Evaluation and Prediction of Groundwater Quality for Irrigation Using an Integrated Water Quality Indices, Machine Learning Models and GIS Approaches: A Representative Case Study

Agriculture has significantly aided in meeting the food needs of growing population. In addition, it boosted economic development irrigated regions. this study, an assessment groundwater (GW) quality for agricultural land was carried out El Kharga Oasis, Western Desert Egypt. Several irrigation water indices (IWQIs) and geographic information systems (GIS) were used modeling development. Two machine learning (ML) models (i.e., adaptive neuro-fuzzy inference system (ANFIS) support vector (SVM)) developed prediction eight IWQIs, including index (IWQI), sodium adsorption ratio (SAR), soluble percentage (SSP), potential salinity (PS), residual carbonate (RSC), Kelley (KI). The physicochemical parameters included T°, pH, EC, TDS, K+, Na+, Mg2+, Ca2+, Cl?, SO42?, HCO3?, CO32?, NO3?, they measured 140 GW wells. hydrochemical facies resources Ca-Mg-SO4, mixed Ca-Mg-Cl-SO4, Na-Cl, Ca-Mg-HCO3, Na-Ca-HCO3 types, which revealed silicate weathering, dissolution gypsum/calcite/dolomite/ halite, rock–water interactions, reverse ion exchange processes. IWQI, SAR, KI, PS showed that majority samples categorized purposes into no restriction (67.85%), excellent (100%), good (57.85%), to (65.71%), respectively. Moreover, selected as safe according SSP RSC. performance simulation evaluated based on several skills criteria, ANFIS model SVM capable simulating IWQIs with reasonable accuracy both training “determination coefficient (R2)” (R2 = 0.99 0.97) testing 0.97 0.76). presented models’ promising illustrates their use IWQI prediction. findings indicate ML methods geographically dispersed hydrogeochemical data, such SVM, be assessing irrigation. proposed methodological approach offers a useful tool identifying crucial components evolution mitigation measures related management arid semi-arid environments.