A Bayesian approach to long-range source localization in the complex propagation environment across the Gulf Stream front

Résumé fourni par la source

Long-range low-frequency acoustic propagation across strong oceanographic fronts presents a challenging source localization problem. During the spring 2023 pilot of the New England Seamounts Calibrated Acoustic Fluctuation Experiment (NESCAFE), a source-receiver mooring pair was deployed on either side of the Gulf Stream with a separation of roughly 150 km. The receiver mooring recorded 23 transmissions from the source mooring, which was at an unknown depth, on two large aperture vertical arrays. The 23 received time fronts exhibit a rich multipath structure with individual arrivals undergoing between 6 and 17 surface and bottom interactions. Arrival time and angle measurements, extracted through deconvolution and beamforming, capture key propagation physics but are subject to ambiguity arising from complex multipath arrivals and environmental uncertainty. We propose a Bayesian estimation method which efficiently solves for an estimate of the posterior mean and variance using a Gaussian observation model of acoustic arrival times and angles. This probabilistic formulation enables the coherent integration of multiple boundary-interacting arrivals and provides an uncertainty-aware approach to localization in a highly complex propagation environment. The results highlight the information inherent in deep-sea multipath arrivals and the utility of Bayesian framing for long-range acoustic source localization across dynamic ocean fronts.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.