Understanding Snow Leopard Home Ranges

Snow Leopard Trust researcher Örjan Johansson recently published a groundbreaking study where he could show that most Protected Areas in the cats’ habitat are too small to hold viable snow leopard populations. In this article, he explains how he and his team calculated snow leopard home ranges using data from cats they tracked with GPS collars.

‘Home range’ is a classic concept in biological research and is generally defined as the area that an animal utilizes in its daily life.

It’s a very intuitive idea, and it’s important if you want to describe how animals use space, both individually and in relation to each other.

This type of information is critical for conservation. For instance, in order to estimate how many snow leopards there may be in the wild, you need to have a good idea of the size of their habitat, and of how many cats can fit in that area.

To figure that out, you need to know the size of an individual snow leopard’s home range, and how much they overlap with one another.

GPS collars allow researchers to calculate home range sizes of snow leopards. Photo: SLCF / SLT
GPS collars allow researchers to calculate home range sizes of snow leopards. Photo: SLCF / SLT

For a species where home ranges overlap completely, you can obviously fit many more individuals in the same area than you can if the home ranges are exclusive (territories).

The concept may be easy enough to understand, but it’s a big challenge to actually calculate or depict a snow leopard’s home range in practice. I would like to try and explain what’s so difficult about it and how we’ve tried to overcome the challenge.

Home or Away?

To determine the size and shape of home ranges, researchers are often restricted to individual location data from tracking devices (e.g. GPS data). For the calculation we need a method that incorporates the area around the locations as accurately as possible – without missing areas that are being used, or including areas that aren’t.

There are several methods to do that, and since we do not know why the animal visited the places it did, and why it didn’t go to other sites, we need to interpret the locations we know as best we can. Depending on how we interpret the locations and on the method we use to calculate the home range, we can arrive at very different sizes and shapes.

It is also important to consider the time-period for which we calculate home ranges. Most commonly, home ranges are described over the span of a year (i.e. annual home ranges). But our calendar year doesn’t mean much to the animals, which are generally more affected by neighboring animals and seasonal changes in external conditions.

For example, a snow leopard could live in one area for half the year, and then, if it realizes that its neighbor has disappeared, it could move into the newly vacant area – perhaps because it will find more prey there, or because that area overlaps with a female’s home range.

Orjan in the reserach ger
The ger at our base camp in Tost is both home and office to Örjan Johansson. Photo: SLT

If we were to calculate this cat’s annual home range, we would include both the old and the new area, even though the snow leopard never actually used both areas at the same time.

Look at it this way: if you move to a new apartment, do you consider yourself to be living in both the old and the new place? Probably not.

An animal’s age is another factor to take into consideration. The behavior of most animals varies with age and breeding status, so we need to differentiate between adult and subadult animals when calculating home ranges.

We recently published a scientific paper on snow leopard home ranges, based on GPS data from 16 cats we had collared in Mongolia over a period of 6 years. In this paper, we aimed at finding the best or biologically most relevant method to calculate home ranges for these cats – i.e., we wanted to identify the method that best describes the area used by our collared snow leopards.

Ariun, one of the snow leopards we’ve tracked with GPS in Mongolia.

Where is the challenge in that? Well, take me as an example. In 2016 I have travelled over a large part of southern Sweden, and I have also visited London, Dubai, Melbourne, Seattle and a fair bit of the South Island of New Zealand. If you tracked me through my phone (hello, Google), all of these travels could be included as my home range for 2016 – but I wouldn’t recognize it. Instead, I would say that a better description of my home range is the area around my house, the office, the town where we buy food, the village with the daycare and the forests surrounding my house. The visits to the other areas would be more like temporary excursions from my regular home range. Snow leopards don’t travel quite as far, but neither do they just stay put all the time – so part of our challenge was to understand which locations were part of their regular home range, and which were just excursions or outliers.

The home ranges of three male snow leopards in Tost Mountains: Shonkor (brown), Aztai (blue) and Tsagaaan (green). Aztai undertook short trips into Tsagaan's home range several times (presumably to find females), but always returned. To calculate his true home range, these excursions need to be taken out of the equation. Illustration: SLT
The home ranges of three male snow leopards in Tost Mountains: Shonkor (brown), Aztai (blue) and Tsagaaan (green). There is very little overlap between the cats. As illustrated by the blue lines, Aztai (blue) undertook short trips into Tsagaan’s home range several times (presumably to find females), but always returned. To calculate his true home range, these excursions need to be taken out of the equation. Illustration: SLT

I’ve talked about the importance of time periods above. Given the challenges that come with using calendar dates when calculating home ranges, we let the cats’ movements decide which time periods to use in our calculations instead.

We also let the movement patterns determine at what point a cat made the transition from subadult to adult. All our snow leopards moved around quite substantially until they were about three years old, after which they settled in more stable home ranges. We therefore classified them as subadults until we saw this change in movement patterns.

