Changes in Community Composition and Functional Traits of Bumblebees in an Alpine Ecosystem Relate to Climate Warming.

Climate warming has been observed as the main cause of changes in diversity, community composition, and spatial distribution of different plant and invertebrate species. Due to even stronger warming compared to the global mean, bumblebees in alpine ecosystems are particularly exposed to these changes. To investigate the effects of climate warming, we sampled bumblebees along an elevational gradient, compared the records with data from 1935 and 1936, and related our results to climate models. We found that bumblebee community composition differed significantly between sampling periods and that increasing temperatures in spring were the most plausible factor explaining these range shifts. In addition, species diversity estimates were significantly lower compared to historical records. The number of socio-parasitic species was significantly higher in the historical communities, while recent communities showed increases in climate generalists and forest species at lower elevations. Nevertheless, no significant changes in community-weighted means of a species temperature index (STI) or the number of cold-adapted species were detected, likely due to the historical data resolution. We conclude that the composition and functionality of bumblebee communities in the study area have been significantly affected by climate warming, with changes in land use and vegetation cover likely playing an additional important role.

A biophysical approach to assess weather impacts on honey bee colony winter mortality.

The western honey bee (Apis mellifera) is one of the most important insects kept by humans, but high colony losses are reported around the world. While the effects of general climatic conditions on colony winter mortality were already demonstrated, no study has investigated specific weather conditions linked to biophysical processes governing colony vitality. Here, we quantify the comparative relevance of four such processes that co-determine the colonies' fitness for wintering during the annual hive management cycle, using a 10-year dataset of winter colony mortality in Austria that includes 266 378 bee colonies. We formulate four process-based hypotheses for wintering success and operationalize them with weather indicators. The empirical data is used to fit simple and multiple linear regression models on different geographical scales. The results show that approximately 20% of winter mortality variability can be explained by the analysed weather conditions, and that it is most sensitive to the duration of extreme cold spells in mid and late winter. Our approach shows the potential of developing weather indicators based on biophysical processes and discusses the way forward for applying them in climate change studies.