Depth estimation from the scaling power spectrum of potential fields?

Depth estimation from potential field power spectra requires a realistic assumption of the statistical properties of the source distributions. Density and susceptibility distributions in the Earth's crust exhibit a long-range dependence, which is adequately described by scaling random fields with a spectral density proportional to some power of the wavenumber. The theoretical power spectrum for a half-space model of scaling sources explains the shape of observed power spectra of real potential field data very well. Minimizing the misfit between the model and the observed power spectrum yields an estimate for the depth to the top of the sources. After demonstrating this approach on synthetic magnetic data, we reinterpret power spectra of gravity and aeromagnetic data from Utah, Hawaii and Saskatchewan, finding depth values that differ significantly from earlier interpretations. All three power spectra are best explained by source distributions starting at surface level, even the power spectrum from an aeromagnetic survey of a sedimentary basin with virtually non-magnetic basin fill. In the latter case, a priori information on the intensity and the scaling exponent of the field caused by the basement had to be included to obtain an approximate estimate of the basin depth. In general, potential field power spectra are dominated by scaling properties of their source distributions and contain only limited depth information.