Acoustic Scattering by Marine Organisms☆

KG Foote, Woods Hole Oceanographic Institution, Woods Hole, MA, United States
© 2018 Elsevier Inc. All rights reserved.

This is an update of K.G. Foote, Acoustic scattering by marine organisms, encyclopedia of ocean sciences (2nd edn.). J. H. Steele (ed.), Academic Press, 2001,
pp. 62–70.

Historical Overview 2
Recent Trends 3
Acoustic Scattering Studies 3
Mensuration 3
Sonar Technology 3
Platforms 3
Physical Basis for Scattering 4
Classification of Marine Organisms as Scatterers 4
Dependences of Scattering 4
Intrinsic Factors 4
Extrinsic Factors 4
Quantification of Scattering 4
Nomenclature 4
Measurement 6
Modeling 6
Fish as Scatterers 6
Swimbladder-Bearing Fish 6
Low Frequencies 6
Single-shell spherical model 6
Double-shell spherical model 7
Applications to swimbladdered fish 7
Intermediate Frequencies 8
High Frequencies 9
Swimbladderless Fish 9
Zooplankton as Scatterers 9
Liquid-Like Bodies 9
Homogeneous Liquidlike Bodies 10
Inhomogeneous Liquidlike Bodies 10
Hard-Shelled Bodies 10
Gas-Bearing Bodies 10
Other Organisms as Scatterers 10
Squid 10
Common Jellyfish 11
Algae 11
Clams 12
Marine Mammals 12
Challenges 12
Acoustic Scattering Studies 12
Target strength dependences 12
Acoustic classification 12
Fish school quantification 12
Mesopelagic fish quantification 12
Quantifying behavior 13
Mensuration 13
Bistatic scattering 13
In situ calibration 13
Sonar Technology 13
Adapting nonscientific echo sonars for scientific use 13
Synthesizing images 13

☆
Change History: August 2018. KG Foote added Abstract and new sections Recent Trends and Challenges and updated sections Historical Overview,
Quantification of Scattering, Fish as Scatterers, and Further Reading and lesser changes have been made throughout the MS.

Encyclopedia of Ocean Sciences, 3rd Edition https://doi.org/10.1016/B978-0-12-409548-9.11486-1 1

2 Acoustic Scattering by Marine Organisms

Platforms 13
Acknowledgment 13
Further Reading 14

Historical Overview

Development of underwater sonar as a tool for navigation and military operations, following sinking of the Titanic in 1912, led
inevitably to applications to marine organisms. By the 1930s, echoes from fish schools had been detected. In the 1940s, the deep
sound-scattering layer was observed. Its biological origin in mesopelagic fish was identified in the 1950s. At the same time,
applications to commercial fish were pursued with vigor, and both scientific echo sounders and fishery echo sounders began to
be manufactured.
Steady improvements in transduction enabled individual fish of certain species and sizes to be detected at ranges of hundreds of
meters. The ultrasonic frequency of 38 kHz was becoming a standard at this time; it was subsequently shown to be near the
optimum for achieving detection of commercially important fish in the presence of attenuation due to spherical spreading and
absorption. Parallel to studies of single-fish scattering at ultrasonic frequencies were studies of scattering at sonic frequencies,
especially to determine the resonance frequency in swimbladder-bearing fish, which is a measure of size.
Echo integration was introduced in 1965 as a tool for quantifying fish aggregations at essentially arbitrary conditions of
numerical density. This was rapidly developed, and it has been used routinely in surveys of fish stock abundance since about
1975. Reintroduction of standard-target calibration in 1979 served the cause of quantification by providing a rapid, high-accuracy
method of enabling the results of echo integration to be expressed in absolute physical units. With few exceptions, standard-target
calibration has become the method of choice.
Sonar, with one or more obliquely oriented or steerable beams, began to find common application in the 1970s for counting
fish schools that might be missed by a vertical echo sounder beam. This was a significant development for acknowledging the
narrowness of the sampling volume of vertically oriented directional echo sounder beams and the possibility of fish avoidance
reactions to the transducer platform, typically a research vessel.
In another parallel development, the Doppler principle was exploited to measure the rate of approach or recession of fish targets.
Both horizontally oriented echo sounder beams and sonar beams were used. Early applications determined the swimming speeds of
schools of small pelagic fish and individual salmon in rivers.
Applications of acoustics to fish in the 1970s were accompanied by notable applications to zooplankton, if pursued less
intensively owing to differences in commercial importance. Because of the enormous diversity of zooplankton species in size,
shape, and composition, it was recognized early that ensonification over a band of frequencies is required, even for routine
observation. This has usually been achieved by the use of multiple resonant transducers, but genuinely broadband sonars are also
proving successful in yielding spectra of individual euphausiids and copepods.
Recognition of the importance of bandwidth in scattering by zooplankton was accompanied by appreciation of the role of
interpretive models. Acoustic scattering models have been developed and applied to fish since the 1950s and to zooplankton since
the 1970s.
The transition from analog to digital technologies from the 1970s facilitated processing of echo data. Digital echo integrators
were introduced in the 1980s. Echo data postprocessing systems followed, with the development of software to assist in data
storage, retrieval, visualization, scrutinizing, and quantification. An important principle was enunciated: to maintain operator
control of all critical decisions through the man–machine interface. This principle was adhered to the 1990s, establishing the
standard for subsequent commercial echo data postprocessing systems.
In the 1980s and 1990s, multiple-element transducer arrays were developed, adapted, or refined for application to marine
organisms. An early example was that of the split-beam transducer array, which was integrated into scientific echo sounders. This
enabled localization of resolved individual fish targets, hence tracking. By knowing the position of a target in the transducer beam,
the beam pattern loss could be estimated, effecting direct measurement of fish target strength.
During the same period, multiple-element transducer technology was developed and applied in high-frequency, narrowband
multibeam and sidescan sonars. These enabled imaging of fish schools, while begging the question of their further quantification.
All of these multiple-transducer-element developments were due to increasing data processing power through faster digital
computations. This development was exploited in ad-hoc processors called beamformers.
Contemporaneous with developments in sonar array technology, capabilities of scientific echo sounders increased significantly.
Dynamic ranges of 140–160 dB were achieved in commercial off-the-shelf (COTS) systems, increasing sensitivity so that individual
zooplankton, for example, euphausiids and some mesozooplankton, could be detected. At the same time, seafloor echoes could be
registered without saturating the receiver. These were remarkable achievements indeed. In parallel, theoretical advances enabled
compensation for acoustic extinction in dense, extended schools or aggregations of fish.
The several developments were notable for enabling acoustic observations of overwintering Norwegian spring-spawning herring
(Clupea harengus) to be quantified during the period of its recovery, when the spawning biomass again exceeded one million tons.
Through similar developments, the collapse of the stock of northern cod (Gadus morhua) in the early 1990s was monitored
 Acoustic Scattering by Marine Organisms 3

convincingly. The management of other fish stocks benefitted from these developments, as in the case of walleye pollock (Theragra
chalcogramma).

