Deconvolution of the modal phase velocity spectrum for source depth estimation in shallow water

Abstract

The modal energies have been used to estimate source depth in shallow water. The performance of source depth estimation based on mode filtering would be severely impacted if the aperture of the horizontal array is not sufficient, given the closeness of modes in the wavenumber domain. In this paper, we use the fact that the phase velocity spectrum of the source does not truly represent the modal energy distribution, but rather one that is convolved with the array response functions. Thus, the source depth is estimated by deconvoluting the phase velocity spectrum. During the deconvolution, modal energy ratios are selected from the replica field, and an iterative process is used to search for phase velocities of modes to minimize the residual error between the measured and reconstructed phase velocity spectra. Simulation results demonstrate that the proposed method provides better performance with relatively small apertures and low signal-to-noise ratios compared to the maximum a posteriori filter. Real data collected from a horizontal array on the seafloor confirms that the proposed method is capable of accurately estimating source depth in a shallow water environment.