COS 135-4 - Bumble bees from different sites vary in their sensitivity to multiple agricultural chemicals

Abstract: Wild bees’ population declines are linked to multiple stressors including the widespread use of pesticides such as insecticides and fungicides which they consume while foraging from both wild and cultivated plants. Although exposure to agricultural chemicals can be harmful to many aspects of bee health, it is unclear whether population-level differences might influence bee response to pesticides in their diet. For instance, it is possible that bees that often encounter these chemicals while foraging may be less sensitive than bees that are naive to agricultural chemicals. Alternatively, bees that have not encountered these chemicals before may be healthier overall and are better able to cope with additional stressors. Here we ask if bumble bees from different sites vary in their sensitivity to exposure to multiple agricultural chemicals. To do this, we collected yellow-faced bumble bees (Bombus vosnesenskii, n=202) from three different sites in Northern Nevada which varied in their degree of urbanization (developed land, evergreen forest, and scrubland on the shore of a eutrophic lake) and ranged from 14.91 km to 30.35 km apart. Bees were transferred to lab and maintained on a 30% sucrose diet containing either the insecticide imidacloprid, the fungicide tebuconazole, the herbicide glyphosate, the combination of all three chemicals, or a water control. We then monitored bee survival for 15 days. Our survival analysis found no overall difference in bee survival by treatment, with the average bee survival being 11.32 days (p=0.68). However, there were site-level differences in bee survival and bees from the forest site survived 3 days longer on average than bees from other sites (p< 0.001). Bees from all three sites had equal survival under the control and herbicide solutions (p=0.5, p=0.74), but they differed when fed the insecticide, fungicide, and combined solution (p=0.04, 0.03, and 0.008 respectively) and bees from the forest site had consistently better within-treatment survival than the other two sites. However, in an added layer of complexity, 15% our bees had been parasitized by conopid flies (Physocephala sp.), which significantly reduced survival (p< 0.001). Further, most (86%) of parasitized bees were from the scrubland site. These results demonstrate that bees from different sites may differ in their baseline level of exposure to stressors (i.e. parasites and pesticides), which may influence their subsequent responses to pesticide exposure in controlled assays. We thus highlight the importance of considering population of origin in studies examining the effects of pesticides on wild bees.