22 Years of Reported Snow Leopard Depredation

Home , Hemis National Park

Factors contributing to a striking shift in human–wildlife dynamics in Hemis National Park, India: 22 years of reported snow leopard depredation

P USHPINDER S. JAMWAL,JIGMET T AKPA and M ICHAEL H. PARSONS

Abstract Hemis National Park of the Trans-Himalayas is Introduction home to a large population of the snow leopard Panthera uncia and increasing numbers of agro-pastoralists. To emis National Park of the Indian Trans-Himalayas is persist in this harsh terrain, farmers have to either farm live- Hknown for its rugged terrain, aridity and relative in- stock or hunt free-ranging, native ungulates. The availability of accessibility to people other than adventure enthusiasts more livestock and fewer natural prey created a dynamic and indigenous populations. Within the Park there is a whereby snow leopards depredated livestock, followed by re- dynamic between wildlife and the community of mostly taliatory killing of snow leopards. In , to assist farmers and Buddhist famers. This National Park is one of the largest wildlife, the government enacted a cost-compensation federally-protected areas in the region and supports a high scheme. Following a decade with marginally fewer depreda- density of Vulnerable snow leopards Panthera uncia (Fox  tion events, in , two additional strategies were implemen- et al., ). However, despite its prominence for wildlife ’ ted: predator-proof holding pens and the Himalayan conservation, the Park s rural economy has traditionally Homestay Programme. We assessed  years (–)of been driven by agro- and nomadic pastoralism, which re- depredation data, comparing the periods before and after quires the import of large numbers of livestock, supported the additional initiatives. Government records showed that by occasional hunting of native wildlife (Namgail et al.,  during –, , livestock were depredated from  a). The Park has become home to increasing numbers sites, with c. USD , paid as compensation. There were of livestock, especially sheep, as a result of the ever-growing  significantly more kills annually before (a mean of ) than demand for pashmina wool (Bhatnagar et al., ). For after (.) the initiatives, and mass killings ($  animals killed similar reasons, the livestock population on the broader  per attack) were significantly reduced from . to . events per Changthang Plateau doubled over a -year period follow- –  year. Goats and sheep (%) and horses (%) comprised the ing the Indo Chinese war in , and has continued to in-  majority of losses. The marked reduction in depredation oc- crease (Bhatnagar et al., ). ’ curred whilst regulations against hunting were being enforced, Not only has the snow leopard s range contracted as a re- probably resulting in an increase in the number of wild prey as sult of human and livestock expansion, but there have been alternative food. We conclude that together, cost-compensa- reductions in its natural prey from overhunting (Bagchi &  tion, tighter hunting regulations, improved holding pens Mishra, ). Some prey species, such as the Tibetan gaz- and the Homestay Programme helped support the well- elle Procapra picticaudata, have undergone a dramatic range being of the community while aiding conservation efforts. reduction in the nearby Ladakh region, and this will impact natural predators (Namgail et al., a). Prior to the re- Keywords Compensation scheme, human–wildlife dynam- gion’s development, snow leopards relied on stealth and sur- ics, indigenous communities, livestock depredation, prise to take exposed or weakened small- to medium-sized Panthera uncia, species prey (Anwar et al., ), including the Asiatic ibex Capra Supplementary material for this article is available at ibex sibirica, Ladakh urial Ovis vignei vignei, bharal https://doi.org/./S Pseudois nayaur and Tibetan argali Ovis ammon hodgsoni (Mallon, ). However, with fewer natural prey available, snow leopards have found alternative sustenance from local  PUSHPINDER S. JAMWAL* Himalayan Biodiversity Foundation, Himachal Pradesh, livestock (Bagchi & Mishra, ), including larger animals India such as cattle and horses (Sangay & Vernes, ).

JIGMET TAKPA Department of Renewable Energy, Leh, Ladakh, India, and Although livestock, particularly small- to medium-sized Department of Wildlife Protection, Leh, Ladakh, India animals, comprise the majority of the diet of most carni- MICHAEL H. PARSONS (Corresponding author) Department of Biological Sciences, vores in the Trans-Himalayan region (Mallon, ; Fordham University, Bronx, New York 10458, USA   E-mail [email protected] Chundawat & Rawat, ; Bagchi & Mishra, ), they have been implicated in up to % of the snow leopards’ *Current address: Department of Biosciences and Territory, Università degli  Studi del Molise, Pesche, Italy diet (Anwar et al., ). Across the Trans-Himalaya, with Received  October . Revision requested  November . few alternative vocations, pastoralism is the norm, and live- Accepted  November . First published online  April . stock are abundant. Therefore, despite its status as a flagship

