Evaluating spatiotemporal groundwater quality changes in the Rhodope coastal aquifer system (NE Greece) employing a GIS-assisted hybrid approach of multivariate statistics and inverse geochemical modeling.

This research employs a GIS-assisted approach of multivariate statistics and inverse geochemical modeling to unravel the processes driving groundwater salinization in a complex aquifer system. Multivariate statistical methods define the end-member water groups, identifying dominant processes explaining hydrogeochemical variance in wet and dry season water chemistry datasets. Mineral saturation indices (SIs) and inverse geochemical modeling (IGM) investigate potential geochemical reactions and mixing processes responsible for the observed groundwater compositions and their spatiotemporal evolution along reversed flow paths caused by overexploitation in the Rhodope aquifer system. Results reveal that a concise set of reactant and product phases, including CO, HO, calcite, gypsum, halite, celestite, plagioclase, K-feldspar, illite, and Ca-montmorillonite, along with ion exchange processes (CaX, MgX, and NaX), explains the hydrogeochemical evolution of groundwater along reversed flow paths between genetically and compositionally different surface and groundwater bodies. Systematic changes in water chemistry along the flow paths are attributed to mixing of surface waters and/or different groundwater end-members, dilution by a freshwater component, water-rock interaction (WRI) processes, and ion exchange involving Ca/Mg- and/or Na-clays. The chemical evolution represented by IGMs initiates with the mixing of Aegean seawater and Aspropotamos River, incorporating WRI and ion exchange processes (Mg- and Na-clays) to produce the water chemistry of Vistonida Lake, the only surface water body with hydraulic interaction with the groundwater system in the study area. Statistically-defined end-member water groups effectively explain the groundwater flow system and evolutionary processes between hydraulically connected surface and groundwater bodies. Overall, the fusion of multivariate statistical analysis (MVSA), inverse geochemical modeling (IGM), and GIS techniques proves potent and comprehensive, enhancing understanding of groundwater dynamics, improving prediction accuracy, aiding proficient management, and facilitating data-driven decision-making within the realm of groundwater assessment and management.