Evaluation of Groundwater Quality for Irrigation in Deep Aquifers Using Multiple Graphical and Indexing Approaches Supported with Machine Learning Models and GIS Techniques, Souf Valley, Algeria

Irrigation has made a significant contribution to supporting the population’s expanding food demands, as well promoting economic growth in irrigated regions. The current investigation was carried out order estimate quality of groundwater for agricultural viability Algerian Desert using various water indices and geographic information systems (GIS). In addition, support vector machine regression (SVMR) applied forecast eight irrigation (IWQIs), such index (IWQI), sodium adsorption ratio (SAR), percentage (Na%), soluble (SSP), potential salinity (PS), Kelly (KI), permeability (PI), residual carbonate (RSC). Several physicochemical variables, temperature (T°), hydrogen ion concentration (pH), total dissolved solids (TDS), electrical conductivity (EC), K+, Na2+, Mg2+, Ca2+, Cl−, SO42−, HCO3−, CO32−, NO3−, were measured from 45 deep wells. hydrochemical facies resources Ca–Mg–Cl/SO4 Na–Cl−, which revealed evaporation, reverse exchange, rock–water interaction processes. IWQI, Na%, SAR, SSP, KI, PS, PI, RSC showed mean values 50.78, 43.07, 4.85, 41.78, 0.74, 29.60, 45.65, −20.44, respectively. For instance, IWQI obtained results indicated that samples categorized into high restriction moderate purposes, can only be used plants are highly salt tolerant. SVMR model produced robust estimates IWQIs calibration (Cal.), with R2 varying between 0.90 0.97. Furthermore, validation (Val.), 0.88 0.95 achieved model, reliable IWQIs. These findings feasibility models evaluation management complex terminal aquifers irrigation. Finally, combination IWQIs, SVMR, GIS effective an applicable technique interpreting forecasting both arid semi-arid