Data sources
The datasets that power the Global Ecosystems Atlas. Open any source to view it at its origin.
The Global Ecosystems Atlas combines existing ecosystem data in a single, standardised synthesis map. This synthesis map licensed under a Creative Commons Attribution 4.0 International Licence (CC-BY 4.0), which allows users reuse the data with the following attribution:
GEO ([Insert year])). Global Ecosystems Atlas ([Insert data version]). Group on Earth Observations. https://globalecosystemsatlas.org/data-sources
We urge users to include the link to this data sources page when citing the Global Ecosystems Atlas to acknowledge all the contributors to this global initiative.
A 10 m resolution land cover map of the Tibetan Plateau with detailed vegetation typesA 10 m resolution land cover map of the Tibetan Plateau with 12 vegetation types and 3 non-vegetation types for the year 2022 (TP_LC10-2022) by leveraging state-of-the-art remote sensing approaches including the Sentinel-1 and Sentinel-2 imagery, environmental and topographic datasets, and Random Forest model using Google Earth Engine platform. | Wuhan University | 2024 | CC BY 4.0 | |
A high-resolution map of Singapore’s terrestrial ecosystemsThe natural and semi-natural areas within cities provide important refuges for biodiversity, as well as many benefits to people. To study urban ecology and quantify the benefits of urban ecosystems, we need to understand the spatial extent and configuration of different types of vegetated cover within a city. It is challenging to map urban ecosystems because they are typically small and highly fragmented; thus requiring high resolution satellite images. This article describes a new high-resolution map of land cover for the tropical city-state of Singapore. We used images from WorldView and QuickBird satellites, and classified these images using random forest machine learning and supplementary datasets into 12 terrestrial land classes. Close to 50 % of Singapore’s land cover is vegetated while freshwater fills about 6 %, and the rest is bare or built up. The overall accuracy of the map was 79 % and the class-specific errors are described in detail. Tropical regions such as Singapore have a lot of cloud cover year-round, complicating the process of mapping using satellite imagery. The land cover map provided here will have applications for urban biodiversity studies, ecosystem service quantification, and natural capital assessment. | Natural Capital Singapore, Campus for Research Excellence and Technological Enterprise, National Parks Board Singapore | 2019 | CC BY 4.0 | |
A new satellite-derived dataset for marine aquaculture in the China's coastal regionThis dataset contains the marine aquaculture map in the China’s coastal region with a spatial resolution of 16 m.
We provided this dataset in KMZ format (.kmz), which can be dierctly opened by Google Earth. Alternative options are also available in ESRI shapefile format (.rar), which can be opened by ArcGIS. | Zhejiang University, University of Leeds, University of Hong Kong | 2020 | CC BY 4.0 | |
Allen Coral AtlasThe Allen Coral Atlas is built by a dedicated team of scientists, technologists, and conservationists, using one-of-a-kind methodologies. The Allen Coral Atlas was conceived and funded by the late Paul Allen’s Vulcan Inc. and is managed by the Arizona State University Center for Global Discovery and Conservation Science. Along with partners from Planet, the Coral Reef Alliance, and the University of Queensland, the Atlas utilizes high-resolution satellite imagery and advanced analytics to map and monitor the world’s coral reefs in unprecedented detail. The partnership together identified the following methods for habitat map creation, dynamic monitoring, and other related coral reef products. | Allen Coral Atlas Partnership, University of New South Wales, University of Queensland, James Cook University, Australian Institute of Marine Science, Coral Reef Alliance, Arizona State University, National Geographic Society, Planet, Vulcan | 2022 | CC BY 4.0 | |
Antarctic Ecosystem Inventory: Spatial data for Ice-free lands v1.0This is Antarctica’s first comprehensive ecosystem map of ice-free lands. The data comprise a spatially explicit 3-tiered hierarchical ecosystem classification with nine Major Environment Types (tier 1), 33 Habitat Complexes (tier 2) and 269 Bioregional Ecosystem Types (tier 3). These Bioregional Ecosystem Types are aligned with ‘level 4’ of the IUCN Global Ecosystem Typology (Keith et al. 2022).
