Deuterium (²H) has been used to track movements of land birds, under the assumption that δ²H in precipitation (δ²H_p) and δ²H in bird feathers (δ²H_f) are correlated across broad geographic gradients. The nature of this correlation has been evaluated only in a small percentage of the birds that breed in North America. We sampled Western Burrowing Owl (Athene cunicularia hypugaea) feathers of known origin (nestling feathers) at 36 locations throughout North America (from southern Canada to central Mexico). We used a modification of the “comparative equilibrium” technique of Wassenaar and Hobson (2003) to measure the δ²H of nonexchangeable hydrogen in feather samples. We characterized the strength of the relationship between δ²H_f and amount-weighted mean annual δ²H_p in a raptor that breeds in arid grasslands and deserts throughout western North America. We used a Bayesian hierarchical approach to model the δ²H_f–δ²H_p relationship, accounting for levels of intrinsic and extrinsic variation in δ²H_f. We found a linear relationship between δ²H_f and δ²H_p (δ²H_f = −13.48 0.78 δ²H_p; 95% credible interval (slope): 0.55–1.01) and used this relationship to construct a feather deuterium isoscape map. We observed relatively high levels of variation in mean δ²H_f across locations (SD = 11.01‰), due in part to variation in the contribution of precipitation to local food webs, and substantial variation among individuals within locations (SD = 6.68‰). Our data demonstrate that δ²H_f of juvenile Burrowing Owls can be used to infer local amount-weighted mean annual δ²H_p from the location of origin. Deuterium remains a valuable tool for tracking continental-scale raptor movements, with the caveat that researchers must identify and model for potential discontinuities in the δ²H_f–δ²H_p relationship in their inferences. However, isotopic discontinuities, coupled with a high relative abundance of individuals in those areas, can seriously hinder the usefulness of deuterium for identifying the origin of individual birds.