Predicting Long-term Dynamics of Soil Salinity and Sodicity on a Global Scale

This dataset globally (excluding frigid/polar zones) quantifies the variability in surface soil (0 – 30 cm) salinity and sodicity for the period between 1980 and 2018. This is realised by developing 4-D predictive models of Electrical Conductivity of saturated soil Extract (ECe) and soil Exchangeable Sodium Percentage (ESP) as indicators of soil salinity and sodicity. These machine learning-based models make predictions for ECe and ESP at different times, locations, and depths. The dataset includes 78 maps showing the annual predicted ECe and ESP between 1980 and 2018 on a global scale.Users are referred to the corresponding "READ_ME" file for more information about this dataset. Article10.1073/pnas.2013771117 also provides more information about the dataset.