Images from space: Better snow maps thanks to artificial intelligence

class = “sc-cffd1e67-0 fmXrkB”>

According to the Swiss Federal Institute of Technology in Zurich (ETH Zurich), data on snow depth provides an important basis for winter tourism, hydroelectric power and winter sports enthusiasts to assess avalanche risk.

To date, snow monitoring in Switzerland has relied mostly on data from measuring stations, ETH Zurich said in a statement. Since there are only around 400 stations in all of Switzerland, snow information for many places is quite inaccurate.

For the measurement based on satellite images, researchers from ETH Zurich and the Swiss company Exolabs used images from the Sentinel 2 satellites of the European Space Agency (Esa). They take pictures of every part of the world every five days. In addition to satellite imagery, AI researchers also provided extensive terrain data from Switzerland.

Using this satellite and terrain data, they trained the AI ​​to extract the snow depth for any given point. To do this, they used a process called supervised learning. To do this, they had the system estimate snow depth and compared the results to actual snow measurements. “At each grid point we determined how much the AI ​​was off in its prediction and gradually adjusted the system so that the errors were smaller,” explains Konrad Schindler in the ETH statement.

“Although the effective resolution of the best available snow maps in Switzerland is approximately 250 x 250 meters, you can zoom our maps down to 10 x 10 meters to read snow depth,” says Schindler. “We assume we have set a new standard for snow depth measurement in Switzerland.”

In addition to satellite images from the Sentinel-2 satellites, the researchers also use images from other satellite missions that are less spatially precise but provide daily images. (SDA)

Source : Blick