Multi-Frequency Acoustics


A combination of multi-frequency acoustics and midwater trawl samples are used to estimate the biomass and distribution of mesopelagic and epipelagic (sardine and anchovy) fishes as well as macrozooplankon and identify areas of enhanced productivity and concentration in the southern California Current.


1. Principle

The use of sound to remotely detect aquatic organisms has been used for more than half a century. As a non-invasive technique, the use of sound provides a real-time high resolution technique to study aquatic organisms, both quantitatively and qualitatively, with the capacity to reveal complex dynamics of the local biology that include their response to micro- and meso-scale oceanographic features, such as fronts (Horne 2000, Lavery et al. 2002). The methods and potential to identify species remotely has improved as the technology has evolved, with the addition of correction factors, development of calibration procedures (Foote et al. 1987), and studies of the relationship between echo amplitude and size (Foote 1987), material properties of organisms (Chu et al. 2000, Chu & Wiebe 2005) and orientation with respect to the acoustic beam (Demer & Martin 1995). A multi-frequency acoustic system takes advantage of the differential response of targets of different sizes and composition to different wave lengths (l), thus enabling us to discriminate small organisms (i.e. zooplankton) from large ones (i.e. fish) and fish with air-filled swimbladders from those without (Simmonds & MacLennan 2005).

2. Data Collection

The multi-frequency acoustic system utilized is a hull mounted SIMRAD EK-60 with 5 frequencies: 18 kHz, 38 kHz, 70 kHz, 120 kHz, 200 kHz.


The EK-60 is calibrated at the beginning of each survey using the standard sphere method (Foote et al, 1987), and collects data throughout the survey.


The transceivers’ operational parameters applied are:

Transmit power

18 kHz2000W
38 kHz2000W
70 kHz1000W
120 kHz300W
200 kHz100W

Pulse duration = 1.024 ms for all frequencies
Ping Rate (sample rate)=1 ping every 2 s
No thresholds are applied when collecting data.


To reduce noise while collecting data the following procedures are used:

  1. The transceivers’ grounds are connected to the ship’s ground to stop short noise
  2. A low-pass filter is added to stop conducted signal noise
  3. Ferrite beads are mounted around the cables to suppress high frequency noise from radar and telecommunication systems
  4. Transducer cables are wrapped with aluminum foil to shelter cables from radiation
  5. A power board is used with Electro Magnetic Interference (EMI) and Radio Frequency Interference (RFI) filters
  6. A UPS is used to connect to ship’s clean power supply
  7. The Acoustic Doppler Current Profiler is slaved to avoid cross talk between acoustic systems


Data are collected in GMT and stored into raw data files every 45 minutes, each 1.025MB size.


Events such as power failure, noise detected in the echogram, or interesting acoustic signals are recorded in a logbook.

3. Data Filtering


Acoustic raw data files are organized by transect and station.


EV files are created using Myriax Software’s Echoview ®. Data collected when the ship is steaming and data collected while on station are stored into different EV files.


Calibration parameters, orientation and position of the EK-60 instrument are input into each EV file.


Environmental data such as sound speed and sound absorption per frequency are calculated using temperature and salinity data collected with the CTD during the survey. A weighted mean (Demer 2004) of sound speed and absorption is then applied to each EV file depending on the location of the station where CTD data were collected.


All noise cleaning and further data processing are carried out on the EV files using virtual echograms created in Echoview ®.


Data collected in the first 10 m and near the bottom are eliminated due to vessel noise produced by the propellers and contamination of data due to bottom backscatter using surface and bottom exclusion lines.


To filter out non-biological (“noise”) signals stronger than the “background-noise” level (noise spikes, periodic noise from ship’s winch) that exists over a significant portion of any given ping, a series of virtual variables are utilized to eliminate the entire ping or the portion of the ping that is affected by this type of noise. This type of noise can be caused by turbulence, breaking waves and cavitations and it is usually stronger than the background noise and biological signals.


To eliminate background noise, an estimate of the background noise level at 1 m is done by scrutinizing the noise level in the data. This is based on the assumption that background noise remains constant over the transmit-receive cycle (De Robertis & Higginbottom 2007). A Time Varied Gain (TVG) curve is then generated and subsequently subtracted from the data; this is because the application of the TVG correction factor amplifies this type of noise. Background noise is an unwanted signal (other than reverberation) present at the receiver output even when the transmitter is turned off and includes ambient ocean noise and electrical noise.


False bottom in the two lowest frequencies was eliminated from echograms.


Pings and data deemed “bad” that were removed during the cleaning process are replaced by an average of the surrounding data using convolution techniques. Data affected more than 40% by noise are discarded from further analyses.


Cleaned EV files are then exported into .hac files for optimization and further processing.

4. Data Analyses

The analyses of the multi-frequency acoustic data are currently under development. A variety of methods are being explored to find those that best suit our needs in Southern California.

