Aeromagnetic noise reduction and interpretation by Multifractal Singular Value Decomposition

Abstract

Abstract The aeromagnetic signal often has a minority of noise in it, which may be composed of random noise caused by the aircraft itself, fringe pattern noise caused by the differences between the aircraft airlines, or linear features aligned along the declination direction after reduction to the pole, which are usually hard to distinguish visually but will make useful information submerged In this paper, the aeromagnetic data of a polymetallic deposit-accumulated area in Inner Mongolia Province, northwest China, is taken as an example, from which the aeromagnetic noise is extracted by multifractal singular value decomposition (MSVD), the singular values are automatically fitted into limited number of fractal straight lines according to their inflection points and minimum fitting error, each line represents a certain kind of noise or signal, the noises are chosen and removed in this way. The last two decomposed components are usually a band pass and a low pass filter that can be used for tectonic faults and geological bodies’ interpretation, respectively.