Models used for Red List assessments underestimate climate-related extinction risk of range-shifting species.

Climate change causes species redistribution and elevates extinction risk, making the early identification of vulnerable species critical for timely conservation. The International Union for Conservation of Nature Red List provides guidelines for assessing climate-related extinction risk using species distribution models and spatially explicit population models. Here we present a systematic evaluation of these guidelines based on simulations of virtual species with diverse life-history traits and range dynamics. We find that, while species distribution models provide adequate warning times for range-contracting species, they consistently underestimate extinction risk for range-shifting species. This is due to a concave relationship between population size and habitat loss, which contradicts the linear assumption in the Red List guidelines. Probabilistic extinction estimates from spatially explicit population models provide a belated warning for all species, particularly for highly threatened ones. Our results reveal key limitations of current Red List guidelines under climate change. Based on our findings, we provide tentative recommendations for updating the International Union for Conservation of Nature Red List guidelines.