Comparative Evaluation of Machine Learning Algorithms for Predicting Soil Wetting Front Dynamics Under Drip Irrigation System

Prediction of Soil Moisture Constants in Coastal Soils of Bangladesh Using Machine Learning Algorithms

The prediction of soil moisture constants (SMCs), including field capacity (FC), permanent wilting point (PWP) and available water content (AWC), is essential for efficient irrigation scheduling and water management in the coastal regions of Bangladesh affected by salinity, flooding and waterlogging. However, the comparative evaluation of machine learning (ML) models for predicting SMCs in these complex coastal soils remains limited. Therefore, this study evaluated five ML models to predict SMCs using soil physicochemical parameters (pH; electrical conductivity; organic matter; bulk density; and sand, silt and clay fractions). The observed ranges were 6.5%–38.4%, 1.3%–15.5% and 5.1%–26.0% for FC, PWP and AWC, respectively. Among the applied models, the multilayer perceptron (MLP) achieved the best predictive performance, with R 2 values of 0.926, 0.911 and 0.906 for FC, PWP and AWC, respectively, whereas MLR showed the weakest performance, with corresponding R 2 values of 0.804, 0.761 and 0.738. Compared with the other models, MLP improved R 2 by 3%–23% and reduced RMSE by 13%–40% across FC, PWP and AWC. Using input combination‐5 (IC‐5), the MLP explained more than 89% of the variability and achieved up to 98% prediction accuracy, highlighting its strong potential for predicting SMCs in coastal soils to support agricultural productivity.

Modelling soil water content in different tillage systems and soil types using machine learning

Soil water availability is one of the major challenges in many rainfed crop production systems of the Global South. Soil water conservation practices are being promoted to enhance climate change adaptation for rainfed cropping systems of southern Africa. However, the cost and time required to develop and test appropriate modelling and simulation tools can be enormous. The objectives of this study were to: (i) test the performance of the decision tree, adaptive boosting (AdaBoost), support vector machine, neural network, stochastic gradient descent, k-nearest neighbours, random forest and linear regression machine learning models in predicting soil water under different tillage practices, soil types and depths, and (ii) assess the soil water classification and prediction capabilities of 8 models under different tillage practices, soil types and depths. The neural network, random forest and decision tree models had the best soil water prediction capabilities. The neural network, random forest and decision tree models were the best algorithms (RMSE = 15.801–16.369; MAE = 11.997–12.315; R2 = 0.822–0.835) for predicting and classifying soil water from different soil types and depth intervals. The support vector machine learning model was the weakest algorithm (RMSE = 36.177; MAE = 30.84; R2 = 0.133) for predicting and classifying soil water. All the algorithms poorly predicted and classified soil water based on tillage practices. All the models closely predicted soil water at 300 and 900 mm depths but poorly predicted soil water at 600 mm depth intervals. Based on this study, the neural network model is the best machine learning tool for predicting soil water in clay and sandy soils under semi-arid agroecological conditions.

Machine Learning Prediction of Soil‐Water Characteristic Curve and Its Application in Lateral Earth Pressure Calculation of Retaining Walls

The soil‐water characteristic curve (SWCC) is a fundamental parameter that governs the hydro‐mechanical behavior of unsaturated soils. Conventional laboratory measurement of SWCC is time‐consuming and labor‐intensive, while traditional lateral earth pressure design for retaining walls frequently relies on the saturated soil assumption, neglecting the effects of SWCC and resulting in significant systematic deviations in calculations. This study develops a statistically rigorous machine learning (ML) framework for efficient SWCC prediction and its application to lateral earth pressure calculations for pile‐supported box counterfort retaining walls. Four ML algorithms with distinct methodological frameworks were employed: extreme learning machine (ELM), least squares support vector machine (LSSVM), projection pursuit regression (PPR), and Bayesian ridge regression (BRR). These algorithms were utilized to construct SWCC prediction models using the cleaned UNSODA database. Model performance was assessed through multi‐metric evaluation, paired t ‐tests for statistical significance, and robustness analysis involving 30 independent runs, with validation conducted on measured silty clay data across 12 suction levels. Results indicate that the ELM model achieves the highest prediction accuracy, demonstrating statistically significant superiority over LSSVM, PPR, and BRR and excellent robustness. Independent validation reveals an average relative error of only 2.25% for ELM‐predicted SWCC. The SWCC‐based earth pressure calculation rectifies the bidirectional deviations of the traditional saturated method and identifies a neutral point at a depth of 17.5 m for a 25 m‐high retaining wall. This study offers a reliable technical approach for rapid SWCC acquisition and refined lateral earth pressure design for retaining structures.

