Ocean bottom seismic scattering

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 August 1989Seismic studies of the oceanic crust, both experimental and theoretical, often assume a flat seafloor and laterally homogeneous crust. This is done regardless of the appearance in seismic data of obvious effects due to scattering from lateral heterogeneities both on and in the seafloor. Detailed fine scale surveys of mid-ocean ridges, where the upper oceanic crust is exposed, have revealed the presence of lateral heterogeneities in the form of complicated topography, extrusive volcanic structure, and abundant fracturing and faulting. These heterogeneities have a significant affect on the propagation of seismo/acoustic energy through the crust, especially in the immediate vicinity of the seafloor. This thesis deals with the problem of scattering of seismo/acoustic energy from a number of forms of lateral heterogeneity in the upper oceanic crust. A common theme throughout this work is that the size of the heterogeneity on or in the seafloor is of the same order of magnitude as the seismo/acoustic wavelength. This is the realm of scattering theory where the wave-like characteristics of seismic energy have a particularly large influence on the outcome of interaction with structure in the media. The work presented here involves the application of the finite difference modeling technique to problems concerning laterally heterogeneous elastic media. This method is a full wave solution to the elastic wave equation and as such includes all wave interactions with the media. The finite difference formulation is used to study four distinct phenomena; scattering from discrete deterministic seafloor features; wave propagation through continuous randomly heterogeneous upper oceanic crust; scattering from more complicated topographic profiles and the limitations of the method for the rough seafloor problem; and the problem of plane acoustic wave scattering from an infinite elastic cylinder. The principal finding of this work is that lateral heterogeneities in the upper oceanic crust can have a dramatic affect on seismo/acoustic wave propagation. Scattering from rough seafloors and/or volume heterogeneities is often quite similar and causes the occurrence of signal generated 'noise' (coda), decorrelation of primary arrivals, and anomalies in arrival travel time and amplitude. Topographic and volume scatterers acting as secondary sources of seismic energy can cause a resonant coupling of body wave energy into interface (Stoneley) waves at the seafloor. This is possibly one mechanism by which natural seismic and storm generated acoustic energy can be coupled into seafloor noise. The applicability of the use of the finite difference method for non-planar water-solid interfaces is also discussed. Models were calculated which approximate sinusoidal seafloors and plane acoustic wave scattering from an infinite elastic cylinder. The discretization of a rectangular difference grid must be extremely fine to accurately accommodate a smoothly varying water-solid interface which does not align with the grid. Regardless of the discretization concerns, the rough seafloor models presented here demonstrate the arrivals expected from larger scale sinusoidal topography as well as the importance of considering quite small (