Oryx, 2019, 53(1), 58–62 © 2018 Fauna & Flora International doi:10.1017/S0030605317001892 Downloaded from https://www.cambridge.org/core. IP address: 43.130.79.121, on 09 Oct 2021 at 02:17:21, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605317001892 Snow leopard depredation 59

species for conservation in Asia (Buzzard et al., ), the time livestock holding pens (Jackson & Wangchuk, ). snow leopard has become one of the most persecuted Concurrently, in , the Himalayan Homestay pro- large felids globally (Bagchi & Mishra, ). gramme was introduced (Jackson, ), offering training Despite the presence of carnivores in this region such as in cooking, hygiene, English, and hospitality and manage- the Tibetan wolf Canis lupus filchneri, lynx Lynx isabellina, ment, while providing loans and start-up funds for renova- wild dog Cuon alpinus and red fox Vulpes vulpes (Namgail tions (Peaty, ). Not only did this programme encourage et al., b), the snow leopard has traditionally been impli- alternative careers, but the presence and international ap- cated in the majority of livestock depredation (Bhatnagar peal of the snow leopard led directly to additional income. et al., ). This is probably because it is the predator The holding pen programme was intended to reduce the that most commonly enters corrals and night-time holding number of livestock kills, and the Homestay programme pens (Bhatnagar et al., ). Pens have typically been com- was intended to lessen the number of retaliatory kills prised of low-lying rock walls intended to retain livestock, while also providing alternative jobs in eco- and adventure with little regard for keeping predators out. Once inside tourism. These programmes were implemented while hunt- such a pen, a snow leopard may slaughter many livestock ing was being more tightly regulated. in a single attack. These losses result in emotional distress These programmes are potentially promising for im- and financial loss for herders, and often ends in retaliatory proving the perception of conservation among famers, ei- killing of snow leopards (Mallon, ). Reprisals against ther on their own or synergistically, and baseline data are carnivores usually include trapping and meat poisoning needed on the trends and extent of depredation by snow leo- (Mallon, ), and some incidents have resulted in five pards to quantify the success of these initiatives (Bhatnagar snow leopards killed for each livestock lost (Sharma et al., et al., ). Here, we quantify the number of reported dep- ). These illicit retaliatory events are one of the primary redation events and mass kills in the first  years of the threats to wildlife conservation in the region (Jackson & compensation scheme, focusing on the -year periods be- Wangchuk, ). fore and after , and identify the species and life-stage Since the end of the Indo-Chinese war there has been an of prey reported as consumed (Mishra, ). In addition, increase of , livestock in Hemis National Park, with we examine whether depredation events varied with metrics  numbers as high as , livestock per km . Even the remot- that have not changed over  years, such as elevation, est pasture is now being utilized for livestock grazing ground cover and distance to nearest water sources. (Bhatnagar et al., ). However, although there are more livestock than formerly, grazing a larger portion of the land, they are collectively owned by a greater number Study area of people. Thus, the individual family unit does not neces-  sarily own more animals, and individual families can be de- The , km Hemis National Park (Fig. ) lies at ,– vastated by a single loss (Bhatnagar et al., ) , m altitude in Ladakh, Jammu and Kashmir, India. In  the local government enacted a compensation In the rain shadow of the Himalayas, this arid region scheme to help protect pastoralists and wildlife. Yet, in receives , . mm precipitation per year. Summer tem- , the compensation only amounted to % of the perature highs range from  °C in the shade to  °C in value of the livestock lost, and also required villagers to the sun, and can drop at night to − °C, with a mean of make long treks to file a report (Bhatnagar et al., ). In  °C during daylight hours. High winter snowfall typically other regions, where cash crops are grown in some pastures, leads to the closure of access roads into the area. to provide additional income, compensation schemes have Vegetation includes riverine scrub Salix–Myricaria– been more welcome (Bagchi & Mishra, ). In Hemis Hippophae, scrub-steppe (Artemisia, Caragana, Juniperus National Park, however, in the absence of other sources of and Lonicera) on gentle slopes, herbaceous and grassy mea- income for farmers, the first  years of the compensation dow species (Potentilla, Festuca—Stipa, Carex and Kobresia) scheme were only marginally effective in lowering the num- along valley bottoms in clay soil, and Draba, Rhodiola, ber of livestock killed and the residents’ tolerance of their Christolea and Saussurea spp. in arid alpine pastures losses. above , m. The snow leopard Panthera uncia and As a result of the poor reception of the compensation Tibetan wolf Canis lupus filchneri are the main carnivores, scheme, alternative strategies to reduce depredation events with the lynx Lynx isabellina, wild dog Cuon alpinus and red and retaliatory killings were sought. Corrals were formerly fox Vulpes vulpes occasionally reported (Namgail et al., designed to keep stock in rather than to keep predators a,b). There are few cash crops, and livestock are utilized out (Mishra, ; Bhatnagar et al., ), and .  kills for wool, dairy and meat. The primary livestock are sheep can occur in a single event inside a single holding pen and goats, with some yaks, yak hybrids (dzo, male; (Jackson & Wangchuck, ). A pilot project was therefore dzomo, female) and horses. Almost all livestock are grazed initiated in  to enhance predator-proofing of night- in pastures at higher elevations in the summer, lower

