Prediction of irrigation water quality indices based on machine learning and regression models

Abstract Assessing irrigation water quality is one of the most critical challenges in improving resource management strategies. The objective this work was to predict index Bahr El-Baqr, Egypt, based on non-expensive approaches that requires simple parameters. To achieve goal, three artificial intelligence (AI) models (Support vector machine, SVM; extreme gradient boosting, XGB; Random Forest, RF) and four multiple regression (Stepwise Regression, SW; Principal Components PCR; Partial least squares regression, PLS; Ordinary OLS) were applied validated for predicting six criteria (soluble sodium percentage, SSP; adsorption ratio, SAR; residual carbonate, RSC; potential salinity, PS; permeability index, PI; Kelly’s KR). Electrical conductivity (EC), (Na + ), calcium (Ca 2+ ) bicarbonate (HCO 3− used as input exploratory variables models. results indicated source not suitable without treatment. A good soil drainage system salinity control measures are required avoid salt accumulation within soil. Based performance statistics root mean square error (RMSE) scatter (SI), SW emerged best (0.21% 0.03%) followed by PCR PLS with RMSE 0.22% 0.21% SAR, respectively. classification SI, all having values less than 0.1 indicate prediction indices except RSC. These highlight using regressions developed machine learning methods quality, can be rapid decision tools modelling quality.