Key points are not available for this paper at this time.
Species distribution models (SDMs) are widely used to understand species-environment relationships and inform conservation strategies. However, when based solely on broad-scale bioclimatic variables, SDMs may fail to capture local environmental heterogeneity relevant to habitat selection. In this study, we modelled the distri bution in Italy of the common bent-wing bat (Miniopterus schreibersii), a troglophile species of conservation and public health relevance. We compared a traditional bioclimatic SDM (bio-SDM) with a hierarchical SDM (h-SDM) that integrates both coarse-scale climatic predictors and fine-resolution ecological variables. We used a maximum entropy algorithm to model the species distribution, calibrating the bio-SDM using bioclimatic predictors, while the h-SDM included variables related to roosting (e.g. distance to karstic structures, distance to urban settlements), foraging ecology (e.g. percentage of canopy cover, percentage of agriculture) and climatic suitability derived from the bio-SDM. Both models performed well under cross validation (bio-SDM AUC = 0.80; h-SDM AUC = 0.79), but the h-SDM showed significantly higher maxTSS values (0.81 vs. 0.42) and lower, more variable Boyce Index values (0.26 vs. 0.86), indicating differences in calibration despite similar discrimination ability, and reduced overpredictions in urban and intensively cultivated areas. Climatic suitability and distance to karstic structures proved essential in refining the predicted distribution for the species. These findings highlight the limitations of climate-only models and demonstrate how integrating local-scale variables can enhance spatial accuracy, providing a more informative tool for conservation planning and epidemiological monitoring in the face of climate and land-use change.
Festa et al. (Fri,) studied this question.