Towards robust smart data-driven soil erodibility index prediction under different scenarios.

Soil erosion is a major cause of damage to agricultural lands in many parts of the world and is of particular concern in semiarid parts of Iran. We use five machine learning techniques—Random Forest (RF), M5P, Reduced Error Pruning Tree (REPTree), Gaussian Processes (GP), and Pace Regression (PR)—under two scenarios to predict soil erodibility in the Dehgolan region, Kurdistan Province, Iran. Our models are based on a variety of soil properties, including soil texture, structure, permeability, bulk density, aggregates, organic matter, and chemical constituents. We checked the validity of the models with statistical metrics, including the coefficient of determination (R2 ), mean absolute error (MAE), root mean squared error (RMSE), T-tests, Taylor diagrams, and box plots. All five algorithms show a positive correlation between the soil erodibility factor (K) and silt, sand, fine sand, bulk density, and infiltration. The GP model has the highest prediction accuracy (R² ¼ 0.843, MAE ¼ 0.0044, RMSE ¼ 0.0050). It outperformed the RF (R² ¼ 0.812, MAE ¼ 0.0050, RMSE ¼ 0.0061), PR, (R² ¼ 0.794, MAE ¼ 0.0037, RMSE ¼ 0.0052), M5P (R² ¼ 0.781, MAE ¼ 0.0043, RMSE ¼ 0.0053), and REPTree (R² ¼ 0.752, MAE ¼ 0.0045, RMSE ¼ 0.0056) algorithms and thus is a useful complement to studies aimed at predicting soil erodibility in areas with similar climate and soil characteristics. [ABSTRACT FROM AUTHOR]