Analysis of modal evolution caused by a weakly range-dependent seabed in shallow water and its application to inversion for geoacoustic properties
MetadataShow full item record
In a shallow water ocean environment, the range-dependent variation of the geoacoustic properties of the seabed is one of the crucial factors affecting sound propagation. Since the local modes of propagation depend on the spatial changes in the bottom sediments, the local eigenvalues of these modes are useful as tools for examining the range dependence of the sediment properties. In order to extract the local eigenvalues from measurements of the pressure field in a laterally inhomogeneous waveguide, the zero-order asymptotic Hankel transform with a short sliding window is utilized. The local peak positions in the output spectra differ from the local eigenvalues due to both the range variation of the local modes and the interference of adjacent modes. The departure due to the former factor is evaluated analytically by using the stationary phase method. In order to reduce the error induced by the latter factor, mode filtering is utilized by incorporating data from a fixed vertical array of receivers. The use of the above zero-order Hankel transform in a three-dimensionally varying waveguide results in an underestimate of the local eigenvalues due to the effect of horizontal refraction. Thus a general asymptotic Hankel transform with a 2-D sliding window is used to correct for the underestimated amount. By expanding the latter transform with respect to the azimuthal angle, it can also be shown that the first term in the Taylor series corresponds to the former transform; the rest of the terms account for the value difference between the underestimated and actual local eigenvalues. In order to obtain the spatial variation of the sediment properties from the rangedependent variation of the extracted local eigenvalues, the analytical relationship between these two variations is derived by using a perturbation method in a horizontally varying, multi-layered bottom model. Upon use of the n2-linear profile in each layer, the relationship can be obtained in closed form. As a result, the range variation of the local eigenvalues may be separated into terms that depend on each geoacoustic parameter. Based on this relation, an inversion method for determining the range-dependent geoacoustic parameters is developed. The methods developed in this thesis are applied to simulated pressure field data as well as experimental field data. It is shown that the evolution with range of the local modes as well as the range-dependent geoacoustic properties can be successfully estimated.
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution September 1993
Suggested CitationThesis: Ohta, Kazuhiko, "Analysis of modal evolution caused by a weakly range-dependent seabed in shallow water and its application to inversion for geoacoustic properties", 1993-09, DOI:10.1575/1912/5510, https://hdl.handle.net/1912/5510
Showing items related by title, author, creator and subject.
Barth, John A. (Massachusetts Institute of Technology and Woods Hole Oceanographic Institution, 1987-10)A two-layer shallow water equation model is used to investigate the linear stability of a coastal upwelling front. The model features a surface front near a coastal boundary and bottom topography which is an arbitrary ...
Sanford, Lawrence P. (Massachusetts Institute of Technology and Woods Hole Oceanographic Institution, 1984-08)Intermittent, shoreward propagating packets of high frequency first mode internal waves are common on the continental shelf when the water column is stratified and may induce large fluctuations in near bottom velocity. ...
Nadakuditi, Rajesh Rao (Massachusetts Institute of Technology and Woods Hole Oceanographic Institution, 2007-02)The first part of the dissertation investigates the application of the theory of large random matrices to high-dimensional inference problems when the samples are drawn from a multivariate normal distribution. A longstanding ...