Complementary use of stable isotopes and fatty acids for quantitative diet estimation of sympatric predators, the Antarctic pack-ice seals

• Published in: Oecologia
• Main Authors: Guerrero, A. I., Pinnock, A., Negrete, J., Rogers, T. L.
• Format: Text
• Language: English
• Published: Springer Berlin Heidelberg 2021
• Subjects: Ecosystem Ecology–Original Research; Antarctic; Hydrurga; The Antarctic; Weddell; Antarc*; Crabeater Seal; Euphausia superba; Hydrurga leptonyx; Leopard Seal; Weddell Seal
• Online Access: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8585811/; http://www.ncbi.nlm.nih.gov/pubmed/34626270; https://doi.org/10.1007/s00442-021-05045-z

The quantitative use of stable isotopes (SIs) for trophic studies has seen a rapid growth whereas fatty acid (FA) studies remain mostly qualitative. We apply the Bayesian tool MixSIAR to both SI and FA data to estimate the diet of three sympatric predators: the crabeater (Lobodon carcinophaga), Weddell (Leptonychotes weddellii) and leopard seal (Hydrurga leptonyx). We used SI data of their vibrissae and FA data of their outer blubber to produce comparable diet estimates for the same individuals. Both SI and FA models predicted the same main diet components, although the predicted proportions differed. For the crabeater seal, both methods identified krill, Euphausia superba, as the main, and almost exclusive, food item, although the FA model estimated a slightly lower proportion, potentially due to the low lipid content of krill compared to the fish species used in the model. For the Weddell seal the FA model identified the fish Pleuragramma antarcticum as the most important prey, whereas the SI model was not able to distinguish among prey species, identifying a ‘fish-squid’ group as the main diet component. For the leopard seal, both models identified krill as the main contributor; however, the predicted proportions for the secondary sources differed. Although vibrissae and outer blubber may not represent the same timeframe, the use of MixSIAR with FA data provides diet estimates comparable to those obtained with SI data, thus, both approaches were complimentary. The use of both biotracers offers a feasible option to study diets of wild animals in a quantitative manner. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00442-021-05045-z.