Holistically calibrating, processing and inverting frequency domain AEM surveys

Abstract

A holistic approach to the calibration, processing and inversion of airborne frequency domain electromagnetic data is proposed. This is achieved by simultaneously determining calibration corrections and a conductivity model. This would avoid the inevitable propagation of errors that occurs when calibration, processing and inversion are tackled in an independent and sequential manner. The method is structured as a large non-linear inverse problem, the solution of which yields a set of parameters from which calibration correction factors and a three-dimensional (3-D) conductivity model can be produced by summation of basis functions. The inversion ensures that the observed airborne data are consistent with data predicted from forward modelling of one-dimensional (1-D) local conductivity models generated from the 3-D model, and with the correction parameters. It also ensures that the conductivity model is consistent with independent observations of subsurface conductivity, and with a priori knowledge of conductivity including the local spatial continuity or smoothness of these values.