ИСТИНА

The trophic position (i.e. diet) of the Arctic fox, being an opportunistic predator, reflects available prey species and can be used as an indicator for monitoring of the state of tundra ecosystems. Such monitoring requires robust methods for diet analysis. Our main objective was to evaluate the two different methods of Arctic fox diet analysis based on faces and stable isotopes (SIA) using samples from three sites in low-arctic tundra (Varanger in Norway; Yamal and Nenetskij in Russia). Faeces were collected during summer 2007 on arctic fox dens with breeding activity. Faeces from each den were analyzed separately and the proportion of dry weight for the different prey categories was calculated. The winter fur, reflecting the diets in previous summer/fall, was subjected to SIA. Samples from prey for SIA was also obtained from each site. Stable isotope ratios of carbon and nitrogen were corrected for isotopic discrimination. A Bayesian mixing model as implemented in SIAR (R software) was used to quantitatively determine the contribution of each source to a mixture. We used different prior information (equal “priors” or our knowledge about the ecosystem) and assessed their influence on the resulting diet proportions. Rodent remains vastly dominated in scats in all three sites, but the proportion of lemmings and voles varied: on Yamal arctic foxes equally used lemmings and voles in their diet; in Nenetskij voles strongly dominated over lemmings and the opposite situation was on Varanger. The configuration of prey in the isotopic space, which is specific for the ecosystem, didn’t allow to estimate their contribution on Yamal and estimated mixture proportions depended strongly on priors. In Nenetskij using a prior corresponding to the relative abundance of rodent species in the faeces modified the estimated mixture proportions, but the general results SIA remain the same. On Varanger stable isotope analysis indicated that reindeer and ptarmigan were important in summer – autumn diet of arctic foxes. These resources may have been underestimated by the faeces analysis. We show that SIA depends on the configuration of prey isotope signatures and in some cases it is impossible to distinguish between prey groups. This configuration is site specific and may limit the usefulness of SIA in certain cases. The main advantage of SIA is that it reflects assimilated prey resources and can show the importance of prey groups not leaving many remains in the faeces.