Conference PaperPDF Available

Finite-amplitude sound propagation effects in fish abundance estimation

Authors:
Proceedings of the 45th Scandinavian Symposium on Physical Acoustics, Online, 31 Jan – 1 Feb 2022
Open Access
1 ISBN 978-82-8123-022-4
Finite-amplitude sound propagation effects
in fish abundance estimation
Audun O. Pedersen1, Per Lunde1,
Rolf J. Korneliussen2, Frank E. Tichy3
1 University of Bergen, Department of Physics and Technology, P. O. Box 7803,
N-5020 Bergen, Norway
2 Institute of Marine Research, P. O. Box 1870, N-5817 Bergen, Norway
3 Kongsberg Maritime AS, P. O. Box 111, N-3191 Horten, Norway
Contact email: audun.pedersen@uib.no
Extended abstract
Quantitative echosounder measurements for fish abundance estimation relies on high
accuracy in measurements of backscattered acoustic power. As scientific echosounder
systems have evolved in terms of power and accuracy, finite-amplitude sound propaga-
tion effects have become a potentially significant error source. The waveform distortion
due to finite-amplitude propagation manifests itself as excess attenuation at the funda-
mental (operating) frequency and generation of harmonic overtones. The excess attenu-
ation in echosounder measurement results, also referred to as “nonlinear loss”, has been
studied since around year 2000 [1–3]. Advice has been given to limit the transmitted
acoustic power so that the nonlinear loss would be insignificant [4, 5]. The advice how-
ever fails to indicate the residual nonlinear loss when the recommended power limits
are used. Nonlinear loss in echosounder sound beams have been measured experimen-
tally [1, 6–8], in both fresh water and seawater, and simulated [2, 3, 6–8] using the “Ber-
gen Code” finite difference solution [9] of the KZK second-order, parabolic equation [10,
11]. Generation of harmonics has also been studied for its implications for simultaneous
measurement at multiple frequencies [6, 12]. Quantitative experimental and numerical
data have only been obtained for 120 kHz and 200 kHz operating frequencies, while
other frequencies and wide-band signals are also important for marine research applica-
tions.
A formal framework has been established so that nonlinear loss can be described
within the power budget terminology commonly used in fisheries research [6, 13, 14].
The power budget equation used, assuming linear sound propagation, is [15]


where is the volume backscattering coefficient of the target volume at range ,
and
are the transmitted and received electrical powers at the echosounder transducer ter-
minals, respectively, is the absorption coefficient, is the axial transducer gain, is
the equivalent two-way beam solid angle, is the acoustic wavelength, is the small-
signal sound speed, is the effective signal duration, and is an electrical termination
factor.
2
To account for nonlinear loss in the forward (incident) direction, both upon echo-
sounder calibration and field measurements, the power budget equation can be
amended to [8, 13]




 
    is the axial finite-amplitude factor, equal to the ratio of the
axial sound pressure amplitude under finite-amplitude propagation conditions,
 , to the theoretical axial sound pressure amplitude   assuming no fi-
nite-amplitude effects but otherwise the same conditions. 
 is the axial finite-
amplitude factor for the case of echosounder calibration, where a reference target with
known target strength is positioned on the sound beam axis at range . The axial finite
amplitude factors are always on the interval (0, 1). 
 
is the beam solid
angle finite-amplitude factor, defined as the ratio between the equivalent two-way beam
solid angles with and without nonlinear loss. 
is greater than unity since the
sound pressure amplitude and thus the nonlinear loss are stronger on the sound beam
axis (accounted for by
) and weaker elsewhere. The applicability of this formal
framework has been tested on field survey measurements of volume backscatter from
fish schools [8]. Fair agreement is found between field measurement data and predic-
tions made with the Bergen Code. The agreement has been improved compared to ear-
lier work [6] by applying the Francois-Garrison [16] absorption formula to all frequency
components in the numerical calculations [8].
Figure 1. Calculated correction factor
  
 for nonlinear loss, assuming the transmit
power setting 1000 W upon calibration and field measurements using an EK60 echosounder sys-
tem. The calibration target is positioned on the sound beam axis at the ranges . Left: 120 kHz
operating frequency. Right: 200 kHz operating frequency.
Example simulations are shown in Figure 1 for the Simrad EK60 system with ES120-7C
and ES200-7C transducers, operating at 120 kHz and 200 kHz frequencies, respectively.
The correction factor

