Combinations of multivariate statistical analysis and analytical hierarchical process for indexing surface water quality under arid conditions

Abstract

Novel methods for water quality indexing increase insight into the fitness of water bodies for different uses. We hypothesized that integrating multivariate statistical analysis (MSA) with the analytical hierarchical process (AHP) may provide a reliable estimation of water quality status. Hence, twenty water samples from canals and drains in the northern Nile Delta, Egypt were collected during summer, autumn, winter, and spring and analyzed. Data were subjected to MSA, including correlation analysis, principal component analysis (PCA), and hierarchal cluster analysis (HCA). The AHP was applied to derive weights of parameters implied in developing water quality indices for irrigation (IWQI) and fish farming (FFWQI). Human health risks due to exposure to potentially toxic elements (PTEs) via dermal contact were also considered. The average concentrations of water constituents were acceptable for irrigation, except sodium adsorption ratio (SAR) and Cl−. The dissolved oxygen, total dissolved solids, Cl−, NO2-N, NO3-N, NH3, and PTEs (except Zn) did not meet standard limits for fish production. The MSA revealed that water contamination resulted from human activities (agriculture, industry, and domestic wastes) and hydrochemical processes. The PCA indicated that SAR, Cu, and pH could adequately represent water quality for irrigation, while temperature, NO2-N, Cr, and Zn could reflect fish farming requirements. The AHP provided consistent weights for the original and shortlisted parameters. The water quality varied from good to poor for irrigation and from excellent to low for fish farming. The minimum IWQI could adequately represent the IWQI (R2 = 0.83) and thus reduce the time, effort, and cost for monitoring water quality. However, the minimum FFWQI showed moderate consistency (R2 = 0.51) with FFWQI, implying that increasing the sampling size is essential for better performance. The hazard quotient of all PTEs was below 1.0 for both adults and children, indicating a safe limit. The potential cancer risk was acceptable (1.36E-06) for adults and safe (8.03E-07) for children. Results of this work would be a start point for efficient quality control programs in arid regions.