Skip to main content
Entomology 2024 Main banner
Bayer Crop Science
Icon Legend
 
Oral Presentation
Oral Presentation
Poster Display
Poster Display
Student Competition
Student Competition
Ticketed Event
Ticketed Event
Back

Program Symposium

Plant-Insect Ecosystems

Systematics, Evolution, and Biodiversity

Empowering Insect Science, Management and Conservation Through Emerging Technologies

Landscape bee-havior: Using automated monitoring to understand pollinator behavioral dynamics in complex landscapes

Tuesday, November 12, 2024
10:35 AM – 10:55 AM MST
Location: Phoenix Convention Center, 124 AB, PCC
Oral Presentation
Bees and other pollinators show remarkable behavioral flexibility that allows them to respond dynamically to environmental conditions and resource availability. Similarly, interactions between plants and pollinators show significant flexibility over rapid time scales (e.g. days or weeks). While flexibility in foraging behavior and interactions has important implications for the structure, function, and dynamics of plant-pollinator networks, generating empirical data on interactions between plants and pollinators at sufficient spatial and temporal resolution to characterize these dynamics remains a significant challenge. However, the rapid advance of automated and scalable computer-vision approaches for detecting and identifying insects holds enormous potential to help fill this knowledge gap. Here, we describe an approach for automated monitoring of floral visitation in the field. Our approach uses deep-learning-based detection to perform scalable monitoring of floral visitation and is implemented on low-cost, open-source, field-deployable hardware (Raspberry Pis). We demonstrate the potential of this approach in both natural meadows and agricultural ecosystems using dozens of cameras monitoring floral visitation over several months. We then use this system to quantify effects of short-term weather (i.e., temperature) fluctuations on activity and ecological interaction patterns such as network-level specialization, as well as explore the importance of microclimatic variation for foraging dynamics in agro-ecosystems. Overall, our work highlights the potential of low-cost AI-based for unraveling the dynamics of bee behavior at a landscape scale. Finally, we discuss the prospects for adopting similar approaches for both large-scale ecological monitoring and experimental approaches to insect behavioral ecology.