The Ambient Sound Data from the ATLAS Mooring: Acoustic Measurements of Wind and Rain

Jeffrey A. Nystuen
Applied Physics Laboratory
University of Washington
Seattle, Washington

Dr. Jeff Nystuen and the Acoustic Rain Gauge (ARG)

Abstract Background Data Description Results Summary Data Access and Contacts FTP Site Points of Contact

Abstract: In the frequency band from 500 - 50,000 Hz, the underwater sound field can be used to identify and quantify physical processes at the ocean surface. These include quantification of wind speed, detection, classification and quantification of rainfall, and the detection and quantification of near-surface bubble layers. During the South China Sea Monsoon Experiment (SCSMEX), two Acoustic Rain Gauges (ARGs) were mounted at 20 m and 22 m depths on a mooring at 20° 22.2' N, 116° 31.2' E. The mooring deployment lasted from April 7, 1998 - June 5, 1998. The acoustic data are presented and analyzed to document rainfall and wind conditions at the mooring. Rainfall statistics are in agreement with limited surface rainfall data. The detection of statiform rainfall occurred less often than expected. Several instances of man-made noise were detected including the apparent pirate attack on May 6, 1998.

Background:

During 1998, a major international field experiment, the South China Sea Monsoon Experiment (SCSMEX) took place. The goal of SCSMEX was to better understand the key physical processes for the onset, maintenance and variability of the monsoon over Southeast Asia and southern China (Lau, 1998). Precipitation observations were a principal part of this experiment and included a dual Doppler radar array, land-based rain gauges and an ATLAS ocean mooring at 20° 22.2' N, 116° 31.2' E, in the northern part of the South China Sea near Dongsha Island. Data from the ATLAS mooring were collected from April 7 - June 6, 1998, although on May 6, 1998 pirates vandalized the mooring and stole the surface instrumentation. The ARGs were not detected and continued to collect data.

The rain gauge mooring was an ATLAS mooring deployed by the National Oceanic and Atmospheric Administration (NOAA - USA) using their new instrumentation package (Milburn et al., 1996). Two Acoustic Rain Gauges (ARGs) were deployed on the ATLAS mooring 20 m below the ocean surface. The two ARGs were mounted on the mooring using two different mounting systems. One of the mounting systems had higher internal noise. During periods of high underwater ambient levels, including rain periods, both ARGs reported the same sound levels. The data reported here are from the ARG with the lower internal noise, ARG B-434.

The ARGs consist of an ITC-1032 hydrophone, signal preamplifers and a recording computer (Tattletale-8). Band-pass filters are present to reduce saturation from low frequency sound (high pass at 2000 Hz) and aliasing from above 50 kHz (low pass at 40 kHz). The ITC-1032 hydrophone sensitivity also rolls off above its resonance frequency, about 40 kHz. The equivalent oceanic background noise level of the pre-amplifier system is about 37 dB relative to 1 m Pa²Hz^-1. A data collection sequence consisted of four 1024 point time series collected at 100 kHz (10.24 ms each) separated by 5 seconds. Each time series was fast Fourier transformed (FFT) to obtain a 512-point (0-50 kHz) power spectrum. These four spectra were averaged together and spectrally compressed to 36 frequency bins, with frequency resolution of 1 kHz from 1-20 kHz and 2 kHz from 20-50 kHz. A correction for the system frequency sensitivity has been applied to the data.

Instrument noise, from the Tattletale-8, was present in the data, but was confined to a few frequency bins. This noise was identified for each ARG in the time period just prior to deployment. The contaminated frequency bins for ARG B-434 are 4400, 8300, 12200, 17100, 38600 and 50300 Hertz. Data from these frequency bins has been removed

The data sampling strategy varied depending on the geophysical conditions present. Each recorded spectrum was evaluated to determine the likely geophysical source. The categories were: wind only, rain, drizzle or noise. If "wind only" conditions were detected, the next acoustic sample was 5 minutes later. If "drizzle" was detected, the next sample was 1 minute later and if "rain" was detected the next sample was 30 seconds later. Thus, the time step for the data reported here is not uniform.

Two types of data files are presented in this archive. The first is the processed sound level (SL) data at 30 frequency bins. The frequencies are given in variable FREQ and Table 1. The sound levels are presented in units of decibels (dB) relative to 1 m Pa² Hz^-1. The data are split into 8 sections for ease of analysis. The start and end of each time period are given in Table 2. These files are named SL1 through SL8 (Sound Levels 1 through Sound Levels 8). The data have been adjusted for system sensitivity and the frequency bins with high instrument noise, as discussed above, have been deleted.

The second type of file is the acoustic interpretation of the sound level (SL) data. Acoustic weather analysis (AWA) has been applied to the SL data (Nystuen and Selsor, 1997). AWA identifies the sound source (wind, rain, drizzle or noise) and then quantifies the wind speed or the rainfall rate if appropriate. The sound source is identified by an acoustic weather code. These codes are given in Table 3. Three noise codes, based on the shape of the underwater sound spectrum, are used. These are "shipping", indicated by relatively high low frequency noise, "unknown", indicated by relatively high noise levels with a relative minimum at 20 kHz, and "spikes", with high levels in narrow frequency bands. Each of these "noises" is inconsistent with known spectra of geophysical noise (wind, rain, drizzle) and is therefore called noise for these data.

