Separating volume and roughness components of scattering from the seafloor using spatial coherence filtering

Seabed scattering is usually associated with two mechanisms, interface roughness and volume heterogeneity, which can operate at the same time. In this paper, a method of separately identifying these two mechanisms is presented, involving measurements of spatial coherence of the field with a vertical linear array located near the source. Analytical expressions are derived for the spatial coherence of interface and volume scattering and used to illustrate the potential for their separating and quantifying. It is shown that the correlation scales for the volume and roughness components can be significantly different, and thus spatial filtering can be used for their separating. A Monte Carlo simulation technique based upon the point scatterer model is also developed that makes it possible to generate random time series for the scattered field. An algorithm for data processing is proposed to estimate spatial correlation as an average over realizations (from separate pings) of the measured time series. The simulations illustrate the ability of the algorithm to distinguish and separate volume and surface components of scattering and also verify the analytical expressions for spatial coherence. It is shown that the convergence to the theoretical prediction for both volume and surface components of scattering indeed takes place, but longer pulses require a larger number of measurements (realizations). [Work supported by ONR Code 321OA.]