In this paper, we discuss the influence of the selfdemagnetization effect on magnetic data and present an alternative means of quantitatively interpreting such data in highly magnetic environments. In particular, we present two important results based on simulation which one might consider in their interpretation of magnetic data when selfdemagnetization is present. First, current methods for estimating total magnetization, which are typically applied to the problem of remanent magnetization, do not reliably recover this parameter when the anomalous source bodies have high magnetic susceptibilities. And second, a single value estimation of total magnetization does not provide adequate information to properly resolve subsurface geology through inversion. Numerical experiments demonstrate that directly inverting amplitude data, calculated from magnetic data yet weakly-dependent on magnetization direction, produces superior results when interpreting data generated in terrain with high magnetic susceptibilities.
It is commonly accepted that adequate knowledge of true magnetization direction of a causative body is crucial in order to accurately interpret magnetic data by quantitative methods such as inversion. Most currently available algorithms require the knowledge of magnetization direction, since it is an essential piece of information for carrying out the forward modeling (e.g., Li and Oldenburg, 1996; Pilkington, 1997). Such a requirement has been the driving force for development of many well recognized approaches for estimating total magnetization when strong remanence or self-demagnetization are present. Example problems to which these methods are commonly applied include the interpretation of magnetic data over ferrous unexploded military ordnance, banded iron formations, nickel deposits, kimberlite pipes, and depth to basement problems. In most practical exploration cases, one can simply assume that there is no remanent magnetization and the selfdemagnetization effect can be neglected. Consequently, the direction of magnetization is assumed to be the same as the current inducing field direction. This is a valid assumption in a majority of the cases, as evidenced by many successful applications. However, there are well-documented cases in which such an assumption is inadequate due to the presence of remanence and selfdemagnetization. The difficulties of interpreting magnetic data for these two distinct problems are similar, in some respect, since the magnetization of the causative body will be rotated away from the Earth’s inducing field – often drastically. However, the similarities stop there. This is because remanent-magnetization and self-demagnetization have entirely different origins. Remanent magnetization is the net magnetization present in a material in the absence of an external field (Merrill et al., 1996). Remanent magnetization can occur for any magnetic geologic unit and commonly does not have strong dependence on geometry of the source body. Self-demagnetization, in contrast, exists when susceptibilities become large and the magnetic field at a location in the source body is significantly affected by the induced magnetization from neighboring domains (Clark & Emerson, 1999). This process is highly dependent upon the source geometry and the magnetization direction can be much more variable. The remanent magnetization problem has been well studied and there are two general approaches which have emerged for interpreting magnetic data affected by it.