# Status of the Snow Leopard Panthera uncia in Qinghai and Gansu Provinces, China > How do you count a cat you almost never see? In the 1980s, George Schaller and Chinese colleagues walked the high mountains of Qinghai and Gansu counting snow leopard scrapes and droppings, and found the cats still widespread but hunted hard, with possibly 650 left in Qinghai. Their droppings showed that marmots matter as much as blue sheep in their diet. Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/1988-schaller-junrang-status-snow-leopard-panthera/ ## Citation and access - **Authors:** George B. Schaller, Ren Junrang and Qiu Mingjiang - **Year published:** 1988 - **Journal or source:** Biological Conservation - **Full citation:** Schaller, G. B., Ren J. and Qiu M. 1988. Status of the snow leopard Panthera uncia in Qinghai and Gansu Provinces, China. Biological Conservation 45: 179–194. - **DOI:** 10.1016/0006-3207(88)90138-3 - **Access status:** Publisher link only - **Where to find it:** https://doi.org/10.1016/0006-3207(88)90138-3 ## Tags - **Species:** Mammals > Alpine musk deer; Mammals > Argali; Mammals > Blue sheep; Mammals > Grey wolf; Mammals > Kiang; Mammals > Red deer; Mammals > Snow leopard; Mammals > Tibetan gazelle; Mammals > White-lipped deer; Mammals > Wild yak - **Region:** Asia > China - **Tracking:** Scat; Sign surveys; Tracking; Tracks; Tracks and signs identification - **Behaviour and ecology:** Behaviour; Feeding; Human disturbance; Movement; Predation - **Methods and technology:** Descriptive statistics; Direct observation; Distribution mapping; Interviews; Population estimation - **Data type:** New field data - **Observer qualification:** Not reported - **Discipline:** Behaviour and ecology; Wildlife tracking - **Year published:** 1988 ## Detailed keywords - **Species:** Studied: snow leopard (Panthera uncia), and their wild ungulate prey, chiefly blue sheep (Pseudois nayaur). Food found in snow leopard droppings: blue sheep, Himalayan marmot (Marmota himalayana), musk deer (Moschus sifanicus), Cervus deer, domestic sheep and goats, domestic yak, woolly hare (Lepus oiostolus), pikas (Ochotona) or small rodents, birds, grass and the twigs of shrubs (Tamarix sp., Salsola arbuscula, Sibiraea angustata). Ungulates counted: blue sheep, Tibetan argali (Ovis ammon hodgsoni), white-lipped deer (Cervus albirostris), red deer (Cervus elaphus macneilli), Tibetan gazelle (Procapra picticaudata), Tibetan wild ass (Equus hemionus kiang), wild yak (Bos grunniens) and musk deer. Wolf (Canis lupus) droppings were also analysed. Mentioned but not studied: lynx (Felis lynx), domestic dogs, brown bear (Ursus arctos), goitred gazelle (Gazella subgutturosa), Przewalski’s gazelle (Procapra przewalskii) and Tibetan antelope (Pantholops hodgsoni); livestock (yak, horses, sheep and goats) counted from county records. - **Location and habitat:** Qinghai Province and the western edge of Gansu Province, northeastern Tibetan Plateau, China. Mountain ranges around the Qaidam Basin (Arjin Shan, Qilian Shan with the Tulai Nanshan, Shule Nanshan and Danghe Nanshan, Kunlun Shan, Anyemaqen Shan), the Tanggula Shan and the limestone massifs of Zadoi, Zhidoi and Yushu in southeastern Qinghai; the Die Shan in southern Gansu; a short visit to the Arjin Shan in Xinjiang. Habitats range from desert and sparse shrub to alpine meadow, juniper and conifer forest, mostly above 3000 m. Nine survey blocks of 45 to 610 km² (1375 km² in all), in Shule Nanshan, east and west Anyemaqen, north and south Zadoi, and Yushu, plus foot transects outside the blocks. - **Topics:** snow leopard distribution and status; relative abundance from scrapes and droppings; scrape-marking sites (passes, cliff bases, creek confluences, ridge crests, gorges, spurs at the base of hills); diet from droppings; marmots as a buffer that takes predation pressure off blue sheep; plant eating; wolf diet overlap and competition; ungulate counts, herd composition and reproduction; livestock numbers and overgrazing; livestock losses to predators; poaching, steel traps, poisoning of marmots, market hunting of blue sheep for export; fur and medicine trade; reserves; Buddhist religious sentiment as a basis for conservation education. - **Statistics or models:** Descriptive statistics: spoor counts per survey day (one person or several together searching for one day equals one survey day), densities per km², sex and age ratios, percent of each food item in droppings. Food needs estimated from Emmons (1987), and prey eaten estimated from droppings with the Floyd et al. (1978) equation y = 0.38 + 0.02x (kg of prey per dropping against average prey weight). Population estimate by assuming 1 snow leopard per 100 km² over 65,000 km². - **Data type:** New field data: about 7 months of fieldwork in Qinghai and Gansu between 1984 and 1987 (blue sheep