landcover

Landcover refers to the physical material present on the Earth's surface. It encompasses natural features such as forests, grasslands, wetlands, bare soil and rock, water bodies, and built environments, as well as agricultural fields. Landcover is distinct from land use, which describes how people modify or utilize the land, for example for farming, housing, or industry.

Classification schemes group surface material into categories. Common broad classes include forest, shrubland, grassland, cropland, urban or built-up, water, wetlands, barren or sparsely vegetated areas, and snow or ice. National and regional maps may use finer or different schemes, and many schemes permit mixed or transitional classes.

Landcover maps are produced primarily from remote sensing and geographic information systems. Satellite imagery (for example Landsat, Sentinel-2, MODIS) is analyzed using supervised or unsupervised classification, or object-based approaches. Maps are validated with field observations and reference data, and accuracy is quantified with confusion matrices or kappa statistics.

Landcover data support climate and biodiversity research, water resources, agriculture, forestry, urban planning, and disaster management. They are used to monitor ecosystem extent, habitat loss, land change dynamics, and the effects of land management on carbon stocks.

Global datasets include GlobCover (ESA), Globeland30, and the Copernicus Global Land Service, as well as national products such as the United States NLCD and European CORINE Land Cover. These products are updated at varying intervals to reflect recent changes.

Limitations include seasonal phenology, sensor resolution, misclassification of mixed pixels, and rapid change in urban or agricultural landscapes. Users should consider accuracy assessments and metadata when applying landcover maps.