The Occitanie Region in southern France has launched a three-year biodiversity mapping programme that will combine satellite observations, field surveys and artificial intelligence to create a continuously updated map of natural habitats across nearly 72,000 km².
The project will cover forests, grasslands, wetlands, agricultural land, coastal ecosystems and urban green spaces. Regional authorities say the initiative will support biodiversity conservation, climate adaptation and land-use planning by providing a more detailed picture of how habitats change over time.
The programme is being carried out by a consortium led by CLS, a subsidiary of the French National Centre for Space Studies (CNES), together with ecological engineering firms BIOTOPE, ECO-MED, ENVOLIS and NATURALIA.
Researchers will combine imagery from the European Space Agency’s (ESA) Sentinel satellites and other Earth observation systems with climate, geological, topographic and environmental datasets. Artificial intelligence models will analyse the information to identify and classify hundreds of habitat types across the region.
Unlike conventional habitat maps, which are updated periodically, the new system is designed to provide a dynamic view of biodiversity and ecosystem change.
Fieldwork remains a central part of the programme. Nearly 30 botanists will spend around 700 field days collecting ecological observations across Occitanie during a six-month campaign. These data will be used to train and validate the AI models, helping ensure that classifications derived from satellite imagery accurately reflect conditions on the ground.
The initiative also incorporates what CLS describes as a “frugal AI” approach, using models developed specifically for environmental monitoring rather than large general-purpose systems.
“Artificial intelligence is often discussed in terms of its environmental footprint. We believe the conversation should also focus on how AI is designed and what purpose it serves,” said Karim Mehah, Head of Service & Delivery, Environment & Climate at CLS.
If successful, the programme could demonstrate how satellite observations, ecological expertise and targeted artificial intelligence can be combined to support biodiversity monitoring and environmental management across large territories.
Disclaimer:
The use of brand names and/or any mention or listing of specific commercial products or services herein is solely for educational purposes and does not imply endorsement by Muser Press or our partners, nor discrimination against similar brands, products or services not mentioned.
Article Source:
Press Release/Material by CLS – Collecte Localisation Satellites
Featured image credit: jcomp | Magnific
