Anthropogenic modifications

We obtained data layers of known and mapped silvicultural treatments on the study areas for each year from the Forest Service Activity Tracking System (FACTS) database (https://data.fs.usda.gov/geodata/edw/datasets.php); data layer had a spatial resolution of 1 m. Management activities included (a) thinning of natural fuels, (b) commercial thinning, (c) full planting without concurrent site preparation, (d) fill-in re-planting without concurrent site preparation, (e) individual tree release and weeding, and (f) pre-commercial thinning of individual or selected trees. We calculated the area of each activity for each female home range or male area of use at the three kernel sizes, and converted to the proportion (m² of activity/km² of home range or area of use) of each activity in each home range during each year. We used m² instead of km² because some management activities were very small. The U.S. Forest Service provided locations of buildings and roads in each study area; spatial resolution of 1 m. Using GIS, we determined building density (# buildings/km²) and road density (m of roads/km²) within each female’s home range, or each male’s area of use at all three kernel sizes. We used meters for road length instead of km because some roads were <1 km. Road density consisted of paved, dirt, and decommissioned roads; the available data layer combined all of these roads into one category.