Recent Trends

The past two decades, roughly the 2000s and 2010s, witnessed an ongoing series of major developments that are continuing to
improve understanding and application of acoustic scattering by marine organisms. Some of these are outlined.

Acoustic Scattering Studies

The need for more and better information on the target strength of marine organisms is constant. Numerical values of target strength
associated with a particular sonar signal for identified species of known size and biological and physical states are especially valuable.
These numbers may be expressed through a target strength–size relationship, typically derived from regression analysis. An explicit
dependence on frequency determines the target strength spectrum. A dependence on depth, that is, pressure, is also very useful.
Recent work to determine new values of target strength have been performed on many species. These include, among many
others, armorhead (Pentaceros wheeleri), black porgy (Acanthopagrus schlegelii), chum salmon (Oncorhynchus keta), Euphausia pacifica,
gelatinous zooplankton (Cyanea nozaki and Aurelia aurita), giant jellyfish (Nemopilema nomurai Kishinouye), lanternfish, family
Myctophidae: Diaphus theta, Symbolophorus californiensis, and Notoscopelus japonicus, sandeel (Ammodytes personatus), southern bluefin
tuna (Thunnus maccoyii). Target strengths of cetaceans and fish schools have also been measured during this period.
Other recent work to determine target strength is based on modeling. Advances are being achieved through application of the
prolate spheroidal model to swimbladder-bearing fish at both high and low frequencies. Numerical methods are also being adapted
for target strength determination. These include boundary-element modeling of scattering by swimbladder-bearing fish at relatively
high frequencies, and finite-element modeling of scattering by the whole fish, with or without a swimbladder.

Mensuration
To verify and quantify sonar performance, protocols for the calibration of sonars have continued to be developed and refined. For
operations requiring efficient and rapid, hence cost-effective, calibration, the standard-target method is in worldwide use. This
method has the additional prominent advantages of being adaptable to sonars with transducer arrays as configured for operations,
while aiming to achieve an accuracy of
0.1 dB in the overall transmit-receive response function. The standard-target method has
been extended in frequency, so that its overall span is now from 1 kHz to several megahertz.
The standard-target method is applicable to the measurement of both echo energy and echo spectrum, now demonstrated to be
rigorously consistent. The acoustic sampling volume, which is a parameter in both kinds of measurements, and which is inherently
stochastic, has been defined for both monostatic and bistatic scattering. The method is the subject of a standard being formulated
under the auspices of an international metrological institution.

Sonar Technology
Traditional sonars continue to be developed to enhance sensitivity, dynamic range, precision, directionality, and echo discrimination.
One example is that of a multibeam sonar with a billboard-type rectangular array of elements. This forms a 20  25 matrix of beams,
with bandwidth 70–120 kHz. This is being applied in studies to determine the three-dimensional structure of fish schools. A second
example is that of sidescan sonar in which a single horizontal line array of elements is being compounded through the addition of
similar, aligned arrays, enabling measurement of phase in the vertical plane. The resulting phase-measuring sidescan sonar can
determine bathymetry in addition to measuring backscatter. A third example is that of multibeam sonars whose outputs are expressed
in calibrated units. Such devices, which are also called multibeam echo sounders, are being applied in fisheries research.
Some other, newer sonars are being developed and applied in fisheries research. These include, among others, acoustic-lens-
based sonars, with instantaneous formation of multiple beams over an angular sector; blazed-array multibeam sonars in which
beams are formed in different directions by a simple changes in the transmit frequency; and parametric sonars, with applications of
difference-frequency beams to the measurement of water-column scatterers.

Platforms
A remarkable proliferation of sonar platforms has occurred, with many ongoing developments. Ordinary, now traditional, sonar
platforms have included surface vessels, research vessels, fishing vessels, fishery survey vessels, remotely operated vehicles (ROVs),
and autonomous underwater vehicles (AUVs). Newer platforms include very shallow-water-draft vessels; acoustic landers, sondes,
and autonomous profilers; drifters; moorings; and ocean observatories. The basic aim in using novel platforms is to make better or
more extensive measurements of acoustic scattering.
4 Acoustic Scattering by Marine Organisms

Physical Basis for Scattering

Acoustic scattering by a marine organism is, in principle, no different from that of any other kind of scattering. Differences in the
physical properties of the causative bodies with respect to the surrounding medium are accompanied by reflection and refraction, or
more generally diffraction, of incident waves. Organisms, with contrasts in mass density or elasticity relative to sea water, are thus
sources of scattering.
The processes of reflection, refraction, and diffraction occur at surfaces, both external and internal, marking discrete changes in
physical properties and throughout the volume or inside embedded inhomogeneities, as characterized by continuous changes in
properties. The net result of the individual processes is a redistribution in space of the incident energy field. Changes in direction and
amplitude characterize the scattering.

Classification of Marine Organisms as Scatterers

Marine organisms are conveniently divided into groups based on considerations of taxonomy and anatomy. Two major groups are
those of fish and zooplankton, but others are also treated.
Fish may be distinguished as cartilaginous or bony. Bony fish may be acoustically distinguished because the fish possesses or
lacks a gas-filled swimbladder. Swimbladders may be closed, with gas exchange effected by the rete mirabile, or open, with gas
exchange effected by gulping air at the surface or by releasing a sphincter muscle on a duct leading to the exterior. The respective
swimbladder types are called physoclists and physostomes. They are illustrated by cod (Gadus morhua) and herring (Clupea
harengus), respectively. Some mesopelagic fish possess gas-filled swimbladders, including a number of myctophid species. Some
other myctophids, as well as the deepwater fish orange roughy (Hoplostethus atlanticus), possess swimbladders that are invested with
wax esters. The whiptail (Coryphaenoides subserrulatus), a macrurid, possesses a swimbladder that contains gas in a spongy matrix of
tissue. Swimbladderless fish are illustrated by mackerel (Scomber scombrus). Cartilaginous fish lack a swimbladder, but their liver is
large and presents a marked density contrast with the surrounding fish flesh.
Zooplankton come in many shapes and sizes, but acoustically their variable physical composition admits of a severe reduction.
Three prominent classes have been identified: the liquidlike, the hard-shelled, and the gas-bearing. These are illustrated by,
respectively, euphausiids, pteropods, and siphonophores.
Other marine organisms have also been detected by scattering. These include squid, gelatinous zooplankton, algae, benthos,
marine mammals, and even diving birds. The first five groups are considered in a separate section in the following.

