Data and code for:Desertification mapping in the Three-North Shelterbelt Project Region of China based on the integration of Sentinel-1/2 Imagery

As a major ecological and environmental issue facing arid and semi-arid regions, the occurrence and evolution of desertification profoundly impact regional ecological security and land-use patterns. Conducting desertification monitoring based on remote sensing data is of great significance for providing objective ecological assessments and a scientific basis for ecological governance. To address the limitations of existing studies, which largely rely on a single spectral dataset, this study combines Sentinel-1 and Sentinel-2 remote sensing data to develop a desertification monitoring index (TDMI) that integrates vegetation condition (modified soil vegetation index MSAVI), albedo, and surface texture (VV backscatter characteristics) by using the spatial distance model (SDM). Subsequently, a Gaussian mixture model (GMM) was employed to conduct monitoring of desertification in the Three-North Shelterbelt Forest Project area for the years 2020, 2022, and 2024. Combining field survey data with accuracy validation using high-resolution historical imagery from Google Earth, the results indicate that the TDMI demonstrated good performance across all years, with an average Overall Accuracy (OA) above 83%. This represents an improvement of 7.34% and 11.11% compared to with the two-dimensional Desertification Index (DMI) and vegetation cover (FVC), respectively. In terms of spatial distribution characteristics, TDMI demonstrates stronger identification capabilities for the wind-sand desertification types in the study area and can more effectively characterize the spatial heterogeneity of desertification severity. Generally, this study demonstrates that the synergistic application of multi-source remote sensing data can effectively enhance the accuracy and stability of desertification monitoring.