The spatial data are available in raster format (TIF) at 100 m resolution in the Polar Stereographic Projected Coordinate System (GCS_WGS_1984) for all known ice-free areas south from latitude -57.330551 decimal degrees South (pdf map shows extent of ice-free areas in relation to terrestrial ice and ice shelves). A value attribute table (VAT) provides text fields containing codes and full names for each unit in each level of the classification hierarchy and the spatial extent of tier 3 units in hectares.
Methods of development, source data and uses of the inventory are detailed by Tóth et al. (2024a). Descriptive profiles for tier 1 and 2 units are available in Tóth et al. (2024b). | University of New South Wales, Australian Antarctic Division, Monash University, Natural Environment Research Council, University of Johannesburg, Millennium Institute Biodiversity of Antarctic and Subantarctic Ecosystems (BASE), University of Pretoria, James Cook University, University of Wollongong, Queensland University of Technology, South Australian Museum, University of Adelaide | 2024 | CC BY 4.0 | |
Aquaculture Land Cover DataClark Labs, in partnership with the Gordon and Betty Moore Foundation and in support of the Foundation’s Oceans and Seafood Markets Initiative, is mapping an inventory of pond aquaculture and coastal habitats around the world. Sixteen countries thus far have been mapped for landcover for the years 1999, 2014, and 2018. Several of these countries have been mapped to include 2020. Along with landcover, for each country a change analysis is performed at the provincial or state level between all years mapped. In addition to downloading the data you can explore the data using an interactive web map.
The landcover maps for 2014 and greater are based on Landsat 8 imagery. Landcover maps for 1999 data are based on Landsat 5 imagery. Three main landcover categories are mapped: mangrove, coastal wetland, and pond aquaculture. When possible, a distinction is made between fresh and brackish pond aquaculture. The study area of the coastal zone for mapping landcover is defined as a zone 10 km on either side of the coastline. Where necessary, the zone is extended to include land areas <= 5 m in elevation to a maximum extension of 60 km inland from the coast. The primary concern is to limit the inland extent to areas where it was likely that pond aquaculture would be dominated by brackish water and thereby have a stronger likelihood of being used for shrimp. | Clark University | 2024 | CC BY 4.0 | |
Atlas de Cobertura del SueloLa cobertura del suelo en Uruguay ha experimentado cambios significativos en las últimas décadas y el monitoreo informado del territorio es fundamental para entender estos cambios y planificar el uso y manejo de los recursos naturales de manera sustentable.
El Atlas de cobertura del suelo de Uruguay es resultado del trabajo en conjunto, que se realiza desde el año 2008, entre el Ministerio de Vivienda y Ordenamiento Territorial (MVOT) y la Organización de las Naciones Unidas para la Alimentación y la Agricultura (FAO).
En esta oportunidad se optó por un formato dinámico que facilita la consulta, comparación y actualización continuas de los datos de cobertura del suelo, así como la descarga de los datos geoespaciales para realizar análisis específicos. Se da continuidad de esta forma a las líneas de trabajo desarrolladas por la Dinot y donde desarrolló un rol fundamental la colaboración y coordinación interinstitucional con organismos como el Ministerio de Ambiente (MA) y el Ministerio de Ganadería, Agricultura y Pesca (MGAP).
Este atlas busca proporcionar información actualizada sobre la cobertura del suelo en Uruguay para tomadores de decisiones, investigadores, académicos y todos los que trabajan en temas relacionados con la gestión territorial y el desarrollo sostenible. | Ministerio de Vivienda y Ordenamiento Territorial | 2024 | Custom | |
Australian saltmarsh and sparsely vegetated saltmarsh map version 1.0The dataset contains a national map of saltmarsh ecosystems produced via a classification of Landsat 8 images. Each pixel was classified into four map
classes: saltmarsh, saltflats, other terrestrial (land), and permanent water.
