Paper No. 4
Presentation Time: 2:15 PM
HYDROCARBON DEGRADATION IN LNAPL OIL POOLS IN TWO OIL-CONTAMINATED AQUIFERS
Two light non-aqueous phase liquid (LNAPL) oils that resulted from spills were analyzed for hydrocarbons to determine changes in the chemical composition of the oils over time. One spill occurred in 1979 from a pipeline break near Bemidji, MN, and the other spill, about 40 km away, was from a gradually leaking valve near Cass Lake, MN. The two oils originated from the same pipeline and reside at the water table, which is 6-9.5 m below land surface. The surficial sediments at both sites are 5-14 m thick outwash of fine to medium-grained sands and silts with some clay and gravel. Volatile and semi-volatile concentrations of hydrocarbons were determined in oil samples obtained over a 21-year period from wells at six locations. The order of loss, primarily due to biodegradation, for 25 hydrocarbons analyzed was toluene> o-xylene and n-C6 and n-C12 alkanes > benzene, > n-C7 to n-C11 alkanes> ethylbenzene, m-xylene, cyclohexane, and dimethylnaphthalenes. Other hydrocarbons including p-xylene, 1,2,4- and 1,3,5- trimethylbenzenes, methyl- and ethyl- cyclohexanes, and naphthalene showed little loss or were slightly depleted in the LNAPLs and 1,2,3-trimethylbenzene and the tetramethylbenzenes were not degraded. Relative to the composition of oil from the pipeline, volatile and semi-volatile hydrocarbon concentration losses in oils from wells were 13-64% at the Bemidji, MN site and 85% at the Cass Lake, MN site. The concentrations of other hydrocarbons, including C11-31 n-alkanes, pristane, and phytane, also, indicate that degradation of the oil is occurring in the aquifer. Controls on the attenuation of hydrocarbons in sub-surface oil pools include: heterogeneity of the aquifer, rate of release of the oil, distribution of the oil in the aquifer, and availability of nutrients, which enhance biodegradation. These results demonstrate that the extent of degradation within an oil pool is not always uniform. This has implications for defining the source term in ground-water modeling, for determining the ages of spills, and for the development of remediation strategies.