Counting the ghosts of the mountains: sampling snow leopard populations at large spatial scales

Effective management of large carnivores requires robust monitoring at all scales. In their latest research, Manvi Sharma and colleagues describe the first systematic effort at estimating snow leopard populations at a large regional scale.

The high-altitude mountains of the Himalaya are important habitats for unique flora and fauna adapted to these regions. The most charming of these species that has taken home here is the snow leopard, famously known as the ghost of the mountains. The snow leopard has captured the imagination of generations of storytellers, artists, and explorers and is deemed as a flagship species of the conservation of the Himalayan ecosystems.

image1
Artwork by © Sartaj Ghuman

Efforts to protect the snow leopard have brought researchers and practitioners from partner organisations and government agencies to work together on a common platform. One major initiative for snow leopard conservation is to assess the population status across its distribution in 12 countries.

So how many researchers does it take to count the snow leopards in an area? Not many, if you only search at the good spots – the prime habitats with high densities of these cats. However, this quickly becomes problematic when estimating population size for large areas that run into thousands of kilometres including not-so-prime habitats, and an estimation exercise based on sampling the prime habitats alone is not reliable for a regional and global estimate of population.

For snow leopards and many other species of conservation importance, regional and global estimates still remain elusive. A study conducted by researchers based at the Nature Conservation Foundation and the Snow Leopard trust demonstrated that current estimates of snow leopard populations are likely to be overestimated because of extrapolation from areas of prime habitats to all types of habitats.

image2
Snow leopards in Kibber, Spiti Valley © Prasenjit Yadav

In our latest research aiming to address these limitations, we demonstrate that a two-step sampling approach that combines distribution surveys with abundance estimation surveys can produce reliable population estimation at large spatial scales.

In this approach, we conducted occupancy-based surveys across the snow leopard habitat in the state of Himachal Pradesh in India, using 15km x 15km grids to sample over 20,000km2 of land. Based on these samples, the entire habitat was stratified into three regions of high, low, and unknown (unsampled habitat) snow leopard occupancy and then representatively selected to conduct camera trapping surveys at a 4km x 4km grid resolution, which is a suitable scale for spatial capture-recapture analysis.

image3
Camera trap installation by local youth volunteers from Kibber

Conducting camera-trapping surveys in these remote landscapes meant a fair share of crossing frozen rivers, trekking in 3 feet of snow for 8 hours a day, climbing steep slopes on all fours and many cups of chai. The entire camera-trapping effort was supported by 8 youth volunteers from Kibber, a small hamlet in the Spiti valley of Himachal Pradesh.

In total, we installed camera traps in 284 locations distributed at 10 sites across the three strata and the cameras detected snow leopards on 282 occasions which were then individually identified.

So how many snow leopards are there?

Snow leopard densities varied across sites from 0.08 to 0.37 snow leopards per 100 km2. The trans-Himalayan region of Himachal Pradesh with Spiti, Tabo, Pin, and Hangrang were regions with high densities. The final estimate of the population size was 51 (95% CI: 34– 73) snow leopards for the whole study area.

Along with estimating snow leopard population size, we also conducted double observer surveys to estimate the population of the snow leopard’s primary prey – blue sheep and ibex. Quantifying the relationship between predator and prey densities across these sites found the slope to be 0.25, a value smaller than the slope of linear relationship between another top predator, the tiger, and its main prey.

Our study cautions against using “guesstimates” that are based on extrapolation from limited and biased sampling, and the data contributes towards the global effort on the Population Assessment of the World’s Snow leopards (PAWS). This trans-boundary conservation effort is not only aimed at assessing the status of the “ghosts of the mountains” but also at engaging in training activities for local forest officials and the local community, bringing together a diverse group of people who share a common goal to protect the snow leopard.

Read the full research: “Estimating snow leopard and prey populations at large spatial scales” in Issue 2:4 of Ecological Solutions and Evidence.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s