30 m resolution dataset of ecosystem classes of the Congo Basin mapped using the method developed by the World Bank Congo Basin Forest Ecosystem Accounts, modified for the Global Ecosystems Atlas
- Creators
- Global Ecosystems Atlas
Description
Purpose
This dataset presents spatial information on ecosystem classes for six countries in the Congo Basin, compiled using a modified method first described by:
- World Bank Group. (2025) Congo Basin Forest Ecosystem Accounts and Policy Recommendations: A Regional Synthesis Report of Ecosystem Extent, Condition, Services, and Asset Accounts 2000-2020. © World Bank Group. https://doi.org/10.1596/43866
This compilation follows the methods used in the World Bank Congo Basin Forest Ecosystem Accounts, with an additional processing step to distinguish between tropical lowland forest and tropical dry forest. This extra step is necessary to cross-reference the data to the IUCN Global Ecosystem Typology.
The output supplements the original information presented in the World Bank Congo Basin Forest Ecosystem Accounts by:
- Improving the spatial resolution to 30m (from 100m resolution)
- Providing a transparent and replicable workflow in Google Earth Engine.
- Updating the input data on Tropical Moist Forest extent to the 2025 version.
- Distinguishing between tropical moist and dry forests.
- Developing a cross-referencing table to level 3 of the Global Ecosystems Typology.
- Licensing the data under an open licence (CC-BY)
Data files
- CongoBasin_EarthEngine.tif (469.40 MB): A GeoTIFF raster file with the 18 ecosystem classes
- CongoBasin_EarthEngine.qml (9.14 KB): An optional legend file for use in QGIS
- CongoBasin_EarthEngine.js (15.21 KB): The code to replicate the processing steps in Google Earth Engine
- dry_forest_mask4.zip (152.44 KB): A zipped shapefile vector, indicating the out limits of dry forests
- congobasin_worldbank_GEA_crosswalk_table.csv (3.26 KB): A cross-referencing table between the 18 classes and the IUCN Global Ecosystem Typology
- congobasin_worldbank_membmat_template.csv (3.85 KB): A cross-referencing membership matrix between the 18 classes and the IUCN Global Ecosystem Typology
Input datasets
Five datasets are included combined into a single map of ecosystem:
1. Global Land Analysis and Discovery (GLAD):
- Reference: Potapov, P., et al. (2022) The global 2000-2020 land cover and land use change dataset derived from the Landsat archive: first results. Frontiers in Remote Sensing. 3: 856903. https://doi.org/10.3389/frsen.2022.856903
- Data access: https://glad.umd.edu/dataset/GLCLUC2020
- Earth Engine Asset: projects/glad/GLCLU2020/v2/LCLUC_2020
- License: CC-BY
2. Tree percent canopy Cover Global Forest Watch
- Reference: Hansen, M. C., et al. (2013) High-Resolution Global Maps of 21st-Century Forest Cover Change. Science 342: 850-53. https://doi.org/10.1126/science.1244693
- Data access: https://storage.googleapis.com/earthenginepartners-hansen/GFC-2023-v1.11/download.html
- Earth Engine Asset: UMD/hansen/global_forest_change_2025_v1_13
- License: CC-BY
3. European Commission Joint Research Centre Tropical Moist Forest
- Reference: C. Vancutsem, et al. (2021). Long-term (1990-2019) monitoring of forest cover changes in the humid tropics. Science Advances. https://doi.org/10.1126/sciadv.abe1603
- Data access: https://forobs.jrc.ec.europa.eu/TMF/data
- Earth Engine asset: projects/JRC/TMF/v1_2025/TransitionMap_Subtype
- License: CC-BY equivalent
4. WWF Hydrosheds DEM
- Reference: Lehner, B., et al (2008). New global hydrography derived from spaceborne elevation data. Eos, Transactions, American Geophysical Union, 89(10): 93–94. https://doi.org/10.1029/2008eo100001
- Data access: https://www.hydrosheds.org/
- Earth Engine asset: WWF/HydroSHEDS/03CONDEM
- License: CC-BY
5. Congo Basin Peatlands
- Reference: Crezee, B., et al. (2022). Mapping peat thickness and carbon stocks of the central Congo Basin using field data. Nature Geoscience. 15, 639–644. https://doi.org/10.1038/s41561-022-00966-7
- Data access: https://congopeat.net/maps/
- Earth Engine asset: Not available; download from source and upload as asset
- License: CC-BY
Mapped classes:
- Lowland forest: Forest ecosystem found in low altitude, commonly below 700-1100 m above sea level.
