Determining geographic patterns of migration and dispersal using stable isotopes in keratins

Abstract

Stable isotopes in metabolically inert tissues of migratory animals can be used to infer migratory and dispersal histories. The general approach for estimating geographic origins of migratory animals based on stable isotope values of their keratinous tissues is to develop or calibrate an assignment model based on tissues of known geographic origin. This paper reviews the general forms and evaluates the application of the 3 assignment approaches. Two of these approaches are considered as nominal assignment frameworks because they require prior declaration of named locations as the set of candidate origins. Individual samples can be sorted into the most likely location using a classification tree or a likelihood-based assignment test. The 3rd and more recent approach is considered a continuous assignment framework because it does not require a predetermined list of candidate locations. This approach depends on an underlying mechanistic geographic model of variation in isotope values. Such models can be developed directly from spatially intensive sampling of keratins or by calibrating a spatial model for isotopes in physical (water or soil) or biological (dietary species) resources. Productive approaches to increase spatial resolution of assignment models will use experiments designed to identify specific geographic-based, variance-generating mechanisms, especially if the contributing factors can be quantified for animals that are released back to the wild.