Tracks from a female snow leopard and cub in the Argut Valley of the Altai Republic (photo: ARKHAR)
Habitat or patch occupancy models document presence/absence and relative abundance of wildlife species. Establishing the absence of a species in a particular area is complicated— is it really absent or did the observer simply fail to find sign, when in fact the animal was in the area during the survey?
Occupancy models involve searches of sample areas (grids or polygons) of designated size over a relatively short time interval (e.g., 2-5 consecutive days) to determine if the area is being used by snow leopards. The researcher tallies the proportion of sample units at which snow leopard sign was detected during each visit to estimate the species’ overall detection probability and occupancy rate. When indexed to relevant habitat factors (e.g., slope steepness, landform ruggedness or prey abundance), these data can be more easily extrapolated to a wider area for deriving a relatively robust index of abundance (including the probabilities of occurrence). Occupancy modeling is well suited to detecting range expansions and contractions.
- Relies on sign, which is generally easily detectable
- While data compilation and interpretation requires skill, community-based monitors can be trained to undertake surveys
- Can be conducted in conjunction with SLIMS surveys
- Repeated surveys produce data analogous to camera trapping
- Does not produce actual numbers of snow leopards present
- Sign is difficult to detect in areas subject to high rainfall or containing poor tracking medium
- Requires moderate computer and statistical expertise to run the necessary algorithms