Wild bees benefit from low urbanization levels and suffer from pesticides in a tropical megacity

Abstract

How urbanization affects crop pollination has scarcely been studied, especially in the tropics. Here, we focus on the richness and abundance of wild bees and their pollination services to 30 small-scale fields of Lablab purpureus, a globally wide-spread grain legume, in the Indian megacity Bangalore. Farms were selected along a gradient of urbanization, measured as percentage of impervious surface (grey area) at the landscape scale, ranging from 0% to 30%. We found that the abundance of lablab-visiting wild bees increased with increasing grey proportion on the landscape scale and that, in particular, ground-nesting and large-bodied bees benefitted, as well as Xylocopa bees. The higher availability of forage and open soils in low-density urban areas appeared to enhance bee populations in Bangalore. When pollinating insects were experimentally excluded, lablab plants produced 36% less and 31% lighter fruits. Yet, we did not detect any changes of pollination outcomes along the urbanization gradient, as lablab seem to receive stable pollination services. Finally, we found that the local bee richness was negatively affected by the number of on-field pesticide applications, resulting in 35% fewer species after 3 application rounds. In summary, we conclude that low density urbanization can be beneficial for wild bees in lablab farms, but intensive pesticide use could counteract this positive effect. Large and ground-nesting farmland bees benefited most from urbanization, but more studies on different crops in tropical cities are urgently needed.