Spatiotemporal patterns of livestock predation across contrasting farming systems in an expanding grey wolf population

Livestock predation by large carnivores is widespread, particularly by the grey wolf (Canis lupus), affecting rural livelihoods and often triggering retaliatory killing that threatens wolf conservation. Understanding the drivers of livestock attacks is essential for identifying conflict hotspots and designing effective prevention strategies. We analyzed wolf depredation in two expansion areas of southwestern Europe with contrasting livestock systems: sheep and cattle. Using officially recorded damage reports from 2016 to 2024, we modeled depredation risk at two spatial scales: municipal (occurrence of damage per municipality per year) and local (individual damage locations), incorporating ecological, topographic, and management variables, including distance to wolf breeding areas, livestock and wild prey density, land-use, and relevant landscape features. Based on models, we generated risk maps at both scales. At municipal scale, sheep livestock system, risk was primarily associated with wolf-preferred habitats, livestock density, and proximity to breeding areas, whereas in cattle livestock system, elevation, water availability, and breeding area proximity were most influential. At local scale, habitat suitability for livestock and closeness to wolf breeding areas consistently predicted predation in both systems. These findings highlight the importance of tailoring conflict-mitigation strategies to local ecological and management contexts, as depredation risk varies even within individual municipalities. Integrating municipal and local scale models provides a robust framework for anticipating high-risk areas in wolf expansion ranges and guiding the implementation of preventive measures adapted to specific livestock systems. Our results offer actionable, scale-dependent insights for improving human–wolf coexistence and informing evidence-based conservation management in expanding large carnivore populations.