FIG. 1 Hemis National Park and associated villages, with the total number of depredation events by snow leopards Panthera uncia,as reported per village during – (circle size is proportional to the number of events).

elevations in the winter and retained in holding pens at night (Bhatnagar et al., ). The snow leopard is the only carnivore routinely reported under the compensation scheme.

Methods

We obtained records of depredation events and remunera- tions through the Department of Wildlife Protection, Jammu and Kashmir. We assessed  logged entries that FIG. 2 Total number of reported depredation events, and total included year, name, village, latitude, longitude, type and number of annual kills (some events included loss of multiple life stage (adult/juvenile) of animal. The data were collected animals) of livestock by snow leopards in Hemis National Park (Fig. ) during –. from Department of Wildlife Protection remuneration sur- veys administered during January–October  by  re-  searchers and wildlife guards who went door to door in Results  villages (Fig. ). Because of the importance of each kill, and the residents’ ability to maintain records of their losses, During the first  years of the compensation scheme (– this data remains available within each village and family ) a total of , livestock were depredated from  kill- unit. We used data from  onwards because that is sites. There was a significant difference between the periods when the compensation scheme was initiated. prior to and following the additional initiatives of , with We used the Kruskal Wallis test (with Minitab v. ,State a mean of  and . kills per year in the  years before and College, USA) to examine differences between number of after the initiatives (H()=.,P, .; Fig. ), and also kills, and number of mass killings (defined as $  animals a significant difference in the number of mass killings per killed at one time), before and after the additional initiatives year before (.) and after (.)(H()=.,P, .). of . We also examined the influence of elevation, land- There were no clusters of kills with elevation, land cover cover and streams on depredation of livestock, using ArcGIS or proximity to streams.  (ESRI, Redlands, USA). DIVA-GIS v. . (Hijmans et al., Approximately USD , was paid as compensation ) was used to extract elevation data from SRTM  m over the  years. The majority of livestock losses (Table ) digital elevation data (Gorokhovich & Voustianiouk, ). were of goats and sheep (%), with horses accounting for Land use and land cover data were converted to raster for- an additional % of all losses. No other livestock species ac- mats, and data for locations of streams were extracted after counted for more than %. Juveniles accounted for .%of processing of the Digital Elevation Model. all losses. More than half of all kills (.%; Supplementary

TABLE 1 Livestock species/type and life stage (adult or juvenile) for each reported depredation event over a -year period (–) by the snow leopard Panthera uncia in Hemis National Park during –.

Livestock Adult/ No. of incidents Mean no. of Max. no. of animals Total no. of % of total by predated juvenile reported animals ± SD per incident animals predated type/species Cattle Bos taurus A 33 1.15 ± 0.36 2 38 6.9 J 6 1.33 ± 0.51 2 8 Goat Capra hircus A 164 4.92 ± 5.61 41 808 30.6 J 9 3.55 ± 4.1 12 32 Sheep Ovis ammon A 140 3.42 ± 3.09 16 479 26.2 J 8 3.38 ± 2.07 7 27 Dzo (hybrid yak/cattle) A 29 1 ± 0 1 29 5.5 J2 1± 01 2 Dzomo (female dzo) A 17 1 ± 0 1 17 3.0 J0 Yak Bos grunniens A 32 1.19 ± 0.39 2 38 5.8 J1 11 1 Demo (female yak) A 16 1.06 ± 0.25 2 17 2.8 J0 Horse Equus caballus A 70 1.11 ± 0.32 2 78 12.9 J3 1± 01 3 Donkey Equus asinus A 35 1.34 ± 0.59 3 47 6.2 J0 Total A 519 2.89 ± 3.84 1551 95.5 J 46 2.51 ± 2.68 73 4.5 1624 100

