NOVEMBER 1997 THE LEADING EDGE 1631 Tools for electromagnetic investigation of the shallow subsurface L OUISE PELLERIN, Lawrence Berkeley National Laboratory, Berkeley, California DAVID L. ALUMBAUGH, Sandia National Laboratories, Albuquerque, New Mexico The U.S. Department of Energy (DOE), Office of Environment Management supported a study, in 1996, to identify electromagnetic methods appro- priate for environmental investigations. The pur- pose was to establish the capabilities of systems presently available and target areas for research. The study, conducted at the Idaho National Engineering Laboratory (INEL) Cold Test Pit (CTP), established a current baseline for commercial and research systems; identified gaps in measurement and interpretation; provided a multidimensional data set for testing high resolution imaging and inversion software; acquired broadband data in a variety of array configurations to enhance the qual- ity and efficiency of the interpretation; and pro- duced a data set available for classroom/laborato- ry studies in environmental geophysics. This paper describes 11 technologies demon- strated and evaluated in the study. The intention is not to promote/criticize any specific instrument, but to evaluate the various technologies. Data, a detailed survey description, and lists of contacts are available at http://vetem.lbl.gov/emid. Eight technologies are illustrated in the accom- panying figures. Figures are not included for ground-penetrating radar (GPR), high frequency ellipticity (HFE), and very early time EM (VETEM). GPR is ineffective at INEL due to unfavorable sur- face conductivities and high magnetic losses of the soils. Different antennas and array configurations were used with little success. Sensors and Software represented the method well, and the poor results should not be assumed to reflect poorly on the methods or the system. GPR is often problematic where conductive soils (such as clay caps over waste pits) are present. In fact, it was not an acci- dent that CTP is a poor GPR site because DOE was interested in investigating technologies that work in areas where GPR is not successful. HFE and VETEM have the potential to fill the crit- ical gap between GPR and low frequency diffusive EM. A fundamental problem in working in this range is that the earth loads the transmitter, changing the calibration of the system and thus limiting interpre- tation. The HFE group took an empirical approach by training neural networks with data acquired at dif- ferent test facilities. The ellipticity parameterization reduces calibration problems and the neural network incorporates the system response into the solution; but problems can occur when the network encounters a data set for which it is not trained. The VETEM team took a theoretical approach, based on numerical mod- eling, to system design and interpretation. Because these systems are unique and not routinely available, results are not shown here. Data are available on the web site. A figure is also not included for electrical resistivity because the data are comparable to those shown under induced polarization. Idaho National Engineering Laboratory, Cold Test Pit The INEL CTP, on the Snake River plain flood basalt in south central Idaho, was constructed to test various retrieval, stabi- lization, and characterization technologies. It was excavated to the underlying basalt, and soil from a nearby quarry was import- ed to build the pit. The cells, engineered to simulate waste pits that are 30-40 years old, contain a mixture of metal and cardboard drums. The engineering drawing locates the targets, but depth estimates are not available; hence, excavation of the CTP is nec- essary to complete the study. The locations of the survey grid coordinates are given in feet with respect to a local origin at the northwest corner of the box cell and the southwest corner of the drum cell (coordinates denoted on the figure).