Identification of subsurface organic contamination, particularly dense nonaqueous phase liquids (DNAPLs) is one of the highest priorities and among the most difficult-for remediation of numerous sites, including those of the Department of Defense (DOD) and the Department of Energy (DOE).  Complex resistivity (CR) is the only geophysical method that has been demonstrated in the laboratory to have high sensitivity to organic compounds, by detecting responses indicative of clay-organic electrochemistry.  However, direct detection of organics in the field has been elusive, in part due to the difficulty of obtaining robust measurements at very low contaminant levels in the presence of heterogeneous geological materials and cultural interference (such as metallic utilities and remediation plumbing).

This project sought to improve the capability to detect DNAPL by (1) better geophysical imaging of geological pathways that control DNAPL movement and (2) direct detection by detailed comparison of CR laboratory to field data using this improved imaging.

For the first goal, algorithms were developed for the joint tomographic imaging of seismic and resistivity data.  The method requires that an empirical relationship can be established between seismic and resistivity; if values are ultimately tied to specific lithologies, then the final tomographic product can be an actual geological cross-section.  Because shallow subsurface investigations are now commonly performed using a cone penetrometer (CPT), a new vibratory seismic source was developed for this platform.  For the second goal, a multistep CR investigation protocol was developed to identify sites with clay-organic reactions measurable in the lab from core samples, perform reconnaissance field surveys, and proceed to detailed 2-dimensional or 3-dimensional cross-hole imaging.

Twenty-nine sites were identified over the duration of this project, but all except one were eventually not useable due to potential or measured interference from previously installed infrastructure, spurious natural signals, inadequate clay-organic signatures, or lack of support from site management.  The A-014 Outfall at the DOE Savannah River Site (SRS) was ultimately chosen for joint cross-well seismic and complex-resistivity imaging.  The seismic-data quality was only fair, because the radiated source waveforms were more narrowband than expected and because of artifacts introduced by incomplete decoupling of the source from the CPT push rods or complications of the radiation pattern.  The CR data also were difficult to interpret due to electrode polarization and nonlinearities in the electrodes and/or geology.  Although both the seismic and electrical tomography were able to identify major contacts, the data quality and quantity were insufficient to demonstrate the utility of joint tomography in the field.  The laboratory CR signatures varied significantly with lithology and duration of DNAPL exposure.  Nonetheless, the field study suggested the presence of DNAPL at several locations on the south and west sides of the study area.

While still a promising tool to identify DNAPL, this project demonstrated that CR is extremely sensitive to site conditions, and the field interpretability of laboratory data remains difficult.  A series of experiments should be devised to separate the effects of electrode material, acquisition system, processing, and geology (particularly nonlinearity).  More on-site surveys for DNAPL are necessary to determine the practical utility of the method.