Implications of magnetic backgrounds for unexploded ordnance detection

Abstract

Detection of buried unexploded ordnance (UXO) requires application of geophysical methods that exploit contrasts in magnetic susceptibility and/or electrical conductivity of the UXO relative to surrounding soil and rock (the geologic background). At many sites, the contrasts are very large, and the detection capability is relatively independent of the geologic background. Some sites, however, are predictably problematic, such as volcanic terrains (e.g., Maui and Kaho'olawe, HI, USA). Other sites that would not intuitively be predicted to pose detection problems are found to have localized high susceptibility zones with wavelengths and magnitudes that can significantly complicate UXO detection. Details of a site in Indiana (USA) are presented that illustrate order of magnitude variation of magnetic susceptibility over distances of 2–3 m. Magnetic susceptibility at the site correlates to topography and soil type. Magnetic power spectral density plots from magnetic surveys at selected areas at the Indiana site are compared to each other and to data from Maui, HI, and demonstrate the key issues in UXO detection with magnetometry. Magnetic susceptibility data from Kaho'olawe are used to illustrate the effects on electromagnetic induction surveys for UXO detection and discrimination. The phenomenological observations suggest approaches for magnetic and electromagnetic induction survey data processing to enhance anomaly detectability and model-based inversion.