The long-term advance in the timing of bird spring migration in the Northern Hemisphere is associated with global climate change. The extent to which changes in bird phenology reflect responses to weather conditions in the wintering or breeding areas, or during migration, however, remains to be elucidated. We analyse the relationships between the timing of spring migration of 9 species of trans-Saharan migratory birds across the Mediterranean, and thermal and precipitation anomalies in the main wintering areas south of the Sahara Desert and in North African stopover areas. Median migration dates were collected on the island of Capri (southern Italy) by standardized mist-netting during 1981 to 2004. High temperatures in sub-Saharan Africa (Sahel and Gulf of Guinea) prior to northward migration (February and March) were associated with advanced migration. Moreover, birds migrated earlier when winter rainfall in North Africa was more abundant. The relationships between relevant meteorological variables and timing of migration were remarkably consistent among species, suggesting a coherent response to the same extrinsic stimuli. All these results were obtained while statistically controlling for the long-term trend towards the earlier timing of spring migration across the Mediterranean that has been documented in previous analyses of the same dataset, a trend that was confirmed by the present analyses. In conclusion, our results suggest that thermal conditions in the wintering quarters, as well as rainfall in North African stopover areas, can influence interannual variation in migration phenology of trans-Saharan migratory birds, although the ecological mechanisms that causally link meteorological conditions to the timing of migration remain a matter of speculation.

We identified hotspots of terrestrial vertebrate species diversity in Europe and adjacent islands. Moreover, we assessed the extent to which by the end of the 21st century such hotspots will be exposed to average monthly temperature and precipitation patterns which can be regarded as extreme if compared to the climate experienced during 1950-2000. In particular, we considered the entire European sub-continent plus Turkey and a total of 1149 species of terrestrial vertebrates. For each species, we developed species-specific expert-based distribution models (validated against field data) which we used to calculate species richness maps for mammals, breeding birds, amphibians, and reptiles. Considering four global circulation model outputs and three emission scenarios, we generated an index of risk of exposure to extreme climates, and we used a bivariate local Moran’s I to identify the areas with a significant association between hotspots of diversity and high risk of exposure to extreme climates. Our results outline that the Mediterranean basin represents both an important hotspot for biodiversity and especially for threatened species for all taxa. In particular, the Iberian and Italian peninsulas host particularly high species richness as measured over all groups, while the eastern Mediterranean basin is particularly rich in amphibians and reptiles; the islands (both Macaronesian and Mediterranean) host the highest richness of threatened species for all taxa occurs. Our results suggest that the main hotspots of biodiversity for terrestrial vertebrates may be extensively influenced by the climate change projected to occur over the coming decades, especially in the Mediterranean bioregion, posing serious concerns for biodiversity conservation.

The European Union has made extensive biodiversity conservation efforts with the Habitats and Birds Directives and with the establishment of the Natura 2000 network of protected areas, one of the largest networks of conservation areas worldwide. We performed a gap analysis of the entire Natura 2000 system plus national protected areas and all terrestrial vertebrates (freshwater fish excluded). We also evaluated the level of connectivity of both systems, providing therefore a first estimate of the functionality of the Natura 2000 system as an effective network of protected areas. Together national protected areas and the Natura 2000 network covered more than one-third of the European Union. National protected areas did not offer protection to 13 total gap species (i.e., species not covered by any protected area) or to almost 300 partial gap species (i.e., species whose representation target is not met). Together the Natura 2000 network and national protected areas left 1 total gap species and 121 partial gap species unprotected. The terrestrial vertebrates listed in the Habitats and Birds Directives were relatively well covered (especially birds), and overall connectivity was improved considerably by Natura 2000 sites that act as stepping stones between national protected areas. Overall, we found that the Natura 2000 network represents at continental level an important network of protected areas that acts as a good complement to existing national protected areas. However, a number of problems remain that are mainly linked to the criteria used to list the species in the Habitats and Birds Directives. The European Commission initiated in 2014 a process aimed at assessing the importance of the Birds and Habitats Directives for biodiversity conservation. Our results contribute to this assessment and suggest the system is largely effective for terrestrial vertebrates but would benefit from further updating of the species lists and field management.

Aim Although much has been said on the spatial distribution of taxonomic and phylogenetic diversity of vertebrates, how this diversity interacts in food webs and how these interactions change across space are largely unknown. Here, we analysed the spatial distribution of tetrapod food webs and asked whether the variation in local food web structure is driven by random processes or by natural and anthropogenic factors. Location Europe. Time period Present. Major taxa studied Tetrapods. Methods We combined an expert‐based food web (1,140 species and 70,601 links) of all European tetrapods with their respective spatial distributions. We mapped 17 different food web metrics representing complexity, chain length, vertical diversity and diet strategy across Europe and tested whether their distribution reflects the spatial structure of species richness using a null model of food web structure. To avoid multicollinearity issues, we defined composite descriptors of food web structure that we related to a set of environmental layers summarizing both natural and anthropogenic influences and tested their relative importance in explaining the spatial distribution of European terrestrial vertebrate food webs. Results Of the 17 metrics, 10 showed a non‐random spatial distribution across Europe and could be summarized along two major axes of variation in food web structure. The first was related to species richness, mean trophic level and the proportion of intermediate species, whereas the second was related to the connectance and proximity of species within the web. Both descriptors varied with latitudinal gradient. The best descriptors of food web structure were mean annual temperature and seasonality (negatively correlated with the first axis), and human footprint (positively correlated with the second axis). Main conclusions We demonstrate the importance of climate and anthropogenic pressure in shaping the spatial structure of European tetrapod food webs.

We thank the many authors of the literature sources consulted for collating this database. Without their careful work, the compilation of this data set would have been impossible. We thank Dominique Gravel for his useful comments while preparing the database.