Landscape context and habitat type as drivers of bee diversity in European annual crops

• Published in: Agriculture, ecosystems & environment ecosystems & environment, 2009-09, Vol.133 (1), p.40-47
• Main Authors: Carré, Gabriel, Roche, Philip, Chifflet, Rémy, Morison, Nicolas, Bommarco, Riccardo, Harrison-Cripps, Jenn, Krewenka, Kristin, Potts, Simon G., Roberts, Stuart P.M., Rodet, Guy, Settele, Josef, Steffan-Dewenter, Ingolf, Szentgyörgyi, Hajnalka, Tscheulin, Thomas, Westphal, Catrin, Woyciechowski, Michal, Vaissière, Bernard E.
• Format: Article
• Language: English
• Subjects: Abundance; Agricultural Science; agroecosystems; Agronomy. Soil science and plant productions; Annual crops; annuals; Apiformes; Apoidea; Bee biodiversity; Biological and medical sciences; Classification; Communities; Crops; Cucumis melo; dryland farming; Ecology; Ekologi; Environmental Sciences; Environmental Sciences related to Agriculture and Land-use; Europe; field crops; Fragaria; fruit crops; Fundamental and applied biological sciences. Psychology; General agroecology; General agroecology. Agricultural and farming systems. Agricultural development. Rural area planning. Landscaping; General agronomy. Plant production; Generalities. Agricultural and farming systems. Agricultural development; geographical variation; grain crops; Habitats; insect communities; Jordbruksvetenskap; land use change; Landscape ecology; Landscapes; Miljö- och naturvårdsvetenskap; Pollinators; population dynamics; regression analysis; shrublands; species diversity; urban areas; vegetation cover; woodlands
• ISSN: 0167-8809; 1873-2305; 1873-2305; 0167-8809
• DOI: 10.1016/j.agee.2009.05.001

To better understand the dynamics of bee populations in crops, we assessed the effect of landscape context and habitat type on bee communities in annual entomophilous crops in Europe. We quantified bee communities in five pairs of crop-country: buckwheat in Poland, cantaloupe in France, field beans in the UK, spring oilseed rape in Sweden, and strawberries in Germany. For each country, 7–10 study fields were sampled over a gradient of increasing proportion of semi-natural habitats in the surrounding landscape. The CORINE land cover classification was used to characterize the landscape over a 3 km radius around each study field and we used multivariate and regression analyses to quantify the impact of landscape features on bee abundance and diversity at the sub-generic taxonomic level. Neither overall wild bee abundance nor diversity, taken as the number of sub-genera, was significantly affected by the proportion of semi-natural habitat. Therefore, we used the most precise level of the CORINE classification to examine the possible links between specific landscape features and wild bee communities. Bee community composition fell into three distinct groups across Europe: group 1 included Poland, Germany, and Sweden, group 2 the UK, and group 3 France. Among all three groups, wild bee abundance and sub-generic diversity were affected by 17 landscape elements including some semi-natural habitats (e.g., transitional woodland-shrub), some urban habitats (e.g., sport and leisure facilities) and some crop habitats (e.g., non-irrigated arable land). Some bee taxa were positively affected by urban habitats only, others by semi-natural habitats only, and others by a combination of semi-natural, urban and crop habitats. Bee sub-genera favoured by urban and crop habitats were more resistant to landscape change than those favoured only by semi-natural habitats. In agroecosystems, the agricultural intensification defined as the loss of semi-natural habitats does not necessarily cause a decline in evenness at the local level, but can change community composition towards a bee fauna dominated by common taxa.