In order to accurately evaluate potential shifts in species interactions under climate change, we need more timely observations in nature. Citizen science platforms provide a unique opportunity for crowd-sourced, fast-updating documentation of species interactions. Using advanced Large Language Models (LLMs), I am leading a collaborative effort on collecting species interactions from observation comments on citizen science platforms such as eBird and iNaturalist. The resulting datasets have broad spatial and temporal coverage and high spatiotemporal resolution, providing new opportunities for understanding species interactions under global change.
This is a collaborative effort with Xiaohao Yang (PhD Candidate, University of Michigan), Thabassum Hashmi Hajamaideen (undergraduate research assistant, University of Michigan), Phoebe Zarneske (Michigan State University), Brian Weeks, Kai Zhu, and an interdisciplinary team from multiple institutions.
The data extraction is supported by a team of excellent undergrads, recuited through University of Michigan’s Undergraduate Research Opportunity Program in 2024 and 2025: Yuyao (Michelle) Fan, Samantha Lukes, Jesse Perrault, Olivia Stein, and Sophia Thomas.
Full list of collaborators: Xiaohao Yang (University of Michigan), Thabassum Hashmi Hajamaideen (University of Michigan), Olivia Stein (University of Michigan), Roxanne Beltran (University of California, Santa Cruz), Benjamin Freeman (Georgia Institute of Technology), Mark Lindquist (University of Michigan), Eliot Miller (American Bird Conservancy), Summer Mengarelli (University of Notre Dame), Charlotte Probst (University of Michigan), Fernanda Valdovinos (University of California, Davis), Derek Van Berkel (University of Michigan), Phoebe Zarnetske (Michigan State University), Brian Weeks (University of Michigan), and Kai Zhu (University of Michigan).