Cryospheric Sciences | Cryo Adventures – Droning on glaciers and ice caps in Kyrgyzstan

The Grigoriev ice cap, seen from drone perspective at about 1 km distance and at 4800 m a.s.l. [Credit: Lander Van Tricht]

Drones are not only a cool toy to play with. They are also a useful instrument to monitor and study glaciers and ice caps. By taking thousands of images flying above the ice, we can make 3D models of ice masses at centimetre resolution. Using drones, we can now survey small to medium sized glaciers and ice caps in unprecedented detail. In recent years, we have conducted several surveys on glaciers and one ice cap in the “Heavenly Mountains”, the Tien Shan, in Kyrgyzstan, Central Asia. Here I share some of our ice-stories with you!

Why do we care about glaciers in Kyrgyzstan?

The mountains of High Mountain Asia are considered Central Asia’s water towers. Their glaciers are a vital source of water for the people and nature in this otherwise arid region, and are also referred to as the globe’s ‘third pole’. During dry periods in summer, more than 75% of the region’s water supply can originate from glacial meltwater. Hence, households, (irrigation) agriculture, industry and electricity (through hydropower) all depend largely on meltwater from the glaciers.

The Arabel-Suu River drains meltwater of the glaciers and ice caps in the Inner Tien-Shan over the Arabel plateau system. [Credit: Lander Van Tricht]

However, as in other mountain ranges around the world, the glaciers in the Kyrgyzstan are rapidly shrinking due to climate change. So, the future water supply in the region is uncertain. In this context, my PhD project is aimed at better understanding the state of glaciers and ice caps in Kyrgyzstan and to model their future evolution.

The Ashu-Tor glacier, one of the glaciers of which we measured the ice thickness and the surface elevation. [Credit: Lander Van Tricht]

Droning to get 3D data

In-situ data, more specifically ice thickness, mass balance and surface height data, are exceedingly scarce in the Tien Shan, compared to some other mountain ranges, such as the Alps. We therefore decided to collaborate with regional partners from the Tien Shan High Mountain Research Center and Kumtor Gold Mine to monitor and study some individual ice masses in detail. For several years now, we have measured different glaciers’ ice thickness (using a ground penetrating radar) and their surface elevation (using GPS and drones).

We use a Phantom 4 Pro and Phantom 4 RTK to image the glaciers. [Credit: Lander Van Tricht]

The most spectacular part of our (annual) fieldwork is the drone flying, and the views we get from the drone perspective. The aim of performing drone surveys is to gather hundreds of pictures of glaciers to combine so as to create 3D models of the selected ice masses. To do this, we need a lot of images because they must overlap and show the glacier from different perspectives at about 200 m above the ice. Having such a set of drone-gathered images allows us to create a detailed 3D model of the glacier.

Adventurous fieldwork

To obtain these images, it is not as easy as just letting the drone up and waiting half an hour for it to return with pictures. Due to the limited battery lifetime and the often-large altitude differences between different parts of the ice masses, the most time-consuming part is getting to the take-off location. Furthermore, the need to distribute control points or check points across the ice surface requires adventurous walks between crevasses, looking for the safest path. To reduce the kilometres travelled, in 2021, we used an RTK drone. The RTK system provides the location of the drone at centimetre accuracy which ensures that no (or only few) control points or check points are needed. This makes our fieldwork significantly more efficient (and comfortable!).

Endless walking on snow and ice to reach a good take-off location and to spread control and validation points. [Credit: Benjamin Vanbiervliet]

Before being able to walk to a good take-off spot, we obviously first need to get to the glacier or ice cap. Since the surveys on certain ice masses last several days and because they are so remote, camping is necessary. The rough terrain near the glacier, where the ice has recently retreated, is not a good camping spot. However, we try to install a “basecamp” as close as possible to the glaciers, and in the vicinity of flowing water, with different tents. To carry all our equipment to elevations above 4000m above sea level, some horses joined us on the trip. I admit that when I started my PhD, I never thought I would be studying glaciers on the back of a horse!

To get to the ice masses to set up a basecamp, we get assistance of our local partners from the Tien Shan High Mountain Research Center and Kumtor Gold Mine, on horseback! [Credit: Lander Van Tricht]

Whereas at home we can consult a precipitation radar or weather forecast, in the field we have to interpret cloud formations. Before noon, there is usually not much to worry about and the drone can be safely sent into the air. After noon however, cumulus clouds form because of convection and grow to the shower stage, sometimes with thunderstorms and hail. The drone is then not allowed to take off, or needs to be taken down from a survey, and of course we have to find a safe place ourselves. Usually, the only option is to go back to the camp. A correct assessment is therefore crucial in several respects.

The western Ashu-Tor glacier, which ends in a proglacial lake. [Credit: Lander Van Tricht]

Analysing data back at home

Back at home (in my case Belgium), we create 3D models of the ice masses, by processing all the images, applying algorithms, and correcting the positions thanks to control points. From these up-to-date representations of the ice surface elevation and thickness, we can then for instance also infer the bedrock elevation. Besides that, when we have 3D models of different years, we can calculate ice thickness differences, and surface velocities by tracking visible features at the surface. Such high-resolution data can be an addition to the typical mass balance stake measurements carried out annually . After 10 months of analysing and using the data, now, in June, new fieldwork preparations and plans are being made. The glaciers are still there, and are waiting for us to measure them anew this new summer!