2026-10-12T13:21:07Z https://www.scidb.cn/oai

10.57760/sciencedb.30651 2026-03-18T17:04:19Z

2026-03-18 Forest Degradation Risk Assessment Dataset of Afforestation Areas in the "Two Rivers and Four Tributaries" Basin doi:10.57760/sciencedb.30651 en This dataset provides a forest degradation risk assessment for afforestation areas in the "Two Rivers and Four Tributaries" (Liangjiang Sihe) Basin, Tibet Autonomous Region, covering the period 2000–2024, aiming to identify the spatial pattern of degradation risk and locate high-risk hotspots.The dataset was built upon the Hansen Global Forest Change v1.12 dataset. Areas with canopy cover exceeding 20% in the treecover2000 band served as the baseline forest mask, and pixels experiencing forest loss during 2000–2024 (identified via the lossyear band) were labeled as "degraded." Driving factors included elevation and slope (SRTM DEM), mean annual temperature and precipitation (TerraClimate), and the global Human Modification index (gHM), all reprojected to Albers Equal-Area projection and resampled to 100 m resolution. A stratified random sampling strategy with a 1:1 positive-to-negative ratio was applied on Google Earth Engine. A Multiscale Geographically Weighted Regression (MGWR) model with an adaptive bisquare kernel and corrected Akaike Information Criterion (AICc) bandwidth selection was used, with all variables Z-Score standardized. Discrete predictions were spatially interpolated via Inverse Distance Weighting (IDW) to produce a continuous risk raster.The dataset is in GeoTIFF format (.tif) at 100 m spatial resolution. Pixel values represent degradation risk predictions: values below 0.3 indicate low risk, 0.3–0.7 medium risk, and above 0.7 high risk. Results show that high-risk areas account for approximately 43.32% (mainly in Nyingchi), medium-risk areas 36.47%, and low-risk areas 20.21%. The data can be accessed using ArcGIS, QGIS, or Python (rasterio, GDAL). Forest degradation risk; Two Rivers and Four Tributaries Basin; Afforestation; MGWR Fu Yingchun RESTRICTED dataset Science Data Bank