Peter A. Meylan

Understanding the geographic distribution of genetic diversity of imperiled species across all life history stages, and identifying the factors that shape those distributions, are key to maintaining long-term genetic diversity and the health of populations. This knowledge is particularly important for highly mobile marine organisms, whose extensive movements can obscure patterns of population structure. We substantially expand the genetic dataset for the critically endangered hawksbill turtle, Eretmochelys imbricata, in the West Atlantic, focusing on the southwest Caribbean. Our dataset comprises nearly 3,000 mtDNA control region sequences (740 bp) assigned to 60 haplotypes: 41 found in rookeries and 47 in foraging grounds, including 17 orphan haplotypes. The Panama metapopulation represents a major center of genetic diversity for hawksbills, with one of the highest recorded diversity values for the species (h = 0.749, π = 0.00782), nine endemic haplotypes, and four additional haplotypes that are endemic to the Southwest Caribbean. Rarefaction analyses indicate that a sample size of at least 100 is necessary to reveal true haplotype richness at most rookeries. Many-to-many mixed stock analyses, which incorporated rookery size and distance priors for 19 rookeries and 15 developmental foraging grounds, suggest that hatchlings from rookeries in the southwest Caribbean are distributed among multiple, widely-spaced foraging grounds across the West Atlantic. These results support a groups-to-soups analogy, in which genetic variability across foraging grounds represents a continuum of genetic diversity that can best be explained by a “current conveyor” model. The dataset shows that philopatry in hawksbills is not absolute, resulting in true biological dispersal and geneflow on local, regional, and ocean-basin scales, likely facilitated by dispersion during the epipelagic stage. The important contribution of oceanographic features to genetic variation at rookeries and foraging grounds is corroborated, as is the concept of oceanographic “dispersal shadows” that limit geneflow between rookeries. This study reinforces the assertion that all range states share responsibility for the recovery of the hawksbill, because foraging grounds, that are often at distant locations, are the source of future generations of reproductive adults. We also document significant movement by hawksbills between regional management units (RMUs) 29 and 30 in the West Atlantic. The Spanish version of the Abstract is available in Supplementary File 1.

Abstract – A rich Rancholabrean local fauna representing a freshwater marsh community with

adjacent uplands is identified from a site in St. Petersburg, Florida. During the winter of 2008,

sediments excavated on the Eckerd College campus produced a previously unknown assemblage

of fossil terrestrial and aquatic vertebrates. All fossils found to date can be assigned to a Late

Pleistocene age. Remains of Holmesina septentrionalis, Sigmodon hispidus and Amphiuma means

provide evidence that the age of the site falls within the Rancholabrean North American Land

Mammal Age. The fossils consist of many aquatic vertebrates with alligators, freshwater turtles,

amphibians, and aquatic snakes being the more common elements found. These suggest deposition

in or near a large source of freshwater, most likely a marsh. Fragmentary fossils of terrestrial

mammals including large herbivores such as horse, tapir, and mammoth were most likely washed

into the site from adjacent upland habitats.

To characterize the movements and habitat use of juvenile green turtles (Chelonia mydas) in benthic developmental habitat, we deployed Fastloc-GPS-enabled satellite transmitters on 16 individuals captured as part of a multi-decade study of green turtles on the Bermuda Platform. We characterized residence areas, distinct use areas within them, and seasonal movements based on an average of 562 Fastloc-GPS positions and 284 tracking days per turtle. We estimated residence area sizes using traditional home range methods, e.g., 90% utilization distribution (UD) (mean 2.29 ±2.71 km2) and 50% UD (mean 0.54 ±0.69 km2). Total residence area size increased significantly over the 8-year study, from <1 km2 before 2013 to ≥3 km2 in 2018 (R2 = 0.51, F1,14 = 14.55, p = 0.0019), corresponding to a period of decline in seagrass habitat and suggesting increased foraging effort. We identified three types of distinct use areas within residence areas where tracked turtles typically exhibited behavioral fidelity: foraging, resting, and cool weather refugia. These distinct use areas were smaller than high-use areas from previous studies; e.g., seagrass meadow foraging areas averaged 0.05 km2. Most turtles made daily transits between foraging and resting sites; for some individuals, these involved crossing frequently used vessel navigation channels. Seasonal variation in behavior suggested that the overwintering strategy for green turtles on the Bermuda Platform involves "optional dormancy," during which turtles spent less time on seagrass meadows and made brief excursions to distinct deeper habitats. Four individuals made directed (mean path straightness = 0.93 ±0.02 SD) developmental migrations away from Bermuda toward known adult foraging range. Results of our study further knowledge of the green turtle life cycle at a high-latitude site; they demonstrate that green turtles show fidelity to distinct use areas within developmental habitats over many years and exhibit seasonal movements.

