A noisy spring: the impact of globally
rising underwater sound levels on fish
Hans Slabbekoorn
1
, Niels Bouton
2
, Ilse van Opzeeland
3
, Aukje Coers
4
, Carel ten Cate
1
and Arthur N. Popper
5
1
Behavioural Biology, Institute of Biology, Leiden University, Sylviusweg 72, 2333 BE Leiden, The Netherlands
2
Evolutionary Ecology Group, University of Antwerp, Groenenborgerlaan 171, B-2020 Antwerp, Belgium
3
Ocean Acoustics Lab, Alfred Wegener Institute, Am alten Hafen 26, 27568 Bremerhaven, Germany
4
Pelagic Regional Advisory Council, Treubstraat 17, 2288 EH Rijswijk, The Netherlands
5
Department of Biology and Center for Comparative and Evolutionary Biology of Hearing, University of Maryland, College Park,
MD 20742, USA
The underwater environment is filled with biotic and
abiotic sounds, many of which can be important for the
survival and reproduction of fish. Over the last century,
human activities in and near the water have increasingly
added artificial sounds to this environment. Very loud
sounds of relatively short exposure, such as those pro-
duced during pile driving, can harm nearby fish. How-
ever, more moderate underwater noises of longer
duration, such as those produced by vessels, could
potentially impact much larger areas, and involve much
larger numbers of fish. Here we call attention to the
urgent need to study the role of sound in the lives of
fish and to develop a better understanding of the eco-
logical impact of anthropogenic noise.
The myth of a silent underwater world
In 1962, Rachel Carson wrote about a ‘silent spring’ in the
context of the detrimental impact of the use of pesticides on
singing birds. Here we call attention to a ‘noisy spring’, and
the possible detrimental impact of increasing levels of
anthropogenic noise on fishes
1
[1,2]. Fish populations have
come under threat for a number of well-known reasons
including fisheries [3], habitat degradation [4] and chemi-
cal pollution [5]. Human-generated underwater noise is
potentially becoming another threat to fish, just as traffic
noise has become a major concern in air with regard to
birds and other terrestrial animals [6,7]. Although humans
have engaged in all sorts of activities in, on, and near water
bodies for a long time, only recently have these activities
expanded in an increasingly noisy manner (Box 1). To date,
underwater noise pollution has primarily attracted atten-
tion in the context of marine mammals [8–10], but it is
increasingly recognized as a factor that may also have
implications for fish [11–14].
In this review we focus on the need for behavioural and
ecological studies on the impact of long-term anthropo-
genic noise on fishes. We take this approach since very
large numbers of fish are exposed to moderate but wide-
spread low-frequency noise, produced by vessels, offshore
wind farms and other coastal activities, and yet we have
the barest insight as to the nature and extent of the
behavioural impact of such sounds on fishes (Figure 1).
While data on fish behavioural responses to the increase in
ambient sound are generally not available, we can use data
derived from other vertebrates to suggest that anthropo-
genic noise may deter fish from important feeding and
reproduction areas, interrupt critical activities, or cause
stress-induced reduction in growth and reproductive
output. The concern about wide-ranging effects is further
heightened because sound is of critical importance in the
lives of many fish species. Impeding the ability of fish to
hear biologically relevant sounds might interfere with
critical functions such as acoustic communication, pred-
ator avoidance and prey detection, and use of the ‘acoustic
scene’ or ‘soundscape’ [15,16] to learn about the overall
environment. Taken together, these potential effects could
Review
Glossary
Active space: the distance from a sound-emitting animal over which the sound
is detectable and recognizable by conspecifics. The active space is influenced
by the source amplitude, receiver sensitivity, attenuation and degradation
during transmission, and interference by ambient noise.
Anthropogenic noise: any sound generated by human activities, which has the
potential to warn fish of the danger of approaching boats or risky water inlets.
It may also be detrimental to fish through deterrence, interference and masking
of biologically relevant sounds, or through physiological stress.
Auditory detection continuum: mechanistic scale of fish hearing, replacing the
traditional and oversimplified categories of generalists and specialists. The
scale ranges from fish species without a swim bladder or other air-filled body
cavities and only able to detect particle motion (e.g. sharks) to fish species with
a so-called otophysic connection between swim bladder and ear and able to
detect motion as well as sound pressure (e.g. goldfish).
Auditory masking: the perceptual interference of one sound (often concerning
a signal) by another (often referred to as noise). The masking impact occurring
at the point of the receiver typically depends on the spectral overlap between
and the amplitude ratio of the signal and the noise.
Cortisol: a corticosteroid hormone or glucocorticoid, often referred to as a
stress hormone, due to its involvement in response to stress and anxiety.
Cortisol serves to increase blood sugar levels, stores sugar in the liver as
glycogen, and also suppresses the immune system.
