A Stabilization Procedure For The Transformation Of Magnetic Data

Abstract

Magnetic anomalies are as a result of the summation of magnetic effects produced by spatial variation of magnetic polarization. Consequently, transformations which attenuate the unwanted components of the observed anomalies have been of great interest in processing of magnetic data. Transforming data into the frequency domain has a problem caused by imperfections introduced in the data spectrum by aliasing and Gibb=s effect, which are caused by the discretization and truncation of data respectively. Stable transformation such as upward continuation, integration and reduction to the pole at high magnetic latitudes either preserves or attenuates the amplitude of these imperfection in the data spectrum. Whereas unstable transformation such as downward continuation, reduction to the pole at low latitudes and first vertical derivatives, greatly enhances the amplitude of the spectrum imperfections so that any direct application of such transformation (without stabilization) will produces poor result. Using equivalent source technique usually produce computations too large for computers to handle and the choice of a good stabilizing factor can be arbitrary. In this paper comparison is first made between the conventional filtering technique and the equivalent source technique using theoretical data and secondly a quantitative method is developed by using an algorithm which uses the correlation coefficient between successive pairs of the transformed maps.

IFE Journal of Science Vol. 9 (1) 2007 pp. 77-86