Abstract: Camera trapping is the most widely used data collection method for estimating snow leopard (Panthera uncia) abundance; however, the accuracy of this method is limited by human observer errors from misclassifying individuals in camera trap images. We evaluated the extent Whiskerbook (www.whiskerbook.org), an artificial intelligence (AI) software, could reduce this error rate and enhance the accuracy of capture-recapture abundance estimates. Using 439 images of 34 captive snow leopard individuals, classification was performed by five observers with prior experience in individual snow leopard ID (“experts”) and five observers with no such experience (“novices”). The “expert” observers classified 35 out of 34 snow leopard individuals, on average erroneously splitting one individual into two, thus resulting in a higher number than true individuals. The success rate of experts was 90 %, with less than a 3 % error in estimating the population size in capture-recapture modeling. However, the “novice” observers successfully matched 71 % of encounters, recognizing 25 out of 34 individuals, underestimating the population by 25 %. It was found that expert observers significantly outperformed novice observers, making statistically fewer errors (Mann Whitney U test P = 0.01) and finding the true number of individuals (P = 0.01). These differences were contrasted with a previous study by Johansson et al. 2020, using the same subset of 16 individuals from European zoos. With the help of AI and the Whiskerbook platform, “experts” were able to match 87 % of encounters and identify 15 out of 16 individuals, with modeled estimates of 16 ± 1 individuals. In contrast, “novices” were 63 % accurate in matching encounters and identified 12 out of 16 individuals, modeling 12 ± 1 individuals that underestimated the population size by 12 %. When comparing the performance of observers using AI and the Whiskerbook platform to observers performing the tasks manually, we found that observers using Whiskerbook made significantly fewer errors in splitting one individual into two (P = 0.04). However, there were also a significantly higher number of combination errors, where two individuals were combined into one (P = 0.01). Specifically, combination errors were found to be made by “novices” (P = 0.04). Although AI benefited both expert and novice observers, expert observers outperformed novices. Our results suggest that AI effectively reduced the misclassification of individual snow leopards in camera trap studies, improving abundance estimates. However, even with AI support, expert observers were needed to obtain the most accurate estimates.

Abstract: Photo-identification of individual snow leopards (Panthera uncia) is the primary data source for density estimation via capture-recapture statistical methods. To identify individual snow leopards in camera trap imagery, it is necessary to match individuals from a large number of images from multiple cameras and historical catalogues, which is both time-consuming and costly. The camouflaged snow leopards also make it difficult for machine learning to classify photos, as they blend in so well with the surrounding mountain environment, rendering applicable software solutions unavailable for the species. To potentially make snow leopard individual identification available via an artificial intelligence (AI) software interface, we first trained and evaluated image classification techniques for a convolutional neural network, pose invariant embeddings (PIE) (a triplet loss network), and compared the accuracy of PIE to that of the HotSpotter algorithm (a SIFT-based algorithm). Data were acquired from a curated library of free-ranging snow leopards taken in Afghanistan between 2012 and 2019 and from captive animals in zoos in Finland, Sweden, Germany, and the United States. We discovered several flaws in the initial PIE model, such as a small amount of background matching, that was addressed, albeit likely not fixed, using background subtraction (BGS) and left-right mirroring (LR) techniques which demonstrated reasonable accuracy (Rank 1: 74% Rank-5: 92%) comparable to the Hotspotter results (Rank 1: 74% Rank 2: 84%)The PIE BGS LR model, in conjunction with Hotspotter, yielded the following results: Rank-1: 85%, Rank-5: 95%, Rank-20: 99%. In general, our findings indicate that PIE BGS LR, in conjunction with HotSpotter, can classify snow leopards more accurately than using either algorithm alone.

Abstract: Population density estimation is integral to the effective conservation and management of wildlife. The snow leopard Panthera uncia is categorized as Vulnerable on the IUCN Red List, and reliable information on its density is a prerequisite for its conservation and management. Little is known about the status of the snow leopard in the central and eastern Qilian Mountains, China. To address this, we estimated the population density of the snow leopard using a spatially explicit capture–recapture model based on camera trapping in Machang in the central and eastern Qilian Mountains during January–March 2019. We set up
40 camera traps and recorded 84 separate snow leopard captures over 3,024 trap-days. We identified 18 individual snow leopards and estimated their density to be 2.26/100 km. Our study provides baseline information on the snow leopard and the first population estimate for the species in the central and eastern Qilian Mountains.

