Remanent Magnetization

This Inversion of magnetic data has long been hampered by the need to specify the direction of magnetization. Our research presents a general approach that utilizes the minimal dependence on magnetization direction of amplitude and total gradient data and thereby overcome the difficulty associated with inversion when unknown remanent magnetization is present. To construct the inversion algorithm for the magnitude of magnetization, we solve a nonlinear minimization problem formulated using Tikhonov regularization. A positivity constraint is also incorporated to improve the solution. The algorithm will be illustrated with both synthetic and field data sets.

Download the full expanded abstract from SEG 2004

 

 

Magnetics is a common geophysical technique used to image subsurface structure though variations of magnetic properties. Three-dimensional inversion has been used successfully to achieve quantitative interpretation of magnetic data. However, a crucial parameter in this process is the direction of magnetization. The total magnetization is a vector sum of two components. Induced magnetization is well approximated by the inducing field direction; remanent magnetization is commonly unknown and can have adirection significantly different from that of the current field or a magnitude large enough to alter the direction of total magnetization. As a result, the magnetization direction becomes an unknown quantity and hampers inversion and interpretation of magnetic data.

I present a general approach for inverting magnetic data in the presence of strong remanent magnetization. Two quantities, the amplitude of the anomalous magnetic field and the total gradient, defined as the magnitude of the gradient vector of magnetic anomaly data, are weakly dependent upon the magnetization direction in three dimensions. Therefore, I invert amplitude and total gradient data directly to recover the magnitude of magnetization without precise knowledge of its direction.

Since amplitude and total gradient data depend nonlinearly upon magnetization, solution of a nonlinear inverse problem is required. Further nonlinearity is introduced by imposing a positivity constraint on the magnitude of magnetization. I formulate the inversion using Tikhonov regularization, impose positivity by using a logarithmic barrier method, and solve the resulting optimization by truncated Gauss-Newton method.

The ability to invert magnetic data with little information about the nature of remanent magnetization increases the areas in which three-dimensional inversion of magnetic data can be applied. In fact, it is now possible to invert any magnetic data to some extent. This newfound ability opens the door to quantitative interpretation of data in a variety of practical problems ranging from archaeological investigations, mineral and resource exploration, and crustal and planetary studies.

Download the full Masters Thesis

 


Department of Geophysics• Colorado School of Mines • Golden, Colorado 80401
Phone (303) 273-3510 • E-mail: cgem@mines.edu

CGEM Web Pages designed by mcelhinneydesign.com