Reply to Hussey et al.: The requirement for accurate diet-tissue discrimination factors for interpreting stable isotopes in sharks

Abstract

We recently published data describing stable isotope turnover rates and urea effects on nitrogen isotope discrimination in elasmobranchs. Given a lack of experimental evidence for urea effects on tissue nitrogen isotope values, we concluded that published nitrogen discrimination factors for teleost fishes may be transferable to elasmobranchs. Hussey et al. (Hydrobiologia, 2010a) challenge this conclusion, citing evidence from recent estimates of diet-tissue discrimination factors in sharks (Hussey et al., Biochem Physiol A 155:445–453, 2010b) and a subsequent modeling analysis of dietary proportion estimates for captive sand tiger sharks using a range of discrimination factors. We agree that researchers should use discrimination factors based on the closest possible model species and thus should use estimates by Hussey et al. (2010b) in studies of large predatory sharks rather than values derived from teleosts. However, since our study showed no effects of urea retention on nitrogen isotope values, we argue that the broad body of literature on teleost discrimination factors currently offers useful information for isotope studies on the diverse assemblage of elasmobranch species with vastly different diets, growth rates, and lifestyles relative to the species examined in the Hussey et al. (2010b) study. Furthermore, we feel that the modeling exercise performed by Hussey et al. (2010a) is misleading since it is based on a subset of the same data used to generate their initial discrimination factor estimates and attempts to estimate diet among sources with similar isotope values. We feel that the modeling exercise highlights another topic relevant to this discussion: the importance of quantifying variability in isotope studies. Published studies have demonstrated variability in isotope discrimination factors across species, tissue types, diets, and metabolic rates. While accurate discrimination factors are important for use in dietary mixing models and trophic position estimates, incorporation of associated variability is critical in order to reveal uncertainty in these estimates. In the absence of relevant elasmobranch data, researchers should use teleost-based discrimination factors but also include estimates of uncertainty in such values. Recently developed isotope mixing models (MixSIR and SIAR) allow such uncertainty to be incorporated in dietary proportion estimates.