An expedition by specialists from the Saylyugem National Park and volunteers from the "In the Footsteps of the Snow Leopard" movement h as concluded in the Kosh‑Agach District of the Altai Republic, on the Saylyugem Ridge. This is the first work carried out under a project implemented by the Interregional Irbis Association, the Mars company and the A. N. Severtsov Institute of Ecology and Evolution of the Russian Academy of Sciences. The project’s goal is to collect data to develop conservation methods for this rare feline species. Expedition members installed 50 automatic cameras in accordance with the methodology: they identified a model study area covering 100 sq. km and placed one camera trap in each 2 sq. km square.

“

The monitoring site is located on the Saylyugem Ridge. The terrain in this area is mountainous, but we also included steppes and foothills — flat areas and rocky slopes. The work turned out to be very labour‑intensive because we had to carefully measure the distances between the camera traps and later calibrate the equipment using a marked pole — a calibration pole that helps estimate the distance to animals, their movement speed and adjust the cameras' detection zone. Just installing and calibrating the cameras took almost a whole week

says Alexey Kuzhlekov, expedition coordinator and researcher at the Saylyugem National Park.

The calibration pole is a rod slightly over one metre long, with contrasting stripes every 20 centimetres. For calibration, scientists place the pole strictly vertically at different points within the camera’s field of view, while the camera takes a series of shots. Later, researchers use computer vision on the images captured by the camera trap to create a model that allows them to estimate the animals' movement speed.

“

There are two methods for determining the number and density of the manul population. The first is the Spatially Explicit Capture‑Recapture (SECR) method, which requires individual identification of animals based on the spots on their fur. The second is the Random Encounter Model (REM), which works without individual identification but is more labour‑intensive. Calibrating the field of view with a calibration pole is very important for processing REM data using computer vision; otherwise, the software simply won’t be able to perform the calculations. Calibration allows the algorithms to estimate the path an animal takes in front of the camera, the manuls' movement speed and their daily range. Since we are not yet sure that the camera trap data will allow us to individually identify manuls — as we do with snow leopards, leopards and tigers — we designed the camera trap matrix architecture to enable the use of both methods

explains José A. H ernández‑Blanco, project coordinator and senior researcher at the IEE RAS.

“

There is a lot of livestock in the fieldwork areas; local nomadic shepherds live there. We mounted the cameras on special poles an d surrounded them with stones to prevent livestock from knocking them over. We also informed the herders about the scientific work, asked them to report any fallen cameras and not to touch the installed equipment

continues Alexey Kuzhlekov.

The next stage of the project involves installing camera traps in the Zabaykalsky Krai and monitoring the manul in the Republic of Tuva and in the south of the Krasnoyarsk Krai. The main partners are the Ubsunur Hollow, Daursky and Sayano‑Shushensky nature reserves, the Tunkinsky and Saylyugem national parks, and the "In the Footsteps of the Snow Leopard" movement.