Spatial Analysis of Groundwater Hydrochemistry through Integrated Multivariate Analysis: A Case Study in the Urbanized Langat Basin, Malaysia

Abstract

Rapid urbanization and industrial development in the Langat Basin has disturbed the groundwater quality. The populations’ reliance on groundwater sources may induce possible risks to human health such as cancer and endocrine dysfunction. This study aims to determine the groundwater quality of an urbanized basin through 24 studied hydrochemical parameters from 45 groundwater samples obtained from 15 different sampling stations by employing integrated multivariate analysis. The abundance of the major ions was in the following order: bicarbonate (HCO3−) > chloride (Cl−) > sodium (Na+) > sulphate (SO42−) > calcium (Ca2+) > potassium (K+) > magnesium (Mg2+). Heavy metal dominance was in the following order: Fe > Mn > Zn > As > Hg > Pb > Ni > Cu > Cd > Se > Sr. Classification of the groundwater facies indicated that the studied groundwater belongs to the Na-Cl with saline water type and Na-HCO3 with mix water type characteristics. The saline water type characteristics are derived from agricultural activities, while the mixed water types occur from water–rock interaction. Multivariate analysis performance suggests that industrial, agricultural, and weathering activities have contributed to groundwater contamination. The study will help in the understanding of the groundwater quality issue and serve as a reference for other basins with similar characteristics.

Assessment of hydrogeochemistry in groundwater using water quality index model and indices approaches

“…The potential release of essential indicators into groundwater may occur as a consequence of the interplay between water, soils, and rocks, as well as the weathering of silicate and sodium-bearing minerals [ 105 ]. Moreover, the existence of said variables may be attributed to human activities that introduce them into the subterranean water, primarily through the utilization of agrochemicals in the agricultural fields surrounding the examined region [ 106 ] The PC2 factor, which solely comprised of two variables (TA and HCO 3 − ) accounted for 20% contribution towards the overall variance, indicating the occurrence of evaporation, weathering, and rock-water interaction [ 3 ]. The PC3, which is the third principal component, accounts for 17.5% of the variance with an eigenvalue of 2.97.…”

Aquifers and Groundwater: Challenges and Opportunities in Water Resource Management in Colombia

“…Some of the common techniques include drilling boreholes, observing water tables, using satellite imagery and Geographic Information Systems (GIS) [182,183], and geophysics, which uses methods such as electrical tomography and seismic surveying to map the geological structure of the subsurface [181,184,185]. Once an aquifer has been identified and located, a detailed analysis of the groundwater is performed to determine its quality and supply capacity; this involves the collection of samples and their analysis in the laboratory to evaluate the presence of pollutants such as heavy metals, nitrates, pesticides, or other chemicals that may affect the potability of the water, as well as physicochemical parameters such as pH and electrical conductivity, among others [186,187]. In addition, the monitoring of these aquifers is provided by a framework of meteorological, hydrological, and oceanographic networks that are a fundamental part of the observation, measurement, and surveillance of aquifers [184].…”

Use of Multivariate Statistical Analysis of Hydrochemical Data for the Identification of the Geochemical Processes in the Tirana-Fushe Kuqe Alluvial Aquifer, North-Western Albania

“…Principal component analysis (PCA) is a popular strategy for reducing dimensions of dataset, thus increasing the interpretability with minimal information loss. The way PCA operates is by generating new uncorrelated variables that enable the maximization of variance [Jolliffe and Cadima, 2016;Zainol et al, 2021;Nakagawa, 2021;Zhang et al, 2021]. Basically, the problem of defining the new dataset dimensions is reduced in a matrix decomposition, with eigenvalues/vectors interpretation described in many works [Beattie et al, 2021;Torokhti and Friedland, 2009;Gewers, 2021].…”