%0 Dataset
%T Long-term daily snow depth dataset for the Three-River Source region from 1980 to 2020
%J National Cryosphere Desert Data Center
%I National Cryosphere Desert Data Center(www.ncdc.ac.cn)
%U http://www.ncdc.ac.cn/portal/metadata/8095027c-03a7-4c6c-9a46-21e024e45374
%W NCDC
%R 10.12072/ncdc.nieer-snow.db6816.2025
%A Zhao Zisheng
%A Xiaohua Hao 
%A Li Hangxuan
%A Zhong Xinyue
%A Wu Xiaodong
%K Deep learning;passive microwave;snow depth;long time series
%X High spatial resolution snow depth is essential for hydrological, ecological and disaster studies. However, passive microwave snow depth products (10/25 km) are no longer able to meet the modern high-precision and high-resolution requirements due to their coarse spatial resolution. In this study, the latest calibrated enhanced-resolution brightness-temperature-to-optical snow area ratio and the number of snow-covered days were integrated to invert the daily snow depth data at 5 km spatial resolution during the snowy period (October to April) at the Three-River Source region based on the deep learning FT-Transformer model. Compared with the long time series snow depth data (25 km) in China, the inverted 5 km snow depth has better accuracy, and the RMSE is usually below 8.5 cm. It provides a reliable data base for monitoring snow resources in the Three-River Source region.