Wooded Semi-Natural Habitats Complement Permanent Grasslands in Supporting Wild Bee Diversity in Agricultural Landscapes

Simple Summary Loss of semi-natural habitats in agricultural landscapes negatively affects wild bees. These pollinators are, however, very important in agricultural landscapes as they enable the pollination of crops and wild plants. The aim of this study was thus to understand the respective roles of different wooded and herbaceous habitats in their ability to support a diversity of wild bees. We first found that wild bee communities differed between wooded and herbaceous habitats, some bee species being found in one type of habitat and not in the other. We also showed that wooded semi-natural habitats provide some species of pollen preferred by the bees. Finally, we found that in wooded habitats there are some interactions between plant and bee species that do not happen in permanent grasslands. However, the latter also plays an important role in the diversity of bees and plants, and these wooded and herbaceous habitats complement each other. Overall, our results underline the importance of maintaining a diversity of semi-natural habitats in agricultural landscapes to maintain a diversity of wild bees and thus promote the pollination of wild plants and crops. Loss of semi-natural habitats (SNH) in agricultural landscapes affects wild bees, often negatively. However, how bee communities respond varies and is still unclear. To date, few studies have used precise descriptors to understand these effects. Our aim was to understand the respective and complementary influences of different wooded and herbaceous habitats on wild bee communities. We selected thirty 500-m radius landscapes on a gradient of a percentage of wooded SNH in south-western France. At each landscape, we sampled wild bees in spring 2016 and plants in spring 2015 and 2016 at the forest edge, in a hedgerow, and in a permanent grassland. Pollen carried by the most abundant bee species was collected and identified. Using beta diversity indices, we showed that wild bee community composition differs between the three SNH types, and especially between herbaceous and wooded SNH. Based on Jacobs' selection index, we showed that pollen of some plant species recorded in wooded SNH are preferentially selected by wild bees. Studying the impact of the loss of each SNH type on the global bee-pollen interaction network, we found that wooded SNH contributed to its resilience, enabling specific plant-bee interactions. Overall, our results underline the non-negligible contribution of wooded SNH to the diversity of wild bees in agricultural landscapes, and thus the importance of maintaining different types of SNH.