Abstract: Snow leopards and agropastoral communities have co-existed in snow leopard range countries for centuries. The vulnerable snow leopard forms and maintains the entire ecosystem, serving as an indicator species of a healthy alpine ecosystem. However, snow leopards, on the other hand, habitually kill livestock, occasionally killing 100 or more livestock in a single night, resulting in snow leopard retaliation. Thus, the snow leopard is becoming more threatened, so more attention should be paid. Therefore, numerous conservation mitigation strategies have been applied to maintain human-snow leopard coexistence in countries of the snow leopard range. However, such implemented conservation strategies lacked a thorough assessment of their achievements or shortcomings in protecting the snow leopard and enhancing community tolerance. Therefore, we systematically examined and evaluated peer-reviewed articles and book chapters on existing and implemented mitigation measures. We use the software Publish or Perish to achieve this, and we assess using the Preferred Reporting of Items for Systematic Review and Meta-Analysis (PRISMA) review approach. We thoroughly analyzed 42 papers and book chapters that were condensed human- snow leopard co-existence-related literature published in English from 2010 to 2023. Almost 90% of the papers were country-specific, with the remaining papers covering regional or snow leopard ranges countries. Nepal had the most papers, followed by China, India, and Mongolia; however, Afghanistan, Bhutan, Pakistan, Russia, and Tajikistan each had<10%, but there was no single document from Kazakhstan or Kyrgyzstan. Predator-proof corral, improved herding practices, and community-based insurance programs were three of the key recommendations that were more than 10 to 22 times proposed interventions. There are site-specific sociocultural situations and environments that require long-term action-oriented research that is area-specific rather than short-term and generic interventions. We identified a large knowledge gap in snow leopard research, specifically a lack of evidence that demonstrates and quantifies the effects of conservation actions, and strongly advise that it be further researched.

Abstract: Livestock depredation is an important source of conflict for many terrestrial large carnivore
species. Understanding the foraging behavior of large carnivores on domestic prey is therefore
important for both mitigating conflict and conserving threatened carnivore populations. Handling
time is an important, albeit often overlooked, component of predatory behavior, as it directly
influences access to food biomass, which can affect predator foraging efficiency and subsequent
kill rates. We used long-term data on snow leopards (Panthera uncia) in Mongolia (Asia) and
Eurasian lynx (Lynx lynx) in Norway (Europe) to examine how large carnivore foraging patterns
varied between domestic and wild prey, and how the different landscape characteristics affected
those patterns. Our results suggest handling time was generally shorter for domestic compared to
wild prey. For snow leopards, rugged terrain was linked to increased handling time for larger
prey. For lynx, handling time increased with terrain ruggedness for domestic, but not wild, prey,
and was greater in closed compared to open habitats. There were also other differences in snow
leopard and lynx foraging behavior, e.g., snow leopards also stayed longer at, and remained closer
to, their kill sites than lynx. Shorter handling time suggests that felids may have utilized domestic
prey less effectively than wild prey, i.e., they spent less time consuming their prey. This could a)
result in an energetic or fitness cost related to decreased felid foraging efficiency caused by the
risk of anthropogenic disturbance, or b) exacerbate conflict if reduced handling time associated
with easy prey results in increased livestock depredation.

Abstract: Community-based conservation efforts represent an important approach to facilitate the coexistence of people and wildlife. A concern, however, is that these efforts build on existing community structures and social norms, which are commonly dominated by men. Some biodiversity conservation approaches may consequently neglect women’s voices and deepen existing inequalities and inequities. This paper presents two community case studies that draw upon the knowledge and experience gained in our snow leopard conservation practice in pastoral and agro-pastoral settings in Mongolia and India to better understand women’s roles and responsibilities. In these settings, roles and responsibilities in livestock management and agriculture are strongly differentiated along gender lines, and significant gaps remain in women’s decision-making power about natural resources at the community level. We argue that context-specific and gender-responsive approaches are needed to build community support for conservation actions and leverage women’s potential contributions to conservation outcomes.

