Climate and land-use interactions shape the future of an endangered butterfly in a warming world

• The modelling software MaxEnt and climate variables with a resolution of 1×1km as well as high resolution land-use and topographic variables with a resolution of 25×25m were used to research the range shift of the endangered butterfly species Brintesia circe in southwest Germany under two climate change scenarios. • The models show a substantial increase of suitable habitat for both climate change scenarios, under RCP 4.5 and 2,585 km² under RCP 8.5. Difference between climate projections are marginal though. • Meadows, protected areas and vineyards have proven the best predictors for land-use variables, while precipitation in winter, radiation in autumn and maximum temperature in winter have been the best predictors for climate variables. We put these variables in context of the species ecology and discuss the implication for future land-use management and conservation. Climate change has significantly impacted butterfly distributions in recent decades, and its influence is expected to increase with ongoing global warming. Future range shifts at the continental scale have been modelled in several recent studies, but very few studies have used high-resolution environmental data to predict butterfly distribution changes at the local and regional scale. We used such data for the first time to model the occurrence probability of the endangered butterfly Brintesia circe under current and future climatic conditions in the federal state of Baden-Württemberg. A total of 364 spatially precise species records were compiled from various sources. Species distribution modelling was conducted with MaxEnt using two types of environmental data: (1) 28 climate variables (1 km × 1 km) for 1991–2020, including projections for 2050 under RCP 4.5 and RCP 8.5; (2) fine-scale surface data (25 m × 25 m) including a digital elevation model and land-use layers. Under the current climate, areas with high suitably are small and patchy. Key climatic predictors were winter precipitation, maximum winter temperature, and autumn solar radiation, while important land-use variables included the distance to meadows, protected areas, and vineyards. Under both climate scenarios, suitable habitat area increased substantially by 2050. Our results suggest that B. circe in SW Germany could potentially benefit from climate change, if suitable grasslands with the species’ host plant remain available and are well-managed.