This work package is dealing with the production of the best possible automated land cover and land cover changes based on a time series of Sentinel-2 data for the study area in Poland and Norway. Verification of the EAGLE concept by interlinking satellite-based land cover (LC) with land use (LU) information and other landscape characteristics (CH) derived from national databases and the products of the Copernicus Land Monitoring Service (CLMS). Coordinated by the Institute of Geodesy and Cartography (IGIK).
Task 1.1: Land cover classification based on Sentinel-2 imagery
Task 1.2: Assessment of the existing databases
Task 1.3: Population of the land cover with the information on the land use
The team at IGiK is currently developing land cover classification for the pilot area at the Łódź province (Poland) and Viken country (Norway). The classification is carried out using a time series of Sentinel-2 data for the reference year 2018 and 2020. The classification is performed using the advance non-parametric machine learning algorithms.
In parallel, the teams at IGIK and NIBIO is evaluated existing land cover databases for Poland and Norway. National databases are assessed in terms of their accuracy, reference year, thematic content, comparability of definitions, reliability, accessibility.
The qualitative and quantitative analysis of national databases performed in task 1.2. is essential for the next steps (task 1.3). the aim of the task 1.3 is to enrich the land cover data (LC) by combining existing information on land use (LU) and other landscape characteristics (CH) available at the national level according to the EAGLE concept.