The map output of this classification and associated map data files are described below:
1. JCU_Australia-saltmarsh-extent_v1-0.tiff: Contains a national map of saltmarsh ecosystems produced via a classification of Landsat 8 images. Pixel value is either 0 (Non Saltmarsh) or 1 (Saltmarsh).
2. JCU_Australia-saltmarsh-probability_v1-0.tiff: Contains the associated probability map associated with the saltmarsh map. Pixel value represents percentage probability with values ranging from 0 – 100 (%).
3. JCU_Australia-saltflat-extent_v1-0.tiff: Contains a national map of saltflats produced via a classification of Landsat 8 images. Pixel value is either 0 (Non Saltflats) or 1 (Saltflats).
4. JCU_Australia-saltflat-probability_v1-0.tiff: Contains the associated probability map associated with the saltflat map. Pixel value represents percentage probability with values ranging from 0 – 100 (%).
5. JCU_MappingRegion_v1-0.geojson: Contains the spatial extent at which the mapping classification was done.
6. JCU_ManualMask_Saltflat_v1-0.geojson: Contains polygons manually created to mask out known misclassification of saltflats.
7. JCU_ManualMask_Saltmarsh_v1-0.geojson: Contains polygons manually created to mask out known misclassification of saltmarsh. | James Cook University, University of New South Whales, Digital Earth Australia | 2023 | CC BY 4.0 | |
Carta Degli Ecosistemi d’ItaliaA second version of the Ecosystems Map of Italy is presented that updates the one realized for the Mapping and Assessment of Ecosystems and their Services (MAES) process. The map represents a renewed reference for the implementation of biodiversity-related policies in the country, including the Red List of Ecosystems, Ecosystem Accounting under the UN System of integrated Environmental and Economic Accounts, EU Nature Restoration Law, and, more in general, for the support of planning initiatives aimed at climate change adaptation and recovery of degraded ecosystems in keeping with restoration ecology principles. The mapping approach recalled the earlier rationale, i.e. that current and potential vegetation are valuable proxies for outlining ecosystems, but basic information on land cover and characterization of types have been updated. Additionally, a detailed description of the mapping procedure is provided that may facilitate replication in time, validation processes and comparison with different maps. The crosswalk between the Italian ecosystem typology and other classification systems, already available for Corine Land Cover and EUNIS habitats, was therefore revised and complemented with respect to the IUCN Global Ecosystem Typology. Finally, future perspectives for a regular updating of the map and further improvement of its geometric and thematic detail are sketched. | Società Botanica Italiana, Sapienza University of Rome, Ministry of Environment and Energy Security Italy | 2021 | CC BY 4.0 | |
Circumpolar Arctic Land Cover for circa 2020 (CALC-2020)Circumpolar Arctic Land Cover for circa 2020 (CALC-2020) is a new baseline land cover product that maps the entire terrestrial Arctic (tree line nowthward) at 10m resolution. This dataset is derived from Sentinel-1/2 and ArcticDEM, using locally adaptive machine learning models trained by sample over 58,000 site locations.