We tracked Aylagch (black), a subadult male cat, as he dispersed from his mother, Kashaa (yellow), and moved across 40km of steppe from Tost Mountains in the South to Nemegt Mountains in the North, where he eventually settled. Illustration: SLT
We tracked Aylagch (black), a subadult male cat, as he dispersed from his mother, Kashaa (yellow), and moved across 40km of steppe from Tost Mountains in the South to Nemegt Mountains in the North, where he eventually settled. Illustration: SLT

Similarly, instead of calculating annual home ranges for ‘our’ cats, we calculated home ranges that were based on their movements instead of a calendar year. If a cat shifted to a new area and settled there, we treated it as a permanent move, and calculated a new, separate long-term home range.

As a result, the home ranges we found vary in time from a couple of months to several years – but we believe that this approach results in a more accurate depiction of reality than any calculations based on fixed calendar dates.

Different Models, Different Results

I’ve mentioned above that there are several methods we have at our disposal to calculate the size and shape of a home range based on a number of individual GPS locations. Each of these methods – statistical models that help make sense of your data – has a set of advantages and disadvantages, depending on the species and the data you are working with.

For our purposes, we ran the location data through three different statistical models that have all been used before for similar exercises, but with different species. This resulted in three different, ‘theoretical’ home ranges for each of our collared snow leopards.

Then, to find out which was the most accurate, we compared the habitat composition in the different ‘model’ home ranges to the actual locations from our collared cats. Almost all the locations we got from the GPS collars were in the mountains, and only very few in steppe areas. Therefore, we decided that the home range ‘model’ that had the highest amount of mountains, and the least amount of steppe, was the most appropriate method to calculate actual home ranges for snow leopards.

With this method (for those of you who are interested, it’s called adaptive local convex hulls, or aLoCoH), we found that the average home range for male snow leopards was around 220km (77 square miles), while females used about 130km2 (50 square miles).

Protected Areas Aren’t Enough

Thanks to this work, we now have a fairly good idea of how large an average snow leopard home range is, and how they overlap (or rather, as we found, don’t overlap). But what does that tell us?

It helps us make better estimates of how many snow leopards there actually are, for starters. It also allows us to make more informed conservation decisions – for instance when it comes to Protected Areas, and the landscapes surrounding them.

To gain some perspective, we used the size and overlap of the home ranges calculated for our collared snow leopards and compared them with the size of the Protected Areas in the snow leopard range to see how many of these Protected Areas were large enough to host one reproducing pair, or a small population of 15 breeding females.

The results are quite telling: 40% of the Protected Areas within snow leopard habitat are too small to host just one breeding pair of cats (i.e., they’re smaller than 220km2).

Depending on calculation methods, only about 3-13% of the Protected Areas are large enough to contain a small population of 15 breeding females.

Just eight Protected Areas could hold 50 females or more.

To put it simply: Protected Areas alone won’t be enough to save the snow leopard. That means we need to focus on helping people and snow leopards co-exist outside of Protected Areas as well. This approach is also reflected in the Global Snow Leopard & Ecosystem Protection Program (GSLEP), which aims to establish 23 large landscapes that are to be secured for snow leopards by 2020.

We stress that Protected Areas are still important for snow leopard conservation as they may function as reservoirs, where damaging activities such as mining and other large-scale development projects can be regulated more easily that in multi-use landscapes. But on their own, they’re not going to cut it.

Disclaimer: while the illustrations in this post show approximate GPS locations of snow leopards (from 2008 to 2012), they are deliberately imprecise, so as to not reveal details about the cats’ movement patterns.


Örjan Johansson is a PhD student at the Swedish University of Agricultural Sciences. He has captured, immobilized and collared a total of 20 wild snow leopards as part of his research in Mongolia’s South Gobi. In total, he has spent more than three years at the remote base camp in Tost Mountains, living in a ger – often with only Friday the camp cat for company. He has somehow managed to maintain the ability to communicate with humans in both Swedish and English (as well as very basic Mongolian) despite all this. 



This study is dedicated to the memory of our dear friend Sumbee Tomorsukh, who had a key role in the field work and data analysis that led to these results. He left us in 2015.

This work is a result of the ongoing long-term ecological study on snow leopards in Mongolia’s South Gobi province that’s been conducted by the Snow Leopard Conservation Foundation, Snow Leopard Trust, and the Mongolian Academy of Sciences since 2008. The conservation organization Panthera helped launch the study and was a partner until 2012.

We are thankful to the Ministry for Environment and Green Development, Government of Mongolia, the Mongolian Academy of Sciences, and the Swedish University of Agricultural Sciences for partnering with us in this research endeavor.

Partnership Funding by Fondation Segré, managed by the Whitley Fund for Nature, has helped tremendously with this work.

We are equally thankful to the Woodland Park Zoo, Cat Life Foundation, Columbus Zoo & Aquarium, David Shepherd Wildlife Foundation, Kolmarden Zoo, Nysether Family Foundation, Twycross Zoo and all other donors and supporters.

Special thanks go to all staff and volunteers who aided in the work.


Leave a Reply