Multivariate Analysis and Origin of Groundwater Quality in Arid Region: Maadher of Hodna (Algeria)

Climate change and overexploitation of aquifers by irrigated agriculture and water consumption by the population caused the decline of water levels and hydrochemical changes in the aquifer. Our study area is the Maadher plain which is located in the north-central of Algeria, with an arid to semi-arid climate. The Mio-Plio-Quaternary aquifer often shows disordered alternations of sedimentary deposits. The aquifer is recharged mainly by wadis and drainage contributions from the Cretaceous and Miocene layers. The sampling campaign in 2019 based on 13 physicochemical parameters was carried out on water from 32 boreholes in the study area, compared to data archives of both sampling campaign in 1996. Multivariate statistical analyses were used to identify and know the groundwater origin from analysis results. Two statistical methods, principal component analysis (PCA), and hierarchical cluster analysis (HCA) are applied in order to define the major control factors that affect the hydrochemistry of Maadher plain. The results of groundwater samples showed that water is mainly hard. The order of abundance of major cations exchangeability for common ions in groundwater was: Ca²⁺ > Mg²⁺ > Na⁺ > K⁺ while the major anions reveal an order as follows SO₄²⁻ > HCO⁻₃ > Cl⁻ > NO₃⁻. The Piper diagram indicates that the hydrochemical facies of sulfate–chloride–nitrate–calcium (SO₄⁻²–Cl⁻–NO₃⁻–Ca₂⁺ type water), which globally characterizes the study area, and these elements are the dominant dissolved ions. It was noted that the tendency of such groundwater is toward salinization because the minimum values of total hardness exceed the WHO standard, which has led to a water quality between average and poor. A total of 62% of the groundwater samples exceeded the Ca²⁺ standard, nearly 65% of the samples exceeded the Cl⁻ standard, and also the sulfate concentrations were all above health guidelines. The electrical conductivity values are considerable, and 22% of the groundwater samples exceed the standard. PCA analysis showed that the first factors accumulated a percent variance of 74.22%. The first factor accounts for 40.45% variance, represented by a large number of mineralization parameters. The HCA analysis with Q-mode provided insight into the mineralization of aquifers (Mio-Plio-Quaternary), so that the main study area can be divided into three zones of different groundwater quality, with a total of three factors of control.