Once a geophysical sound source (wind, rain or drizzle) is identified, it is quantified. For wind speed, the algorithm of Vagle et al. (1990) is applied using the sound level at 10.3 kHz. This algorithm is given by

where SL₁₀ is the sound level at 10 kHz and U₁₀ is the estimated wind speed at 10 m. Acoustic wind speed estimates are not available during periods of rain, Weather Code = 1. When the AWA code indicates rain or noise, the wind speed estimate is set to zero. This is "NO DATA", not a zero wind speed. Furthermore, the instrument noise is equivalent to a sound level of 37 dB relative to 1 m Pa² Hz^-1 which corresponds to a wind speed of 4 m/s. Thus, the minimum wind speed reported by the ARGs is 4 m/s. Finally, the wind speed estimate may be contaminated by undetected rain or drizzle. Wind speed measurements near to times when rain is detected are most likely to be contaminated. An estimate of wind speed is made during "drizzle" conditions, Weather Code = 2. The likelihood of contamination is high when the Weather Code = 2.

The rainfall rate is quantified using two different acoustic algorithms. The first is an inversion of the sound field to identify the drop size distribution in the rain (Nystuen, 1996). Once the drop size distribution is obtained, rainfall rate and equivalent radar reflectivity can be calculated directly from the drop size distribution. A second, empirical algorithm based on the sound level in the frequency band from 4-10 kHz can also used to estimate rainfall rate. This algorithm in given by

where SL_4-10 is the sound level from 4-10 kHz and R is the rainfall rate in mm/hr.

These processed data files are presented for the same 8 sections identified in Table 2. These files are named AWA1 through AWA8 (Acoustic Weather Analysis 1 through 8). Each AWA file includes the time of each measurement, the weather code, the wind speed, the rainfall rate using the drop size distribution inversion, the rainfall rate using Eq. 2 and the equivalent radar reflectivity using the drop size distribution inversion. When calculating rainfall rate, Eq. 2 is inappropriate for drizzle, Weather Code = 2. When the Weather Code = 2, the rainfall rate has been arbitarily set to R = 1 mm/hr. The format for the AWA files is summarized in Table 4.

Table 5 shows a daily summary of rainfall events detected by the ARG. Overall, non-geophysical noise (Weather codes 3, 4 and 5) was detected in 7.1% of the data. Heavy rain was detected 1.3% of the time. Drizzle, or stratiform rain, was detected 1.4% of the time. The heaviest rainfall event occurred on April 19th (Julian Day 109) with an estimated 101 mm of rain. The monsoon onset was detected on May 15th (Julian Day 135).

The underwater sound spectra recorded by an acoustic rain gauge (ARG-B434) mounted on the ATLAS mooring at 22 m depth are presented in two forms. The underwater sound levels (SL files) at 30 frequency bands are presented in 8 sections. These data files are named SL1, SL2, SL3, SL4, SL5, SL6, SL7 and SL8. The time periods covered by each file are given in Table 3.

The underwater sound field can be interpreted to provide quantitative measurements of wind speed and precipitation. This interpretation is given in data files AWA1, AWA2, AWA3, AWA4, AWA5, AWA6, AWA7 and AWA8. Each of these data files also corresponds to the time periods given in Table 3. Each file has a time variable, weather code, wind speed, two rainfall rate estimates and an equivalent radar reflectivity. The format is given in Table 4.

Acknowledgments: The mooring was deployed by the Pacific Marine Environmental Laboratory (PMEL) of the National Atmospheric and Oceanic Administration (NOAA). Funding for the mooring was provided by the Tropical Rain Measuring Mission (TRMM) Project Office of NASA. Permission to deploy the acoustic rain gauges on the mooring was granted by Prof. D. Tang, National Technical University in Taiwan.

References:

Lau, K.M., 1998: The South China Sea Monsoon Experiment (SCSMEX). EOS 78, 599, 603.

Milburn, H.B., P.D. McLain and C. Meinig, 1996: ATLAS buoy - Reengineered for the next decade. In Proceedings of IEEE Oceans'96, Fort Lauderdale, FL, 698-702.

Nystuen, J.A., 1996: Acoustical rainfall analysis: Rainfall drop size distribution using the underwater sound field. J.Acoust. Soc. Am. 13, 74-84.

Nystuen, J.A. and H.D. Selsor, 1997: Weather classification using passive acoustic drifters. J. Atmos. and Oceanic Tech., 14, 656-666.

Vagle, S., W.G. Large and D.M. Farmer, 1990: An evaluation of the WOTAN technique for inferring oceanic wind from underwater sound. J. Atmos. and Ocean. Tech. 7, 576-595.

Data Access and Contacts

FTP Site

The Rain Gauge data from SCSMEX resides on DISC anonymous FTP. You may access it from this document,

Acoustic Rain Gauge Data (ASCII)

or directly via FTP at

ftp disc2.nascom.nasa.gov

login: anonymous

password: < your internet address >

cd data/scsmex/rain_gauge/

Points of Contact

The Principal Investigator for the Acoustic Rain Gauge data is

Jeffrey A. Nystuen
Applied Physics Laboratory
University of Washington
Seattle, Washington
nystuen@apl.washington.edu

For information about or assistance in using DISC data, contact

Hydrology Data Support Team
EOS Data and Information Services Center (DISC)
Code 610.2
NASA Goddard Space Flight Center
Greenbelt, Maryland 20771
hydrology-disc@listserv.gsfc.nasa.gov
301-614-5165 (voice)
301-614-5268 (fax)

Last update:Fri Jan 30 08:26:15 EST 2004 Page Author: Hydrology Data Support Team -- hydrology-disc@listserv.gsfc.nasa.gov Web Curator: -- Website Curator: Stephen W Berrick NASA official: Steve Kempler, DISC Manager -- kempler@disc.gsfc.nasa.gov