Results
A total survey effort of 3,244 km transect walk was carried out to record signs of leopard, their prey, and human disturbances. The leopard signs were detected from 70 grids with a naïve occupancy of 0.31. Wild boars were the most abundant among the prey species with records from 104 grids (48%). They were present in 49% of the grids where leopards were detected. Other prey species combined (chital, sambar, rhesus, barking deer) were present in 111 grids (52%). Lopping and encroachment were recorded on 97 grids (45%) whereas livestock was found in 117 grids (55%).
The detection probability was negatively influenced by the location of the grid (inside the protected areas) and vegetation cover (NDVI) whereas positively influenced by the presence of the livestock (Table 1, 3). The top model obtained after model averaging (table 2,3) for the occupancy included wild boar (WB), human population density (PD), ruggedness (R) and livestock (L). The model-averaged leopard occupancy (Ψ̂) in Chure was 0.5732 (0.0082 SD) with the detection probability 0.2554 (0.1142).  The β coefficient estimate indicated that the leopard occupancy was highly influenced by the presence of the livestock and the wild boar (Table 2, 3).
[Table 1,2,3,4]