- Submontane forest: Transition zone between montane and lowland forests, typically starting from around 700-1100 m above sea level up to around 1500 m.
- Montane forest: Forest ecosystem found in higher elevation mountainous areas, typically above around 1500 m
- Swamp forest (non-peat): Seasonally and permanently inundated forests that occur along rivers and poorly drained areas.
- Peat forest: A type of swamp forest that is underlain by a thick layer of peat, dynamic and influenced by water level fluctuations.
- Woodland: Less dense than a forest with deciduous trees and drier conditions.
- Savanna: A mix of scattered trees and shrubs and open grasslands with distinct wet and dry seasons.
- Wetland: Areas where land is saturated with water throughout the year or seasonally, often found along rivers and lakes.
- Sparse vegetation: Areas with little to no plant cover.
- Forest-farm mosaic: A mixture of secondary forest regrowth, active and fallow agricultural fields, home gardens and village plantations.
- Cropland: Areas dominated by annual crops with little to no natural vegetation remaining.
- Plantation: Tree crops established by planting and/or seeding, typically monospecific stands with exotic species or intensively managed stands of indigenous species.
- Built-up: Areas dominated by built infrastructure, including roads.
- Mangroves: Coastal forest made up of salt-tolerant trees and shrubs.
- Open water: Areas of water, such as lakes or rivers, where the water surface does not have significant vegetation.
- Ocean: Ocean
- Dry Forest*: Closed canopy forest with seasonal periods of water deficit and deciduous tree species
- Ice*: Land covered by permanent ice or snow
* Ecosystem classes not originally included in the World Bank Congo Basin Forest Ecosystem Accounts
Modifications to delineate Tropical Dry Forest
A mask was created to delineated tropical forest into two classes: tropical lowland forest (Class 1), and tropical dry forest (Class 17).
To differentiate these classes a range of authorities on dry forest distribution in the northern and southern Congo Basin were consulted; including: Réjou-Méchain et al. (2021); White (1983), Shapiro et al (2021) and Sayre et al (2013).
Bioclimatic data from Hijmans et al (2005) allowed key precipitation and seasonality thresholds for ecosystem functional group T1.2 (Dry forests: see Keith et al., 2022) to be tested and ensure that the resulting Dry forest (T1.2) units identified conformed to the Global Ecosystem Typology (i.e. ~1800 mm per year mean annual precipitation, with a period of 3 to 6 months receiving less than 100 mm per month).
A dry forest mask was created as a polygon by heads-up digitizing the probable southern and northern boundary of the Lowland forest type, guided by the above-mentioned data sources.
Citations
- Hijmans, R.J., et al. (2005) Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology, 25,1965-1978. https://doi.org/10.1002/joc.1276
- Réjou-Méchain, M., et al. (2021) Unveiling African rainforest composition and vulnerability to global change. Nature, 593, 90–94. https://doi.org/10.1038/s41586-021-03483-6
- White, F.A. (1983) The Vegetation of Africa; a descriptive memoir to accompany the UNESCO/AETFAT/UNSO vegetation map of Africa. https://unesdoc.unesco.org/ark:/48223/pf0000058054
- Sayre, R., et al. (2013). A new map of standardized terrestrial ecosystems of Africa. Washington, DC: Association of American Geographers. ISBN 978-0-89291-275-9
- Shapiro, A C., et al. (2021) Forest condition in the Congo Basin for the assessment of ecosystem conservation status. Ecological Indicators, 122, 107268 https://doi.org/10.1016/j.ecolind.2020.107268
License
The dataset and Earth Engine scripts are available under a CC-BY 4.0 which allows users reuse the data with attribution.