Temporal consistency and individual specialization in resource use by green turtles in successive life stages.

Not all individuals in a population use the same subset of dietary and habitat resources. Patterns of individual specialization have been documented in an increasing number of organisms, but often without an associated time scale over which niche specialization was observed. We examined the patterns in individual resource use through time and in relation to the population with metrics of temporal consistency and degree of individual specialization. We used stable isotope analysis of carbon and nitrogen in successive subsections of scute tissue from the carapace to compare foraging patterns in three successive life stages of the green turtle (Chelonia mydas). Temporal consistency was measured as the mean within-individual variation in stable isotope values through time, whereas the degree of individual specialization was a ratio of the individual variation to that of the population. The distinction between these two parameters is important, as the metric of temporal consistency quantifies the regularity of individual resource use, and the degree of individual specialization indicates what proportion of the population niche an average individual uses. The scute record retains a chronological history of resource use and was estimated to represent a minimum 0.8 years in juveniles to a maximum of 6.5 years in adults. Both temporal consistency and individual specialization varied significantly among life stages. Adults were highly consistent in resource use through time and formed a generalist population with individual specialists maintaining long-term patterns in resource use. Oceanic and neritic juvenile life stages exhibited less temporal consistency in resource use with less individual specialization than adults. These observations are important when considering the ecological roles filled by green turtles in each life stage; also, individual differences in resource use may result in differential fitness consequences.

Inherent variation in stable isotope values and discrimination factors in two life stages of green turtles.

We examine inherent variation in carbon and nitrogen stable isotope values of multiple soft tissues from a population of captive green turtles Chelonia mydas to determine the extent of isotopic variation due to individual differences in physiology. We compare the measured inherent variation in the captive population with the isotopic variation observed in a wild population of juvenile green turtles. Additionally, we measure diet-tissue discrimination factors to determine the offset that occurs between isotope values of the food source and four green turtle tissues. Tissue samples (epidermis, dermis, serum, and red blood cells) were collected from captive green turtles in two life stages (40 large juveniles and 30 adults) at the Cayman Turtle Farm, Grand Cayman, and analyzed for carbon and nitrogen stable isotopes. Multivariate normal models were fit to the isotope data, and the Bayesian Information Criterion was used for model selection. Inherent variation and discrimination factors differed among tissues and life stages. Inherent variation was found to make up a small portion of the isotopic variation measured in a wild population. Discrimination factors not only are tissue and life stage dependent but also appear to vary with diet and sea turtle species, thus highlighting the need for appropriate discrimination factors in dietary reconstructions and trophic-level estimations. Our measures of inherent variation will also be informative in field studies employing stable isotope analysis so that differences in diet or habitat are more accurately identified.

Annual variation in source contributions to a mixed stock: implications for quantifying connectivity.

Connectivity among populations of highly migratory species is an area of active research and is often quantified with genetic markers. We determined mitochondrial DNA (mtDNA) sequences in 350 green turtles, Chelonia mydas, in 10 annual samples over a 12-year period from an aggregation of immature green turtles in the southern Bahamas. We found significant temporal structuring in haplotype frequencies among years for all turtles and for recruits. These significant differences were reflected in substantial variation in the relative contributions from different rookeries among years estimated by a Bayesian hierarchical model. Because this foraging aggregation has been the subject of a demographic study for over 30 years, we were able to determine that, among the three potential causes of temporal structuring-differential recruitment, mortality and emigration-recruitment accounts for most of this variation. We found that estimates of connectivity and genetic diversity in sea turtle populations are affected by the level of temporal variation reported here. More studies on the extent of temporal variation in composition of mixed stocks of other migratory species are needed to determine how this affects measures of connectivity.

Intraspecific application of the mid-domain effect model: spatial and temporal nest distributions of green turtles, Chelonia mydas, at Tortuguero, Costa Rica.

The mid-domain effect (MDE) model was developed to evaluate patterns of species richness. We applied the MDE model to intraspecific distribution patterns - the spatial and temporal nest distributions of green turtles, Chelonia mydas, at Tortuguero, Costa Rica, from 1972 to 2000. Spatial and temporal distributions of green turtle nests at Tortuguero did not exhibit significant annual variation over this time period. The spatial and temporal distribution of nests largely conformed to the predictions of the MDE model, although the spatial model has a better fit. Environmental factors that may cause deviations from the MDE model are discussed. The model also indirectly provided a first estimate of the mean spatial nesting range of individual green turtles at Tortuguero: 10.1 km (SD 8.7 km). The MDE model provides insight into intraspecific as well as interspecific distribution patterns.