Euler deconvolution applied to potential field gradients

Abstract

Euler deconvolution is a useful tool for providing initial estimates of the locations and depths of magnetic and gravity sources. When it has been used with aeromagnetic data, it has traditionally been applied to the total magnetic field intensity. In recent work it has been also applied to the vertical gradient of the field with much success. Its application to the horizontal gradient has been less common. This paper shows that a simultaneous deconvolution of both the horizontal and vertical gradients is robust with respect to noise. Its application to the analytic signal, a complex quantity, is also demonstrated, and shown to give superior results than those obtained from the conventional approach of the deconvolution of the amplitude of the analytic signal. Strategies for the clean computation of gradients are discussed, and a new method for rejecting invalid Euler solutions is suggested. The modified Euler deconvolution algorithm is demonstrated both on synthetic models and on aeromagnetic data from the Eastern Bushveld complex of South Africa.