Africa Open DEAL
Parameter
Land Use
Represented classes
Forest
Grassland
Cropland
Wetland
Settlement
Otherland
No data
Description
Land Use change map using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Forest subdivisions
Broadleaf evergreen
Broadleaf decidious
Broadleaf mixed
Mixed forest (brod-con)
Coniferous evergreen
Coniferous decidious
Coniferous mixed
Coniferous plantation
Palm
Other plantation
Mangrove forest
Riparian forest
Eucalyptus plantation
Acacia plantation
Non forest
Forest in Africa by subdivision, using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Grassland subdivisions
Grassland
Grassland with shrubs
Grassland with trees
Grassland with trees and shrubs
Shrubland
Shrubland with trees
Lichens and mosses
Non grasslands
Grassland in Africa by subdivision, using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Cropland subdivisions
Land under temorary crops
Orchard
Palm
Rice paddy
Greenhouse
Land under permanent crops
Non cropland
Croplands in Africa by subdivision, using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Land Use Change - Cropland transition Heatmap
Cropland extension: low - high
Land Use Change hotspots of land that has been transformed to CROPLAND, using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Land Use Change - Deforestation Heatmap
Deforestation: low - high
Land Use Change hotspots of forest land that has been transformed another land use (deforestation), using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Tree Cover outside of Forest
Tree Cover % outside Forest:
0% - 91-90%
Percentage of trees cover in areas that are not classified as forest, using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
Trees/ha outside of Forest
Trees/ha outside of Forest:
1tree/ha - >30 trees/ha
Number of single trees per hectare in areas that are not classified as forest, using the data collected through Africa Open DEAL more than 300k plots collected through Collect Earth between 2019 and 2020. Timeframe of the assessment is 2000-2019.
FAO- The African Union
Source data
FAO/ African Union
Methodology
More than 350 African experts with knowledge of landscapes, GIS and land uses conducted the data collection using the free open-source software tool Collect Earth. The interpretation was made throughout 2019‒2020 in 16 nationally and regionally focused workshops (Collect Earth trainings and group data collection, called Mapathons) convened by FAO in collaboration with government and regional institutions including the Panafrican Agency of the Great Green Wall (PA-GGW) and the Southern Africa Development Community (SADC).
Sampling Design
The assessments draw on information from more than 300 000 sampling plots in Africa. The plots were distributed over stratified systematic grids for which the continent has been divided in (i) the hyper-arid zones, which were sampled at a lower intensity (20x20km) because of the relative homogeneity of the landscape, and (ii) the nonhyper arid areas with a sampling intensity of 10x10 km. In addition, the Great Green Wall area has had a sampling design with a denser grid (6.5x6.5 km). At the national level, some of the countries opted for higher sampling density (e.g. Tunisia 4x4 km or Eswatini 2x2 km) in order to improve the accuracy of collected data
Collected Data
Around 120 different environmental variables and land parameters were collected, including numbers and density of trees, existence of infrastructures, wild fires etc.
Data Accurracy
3 200 plots (or 1 percent of the total sampling plots) were randomly selected throughout the continent, and were re-interpreted by FAO experts to quantify the measurement error. The overall accuracy, also including the sampling error, is estimated at over 90 percent over the whole sample with specific parameters, with uncertainties below 3 percent.