Electromagnetic inversion for environmental site characterization: Data quality versus image resolution

Abstract

Non-linear electromagnetic inversion schemes have been developed to produce 2D and 3D images of subsurface conductivity structure from electromagnetic geophysical data. The solutions are obtained by successive linearized model updates where full forward modeling is employed at each iteration to compute model sensitivities and predicted data. Regularization is applied to the problem to provide stabiity. The use of the inversion is demonstrated on a data set collected with the Apex Parametrics tMaxMin I-8S' over a section of conductive waste at the Idaho National Laboratory's Cold Test Pitt . The out-of phase data are of very good quality while the in-phase are rather noisy due to slight mispositioning errors. A resolution study on synthetic data indicates that the enor present in the in-phase data causes images of far lower resolution with more artifacts than if the in-phase and out-of-phase components are of similar quality. Better resolution images result if the data are weighted proportional to frequency; this gives each frequency equal weight. The loss of resolution due to poor quality in-phase data is demonstrated in a 3D inversion of the MaxMin data which shows both artifacts forming outside of the area known to contain the buried waste, as well as an inability to resolve depths.