Foraging habitat selection and diet quality are key factors that influence individual fitness and meta-population dynamics through effects on demographic rates. There is growing evidence that sea turtles exhibit regional differences in somatic growth linked to alternative dispersal patterns during the oceanic life stage. Yet, the role of habitat quality and diet in shaping somatic growth rates is poorly understood. Here, we evaluate whether diet variation is linked to regional growth variation in hawksbill sea turtles (Eretmochelys imbricata), which grow significantly slower in Texas, United States versus Florida, United States, through novel integrations of skeletal growth, gastrointestinal content (GI), and bulk tissue and amino acid (AA)-specific stable nitrogen (δ15N) and carbon (δ13C) isotope analyses. We also used AA δ15N ΣV values (heterotrophic bacterial re-synthesis index) and δ13C essential AA (δ13CEAA) fingerprinting to test assumptions about the energy sources fueling hawksbill food webs regionally. GI content analyses, framed within a global synthesis of hawksbill dietary plasticity, revealed that relatively fast-growing hawksbills stranded in Florida conformed with assumptions of extensive spongivory for this species. In contrast, relatively slow-growing hawksbills stranded in Texas consumed considerable amounts of non-sponge invertebrate prey and appear to forage higher in the food web as indicated by isotopic niche metrics and higher AA δ15N-based trophic position estimates internally indexed to baseline nitrogen isotope variation. However, regional differences in estimated trophic position may also be driven by unique isotope dynamics of sponge food webs. AA δ15N ΣV values and δ13CEAA fingerprinting indicated minimal bacterial re-synthesis of organic matter (ΣV < 2) and that eukaryotic microalgae were the primary energy source supporting hawksbill food webs. These findings run contrary to assumptions that hawksbill diets predominantly comprise high microbial abundance sponges expected to primarily derive energy from bacterial symbionts. Our findings suggest alternative foraging patterns could underlie regional variation in hawksbill growth rates, as divergence from typical sponge prey might correspond with increased energy expenditure and reduced foraging success or diet quality. As a result, differential dispersal patterns may infer substantial individual and population fitness costs and represent a previously unrecognized challenge to the persistence and recovery of this critically endangered species.

To understand the demographic responses of green turtles to seagrass decline, we examined a data set from study of a mixed-stock foraging aggregation of immature green turtles, Chelonia mydas, collected in Bermuda (32(o)18'N, - 64(o)46'W) over five decades. Average turtle size (SCLmin) and mass declined by 22.3% and 58.2%, respectively. Aggregation size structure shifted to smaller sizes and now consists of more small turtles and fewer large turtles. Density (turtles ha(-1)) increased significantly but biomass (kg ha(-1)) remained unchanged and low compared to C. mydas biomass observed elsewhere. Green turtles exhibited reduced site fidelity during two portions of the study period, suggesting increased foraging effort. Reduction in turtle body condition index and seagrass coverage occurred from offshore to inshore. Changes in aggregation composition and behavior were consistent with expectations given a documented decline in seagrass availability, combined with increased output from source rookeries. Apparent response to resource decline is traced back to 1976, well before seagrass loss was first documented. Green turtles and their primary food source (Thalassia testudinum) are at the northern limit of their range in Bermuda, where seagrasses would be expected to have a reduced tolerance for natural grazing pressure and increased susceptibility to synergistic stressors, especially temperature, bioturbation and phosphorus limitation. Our results suggest that synergistic stressors, and not green turtles alone, have produced the observed reduction in seagrasses on the Bermuda Platform. Given that seagrass declines have been reported worldwide, our findings may suggest how green turtles will respond elsewhere.

A miniaturized species is one that has endured ecological, physiological or life history costs due to small size and has implemented discrete strategies to compensate for those costs. We studied the impact of small size on the reproductive biology of the miniaturized turtle, Sternotherus minor (Kinosternidae: Kinosterninae), by exploring two alternative hypotheses that explain within-clutch trade-offs: the Optimal Egg Size Theory (OEST) and the Morphological Constraint Hypothesis (MCH). Female S. minor in this study showed a combination of reproductive parameters that support both the MCH and the OEST. Small individuals follow the MCH and larger individuals follow the OEST, fitting the previously proposed 'threshold size-constrained' model of egg size to female size. The large proportion of suboptimally-sized eggs (31.1%) produced in our study population is evidence that a novel strategy that compensates for very small size exists in this kinosternine turtle. Early reproduction in our study population, as well as a mobile plastron made up of a reduced number of bones and scutes in all members of this subfamily, is strong evidence of paedomorphosis, a frequent consequence of miniaturization. Re-examination of reproduction in other kinosternines will further test how this life history strategy facilitated miniaturization in testudines.