Wolverines in winter: indirect habitat loss and functional responses to backcountry recreation

Outdoor recreation is increasingly recognized to impact nature and wildlife, yet few studies have examined recreation within large natural landscapes that are critical habitat to some of our most rare and potentially disturbance-sensitive species. Over six winters (2010-2015) and four study areas (>1.1 million ha) in Idaho, Wyoming, and Montana, we studied the responses of wolverines (Gulo gulo) to backcountry winter recreation. We fit Global Positioning System (GPS) collars to 24 individual wolverines and acquired >54,000 GPS locations over 39 animal-years during winter (January-April). Simultaneously, we monitored winter recreation, collecting similar to 6000 GPS tracks (similar to 200,000 km) from backcountry recreationists. We combined the GPS tracks with trail use counts and aerial recreation surveys to map the extent and relative intensity of motorized and non-motorized recreation. We integrated our wolverine and backcountry recreation data to (1) assess patterns of wolverine habitat selection and (2) evaluate the effect of backcountry recreation on wolverine habitat relationships. We used resource selection functions to model habitat selection of male and female wolverines within their home ranges. We first modeled habitat selection for environmental covariates to understand male and female habitat use then incorporated winter recreation covariates. We assessed the potential for indirect habitat loss from winter recreation and tested for functional responses of wolverines to differing levels and types of recreation. Motorized recreation occurred at higher intensity across a larger footprint than non-motorized recreation in most wolverine home ranges. Wolverines avoided areas of both motorized and non-motorized winter recreation with off-road recreation eliciting a stronger response than road-based recreation. Female wolverines exhibited stronger avoidance of off-road motorized recreation and experienced higher indirect habitat loss than male wolverines. Wolverines showed negative functional responses to the level of recreation exposure within the home range, with female wolverines showing the strongest functional response to motorized winter recreation. We suggest indirect habitat loss, particularly to females, could be of concern in areas with higher recreation levels. We speculate that the potential for backcountry winter recreation to affect wolverines may increase under dimate change if reduced snow pack concentrates winter recreationists and wolverines in the remaining areas of persistent snow cover.