Scalgo Live land cover map expands to France
- Nye data
France is next in our ambitious roll-out of nationwide land cover maps. With a 25 cm resolution and 11 distinct land cover classes, this is the most detailed land cover map of France available.
In early 2024, we released the Scalgo land cover map for Germany, building on previous work in Denmark, Sweden, Finland and Norway. The launch of the land cover map in France shows new advancements in our AI model, which has been under continuous development since 2019.
A detailed landcover map is now available for the entire France.
“Our AI model is very powerful because it keeps learning and getting better every time we apply it,” Jonas Tranberg, data scientist at Scalgo, says. “And the initial results we got for France, using a model trained on the Nordic countries and Germany, were already very accurate. The new thing we trained the model to identify is dry vegetation, since vegetation in the south of France dries out in summer to a degree that is uncommon in northern Europe.”
Land cover has immense impact on the amount of runoff on the surface, and the map’s accurate description of various land covers in high resolution provides valuable input for local as well as regional analysis, ultimately enabling improved planning of surface water.
Our AI model is now trained to identify dry vegetation which will be very beneficial when we move more south from the Nordic countries.
Powerful AI methodology
The Scalgo land cover map distinguishes between 11 land cover classes in 25 cm resolution.
To create these maps, we use Artificial intelligence (AI), more specifically a variation of the UNET model, which is trained to identify land cover from orthophotos. In France, we use BD ORTHO aerial images provided by the Institut national de l'information géographique et forestière (ING).
As Jonas explains, “We create the AI output by alternating between training and evaluation phases. We start with a model trained on orthophotos outside of France to identify areas where performance is insufficient. Next, we add new training data from these areas and re-train. We repeat the process until the model is sufficiently accurate, which entails at least 95% accuracy on randomly sampled locations.”
The AI model creates four land cover classes: impervious areas, bare soil (i.e., sand and gravel), shallow vegetation and dense vegetation. Finally, the AI output is combined with auxiliary data, specifically BD TOPO and Registre Parcellaire Graphique (RPG) data from the Institut National de l'Information Géographique et Forestière (IGN). This results in more land cover classes including buildings, roads, railroads, water and agricultural fields.
The land cover has a high accuracy and can distinguish between various natural and artificial surfaces.
Using the map for improved assessments
Over the past few decades, the degree of surface artificialisation (land that humans have ended up transforming) has increased, although many policies have been aimed at halting or reversing this trend. In France, we have combined the land cover map with the official administrative boundary maps to provide detailed land cover statistics. As a result, you can see the current degree of artificialisation in your arrondissement, commune, department or region.
What commune has the highest forest coverage? Where would you find the greatest density of buildings? How are the most environmentally efficient regions planned? This and more you can get answers to using the land cover map.
Calculations of land cover statistics for each commune in France have been made. In Bordeaux, artificial surfaces cover 53 %.
In addition, land cover statistics is now included in the watershed info box. Click anywhere in France with the interactive Watershed tool to query a detailed breakdown of land cover within the watershed.
Land cover information is integrated into the Watershed query tool. Here is an example of a highly paved watershed in Paris.
Integrating the land cover to your flood analysis
To refine Flash Flood Map in Scalgo Live, you can now assign a runoff function to each of the land cover classes or to groups of classes using the infiltration module. It’s also easy to add your own land cover classes for special cases such as, for example, a green roof or a pervious parking lot. Check out this getting started video for a detailed description on how to include infiltration to Scalgo Live.
Infiltration changes surface runoff in the Flash Flood Map.
Discover Scalgo Live
Contact us at info@scalgo.com with any questions or click here to request a free 30-day trial period.