Table ) were reported in January–March, and the fewest economic interest of indigenous communities and the wel- kills (.%) were reported in October–December. fare of their livestock (Jackson & Wangchuk, ). Among the  villages (Fig. ) there was one hotspot of Although the success of the initiatives could be attributed kills, in the central valley, in the villages of Markha and to improvements of the night-time holding pens, our meas- Sku, where almost half of all kills took place (Supplementary ure was animals reported as depredated for compensation Table ). During –, the numbers of reported kills purposes. Thus, as simultaneously there was increased in- rose slightly, and goats were replaced by sheep as the prey come from the Homestay programme, it is possible that most often reported consumed (data not shown). more livestock were killed than were reported. For example, villagers may not wish to publicize a depredation event for Discussion fear of encouraging retaliatory killing of a species that is now generating income. We report here a striking reduction in reported depredation Additionally, we do not know how the stricter regula- of livestock by the snow leopard following initiatives in  tions controlling hunting of snow leopard prey influenced that were intended to support the compensation scheme. depredation. For instance, the relationship between the Reports of the numbers of livestock depredated in the pre- numbers of wild ungulates and intensity of predation can ceding decade, compared to the decade following these in- be complex: increased populations of wild ungulates may itiatives, differed by an order of magnitude. Almost %of initially result in an increase, and later a stabilization, of po- all kills occurred prior to the  initiatives, with only two pulations of snow leopards (Suryawanshi et al., ). mass killings reported since. We were initially surprised that the majority of livestock Data on the extent of retaliatory killing are unavailable. reported consumed were adults, because, like most leopards, There is a moratorium on killing carnivores, and thus any snow leopards prefer smaller prey (Lyngdoh et al., ). individuals reporting this information would either indict However, killing of juveniles receives less compensation themselves or be complicit in the possible prosecution of than killing of adults, and thus many killings may have other pastoralists (Mallon, ). However, it is reasonable gone unreported. The shift in species depredated during to conclude that as fewer depredation events occur, public – may have been related to a shift in the number perception regarding conservation will improve, and there and types of livestock. For example, there may now be will be fewer retaliations against the snow leopard. Thus, fewer sheep available as farmers move to larger pack animals the  initiatives appear to have been successful in redu- to support adventure tourism, and goats are the next smal- cing the numbers of killings whilst also protecting the lest potential prey available.