Dependences of Scattering

In general, scattering by marine organisms is affected by a number of factors. Some are listed here.

Intrinsic Factors
Intrinsic to the scatterers are size, shape, internal composition, and condition. Condition may be affected by the stage of
development, presence of reproductive products, and degree of stomach filling. Behavior is another intrinsic factor, if often directly
affected or determined by the external environment, including the conditions of observation. It is typically quantified through the
attitude, or orientation, of the organism and its velocity of movement.

Extrinsic Factors
Scattering is affected by the ensonification signal, hence by its spectral composition. For impulsive signals, the spectrum may be
broadly continuous. For a typical pulsed sinusoid containing many wavelengths, the spectrum will be narrow, and the signal can be
characterized by the center frequency, pulse duration, and amplitude. Depth and history of depth excursion may also influence the
scattering, as in the case of rapid depth changes for physoclists. For swimbladdered fish lacking rete mirabile, depth excursions will
necessarily affect the swimbladder form, with the volume changing in accordance with Boyle’s law, thus inversely with the ambient
pressure.

Quantification of Scattering
Nomenclature
Scattering properties of organisms are distinguished as belonging to individual organisms or to aggregations of organisms. The
fundamental scattering property of a single organism is the scattering amplitude.
Traditionally, this has been described for the idealization of a plane harmonic wave incident on a finite scattering body. At a great
distance r from the body, the scattered pressure field or amplitude psc is related to the incident pressure amplitude pinc by Eq. (1).
 Acoustic Scattering by Marine Organisms 5

lim r!1 psc ¼ pinc f exp ðikr Þ=r: (1)

In Eq. (1), f is the far-field scattering amplitude, r is the distance from the scatterer, k is the wavenumber 2p/l, and l is the acoustic
wavelength. The scattering amplitude f describes the angular characteristics of the scattered field.
The limiting condition in Eq. (1) is generally not satisfied, as in the case of a point source at finite distance r0 from the target and
in which the measurement of the scattered field is performed at position r. For this geometry, in which both the source and receiver
are generally in the nearfield of the target, the scattered pressure field is described by Eq. (2).

psc ¼ q f exp ðikRÞ=R: (2)

where q is the source strength, f is the generalized bistatic scattering amplitude, and R is the distance between the source and receiver,
namely |r–r0 |. In order that Eq. (2) reduces to Eq. (1) in the limits that both the source distance r0 and receiver distance r, become
very large, pinc ¼ q exp(ikr0)/r0. Choosing q ¼ pinc r0 exp( ikr0) is tantamount to requiring that the two incident fields have equal
amplitudes at the origin r ¼ 0. Other limiting cases, as well as that of distributed sources and receivers, can be imagined, enabling a
quite general and powerful definition for the bistatic scattering amplitude.
The differential or bistatic scattering cross-section is | f |2. In the backscattered direction f ¼ fb, and the backscattering cross-section
is given by Eq. (3), where the dual convention of using both sbs and s is shown.
s
sbs ¼ jfb j2 ¼ : (3)
4p

The target strength TS is a logarithmic measure. It is defined in Eq. (4), where r0 now denotes the reference distance,
typically 1 m.
sbs
TS ¼ 10 log : (4)
r02

When many scatterers are concentrated in a volume in which individual scatterers cannot be distinguished by their echoes, a
collective standard measure of scattering is used. This is the volume scattering coefficient. In the backscattered direction, the volume
backscattering coefficient sv is given by Eq. (5), where fb,i is the backscattering amplitude for the ith scatterer of N, and V is the
volume.

X
N  
sv ¼ V1 fb, i 2 : (5)
i¼1

The volume backscattering strength is given by Eq. (6).

Sv ¼ 10 log ðr0 sv Þ: (6)

A quantity useful in echo integration is the area or column backscattering coefficient sa, (Eq. (7)), where the integration is
performed over the range interval [r1, r2].
ðr2
sa ¼ sv dr: (7)
r1

In scattering by fish, a numerically more convenient measure of sa is Eq. (8), which refers the backscattering to the reference area
of one square nautical mile.

sA ¼ 4p18522 sa : (8)

This form is particularly useful, for the fundamental equation of echo integration is simply Eq. (9), where rA is the numerical density
of fish referred to the same area of one square nautical mile, and s is the characteristic or mean backscattering cross-section.

sA ¼ rA s: (9)

Another measure of scattering is the extinction cross-section. This measures the relative loss of energy due to scattering and
internal absorption. It may be defined for an individual scatterer, but is generally applied to aggregations of organisms if they are
sufficiently numerous.
With few exceptions, the issue of calibration must be addressed when making measurements. Standard methods are available for
this, the aim being to define the system characteristics so that the result of a measurement, a voltage signal for instance, can be
expressed relative to a comparable quantity associated with the transmit signal.
6 Acoustic Scattering by Marine Organisms

Measurement
There are dozens of techniques for measuring the scattering properties of individual organisms and aggregations of organisms.
These are commonly distinguished as being in situ, without constraint in the natural environment of the organisms, or ex situ, hence
constrained in some way, wherever this might be.
Target strength is a key quantity in many investigations. It may be determined with a single-beam echo sounder; for example, by
repeated measurement of similar organisms that are acoustically resolved and by appropriate statistical reduction of these
measurements. Alternatively, it may be measured directly with a dual- or split-beam echo sounder, in which the beam pattern
can be determined in the direction of the organism, enabling the backscattering cross-section to be extracted from each
individual echo.
Similar measurements can be performed on single organisms ex situ with greater control and hence knowledge of their state
during measurement. Measurements on tethered organisms, constrained to maintain a given orientation during insonification, are
popular.
Aggregations of organisms are frequently quantified acoustically through the volume backscattering coefficient. If the number
and occupied volume of the organisms are known, then the characteristic target strength can be inferred through Eq. (10).

Sv ¼ 10 log n þ TS: (10)

Here n is the numerical density of organisms, and TS is the so-called mean target strength corresponding to a single organism, but
derived as the logarithmic measure of the mean backscattering cross-section.
Cages are often employed to confine a known or knowable number of organisms to a fixed volume. Measurement of Sv can then
yield a value for TS.

Modeling
The importance of target strength in many studies involving scattering by marine organisms is so great that recourse is frequently
made to theoretical models. On the basis of assumptions about the shape and internal composition of subject organisms,
mathematical expressions may be derived that can be evaluated for particular conditions of concentration or frequency that
might not be realistically explored through measurement. Ultimately, measurements may be used to refine models, and models
to interpret measurements.

Fish as Scatterers
Swimbladder-Bearing Fish
The swimbladder shape varies with species and with condition of the individual specimen. An example of a swimbladder in corpus is
shown in Fig. 1. Here the swimbladder of an Atlantic herring (Clupea harengus) has been exposed by careful dissection.

