The global map of forest types for year 2020 (GFT 2020) combines available global datasets (wall-to-wall or global in their scope) that indicate or are proxies for the four main forest types (primary forest, naturally regenerating forest, planted forest, plantation forest) that are set out by the EU Regulation on deforestation-free supply chains (EUDR, Regulation (EU) 2023/1115). The definitions in this regulation generally follow definitions on forest characteristics and specific forest categories from the FAO Global Forest Resources Assessment.
Data for deriving a harmonized, globally consistent representation of forest types are scarce, and mapping of the respective forest types from Earth Observation data is challenging. This map of global forest types is released as a preliminary version (version 0) for open consultation by the user community which can be concerned or interested in the issue of forest degradation under the EUDR. Based on expected feedback and future additional or improved data sets, the JRC aims to produce a consolidated map during year 2025.
The base layer for mapping these forest types is the extent of forest cover of version 1 of the Global Forest Cover map for year 2020 (GFC 2020). The definition of forest degradation under the EUDR implies that it is only necessary to assess the presence of primary forests, naturally regenerating forests (other than primary forests) and planted forests in the year 2020. As the GFT 2020 map is intended for the phase of risk assessment only, the plantation forests are not identified as sub class within planted forests. Moreover “other wooded land” exists only outside the extent of forest cover and is therefore not mapped in GFT 2020.
The main data layers that are used to delineate primary forests in GFT 2020 are (i) the Forest Landscape Integrity Index in 2019, (ii) map of undisturbed tropical mangroves in 2020 and degradation and deforestation in tropical moist forests from 1990 to 2020 (from JRC-TMF), (iii) map of Intact Forest Landscapes 2020, (iv) World Database on Protective Areas, (v) the European Primary Forest Dataset from Sabatini et al. 2021, (vi) global tree cover loss from 2001 to 2020 in combination with the map of drivers of global forest loss and (vii) a map on global mining land use. The main layers to map planted forests (including plantation forests) are (i) the WRI Spatial Database on Planted Trees (version 2.1), (ii) IIASA Forest Management classes on planted/plantation forest, (iii) the Global Forest Canopy Height dataset in 2019 and (iv) global tree cover in 2000 and loss from 2001 to 2020.
A JRC science for policy report, expected to be published early next year, will provide a detailed description of the input layers and mapping approach used for this preliminary GFT 2020 map.
The following tables show a non-exhaustive list of issues that have not been corrected in the map. The table “by region” lists issues of broader, more general nature. Specific issues are listed in the table “by coordinates”, where the coordinates point approximately to the centre location where the issue was found. In many cases coordinates point to a location as an example for several other cases. Occasionally, these lists may be updated. Improvements of the map (higher versions) will aim to alleviate some of the known issues.
| Region | Description of the issue | Date | Source |
|---|---|---|---|
| Brazilian Amazon | Patches of primary forests may not meet the MMU (lower due to buffer around selective logging pixels) | 15/11/2024 | Producer |
| Global | Delineation of planted/plantation looking coarser than 10m resolution | 15/11/2024 | Producer |
| North America | Planted forest appears in blocks of ca. 100m. However, in some cases this general pattern is modulated with 10m data (10m holes, adjustments along boundaries) | 15/11/2024 | Producer |
| North America | The origin from a vector dataset is still visible for primary forests by having smooth forms or circles | 15/11/2024 | Producer |
| Mexico, Sierra Madre Occidental | Blocky pattern of 100m grid cells for primary forests when separated from naturally-regenerating forest, This effect is also visible in other regions but particularly notable here | 15/11/2024 | Producer |
| Canada, Alaska | There is a fringe of 100-200 km between primary forest in the boreal and the tundra non-forest area mapped as naturally-regenerating | 15/11/2024 | Producer |
| Kenya- Tanzania | Buffer on primary forest creates lots of fragmented patches | 15/11/2024 | Producer |
| Cote d'Ivoire | Patches of primary forests may not meet the MMU (lower due to buffer around selective logging pixels) | 15/11/2024 | Producer |
| France - Hautes Alpes | Old plantations / planted forests are mostly not depicted. Small patches of plantations (one or few 100m res pixels) look even more strange when located within primary forest's extent (e.g. in Parc des Ecrins) | 15/11/2024 | Producer |
| Boreal zone | Forests at highest latitudes appear as naturaly regenerating forests (very large areas in Canada and Russia) just above a long belt of primary forests. It might be the effect of a low forest integrity index | 15/11/2024 | Producer |
| Tropical zone - Northern Thailand and Northern Laos | The filtering of primary forests with a 250m buffer around tropical moist forest degraded pixels leads to very fragmented patches. Specifically in Northern Thailand and Laos where there is a lot of shifting cultivation, the map seems to overestimate forest areas and primary forests | 15/11/2024 | Producer |
