ABSTRACT
Natural river flooding events can mobilize contaminants from the vadose zone and lead to increased concentrations in groundwater. Characterizing the mass and transport mechanisms of contaminants released from the vadose zone to groundwater during these recharge events is particularly challenging. Therefore, conducting highly-controlled in-situ experiments that simulate natural flooding events can help increase the knowledge of where contaminants can be stored and how they can move between hydrological compartments. This study specifically targets uranium pollution, which is accompanied by high sulfate levels in the vadose zone and groundwater. Two novel experimental river flooding events were conducted that utilized added non-reactive halides (bromide and iodide) and 2,6-difluorobenzoate tracers. In both experiments, about 8 m3 of traced water from a nearby contaminant-poor river was flooded in a 3-m diameter basin and infiltrated through the vadose zone and into a contaminant-rich unconfined aquifer for an average of 10 days. The aquifer contained 13 temporary wells that were monitored for solute concentration for up to 40 days. The groundwater analysis was conducted for changes in contaminant mass using the Theissen polygon method and for transport mechanisms using temporal moments. The results indicated an increase in uranium (21 and 24%), and sulfate (24 and 25%) contaminant mass transport to groundwater from the vadose zone during both experiments. These findings confirmed that the vadose zone can store and release substantial amounts of contaminants to groundwater during flooding events. Additionally, contaminants were detected earlier than the added tracers, along with higher concentrations. These results suggested that contaminant-rich pore water in the vadose zone was transported ahead of the traced flood waters and into groundwater. During the first flooding event, elevated concentrations of contaminants were sustained, and that chloride behaved similarly. The findings implied that contaminant- and chloride-rich evaporites in the vadose zone were dissolved during the first flooding event. For the second flooding event, the data suggested that the contaminant-rich evaporites continued to dissolve whereas chloride-rich evaporites were previously flushed. Overall, these findings indicated that contaminant-rich pore water and evaporites in the vadose zone can play a significant role in contaminant transport during flooding events.
ABSTRACT
The recharge of stream water below the baseflow water table can mobilize groundwater contaminants, particularly redox-sensitive and sorptive metals such as uranium. However, in-situ tracer experiments that simulate the recharge of stream water to uranium-contaminated groundwater are lacking, thus limiting the understanding of the potential mechanisms that control the mobility of uranium at the field scale. In this study, a field tracer test was conducted by injecting 100 gal (379 l) of oxic river water into a nearby suboxic and uranium-contaminated aquifer. The traced river water was monitored for 18 days in the single injection well and in the twelve surrounding observation wells. Mobilization of uranium from the solid to the aqueous phase was not observed during the tracer test despite its pre-test presence being confirmed on the aquifer sediments from lab-based acid leaching. However, strong evidence of oxidative immobilization of iron and manganese was observed during the tracer test and suggested that immobile uranium was likely in its oxidized state as U(VI) on the aquifer sediments; these observations ruled out oxidation of U(IV) to U(VI) as a potential mobilization mechanism. Therefore, desorption of U(VI) appeared to be the predominant potential mobilization mechanism, yet it was clearly not solely dependent on concentration as evident when considering that uranium-poor river water (<0.015 mg/L) was recharged to uranium-rich groundwater (≈1 mg/L). It was possible that uranium desorption was limited by the relatively higher pH and lower alkalinity of the river water as compared to the groundwater; both factors favor immobilization. However, it was likely that the immobile uranium was associated with a mineral phase, as opposed to a sorbed phase, thus desorption may not have been possible. The results of this field tracer study successfully ruled out two common mobilization mechanisms of uranium: (1) oxidative dissolution and (2) concentration-dependent desorption and ruled in the importance of advection, dispersion, and the mineral phase of uranium.
