Species‐Specific Responses to Multiple Climatic Variables Predict Diverging Locations of Future Climate Change Refugia

ABSTRACT Aim We assessed the geographic and hydroclimatic patterns of species‐specific habitat suitability change for 81 plant species under diverging climate models. Location California Floristic Province, USA. Time Period 1981–2010 and 2070–2099. Major Taxa Studied 81 plant species. Methods We used species distribution models calibrated with baseline climate and species occurrence data to project habitat suitability under two climate models that reflect different magnitudes of temperature and precipitation change in California. We related species‐specific predicted changes in habitat suitability at the pixel level to four geographic features (latitude, distance to coast, elevation and topographic heterogeneity) and the projected change in five hydroclimatic variables that drive plant distributions in Mediterranean‐type ecosystems. Results Spatial patterns of projected habitat suitability change varied across species and climate models, with 59% of species exhibiting a positive relationship between habitat suitability and elevation (high elevation refugia) under a hot, dry climate model, while only ~33% showed this pattern under the warm, wet model. Habitat suitability responses to changing hydroclimatic conditions were similarly variable, but species' predicted responses to changes in minimum temperature were strong determinants of elevational and topographic positions of their refugia, regardless of climate model. Species predicted to experience habitat declines in response to increasing temperature were more likely to be associated with topographically complex refugia and high elevation habitat persistence. Main Conclusions Our results support empirical observations that conventional assumptions about climate‐driven range shifts may not hold for many plant species in California. Areas of persistent high habitat suitability for many species are predicted outside traditionally expected refugia, including lower elevations. By linking modelled habitat suitability to both geographic and hydroclimatic gradients, we provide a novel approach for characterising species‐specific climate refugia. Integrating species‐level projections with climate model uncertainty can guide the prioritisation of resilient habitats and improve conservation planning under climate change.