2026-10-14T10:42:21Z https://www.scidb.cn/oai

10.57760/sciencedb.37659 2026-05-21T13:59:29Z

2026-05-21 0.8 DEM and Multispectral Images for Retrogressive Thaw Slump Susceptibility Mapping doi:10.57760/sciencedb.37659 en The dataset provides high spatial resolution raster data for RTS susceptibility analysis in four representative permafrost regions of the central Tibetan Plateau: Beiluhe, Chumarhe, Wulanwula, and Gaolinggou. Each study area covers approximately 28,224 ha and contains 21,000 × 21,000 pixels at 0.8 m spatial resolution. The Beiluhe, Chumarhe, and Wulanwula scenes were acquired on 12 December 2022, and the Gaolinggou scene was acquired on 26 November 2023. The dataset includes pan-sharpened GaoFen-7 multispectral images and corresponding 0.8 m digital elevation models in GeoTIFF format.The multispectral images were produced from GaoFen-7 optical observations through orthorectification, geometric correction, radiometric calibration, atmospheric correction, and panchromatic sharpening. The DEMs were derived from the forward-looking and backward-looking GaoFen-7 panchromatic stereo images. The stereo workflow included orbit correction, orthorectification, tie-point matching, epipolar image generation, dense stereo matching, disparity-to-elevation conversion, interpolation to a regular grid, and quality checking to reduce mismatches, abnormal elevation values, and local artifacts. RTS sample polygons were manually delineated from the Beiluhe scene and cross-checked by experienced interpreters; binary ground-truth rasters were then generated at the same 0.8 m resolution.No systematic missing data are expected in the submitted raster files. Possible uncertainties mainly arise from stereo matching in weak-texture terrain, residual geometric misregistration, radiometric correction, and manual interpretation of RTS boundaries. All files are standard GeoTIFF rasters and can be opened in QGIS, ArcGIS, ENVI, GDAL, or Python-based geospatial libraries. Susceptibility Mapping; Retrogressive Thaw Slump; Tibetan Plateau; Deep Learning; Digital Elevation Model; Permafrost Degradation Yi Yuan Guiyun Zhou PUBLIC https://creativecommons.org/licenses/by-sa/4.0/ dataset Science Data Bank