Discussion
This is the first comprehensive survey of leopard occupancy covering the entire Chure range (18982 km2) of Nepal. We found the spatial replicate model performed better than the standard occupancy model (supplementary 1). Our result showed that more than half of the Chure range was occupied by leopards. The leopard occupancy was primarily associated with the presence of wild boar (one of the prey species), human population density, terrain ruggedness and the presence of livestock. The detection probability of the leopard was higher outside the protected areas, less in the densely vegetated areas and higher in the areas with the presence of livestock.
The reliability of the occupancy is dependent on the detection probability of the leopard sign on the replicates (Hines et al., 2010). The value of naïve estimate occupancy (0.31) through the conventional presence-absence approach created biased in the actual occupancy because it did not consider the false absences (Fig 3). The prior consideration of leopard home range, their behaviour, the prior identification of associated covariates while designing the survey, and formation of the representative global model has helped us to obtain robust detection function and explain the pattern of leopard occupancy as well as associated environmental and ecological factors (Karanth et al., 2011).
In our result, the probability of leopard presence on 1st level replicate was more than the probability of leopard presence in a replicate occupied and which was absent in the previous replicate (θ0). Similarly, the leopard presence on a replicate was higher when the previous replicate was occupied (θ1) by the leopard. The result was consistent to Barber‐Meyer et al., 2013.
The probability of leopard occurrence in the Chure range in Terai Arc Landscape (TAL) (between Parsa National Park (PNP) in the east and Shuklaphanta National Park (ShNP) in the west (Fig 1, 2) was higher. Within TAL, there are 5 national parks with source populations of leopards. Leopards are highly adaptable in terms of foraging strategy and flexible for habitat selection in the rugged Chure area (Balme et al., 2007; Dutta et al., 2013). Similarly, all 5 national parks are the home for tiger, the apex carnivore. The tiger focused conservation activities in these protected areas has increased their number nearly twice since 2010 (DNPWC & DFSC, 2018). The increasing number of tigers in these national parks may have pushed leopards to the adjacent Chure range (Odden et al., 2010; Thapa & Kelly, 2017; Lamichhane et al., 2019a). Compared to protected areas, the Chure forests outside has a lower density of prey (Shrestha, 2004). Leopard probably avoids interspecific encounters with tigers by choosing these marginal areas (Woodroffe & Ginsberg, 1998; Lamichhane et al., 2019a). Besides the TAL area, in the eastern part of Nepal, there is only a small protected area, i.e. Koshi Tappu Wildlife Reserve (area: 175 sq km) which touches a small portion of the Chure range in the north (Fig 1). Due to this, the wildlife conservation activities are low in the eastern part. Similarly, the average forest cover in the central, western, and far-western grid of the Chure range is greater than the eastern grid. It may have reduced the prey availability and subsequently reduced the leopard occupancy in the eastern part compared to the TAL area. Hence, this study of leopard occupancy distribution helps wildlife managers and policymakers to guide for identifying locations to focus on leopard conservation in the Chure range.
Our results did not correspond to earlier findings that the leopards avoid wild boars (Eisenberg & Lockhart, 1972; Ramakrishnan et al., 1999; Hayward et al., 2006). In our study, it positively influenced the leopard occupancy in the Chure range. The leopard consuming wild boar as a diet was also observed by (Kandel et al., 2020) in the Kamdi forest corridor of the western part of the Chure range. The wild boar occurred in almost half of the surveyed grids in the Chure range, the highest among the mammal species surveyed. Leopard and wild boar co-occurred in 49 (22%) grids. (Karanth, 1999) also reported the occurrence of the leopard proportional to the wild boar. Our study showed the importance of wild boar as prey species on the occurrence of leopard. Some previous studies excluded prey species in their analysis which could have biased the result (Gavashelishvili & Lukarevskiy, 2008; Maharjan et al., 2017).
