We evaluated the space use of leopards across the study site for 2010 and 2011 using occupancy models. Animals may
be present at a location but not detected by a camera (i.e., imperfect detection) due to a number of reasons, including
the orientation of the animal relative to the camera or obstruction by vegetation. The effects of imperfect detection on
estimates of space use and other measures, like density, can be significant (MacKenzie et al., 2002). Occupancy models
are ideal for camera trap data, because they formally account for imperfect detection and allow both the probability of an
animal occurring and being detected at a location to vary in response to covariates. Given our fine spatial scale camera
trapping grid (i.e., multiple camera trap locations within leopard home range), true absences (no detection because camera
trap not within leopard home range) and pseudo-absences (no detection though camera trap is within leopard home range)
cannot be differentiated. Thus, for our study, ‘occupancy’ is better interpreted as ‘usage’ (MacKenzie and Royle, 2005; Linkie
et al., 2006). We assume animals move randomly between the fine-scale sampling sites, which relaxes the assumption of
population closure typically required for occupancy models.