Abstract
White‐nose syndrome (WNS), caused by the fungus Pseudogymnoascus destructans (Pd), ranks among the most devastating wildlife diseases in recent history and is driving severe declines in North American bats. Identifying environmental factors that influence both the severity of population crashes and the potential for population persistence is essential for effective mitigation. We used a long‐term dataset from hibernacula across Michigan's Upper Peninsula to evaluate how temperature variation affected M. lucifugus populations before and after Pd introduction. Our analysis incorporated 350 surveys from 48 hibernacula, and we tested 4 hypotheses: M. lucifugus populations move to cooler roosts (mean winter temperature 2–5°C) following Pd introduction; cooler hibernacula have less severe WNS‐related population crashes; sites with less severe population crashes exhibit more positive population growth trajectories; and population trajectories after Pd introduction are influenced by site characteristics, including interior mean temperature. Consistent with earlier studies documenting shifts from warm to cool microclimates, our results showed broadscale redistribution of bats among hibernacula. Historically warm sites that once supported most bats were increasingly abandoned in favor of cooler hibernacula. Cooler mean winter temperature was the strongest predictor of this redistribution (−0.023, 95% CI −0.042 to −0.003); sites with lower temperatures had less severe population crashes during initial WNS infections (0.382, 95% CI 0.067–0.674) and exhibited more positive current population trajectories (0.16, 95% CI 0.06–0.26). These results highlight temperature as a key modulator of WNS severity and M. lucifugus persistence. Regions with limited thermal diversity and dominated by warm hibernacula (>7°C) may face heightened risk of severe declines. Our findings provide actionable insights for targeted management approaches, including microclimate‐focused manipulations, that can enhance WNS mitigation and support long‐term population stability.