1999 GSA Annual Meeting -- Denver, Colorado
1999 GSA Annual Meeting -- Denver, Colorado
Author(s): JOHNSON, Raymond, H., Dept. of Geology and Geological Engineering, Colorado School of Mines, Golden, CO, 80401, rhjohnso@mines.edu; SNEDDON, Kristen, and OLHOEFT, Gary, Geophysics Dept., Colorado School of Mines, Golden, CO, 80401; POETER, Eileen, and MCCRAY, John, Dept. of Geology and Geological Engineering, Colorado School of Mines, Golden, CO, 80401; POWERS, Michael, U. S. Geological Survey, P. O. Box 25046, MS-964, Denver, CO 80225.
Keywords: DNAPLs, inverse modeling, ground-penetrating radar
Toxic organic contaminants in the form of dense non-aqueous phase liquids (DNAPLs) are a well-know threat to subsurface drinking water supplies. Reliable predictions of the subsurface movement of DNAPLs and the performance of alternative remedial actions require accurate values for hydraulic parameters and conceptual models. The lack of reliable values for the pertinent parameters and cost of obtaining such values are limiting factors in predicting DNAPL movement. Researchers have successfully delineated the subsurface distribution of DNAPLs using non-invasive measurements such as ground penetrating radar (GPR). This study presents a methodology that uses only post-spill, ground penetrating radar data describing the DNAPL distribution as observations for the inversion of multiphase fluid-flow models. We demonstrate a methodology to: 1) interpret and evaluate the GPR data; 2) ascertain the type and location of data that are most effective in characterizing the system; 3) determine the necessary parameter values for simulation of the system; 4) identify representative conceptual models; and 5) quantify the uncertainty associated with model predictions. This methodology enables the successful use of DNAPL flow models for design of remedial action. GPR data from a DNAPL spill in the Borden aquifer (9m x 9m cell experiment, University of Waterloo) is used in the development of the methodology. The experience gained from this study can be used to guide the determination of data needs and the selection of conceptual models at DNAPL field sites. The methodology reduces the cost and risk associated with evaluating DNAPL spills and designing effective remedial action.
© Copyright 1999 The Geological Society of America. All rights reserved.