Low Frequencies
At low frequencies, with acoustic wavelengths much greater than characteristic swimbladder dimensions, the effect of a pressure
wave on the swimbladder is essentially that of uniform compression and rarefaction. Consequently, the fish can be modeled as a
spherical swimbladder. The first model, published in 1964, described a swimbladder-bearing fish as a single spherical bubble. This
model was developed and expanded, eventually resulting in two shelled-spherical models. Both are current.

Single-shell spherical model

The current single-shell spherical model was developed by Love in 1978. This describes scattering of a plane acoustic wave by an
immersed swimbladder-bearing fish as due entirely to the swimbladder modeled as a small spherical shell enclosing an air cavity.
The shell has finite thickness, with the physical properties of fish flesh. Under assumption that the ratio of specific heats is the same
in fish flesh and water, that the temperature of the fish flesh is the same as that in the immersion medium, and that the viscosities of

Fig. 1 Drawing of a specimen of Atlantic herring (Clupea harengus), female, 36.0 cm long, 453 g, with exposed swimbladder. Drawing by H. T. Kinacigil, used
with permission.
 Acoustic Scattering by Marine Organisms 7

water and air are negligible compared to the viscosity of fish flesh, the basic equations can be reduced to five linear equations. This is
the same as the number of boundary conditions, which are the following: continuity of the normal components of particle velocity
and pressure or stress on the two boundaries, or interfaces, with the addition of continuity of temperature at the inner boundary.
The resulting solution has been reduced and simplified further by restriction of the depth to 1000 m and sonar frequency to the
band 0.1–40 kHz. This has enabled simplification and use of earlier functional forms, as in Eq. (11) for the backscattering cross
section s and frequency n.
 2
4pa2 rw =rf
s¼  2 , (11)
½v0 =ðvHÞ2 þ ðv0 =vÞ2  1

where rw is the ambient mass density of water, rf is that for fish flesh, n0 is the resonance frequency given by Eq. (12), and H is a
damping factor given by Eq. (13).
  1=2  
1=2
n0 ¼ ð2pÞ1 3ga P= ra a2 1  2s= 3ra ca 2 a (12)

where a is the equivalent spherical radius for the swimbladder volume, ga is the ratio of specific heats of air, P is the ambient
pressure, ra is the mass density of air, s is the surface tension in the swimbladder wall, hence in the surrounding viscous shell, and ca
is the speed of sound in air.
1 1 1 1
¼ þ þ (13)
H Hrad Hvis Hth

where the successive factors in H describe radiation, viscous, and thermal damping. Model parameters include the sound speeds,
mass densities, and ratios of specific heats, thermal conductivity of air, and shear viscosity of fish flesh.

Double-shell spherical model

The single-shell spherical model has proved to be successful in many instances. However, it has failed to predict the observed
disproportionate increase in resonance frequency with depth of at least one physoclistous fish, namely cod (Gadus morhua), with
speculation that this is due to a controlled increase in tonus of the swimbladder wall to maintain stable hearing during depth
excursions. To remedy this shortcoming, Feuillade and Nero developed a double-shell model in 1998.
The double-shell model consists of a spherical gas-filled cavity surrounded by a solid elastic inner shell surrounded by a viscous
fluid outer shell immersed in water. Each shell has a finite thickness. The relevant boundary conditions are essentially continuity of
normal components of particle velocity and normal component of pressure or stress at all interfaces, supplemented by vanishing of
tangential stress on the inner wall of the elastic shell, that is, the interface between cavity and inner shell, and between the outer shell
and immersion medium, and continuity of tangential stress and continuity of tangential particle velocity at the interface between
the two shells. Thus there is a total of 10 boundary conditions, which when applied yield 10 linear equations in the 10 unknown
coefficients used for the n-th term in the expansion of the corresponding stress and strain fields.
Solving the set of equations is straightforward but sufficiently cumbersome that it is not elaborated in the primary reference.
However, this double-shell model supports a depth dependence in shear rigidity modulus in the elastic inner shell and viscosity in
the outer viscous shell that explains the mentioned observation of the disproportionately increasing resonance frequency with
increasing depth.

Applications to swimbladdered fish

For gadoids and clupeoids in the size range 8–30 cm, n0 varies over 2.2–0.3 kHz. Given the inverse relationship of resonance
frequency and size in Eq. (12), smaller fish will have higher resonance frequencies. Thus, mesopelagic fish with partially wax-
invested swimbladders may have resonance frequencies in the low ultrasonic range. Very large swimbladdered fish, say with a total
fish length exceeding 1 m, will have resonance frequencies of the order of hundreds of hertz. The corresponding backscattering
cross-section, hence target strength, can be computed from Eq. (11) or that described in the Feuillade and Nero work.
It is useful to note that the quality factor of the resonance condition, Eq. (14), where Dn describes the range in frequency over
which s decreases to one-half its maximum value, may be of the order of 1.5–3.

Q ¼ v0 =Dv: (14)

Implicit in the low-frequency condition of the model is that s is independent of orientation. Averages of s with respect to
arbitrary orientation distributions will be identical to s itself.
When computing average values of s for aggregations of swimbladdered fish of varying size, s must be averaged with respect to
the size distribution. The characteristic target strength is determined in accordance with Eq. (4).
8 Acoustic Scattering by Marine Organisms

Intermediate Frequencies
As the acoustic wavelength decreases toward characteristic swimbladder dimensions, the scattering becomes markedly directional,
and the backscattering begins to depend sensitively on the orientation of the fish. From measurements made both in situ and ex situ,
the empirical relationship of Eq. (15) between mean target strength TS at 38 kHz and total fish length l in centimeters has been
derived for a number of gadoids.

TS ¼ 20 log l  67:5: (15)

Eq. (16) applies for clupeoids

TS ¼ 20 log l  71:9: (16)

The average backscattering cross-section s may be determined immediately from Eq. (3). For a cod of length l ¼ 50 cm,
TS ¼  33.5 dB and s ¼ 56 cm2. For a herring of length l ¼ 30 cm, TS ¼  42.4 dB, and s ¼ 7.2 cm2.
Blue whiting is an important commercial stock in both hemispheres, and it is routinely surveyed by acoustics. To convert
measurements of acoustic density at 38 kHz to numerical density in accordance with the echo integration Eqs. (9), (17), where l is
the fork length in centimeters. is used for the northern-hemisphere blue whiting (Micromesistius poutassou):

TS ¼ 21:7 log l  72:8: (17)

Eq. (18) applies for the southern-hemisphere southern blue whiting (Micromesistius australis), where l is again the fork length in
centimeters.