The methods that are being currently assessed are:

a) The use of theoretical models that encompass organismal size, material properties and orientation to predict the echo amplitude reflected by the different types of organisms present in the region. These models include Approximate Deformed Cylinder and Prolate Sphere models for midwater fish (Yasuma et al. 2003), Distorted Wave Born Approximation for small fluid-like zooplankton (Chu et al. 1993, Stanton et al. 1993), and Fluid Sphere for strong zooplankton scatterers (Anderson 1950). It is essential to collect biological samples to determine the species composition and size distributions to use this method.

b) Mean Volume Backscattering Strength difference between frequencies is a different method that uses only acoustic data and uses information on the frequency response from different organisms, i.e. zooplankton versus fish (Madureira et al. 1993, Miyashita et al. 1997). This method is mainly used for preliminary analyses.

c) Inverse approach, which uses the measured backscatter from all frequencies and extracts information on the numbers by size and characteristics (i.e. presence of gas) of organisms in the area ensonified (Holliday 1977, Holliday & Pieper 1995). It is important to know the composition of the organisms responsible for the backscatter.

5. Equipment/Software

  • Multifrequency acoustic system EK-60 with 5 frequencies (18kHz, 38kHz, 70 kHz, 120 kHz, 200kHz)
  • Echoview ®
  • Matlab ®
  • Microsoft Office Suite

8. References

  • Anderson, V.C. 1950. Sound scattering from a fluid sphere. Journal of the Acoustical Society of America 22: 426-431.
  • Chu, D., K.G. Foote, and T.K. Stanton. 1993. Further analysis of target strength measurements of Antarctic krill at 38 and 120 kHz: Comparison with deformed cylinder model and inference of orientation distribution. Journal of the Acoustical Society of America 93: 2985-2988.
  • Chu, D. and P.H. Wiebe. 2005. Measurements of sound-speed and density contrasts of zooplankton in Antarctic waters. Ices Journal of Marine Science 62: 818-831.
  • Chu, D., P. Wiebe, and N. Copley. 2000. Inference of material properties of zooplankton from acoustic and resistivity measurements. Ices Journal of Marine Science 57: 1128-1142.
  • De Robertis, A. and I. Higginbottom. 2007. A post-processing technique to estimate the signal-to-noise ratio and remove echosounder background noise. Ices Journal of Marine Science 64: 1282-1291.
  • Demer, D.A. 2004. An estimate of error for the CCAMLR 2000 survey estimate of krill biomass. Deep-Sea Research Part II 51: 1237-1251.
  • Demer, D.A., and L.V. Martin. 1995. Zooplankton target strength – volumetric or areal dependence. Journal of the Acoustical Society of America 98: 1111-1118.
  • Foote, K.G. 1987. Fish target strengths for use in echo integrator surveys. Journal of the Acoustical Society of America 82: 981-987.
  • Foote, K.G., H.P. Knudsen, G. Vestnes, D.N. MacLennan, and E.J. Simmonds. 1987. Calibration of acoustic instruments for fish density estimation: A practical guide. 69 pgs.
  • Holliday, D.V. 1977. Extracting bio-physical information from the acoustic signatures of marine organisms. In Ocean Sound Scattering Prediction. N.R. Anderson and B.J. Zahuranec (Eds.) Marine Science Series 5: 619-624. Plenum Press, New York, NY.
  • Holliday, D.V. and R.E. Pieper. 1995. Bioacoustical oceanography at high frequencies. Ices Journal of Marine Science 52: 279-296.
  • Horne, J.K. 2000. Acoustic approaches to remote species identification: a review. Fisheries Oceanography 9: 356-371.
  • Lavery, A.C., T.K. Stanton, D.E. McGehee, and D.Z. Chu. 2002. Three-dimensional modeling of acoustic backscattering from fluid-like zooplankton. Journal of the Acoustical Society of America 111: 1197-1210.
  • Madureira, L.S.P., I. Everson, and E.J. Murphy. 1993. Interpretation of acoustic data at 2 frequencies to discriminate between Antarctic krill (Euphausia superba Dana) and other scatterers. Journal of Plankton Research 15: 787-802.
  • Miyashita, K, I. Aoki, K. Seno, K. Taki, and T. Ogishima. 1997. Acoustic identification of isada krill, Euphausia pacifica Hansen, off the Sanriku coast, north-eastern Japan. Fisheries Oceanography 6: 266-271.
  • Simmonds, J. and D. MacLennan. 2005. Fisheries Acoustics: Theory and Practice, 2nd edition, Blackwell Publishing, Oxford.
  • Stanton, T.K., D. Chu, P. Wiebe, and C.S. Clay. 1993. Average echoes from randomly oriented random-length finite cylinders: Zooplankton models. Journal of the Acoustical Society of America 94: 3463-3472.
  • Yasuma, H., K. Sawada, T. Olishima, K. Miyashita, and I. Aoki. 2003. Target strength of mesopelagic lanternfishes (family Myctophidae) based on swimbladder morphology. Ices Journal of Marine Science 60: 584-591.