We have developed an algorithm to obtain the model parameters for two co-axial structures from self-potential data. The method uses the first numerical horizontal derivatives calculated from the observed self-potential anomaly employing filters of sequential window lengths (s-values) so as to gauge the model constraints for the shallow and deep structures. In addition, this algorithm uses a standard inversion method for solving a non-linear equation based on the lowest root-mean-square (RMS) error of the estimated model parameters. The body constraints are the depth, polarization angle and electric dipole moment of each structure. Our approach models the self-potential dataset as an aggregation of spheres, horizontal cylinders, and vertical cylinders. These simple bodies are used to approximate, without a priori expectations, the furthermost plausible position and/or area of intersection. In other words, the bodies are used to estimate the true values of the source parameters for the two-co-axial bodies at different s-values. Minimizing the RMS error has the advantage of optimizing all model factors. The proposed technique is tested using a numerical model with and without noise and on self-potential field data acquired at a site in Germany. In all cases, the assessed body parameters are reasonable approximations of the known values.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.