Abstract: There is an increasing emphasis in conservation strategies for large carnivores on facilitating their coexistence with humans. Justification for coexistence strategies should be based on a quantitative assessment of currently remaining large carnivores in human-dominated landscapes. An essential part of a carnivore’s coexistence strategy has to rely on its prey. In this research, we studied snow leopards Panthera uncia whose habitat mainly comprises human-dominated, unprotected areas, to understand how a large carnivore and its primary prey, the bharal Pseudois nayaur, could coexist with human land use activities in a large proportion of its range. Using a combination of livestock census, camera trapping and wildlife surveys, across a broad gradient of livestock grazing intensity in a 363 000 km2 landscape on the Tibetan Plateau, we found no evidence of livestock grazing impacts on snow leopard habitat use, bharal density and spatial distribution, even though livestock density was 13 times higher than bharal density. Bharal were found to prefer utilizing more rugged habitats at higher elevations with lower grass forage conditions, whereas livestock dominated in flat valleys at lower elevations with higher productivity, especially during the resource-scarce season. These findings suggest that the spatial niche separation between bharal and livestock, together with snow leopards’ specialized bharal diet, minimized conflicts and allowed snow leopards and bharal to coexist in landscapes dominated by livestock grazing. In recent years, reduced hunting and nomadic herder’s lifestyle changes towards permanent residence may have further reinforced this spatial separation. Our results indicated that, for developing conservation strategies for large carnivores, the niche of their prey in relation to human land-use is a key variable that needs to be evaluated.

Abstract: The cycle of livestock depredation and retaliatory killing constitutes a major threat to large carnivores worldwide and imposes considerable hardships on human communities. Mitigation efforts are often undertaken with little knowledge of ecological underpinnings and patterns of depredation, limiting conservationists' ability to develop, prioritize, and evaluate solutions. Carnivore detection and depredation data from interviews in affected communities may help address this gap, but such data are often prone to false-positive uncertainty. To address these challenges in the Pamir Mountains of Tajikistan we collected snow leopard, lynx, wolf, and bear detection and depredation reports from local communities via semi-structured interviews. We used a novel hierarchical multi-species multi-state occupancy model that accounted for potential false-positives to investigate carnivore site use and depredation concurrently with respondents' apparent vulnerability to that risk. Estimated false-positive probabilities were small, but failure to account for them overstated site use probabilities and depredation risk for all species. Although individual vulnerability was low, depredation was nonetheless commonplace. Carnivore site use was driven by clear habitat associations, but we did not identify any clearly important large-scale spatial correlates of depredation risk despite considerable spatial variation in that risk. Respondents who sheltered livestock in household corrals reinforced with wire mesh were less likely to report snow leopard depredations. Reducing depredation and retaliation at adequately large scales in the Pamirs will likely require a portfolio of species-specific strategies, including widespread proactive corral improvements. Our approach expanded inference on the often-cryptic processes surrounding human-carnivore conflict even though structured wildlife data were scarce.

Abstract: This paper uses remote sensing data from the Sanjiangyuan National Park (SNP) to explore the divergence between the boundaries of national parks and the distribution of natural habitats. Results are used to argue that these discrepancies evolve along with the potential impact of global warming. Using the example of the habitat change of snow leopards and the conflicts between local people and snow leopards, we reflect on the consequences of this divergence. Results show that divergence between the political boundaries and natural habitats as well as the consequent influence on the living conditions of local people are strikingly visible, and the effects of global warming on such conflicts are apparent. The authors conclude that both notions of ‘political boundaries’ and ‘natural habitats’ are expected to come together as the SNP region is spatially configured, while ‘global warming’ seems to be relevant as an essential reference when delimiting the region in the future. Finally, the proposal for the establishment of cooperative conservation areas is presented,
emphasizing the role of cooperative governance in/around national parks.

Abstract: Mammalian communities inhabiting temperate grasslands are of conservation concern globally, especially in Central Asia, where livestock numbers have dramatically increased in recent decades, leading to overgrazing and land-use change. Yet, how this pervasive presence of livestock herds affects the community of wild mammals remains largely unstudied. We used systematic camera trapping at 216 sites across remote, mountainous areas of the Mongolian Altai Mountains to assess the spatial and temporal patterns of occurrence and the interspecific relationships within a mammalian community that includes different categories of livestock. By adopting a recently proposed multispecies occupancy model that incorporates interspecific correlation in occupancy, we found several statistically strong correlations in occupancy among species pairs, with the majority involving livestock. The sign of such associations was markedly species-dependent, with larger wild species of conservation concern, namely, snow leopard and Siberian ibex, avoiding livestock presence. As predicted, we found evidence of a positive correlation in occupancy between predators and their respective main prey. Contrary to our expectations, a number of intraguild species pairs also showed positive co-occurrence, with no evidence of spatiotemporal niche partitioning. Overall, our study suggests that livestock encroaching into protected areas influences the whole local community of wild mammals. Though pastoralism has coexisted with wildlife for millennia in central Asian grasslands, our findings suggest that policies and practices to decrease the pressure of livestock husbandry on wildlife are needed, with special attention on large species, such as the snow leopard and its wild prey, which seem to be particularly sensitive to this pervasive livestock presence.