TS ¼ 25:0 log l  81:4: (18)

Coincidentally, perhaps, the target strength of yellowfin tuna (Thunnus albacares) at 38 kHz is nearly identical to that of
Micromesistius australis and is given by Eq. (19).

TS ¼ 25:3 log l  80:6: (19)

The target strength of bigeye tuna (Thunnus obesus) under similar conditions is given by Eq. (20).

TS ¼ 24:3 log l  73:3: (20)

These relations were established from specimens in the approximate size range 50–130 cm and 3–50 kg.
The whiptail (Coryphaenoides subserrulatus), with a swimbladder containing gas-filled spongy tissue, seems to have a mean in situ
target strength at 38 kHz that is consistent with the equation developed for another macrurid, the blue grenadier or hoki
(Macruronus novaezelandiae) (Eq. (21), where l is the total length in centimeters).

TS ¼ 20 log l  72:7: (21)

Some stocks of orange roughy (Hoplostethus atlanticus) are being surveyed about their seamount habitats. Determination of the
target strength of this deepwater fish with fat-invested swimbladder is admittedly problematical. Some work suggests convergence of
the mean target strength of a 35 cm long orange roughy at 38 kHz to about 48 dB. If the standard equation for mean target
strength–length were used, namely Eq. (22),

TS ¼ 20 log l þ b: (22)

the coefficient b would be 79 dB.

For modeling scattering by swimbladdered fish at these frequencies, the Kirchhoff approximation model can be used. This
assumes that the fish is represented by the swimbladder, which acts as a pressure-release surface where it is directly insonified, and as
a surface without response otherwise.
A more general scattering model is that of the boundary-element method. The swimbladder is represented by a mesh of points,
called nodes, spanning the surface, illustrated in Fig. 2. The harmonic wave equation is solved numerically, assuming continuity of
pressure and normal component of velocity at each node. It is thus possible to model the effects of internal gas density and pressure.
To convert modeled values for s as a function of orientation to an average value, an orientation distribution is required. Ideally,
this is done on the basis of in situ observations, but often such data are lacking and an orientation distribution must be assumed.
Some orientation distributions are described in the literature. In some special circumstances it has been possible to infer the
orientation distribution by a combination of acoustic measurement and modeling.
The relationship of maximum and average measures of s is given approximately by Eq. (23).
 Acoustic Scattering by Marine Organisms 9

Fig. 2 Boundary element model of the swimbladder of a specimen of pollack (Pollachius pollachius), 34.5 cm in length, with anterior end to the lower right
(y direction). Model by D. T. I. Francis, used with permission.

smax  7save : (23)

Alternatively, Eq. (24) can be used.

TSmax  TSave þ 5 dB: (24)

Measures of the extinction cross-section are relatively rare, there being few occasions when it is necessary to compensate for
scattering losses. However, measurement or inference suggest that the extinction cross-section is very roughly 1–3 times the
backscattering cross-section at intermediate frequencies. Ultimately, the cross-sections and their ratio must depend on the behavior
of the organism, as quantified through the orientation distribution.

High Frequencies
When the acoustic wavelength becomes very small compared to the swimbladder size, scattering by other tissues may become
important. The contributions of head structure, vertebrae, and even scales at very high frequencies have been established through ex
situ measurement. Modeling of scattering by such structures can be computationally excessive, suggesting the advantages of
stochastic modeling if direct measurement is not possible or convenient.

Swimbladderless Fish
The mackerel is a prominent example of a swimbladderless fish. Its target strength must be attributed to the nonswimbladder
structures and hence is intrinsically complicated at nearly all frequencies. At intermediate frequencies, the mean target strength is
roughly 10 dB less than that of a gadoid of comparable size (Eq. 25).

TSmackerel  TSgadoid  10 dB: (25)

For cartilaginous fish, such as sharks, the liver may be very large. In pelagic sharks, this may be of the order of 7%–23% by
weight; in demersal sharks, 3%–6%. The specific gravity of lipids is of the order of 0.87–0.92 in pelagic sharks and 0.93–0.94 in
demersal sharks, further suggesting the role of the liver in buoyancy and its significance in acoustic scattering. At least for the pelagic
sharks, the size and difference in mass density may explain much of the target strength. Were a model to be constructed, a pelagic
shark might be represented by a body with the size, shape, and physical properties of the liver.

Zooplankton as Scatterers
Liquid-Like Bodies
A number of prominent and abundant zooplankton can be classified as liquidlike in their acoustic properties. Extensive modeling
and measurement have demonstrated that internal shear waves have negligible influence in scattering by such organisms.
The animals are thus generally fluidlike in their properties. If the same animals lack sizable organs or other tissue presenting
large contrasts in mass density or compressibility relative to the sea water immersion medium, then the acoustic properties of the
organisms are more particularly liquidlike, and their acoustic scattering is consequently relatively weak. Two examples of zoo-
plankton with liquidlike properties are euphausiids and copepods. These are also representative of homogeneous and inhomoge-
neous scatterers, respectively.
10 Acoustic Scattering by Marine Organisms

Homogeneous Liquidlike Bodies

The expectation of relatively weak scattering by euphausiids has been confirmed by measurement. For example, the target strength
of Antarctic krill (Euphausia superba) of mean lengths 30–39 mm is in the range from 88 to 83 dB at 38 kHz and from 81 to
74 dB at 120 kHz. The respective acoustic wavelengths are 39 and 12.5 mm.
For a scattering body that is relatively long compared to the wavelength, the scattering will be inherently directional. Laboratory
measurement has demonstrated strong effects of orientation on scattering by euphausiids in the size range 30–42 mm at frequencies
of 120 kHz and higher.
In modeling scattering by homogeneous liquidlike zooplankton, there are just two significant material properties, the mass
density and compressibility, or longitudinal-wave sound speed. A variety of models can be used to represent shape. At low
frequencies, a single euphausiid can be represented by a finite circular cylinder or even a sphere, with volume equal to that of the
animal. At higher frequencies, the same animal might be represented as a finite, bent, tapered cylinder or, better, by the actual shape
of the exoskeleton.
Scattering models for euphausiids have demonstrated the sensitive dependence of target strength on both the material properties
and orientation of the organism. Given the rarity of measurements of material properties, their seasonal and individual variability,
and the generally unknown orientation, there has been little systematization of measured values of target strength.
Theoretical understanding of scattering by euphausiids has succeeded in associating large angular lobes with the echo spectrum
at rather short acoustic wavelengths. When these are combined with knowledge of the target strength to within about an order of
magnitude, it is possible to classify euphausiids by their acoustic signature.