classified in July 1984, September and October 1985, and August and September 1986; droppings collected in August, September, October and December). Livestock numbers from 1984 county records. The authors note that their densities are maxima, because survey blocks were chosen where people said wildlife was most abundant. Small points to check in the paper: the text gives blue sheep density as 1 to 5 animals per km², while Table 2’s figure of 1.4 to 5.4 per km² is for all wild ungulates; the “Amount spoor/survey day” column in Table 1 counts scrapes and droppings only (the east Kunlun track and the 2 fresh tracks at Yushu are mentioned only in the text); the text gives livestock as about 15 to 45 animals per km², which matches Table 4’s livestock units (13 to 45) rather than its animals per km² of pasture (28 to 70). ## Synopsis ### 1. What they did Little had been published about snow leopards in China, where their range is larger than in any other country. As part of a project to assess China’s high-altitude wildlife, the authors spent about 7 months in the field in Qinghai and Gansu provinces between 1984 and 1987. They wanted to know the status of snow leopards, their prey and other large mammals. They used three survey methods: - Asking and looking. They asked local people about wildlife, looked at the terrain and habitat, and noted every animal seen or reported. - Transects. They crossed the country by car, horse, camel or on foot, always on foot when searching for snow leopard spoor. - Survey blocks. In nine blocks (1375 km² in all), they tried to count every wild ungulate by searching all drainages, sorting animals by age and sex where possible. The blocks were chosen because people said snow leopards and other wildlife were more common there, so the densities are a maximum rather than typical. Snow leopard tracks are hard to find without snow, but the cats rake the ground with their hind paws to make scrapes at conspicuous places such as mountain passes, the base of cliffs and where creeks meet. The team searched these likely places and counted every scrape and dropping as a rough measure of abundance. They worked out the cats’ diet from the hair, bones, claws and other remains in droppings, leaving out any dropping that might have come from a lynx, dog or wolf. ### 2. What they found Distribution and numbers - Snow leopards live in all the major mountain ranges of Qinghai and in many small massifs, roughly 65,000 km² or 9% of the province. Viable populations cover less than that: herders have wiped out the cats in some areas or left only rare stragglers. - In Gansu, the authors found signs only in the Qilian Shan along the Qinghai border and in the small Die Shan, which, with an unconfirmed population in Sichuan, marks the eastern edge of the snow leopard’s range in China. - Spoor counts ranged from 0.3 to 16.3 per survey day. Only the Shule Nanshan block had a substantial population (16.3 per survey day), with fresh signs in every valley, even though 12 cats had been trapped there for the Xining zoo in the winter of 1983 to 1984. Tibetan herders killed at least 11 snow leopards in the winter after the survey. - From the signs, four blocks held about 1 cat per 25 to 35 km², and Shule Nanshan probably more. At 1 cat per 100 km², Qinghai might hold about 650 snow leopards. Gansu has few. Reading the signs - Scrapes and droppings were usually found apart: only 10 to 21% of scrapes had a dropping within 2 m. - Where the cats travelled showed in their signs: in the Shule Nanshan, 19% of scrapes were on ridge crests and 75% in gorges where prey came down to drink. In the western Kunlun, scrapes lay along the base of hills beside broad valleys, where the cats apparently patrolled for blue sheep and gazelles. - Scrapes outnumbered droppings in some places and the other way round in others. In the Yushu block the team found 46 droppings and 2 fresh tracks but not a single scrape, so counting scrapes alone could give a skewed picture. Prey - Blue sheep were the most abundant ungulate and the only one sharing the snow leopard’s habitat throughout both provinces. All wild ungulates together came to 1.4 to 5.4 animals per km² in the blocks. - Argali, Tibetan gazelles, red deer, wild ass and wild yak were rare or gone from the blocks; only two argali were seen in all the months of fieldwork. Livestock were far denser, an average of about 15 to 45 animals per km² in southeastern Qinghai, often more than the range could carry. Diet - Wild ungulates, mainly blue sheep, made up about 30 to 45% of the droppings’ contents. Marmots were just as important and sometimes more (37 to 65%). Livestock were minor, except in the Yushu block (22%). - The droppings were mostly from summer and autumn, when marmots are active. The authors expect a winter sample would show more blue sheep and