 


is plotted against range for various calibration ranges  when a 1000 W power setting
has been used for both calibration and subsequent measurements. Seawater parameters
3
from a survey cruise in the North Sea [6, 8] have been applied in the numerical calcula-
tions.
Measurement errors due to the nonlinear loss are largely systematic. Given a speci-
fied bound on the acceptable measurement uncertainty, illustrated as an example as
0.2 dB by coloured areas in Figure 1, one can balance the applied transmit signal ampli-
tude against the need to correct for nonlinear loss. Such considerations depend on quan-
titative knowledge of the parameters of the specific echosounders considered and of the
seawater as propagation medium.
References
[1]
F. E. Tichy, H. Solli and H. Klaveness, “Nonlinear effects in a 200-kHz sound beam and consequences
for target strength measurement,” ICES J. Mar. Sci., 60, pp. 571-574, 2003.
[2]
A. C. Baker and P. Lunde, “Nonlinear effects in sound propagation from echosounders used in fish
abundance estimation. Numerical simulation results. CMR Technical Note CMR-TN02-A10008-rev-01,
de-classified version of CMR-TN02-F10008 (2002),” Christian Michelsen Research AS, Bergen, 2011.
[3]
A. C. Baker and P. Lunde, “Nonlinear propagation from circular echosounder transducers. Numerical
simulation results. CMR Technical Note CMR-TN01-A10010-rev-01, de-classified version of CMR-
TN01-F10010 (2001),” Christian Michelsen Research AS, Bergen, 2011.
[4]
Kongsberg Maritime, “Non-linear effects. Recommendation for fishery research investigations,”
March 2002. [Online]. Available: https://www.kongsberg.com/no/maritime/products/commercial-
fisheries/ekkoloddsvingere/non-linear-effects/. [Accessed 30 November 2021].
[5]
R. J. Korneliussen, N. Diner, E. Ona, L. Berger and P. G. Fernandes, “Proposals for the collection of
multifrequency acoustic data,” ICES J. Mar. Sci., 65, pp. 982-994, 2008.
[6]
A. Pedersen, Effects of nonlinear sound propagation in fisheries research. Ph.D. thesis, Bergen:
University of Bergen, 2006. https://hdl.handle.net/1956/2158
[7]
A. Pedersen, M. Vestrheim and P. Lunde, “Non-linear sound propagation effects in fisheries research
echo sounders - Measurements and simulations in fresh water,” in Proc. 28th Scandinavian Symposium
on Physical Acoustics, Ustaoset, Norway, 2005.
[8]
A. O. Pedersen, P. Lunde, F. E. Tichy and R. J. Korneliussen, “Finite-amplitude sound propagation
effects in volume backscattering measurements for fish abundance estimation,” Acta Acustica, 6, 14,
2022.
[9]
S. I. Aanonsen, T. Barkve, J. Naze Tjøtta and S. Tjøtta, “Distortion and harmonic generation in the
nearfield of a finite amplitude sound beam,” J. Acoust. Soc. Am., 75(3), pp. 749-768, 1984.
[10]
E. A. Zabolotskaya and R. V. Khokhlov, “Quasi-plane waves in the nonlinear acoustics of confined
beams,” Sov. Phys. Acoust., 15(1), pp. 35-40, 1969.
[11]
V. P. Kuznetsov, “Equations of nonlinear acoustics,” Sov. Phys. Acoust., 16(4), pp. 467-470, 1971.
[12]
B. Khodabandeloo, E. Ona, G. J. Macaulay and R. Korneliussen, “Nonlinear crosstalk in broadband
multi-channel echosounders,” J. Acoust. Soc. Am., 149(1), pp. 87-101, 2021.
[13]
P. Lunde, “Finite-amplitude power budget equations for acoustic fish abundance estimation,” J. Mar.
Sci. Eng., 8, 98, 2020.
[14]
P. Lunde and A. O. Pedersen, “Sonar and power budget equations for backscattering of finite
amplitude sound waves, with implications in fishery acoustics for abundance estimation of marine
resources,” in 35th Scand. Symp. Phys. Acoust., Geilo, Norway, Jan 29-Feb 1, 2012.
[15]
P. Lunde, A. O. Pedersen, R. J. Korneliussen, F. E. Tichy and H. Nes, “Power-budget and echo-
integrator equations for fish abundance estimation,” Fisken og havet, 10, 2013.
[16]
R. E. Francois and G. R. Garrison, “Sound absorption based on ocean measurements. Part II: Boric acid
contribution and equation for total absorption,” J. Acoust. Soc. Am., 72(6), pp. 1879-1890, 1982.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Scientific advice for governmental management of marine resources relies on acoustical observation methods. Quantification and identification of fish and plankton species are often achieved using multi-frequency acoustic data. Accurate measurements of backscattering cross-sections and volume backscattering coefficients are essential. Systematic errors from finite-amplitude sound propagation are demonstrated in experimental survey measurements on Atlantic mackerel using 120 kHz and 200 kHz echosounders and high power settings. Finite-amplitude signal distortion causes excess transmission loss that is not accounted for in fisheries acoustics today, other than by fixed limits on the maximum transmitted power. The demonstrated errors are of a magnitude that can seriously bias abundance estimation and species identification. It is shown how the finite-amplitude effects can be modelled and predicted quantitatively, within a framework of electroacoustic power budget equations. A method is provided to calculate related errors in echosounder calibration and oceanic measurement of acoustic volume backscattering. When accounting for finite-amplitude effects in echosounder signal processing, higher transmit powers can be used when needed to improve signal-to-noise ratio or extend measurement range. The results indicate that historical survey data can be adjusted for such errors using numerical simulations. The echosounder characteristics relevant to finite-amplitude effects can be determined by laboratory measurements.
