See the complete article: Publicly available global environmental layers

  • Edited by: T. Hengl

Land cover maps are categorical-type maps, commonly derived using semi-automated methods and remote sensing images as the main input. There are at least four global land cover mapping projects in the world where such data can be found (they differ in legends, resolution, temporal coverage etc), from which the following two need to be emphasized. The most detailed and up-to-date global land cover maps are the ones produced by the European Space Agency — the GlobCover Land Cover version V2. This map was produced using a time-series of ENVISAT MERIS images and was the highest resolution (300 meters) Global Land Cover product in 2012. The second important source of land cover data is the MODIS12C1 Land Cover Type Yearly L3 Global product available in resolution from 500 m to 0.05 arcdegrees. The advantage of using the MODIS Land cover maps (17 land cover classes based on the International Geosphere Biosphere Programme IGBP classification system) is that this is a temporal dataset so that one can also derive various change indices and quantify the land cover dynamics (Friedl et al. 2002).

A detailed Water mask of the world is the GLCF watermask (Carroll et al. 2009). The 250 m resolution tiles can be obtained from the GLCF ftp server.

ESA produced in 2012 a 100 m resolution land cover map of the World from the Proba-V remote sensing system. The National Oceanic and Atmospheric Administration NOAA is also maintaining a global land cover map taken by the Suomi NPP satellite. Vegetation Imagery from the Soumi NPP Satellite can be obtained via the NOAA website.

FAO's Global Land Cover Network released in 2014 Global Land Cover-SHARE product (approximately 1 km resolution; a merge from various other land cover databases and following the ISO standard Land Cover Classification System), which focuses only on a number of classes of general interest for land management such as: cropland, grassland, mangroves, bare soil, mangroves etc (Latham et al. 2014). The percentage cover maps per class available for download via the FAO GeoNetwork. The classification accuracy for this product is about 80% (assessed using 1087 ground locations). Even more ground truth observations of land cover can be obtained from geo-wiki.org crowd sourcing project.

Global land cover / land cover change maps can also be derived from the Landsat imagery but these require significant storage and computing capacities. For example, the GlobalForestWatch.org imagery showing deforestation/reforestation were derived from the 30 m resolution Landsat images (Hansen et al. 2013). The global Landsat mosaics (Global Land Survey, GLS) are available for download as harmonized scenes from the University of Maryland GLCF (Global Land Cover Facility, www.landcover.org), but these datasets are ca 1-2.5TB in size and require significant processing facilities.

In 2014, Chinese government (National Geomatics Center of China) produced two global high-resolution (30 m) full coverage land cover maps for years 2000 and 2010. This product is known as GlobeLand30 and is publicly available for download (non-commercial use only; no derivative products permited) via www.globallandcover.com. The two 30-meter resolution land cover maps show global distribution of ten major land cover classes: Water bodies, Wetland, Artificial Surfaces, Cultivated land, Permanent snow and ice, Forests, Grasslands, Shrublands, Bareland and Tundra. GlobeLand30 products are free of charge for scientific research or public welfare undertakings.

Japan Aerospace Exploration Agency (JAXA) has also released in 2014 a global 25m-resolution PALSAR mosaic and forest/non-forest map (2007-2010), which apparently provides an average classification accuracy of 90%. This data is freely available for download via the JAXA pages.

Other known global land cover maps are the Global Land Cover map for the year 2000 (GLC2000) at 1 km resolution is distributed by the Joint Research Centre in Italy (Bartholome et al., 2002). A slightly outdated (1998) global map of land cover is provided by the AVHRR Global Land Cover Classification at resolutions of 1 and 8 km (Hansen et al. 2000). Likewise, International Steering Committee for Global Mapping provides access to the Global Land Cover by National Mapping Organizations (GLCNMO) map, produced using MODIS data observed in 2003.

The Global Land Cover Map for the Year 2000, 2003. GLC2000 database, European Commision Joint Research Centre.

In addition to standard land cover maps, Ellis and Ramankutty (2008) prepared the first global map of the anthropogenic biomes (18 classes; read more) showing dense settlements, villages, croplands, rangelands, forested lands and wildlands. FAO distributes a number of global thematic maps (usually at 5 arcmin resolution) including: suitability for various type of land use, soil and water resources, maps of environmental conditions, and similar. This can be downloaded directly from the FAO's GeoNetwork server.

The International Water Management Institute also produced the Global map of Irrigated Areas (GMIA; 28 classes) and the Global map of Rainfed Cropped Areas (GMRCA), both at 10 km resolution, and based on the twenty years of AVHRR images, augmented with higher resolution SPOT and JERS-1 imagery. A Global Map of Irrigation Areas is also available for download (ArcInfo ASCII grids and shape files) via the FAO's Information System on Water and Agriculture website.

Center for Sustainability and the Global Environment (SAGE), a research center of the Nelson Institute for Environmental Studies at the University of Wisconsin-Madison, distributes a number of global Land Use products: (1) Crop Calendar Dataset, (2) harvested area and yields of 175 crops, (3) historical and current cropland maps and similar. 1—arcmin resolution global yield estimates for 175 crops (year 2000) can be obtained directly from the Department of Geography, McGill University website. A number of human impacts, land use, water resources, urban extent (Schneider et al. 2009), and ecosystem maps can be browsed more systematically via the SAGE's AtlasBiosphere server.