livestock. - Plant matter made up 2 to 13% of the droppings. The cats sometimes bit off and swallowed 2 to 4 cm twigs of particular shrubs, enough to fill whole droppings, for reasons unknown. - Wolves ate much the same food, mainly blue sheep and marmots. - By their calculation, a 35 kg snow leopard needs to kill at least 822 kg of prey a year. Over the 6 months marmots are active, one cat might kill about 35 marmots and 5.6 blue sheep. Threats - Snow leopards have been fully protected in China since 1983, but there were no staff or funds to enforce the law. Many herders carried rifles, steel traps took everything from snow leopards and brown bears to gazelles, and skins and bones were sold in town shops. - Snow leopards were killed mainly for preying on livestock, although losses were small: 5 households in east Anyemaqen lost 12 of 2350 animals (0.5%) to wolves and snow leopards in a year. - Marmots were killed for pelts and poisoned. From 1958, Qinghai exported 100,000 to 200,000 kg of blue sheep meat a year, about 5000 to 10,000 animals. ### 3. What they concluded - Snow leopards are still fairly widespread in Qinghai but have been decimated or wiped out in many areas, as has their prey. Their future in Qinghai and Gansu is far from secure. - With wild prey reduced to remnants, snow leopards and wolves depend mainly on blue sheep, and hunting by people on top of predation probably pushes blue sheep numbers down. Marmots act as an important buffer that takes pressure off blue sheep. - Where habitat is fragmented, losing even two or three cats could wipe out a local population. - Qinghai had no reserves for large mammals. It could set up reserves next to the Arjin Shan Reserve in Xinjiang and the Yanchiwan Reserve in Gansu. - Protecting blue sheep, managing market hunting sustainably, stopping illegal steel traps and ending the sale of skins and bones would be first steps. Because most herders are Buddhist Tibetans, religious sentiment could be the basis for education, encouraging herders to lay down their guns and communes to set up their own reserves, especially around temples. ### 4. How tracking was or was not used Tracking was the main way the authors judged snow leopard presence and abundance. They searched on foot at likely places (passes, cliff bases, creek confluences, ridges and gorges) and counted scrapes and droppings per survey day, also noting tracks. They read travel routes from where scrapes were found. Droppings were identified to species and analysed for prey hair, bones and claws; droppings that could have come from lynx, dogs or wolves were left out. Spoor and sightings were also used to judge musk deer abundance, and the dots on their distribution map mark places where snow leopard spoor was found outside the survey blocks. No trailing was reported. ### 5. Tracker or observer certification Not reported. The authors did the searching themselves and say nothing about who identified the signs or how; the abundance and diet results depend on telling snow leopard scrapes and droppings apart from those of other carnivores. ### 6. Other technologies, skills or innovations - Total counts of ungulates in survey blocks by searching every drainage, with animals sorted by age and sex. - Estimating prey eaten from droppings using the Floyd et al. (1978) method, which corrects for small prey leaving more indigestible matter than large prey. - Travel on foot, horseback and camel across remote mountain country. ### 7. Implications for conservation (from the paper) - Legal protection without enforcement does little: local people kill snow leopards almost with impunity. - Snow leopard conservation depends on their prey. Protecting blue sheep, managing market hunting sustainably and stopping the poisoning of marmots matter as much as protecting the cats. - Reserves next to existing ones across the provincial borders, and control of the few herders living there, could help (only 30 households caused most of the damage to wildlife around the Shule Nanshan block). - Education rooted in local Buddhist beliefs might help, including community reserves around temples. ### 8. Implications for tracking (Original Wisdom's observations) _These are Original Wisdom's observations, not the authors'._ - Where snow is absent and tracks are hard to see, a cat’s scrapes and droppings at predictable places can still show where they live, how they travel and what they eat. - The Yushu block, with 46 droppings and no scrapes, shows why counting only one kind of sign can mislead: the signs an animal leaves vary from place to place. ## Cite this summary Lawrence, K. 2026. Summary of: Status of the Snow Leopard Panthera uncia in Qinghai and Gansu Provinces, China (Schaller and Junrang et al., 1988). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/1988-schaller-junrang-status-snow-leopard-panthera/