Movement of animals on a daily, seasonal, or annual basis is a fundamental component of ecology and evolutionary biology.  Until recently, methods of studying movement primarily involved trapping and tracking extrinsically-marked individuals (e.g., radio or satellite transmitters).  Recent technological advancements and new analytical techniques have allowed researchers to use various intrinsic markers, such as biological (e.g., genetic) and biogeochemical (e.g., stable isotopes and trace elements) signatures as a means of tracking movement and migration. Like stable isotopes, trace elements are incorporated into animal tissues through  consumed prey.  However, unlike stable isotopes, most trace elements vary with the underlying geology, vegetation and land-use practice of an area.  Therefore, trace elements in tissues may reflect an individual’s movement history as it passes through chemically distinct environments. Very little empirical work has been performed to support this assertion, and has been limited to investigations of avian migration.  To determine if trace elements can resolve the movement history of non-avian animals, we studied claw keratin to determine the relationship between geography and biogeochemistry for a model organism, American badgers (Taxidea taxus).  

Results/Conclusions

We obtained claw samples from 29 badger specimens killed inadvertently in southern Ontario, Canada; all with known collection coordinates. We identified 13 elements that comprise 98% of the chemical variability along the claw length (Ca, Mg, P, Na, Al, Ba, Cr, Cu, Mn, Sr, S, Zn, Fe: n=5).  To examine potential movement paths, we collected soil samples (n=388) from areas within 10km of the collection coordinates (representing 7 habitat and 6 soil-texture types) and assayed these soils for the 13 elements of putative importance.  Only 4 (Ca, Mg, Al, Fe) of the 13 elements assayed were included in subsequent analyses do to high multicollinearity and skewness.  After log-transforming these data, we used a discriminant function analysis to determine which elements would discriminate between groups and would ultimately permit assignment of individuals to geographic origin.  The first discriminant function explained 83% and 92% of the between-group variance in habitat and soil-texture types respectively.  Significant differences in trace element composition were observed between habitat classes (Wilk’s λ=0.74, df=6, p<0.001) and soil textures (Wilk’s λ=0.47, df=5, p<0.001).  Ca and Fe contributed most to the discrimination between habitat types whereas Al contributed most to the discrimination between soil types.   The variability in soil and claw chemistry indicates that trace elements may be useful for tracking animal movement at small spatial scales, a dynamic which we illustrate visually.