PS 24-81 - Habitat selection and connectivity of mountain lions (Puma concolor) across human-made barriers on the Olympic Peninsula: a multi-method comparison

Abstract: Designing and maintaining effective wildlife corridors is an increasingly important conservation challenge in today’s fragmenting landscapes. GPS locations from collared animals and trail camera detections of target species are two of the most common data sources for connectivity analysis. GPS data are valuable for their high accuracy and resolution but require considerably more funding and effort and can cause substantial stress to individual study organisms. Camera trapping is less invasive and less cost and effort intensive than GPS tagging but tends to provide a much coarser, population-level assessment of animal space use. Conservation and management organizations worldwide rely heavily on camera-based occupancy modeling to plan wildlife corridors, especially where resource limitations or ethical concerns prohibit the use of GPS tags. Recent studies comparing these methods suggest that both have utility for corridor planning. However, high variation in model agreement between species, season, and study area indicates a need for further study. To compare these two methods directly, I am developing GPS-based integrated step selection models and camera-based occupancy models for a population of mountain lions (Puma concolor) at risk of genetic isolation on the Olympic Peninsula in Washington State. I will map and compare connectivity estimates derived from the models and use these findings to identify potential locations for new wildlife crossing structures on and around the peninsula. Preliminary results from step selection analyses indicate that mountain lions on the Olympic Peninsula select for areas of high NDVI value and mid to low elevation. Patterns of resource selection do not appear to differ between males and females.