Effects of anisotropic magnetic susceptibility in data interpretation and potential in application

Abstract

Anisotropic magnetic susceptibility (AMS) contains information about the internal structure of magnetic geologic units and may have the potential to provide information about the formation history and subsequent geologic processes to which these units were subjected. Its application in the geological study has been established as a reliable tool in formation analyses. However, in the geophysical study, especially in magnetic data interpretation, there has not been much work on AMS, although its potential has been known for decades. Based on the current research status, the goal of this thesis is to assess the influence of AMS on induced magnetization and to explore the possibility of extracting structural information from magnetic data containing the influence of AMS. In this work, I review the basic mathematical expression of AMS and examine the rotation matrix through the derivation of Euler parameters. I show that the matrix can be directly formed by using the principal directions of the AMS within the user-defined coordinate system. Forward modeling of magnetic response over a synthetic AMS model shows that even at low susceptibility values, the induced magnetization direction can be rotated significantly away from the inducing field direction. A series of numerical experiments demonstrate that estimating source parameters directly from magnetic data in the presence AMS can recover the information of AMS and help to identify different magnetic units. The comparison between different constructions of rotation matrix provides us with a direct means to rotate AMS tensor for modeling purpose, while the forward modeling result shows that the interpretation of AMS-influenced data is faced with similar challenges as found in the cases of self-demagnetization due to high magnetic susceptibility or strong remanent magnetization. Parameter estimation studies show that it is possible to recover structural information through AMS from magnetic data without the assistance of laboratory measurements. This research opens the door for future work on magnetic data interpretation