Predicting long-term dynamics of soil salinity and sodicity on a global scale

Significance Land degradation due to soil salinization has detrimental impacts on vegetation, crops, and human livelihoods, leading to a need for a methodologically consistent analysis of the variability of different aspects of salt-affected soils. However, previous studies on the soil salinity issue have been primarily spatial and localized, leaving the large-scale spatiotemporal variations of soil salinity widely ignored. To address this gap, we present a globally validated analysis quantifying the long-term variations (40 y) of topsoil salinity at high spatial resolutions using machine-learning techniques. The results have significant implications for agroecological modelling, land assessment, crop growth simulation, and sustainable water management.