Characterizing stable isotope relationships between green turtle (Chelonia mydas) skin and unhatched eggs.

ABSTRACT

RATIONALE Stable isotope analysis is used to understand the foraging habits and movements of a diverse set of organisms. Variability in stable isotope ratios among tissues derived from the same animal makes it difficult to compare data among study results in which different tissue types are evaluated. Isotopic relationships between two green turtle (Chelonia mydas) tissue types, skin and unhatched egg contents are unknown. Similarly, few data exist to evaluate the influence of time elapsed after oviposition (as a proxy for decomposition) on isotopic variability among unhatched eggs within the same nest. METHODS: Skin and unhatched egg contents were collected from 69 adult female green turtles and associated nests at the Archie Carr National Wildlife Refuge in Florida, USA. Values of δ13 C, δ15 N, and δ34 S were measured for both tissue types using a continuous flow isotope ratio mass spectrometer. Standardized major-axis (SMA) regression was used to generate conversion equations of carbon, nitrogen, and sulfur isotope ratios between the two tissue types. Model selection frameworks consisting of single-factor linear models were employed per isotope ratio to assess how egg time-in-nest affected intraclutch isotopic variability. RESULTS: Conversion equations for all three isotope ratios indicated significant relationships between skin and unhatched egg values, although model fits were lower than found in some studies examining similar patterns in other marine turtle species. The probability of increased intraclutch variability was significantly higher among eggs collected at longer intervals after deposition. CONCLUSIONS: This study reports the first-ever δ13 C and δ15 N conversion equations between skin and unhatched eggs for green turtles, and the first δ34 S conversion equation for any marine turtle species. SMA regression was used to directly convert tissue values bidirectionally, unlike equations generated using ordinary least-squares regression. Issues with increased intraclutch variability at later excavation dates highlight the importance of collecting unhatched eggs as soon as possible after hatchling emergence.