The snow leopard Panthera uncia, a flagship species of mountainous central Asia, is categorized as Endangered on the IUCN Red List (Jackson et al., Reference Jackson, Mallon, McCarthy, Chundawat and Habib2008). China has the largest number of snow leopards and the largest area of potential habitat of the 12 range countries (McCarthy & Mallon, Reference McCarthy and Mallon2016). China is thus crucial to snow leopard conservation, but there are many challenges to the conservation of the species, including poaching, mining, and infrastructure development. More research is needed for effective conservation of the species (Alexander et al., Reference Alexander, Zhang, Shi and Riordan2016; Li et al., Reference Li, Xiao and Lu2016). To address this we provide data from camera traps at four areas across the Tianshan Mountains, including some areas not previously systematically surveyed. The Tianshan Mountains stretch across Xinjiang province in north-west China and provide a potential link between snow leopard populations in Kazakhstan/Kyrgyzstan and Mongolia. These populations have been recognized as the subspecies P. uncia uncia and P. uncia irbis, respectively (Janecka et al., Reference Janecka, Zhang, Li, Munkhtsog, Bayaraa and Galsandorj2017). Given their geographical location, the Tianshan Mountains are potentially important to snow leopard conservation, and it is a priority to determine the snow leopard's status there (Snow Leopard Network, 2014).

In China, snow leopard conservation efforts increased in the late 1980s when the snow leopard was protected as a class 1 species and several nature reserves were established (e.g. Tomur Feng (250 km²); Li et al., Reference Li, Xiao and Lu2016). Despite these positive steps, early work by Schaller et al. (Reference Schaller, Hong, Talipu, Junrang and Mingjiang1988) in the Tianshan Mountains and other parts of Xinjiang found relatively few signs of snow leopards. In Tomur Feng reserve in the western Tianshan, adjacent to the Krygyzstan border, there was more evidence of snow leopard presence than in the Harendanban Shan of the central Tianshan (Schaller et al., Reference Schaller, Hong, Talipu, Junrang and Mingjiang1988). Subsequent work at Tomur Feng confirmed its importance for snow leopards (e.g. Ma et al., Reference Ma, Xu, Chundawat, Jumabay, Wu, Aizezi and Zhu2006; MaMing et al., Reference MaMing, Xu and Cheng2013), and more recent work confirmed snow leopard presence west of Jinghe in the western Tianshan, near Kazakhstan (Pan et al., Reference Pan, Alexander, Riordan, Shi, Kederhan and Yang2016). Wu et al. (Reference Wu, MaMing, Xu, Zhu and Buzzard2015) reported a relatively high density of Siberian ibex Capra sibirica (154 ± SD 23 ibex per 100 km²) in the Tianshan south of Urumqi and estimated 1.3–2.6 snow leopards per 100 km² could be supported.

To detect the presence of snow leopards and potential prey, including Siberian ibex, red deer Cervus elaphus and wild boar Sus scrofa, in Tianshan we set heat/motion-sensing infrared camera traps that recorded both photographs and videos (Bushnell/Reconyx). To maximize the area covered and the number of snow leopards captured we set cameras singly near snow leopard sign or in areas likely to be used by snow leopards, such as along valleys. We set cameras at 1,995–5,000 m altitude and at least 500 m apart or in different valleys. We set 44 cameras during December 2014–June 2015 and/or January–May 2016 at four study areas c. 65–200 km apart: (1) south of Urumqi, (2) four valleys 20–30 km apart (Maolugou, Bayinggou, Toudagou and Dianzhangou) near Wusu, (3) east of Jinghe, and (4) around the Buzarde valley near Ili and the Kazakhstan border (Table 1; Fig. 1).

Fig. 1 The snow leopard Panthera uncia range (Jackson et al., Reference Jackson, Mallon, McCarthy, Chundawat and Habib2008), and the four study areas in the Tianshan Mountains of Xinjiang, China.

Table 1 Numbers of camera sites, camera-days, and independent images or events recorded (adult snow leopard Panthera uncia individuals identified), and camera-trapping rates (events per 100 camera-days), for the four study areas in the Tianshan Mountains of Xinjiang, China (Fig. 1).

Consecutive photographs of snow leopards and other species at the same site were deemed independent events when there was at least a 1-hour interval between them (Rovero & Marshall, Reference Rovero and Marshall2009). To determine survey effort we calculated the number of camera trap-days from the date of each camera's deployment until the date of its retrieval or until the last photograph was taken. To compare with other snow leopard studies we calculated camera trapping rates (events per 100 camera-days), and to demonstrate the relative importance of areas we distinguished among snow leopard individuals in the photographs and videos by comparing spot patterns and scars and other distinguishing marks. Although the cameras were set singly, the combination of pictures and videos often provided clear views of both sides of individuals, and the identification of individuals was confirmed by several observers.

At each site potential prey of the snow leopard, particularly Siberian ibex, were common; for example, ibex herds of 20–60 individuals were observed frequently while setting up cameras near Wusu, and at each study area we captured multiple events of ibex in groups of at least 3–20 individuals (from a low of six events at Jinghe to 86 events at Wusu). We also captured 11 events of wild boar at Ili and three at Wusu, as well as one event of red deer at Urumqi and six at Wusu. Urumqi and Jinghe had the highest camera-trapping rates for snow leopards, and Wusu had the lowest (Table 1); snow leopards were recorded in only two of the valleys surveyed in Wusu (Maolugou and Bayinggou). We identified at least two or three individual adult snow leopards at each site (Table 1). At Maolugou valley, Wusu, two adult snow leopards were photographed together, and in Urumqi two adults were recorded within a 1-minute interval.

Given their geographical location between Central Asia and Mongolia (Snow Leopard Network, 2014), the Tianshan Mountains are a conservation priority for snow leopards and our study demonstrates the importance of the Tianshan Mountains for snow leopard conservation. We confirmed the presence of several adult snow leopards and large numbers of potential prey, such as Siberian ibex, at multiple areas across the east, central and western Tianshan. In addition, our study supports previous findings in the western Tianshan, where Pan et al. (Reference Pan, Alexander, Riordan, Shi, Kederhan and Yang2016) recorded 11–15 individual adult snow leopards in 192 km². Given this potential for large numbers of snow leopards, more work is necessary at our study sites and across the Tianshan to clarify snow leopard density and status. For example, we recorded three individuals (2 males and 1 female) south of Urumqi with relatively limited trapping effort. These same individuals were also recorded in a much greater survey effort adjacent to our study area by Xinjiang Wild Watchers, who obtained photographic evidence of at least 22 individual snow leopards, including 12 adults, in c. 100 km² (Xing, Reference Xing2016).

To enhance snow leopard conservation in Xinjiang there is an urgent need for more protected areas (Xu et al., Reference Xu, Maming, Buzzard and Blank2014), and the Tianshan Mountains were the focus of a conservation and management plan workshop organized by the Xinjiang Forestry Bureau (Tianshan Snow Leopard Conservation Conference, 24–26 August 2016). Our data confirm the importance of several areas for snow leopards, and we have used these data as the basis for formal applications to the Chinese and Xinjiang Governments for multiple protected areas (Table 2).

Table 2 Details of the protected areas for which formal applications were made to the Chinese government based on snow leopard data from the Tianshan Mountains (Fig. 1).