Rabbits on the road: Disentangling the factors driving the warren's abundance on motorways

The European rabbit ( Oryctolagus cuniculus ) is a keystone species in Mediterranean ecosystems but also considered a pest in some agricultural areas. Despite its threatened status due to diseases and habitat loss, rabbit populations thrive in motorway verges, causing conflicts with human activities. In this study we examine the factors affecting rabbit warren abundance in motorway verges in central Spain, with implications for conservation and management. The research aimed to assess the importance of infrastructure (e.g. motorway slopes) and landscape (e.g. land use, soil depth) factors on rabbit warren abundance along 1631 km of motorway verges and to develop an index for broader-scale abundance and risk assessment. Using generalized linear mixed models, the study revealed that both infrastructure (slope) and landscape factors (soil depth, vegetation structure and land cover gradients) significantly influenced warren abundance. Rabbit warrens were more abundant in agricultural landscapes with deep soils and in intermediate slope ranges. The findings suggest that rabbit abundance in motorway verges is driven by a combination of factors involving both infrastructure features but also land use in surrounding areas. The derived model predictions were able to correctly discriminate between crop damaged and non-damaged areas, highlighting its potential as a tool for conflict mitigation and conservation planning. The study underscores the need to integrate landscape and infrastructure features into wildlife management strategies to address human-wildlife conflicts effectively. Future work should include direct population monitoring and explore broader ecological impacts, such as predator dynamics and wildlife-vehicle collisions. • Rabbit warrens thrive on motorway verges. • Infrastructure intrinsic but also extrinsic factors influenced warren density. • Agricultural areas with deep soils and intermediate slopes favor warren abundance. • Our model predicts crop damage risk, aiding conflict mitigation. • Integrating landscape and infrastructure features aids wildlife management.