Enhancing forest soil water retention curve prediction accuracy and interpretability through CatBoost, artificial ecosystem optimization, and SHAP analysis

Machine Learning-Driven Prediction and Interactive Nonlinear Analysis of Compaction Parameters for Fine-Grained Soils

Compaction parameters of soil material, maximum dry density (MDD) and optimum moisture content (OMC), are critical control indicators for highway embankment construction. In this study, a dataset containing 199 compaction test results for fine-grained soils was collected. Using MDD and OMC as prediction targets, Random Forest (RF), Gradient Boosting (GB), Extreme Gradient Boosting (XGBoost), and Support Vector Regression (SVR) were developed and optimized using the Wild Horse Optimization (WHO) algorithm. Five universal evaluation metrics, combined with radar charts, were used to comprehensively compare the predictive performance of each model. Based on univariate sensitivity analysis and SHapley Additive exPlanations (SHAP) global feature interpretation, two-way partial dependence plots (PDPs) were applied to reveal the pairwise nonlinear relationships between soil physical indices and compaction indicators. The results demonstrate that WHO-tuned XGBoost achieves optimal comprehensive predictive performance for both MDD (test set R2 = 0.8226) and OMC (test set R2 = 0.7486), outperforming SVR, GB, and RF in terms of fitting accuracy and generalization under small-sample conditions. Plastic limit (PL) exerts a significant influence on compaction performance. By further comparing WHO, Particle Swarm Optimization (PSO), Gray Wolf Optimizer (GWO), Random Search, Bayesian Optimization and Grid Search on the optimal model, the applicability and feasibility of WHO-XGBoost in predicting compacted-soil compaction parameters were validated. The proposed data-driven compaction evaluation framework (WHO-XGBoost-SA-SHAP-PDPs) acts as an auxiliary tool to lower the workload and cost of laboratory Proctor tests, offering theoretical support and technical guidance for rapid refined embankment compaction control in green transportation infrastructure.

A Stepwise Correction Model for Operational Forecasting of Surface Soil Moisture During the Spring Sowing Period

Accurately predicting soil moisture conditions and determining the optimal sowing timing during the spring sowing period play a crucial role in ensuring high and stable grain yields. To address the limitations of existing soil moisture prediction models, i.e., the complexity and parameterization challenges of hydrological models, the weak theoretical foundations of statistical models, and the data dependence, lack of interpretability, and poor generalization of machine learning approaches, a prediction model for surface soil water content (SWC) was developed in this study. The model is based on the water balance principle and uses a stepwise correction approach with normalized functions of key influencing factors. The results are as follows: (1) The model requires readily available parameters from public databases and is driven by daily scale meteorological variables (precipitation, temperature, wind speed, and vapor pressure deficit), facilitating its integration into existing operational weather forecasts. (2) After parameterization, only the moisture exchange between surface and deep soil layers needs optimization, resulting in low computational demand. (3) A trial in Shenyang region showed that the model explains 82.1% of the SWC variance, with an RMSE of 1.7% for 1–7 day lead predictions. (4) When applied to regions without initial soil moisture observations, the model achieves satisfactory accuracy after an initial condition sensitivity period of approximately 40 days. These results provide a methodological reference for soil moisture prediction studies and offer technical support for meteorological services to integrate soil moisture forecasting into their operational frameworks.