Skip to main content
Entomology 2021 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
In-Person
In-Person
Livestream
Livestream
On-Demand
On-Demand
Student Competition
Student Competition
Video Chat
Video Chat
Ticketed Event
Ticketed Event
Poster Icons
Student Competition
Student Competition
In-Person
In-Person
Poster
Poster
On-Demand
On-Demand
Q&A ViewAttendees 8

Plant-Insect Ecosystems

10-Minute Paper

P-IE: Pollinators 1

Toward automated monitoring of plant-pollinator interactions using low-cost, field-deployable cameras

Sunday, October 31, 2021
8:36 AM - 8:48 AM MT
Location: Colorado Convention Center, Meeting Room 102-106
In-Person

Interactions between plants and pollinators vary significantly in both space and time. Recent work has shown, for example, that plant pollinator networks show significantly flexibility over short time scales (e.g. days or weeks). Such fine-scale investigations of interaction networks could impact our understanding of their structure, function, and dynamics. However, generating empirical data on interactions between plants and pollinators at high spatial and temporal resolution remains a significant challenge. Here, we describe an approach for automated monitoring of floral visitation in the field. Our system uses deep-learning-based detection to perform real-time monitoring of floral visitation and is implemented on low-cost, open-source, field-deployable hardware. We used this system to monitor plant-pollinator interactions for nearly two months in an urban meadow in Boston, MA using 60 cameras. Combined with high-resolution mapping of dynamic changes in floral resources using drone imagery, we use this system to describe fine-scale spatial and temporal variation in interactions between pollinators and flowering plants, including the impacts of local resource abundance and fluctuating weather conditions. Finally, we assess the potential for using rapidly emerging tools in deep learning, computer vision, and open-source hardware for improving our understanding of the structure and function of plant-pollinator interaction networks.