The ESRI Land Cover project offers an annual high-resolution map of Earth's land surface from 2017-2022, created using advanced AI land classification models. These models utilize a vast dataset of human-labeled image pixels and were applied to over 2 million Earth observations from Sentinel-2's six spectral bands. The maps, which feature 9 classes like vegetation, water, and built areas, aim to assist decision-makers in various industries and governments by providing valuable insights into land use, natural capital, and resource management. Available through the ArcGIS Living Atlas, these maps enable better understanding and informed decision-making regarding geographical changes over time.
- Trees
- Water
- Flooded Vegetation
- Bare Ground
- Crops
- Rangeland
- Snow/Ice
- Built Area