Climate change is ostensibly one of the greatest modern selective pressures, and species with sensitive life histories or physiologies must adapt, migrate, or buffer its effects to persist. Some 15–37% of species are expected to be endangered or extinct by 2050. The most vulnerable include habitat specialists, local endemics, and species with low intrinsic growth rates. Yosemite toads (Anaxyrus canorus) are one such alpine endemic, having been extirpated from up to 69% of their historical range. Several features of their natural history make them vulnerable: small population sizes, high larval mortality, infrequent breeding, and specialized, patch-limited habitat prone to premature desiccation. In addition to their role as ecosystem flagships, Yosemite toads provide a model system for the many other specialists with similar life histories that are challenged by environmental change. The goal of this dissertation is to understand how historical evolutionary processes such as lineage divergence and secondary admixture, along with current levels of genetic connectivity, are expected to shape the future of Yosemite toad persistence in the face of climate change. The first chapter reconstructs phylogeographic patterns of lineage formation and fusion during repeated bouts of Pleistocene glaciation, and showcases a role for refugia in ecological divergence. The second chapter examines three contact zones as replicate tests of the hypothesis that loci associated with incipient speciation are distinct from those that readily cross ancient lineage boundaries. The third chapter models modern genetic connectivity as a network of environmental and climatic interactions, using a novel approach that incorporates phylogeographic structure. The fourth chapter forecasts the future selective pressure of climate change, and predicts where connectivity may be a mitigating force to restore genetic diversity. My dissertation provides an example of how conservation strategies can incorporate the many temporal processes (ancient, recent, and current) that have shaped current genetic diversity patterns, and use a “total evidence” approach to predict future adaptive potential.