Multivariate Analysis and Hydrogeochemical Evolution of Groundwater in a Geologically Controlled Aquifer System: A Case Study in North Central Province, Sri Lanka

This study investigates the coupled relationship between groundwater chemistry, lithology, and structural features in the dry zone of Netiyagama, Sri Lanka, within a fractured crystalline basement. Groundwater chemistry fundamentally reflects geological conditions determined by rock-water interactions, we hypothesized that the specific spatial patterns of groundwater chemistry in heterogeneous fractured systems are distinctly controlled by integrated effects of lithological variations, structurally driven flow pathways, aquifer stratification, and geochemical processes, including cation exchange and mineral-specific weathering. To test this, we integrated hydrogeochemical signatures with mapped hydrogeological data and applied multi-stage multivariate analyses, including Piper diagrams, Hierarchical Cluster Analysis (HCA), and Principal Component Analysis (PCA), and various bivariate plots. Piper diagrams identified five distinct hydrochemical facies, but these did not correlate directly with specific rock types, highlighting the limitations of traditional methods in heterogeneous settings. Employing a multi-stage multivariate analysis, we identified seven clusters (C1–C7) that exhibited unique spatial distributions across different rock types and provided a more refined classification of groundwater chemistries. These clusters align with a three-unit aquifer framework (shallow weathered zone, intermittent fracture zone at ~80–100 m MSL, and deeper persistent fractures) controlled by a regional syncline and lineaments. Further analysis through bivariate diagrams revealed insights into dominant weathering processes, cation-exchange mechanisms, and groundwater residence times across the identified clusters. Recharge-type clusters (C1, C2, C5) reflect plagioclase-dominated weathering and short flow paths; transitional clusters (C3, C7) show mixed sources and increasing exchange; evolved clusters (C4, C6) exhibit higher mineralization and longer residence. Overall, the integrated workflow (facies plots + PCA/HCA + bivariate/process diagrams) constrains aquifer dynamics, recharge pathways, and flow-path evolution without additional drilling, and provides practical guidance for well siting and treatment.