The potential responses of animal species to climate change often are assessed by correlating species occurrence or density with long-term average temperature or precipitation. These approaches overlook the effects on species’ distributions and abundances of climate extremes and the indirect effects of climate. We developed an approach for projecting responses of wildlife to future climate that explicitly accounted for the direct effects of climate extremes and the indirect effects of climate via changes in the timing and magnitude of primary productivity (henceforth phenology). We used historical climate data and remotely sensed data on phenology to develop predictive models of climate-phenology relations in desert grasslands of the southwestern United States. We also projected phenology under future climate conditions. We modeled relations between occurrence or density of four wildlife species associated with these grasslands (Gambel’s Quail [Callipepla gambelii], Scaled Quail [C. squamat], Gunnison’s prairie dog [Cynomys gunnisoni], and American pronghorn [Antilocapra americana]) and both climate and phenology variables. We then used climate and phenology projections to predict how these species will respond to changes in climate and phenology over time. Results of our analysis suggest that climate extremes and indirect effects via phenology may be equally if not more relevant than climate averages in predicting wildlife responses to climate change. However, the amount, quality, and distribution of biological and other physical dimensions of habitat complicates inference. As the resolution, accuracy, and availability of climate and phenology data increase, it likely will become more feasible to incorporate climate extremes and indirect climate effects into assessments of vulnerability or adaptive capacity. Management actions that address well-known threats to native species, including changes in land use, also may continue to provide substantive benefits despite rapid changes in climate.
