1D Imaging of Central Loop Transient Electromagnetic Soundings

Abstract

The success of geophysical investigations in general is to a large extent determined by the density of the measurements and the quality of the interpretations. The development in geophysical data acquisition has gone in the direction of covering larger and larger areas with more and more dense grids of measurements. This development has been especially pronounced in the field of transient electromagnetic data acquisition (TEM) for mineral prospecting, where airborne systems collect large volumes of data, and also in the field of environmental geophysics, where dense measurements of transient soundings have proved very useful in connection with hydrogeological investigations. An ordinary 1D least squares iterative inversion of TEM sounding, data requires that the interpreter supply an initial model, and the computation time is not at all negligible even on present‐day computing platforms. With a large daily production of soundings this procedure is slow, and there is need for fast approximate ways of interpretation. This paper outlines the development of an algorithm for imaging of TEM soundings in the central loop configuration based on the Born approximation and the Frechet kernel for the step response of the vertical magnetic dipole. The Frechet kernel depends on the resistivity of the halfspace, but instead of using one Frechet kernel for a certain halfspace resistivity for all data, the Frechet kernel is scaled according to the all‐time apparent resistivity of the measurements at every delay time. In this way, the actual resistivity structure of the halfspace is taken into account, and the inversion problem is linearized resulting in better imaged models. The imaging procedure produces models with 20–40 layers which fit the original data, typically within 5–10%. No initial model is required, and the algorithm is therefore well suited for automatic inversion. The algorithm makes it possible to see the results of a day's work in a matter of minutes and to implement on‐line inversion simultaneous with the measurements.