Underwater ambient sound in tropical cyclones

Underwater ambient sound measurements were made in three tropical cyclones: Hurricane Gustav (2008) in the Gulf of Mexico, and Typhoons Fanapi (September 2010) and Megi (October 2010) in the western Pacific Ocean as part of the ITOP (Impact of Typhoons on the Ocean in the Pacific) program.  Measurements were made from eight Lagrangian floats, each equipped with one hydrophone, air deployed ahead of these storms by WC-130J aircraft operated by the U.S. Air Force Reserve 53rd Weather Reconnaissance squadron Hurricane Hunters.  Floats 50 and 51 were in Gustav, 60, 61 and 62 in Fanapi, and 66, 67 and 68 in Megi. After the storm passage, the Lagrangian floats were recovered by a research vessel.  Float positions were determined by interpolating between a few GPS positions taken during the storm passage guided by time-integrated velocity measurements from Electromagnetic Autonomous Profiling Explorer (EM-APEX) floats deployed at the same time.

During the passage of tropical cyclones, the hydrophone switched between the work and sleep modes every 30 minutes due to limited data storage.  In the work mode, the hydrophones sampled underwater ambient sound twice per second.  The sound measurements thus are in 30-min segments.  There are 39 (50), 38 (51), 60 (60), 56 (61), 56 (62), 55 (66), 53 (67) and 51 (68) segments (float serial numbers are in brackets), respectively, giving a total of 408 data segments and about 190 hours of sound measurements.  Each raw time series has been Fourier transformed to obtain a power spectrum from 40 Hz to 50 kHz, with a spectral resolution of 40 Hz.  The sound pressure level (SPL) in decibels (dB) is defined as \(\textrm{SPL} = 20\cdot \textrm{log}(\textrm{P}/\textrm{P}_\textrm{ref}),\) where \(\textrm{P}\) is the hydrophone measured sound pressure, and \(\textrm{P}_\textrm{ref}\) is the reference pressure 1 \(\mu \textrm{Pa}^{2} \textrm{Hz}^{-2}\).  The hydrophones were inter-calibrated in laboratory before and after the deployments, and agreed with a root-mean-square difference of 1–2 dB. The raw sound measurements are labeled SpdbP_raw.

Each Lagrangian float carried a variety of instruments including a pumped CTD (conductivity, temperature and depth) sensor, a pumped GTD (gas tension device), a motor to control drogue, and another motor to control the float's buoyancy.  These instruments generated noise of different temporal and spectral features.  Sound measurements contaminated by noise were removed as described in Zhao et al. (2014 JPO). One exception is GTD, which ran for 90% of the time for floats 50 and 51 (Gustav) and 66, 67, and 68 (Megi), and caused significant contamination on the < 5-kHz sound data.  However, the > 5-kHz sound measurements are not affected by the GTD noise, and thus kept for future studies (detect rain events and breaking waves). The cleaned sound measurements are labeled SpdbP_clean.

We decomposed the underwater ambient sound into three components according to their time scales.  First, we calculate background sound, defined as the mean of the lowest 10% sound level over the 30 min period.  The background sound generally rises and falls with increasing/decreasing wind speed and the presence of bubble clouds.  Second, sound fluctuation is obtained by removing the background sound from the original data. Third, the sound fluctuation is divided into two components using two matched temporal filters. The second-scale fluctuation is obtained by high-pass filtering the sound fluctuation using 20-second running mean.  The minute-scale fluctuation is obtained by low-pass filtering the sound fluctuation. By this method, the original underwater ambient sound is decomposed into three components: background (SpdbP_background), minute-scale fluctuation (SpdbP_MidFreq), and second-scale fluctuation (SpdbP_HighFreq).  We applied the above decomposition method to all 408 30-min sound segments from eight Lagrangian floats, and created 408 figures with the same format. We share here the raw, de-noised, and decomposed sound data for all eight Lagrangian floats (eight Matlab data files) and demonstrate their decomposed sound data (eight PDF files). Fig-5-Q101623.pdf  and Fig-6-J091801.pdf are Figures 5 and 6 in a recent paper (https://journals.ametsoc.org/view/journals/atot/aop/JTECH-D-22-0078.1/JTECH-D-22-0078.1.xml).