Predicting Irrigation Water Quality Indices Based on Data-Driven Algorithms: Case Study in Semiarid Environment

Ascertaining water quality for irrigational use by employing conventional methods is often time taking and expensive due to the determination of multiple parameters needed, especially in developing countries. Therefore, constructing precise adequate models may be beneficial resolving this problem agricultural management determine suitable classes optimal crop yield production. To achieve objective, five machine learning (ML) models, namely linear regression (LR), random subspace (RSS), additive (AR), reduced error pruning tree (REPTree), support vector (SVM), have been developed tested predicting six irrigation (IWQ) indices such as sodium adsorption ratio (SAR), percent (%Na), permeability index (PI), Kelly (KR), soluble percentage (SSP), magnesium hazards (MH) groundwater Nand Samand catchment Rajasthan. The accuracy these was determined serially using mean squared (MSE), correlation coefficients (r), absolute (MAE), root square (RMSE). SVM model showed best-fit all during testing, that is, RMSE: 0.0662, 4.0568, 3.0168, 0.1113, 3.7046, 5.1066; r: 0.9364, 0.9618, 0.9588, 0.9819, 0.9547, 0.8903; MSE: 0.004381, 16.45781, 9.101218, 0.012383, 13.72447, 26.078; MAE: 0.042, 3.1999, 2.3584, 0.0726, 2.9603, 4.0582 KR, MH, SSP, SAR, %Na, PI, respectively. KR SAR values were predicted accurately comparison observed values. As a result, algorithms can improve characteristics, which critical farmers various procedures. Additionally, findings research suggest ML are effective tools reliably general acquired directly on periodical basis. Assessment also help deriving strategies utilise inferior conjunctively with fresh resources water-limited areas.