Analysis of underwater noise due to hurricanes in the northern Gulf of Mexico

Abstract

The proposed study will use broadband deep-water ambient sound field information collected by stationary and mobile passive acoustic monitoring platforms during a hurricane passage in the northern Gulf of Mexico. EARS buoys will be used for stationary data collection and reconnaissance flights of Slocum gliders equipped with hydrophones for mobile collection. The goal is to use these data to reconstruct storm wind speed distributions and to quantify changes in the ocean environment during hurricanes and tropical storms. The majority of this research is the signal processing involved in analyzing the collected noise data. Signal processing techniques employed include Fourier transform frequency analysis, Wavelet transform signal analysis, Bayesian signal processing, and machine learning, all of which have been applied to similar acoustic data in the past. Machine learning techniques are used to compare and correlate analyses from both bottom-moored EARS and moving glider data. Comparisons of data from gliders and from stationary EARS buoys can be made with similar analyses of recorded stationary EARS buoy data from the same and nearby locations over several recent years, now with a focus on estimating wind speeds from acoustic noise analysis. [This material is based upon the work supported by the Office of the Under Secretary of Defense for Research and Engineering under Award No. FA9550-21-1-0215.]