Combining camera trap and fitness app data to assess mammal response to hiking and mountain biking trail use

Abstract

Managing lands for both wildlife conservation and accessible recreational opportunities can be a delicate balance. General trail use can disturb a range of species in a variety of well-described ways, but the specific effects of mountain biking on wildlife – an important management question – remains controversial in the literature. In the past, collecting data on recreational uses was time-consuming and expensive. Today, on-line recreation apps automatically collect these data. This study used data from a social fitness app, Strava Metro, to assess the differential effects of hiking and mountain biking on wildlife in parks in Marin County, California, USA. Combining user data with mammal frequency data from a community science camera trap project, we compared generalized linear mixed models to assess how mammals spatially and temporally responded to distance from trails and to levels of hiking and mountain biking activity. Of the 5 primarily non-nocturnal mammals in our study area, 4 were either spatially or temporally less frequent near trails: western gray squirrels (Sciurus griseus), brush rabbits (Sylvilagus bachmani), coyotes (Canis latrans), and mule deer (Odocoileus hemionus). Mule deer were also moderately sensitive to sites with high levels of mountain bike use. When accounting for both hiking and biking use, both mule deer and brush rabbits were less frequent at high-use mountain bike sites. Strava Metro provided useful data for these analyses, showing potential as a resource for managing mountain biking effects on public lands as mountain and e-bike prevalence increases.