Search for a command to run...
Global warming is predicted to cause distributional changes in organisms whose geographic ranges are controlled by temperature. We report a recent latitudinal and altitudinal expansion of the pine processionary moth, Thaumetopoea pityocampa, whose larvae build silk nests and feed on pine foliage in the winter. In north-central France (Paris Basin), its range boundary has shifted by 87 km northwards between 1972 and 2004; in northern Italy (Alps), an altitudinal shift of 110–230 m upwards occurred between 1975 and 2004. By experimentally linking winter temperature, feeding activity, and survival of T. pityocampa larvae, we attribute the expansions to increased winter survival due to a warming trend over the past three decades. In the laboratory we determined the minimum nest and night air temperatures required for larval feeding and developed a mechanistic model based on these temperature thresholds. We tested the model in a translocation experiment that employed natural temperature gradients as spatial analogues for global warming. In all transects we transferred colonies of T. pityocampa larvae to sites within zones of historical distribution, recent distribution, and outside the present range. We monitored air and nest temperature, incoming solar radiation, larval phenology, feeding activity, and survival. Early-season temperature effects on phenology were evident, with delayed development of colonies in the more extreme (colder) sites. In the coldest months, our model was consistent with the observed patterns of feeding activity: Feeding was progressively reduced with increasing latitude or elevation, as predicted by the lower number of hours when the feeding threshold was reached, which negatively affected final survival. Insolation raised nest temperature and increased feeding activity on the south but not the north aspect. Prolonged temperature drops below the feeding thresholds occurred at all sites, leading to starvation and partial mortality. Nonetheless, even the most extreme sites still allowed some feeding and, consequently, up to 20% colony survival and successful pupation. Given that the present distribution of the oligophagous T. pityocampa is not constrained by the distribution of its actual or potential hosts, and that warmer winters will cause the number of hours of feeding to increase and the probability of the lower lethal temperature to decrease, we expect the trend of improved survival in previously prohibitive environments to continue, causing further latitudinal and altitudinal expansion. This work highlights the need to develop temperature-based predictive models for future range shifts of winter-limited species, with potential applications in management.