LAUSR

We present a wavenumber-domain iterative approach for rapid 3-D imaging of gravity anomalies and gradients data, which is based on the 3-D mesh model with a flat observational surface. The approach deconvolves the spectra of gravity anomalies or gradients by a 2-D deconvolution filter describing the spectrum of the imaging operator and then transforms the resultant spectra into the space domain to derive the density distribution. This 2-D deconvolution filtering is operated layer by layer from top to bottom in the subsurface and finally, all the results are merged to generate the 3-D density distribution. We improve previous 2-D deconvolution filters by involving a depth-scaling factor and utilize a priori constraint and the iteration algorithm for imaging, enable the presented approach to produce a density model with a considerable resolution and accuracy. The wavenumber-domain algorithm makes the imaging faster than the conventional space-domain inversions. Tests on the synthetic data and the real data from a metallic deposit area in Northwest China verified the feasibility and high efficiency of the presented approach.