The Basin Acoustic Seamount Scattering Experiment (BASSEX) of 2004 was conducted to measure forward-scattering around the Kermit-Roosevelt Seamount Complex in the Northeast Pacific. The BASSEX experiment was focused on the bathymetric effects on acoustic propagation, in particular, on direct blockage, horizontal refraction, diffraction, and scattering by the seamounts. A towed hydrophone array, with 64 sensors cut for 250Hz (3m spacing), was used to measure the signals transmitted from the aforementioned broadband sources at many locations around the Kermit-Roosevelt and Elvis seamounts. Utilizing the measured broadband signals from the towed array, the size of the shadow zone was obtained. The measured data in the BASSEX experiment strongly support the understanding of the complicated phenomena of sound propagation around the seamounts. In addition, the experimental data could be used to validate current 2D and 3D theoretical models and develop new models to properly realize the sound propagation with such complicated phenomena.In this thesis, the reconciliation between the measured pulse arrivals from the BASSEX experiment and various two-dimensional (2D) and three-dimensional (3D) theoretical models is carried out to investigate the physical characteristics of the sound propagation around seamounts: First, the 2D Parabolic Equation (PE) model and the 2D ray tracing model are used to reconcile each ray arrival with the BASSEX experiment in terms of arrival time and grazing angle. We construct a sound speed field database based on the sound speed profiles from the BASSEX experiment, World Ocean Atlas (WOA)2005, and CTD casts using the objective analysis.Second, 3D broadband sound propagation around a conical seamount is 4 investigated numerically using the 3D spectral coupled-mode model (W. Luo, PhD Thesis, MIT, 2007). Since the calculation of 3D broadband pulses with the spectral coupledmode model requires extensive computation time, a parallel program is developed with a clustered computing system to obtain results in reasonable time. The validation of the 3D spectral coupled-mode model is performed by a series of comparisons between the various 2D and 3D models for a shallow-water waveguide. The Kermit-Roosevelt seamount is modeled by a simple conical seamount for the 3D model. The computed 3D broadband pulses for the modeled conical seamount are compared with those from the BASSEX experiment and the 2D PE simulation.Through this analysis, we examine the limit of the application of the sound propagation models and improve the efficiency of the 3D sound propagation model using parallel computing to obtain a broadband pulse in a reasonable amount of time.

The SPICEX-LOAPEX-BASSEX experiments measured forward scattering from the Kermit-Roosevelt seamounts. LFM and M-sequence signals from 250-Hz moored sources and a 75-Hz ship-deployed source were received with Penn State University’s 162-element FORA array. Receptions were obtained from these sources at ranges of hundreds to 1600 km. Previous work [J. Acoust. Soc. Am. 117, (2549)] discussed the BASSEX forward shadow and patterns of horizontal refraction around the seamount. Additional work [J. Acoust. Soc. Am. 118, (1936)] compared experimental results with theory using a conical section model of a seamount. This presentation concerns the pulse compressed and array processed time series of M-sequence signals in the forward scattered field of the seamounts. Both travel times and angles of arrival are reconciled with 2-D ray-trace models. Results are comparable to refracted rays, which mainly propagate over the seamount. These, however, do not explain additional echoes, which appear in the experiment, possibly due to reflection and/or diffraction. Normal mode and PE methods are used to confirm ray models and identify additional echoes.