A dataset of summer surface water distribution in Qinghai Province based on multi-source data (2019–2023)

This study, based on the Google Earth Engine (GEE) cloud platform, utilizes Sentinel-1 data, Sentinel-2 data, and a digital elevation model combined with a threshold segmentation method to extract the summer surface water distribution data (including water bodies and glacial snow) in Qinghai Province from 2019 to 2023. Additionally, the surface water area for each year was calculated, providing five years of statistical information, including the area of water bodies and glacial snow (in tens of thousands of hectares) for each year. A confusion matrix was constructed by selecting sample points from historical high-resolution images to evaluate the accuracy of the dataset. The results indicate that the accuracy of the surface water extraction for Qinghai Province in the summers of 2019–2023 is relatively high, with an overall average accuracy of 0.90 and an average Kappa coefficient of 0.85. Due to incomplete data from Sentinel-2 (surface reflectance products), the time series for this dataset spans five years. For the issue of local pixel misclassification, a method based on local statistical information, including a slope-elevation mask and logical operations, was applied.The naming format for the raster data is YYYY_qh.tif, where YYYY represents the year and qh is the abbreviation for the study area. The dataset is stored in both TIFF and XLSX formats. It consists of 26 files, compressed into a 245 MB archive.