BEYOND THE GPS TRACK: STATISTICAL MODELS LINKING ANIMAL BEHAVIOR AND HABITAT SELECTION TO INFORM WILDLIFE CONSERVATION

Matthew D Johnson; Cal Poly Humboldt; mdj6@humboldt.edu; Ayla K. Zolwik, Lauren M. Jackson, Jadzia M. Rodriguez, Ivonne Romero, Alyssa Roddy

Advances in GPS tracking technology allow researchers to quantify wildlife movements at increasingly fine spatial and temporal resolutions. Meanwhile, analytical advances provide new opportunities to translate movement trajectories into insights about animal behavior and habitat selection. Hidden Markov models (HMMs) infer latent behavioral states from movement characteristics, while integrated step selection analyses (iSSAs) quantify how animals respond to habitat features as they move through landscapes. Combining these approaches can reveal behavior-specific patterns of habitat selection that might otherwise be obscured. However, the sophistication of these tools also introduces risks for misapplication and underscores the importance of careful data preparation, appropriate model assumptions, and alignment between ecological questions and analytical approaches. This presentation will introduce a conceptual workflow for combining HMMs and iSSAs, illustrated with case studies involving hawks, swans, owls, and martens. Emphasis will be placed on the ecological and management questions these approaches can address, practical considerations for their application, and important caveats for interpreting results.

Wildlife Movement Ecology Across Taxa 
Wednesday 1:05 PM
 

Speaker Bio:

Matt Johnson is a professor in the Dept of Wildlife at Cal Poly Humboldt (formerly Humboldt State Univ) in northern California. He earned his undergraduate degree at UC Davis in Wildlife, and a PhD in Ecology at Tulane Univ. He and his students study habitat selection and habitat ecology, with a particular emphasis on birds in agricultural landscapes.