The CALC-2020 product is delivered as six independent TIFF files. Each of them spatially covers one of Arctic countries/regions including Alaska (AK), Canada (CA), Greenland (GR), Iceland (IC), Norway (NO), and Russia (RU). The CALC-2020's legend includes ten generic classes that appropriately describe landscapes across the Arctic: Cropland (01), Forest (02), Graminoid tundra (03), Shrub tundra (04), Wetland (05), Open water (06), Lichen/moss (07), Man-made imperviousness (08), Barren land (09), Ice/snow (10). | Sun Yat-sen University | 2022 | CC BY 4.0 | |
Cocoa plantations are associated with deforestation in Côte d’Ivoire and GhanaCôte d’Ivoire and Ghana, the world’s largest producers of cocoa, account for two thirds of the global cocoa production. In both countries, cocoa is the primary perennial crop, providing income to almost two million farmers. Yet precise maps of the area planted with cocoa are missing, hindering accurate quantification of expansion in protected areas, production and yields and limiting information available for improved sustainability governance. Here we combine cocoa plantation data with publicly available satellite imagery in a deep learning framework and create high-resolution maps of cocoa plantations for both countries, validated in situ. Our results suggest that cocoa cultivation is an underlying driver of over 37% of forest loss in protected areas in Côte d’Ivoire and over 13% in Ghana, and that official reports substantially underestimate the planted area (up to 40% in Ghana). These maps serve as a crucial building block to advance our understanding of conservation and economic development in cocoa-producing regions. | ETH Zurich | 2023 | CC BY 4.0 | |
Conjunto de datos vectoriales de uso del suelo y vegetación. Escala 1:250 000. Serie VII. Conjunto NacionalThis map contains information on Land Use and Vegetation obtained from the application of photointerpretation techniques with Geomedian images with a base year of 2018, generated from the LANDSAT satellite constellation in multispectral format. This interpretation is supported by field work. The Data Set contains the location, distribution and extent of different plant communities and uses, with their respective variants in vegetation types, agricultural uses, and relevant ecological information. Said digital geographic information contains data structured in vector form coded in accordance with the Dictionary of Vector Data on Land Use and Vegetation Scale 1:250,000 (version 3) applicable to the different ecological units (plant communities and anthropic uses) contained in the data set. | Instituto Nacional de Estadística y Geografía (INEGI) | 2021 | Ok to use for POC - check again for main Atlas | |
Cote d'Ivoire BNETD 2020 Land Cover MapThe Cote d'Ivoire BNETD 2020 Land Cover Map was produced by the Ivorian Government through a national institution, the Center for Geographic Information and Digital from the National Study Office Techniques and Development (BNETD-CIGN), with technical and financial support from the European Union. The methodology used to produce the map was transparent, participatory and in line with international standards.
To develop this map, a mosaic of satellite images (Sentinel 2) from 2020 was processed via Google Earth Engine and supplemented with data collected in the field, to train a supervised classification algorithm (Random Forest). Two field campaigns were conducted, from 10 November to 9 December 2022 and from 26 January to 13 February 2023, throughout the country. These missions involved 33 people from multiple partner organizations because the data collection methods and definitions of certain land use classes adopted by stakeholders may sometimes differ.
As part of the EUDR due diligence process, geolocation data for plots of land producing EUDR-relevant products could be overlaid with 2020 forest cover data in order to assess the risk that the plot is located in an area that was forested before the 2020 cut-off date. To do this, forest cover data aligned with the FAO definition of forests and the 2020 cut-off date is required. The 2020 land cover map of Cote d'Ivoire meets these needs. Indeed, the classes in the land cover map can be combined to create a forest/non-forest map that is aligned with the FAO definition of forests. | BNETD CIGN | 2020 | CC BY 4.0 | |
Dynamic World V1Dynamic World is a 10m near-real-time (NRT) Land Use/Land Cover (LULC) dataset that includes class probabilities and label information for nine classes.
Dynamic World predictions are available for the Sentinel-2 L1C collection from 2015-06-27 to present. The revisit frequency of Sentinel-2 is between 2-5 days depending on latitude. Dynamic World predictions are generated for Sentinel-2 L1C images with CLOUDY_PIXEL_PERCENTAGE <= 35%. Predictions are masked to remove clouds and cloud shadows using a combination of S2 Cloud Probability, Cloud Displacement Index, and Directional Distance Transform. | Google, World Resources Insitute, Nat Geo Society | 2022 | CC BY 4.0 | |
Ecossistemas Marinhos - MozambiqueData used for MPA Planning in Mozambique | Wildlife Conservation Society | 2022 | CC BY 4.0 |