In the face of predicted climate change, a broader understanding of biotic responses to varying environments has become increasingly important within the context of biodiversity conservation. Local adaptation is one potential option, yet remarkably few studies have harnessed genomic tools to evaluate the efficacy of this response within natural populations. Here we show evidence of selection driving divergence of a climate change-sensitive mammal, the American pika (Ochotona princeps), distributed along elevation gradients at its northern range margin in the Coast Mountains of British Columbia (BC), Canada. We employed amplified fragment length polymorphism-based genomic scans to conduct genome-wide searches for candidate loci among populations inhabiting varying environments from sea-level to 1500m. Using a highly stringent approach to outlier locus detection, we identified 68 candidate loci putatively under selection (out of a total 1509 screened), 15 of which displayed significant associations with environmental variables including annual precipitation and maximum summer temperature. These candidate loci may represent important targets for predicting pika responses to climate change and informing novel approaches to wildlife conservation in a changing world.