Using Multivariate Statistical Analysis, Geostatistical Techniques and Structural Equation Modeling to Identify Spatial Variability of Groundwater Quality

Multivariate statistical analysis, geostatistical techniques and structural equation modeling were used to determine the main factors and mechanisms controlling the spatial variation of groundwater quality in the Ain Azel plain, Algeria. Cluster analysis grouped the sampling wells into two statistically significant clusters based on similarities of groundwater quality characteristics. Principal component and factor analyses (PCA/ FA) revealed that two factors explained around 85 % of the total variance, which water-rock interaction and anthropogenic impact as the dominant factors affecting the groundwater quality. The distribution of factor score one represents high loading for EC, Ca, Mg, Na, K, and SO₄ in the western side and south eastern side of the plain, where water-rock interactions are dominate factors influence groundwater quality. Spatial distribution map of factor score 2 indicate that NO₃, NO₂, NH₄, and COD show high concentration in central and southern side of the plain, where anthropogenic impact reduce groundwater quality. Further, one-way analysis of variance (one-way ANOVA) showed that the mean differences between cluster one and two show significantly differences for some water quality parameters including EC, Ca, Mg, Na, K, Cl, and SO₄. Structural equation modeling (SEM) confirmed the finding of multivariate analysis. This study provides a new technique of confirming exploratory data analysis using SEM in groundwater quality.