Inhomogeneous Liquidlike Bodies

Copepods, like euphausiids, also display relatively weak acoustic scattering. Unlike euphausiids, however, their internal structure is
acoustically distinct, being composed of two dominant scatterers, a prosome and an embedded oil sac. Because of the low density of
lipids in the oil sac, of the order of 900 kg m3, the prosome must be correspondingly denser and more massive. Because the
copepod body as a whole is close to neutral buoyancy in sea water, the target strength is due to the internal contrast in mass density
and compressibility, or longitudinal-wave sound speed, between the prosome and oil sac.
Measurement has shown that the target strength of a 2 mm long copepod, Calanus finmarchicus, is in the approximate range from
95 to 90 dB over the frequency range 1600–2400 kHz.
Copepods have been modeled as composite two-liquid-body structures. Numerical values for the mass density and longitudinal-
wave sound speed have been derived from measurements or have been assumed. The shapes of embedded oil sac and encompassing
prosome, illustrated in Fig. 3, have been determined from videomicroscopic cross-sections in dorsal and lateral views. Results of
modeling of copepods have shown the expected weak dependence on orientation at low or moderate frequencies, and an overall
mean target strength that is in line with measured values.

Hard-Shelled Bodies
An example of a hard-shelled zooplankton is the pteropod Limacina retroversa, a marine snail with a spiral shell, opercular opening,
and wings that propel it through the mid-water column. The target strength of specimens of shell length 2 mm has been measured
over the approximate frequency range from 350 to 750 kHz. The target strength varies between 80 and 60 dB, depending on
both frequency and orientation.
The pteropod has been modeled as a rough spherical shell with a circular opening. Predictions of scattering have been in
reasonable agreement with measurements at wavelengths roughly comparable to the maximum shell dimension.

Gas-Bearing Bodies
Siphonophores are representatives of gas-bearing zooplankton, with gas inclusions in the pneumatophores. These are generally
small compared to overall dimensions of specimens, and the target strength varies widely over the frequency range 350–750 kHz.
In particular, the target strength varies over the range from 90 to 60 dB, but with no apparent systematic dependence on
frequency. This wide range is suggestive of interference between echoes from the gas inclusions and the nongaseous tissue, the basis
of an acoustic model.

Other Organisms as Scatterers

Squid
A number of specimens of squid have been observed by acoustics. These include Todarodes pacifica, Loligo opalescens, and Loligo
vulgaris reynaudii. In a survey of the second species, performed at 120 kHz, the target strength of specimens of mean dorsal mantle
length 11.6 cm and mean mass 23.7 g was about 59 dB. Thus in the standard target strength–length Eq. (21), but with l
representing the mean dorsal mantle length, b is about 80 dB. For Todarodes pacifica of mean dorsal mantle length 16 cm and
 Acoustic Scattering by Marine Organisms 11

Fig. 3 Boundary element models of the prosome and oil sac of a specimen of Calanus finmarchicus, stage 6 female, 2.74 mm in length, with anterior end to the
lower left (x direction). Models by D. T. I. Francis, used with permission.

mean mass 95 g, the target strength is about 51 dB at 28.5 kHz and 55 dB at 96.2 kHz, corresponding to values of b of 75 and
79 dB, respectively. For Todarodes pacifica of mean dorsal mantle length 23.7 cm and mean mass 340 g, the respective mean target
strengths at 28.5, 50, 96.2, and 200 kHz are about 45.7, 46.5, 48.0, and 47.6 dB, with respective values of b of 75, 74,
76, and 76 dB. For Loligo vulgaris reynaudii, the target strength was measured at 38 kHz for sufficiently dispersed animals of mean
mass 300 g. The target strength when referred to 1 kg was 42.5 dB. This compares favorably with the measurements on Loligo
opalescens at 120 kHz and Todarodes pacifica at 28.5 kHz. When expressed relative to 1 kg, the respective target strengths are 42.3
and 41.1 dB.

Common Jellyfish
In anticipation of acoustic surveying of the ctenophore Mnemiopsis leidyi and other gelatinous zooplankton, namely Aurelia aurita
and Pleurobrachia pileus, in the Black Sea, measurements have been made of the target strength of the common jellyfish Aurelia aurita.
Functional regression equations have related the mean target strength in decibels to the disk diameter d in centimeters. At 120 kHz,
the relation is Eq. (26).

TS ¼ 14:7 log d  74:6: (26)

At 200 kHz it is Eq. (27).

TS ¼ 39:6 log d  104:4: (27)

Thus for a specimen with mean diameter 10 cm, TS ¼  59.9 and 64.8 dB at the respective frequencies.

Algae
Algae, such as kelp, are being surveyed by acoustics. For purposes of quantification, the acoustic properties of the plants themselves
are being studied, both by experiment and by theoretical modeling. Measurements have been performed on leaves of Laminaria
saccarina and Laminaria digitata at three ultrasonic frequencies. The lengths of these span the range 0.7–2 m; the widths 0.4–0.9 m;
12 Acoustic Scattering by Marine Organisms

the thicknesses 1–5 mm; and the masses 0.33–0.8 kg. Target strengths expressed relative to 1 kg of biomass vary from 35 to
28 dB at 50 kHz, from 33 to 24 dB at 70 kHz, and from 29 to 22 dB at 200 kHz.
Smaller algae, the phytoplankton Prorocentrum micans, Peridinium triquetrum, Olistodiscus luteus, Dunaliella salina, Platimous viridis,
and Phaeodactylum tricornutum, are also being studied by acoustics. Measurements of reverberation, in particular, are being used in
attempts to quantify the volume of gas vacuoles.

Clams
Both the razor clam (Tagelus dombeii) and the surf clam (Mesodesma donacium) have been surveyed by acoustics. Beds of the razor
clam have been surveyed in shallow water over a flat bottom. Echograms that show the bottom–surface–bottom reflection in
addition to the first bottom reflection show an enhanced registration above the so-called second bottom echo. Counting of its
characteristic serrations provides a quantitative measure of clam density.

Marine Mammals
A few measurements have been reported on the target strength of the sperm whale (Physeter catodon) and the humpback whale
(Megaptera novaeangliae) in situ. Measurements have been made of the Atlantic bottlenose dolphin (Tursiops truncatus) in captivity.
Measurements made on a 2.2 m long 126 kg female dolphin in broadside aspect at the surface revealed a mean target strength that
decreased from about 10 dB at the lowest measurement frequency of 23 kHz to about 24 dB at 45 kHz, rising to about 20 dB at
65 kHz, then falling to 25 dB at 80 kHz. The observed degree of variability about these nominal values due to repeated
insonification was 4–11 dB to within the first standard deviation to either side.

Challenges

Applications of acoustic scattering in investigations of marine organisms continue to be developed, expanded, and refined. These
are meeting challenges, several of which are mentioned here. The organization follows that of the earlier section on recent trends.

