No abstract

We compared the effect of general circulation models and greenhouse gas emission scenarios on the uncertainty associated with models predicting changes in areas favourable to animal species. Given that mountain species are particularly at risk due to climate warming, we selected one amphibian (Baetic midwife toad), one reptile (Lataste's viper), one bird (Bonelli's eagle), and one mammal (Iberian wild goat) present in Spanish mountains to model their distributional response to climate change during this century. Climate forecasts for the whole century were provided by the Agencia Estatal de Meteorología (AEMET; National Meteorological Agency) of Spain, which adapted the general circulation models CGCM2 and ECHAM4 and produced expected temperature and precipitation values for Spain according to the A2 and B2 emission scenarios. We constructed separate models of the species response to spatial, topographic, human, and climate variables using current values of the corresponding variables. We predicted future areas favourable to the species by replacing the current climate values with those expected according to each climate change scenario, while keeping spatial, topographic and human variables constant. Fuzzy logic was used to compute the coincidence between predictions for different emission scenarios in the same global circulation model, and the consistency between predictions for the same emission scenario applying different general circulation models. In general, coincidences were higher than consistencies and, thus, discrepancies between predictions were more attributable to uncertainty in global circulation models, i.e. our insufficient knowledge concerning the effect of the oceans and atmosphere on climate, than to the putative effect of different emission scenarios on future climates. Our conclusion is that species distribution models in climate warming scenarios are still not useful for informing emission policy planning, although they have great potential as tools once consistencies become higher than coincidences.

Aim Understanding the factors that govern species' geographical ranges is of utmost importance for predicting potential range shifts triggered by environmental change. Species ranges are partially limited by their tolerances to extrinsic environmental conditions such as climate and habitat. However, they are also determined by the capacity of species to disperse, establish new populations and proliferate, which are in turn dependent on species intrinsic life‐history traits. So far, the contribution of intrinsic factors driving species distributions has been inconclusive, largely because intrinsic and extrinsic factors have not been examined simultaneously in a satisfactory way. We investigate how geographical ranges of plants are determined by both extrinsic environmental factors and species intrinsic life‐history traits. Location Europe. Methods We compiled a database on plant geographical ranges, environmental tolerances and life‐history traits that constitutes the largest dataset analysed to date (1276 species). We used generalized linear modelling to test if range size and range filling (the proportion of climatically suitable area a species occupies) are affected by dispersal distance, habitat breadth and 10 life‐history traits related to establishment and proliferation. Results The species characteristics that were most linked to range limitations of European plant species were dispersal potential, seed bank persistence and habitat breadth (which together explained ≥ 30% of deviance in range filling and range size). Specific leaf area, which has been linked to establishment ability, made a smaller contribution to native range limitations. Main conclusions Our results can be used to improve estimates of extinction vulnerability under climate change. Species with high dispersal capacity, that can maintain viable seed banks for several years and that can live in an intermediate number of habitats have the fewest non‐climatic limitations on their ranges, and are most likely to shift their geographical ranges under climate change. We suggest that climate‐change risk assessments should not focus exclusively on dispersal capacity.