The data we have summarised here appear to support Plateau Institute of Biology (eds J.L. Fox & D. Jizeng), pp. –. the overall strategic approach of reducing hunting of snow leop- International Snow Leopard Trust, Seattle, USA.  ard wild prey and retaliatory killing of snow leopards through a FOX, J.L., SINHA, S.P., CHUNDAWAT, R.S. & DAS, P.K. ( ) Status of the snow leopard Panthera uncia in northwest India. Biological combination of improved regulations and night-time holding Conservation, , –. pens, and the Homestay Programme. However, other factors GOROKHOVICH,Y.&VOUSTIANIOUK,A.() Accuracy assessment such as numbers of livestock, species composition of flocks, of the processed SRTM-based elevation data by CGIAR using field variation in numbers of natural snow leopard prey, levels of il- data from USA and Thailand and its relation to the terrain  – licit hunting, and tourism pressure may also have influenced characteristics. Remote Sensing of Environment, , . HIJMANS, R.J., GUARINO, L., BUSSINK, C., MATHUR, P., CRUZ, M., our findings. Future research could seek to disambiguate BARRENTES,I.&ROJAS,E.() DIVA-GIS v. .. Http://www. these factors as well as carrying out more in-depth socio- diva-gis.org [accessed  January ]. ecological studies to provide additional context for our findings. JACKSON,R() Fostering community-based stewardship of wildlife in Central Asia: transforming snow leopards from pests into valued assets. In Rangeland Stewardship in Central Asia: Balancing Acknowledgements Improved Livelihoods, Biodiversity Conservation and Land Protection (ed. V. Squires), pp. –. Springer, Dordrecht, The Netherlands. We thank the Department of Wildlife Protection, Jammu JACKSON,R.&WANGCHUK,R.() Linking snow leopard and Kashmir, for permission to conduct this study and conservation and people–wildlife conflict resolution: grassroots for field support, Prameek M. Kannan of Pace University measures to protect the endangered snow leopard from herder retribution. Endangered Species Update, , –. for his valuable support, and Yash Veer Bhatnagar and two JACKSON, R.M. & WANGCHUK,R.() A community-based anonymous reviewers for their constructive suggestions. This approach to mitigating livestock depredation by snow leopards. project was self-funded by the authors. Amir Javaid of the Human Dimensions of Wildlife, , –. Institute of Space Technology, Pakistan, and Faith LYNGDOH, S., SHROTRIYA, S., GOYAL, S.P., CLEMENTS, H., HAYWARD,  E. Parsons of Columbia University, USA provided GIS ser- M.W. & HABIB,B.( ) Prey preferences of the snow leopard (Panthera uncia): regional diet specificity holds global significance vices. All authors declare they have no conflicts of interest. for conservation. PloS One, ,e. MALLON, D.P. () Status and conservation of large mammals in Ladakh. Biological Conservation, , –. Author contributions MISHRA,C.() Livestock depredation by large carnivores in the Indian trans-Himalaya: conflict perceptions and conservation PSJ and JT designed the study and collected all data. MHP prospects. Environmental Conservation, , –. analyzed the data and wrote the article. NAMGAIL, T., BHATNAGAR, Y.V., MISHRA,C.&BAGCHI,S.(a) Pastoral nomads of the Indian Changthang: production system, landuse and socioeconomic changes. Human Ecology, , –. Data NAMGAIL, T., FOX, J.L. & BHATNAGAR,Y.V.(b) Habitat shift and time budget of the Tibetan argali: the influence of livestock grazing. The data are available at https://doi.pangaea.de/./ Ecological Research, , –. PANGAEA. PEATY,D.() Community-based tourism in the Indian Himalaya: homestays and lodges. Journal of Ritsumeikan, Social Sciences and Humanities, , –. References SANGAY,T.&VERNES,K.() Human–wildlife conflict in the kingdom of Bhutan: patterns of livestock predation by large ANWAR, M.B., JACKSON, R., NADEEM, M.S., JANEČKA, J.E., HUSSAIN, mammalian carnivores. Biological Conservation, , –. S., BEG, M.A. et al. () Food habits of the snow leopard Panthera SHARMA, R.K., BHATNAGAR,Y.V.&MISHRA,C.() Does livestock uncia (Schreber, ) in Baltistan, Northern Pakistan. European benefit or harm snow leopards? Biological Conservation, , –. Journal of Wildlife Research, , –. SURYAWANSHI, K.R., REDPATH, S.M., BHATNAGAR, Y.V., BAGCHI,S.&MISHRA,C.() Living with large carnivores: RAMAKRISHNAN, U., CHATURVEDI, V., SMOUT, S.C. & MISHRA,C. predation on livestock by the snow leopard (Uncia uncia). Journal of () Impact of wild prey availability on livestock predation by Zoology, , –. snow leopards. Royal Society Open Science, , . BHATNAGAR, Y., STAKREY,R.&JACKSON,R.() A survey of depredation and related wildlife–human conflicts in the Hemis National Park, Ladakh (India). Bulletin of the International Snow Leopard Trust, , –. Biographical sketches BHATNAGAR, Y.V., WANGCHUK, R., PRINS, H.H., VAN WIEREN, S.E., &MISHRA,C.(). Perceived conflicts between pastoralism and The authors are an international group of conservation biologists conservation of the kiang Equus kiang in the Ladakh working for the Department of Wildlife Protection, Jammu and Trans-Himalaya, India. Environmental Management, , –. Kashmir, India, and Fordham University, USA, with a vested interest BUZZARD, P., LI,X.&BLEISCH,W.() The status of snow leopards in exploring, documenting, and conserving nature in under- Panthera uncia, and high altitude use by common leopards P. documented, but critically important, landscapes such as the pardus, in north-west Yunnan, China. Oryx, , –. Western Himalayas. All authors are particularly interested in promot- CHUNDAWAT,R.&RAWAT,G.() Food habits of the snow leopard ing land use by the rural and traditional communities while positively in Ladakh. In International Snow Leopard Trust and Northwest influencing human–wildlife dynamics.