Abstract: 1. Spatial responses to risk from multiple predators can precipitate emergent consequences for prey (i.e. multiple-predator effects, MPEs) and mediate indirect interactions between predators. How prey navigate risk from multiple predators may therefore have important ramifications for understanding the propagation of predation-risk effects (PREs) through ecosystems.
2. The interaction of predator and prey traits has emerged as a potentially key driver of antipredator behaviour but remains underexplored in large vertebrate systems, particularly where sympatric prey share multiple predators. We sought to better generalize our understanding of how predators influence their ecosystems by considering how multiple sources of contingency drive prey distribution in a multi-predator–multi-prey system.
3. Specifically, we explored how two sympatric ungulates with different escape tactics—vertically agile, scrambling ibex Capra sibirica and sprinting argali Ovis ammon—responded to predation risk from shared predators with contrasting hunting modes—cursorial wolves Canis lupus and vertical-ambushing, stalking snow leopards Panthera uncia.
4. Contrasting risk posed by the two predators presented prey with clear trade-offs. Ibex selected for greater exposure to chronic long-term risk from snow leopards, and argali for wolves, in a nearly symmetrical manner that was predictable based on the compatibility of their respective traits. Yet, acute short-term risk from the same predator upended these long-term strategies, increasing each ungulates' exposure to risk from the alternate predator in a manner consistent with a scenario in which conflicting antipredator behaviours precipitate risk-enhancing MPEs and mediate predator facilitation. By contrast, reactive responses to wolves led ibex to reduce their exposure to risk from both predators—a risk-reducing MPE. Evidence of a similar reactive risk-reducing effect for argali vis-à-vis snow leopards was lacking.
5. Our results suggest that prey spatial responses and any resulting MPEs and prey-mediated interactions between predators are contingent on the interplay of hunting mode and escape tactics. Further investigation of interactions among various drivers of contingency in PREs will contribute to a more comprehensive understanding and improved forecasting of the ecological effects of predators.

Abstract: The snow leopard (Panthera uncia Schreber, 1776) population in Russia and Mongolia is situated at the northern edge of the range, where instability of ecological conditions and of prey availability may serve as prerequisites for demographic instability and, consequently, for reducing the genetic diversity. Moreover, this northern area of the species distribution is connected with the western and central parts by only a few small fragments of potential habitats in the Tian-Shan spurs in China and Kazakhstan. Given this structure of the range, the restriction of gene flow between the northern and other regions of snow leopard distribution can be expected. Under these conditions, data on population genetics would be extremely important for assessment of genetic diversity, population structure and gene flow both at regional and large-scale level. To investigate large-scale and fine-grain population structure and levels of genetic diversity we analyzed 108 snow leopards identified from noninvasively collected scat samples from Russia and Mongolia (the northern part of the range) as well as from Kyrgyzstan and Tajikistan (the western part of the range) using panel of eight polymorphic microsatellites. We found low to moderate levels of genetic diversity in the studied populations. Among local habitats, the highest heterozygosity and allelic richness were recorded in Kyrgyzstan (He = 0.66 ± 0.03, Ho = 0.70 ± 0.04, Ar = 3.17) whereas the lowest diversity was found in a periphery subpopulation in Buryatia Republic of Russia (He = 0.41 ± 0.12, Ho = 0.29 ± 0.05, Ar = 2.33). In general, snow leopards from the western range exhibit greater genetic diversity (He = 0.68 ± 0.04, Ho = 0.66 ± 0.03, Ar = 4.95) compared to those from the northern range (He = 0.60 ± 0.06, Ho = 0.49 ± 0.02, Ar = 4.45). In addition, we have identified signs of fragmentation in the northern habitat, which have led to significant genetic divergence between subpopulations in Russia. Multiple analyses of genetic structure support considerable genetic differentiation between the northern and western range parts, which may testify to subspecies subdivision of snow leopards from these regions. The observed patterns of genetic structure are evidence for delineation of several management units within the studied populations, requiring individual approaches for conservation initiatives, particularly related to translocation events. The causes for the revealed patterns of genetic structure and levels of genetic diversity are discussed.