Résumé
Highlights: • We model the geographic range of the pine processionary moth (PPM). • Three species distribution models are constructed using tree-based methods. • Interpretable machine learning reveals climate variables limiting the PPM's distribution. • A significant northward expansion is projected for the coming decades. • A contraction of suitable regions is expected in North Africa by 2040–2060.Abstract: Assessing the species ecological responses to ongoing climate change is a critical challenge in environmental science. Rising temperatures, particularly in winter, are altering the distribution patterns of many species, including the pine processionary moth (PPM), Thaumetopoea pityocampa (Denis & Schiffermuller, 1775). This Mediterranean species, a significant defoliator of conifers, is expanding its range northward as winter temperatures increase. The larvae of PPM also pose serious public health risks due to their ability to induce allergic reactions in humans, pets, and livestock. To better understand these ecological shifts, we calibrated three distribution models (Bayesian Additive Regression Trees, Boosted Regression Trees, and Random Forest) based on historical and modern occurrence data compiling of 1769 points, and assessed climate suitability under historical, current and future conditions. Our results show that winter minimum temperatures, summer maximum temperatures, and solar radiation significantly influence the life cycle, and shape the geographical distribution of PPM. Under current conditions, PPM could extend its range further north, but its limited flight capabilities hinder its ability to keep up with the pace of climate change. Future projections suggest continued northward expansion, although solar radiation is expected to limit the northernmost range of PPM. Certain host tree species of PPM are frequently used as ornamental plants, particularly in urban areas, which makes the careful selection of these species a potentially valuable tool for management. Our findings identify regions that are likely to become suitable for PPM colonization, where proactive measures could be implemented.