Article
Full-text available
Finiteamplitude (nonlinear) sound propagation effects in seawater may cause measurement errors in fish and zooplankton abundance estimation and species identification for accessible echo sounder transmit electrical power levels and operating frequencies of about 100 kHz and higher. A sufficiently validated framework to quantify, control, and compensate for such errors in these applications is not available. The conventional power budget equations in fisheries acoustics are valid for smallamplitude signals only. The study aims to fill this “gap”. The conventional theory is generalized to account for finiteamplitude incident sound propagation, arbitrary electrical termination, and the range of electrical and acoustical echo sounder parameters. Equations for use in calibration and oceanic surveying are derived in terms of the backscattering cross section,, and the volume backscattering coefficient,. The “finiteamplitude terms” in these expressions can—for relevant transmit electrical power levels of relevant echo sounders—be measured in controlled tank experiments. Alternatively, they can be calculated using numerical models. The resulting equations enable estimation of finiteamplitude measurement errors in these applications; development of recommended upper limits for echo sounder power levels; controlled reduction of finiteamplitude errors in calibration and surveying; and development of correction factors for survey data already subjected to such measurement errors.
Conference Paper
Full-text available
Fax number +47 55574041. Audun.Pedersen@cmr.no. Abstract: In estimation of fish stock abundance and discrimination between species, the effect of excess attenuation due to nonlinear sound propagation in water has been disregarded until recent years. It has been shown earlier that this may be significant to target strength measurements along the axis of the sound beam. Nonlinear distortion depends on several factors such as amplitude, frequency, propagation distance, temperature, salinity, bubble and particle contents of the water. In the present paper, measurement errors due to nonlinear sound propagation from fisheries research echo sounders are studied quantitatively. Measurements are at this stage made in fresh water, using a calibrated hydrophone and a calibration sphere. Simulations based on a finite difference solution of the KZK equation are compared with the measurement results.
Article
Full-text available
Korneliussen, R. J., Diner, N., Ona, E., Berger, L., and Fernandes, P. G. 2008. Proposals for the collection of multifrequency acoustic data. – ICES Journal of Marine Science, 65: 982–994. Acoustic surveys are used to estimate the abundance and distribution of many fish species, and have been based traditionally on data collected at a single acoustic frequency. Although it has been known for some time that the use of additional frequencies can provide information on the nature of the acoustic target, the knowledge and technology required to combine the so-called “multifrequency data” in an appropriate manner has been limited. The use of several transducers of different frequencies is now common on board research vessels and fishing vessels, so multifrequency data are often collected. In order for these data to be combined appropriately, their physical and spatial characteristics from each frequency should be as similar as possible. We detail the requirements deemed necessary to collect multifrequency data in an appropriate manner. They can be stringent and may not always be achievable, so we also consider the consequences of combining acoustic data originating in transducers with varying degrees of spatial separation and with different beam widths.
Article
