Multiple environmental tracers combined with a constrained Bayesian isotope mixing model to elucidate nitrate and sulfate contamination in a coastal groundwater system.

Several groundwater quality investigations have been conducted in coastal regions that are commonly exposed to multiple anthropogenic stressors. Nonetheless, such studies remain challenging because they require focused-diagnostic approaches for a comprehensive understanding of groundwater contamination. Therefore, this study integrates a multi-tracer approach to acquire comprehensive information allowing for an improved understanding of the origins of groundwater contamination, the relative contribution of contaminants, and their biogeochemical cycling within a coastal groundwater system. This multi-tracer approach, focusing on nitrate (NO) and sulfate (SO) groundwater contamination, is applied to a Mediterranean coastal aquifer underlying an important economically strategic agricultural area. Dissolved NO in groundwater has concentrations up to 89 mg/L, whereas SO concentrations in groundwater are up to 458 mg/L. By integrating isotope tracers (i.e., δN, δO, δB, δS, and δO), NO and SO in the groundwater are found to have originated from multiple anthropogenic and natural sources including synthetic fertilizers, manure, sewage, atmospheric deposition, and marine evaporites. Chemical and isotopic data are coupled to identify the dominant hydro(geo)logic processes and the major subsurface biogeochemical reactions that govern the NO and SO occurrences. Nitrate and SO concentrations are identified to be respectively controlled by nitrification/denitrification and by bacterial dissimilatory SO reduction. Identifying these subsurface biogeochemical processes constrained the Bayesian isotope MixSIAR model, that is used for apportioning the relative contributions of the identified groundwater contamination sources, by informed site-specific isotopic fractionation effects. Results from MixSIAR indicate that manure is distinguished as the predominant source for NO (61 %), whereas SO in groundwater is mostly supplied from two sources (i.e., synthetic fertilizers and soil-derived sulfate) identified with similar contributions (30 %). This study particularly demonstrates the utility of initially describing the subsurface processes, not only to predict the fate of NO and SO concentrations within the groundwater system, but also to constrain the MixSIAR model with justified site-specific isotopic fractionation effects for subsurface transformation processes affecting NO and SO.