Awards issued prior to january 20, 2025, were funded under previous administrations and may not reflect the priorities and policies of the current administration.** bumble bees are important native…

Awards issued prior to january 20, 2025, were funded under previous administrations and may not reflect the priorities and policies of the current administration.** bumble bees are important native pollinators that play a critical role in supporting biodiversity and agricultural crop yields in the USA.mounting evidence demonstrates that exposure to common insecticides affect behavior, physiology, and performance of bees at sublethal levels well below acute toxicity. However, significant knowledge gaps remain in how these proximate impacts drive relevant outcomes in the field, such as pollination services or colony fitness. Computer vision and deep learning offer novel approaches that can easily integrate across scale (e.g., from individual to colony) and improve our understanding of complex, real-world impacts of pesticides. In this project, we will develop a low-cost, high-throughput tracking system for quantifying behavior and growth of entire colonies that can be used in both the lab and field. Using this system, we will characterize the impacts of exposure to three common insecticides on behavior in bumble bees, and how they interactwith two other common environmental stressors: temperature and nutrition. We will then assess how these same interactions combine to drive behavior and delivery of pollination services under field conditions, focusing on a single common neonicotinoid insecticide (imidacloprid). Finally, we will examine how temporal fluctuations in pesticide exposure and floral resources interact to drive behavior and growth of free-foraging bumble BEE colonies across a landscape gradient in agricultural cultivation intensity. This project will help address key knowledge gaps infactors that influence the abundance and health of a key groups of pollinators, as well developing novel methodological approaches that will be broadly applicable for studying pollinator health.