A comparative analysis of crop pollinator survey methods along a large-scale climatic gradient

Safeguarding crop pollination services requires the identification of the pollinator species involved and the provision of their ecological requirements at multiple spatial scales. However, the potential for agroecological intensification of pollinator-dependent crops by harnessing pollinator diversity is limited by our capacity to characterise the community of pollinator species for each crop, and to determine how it is influenced by the different survey methods used, as well as by climatic variables at larger geographic scales. Here, we surveyed wild bees using a standardised protocol at an unprecedented scale including 62 commercial apple orchards in Western and Central Europe (i) to validate recent findings on pollinator community divergence as measured by common survey methods (netting and pan trapping) using conventional and alternative biodiversity metrics (phylogenetic and functional diversity), and (ii) to investigate the impact of climatic variation on the patterns observed. Our results confirm the significant divergence in pollinator communities measured using the two common methods at the larger, sub-continental scale, and we provide evidence for a significant influence of climate on the magnitude of pollinator community divergence (beta diversity and its turnover component) between survey methods, particularly when comparing colder to warmer sites and regions. We also found that warmer sites are more dissimilar than colder sites in terms of species composition, functional traits, or phylogenetic affinities. This result probably stems from the comparatively larger species pool in Southern Europe and because apple flowers are accessible to a wide spectrum of pollinator species; hence, two distant survey localities in Southern Europe are more likely to differ significantly in their pollinator community. Collectively, our results demonstrate the spatially-varying patterns of pollinator communities associated with common survey methods along a climate gradient and at the sub-continental scale in Europe. •Netting and pan trapping showed distinct communities at a large-spatial scale.•Contrasting communities characterised by taxonomic, functional, and phylogenetic diversities.•Climate influences divergences between communities of both methods.•Divergence between communities increased in warmer regions (increasing temperatures).•A " filtering " effect is occurring in apple orchards in Europe.