recover? Are they going to recover? Threats to leatherback survival have been well-documented in many places. Chief among them have been human consumption of eggs and meat and incidental mortality in fishing gear. But something else might be compounding the negative effects of high mortality and low recruitment. Something else might be holding back the leatherback.

Perhaps that something is heightened sensitivity of the species to fluctuations in the marine environments the turtles depend on to survive and thrive. Like those of all animals, leatherback populations are driven by environmental ups and downs. Research in the early 2000s showed that eastern Pacific leatherbacks encountered much less predictable, much less favorable oceanographic conditions than did their Northwest Atlantic counterparts, making them smaller and more vulnerable to threats (see SWOT Report, vol. IV, pp. 8–11). This change caused divergent population trajectories between the two RMUs. Might the same one-two punch of poor ocean conditions and threats now be knocking down the Northwest Atlantic and other RMUs as well? Maybe we have underestimated the unique—and uniquely weird—sensitivity of the species to environmental conditions.

Digging further into available data might provide some clues. Mark-recapture analyses of flipper tags and microchips on individual leatherbacks can reveal how long leatherbacks take to return to beaches between nesting seasons and can shed light on the probability that leatherbacks survive from one season to the next. In declining populations like Pacific Mexico, Pacific Costa Rica, and French Guiana, adult female survival rates (lower than 80 percent per year) are indeed much lower than they should be for long-lived vertebrate populations with stable trends (typically above 90 percent per year), thereby pointing to high adult mortality. However, those analyses have also revealed an interesting and often overlooked phenomenon: the surprising prevalence of so-called transient turtles—turtles that are tagged in one season and then never seen again.

Transients are common in animal population biology and are usually explained as individuals that are merely “passing through” a study area, rather than long-term residents. However, given robust coverage of leatherback nesting in most regions, the proportion of these one-and-done nesters is much higher than expected in multiple populations. For example, of roughly 8,000 turtles tagged across 30 years in Mexico and Costa Rica, perhaps 30 percent were tagged in one season and never seen again, and adult female survival probability was quite low—less than 80 percent per year. Similarly, in French Guiana, the high number of one-and-done turtles strongly influenced survival rates, which were also less than 80 percent per year.

Why would there be so many transients, and how might this be a clue to what is happening with leatherback populations globally? Let’s assume that a turtle receiving her first tags at a long-term monitoring project is a newly reproductive adult. Given what we know about sea turtles’ site fidelity, we should expect to see her again in another nesting season. So, if she does not return, it is possible that she has been unable to find the food resources she needs to remigrate and reproduce. Perhaps this just means that she needs a prolonged remigration interval to gather what she needs. Maybe some turtles are swimming around for 10 years or more trying to accumulate enough fat stores to make the return trip. In fact, this is exactly what researchers in Australia have observed in green turtles tagged on nesting beaches and recaptured in foraging areas without being resighted on beaches over prolonged periods.

Or maybe the cost of being a reproductive adult is just harder on the neophytes. It’s a huge physiological shift from being a juvenile to actually “adulting,” which for a leatherback requires migrating across oceans to make and lay 30 kilograms of eggs each nesting season. This shift is particularly costly if neophytes haven’t honed their foraging and migration skills like older, more experienced turtles. Maybe this physiological cost is simply too high for many new adults, resulting in proportionally higher mortality in this age class compared to older remigrants.

The large number of one-and-done leatherbacks across multiple populations suggests that, in general, leatherbacks might just be more sensitive to environmental conditions than we appreciate. For example, according to tag return data, female leatherbacks in French Guiana are expected to reproduce no more than three times in their entire lives. Normal sea turtle life history assumes reproduction occurs every few years over multiple decades to compensate for high mortality of eggs, hatchlings, and juveniles. In comparison, the oddly brief reproductive lifespan of leatherbacks makes them more like salmon than sea turtles.

The possibility of environmental sensitivity becomes more intriguing when we look beyond the data from recent monitoring programs. For example, elder residents in communities of Pacific Costa Rica have recounted how, in the 1960s, very few leatherbacks were on the very beaches that by the 1980s were crawling with so many turtles that locals referred to these behemoths as “ants.” Colleagues in Mexico confirmed that elders there recalled similar patterns—not many leatherbacks in the 1960s, but tons by the 1980s.

What if the historical absence (or comparatively low abundance) of rookeries we now know to be highly abundant (for instance, Trinidad and Gabon) suggests that those patterns also occurred elsewhere in the world? The smaller numbers of leatherbacks were probably not due to human threats; it seems that there simply were not many leatherbacks around at the time of the early global assessments. Perhaps presciently, Pritchard wrote in 1971 that “there is no evidence that present numbers are yet substantially reduced from primordial, equilibrium population levels.” Is it possible that historical leatherback abundance has fluctuated over time, often without human intervention?

A Future for Leatherbacks

Despite all our data—and all our supposed knowledge of how populations work—perhaps the forces truly driving leatherback population dynamics or limiting their recovery are mainly environmental in nature. This hypothesis contends that the rapid increase in leatherback numbers in the 1970s and 1980s in the eastern Pacific (perhaps later in other regions) would have been due to long-term cycles in environmental conditions that favored leatherback growth, recruitment, survival, and reproductive output for a period.

On the flip side, when times get tough, leatherback fecundity might be significantly depressed, slowing population growth and making them less resilient to threats. Maybe this boom and bust cycle characterizes leatherback population dynamics over long time periods and perhaps more so than other sea turtle species. Of course, human-caused mortality from commercial egg harvesting and fisheries bycatch is sufficient to reduce turtle numbers on its own, regardless of environmental conditions. But if unsustainably high mortality is coupled with unfavorable environments, this is a recipe for disaster for leatherbacks.

But there is a positive side to this paradigm: With effective threat reduction and favorable environmental conditions, leatherbacks should recover eventually. Their numbers apparently increased on their own many decades ago, and they have contracted and expanded globally with the wax and wane of glaciers over geological time. Though there isn’t much we can do to produce more reliable food sources for leatherbacks, there is a lot we can do to bolster resilience in leatherback populations through effective conservation efforts.

Fortunately, people are rising to the challenges by reorganizing and redoubling efforts to reduce threats that leatherback populations currently face. In the eastern Pacific, the Eastern Pacific Leatherback Conservation Network (Red Laúd OPO in Spanish) is coordinating and supporting members’ efforts everywhere they need to happen, whether on beaches or on boats or in conference rooms. In the Wider Caribbean, a hot-off-the-press regional action plan developed by key actors across the region highlights priority actions that must be implemented to promote leatherback recovery. And a tri-national plan involving Indonesia, Papua New Guinea, and Solomon Islands is in place to promote West Pacific leatherback conservation. There are many comparable examples in other parts of the world.

Good conservation is a long, hard slog that involves long-term collaborations among multiple stakeholders to implement the best possible interventions. And we have lots of examples of effective, long-term conservation providing positive results forsea turtles. Despite all the bad news, our conservation efforts will work in the long run—especially when Mother Nature lends her hand by providing leatherback-friendly ocean conditions. This is not the first time people have thought leatherbacks were headed toward extinction, and each time, there was more to the story. This time will be no exception.

So let’s keep slogging.

This article originally appeared as part of a larger special feature on leatherbacks in SWOT Report, vol. 18 (2023). Click here to download the complete article as a PDF.