Electromagnetic coupling in frequency domain induced polarisation data

Frequency domain induced polarization (IP) surveys are commonly carried out to provide information about the chargeability structure of the earth. The goals might be as diverse as trying to delineate a mineralized and/or alteration zone for mineral exploration, or to find a region of contaminants for an environmental problem. Unfortunately, the measured responses can have contributions from inductive and galvanic effects of the ground. The inductive components are called EM coupling effects. They are considered to be "noise" and much of this thesis is devoted towards either removing these effects, or reformulating the inverse problem so that inductive effects are part of the "signal". If the forward modeling is based on galvanic responses only, then the inductive responses must first be removed from the data. The motivation for attacking the problem in this manner is that it is easier to solve D.C. resistivity equation than the full Maxwell's equation. The separation of the inductive response from the total response is derived by expressing the total electric field as a product of an IP response function, and an electric field which depends on EM coupling response. This enables me to generate formulae to obtain IP amplitude (PFE) and phase response from the raw data. The data can then be inverted, using a galvanic forward modeling. I illustrate this with 1D and 3D synthetic examples. To handle field data sets, I have developed an approximate method for estimating the EM coupling effects based upon the assumption that the earth is locally 1D. The 1D conductivity is obtained from a 2D inversion of the low frequency DC resistivity data. Application of this method to a field data set has shown encouraging results. I also examine the EM coupling problem in terms of complex conductivity. I show that if the forward modeling is carried out with full Maxwell's equation, then there is no need to remove EM coupling. I illustrate this with 1D synthetic example. In summary, I have investigated the EM coupling problem in IP and developed a practical removal methodology that can be applied to data sets from 1D, 2D and 3D earth structures.Science, Faculty ofEarth, Ocean and Atmospheric Sciences, Department ofGraduat