UC San Diego

Deep Sound is an un-tethered, free-falling acoustic platform designed to profile the ambient noise field in the ocean from the surface to a pre-programmed depth, at which point a ballast weight is dropped and the instrument returns to the surface under its own buoyancy. Three iterations of the instrument, Mk I, II and III, have been designed, built and tested, the first two rated to descend to 9 km and the third to a full ocean depth of 11 km. During a deployment of the instrument, vertically and horizontally spaced hydrophones continuously record the ambient noise pressure time series over a large bandwidth (5 Hz - 40 kHz), returning the power spectral density, vertical and horizontal coherence as a function of depth. Deep Sound Mk I and Mk II have been deployed down to 9 km depth in the Mariana Trench and Mk I has descended three times to 5 km, 5.5 km and 6 km in the Philippine Sea. The data reported here examines the depth-dependence of the power spectrum, vertical coherence and directionality of rain and wind noise in the Philippine Sea. Acoustic estimates of rainfall rates and wind speeds are made from the surface to 5.5 km and 6 km respectively and compared to surface meteorological measurements. The depth- dependence of the accuracy of these estimates is relatively small and found to improve with depth. A coherence fitting procedure is employed to return ambient noise directionality and provide information on the spatial variability of an overhead rainstorm. With moderate 7-10 m/s winds, downward propagating noise from directly overhead dominates the noise field directionality from the surface to 6 km. Using the wind generated surface noise and the depth dependence of the spectral slope over the band 1 - 10 kHz, the frequency dependence of the absorption due to sea water is estimated and used to infer a mean water column value of pH