3-D spectral IP imaging: Non-invasive characterization of contaminant plumes. 1998 annual progress report

Abstract

'The overall objective of this project is to develop the scientific basis for characterizing contaminant plumes in the earth''s subsurface using field measurements of induced polarization (IP) effects. Three specific objectives towards this end are: (1) understanding IP at the laboratory level through measurements of complex resistivity as a function of frequency in rock and soil samples with varying pore geometries, pore fluid conductivities and saturations, and contaminant chemistries and concentrations; (2) developing effective data acquisition techniques for measuring the critical IP responses (time domain or frequency domain) in the field; (3) developing modeling and inversion algorithms that permit the interpretation of field IP data in terms of subsurface geology and contaminant plume properties. The authors laboratory experiments to date are described in Appendices A and B, which consist of two papers submitted to the annual SAGEEP conference (Frye et al., 1998; Sturrock et al., 1998). The experiments involved measurements of complex resistivity vs. frequency on a suite of brine saturated sandstone samples. In one set of experiments, the fluid chemistry (pH, ionic strength, and cation type) was varied. In a second set of experiments, the microgeometry of the rock matrix was varied. The experiments showed that spectral IP responses are sensitive to subtle variations in both the solution chemistry and rock microgeometry. The results demonstrate that spectral IP responses have the potential of being sensitive indicators of in-situ chemistry and microgeometry, the latter of which may be related to the hydraulic properties. Data Acquisition The authors have been looking in some detail at the effects of electromagnetic coupling and how to practically deal with it. In this area, the results to date are summarized in Vandiver (1998). The progress in the development of modeling and inversion algorithms for IP is described in Appendix C, a paper submitted to the annual SAGEEP conference (Shi et al., 1998). The authors have developed algorithms for forward modeling and inversion of spectral IP data in 3-D media. The algorithms accommodate a general earth model with a complex electrical conductivity as a function of frequency and 3-D spatial position. Using regularization and optimization techniques, the inversion algorithm obtains a 3-D image of resistivity amplitude and phase for each frequency contained in the data set. They have begun testing their algorithms on synthetic data generated from a simple model of a contaminant plume. The complex resistivity parameters of the background medium and plume are based on the laboratory results described above.'