A 100 m annual soil salinization dataset from 1985 to 2024 in the Western Songnen Plain, China

Using machine learning algorithms, we developed a novel, high-resolution soil salinization dataset by integrating ground survey data, and remote sensing imagery to classify soil salinity over the period from 1985 to 2024. Also, we created a regional soil salinity prediction model which was further refined to account for soil moisture effects, enabling the generation of annual, 100 m resolution soil salinization maps.