This dataset integrates environmental factors related to avalanche formation and corresponding susceptibility results in southeastern Tibet. The dataset includes multi-source environmental variables such as topography, meteorology, snow conditions, and vegetation, all resampled to a spatial resolution of 30 m. Based on these factors, a Support Vector Machine (SVM) model was employed to generate monthly avalanche susceptibility maps for the 2021–2022 snow season (December to May). This dataset can support studies on avalanche formation mechanisms, susceptibility assessment, and regional hazard analysis.
Topographic factors were derived from ALOS DEM. Vegetation index (NDVI) was obtained from Landsat-8 imagery. Meteorological factors (precipitation, temperature, wind speed) were sourced from the TPMFD monthly dataset. Snow-related variables include MODIS snow cover products and long-term snow depth datasets in China.
(1) Multi-source data were reprojected and resampled to a spatial resolution of 30 m;
(2) Dynamic factors such as meteorological and snow variables were organized on a monthly basis;
(3) A comprehensive environmental factor system including topography, meteorology, snow, and vegetation was constructed;
(4) A Support Vector Machine (SVM) with RBF kernel was applied for susceptibility modeling;
(5) Historical avalanche hazard points were used as positive samples, and non-avalanche areas as negative samples, with a 7:3 split for training and validation to produce monthly susceptibility maps.
Model performance was evaluated using the AUC metric, with all monthly results exceeding 0.93, indicating high predictive accuracy. Environmental variables were standardized to a uniform spatial resolution, ensuring consistency and comparability across datasets.
| # | number | name | type |
| 1 | 42371085 | National Natural Science Foundation of China | |
| 2 | 2024YFC3013403 | National key R & D plan |
This work is licensed under
CC BY 4.0 (Creative Commons Attribution 4.0 International License).
CSTR:11738.11.NCDC.SNOW.DB7232.2026
10.12072/ncdc.snow.db7232.2026