Acoustic Scattering Studies

Target strength dependences
Target strength is a measure of scattering. As a number, it applies to an individual of a particular species, size, orientation, and
biological and physical state, when ensonified by a particular sonar signal. To improve the measurement of numerical density, rA in
Eq. (9), dependences of the backscattering cross section s, or equivalently the target strength per Eq. (4), continue to inspire both
experimental and theoretical work.

Acoustic classification
The overall complexity of the problem of describing the character of marine organisms, including size, on the basis of acoustic
scattering without physical capture has motivated researchers to pursue ad-hoc solutions that apply to particular species or
scenarios. Methods include, among others, matching measured target strength spectra with tabulated reference spectra derived
from controlled measurements or modeling; applying discriminant analysis; training neural networks; and machine-learning. If
successful to almost any degree, the methods potentially can be applied in fishing operations to mitigate bycatch.

Fish school quantification

The bulk of work done on acoustic scattering by marine organisms applies to individuals. Work is now underway to quantify fish in
schools or other dense aggregations. Both low- and high-frequency sonars are being developed or applied to the problem, often with
rather large bandwidths, allowing application of the powerful signal processing method of cross-correlation with the transmit
signal, that is, pulse compression. Low-frequency ensonification is especially advantageous for swimbladder-bearing fish, to excite
the swimbladder resonance, but the typically much narrower beamwidths at high frequencies are useful for resolving the three-
dimensional structure, which may be characteristic of the subject organism.

Mesopelagic fish quantification

The diurnal vertical movements of mesopelagic fish are said to be the greatest migration on Earth. Beyond the observations that the
corresponding deep sound scattering layer is worldwide and is presumed due to the possession of a swimbladder by at least some
ubiquitous species, surprisingly little is known. The likely vastness of mesopelagic fish as a cause of formidable acoustic scattering
and as a biological resource subject to fishing is motivating acoustic research to learn more about the resource, including its exact
composition and details of its geographical distribution.
 Acoustic Scattering by Marine Organisms 13

Quantifying behavior
This problem has entered the classic category, the intention being to avoid disturbing the marine organisms being observed.
An alternative approach is to observe and quantify the behavioral effect so that compensation can be made, if possible, to the
measured scattered signal. Certainly this second approach can be argued to be more affordable than the first, given the expense of
noise-quieting measures, hence the emphasis on quantifying behavior.

Mensuration
Two challenges are noted: bistatic scattering and in situ calibration.

Bistatic scattering
Opportunities to measure bistatic scattering are becoming more frequent, as with operation of multiple AUVs or other mobile
platforms in the same area, but how can this additional information be used? Might it be sufficient to derive or infer bistatic
scattering properties of marine organisms on the basis of theory, without the used to undertake measurements too?

In situ calibration
Advantages of the standard-target sonar calibration method have already been mentioned. For the most part, with some exceptions,
this calibration method can be performed at a special facility or on a stationary platform. There are situations, however, where the
calibration should be performed in situ. Cogent reasons may include the lowness of the sonar frequency, with correspondingly large
nearfield, or circumstances requiring calibrations to be performed expeditiously at the time or venue where organisms are being
measured. Calibration protocols need to be developed when the standard target is deployed away from the sonar platform, that is,
independently of the platform. Knowing where the target is relative to the sonar transducer becomes a challenge that must be
addressed explicitly if in situ calibration is going to succeed.

Sonar Technology
The push and pull of science and technology is constant in the long term, but progress is irregular in the short term, depending as it
does on opportunity or chance discovery. Thus the following challenges are inevitably subjective.

Adapting nonscientific echo sonars for scientific use

Opportunities to measure acoustic scattering by marine organisms would be very much enhanced if fishery echo sounders, acoustic
Doppler current profilers (ADCPs), and many commercial sidescan sonars could be used scientifically. The basic problem, which is
common to these, is that the output signal cannot be calibrated with respect to amplitude. (i) In principle, the automatic gain
control (AGC) that is applied to the echo signal in some of the most widely sold fishery echo sounders could be measured and used
to retrieve what was earlier a quantitative signal. This is very cumbersome. (ii) In the case of ADCPs, often the amplitude part of the
signal is neglected in the device for being extraneous to the phase measurement that is used to determine velocity. (iii) A problem
with many commercial sidescan sonars is that application of range compensation, as through time-varied gain, is quite variable,
with abrupt changes in gain made at different ranges. The dynamic range of sidescan sonars, as with some ADCPs, is often
fundamentally limiting, and may require adaptive or dynamic adjustment.
It is to be noted that all of the devices are fully capable of being developed to provide quantitative, calibrated output signals. The
basic adaptations are best done by the manufacturer, not by the individual researcher.

Synthesizing images
Integrating or synthesizing images is necessary when a target is being ensonified at different aspects, as by a sonar transducer
mounted on a moving platform. Such a synthesis can be useful in applications to marine organisms for both classification and
quantification, but also for studies involving the habitat, for example, bottom and bottom structures.

Platforms
The proliferation of novel platforms has already been noted. Apropos of sonar applications, avoidance of flow noise over the sonar
transducer will remain a design consideration if not a challenge.

Acknowledgment

Support by the Investment in Science Program at the Woods Hole Oceanographic Institution is gratefully acknowledged.
14 Acoustic Scattering by Marine Organisms

