Raymond H. Johnson and Eileen P. Poeter
Colorado School of Mines, rhjohnso@mines.edu, epoeter@mines.edu, Golden, CO, USA
and United States Geological Survey, Denver, CO, USA

Abstract

DNAPL saturations estimated from ground penetrating radar (GPR) data are used as observations to invert multiphase flow simulations to estimate intrinsic permeability values. Synthetic cases show that saturation data alone are sufficient to estimate optimal intrinsic permeability values, but the character and magnitude of error in the saturation data are critical to accurately estimating permeability values. For the Borden PCE spill, inversion of flow simulations reveals bias in the difference of simulated and GPR interpreted DNAPL saturations. The data cannot distinguish whether the bias is due to error associated with the GPR interpretation or error in the conceptual models, which leads to reinterpretation of the GPR data, creation of alternate conceptual models and revisions to the multiphase flow code. The resulting reinterpretations produce improved, but still biased, residual statistics. Determining the most appropriate conceptual model and flow code formulation requires quantification of the error associated with saturation observations through more sophisticated GPR interpretation.