What's new – New Danish land cover map with more classes

We have released a major upgrade to our nationwide Danish land cover map. The previous map was based on a machine learning model that could distinguish between pervious and impervious areas, on top of this we added data from GeoDanmark to enrich the map with forest, building and surface water (lakes) classes.

The new map is based on a much more powerful machine learning model that allows us to classify each pixel into the following categories: impervious, shallow vegetation, dense vegetation and bare land. On top of this we have added an expanded set of GeoDanmark classes to mark buildings, agricultural fields, surface water (now also from streams) as well as paved and unpaved roads. The result is a high-resolution national land cover map with 9 classes, as summarised in the below graphic.

We have also upgraded the Flash Flood Map to use the new classes, and in particular we are excited about the dense vegetation class that replaces the forest polygons, providing a much better mapping of those high-infiltration areas.

For more information please read the accompanying blog post (in Danish).