干旱区生态环境知识资源中心

Climatic shifts may increase the extinction risk of populations, especially when they are already suffering from other anthropogenic impacts. Our ability to predict the consequences of climate change on endangered species is limited by our scarce knowledge of the effects of climate variability on the population dynamics of most organisms and by the uncertainty of climate projections, which depend strongly on the region of the earth being considered. In this study, we analysed a long-term monitoring programme (1988-2009) of Hermann’s tortoise (Testudo hermanni) aimed at evaluating the consequences of the drastic changes in temperature and precipitation patterns predicted for the Mediterranean region on the demography of a long-lived species with low dispersal capability and already suffering a large number of threats. Capture-recapture modelling of a population in the Ebro Delta (NE Spain) allowed us to assess the effect of climate variability on the survival of tortoises. Winter rainfall was found to be the major driver of juvenile and immature survival, whereas that of adults remained high and constant across the study. Furthermore, local climate series obtained ad hoc from regional climate simulations, for this and 10 additional Mediterranean locations where tortoises occurred, provided us with reliable future climate forecasts, which were used to simulate the fate of these populations under three precipitation scenarios (mean, wet and dry) using stochastic population modelling. We show that a shift to a more arid climate would have negative consequences for population persistence, enhancing juvenile mortality and increasing quasiextinction risk because of a decrease in recruitment. These processes varied depending on the population and the climate scenario we considered, but our results suggest that unless other human-induced causes of mortality are suppressed (e. g. poaching, fire, habitat fragmentation), climate variability will increase extinction risk within most of the species’ current range.