Spatial-multivariate statistical analyses to assess water quality for irrigation of the central part of Kuwait

Multivariate statistical analysis and ArcGIS tools have been used to assess groundwater quality in the central part of Kuwait. The chemical data for groundwater samples collected from 71 bore wells are assessed to determine their suitability for drinking and irrigation purposes using the irrigation parameter indices. The WATEQ4F program is used to investigate the degree of saturation of groundwater with respect to the minerals calcite, dolomite, aragonite, halite, gypsum, magnesite, and brucite. In multivariate statistical analysis, principal component analysis (PCA) and R-mode and Q-mode cluster analysis are used to classify the groundwater into different clusters. In the PCA, the cumulative variance percentages of the first three components are 89.5%, expressed for four main chemical compounds, of which sodium chloride and sodium sulphate are the first principal component (PC1)compounds, with a variance of 65%, followed by the carbonates and silicates in PC2 and PC3, which are 15% and 9%, respectively. The Q-mode cluster mean value analysis identified four distinct groundwater chemical types, in which cluster I is mainly NaCl and SiO₂, followed by Na₂SO₄ in cluster II, and the concentration of HCO₃⁻ revealed that clusters III and IV were mainly carbonate groups. In addition, the TDS and the TH clusters' mean values decrease from cluster I to cluster IV, indicating that the quality of the groundwater improves sequentially from cluster I to cluster IV. The results reveal that the groundwater in the study area is not suitable for drinking, but it is suitable for irrigation purposes.