In the Zeroud basin, a diverse array of methodologies were employed to assess, simulate, and predict the quality of groundwater intended for irrigation. These methodologies included the irrigation water quality indices (IWQIs); intricate statistical analysis involving multiple variables, supported with GIS techniques; an artificial neural network (ANN) model; and an XGBoost regression model. Extensive physicochemical examinations were performed on groundwater samples to elucidate their compositional attributes. The results showed that the abundance order of ions was Na⁺ > Ca²⁺ > Mg²⁺ > K⁺ and SO₄²⁻ > HCO₃⁻ > Cl⁻. The groundwater facies reflected Ca-Mg-SO₄, Na-Cl, and mixed Ca-Mg-Cl/SO₄ water types. A cluster analysis (CA) and principal component analysis (PCA), along with ionic ratios, detected three different water characteristics. The mechanisms controlling water chemistry revealed water–rock interaction, dolomite dissolution, evaporation, and ion exchange. The assessment of groundwater quality for agriculture with respect IWQIs, such as the irrigation water quality index (IWQI), sodium adsorption ratio (SAR), sodium percentage (Na%), soluble sodium percentage (SSP), potential salinity (PS), and residual sodium carbonate (RSC), revealed that the domination of the water samples was valuable for agriculture. However, the IWQI and PS fell between high-to-severe restrictions and injurious-to-unsatisfactory. The ANN and XGBoost regression models showed robust results for predicting IWQIs. For example, ANN-HyC-9 emerged as the most precise forecasting framework according to its outcomes, as it showcased the most robust link between prime attributes and IWQI. The nine attributes of this model hold immense significance in IWQI prediction. The R² values for its training and testing data stood at 0.999 (RMSE = 0.375) and 0.823 (RMSE = 3.168), respectively. These findings indicate that XGB-HyC-3 emerged as the most accurate forecasting model, displaying a stronger connection between IWQI and its exceptional characteristics. When predicting IWQI, approximately three of the model’s attributes played a pivotal role. Notably, the model yielded R² values of 0.999 (RMSE = 0.001) and 0.913 (RMSE = 2.217) for the training and testing datasets, respectively. Overall, these results offer significant details for decision-makers in managing water quality and can support the long-term use of water resources.

This paper presents an integrated study of groundwater quality for irrigation and its spatial distribution in the Bou Arada-El Arroussa plain to guide decision-makers in the development of sustainable groundwater resource management plans. Assessing water quality for irrigation in this plain involves a process that begins with statistical analysis to determine similarities between samples and group them into clusters. This is followed by principal component analysis on physico-chemical data to determine relationships between variables and visualize their independence on a correlation matrix. Statistical analyses combined with a graphical approach through diagrams such as Piper, Gibbs, Wilcox and USSL are used to better assess the properties, facies and interactions between water and minerals within the aquifer. In addition, geographic information system (GIS) tools were used to better combine water quality indices and create a map of spatial quality distribution. Clustering results and principal component analysis reveal three types of water: the first type includes samples F04, F05 and P01, characterized by high concentration of TDS. The second type consists solely of well P23, distinguished by elevated level of K+, Mg2+ and NO3– ions. The third type can be divided into two subsets: one dominated by SO42– and the other by Cl–, with additional ions such as HCO3–, Ca2+ and Na+. Major ions analyses of the Bou Arada-El Arroussa groundwater show that the order of ions abundance was SO42–>Cl–>HCO3–>Ca2+>Mg2+>NO3–>K+. Piper diagrams and ion ratios revealed three water facies similar to the results obtained from the cluster analysis dendrogram: Ca-Mg-SO4, mixed and Na-Cl facies. The water hardness index (TH) classifies water samples into hard-brackish water category due to their high magnesium and calcium content, which is crucial for plant growth especially in agricultural area. The investigation of the relationship between water chemistry and mineralogical composition of the studied aquifer system reveals that all groundwater samples fall in the field corresponding to the water-rock interaction, with the exception of F04 and F05 showing a predominant evaporation. The Chadha diagram shows that the majority of samples are located in the reverse ion exchange zone, reflecting the exchange between groundwater and clay minerals. The USSL diagram shows that the majority of samples belong to the C3-S1 category, reflecting that the water is not suitable for irrigation due to the high salinity content. Evaluation of irrigation water quality using quality indices revealed that the majority of samples are within the standards for agricultural use. A multi-criteria analysis on the indices layers (SAR, SSP, PS, RSC, Na% and IWQI) created a synthesis map of the spatial distribution of the water quality. This map reveals that the western part of the Bou-Arada El Arroussa plain contains good-quality water, while the Eastern part contains the poorest quality related to the upwelling of poor-quality deep water.