Speaker
Description
Three-dimensional acoustic propagation in the open ocean for low frequency sound is an observed phenomenon (Heaney and Murray, JASA 2008), and is known to lead to biases in the localisation of long-range sources (Munk et. al., JASA, 1988). In this paper, a high-fidelity fully 3D global acoustic Parabolic Equation model is applied to the problem to demonstrate the impact of 3D acoustic propagation due both to bathymetric diffraction/refraction and to mesoscale oceanographic variability. The mesoscale/seasonal angle-of-arrival variability (for an open water path) for a 10,000 km source/receiver range through dynamic oceanography is shown to be on the order of 2 degrees over the annual cycle. For regions with traditional bathymetric blockage, 3D propagation exhibits significant diffraction, leading to a large area of the ocean that is covered by the hydroacoustic network, but is currently considered shadowed. Three-dimensional back-propagation from each hydroacoustic station can be computed and used as a look-up table to directly improve the localisation accuracy of hydro-acoustic only detections. Inclusion of these maps in the Bayesian localisation approach used at CTBTO should lead to significant reduction in the area of uncertainty associated with hydro-only detections and localisations, permitting a significant improvement in the sensor networks overall detection/classification/localisation capability.
