Continental-Scale Land Cover Mapping at 10 m Resolution Over Europe (ELC10)

Land cover maps are important tools for quantifying the human footprint on environment and facilitate reporting accounting to international agreements addressing Sustainable Development Goals. Widely used European land such as CORINE (Coordination of Information Environment) produced at medium spatial resolutions (100 m) rely diverse data with complex workflows requiring significant institutional capacity. We present a 10 m resolution map (ELC10) Europe based satellite-driven machine learning workflow that is annually updatable. A random forest classification model was trained 70K ground-truth points from LUCAS (Land Use/Cover Area Frame Survey) dataset. Within Google Earth Engine cloud computing environment, ELC10 can be generated approx. 700 TB Sentinel imagery within 4 days single research user account. The achieved an overall accuracy 90% across eight classes could account statistical unit proportions 3.9% (R2 = 0.83) actual value. These accuracies higher than other including S2GLC FROM-GLC10. Spectro-temporal metrics capture phenology were most in producing high mapping accuracies. found atmospheric correction Sentinel-2 speckle filtering Sentinel-1 had minimal effect enhancing (<1%). However, combining optical radar increased by 3% compared alone 10% alone. addition auxiliary (terrain, climate night-time lights) additional 2%. By using centroid pixels Copernicus module polygons we <1%, revealing forests robust against contaminated training data. Furthermore, requires very little achieve moderate accuracies—the difference between 5K 50K only (86% vs. 89%). This implies significantly less resources necessary making situ survey (such LUCAS) suitable satellite-based classification. At resolution, distinguish detailed landscape features like hedgerows gardens, therefore holds potential aerial statistics city borough level monitoring property-level environmental interventions (e.g., tree planting). Due reliance purely input data, continuously updated independent any country-specific geographic datasets.