ABSTRACT Climate change is reshaping biodiversity patterns, yet conservation planning for amphibians, particularly Caudata, remains limited in both geographic coverage and the integration of multiple biodiversity dimensions. In this study, we combined species distribution models (SDMs), taxonomic, phylogenetic, and functional dimensions of biodiversity, and Zonation‐based spatial prioritization to identify conservation priorities for Caudata in China. The analysis includes species with limited occurrence data, allowing a more complete representation of the group. Multidimensional diversity shows a clear spatial structure, forming three distinct biogeographic clusters, with hotspots mainly concentrated in mountainous and karst regions. Phylogenetic and functional diversity display strong spatial congruence, suggesting that functional differentiation is largely conserved across evolutionary lineages. In contrast, overlap among priority areas based on different biodiversity dimensions is limited, and only 20.23% of grid cells are identified as high priority across all three dimensions. Current protected areas cover only a small proportion of high‐diversity regions, ranging from 7.35% to 12.57%, and this coverage is projected to decline under future climate scenarios. Although priority areas shift to some extent, their spatial patterns remain largely consistent across scenarios, with an overlap of 76.79%–84.99%. Based on this stability, we identify 14 climate‐stable refugia that represent key areas for long‐term conservation, yet only about 9% of these areas are currently protected. These results indicate that mountain systems play a dual role as centers of present‐day diversity and as refugia under future climate change. Identifying spatially stable priority areas provides a practical basis for improving protected area networks, strengthening connectivity along elevational gradients, and integrating climate change into conservation planning. This framework may also support the prioritization of other climate‐vulnerable and data‐poor taxa.
Sui et al. (Fri,) studied this question.