Quantifying variation in δ13C and δ15N isotopes within and between feathers and individuals: Is one sample enough?

Studies of avian migration increasingly use stable isotope analysis to provide vital trophic and spatial markers. However, when interpreting differences in stable isotope values of feathers, many studies are forced to make assumptions about the timing of moult. A fundamental question remains about the consistency of these values within and between feathers from the same individual. In this study, we examine variation in carbon and nitrogen isotopes by sub-sampling feathers collected from the wings of adults of two small congeneric petrel species, the broad-billed Pachyptila vittata and Antarctic prion P. desolata. Broad-billed prion feather vane material was enriched in 15N compared to feather rachis material, but there was no detectable difference in δ13C. Comparison of multiple samples taken from Antarctic prion feathers indicated subtle difference in isotopes; rachis material was enriched in 13C compared to vane material, and there were differences along the length of the feather, with samples from the middle and tip of the feather depleted in 15N compared to those from the base. While the greatest proportion of model variance was explained by differences between feathers and individuals, the magnitude of these within-feather differences was up to 0.5 ‰ in δ15N and 0.8 ‰ in δ13C. We discuss the potential drivers of these differences, linking isotopic variation to individual-level dietary differences, movement patterns and temporal dietary shifts. A novel result is that within-feather differences in δ13C may be attributed to differences in keratin structure within feathers, suggesting further work is required to understand the role of different amino acids. Our results highlight the importance of multiple sampling regimes that consider both within- and between-feather variation in studies using stable isotopes.