Protecting bees from pesticides is a complex and hotly debated task. On the one hand, we hail pesticides as critical to crop production, and on the other, a key stressor in bee decline. But before we can assess these serious pros and cons, we first must determine how bees are exposed to pesticides in agricultural landscapes - a prerequisite for effect.
Basic ecological theory predicts that organisms will encounter pesticides as their activity intersects pesticide use. Bees' life-history traits determine this activity, and pesticide exposure and resulting risk are likely to vary among species. Slowly, pesticide risk assessment is evolving to capture this ecological complexity. However, this requires a better understanding of how pesticide exposure and risk vary among bee species and landscape contexts.
To tackle this, we used honey bees (Apis mellifera), buff-tailed bumble bees (Bombus terrestris) and red mason bees (Osmia bicornis) as umbrellas for species with similar ecological traits. We quantified their pesticide exposure in different landscape contexts representing various focal crops (apple, oilseed rape and red clover) and the proportion of agricultural land in the landscape. Our approach was field-intensive, sampling multiple colonies and trap nests of our focal species during and after crop bloom at 24 sites across southern Sweden.
Our study showed that bees' ecological traits interact with the landscape context to determine pesticide exposure and risk – the latter defined by toxicity-weighted exposure. However, the landscape only modified risk for limited and intermediate foragers (here, the solitary bee Osmia bicornis and the bumblebee Bombus terrestris, respectively), highlighting the potential for semi-natural habitats to buffer pesticide-related risks for wild bees.
For honey bees, there was no relationship between pesticide risk and landscape context, a result we expected given their extensive foraging ranges and ability to find and communicate mass-flowering resources. Consequently, a honey bee-based pesticide risk indicator could be a promising metric for pesticide monitoring, particularly in less intensively managed agricultural landscapes. However, questions remain as to how pesticide exposure affects individuals and, ultimately, populations of bees – tasks for a more holistic and realistic environmental risk assessment that we are currently working toward in IPol-ERA.
Within our project, MixToxBee, we are just getting started - quantifying bees' pesticide exposure is the first step in scaling farmers' pesticide use to bees' population-level effects. In doing so, we stress the "eco" in ecotoxicology, extending from individuals to populations, from the lab to the field and from single compounds to realistic pesticide mixtures. We hope our data will support the long-term safeguarding of bee populations and their pollination services and enable us to provide evidence-based and balanced information on how and when agricultural pesticide use affects bees and other pollinators.
Read the paper here.