2D TEM Modeling and Inversion by Adaptive Born Forward Mapping

Abstract:

In the airborne electromagnetic survey, vast data are acquired with the development of precise measuring equipment and the automation of data acquisition. In this study we developed fast and accurate two-dimensional (2D) modeling and inversion algorithm based on the adaptive born forward mapping (ABFM) method, which is recently emerging for fast time-domain electromagnetic (TEM) modeling. The ABFM method is an approximation method that takes into consideration the true electrical conductivity distribution of subsurface media and is different from the conventional Born approximation that uses the constant electric conductivity. One of the most important points of the ABFM method is how to set a suitable sensitivity function. In this study, the known 1D sensitivity function was expanded into 2D sensitivity function to effectively approximate the dispersive behavior of electromagnetic field. By comparing the analytic solution and approximate ABFM solution for layered earth models, we found that the two solutions correspond to each other well. This implies that the 2D sensitivity function suggested in this study is suitable and that the ABFM method has very excellent accuracy in 2D TEM modeling even though it is an approximation method. Furthermore, a 2D inversion algorithm was developed with respect to the apparent conductivity data of TEM based on ABFM. To enhance the resolution and stability, the smoothness-constrained least-squares method with ACB constraint was employed. The inversion of calculated data for various models produced a reasonable model close to the true model. It is expected that the method will be extensively applicable to TEM modeling and inversion without difficulty in the future.