Paper No. 8
Presentation Time: 10:30 AM
UNPREDICTED PERSISTENCE OF A URANIUM PLUME AT A FORMER MILL TAILINGS SITE IN RIFLE, COLORADO: EVIDENCE FOR MULTIPLE SOURCES OF URANIUM
LONG, Philip E.1, WILLIAMS, Kenneth H.
2, PEACOCK, Aaron D.
3, BARGAR, John R.
4, BUSH, Richard P.
5, DAVIS, James A.
2, FOX, Patricia M.
6, HATFIELD, Kirk
7, NEWMAN, Mark A.
7 and CAMPBELL, Kate M.
8, (1)Energy and Environment Directorate, Pacific Northwest National Laboratory, PO Box 999, Richland, WA 99354, (2)Earth Sciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Rd. MS-90-1116, Berkeley, CA 94720, (3)Microbial Insights, Rockford, TN 37853-3044, (4)Stanford Synchrotron Radiation Lightsource, 2575 Sand Hill Rd, Menlo Park, CA 94025, (5)U.S. Department of Energy, Legacy Management, Grand Junction, CO 81502, (6)Earth Sciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Rd, MS 74R316C, Berkeley, CA 94720, (7)Civil and Coastal Engineering, University of Florida, Gainsville, FL 32611, (8)Water Resources Division, U.S. Geological Survey, Boulder, CO 80303, philip.long@pnl.gov
The uranium plume at a former vanadium and uranium mill site in Rifle, Colorado was predicted to attenuate below MCL’s in 10 years based on a K
d model coupled to groundwater flow. Observations of U concentration in multiple wells since that prediction in 1998 show that uranium has decreased or remained constant at low levels in some places. However, the core of the plume has changed little over more than a decade and in one area, U concentrations have actually increased. We identify five possible “sources” of U to account for these observations, 1) Downward flux of U from the vadose zone that is entrained into the aquifer during times of high water table and high runoff in the Colorado River, 2) Mobilization of reduced U from naturally reduced zones (NRZ’s), 3) Slow oxidation of U(IV) dispersed in the saturated zone, 4) Influx of natural background U(VI) from upgradient groundwater, and 5) Dispersal of U from residual tailings remaining at the site.
We are currently quantifying the contribution of each of these sources of U and have focused on sources 1) and 2) above. Data collected in early summer 2011 during the highest and most sustained runoff in the Colorado River (adjacent to the site) since 1984 indicate that the vadose zone constitutes a relative minor contributor of U to the plume. In contrast, persistently high Fe(II) concentrations in alluvial aquifers at former uranium mill sites suggest that natural reduction may be a common feature of alluvial aquifers similar to the one at Rifle. One NRZ at Rifle contains up to 50 times more U than the typical alluvial sediment at the site. While speciation is likely to be an important control on the release of uranium from NRZ’s to groundwater, other factors may be equally important, including redox-poising constituents and permeability interfaces along boundaries of NRZ’s, which give rise to diffusion-limited reaction rates. In addition to U, NRZ’s contain elevated concentrations of NOM, Fe(II) and reduced inorganic sulfides, all of which can scavenge oxidants introduced following seasonal incursion of (sub)oxic groundwater, promoting retention of U(IV). Consideration of factors controlling the evolution of NRZ’s suggests that defining “biogeofacies” in contaminated aquifers may be useful in predicting contaminant mobility and persistence of redox-sensitive metals in groundwater.