Further Reading
Bassett C, De Robertis A, and Wilson CD (2018) Broadband echosounder measurements of the frequency response of fishes and euphausiids in the Gulf of Alaska. ICES Journal of
Marine Science 75: 1131–1142. https://doi.org/10.1093/icesjms/fsx204.
Becker KN and Warren JD (2015) Material properties of Pacific hake, Humboldt squid, and two species of myctophids in the California Current. The Journal of the Acoustical Society of
America 137: 2522–2532. https://doi.org/10.1121/1.4919308.
Benoit-Bird KJ and Au WWL (2003) Echo strength and density structure of Hawaiian mesopelagic boundary community patches. The Journal of the Acoustical Society of America
114: 1888–1897. https://doi.org/10.1121/1.1612484.
Butler JL and Sherman CH (2016) Transducers and arrays for underwater sound, 2nd edn. Cham: ASA Press and Springer.
Chu D and Stanton TK (1998) Application of pulse compression techniques to broadband acoustic scattering by live individual zooplankton. The Journal of the Acoustical Society of
America 104: 39–55. https://doi.org/10.1121/1.424056.
Clay CS and Horne JK (1994) Acoustic models of fish: The Atlantic cod (Gadus morhua). The Journal of the Acoustical Society of America 96: 1661–1668. https://doi.org/
10.1121/1.410245.
Craig RE (ed.) (1984) Fisheries Acoustics, A symposium held in Bergen, 21–24 June 1982, Rapports et Proces-Verbaux des Reunions, vol. 184, International Council for the
Exploration of the Sea: Copenhagen.
Demer DA and MacLennan DN (eds.) (2009) The ecosystem approach with fisheries acoustics and complementary technologies, Proceedings of an ICES International Symposium held
in Bergen, Norway, 16–20 June 2008, Oxford: Oxford University Press. ICES Journal of Marine Science, vol. 66, no. 6.
Feuillade C and Nero RW (1998) A viscous-elastic swimbladder model for describing enhanced-frequency resonance scattering from fish. The Journal of the Acoustical Society of
America 103: 3245–3255.
Foote KG (1997) Target strength of fish. In: Crocker MJ (ed.) Encyclopedia of acoustics, vol. 1, pp. 493–500. New York: Wiley.
Foote KG (2014) Discriminating between the nearfield and the farfield of acoustic transducers. The Journal of the Acoustical Society of America 136: 1511–1517. https://doi.org/
10.1121/1.4895701.
Foote KG (2017) Standard-target calibration of active sonars used to measure scattering: Principles and illustrative protocols. IEEE Journal of Oceanic Engineering 15. https://doi.org/
10.1109/JOE.2017.2713538.
Foote KG and Stanton TK (2000) Acoustical methods. In: Harris RP, et al. (eds.) ICES zooplankton methodology manual, pp. 223–258. London: Academic Press.
Foote KG, Hanlon RT, Iampietro PJ, and Kvitek RG (2006) Acoustic detection and quantification of benthic egg beds of the squid Loligo opalescens in Monterey Bay, California. The
Journal of the Acoustical Society of America 119: 844–856.
Francis DTI and Foote KG (2003) Depth-dependent target strengths of gadoids by the boundary-element method. The Journal of the Acoustical Society of America 114: 3136–3146.
Freon P and Misund OA (1999) Dynamics of pelagic fish distribution and behaviour: Effects on fisheries and stock assessment. Oxford: Fishing News Books.
Furusawa M (1988) Prolate spheroidal models for predicting general trends of fish target strength. Journal of the Acoustical Society of Japan (E) 9: 13–24. https://doi.org/10.1250/
ast.9.13.
Furusawa M (1991) Designing quantitative echo sounders. The Journal of the Acoustical Society of America 90: 26–36. https://doi.org/10.1121/1.401297.
Godø OR, Foote KG, Dybedal J, Tenningen E, and Patel R (2010) Detecting Atlantic herring by parametric sonar. The Journal of the Acoustical Society of America 127: EL153–EL159.
Gorska N, Ona E, and Korneliussen R (2005) Acoustic backscattering by Atlantic mackerel as being representative of fish that lack a swimbladder. Backscattering by individual fish.
ICES Journal of Marine Science 62: 984–995. https://doi.org/10.1016/j.icesjms.2005.03.010.
Jones BA, Stanton TK, Colosi JA, Gauss RC, Fialkowsi JM, and Jech JM (2017) Broadband classification and statistics of echoes from aggregations of fish measured by long-range,
mid-frequency sonar. The Journal of the Acoustical Society of America 141: 4354–4371. https://doi.org/10.1121/1.4983446.
Karp WA (ed.) (1990) Developments in Fisheries Acoustics, A symposium held in Seattle, 22–26 June 1987. Rapports et Proces-Verbaux des reunions, vol. 189, Copenhagen:
International Council for the Exploration of the Sea.
Love RH (1978) Resonant acoustic scattering by swimbladder-bearing fish. The Journal of the Acoustical Society of America 64: 571–580.
Lucifredi I and Stein PJ (2007) Gray whale target strength measurements and the analysis of the backscattered response. The Journal of the Acoustical Society of America
121: 1383–1391. https://doi.org/10.1121/1.2436643.
Margetts AR (ed.) (1977) Hydro-acoustics in Fisheries Research, A symposium held in Bergen, 19–22 June 1973. Rapports et Proces-Verbaux des Reunions, vol. 170, International
Council for the Exploration of the Sea: Copenhagen.
Medwin H and Clay CS (1998) Fundamentals of acoustical oceanography. San Diego, CA: Academic Press.
Mukai T and Iida K (1996) Depth dependence of target strength of live kokanee salmon in accordance with Boyle’s law. ICES Journal of Marine Science 53: 245–248. https://doi.org/
10.1006/jmsc.1996.0029.
Nakken O and Venema SC (eds.) (1983) Symposium on Fisheries Acoustics, Selected papers of the ICES/FAO Symposium on Fisheries Acoustics, Bergen, Norway, 21–24 June 1982,
FAO Fisheries Report no. 300, Rome: Food and Agriculture Organization of the United Nations.
Ona E (1990) Physiological factors causing natural variations in acoustic target strength of fish. Journal of the Marine Biological Association of the United Kingdom 70: 107–127.
Ona E (2003) An expanded target-strength relationship for herring. ICES Journal of Marine Science 60: 493–499.
Ona E, Godø OR, Handegard NO, Hjellvik V, Patel R, and Pedersen G (2007) Silent research vessels are not quiet. The Journal of the Acoustical Society of America 121: EL145–EL150.
Pedersen G, Godø OR, Ona E, and Macaulay GJ (2011) A revised target strength-length estimate for blue whiting (Micromesistius poutassou): Implications for biomass estimation. ICES
Journal of Marine Science 68: 2222–2228.
Physics of Sound in the Sea (1969) Reprint of the 1946 edition. Washington, DC: Department of the Navy.
Progress in Fisheries Acoustics (1989) Proceedings of the institute of acoustics. vol. 11. St. Albans: Institute of Acoustics.
Rose GA (2009) Variations in the target strength of Atlantic cod during vertical migration. ICES Journal of Marine Science 66: 1205–1211.
Sawada K, Furusawa M, and Williamson NJ (1993) Conditions for the precise measurement of fish target strength in situ. ICES Journal of Marine Science 20: 73–79. https://doi.org/
10.3135/jmasj.20.73.
Simmonds EJ and MacLennan DN (eds.) (1996) Fisheries and plankton acoustics, Proceedings of an ICES international symposium held in Aberdeen, Scotland, 12–16 June 1995.
ICES Journal of Marine Science, vol. 53, no. 2, London: Academic Press.
Trevorrow MV, Mackas DL, and Benfield MC (2005) Comparison of multifrequency acoustic and in situ measurements of zooplankton abundances in Knight Inlet, British Columbia. The
Journal of the Acoustical Society of America 117: 3574–3588. https://doi.org/10.1121/1.1920087.
Urick RJ (1983) Principles of underwater sound, 3rd edn. New York: McGraw-Hill.