| Longitude | Latitude | Description of the issue | Date | Source |
|---|---|---|---|---|
| 2.154516711 | 6.980361744 | Teak plantation in Benin, some part are not identified as forest in GFC 2020 and therefore are not classified as planted forest. The planted area is a bit patchy even if from the VHR it seems correct. Part of forest plantations are classified as regenerating forests | 15/11/2024 | Producer |
| -4.795415532 | 7.841880394 | Identified as 'primary forest' in GFT 2020 and as tree plantation/regeneration in national data of Cote d'Ivoire | 15/11/2024 | Producer |
| -4.69844158 | 7.541979696 | Identified as 'primary forest' in GFT 2020 and as tree plantation/regeneration in national data of Cote d'Ivoire | 15/11/2024 | Producer |
| -9.064815345 | 7.401115916 | Patchy effect of planted/plantation forest | 15/11/2024 | Producer |
| -59.984629 | -2.438423812 | Naturally regenerating forests north of Manaus are not so extensive as suggested by the map | 15/11/2024 | Producer |
| -54.37169801 | -4.438179672 | Naturally regenerating forests east of FLONA Tapajos are also not as extensive as indicated in the map | 15/11/2024 | Producer |
| -62.87939303 | -0.340967828 | I don 't see any evidence of naturally regenerating forests in this region - not sure why they are not being mapped as primary forest | 15/11/2024 | Producer |
| 109.8225 | 19.1256 | Wrong classification of plantation-planted forest | 15/11/2024 | Producer |
| 102.6654 | 0.4829 | Stripe classified as planted forest | 15/11/2024 | Producer |
| 98.9013 | 15.0235 | Intense buffering effect around primary forest extent | 15/11/2024 | Producer |
| 111.69568 | -2.46621 | Potential overestimation of planted/plantation forest | 15/11/2024 | Producer |
| -157.9121 | 21.3992 | Isolated planted/plantation pixels that should be classified in the forest type as their nearest neighbouring pixels | 15/11/2024 | Producer |
| 32.92007 | -26.46126 | Potential overestimation of planted/plantation forest | 15/11/2024 | Producer |
| 80.59996 | 5.9814 | Potential overestimation of planted/plantation forest | 15/11/2024 | Producer |
| -123.6859954 | 41.35208693 | Primary forest interfringes with planted forest | 15/11/2024 | Producer |
| -8.265423949 | 41.84438908 | Area where primary and planted forest are mixed | 15/11/2024 | Producer |
| 10.87935666 | 64.176489 | Isolated pixel of panted forest in the middle of primary forest | 15/11/2024 | Producer |
| -78.07976945 | 8.648594197 | Potential overestimation of planted/plantation forest | 15/11/2024 | Producer |
| -78.29487136 | 8.536957665 | Potential overestimation of planted/plantation forest | 15/11/2024 | Producer |
The preliminary global map of forest type for year 2020 (version 0) is available for download by 10°x10° tiles. The tiling starts at 180°W and 80°N with tile ID “N80” and “W180”, respectively. Tiles over the ocean are not provided. The file format is GEOTIFF.
The naming convention is as follows JRC_GFT2020_V0_[lat]_[lon].tif, where:
Map coordinates are in latitude and longitude, referenced to the WGS84 datum.
Click on the tile in the image below to start the download.
Each of the downloadable files can be displayed in desktop GIS tools. The table below contains symbology files to display the colourmap and labels for the values in QGIS and ArcGIS.
| Dataset | QGIS | ArcGIS |
|---|---|---|
| Global forest cover map 2020, Version 1 | JRC_GFT2020_V0.qml | JRC_GFT2020_V0.lyr |
The files for download are geolocated (latitude and longitude with WGS84 datum) but do not contain any metadata information. Please right-click in the table below to download the metadata file.
| Dataset | ISO 19139 Metadata file |
|---|---|
| Global map of forest types 2020, Version 1 | JRC_GFT2020_V0.xml |
Interactive visualization of the global forest type map along with frequently asked questions and other information can be found at the EU Observatory on deforestation and forest degradation website.
The global forest type map of 2020 can also be viewed in other websites or GIS clients as WMS (Web Map Service). They are not suitable for analysis as the data are represented only as RGB images. The WMS URL is: https://ies-ows.jrc.ec.europa.eu/iforce/gft2020/wms.py?
The Global map of forest types for year2020 (Version 0) can also be accessed and used in the Google Earth Engine platform: https://developers.google.com/earth-engine/datasets/catalog/JRC_GFC2020_subtypes_V0
All data here are produced under studies funded by the Directorate-General for Environment of the European Commission through the Dev-ForObs and DevForObs2 projects. All data are provided free-of-charge, without restriction of use. For the full license information see the Copernicus Regulation of the European Commission.
Publications, models and data products that make use of this dataset must include proper acknowledgement, including citing dataset as in the following citation:
Bourgoin, Clement; Verhegghen, Astrid; Carboni, Silvia; Ameztoy, Iban; Ceccherini, Guido; Colditz, Rene; Achard, Frederic (2024): Global map of forest types 2020 - version 0. European Commission, Joint Research Centre (JRC) [Dataset] PID: http://data.europa.eu/89h/037ca376-ba92-49db-a8f7-0c277c1e5436
The designations employed and the presentation of material on this map do not imply the expression of any opinion whatsoever on the part of the European Union concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.
