Modelling geographic distribution and detecting conservation gaps in Italy for the threatened beetle Rosalia alpina

Presence-only models can aid conservation and management of threatened, elusive species. We developed a Maxent model for the rare cerambycid beetle Rosalia longicorn Rosalia alpina L. in Italy and neighbouring regions and identified the variables best explaining the species' occurrence on a large scale. Once successfully validated, we used the model to (a) evaluate the current degree of fragmentation of R. alpina range in Italy; and (b) quantify the amount of the Italian territory with the highest probability of beetle presence within the existing national conservation areas (Natura 2000 network, parks and reserves). Low (<0.5) probability scores of R. alpina presence corresponded to 89% of the total area considered, whereas high scores (>0.9) covered only 2.5%. R. alpina was predicted to occur mostly in broadleaved deciduous forest at 1000-1700 m a.s.l. with warm maximum spring temperatures and May and November precipitation >80 mm. We found a high degree of fragmentation; gaps were mainly covered with farmland or other unsuitable habitat. Over 52% of potential habitat is unprotected. While the Natura 2000 network protects 42% of potential habitat, parks and reserve covers less than 29%. To preserve R. alpina, we urge to create, or restore, forest corridors to bridge the otherwise impermeable gaps our model detected and grant protection to the still largely unprotected area of the Italian territory e. g. by including it in further Natura 2000 sites. Models such as ours may also help focus field surveys in selected areas to save resources and increase survey success. (C) 2012 Elsevier GmbH. All rights reserved.