Skip to main content

Copernicus Global Land Cover 2019

Whether you’re monitoring crops, modelling green energy installations or soil sealing, combatting loss of natural resources or just helping countries meet their Sustainable Development Goals, chances are high that you’ll need an accurate and spatially detailed map on land cover and land use. Earth Observation satellites, like those from EU’s flagship programme Copernicus, are key to providing such maps, at a global scale, with free and open access. Land cover maps represent spatial information on different types (classes) of physical coverage of the Earth's surface, e.g. forests, grasslands, croplands, lakes, wetlands. Dynamic land cover maps include transitions of land cover classes over time and hence captures land cover changes. This dataset shows the land cover for the baseline year 2019 with a discrete classification in 23 classes aligned with UN-FAO's Land Cover Classification System.

Buchhorn, M. ; Smets, B. ; Bertels, L. ; De Roo, B. ; Lesiv, M. ; Tsendbazar, N. - E. ; Herold, M. ; Fritz, S. Copernicus Global Land Service: Land Cover 100m: collection 3: epoch 2019: Globe 2020. DOI 10.5281/zenodo.3939050


The version 3.0 annual 100m land cover classes were mapped with high temporal stability across years and an overall mapping accuracy just over 80% (80.6% in 2015, 80.3% in 2019), when compared to the 28K independent validation points. The statistical validation meets the CEOS Land Product Validation Stage 4 requirements. The updated 2019 classification map shows slightly higher accuracy than the previous version and improvements in characterizing forest, cropland and permanent water classes. Spatial accuracy assessment for three aggregated classes (forest, crops and natural vegetation) against a set of over 200K independent validation points revealed a high level of accuracy in different regions of the world and highlighted areas for further improvement.

Select for mapping