HR: 11:45h
AN: H22A-06    [Abstracts]
TI: Large-Scale Aquifer Test at the Bemidji, Minnesota, Oil-Spill Site: Implications for Modeling Multiphase Flow, Natural Attenuation, and LNAPL Remediation
AU: * Herkelrath, W N
EM: wnherkel@usgs.gov
AF: U.S. Geological Survey, 345 Middlefield Road, MS 496, Menlo Park, CA 94025-3561 United States
AU: Delin, G N
EM: delin@usgs.gov
AF: U.S. Geological Survey, 2280 Woodale Drive, Mounds View, MN 55112-0049 United States
AB: A large-scale aquifer test was carried out at a crude oil spill site near Bemidji, Minnesota. The spill occurred in 1979 when a pipeline ruptured, spreading oil over a large area and creating three subsurface "pools" of high oil saturation near the water table. USGS scientists, in cooperation with researchers from several universities, have investigated the fate and transport of separate phase oil and hydrocarbons dissolved in ground water at this site since 1983. The primary goal of the aquifer test was to estimate parameters used in modeling processes such as subsurface flow of oil and water as well as natural attenuation of dissolved hydrocarbons in the plume. A secondary goal was to evaluate the effects of the oil on the parameters. Our aquifer test was carried out in July 2005 beneath the "north" oil pool, which occupies a 20x100 meter footprint. Prior to the test, the water table was about 6 meters below land surface, and the oil thickness in wells at the center of the pool was about 0.4 meters. A pumping well was installed near the center of the oil pool and screened 4-10 meters below the floating oil. During the test, water was pumped out at about 240 liters/min for 48 hours. Water levels were monitored in 21 wells that were screened below the water table and did not contain oil. Data loggers and pressure transducers were used to monitor 17 of these wells, and 4 wells were measured by hand using a tape. In 20 other wells that were screened at the water table and contained oil, depths to the oil-air and oil-water interfaces were monitored by hand using an oil-interface meter. Preliminary results indicate that oil thickness in wells within about 5 meters of the pumped well increased rapidly during the test to more than a meter. Oil also entered the top of the pumped well screen and filled the well bore to a thickness of about 3 meters. Preliminary analysis of water table drawdown vs. time data implies that the horizontal hydraulic conductivity is about 60 m/day, which is an order of magnitude greater than estimates based on inverse modeling of BTEX transport in the dissolved plume. The apparent ratio of horizontal to vertical hydraulic conductivity is about 10:1. The apparent specific yield is about 0.01, much lower than the typical range of 0.15-0.25 for a water-table aquifer in this sandy glacial outwash formation. This result suggests that the oil may have restricted vertical flow, causing an apparent "semi-confined" aquifer response.
DE: 1828 Groundwater hydraulics
DE: 1829 Groundwater hydrology
DE: 1832 Groundwater transport
DE: 1847 Modeling
SC: Hydrology [H]
MN: Fall Meeting 2005