Preliminary investigation of groundwater flow and trichloroethene transport in the Surficial Aquifer System, Naval Industrial Reserve Ordnance Plant, Fridley, Minnesota

Abstract

Industrial practices at the Naval Industrial Reserve Ordnance Plant, in Fridley, Minnesota, caused soil and groundwater contamination. Some volatile organic compounds from the plant might have discharged to the Mississippi River, forced by the natural hydraulic gradient in the surficial aquifer system. The U.S. Environmental Protection Agency included the Naval Industrial Reserve Ordnance Plant on the Superfund National Priorities List in 1989. This report describes a preliminary characterization of trichloroethene transport in the surficial and Cambrian-Ordovician aquifer systems at the Naval Industrial Reserve Ordnance Plant. The characterization first involved simulation of 2001 conditions using a model, followed by an application of this 2001 simulator to 2011 conditions. The U.S. Geological Survey, in cooperation with the U.S. Department of the Navy, used a steady-state, uniform-density groundwater flow model to simulate measured potentiometric heads in aquifer systems on August 20, 2001, and a single-phase, conservative, non-reactive, miscible transport model to simulate trichloroethene concentrations in aquifer systems measured in 2001. The U.S. Department of the Navy furnished trichloroethene source areas and trichloroethene source area concentrations to the U.S. Geological Survey for this model simulation. Furnished delineations were postulated and informed by data collected from 1995 to 2011. The groundwater flow simulation of August 20, 2001, was superior to the trichloroethene transport simulation at replicating measurements; simulated potentiometric heads matched 90 percent of measured potentiometric heads on August 20, within 2 feet at selected locations whereas simulated trichloroethene concentration contours of 3, 10, 100, 1000, and 10,000 micrograms per liter (µg/L) correctly bounded 52 percent of measured concentrations in 2001 at selected locations. The degree to which the simulated trichloroethene plume does not match trichloroethene measurements in the surficial aquifer system during the 2001 simulation may suggest that furnished trichloroethene source areas and trichloroethene source area concentrations did not accurately represent all trichloroethene sources in the hydrogeologic system. During the model simulation of 2001, trichloroethene discharged to the Mississippi River. A simulated 900-foot-long zone of benthic trichloroethene discharge flux existed in the shallow flow zone, across which simulated trichloroethene discharged from the surficial aquifer system to the Mississippi River at simulated trichloroethene concentrations that ranged from 3 µg/L to more than 100 µg/L. The Mississippi River was not sampled for volatile organic compounds in Fridley, Minn., from 1999 to 2016 (the publication of this report). Trichloroethene concentrations were measured in wells close to the Mississippi River in the surficial aquifer system on the downgradient side of the Naval Industrial Reserve Ordnance Plant groundwater flow field; for example, at well MS–43 in the shallow flow zone of the surficial aquifer system 280 feet east of the Mississippi River between December 1999 and August 2012, trichloroethene concentrations ranged from 130 to 220 µg/L. The 220-µg/L maximum concentration was reached in March 2003 and October 2006. The August 2012 concentration was 140 µg/L. The August 20, 2001, groundwater flow model simulator and the 2001 trichloroethene transport simulator were applied to a groundwater extraction and treatment system that existed in 2011. Furnished trichloroethene source areas and concentrations in the 2001 simulator were replaced with different, furnished, hypothetical source areas and concentrations. Forcing in 2001 was replaced with forcing in 2011. No trichloroethene concentrations greater than 3 µg/L were simulated as discharging to the Mississippi River during applications of the 2001 simulator to the 2011 groundwater extraction and treatment system. These applications were not intended to represent historical conditions. Differences between furnished and actual trichloroethene sources may explain differences between measurements and simulation results for the 2001 trichloroethene transport simulator. Causes of differences between furnished and actual trichloroethene sources may cause differences between hypothetical application results and the performance of the actual U.S. Department of the Navy groundwater extraction and treatment system at the Naval Industrial Reserve Ordnance Plant. Other limitations may also cause differences between application results and performance.