| Camera trapping is a useful tool for monitoring wildlife across landscapes. In areas that support both recreation and wildlife, it is important to be able to estimate the amount of human recreational use as well as wildlife abundance. Our two study areas covering 11.75 sq. km included comparable public lands in the Lagunitas Creek watershed of Marin County, California. These two study areas were monitored concurrently by a total of 47 camera traps set at 0.5 km intervals. We are interested in how recreational use may influence wildlife abundance and already suspected from some preliminary analyses that one study site had greater pedestrian use, and in this study we wanted to explore metrics for estimating human abundance. Typically detection rates (detections per 100 trap nights) are used in estimating abundance using camera traps. We used occupancy modeling for wildlife but human trail use violates an assumption under occupancy modeling. Nonetheless, we were curious to see how using occupancy performed when applied to human trail use in this case and compared to rates of detection. As we suspected, pedestrian detection rates were substantially greater at one of our study sites and greater than other types of human recreational use. However, human occupancy estimates were similar between both sites, which may indicate that the area (size) of use, rather than the intensity of use, was similar. This suggests that using rates of detection as the only metric for human use could be misleading. Interestingly, wildlife occupancy estimates were similar at both study areas indicating that the area in which pedestrians are active (extent) may be more important than the actual number of pedestrians present within a site. Further analysis using covariates such as distance from trails would help in continuing to explore whether human recreation influences wildlife abundance. |