The use of multivariate statistical techniques for the analysis and display of AEM data

Abstract

Images of multiple AEM channels can be combined to produce RGB composites that display information about the shape of the EM spectrum (or its time domain equivalent). The very high between-channel correlation, and the wide dynamic range of AEM data mean that careful pre-processing must be performed in order to extract all the useful information in the data. In addition, while data compression techniques such as Principal Components Analysis provide good enhancement of variability within the data set, they produce images that are difficult to relate to geophysical parameters such as conductivity and depth. These limitations can be overcome to some extent by nonlinear rescaling on the AEM data to take into consideration the inherent non-linear relationships arising from the physics of the EM diffusion process. This paper illustrates each of the above problems and discusses new processing methods to overcome them.