Distortion of acoustic wave caused by nonlinear propagation transfers acoustic energy into higher harmonics of the transmitted signal. When operating several broadband echosounders with non-overlapping frequency bands to cover a wide frequency range, higher harmonics generated by one band may interfere with the fundamental band of others. This interference (i.e., crosstalk) can adversely affect the measured backscattered amplitude frequency response and in some circumstances, appears as spurious targets above and/or below the main target in pulse-compressed echograms. The nonlinear propagation of frequency-modulated acoustic waves in a directional beam was modeled and used to assess methods to reduce the deleterious effects of harmonic components in the signal, and was also compared to field experiments using the seabed echo and a metallic target sphere, with good agreement. Two methods are shown to materially reduce crosstalk: (1) reduction in transmit power, which reduces crosstalk amplitude by a larger amount than the associated reduction in transmit power, and (2) selection of a proper Fourier window length in the processing stage. The effect of crosstalk was small (<0.4 dB or 10%) for area backscattering measurements, but could be several dB for target strength measurements at different frequencies, depending on the transmit signals and processing parameters.
Article
Distortion and harmonic generation in the nearfield of a finite amplitude sound beam are considered, assuming time‐periodic but otherwise arbitrary on‐source conditions. The basic equations of motion for a lossy fluid are simplified by utilizing the parabolic approximation, and the solution is derived by seeking a Fourier series expansion for the sound pressure. The harmonics are governed by an infinite set of coupled differential equations in the amplitudes, which are truncated and solved numerically. Amplitude and phase of the fundamental and the first few harmonics are calculated along the beam axis, and across the beam at various ranges from the source. Two cases for the source are considered and compared: one with a uniformly excited circular piston, and one with a Gaussian distribution. Various source levels are used, and the calculations are carried out into the shock region. The on‐axis results for the fundamental amplitude are compared with results derived using the linearized solution modified with various taper functions. Apart from a nonlinear tapering of the amplitude along and near the axis, the results are found to be very close to the linearized solution for the fundamental, and for the second harmonic close to what is obtained from a quasilinear theory. The wave profile is calculated at various ranges. An energy equation for each harmonic is obtained, and shown to be equivalent within our approximation to the three‐dimensional version of Westervelt’s energy equation. Recent works on one‐dimensional propagation are reviewed and compared.
Article
When estimating target strength (TS), sound pressure in the beam where the target is located has to be measured accurately. Sound pressure is normally calculated from the source level, and transmission loss is based on geometric spreading and absorption loss. Additional losses caused by non-linear acoustic propagation may be important, especially in the case of high-power, high-frequency, and highly directive sources. ‘Non-linear loss’ from the fundamental frequency is due to energy from the fundamental harmonic being transferred into higher harmonics. This loss affects the beam pattern in ways that will depend on both power and range, since the non-linear loss depends, in turn, on sound pressure. We present the results of sound-pressure measurements and simulations from a 200-kHz transducer with a beam width of 7°. Sound pressure was measured at different ranges and power levels using a broadband hydrophone to detect some of the higher harmonic frequencies that occur in non-linear acoustic propagation. The TS of a solid copper sphere was measured using a standard echosounder with no correction for non-linear loss. Our study illustrates that this can be significant and that a corresponding correction needs to be considered when estimating TS.
Article
The absorption of sound in seawater is considered as the sum of three contributions: those from pure water, magnesium sulfate, and boric acid. Contributions from other reactions are small and are not included. The pure water and magnesium sulfate contributions obtained from analyses of extensive oceanic measurements, including many in the Arctic, were discussed in Part I. In Part II, an analysis is made of all reported measurements in the low-frequency region (0. 2-10 kHz) to evaluate contribution of boric acid. This is done by subtracting the pure water and magnesium sulfate contributions determined in Part I from the total absorption to give a more accurate estimate of the boric acid contribution than previously obtained. The three contributions are then combined to form an equation with both a theoretical basis and a satisfactory empirical fit that will be useful to researchers and engineers in the field of underwater sound. The equation applies to all oceanic conditions and frequencies from 200 Hz to 1 MHz.