Three-dimensional inverse scattering versus imaging methods

An efficient iterative method is discussed for solving the 3-D nonlinear seismic inverse problem. The method is based on high-frequency asymptotic methods for both computing the data misfit function and accelerating the rate of convergence by means of preconditioning. The method is applied to synthetic and real data and, for the simple models considered here, it is concluded that it appears to be capable of resolving both background velocity and contrast parameters. For a typical three-dimensional data-acquisition geometry, the method seems to be :ess sensitive to sparse-sampling effects than imaging methods