River water intrusion and uranium capture from the vadose zone near the Columbia River at the Hanford Site, Washington

We investigated the effects of river water intrusion into the 300 Area Interdisciplinary Field Research Challenge (IFRC) site, approximately 200 m west of the Columbia River. The IFRC consists of 36 wells in a triangular array, pointing to the east, with wells on 10 m spacing. The site experiences seasonal changes in water table elevation of 2 m due to the influence of the river during the increase in river stage at spring snow melt. Shorter-term (daily to weekly) fluctuations result from river-stage management for power generation at upstream dams. The IFRC wells were screened over the uppermost 3 m of the aquifer, and were sampled daily by pumps central to the screened interval, from May 12 to July 30. Samples were analyzed for anion, cation, carbon, and uranium concentrations, and the elevation of the aquifer was measured across the site. River water arrival was determined by a negative inflection in chloride concentrations, and occurred 6 days after significant coupled river water and water table rises. The influx of river water progressed to a maximum after 18 days, reaching a maximum on June 29: river water comprised a maximum of 75% of the groundwater at the eastern edge of the IFRC, with a gradient in concentration across the 60 m-wide site down to 0% in the west. Tracer solutions were introduced just prior to the river water influx, and showed a rapid movement of water off the site to the west during the influx, against the regional hydraulic gradient, and returning to the western edge of the site as the river water retreated, approximately 25 m south of the point of injection. Uranium concentrations were uniform at approximately 30-50 μg/L before river water intrusion. As the water table rose, the uranium concentration increased within 7 days to 330 μg/L at the south corner of the site. Uranium was contributed heterogeneously: none was contributed at the east corner, and uranium concentration increased to 160 μg/L at the north corner only during the last week of the water table rise. Also, concentrations of dissolved ions and uranium in 6 wells completed deeper in the aquifer were constant; river water intruded only in the upper aquifer, and uranium originated from the vadose zone. These results showed that the hyporheic zone, if defined as the zone over which river water intruded and mixed with groundwater, extended 200 m from the river shore. The vadose zone, including the interval between low and high water table elevations (the smear zone) was known from bicarbonate extractions of drill cuttings to contain contaminant uranium, while the sediment within the saturated zone was essentially uranium free. Contaminant uranium was therefore contributed to the aquifer by desorption as the aquifer extended into the vadose zone. This source of contaminant uranium could sustain a uranium plume in the aquifer through exposure to aquifer groundwater, maintaining an only ephemerally accessed source of uranium. This work was supported by the US Department of Energy Office of Biological and Energy Research.