The effect of landscape characteristics on mesocarnivore occupancy in forest fragments

Knauss, Devon
Journal Title
Journal ISSN
Volume Title
University of Delaware
The objective of this study was to identify landscape characteristics which predict the presence of mesocarnivore species in forests fragmented by urban development. I used remote cameras to detect mesocarnivores in 21 forest patches in Newark, Delaware in June and July 2012. Using the program PRESENCE, I determined which of the following seven landscape characteristics were the most likely predictors of occupancy for the four species detected: proportion of multiflora rose (Rosa multiflora) cover within the plot; average distance of the plot to the nearest road and to the nearest stream; patch size; and proportions of forest, agriculture, and non-forest cover within a 100-m buffer surrounding the plot. The proportion of multiflora rose cover and proportion of surrounding forest cover were common predictors of occupancy for domestic cats (Felis catus), red foxes (Vulpes vulpes), and Virginia opossums (Didelphis virginiana). For foxes and opossums, occupancy probability increased with forest cover and decreased with rose cover. The reverse was true for cats. No landscape variables could be identified as effective predictors of occupancy for raccoons (Procyon lotor), possibly reflecting the species’ ability to adapt to a variety of habitat types. While not applicable for all mesocarnivores, the models derived from my research may be useful in predicting the occupancy of select species in forest fragments based solely on landscape characteristics.