We also used other prey species (Barking Deer, Rhesus, and Chital) as covariates but their influence in the model was weak. Scarcity of prey other than wild boar in the Chure range could be the reason for this. The opportunistically placed camera traps along with this survey also photographed poachers with guns in various locations. It indicates the widespread hunting of wild prey species which have probably contributed to reducing the prey abundance.
The positive influence of the ruggedness index on leopard occupancy indicates the extensive use of rugged Chure hills by leopards. The rugged terrain provides an opportunity for ambush predators to hunt (Sharma et al., 2015). Leopards are excellent climbers and rugged terrain probably does not limit their movements/ use of the habitat. Generally flat and less rugged areas are occupied by human settlements and the rugged hills are still covered with forest providing habitat for leopards, their prey and other wildlife. However, we didn’t find the relation between vegetation cover (NDVI) and leopard occupancy. Instead, the detection probability was inversely related to NDVI. In intact forests (high NDVI value) generally, there are fewer and less visible animal trails. Detecting the leopard sign in such a forest is comparatively difficult which reduces the detection probability. The survey was conducted in the post-monsoon season, the time the leaves start shading from the deciduous trees. The fallen leaves covering the forest floor also reduces the chances of detecting the leopard sign in densely vegetated areas.
We found the positive influence of human population density and livestock on leopard occupancy. It indicates that leopards can persist in the highly modified landscape with high human population density. The findings correspond to (Athreya et al., 2013, 2016; Kuhn, 2014). The majority of the Nepalese rural community is based on agriculture and livestock is an integral part of their farm (Lamichhane et al., 2019b). Thus, livestock can be used as a proxy of human pressure in this landscape. Livestock was present in ~55% of the surveyed grid and leopard occurred in 19% of the grids with livestock presence.  It increases the chances of livestock encounters by leopards. Leopards may be depending potentially for their diet in the livestock (Kandel et al., 2020).
Leopards are specialized solitary hunters primarily hunting wild ungulates, but also kill livestock if opportunity arises (Treves & Karanth, 2003; Kandel et al., 2020).In the presence of the sufficient natural prey base, leopards tend to avoid livestock (Kolowski & Holekamp, 2006). We do not have the data on the density of prey in the Chure range but the low detection of prey signs (except the wild boar) indicates their low abundance (Smallwood & Fitzhugh 1995; Stander, 1998). In absence of enough wild prey leopards shift to livestock for diet (Khorozyan et al., 2015) which is also observed in our study with the positive influence of the livestock on leopard occupancy. So, this opportunistic predator may have followed the optimal foraging theory to minimize their search time, encounter rates and the energy cost to capture prey (Sunquist & Sunquist, 1989; Lamichhane & Jha, 2015). It indicates the possibilities of human-leopard conflict in the Chure range. We suggest that maintaining a sufficient natural prey base can contribute to minimize the livestock depredation and hence, decrease the human-leopard conflict in the Chure range. Similarly, the detection of the leopard sign was higher in the Chure range that falls outside the protected areas. It may be because the vegetation cover (NDVI) inside the national park is high in comparison to the outside area which reduces the chances of leopard sign detection.
More than half of the Chure range is occupied by leopards. We identified wild boar, human population density, ruggedness and livestock presence as top covariates influencing their occupancy that would support the policymakers, researchers, and wildlife managers to search possibilities to increase the leopard occupancy in the range. The grid wise occupancy estimate provides insight to identify the area that needs conservation actions. The positive influence on the occupancy of leopard with the presence of wild boar and livestock has indicated the importance of wild ungulates and pointed the possibilities of human-leopard conflict. The activities focusing to increase the wild prey base in the Chure range through better protection would help to reduce the livestock depredation by leopards and their retaliatory killing.
Sign based occupancy survey is a suitable method for landscape-level studies of large-ranging species like leopards. We recommend carrying out an occupancy survey in the Chure range periodically to understand leopard status as done for tigers in TAL. In future research, the exploration of the livestock depredation and human-leopard conflict data add value to understanding the dynamics of the conflict.