Global Forest Attribute Dataset (GFAD): high-resolution spatial layers and statistics for monitoring forest morphology, connectivity, and accounting

1. Dataset Overview The Global Forest Attribute Dataset (GFAD) provides high-resolution (100 m) maps and statistics derived from GFC2020 forest map v.3 (Bourgoin et al. 2025) characterising forest connectivity, morphology (MSPA), and accounting at the global level and across 259 countries and territories. The indicators used are normalised on a scale of [0-100] and aggregated by country, facilitating neutral reporting and direct cross‑country comparisons of forest conditions. This dataset directly supports the monitoring requirements of the EU Regulation on Deforestation-free Products (EUDR). Further details on followed methodologies to derive GFAD and the full data description are available in the following data paper: Caudullo G, Vogt P, 2026. Global Forest Attribute Dataset (GFAD): high-resolution spatial layers and statistics for monitoring forest morphology, connectivity, and accounting. Annals of Forest Science 83, 13. 10.1186/s13595-026-01332-y 2. ZIP archives GFAD_results.zip Final products (maps and statistics) Source_data.zip Input layers for replication Intermediate_data.zip Pre-processing transition layers 3. Data scructure GFAD_results.zip 📁 Summary_Indices_Spreadsheets: Contains 29 FAL indices for 260 reporting units (Countries + EU27 + Global) in .CSV and .XLSX. 📁 FAL_maps: Final global and national GeoTIFF raster layers: 📁 Accounting: global acccounting map and statistics. Accounting count and classifies forest patches by size (8-connected). Patches are grouped into six size classes with thresholds at 100, 1k, 10k, 100k, 1M ha. Ref: GTB-Objects-Accounting.pdf 📁 CNTR: 259 national accounting maps 📁 Fragmentation: global fragmentation map and statistics, using Forest Area Density (FAD) method and a moving window: 31x31 pixels (~10 km2). Ref: GTB-Fragmentation-FADFOS.pdf 📁 CNTR: 259 national fragmentation maps 📁 Pattern-MSPA: global MSPA map and statistics. MSPA classifies forest pixels based on spatial context (Core, Islet, Edge, etc.). Ref: GTB-Pattern-Morphology.pdf 📁 CNTR: 259 national MSPA maps 📁 Restoration_status_summary: global RSS statistics with 9 indices describing network connectedness (Coherence) and restoration potential. Ref: GTB-RestorationPlanner.pdf 📁 CNTR: 259 national RSS statistics. 📁 Forest_map: The GFC2020 map at 100 m of resolution in World Cylindrical Equal Area projection (EPSG:54034) used as the primary input for all FAL computations. Source_data.zip 📁 country_map: GISCO global map of countries/territories (258 units). IDs, names, and codes are provided in accompanying CSV files. 📁 land-polygons-split-4326: Land polygon layers based on OpenStreetMap. Intermediate_data.zip 📁 Intermediate_data: Layers generated during the conversion of 355 GFC2020 tiles (~10 m resolution, geographic projection) to the final 100 m global forest map in Equal Area. 4. Technical specifications Coordinate Reference System: World Cylindrical Equal Area (EPSG:54034) Spatial Resolution: 100 meters Output Raster Data Type: uint8 (Byte) Maximum Extent (Bounding Box): Left: -20,037,500 Bottom: -6,729,000 Right: 20,037,500 Top: 8,585,000 Dataset Volume (Uncompressed): GFAD_results: ~12.4 GB Source_data: ~2.4 GB Intermediate_data: ~2.2 GB 5. Software & reproducibility The analysis was performed using 'gfad-analysis', an open-source Python pipeline designed to reproduce the entire dataset. GitLab Repository: https://code.europa.eu/jrc-forest/reproducibility/gfad-analysis 6. References Bourgoin C., Verhegghen A., Ameztoy I., Carboni S., Achard F., Colditz R. 2025. Global map of forest cover 2020 - version 3. European Commission, Joint Research Centre (JRC) [Dataset] http://data.europa.eu/89h/8c561543-31df-4e1b-9994-e529afecaf54 Bourgoin C., Verhegghen A., Carboni S., Ameztoy I., Degreve L., Fritz S., Herold M., Tsendbazar N., Lesiv M., Achard F., Colditz R., 2025. GFC2020: A Global Map of Forest Land Use for year 2020 to Support the EU Deforestation Regulation. Earth System Science Data 18(2), 1331–1365. https://doi.org/10.5194/essd-2025-351 Vogt P., Riitters K., Rambaud P., d'Annunzio R., Lindquist E., Pekkarinen A., 2022. GuidosToolbox Workbench: spatial analysis of raster maps for ecological applications. Ecography, 2022(3), e05864. https://doi.org/10.1111/ecog.05864 7. Contact & support Technical Contacts: Giovanni Caudullo - [e-mail] Peter Vogt - [e-mail] For further information, please visit the European Forest Observatory (EUFO) Web Portal: https://forest-observatory.ec.europa.eu/forest/gfa