Predation by Himalayan Wolves: Understanding conflict, culture and co-existence amongst Indo-Tibetan community and large carnivores in High Himalaya

The wolves in the Hindukush-Himalayan region belong to one the most basal lineages within Canis lupus, yet little is known about its ecology, distribution, and behavior. To understand ecological aspects of wolves in this landscape, we predict wolf distribution, diet patterns and conflict perception in Spiti, India using field and remotely sensed information. We collected scats (n = 283) of canid species namely, Wolves, and other predators over a period of 3 years (2014-17) [66]. Wolf diet constituted mostly of domestic prey (79.02 %) while wild prey constituted to 17.80% of wolf diet over the three years. Village surveys recorded only 4% of the respondents confirmed wolf presence and perceived them as a possible threat to various livestock. Over, 98% of the respondents claimed that wolves were not safe for livestock and were averse to its presence. Marginal response curves depicted the model to have positive responses to animal location, LULC, village population, village density and wolf depredation. We found perceived presence/threat distribution wolves in the area significantly differed from actual ecological presence and distribution of wolves. The Himalayan wolf is an apex flagship predator in this fragile high altitude system, whose role is intricately linked with the ecology of the region. The use of such methods can aid in understanding such aspects as well as designing effective long-term conservation strategies for the species.

the United states and Canada. In Europe, wolves have been illegally hunted. In 1966, the wolf was declared functionally extinct in the Scandinavian Peninsula. [9]. Nearly, all mortalities in 48 Scandinavia, Italy, Germany and England were reported from poaching or vehicle strikes [7]. 49 In, the Asian context, wolves seemed to prey largely on domestic livestock [2,10,11]. Probable 51 reasons are, reduced availability of wild prey and changing land-use into livestock grazing grounds 52 leading to increased conflicts and retaliations towards wolves [12]. In India, however, attitudes 53 towards wolves have been less destructive. Unlike, many western notions, wildlife and wolves 54 have been tolerated in India due to many religio-cultural sentiments. Yet, persecution remains as 55 one of the biggest obstacles to wolf recovery around the world [13], including the Himalayan wolf 56 [10,14]. In many Indian landscapes, wolves continue to survive vis-a-vis disturbance and other 57 human induced factors [15]. Tolerance of communities has aided survival and depredation is seen 58 to be part of the occupation [10,16]. With unavailability of wild prey and natural habitats wolves 59 now subsist mainly on livestock and continue to live outside protected areas [17]. There have been several studies on wolves worldwide, however the Himalayan wolf is one of the 62 least studied wolf in spite of its genetic distinction and ancient lineage [2,18,19]. The current study 63 thus aims to understand aspects of Himalayan wolf ecology in context of human-wolf interaction 64 with the use of telemetry, scatological and distribution modelling techniques [20][21][22]

Data on wolf food habits
Sign surveys were conducted throughout the landscape with more than 1000 km tracked and 300 117 man days of effort. Trails, river banks, hill tops, village periphery and grazing pastures were 118 searched for signs of a large predator as well as other meso-predators. We collected scats (n = 283) occurrence [26] of species were obtained. Biomass consumed was calculated using Consumed 124 mean prey mass (kg) per wolf to excrete one collectable scat as a function of mean prey body mass 125 (x kg) provided per feeding experiment by [27] correction factor 1 (CF1), y = 1.798(1-exp(-126 0.008x)) as well as the conventional correction factor , y = 0.439 + 0.008x [28]. We compared data 127 on diet and perceived depredation by large carnivores as well as other carnivores using χ 2 test. We   We used a presence only modelling Maximum Entropy (Maxent) algorithm, which finds the 160 probability distribution of maximum entropy, that is the most spread out or closest to uniform with limited information about the target species [29][30][31]. Maxent has a potential to map the spatial 162 distribution of species with fewer locations and has performed well as compared to other available 163 presence only models [32][33][34][35].

165
Wolf presence records were obtained from data of three wolf packs that were GPS collared from  Cattle contributed the most while rodent species and birds contributed the least (Fig 3).

223
Perceived Human-Wolf conflict 224 Village surveys recorded only 4% of the respondents confirmed wolf presence and perceived them 225 as a possible threat to various livestock (Fig 2 & Fig 4). Cultures that raise livestock have strongly disliked wolves [38]. We found that wolves of Spiti 258 largely predated on large domestic prey. Similar studies from the adjoining landscape of the 259 regions reflect comparable patterns [10,16]. A review of dietary habits of grey wolves [13] showed unclear perceptions that the snow leopard in responsible for most kills of livestock [39], but this 273 practice has slowly been discontinued largely due to conservation awareness, enforcement as well 274 as religio-social sentiments of the predominant Buddhist communities of the region [11,24]. In comparison, attitudes due to perceived losses due to wolf is similar to examples from North 276 America, Finland [14,38,40].

277
Our results showed that perceived levels of livestock depredation did not emulate actual levels of 278 depredation in concurrence with earlier studies [12,14]. We found disparities in terms of type of 279 prey and quantity of livestock consumed as claimed by respondents against scatological analysis.

280
Similar findings have been report from the Northern Rocky Mountains where wolf depredation is 281 lower than expected given its exposure to domestic livestock. Wolves accounted for less than 282 0.04% of the total losses or 0.01% of all predator caused mortalities [38].

284
The simplicity and ease of use of Maxent has prompted many researchers to use the software [33]. 285 We avoided default settings and approached the modelling process to arrive at an optimal 286 regularization parameter, which may differ if push button approaches are used. The three variables 287 that were important permutation terms were animal location, village density and slope. Scat 288 locations, which correlate, well with animal location were thus a very good indication of conflict 289 hotspots that could serve as reliable data in cases where animal locational data is not available. In 290 terms of percent area of conflict, scale appeared less significant. However, it is important to note 291 that the scat presences correlated well with village population as well. Wolves operated at peaks 292 of 3300 mts with gentle slopes, which was expected. This may be also due to competition with 293 sympatric species such as snow leopard which prefer more rugged and cliff like terrain [41].

295
Therefore, conflict hotspots reflected were in areas that had low topographic heterogeneity and, 296 decreasing ruggedness. Wolves preferred areas that were undulating and gentler in their slopes.

297
Whether this is driven by competing species or niche exclusion in terms or space and prey is to be