based on the area under the curve of a receiver-operating characteristic plot (ROC plot); TI, transferability index. ABSTRACTAim To assess the geographical transferability of niche-based species distribution models fitted with two modelling techniques.Location Two distinct geographical study areas in Switzerland and Austria, in the subalpine and alpine belts.Methods Generalized linear and generalized additive models (GLM and GAM) with a binomial probability distribution and a logit link were fitted for 54 plant species, based on topoclimatic predictor variables. These models were then evaluated quantitatively and used for spatially explicit predictions within (internal evaluation and prediction) and between (external evaluation and prediction) the two regions. Comparisons of evaluations and spatial predictions between regions and models were conducted in order to test if species and methods meet the criteria of full transferability. By full transferability, we mean that: (1) the internal evaluation of models fitted in region A and B must be similar;(2) a model fitted in region A must at least retain a comparable external evaluation when projected into region B, and vice-versa; and (3) internal and external spatial predictions have to match within both regions.Results The measures of model fit are, on average, 24% higher for GAMs than for GLMs in both regions. However, the differences between internal and external evaluations (AUC coefficient) are also higher for GAMs than for GLMs (a difference of 30% for models fitted in Switzerland and 54% for models fitted in Austria). Transferability, as measured with the AUC evaluation, fails for 68% of the species in Switzerland and 55% in Austria for GLMs (respectively for 67% and 53% of the species for GAMs). For both GAMs and GLMs, the agreement between internal and external predictions is rather weak on average (Kulczynski's coefficient in the range 0.3-0.4), but varies widely among individual species. The dominant pattern is an asymmetrical transferability between the two study regions (a mean decrease of 20% for the AUC coefficient when the models are transferred from Switzerland and 13% when they are transferred from Austria). Main conclusionsThe large inter-specific variability observed among the 54 study species underlines the need to consider more than a few species to test properly the transferability of species distribution models. The pronounced asymmetry in transferability between the two study regions may be due to peculiarities of these regions, such as differences in the ranges of environmental predictors or the varied impact of land-use history, or to species-specific reasons like differential phenotypic plasticity, existence of ecotypes or varied dependence on biotic interactions that are not properly incorporated into niche-based models. The lower variation between internal and external evaluation of GLMs compared to GAMs further suggests that overfitting may reduce transferability.

No abstract

Continental-scale assessments of 21st century global impacts of climate change on biodiversity have forecasted range contractions for many species. These coarse resolution studies are, however, of limited relevance for projecting risks to biodiversity in mountain systems, where pronounced microclimatic variation could allow species to persist locally, and are ill-suited for assessment of species-specific threat in particular regions. Here, we assess the impacts of climate change on 2632 plant species across all major European mountain ranges, using high-resolution (ca. 100 m) species samples and data expressing four future climate scenarios. Projected habitat loss is greater for species distributed at higher elevations; depending on the climate scenario, we find 36-55% of alpine species, 31-51% of subalpine species and 19-46% of montane species lose more than 80% of their suitable habitat by 2070-2100. While our high-resolution analyses consistently indicate marked levels of threat to cold-adapted mountain florae across Europe, they also reveal unequal distribution of this threat across the various mountain ranges. Impacts on florae from regions projected to undergo increased warming accompanied by decreased precipitation, such as the Pyrenees and the Eastern Austrian Alps, will likely be greater than on florae in regions where the increase in temperature is less pronounced and rainfall increases concomitantly, such as in the Norwegian Scandes and the Scottish Highlands. This suggests that change in precipitation, not only warming, plays an important role in determining the potential impacts of climate change on vegetation

The expected upward shift of trees due to climate warming is supposed to be a major threat to range-restricted highaltitude species by shrinking the area of their suitable habitats. Our projections show that areas of endemism of five taxonomic groups (vascular plants, snails, spiders, butterflies, and beetles) in the Austrian Alps will, on average, experience a 77% habitat loss even under the weakest climate change scenario (11.8 1C by 2100). The amount of habitat loss is positively related with the pooled endemic species richness (species from all five taxonomic groups) and with the richness of endemic vascular plants, snails, and beetles. Owing to limited postglacial migration, hotspots of highaltitude endemics are situated in rather low peripheral mountain chains of the Alps, which have not been glaciated during the Pleistocene. There, tree line expansion disproportionally reduces habitats of high-altitude species. Such legacies of climate history, which may aggravate extinction risks under future climate change have to be expected for many temperate mountain ranges.

Aim Assessing potential response of alpine plant species distribution to different future climatic and land-use scenarios.Location Four mountain ranges totalling 150 km 2 in the north-eastern Calcareous Alps of Austria.Methods Ordinal regression models of eighty-five alpine plant species based on environmental constraints and land use determining their abundance. Site conditions are simulated spatially using a GIS, a Digital Terrain Model, meteorological station data and existing maps. Additionally, historical records were investigated to derive data on time spans since pastures were abandoned. This was then used to assess land-use impacts on vegetation patterns in combination with climatic changes.Results A regionalized GCM scenario for 2050 (þ 0.65°C, )30 mm August precipitation) will only lead to local loss of potential habitat for alpine plant species. More profound changes (þ 2°C, )30 mm August precipitation; þ 2°C, )60 mm August precipitation) however, will bring about a severe contraction of the alpine, non-forest zone, because of range expansion of the treeline conifer Pinus mugo Turra and many alpine species will loose major parts of their habitat. Precipitation change significantly influences predicted future habitat patterns, mostly by enhancing the general trend. Maintenance of summer pastures facilitates the persistence of alpine plant species by providing refuges, but existing pastures are too small in the area to effectively prevent the regional extinction risk of alpine plant species. Main conclusionsThe results support earlier hypotheses that alpine plant species on mountain ranges with restricted habitat availability above the treeline will experience severe fragmentation and habitat loss, but only if the mean annual temperature increases by 2°C or more. Even in temperate alpine regions it is important to consider precipitation in addition to temperature when climate impacts are to be assessed. The maintenance of large summer farms may contribute to preventing the expected loss of non-forest habitats for alpine plant species. Conceptual and technical shortcomings of static equilibrium modelling limit the mechanistic understanding of the processes involved.