Understanding how species and species assemblages will respond to climate shifts is critical for predicting the impacts of modern warming on biodiversity. Addressing this question, we collected location data for four sympatric Sceloporus lizard species across a species-rich region of southern California, constructed habitat suitability models for each, and simulated temperature shifts to predict distributions at the last glacial maximum (LGM) and for a gradient of predicted end-of-century levels of warming. Current habitat modeled for Sceloporus magister comprises 243,204ha, was 357,100ha at the LGM, and may be reduced to 99,780ha at a +3°C future scenario; current habitat modeled for Sceloporus occidentalis comprises 150,870ha, was 222,590ha at the LGM, and could be reduced to 62,800ha at a +3°C; for current habitat modeled for Sceloporus vandenburgianus comprises 18,990ha, was 78,830ha at LGM, and may be 8210ha at +3°C; and habitat modeled for Sceloporus orcutti currently comprises 57,830ha, was 28,660ha at the LGM, and may be 35,970 at +3°C. Having undergone multiple climate shifts, regions recognized for high biodiversity may have intrinsic characteristics supporting that biodiversity. Our results indicate that the elevational heterogeneity of this landscape provided suitable habitat for these species throughout a past cold climate extreme and will likely continue to do so under predicted future warming. We identified climate refugia, areas providing suitable habitat under both current and shifted climates, obviating the challenge for species with low dispersal capabilities to track rapidly changing climate envelopes.