Inner ear: the major structure in fish for detection of sound. The inner ear is
located in the cranial cavity of fish and its basic structure is the same as the
inner ears of sharks and all terrestrial vertebrates, including humans.
Lateral line system: a sense organ used to detect movement and vibration in
fish. Lateral lines are usually visible as faint lines running lengthwise down
each side of the body and sometimes as a faint network of dots on the head.
Swim bladder: an internal gas-filled organ found in most fish species (but for
example not in sharks and rays amongst others) that contributes to the ability
to control buoyancy control and allows a fish to stay at a particular water
depth. The swim bladder can also serve as a resonating chamber and aid in
sound production and sound perception.
Corresponding author: Slabbekoorn, H.
(H.W.Slabbekoorn@Biology.LeidenUniv.NL)
1
Unless otherwise specifically noted, ‘fish’ in this paper refers to bony fishes of the
Osteichthyes taxonomic subclass, the Actinopterygii (or ray-finned fishes).
.
TREE-1243; No. of Pages 9
0169-5347/$ – see front matter ß 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.tree.2010.04.005 Available online xxxxxx
1
have a significant impact on survival of individuals and
populations and affect whole ecosystems.
It is important to make three points at the start of this
review. First, our emphasis is on behavioural effects of
human-generated sound s because these are likely to be
the most significant for fish. At the same time we recog-
nize that there is a great concern by many investigators,
regulators and various industries about high impact
sounds. Underwater explosions, p ile driving, o r seismic
surveys, can all have dramatic effects on nearby fish,
including physical damage and death. While not covered
here, these immediate effects involving relatively few
individual animals have recently been reviewed in great
detail [14].
Second, we also emphasize in this review that research-
ers of the noise impact on fish can get valuable insights
from investigations that have been concerned with similar
issues in terrestrial animals and, to a lesser extent, marine
mammals. In particular, studies on birds might provide
guidance in experimental design and in asking questions
that are the most useful in gaining a better understanding.
Considering the striking similarities in the auditory
system and perceptual abilities of all vertebrates [15],it
would also not be surprising to find congruencies in the
Box 1. Human invasion of the underwater acoustic
environment
Underwater sounds generated by human activities can be sub-
divided in two categories: sounds that are an unintentional by-
product and sounds that are used as a measurement tool. Dominant
in the first category are low-frequency noises from vessels for
container shipping, p ublic transp ort, fishing an d recreational
activities [1,2,67,68, 88]. For exampl e, >80% of gl obal freight
transport takes place over water by motorized shipping, while
passenger crossing occurs on many rivers, lakes and seas, often on
noisy ferries that shuttle between harbours at frequent intervals.
Moreover, fishing vessels typically have strong and noisy motors for
towing gear. Although the global fishing fleet has not grown much
since the early 1990s it still includes about 1.2 million vessels. The
number of recreational vessels is still on the rise, with a growing
impact on coastal and in-shore waters [32,42,89]. Another signifi-
cant source of anthropogenic noise of the first category is that
associated with construction and exploitation of offshore platforms.
The first submerged oil wells were drilled in a fresh water lake in
Ohio (USA) around 1891, and five years later, the first marine oil
wells were drilled near Santa Barbara, California (USA). Today, there
are thousands of offshore oil and gas platforms worldwide. In
addition, the more recent development of exploitation of renewable
sources, such as wind, wave, tidal or current energy, also generates
noise during construction and operational phases [90–92].
The second category of human-generated sounds for various
types of underwater measurements involves both low and high
frequencies. Underwater sound is used by navies, fisheries, the oil
and gas industry, oceanographers, geologists, as well as meteorol-
ogists. The first time that sound was used by humans to locate
objects underwater was shortly after the Titanic sank in 1912. After
that, the use of mid- and later low-frequency sonar has become
widespread for navigation and localization of submarines and other
objects. In the context of fisheries, the first acoustic study concerned
the localization of spawning cod at the Lofoten Islands in 1935 [93].
Then, by 1950, fish-finding echo sounders had become an essential
aid to all commercial fishing vessels. Other acoustic measurement
applications include seismic reflection profiling using high-intensity
airguns to obtain information about the geological structures
beneath the seafloor, and acoustic thermography of ocean climate
(ATOC). The second of these was launched in the 1990 s using
relatively high-intensity sound transmission for long periods to
determine ocean temperature [9].
Figure 1. Four main domains of research to assess the potential impact of
moderate but widespread anthropogenic noise conditions on fish (see Box 4).
Box 2. Underwater sound – an overview
The basic principles of sound propagation in air and in water are the
same, but there are a number of features peculiar to underwater
acoustics [94,95] . Water is an excellent medium for sound transmis-
sion because of its high molecular density. Sound travels about five
times faster in water than in air (about 1500 vs. 300 m/s), and this
means that wavelengths are about five times longer in water than in
air (e.g. for a 100 Hz signal: 3 m in air, 15 m in water). Sound also
attenuates less over the same distance in water than in air. As a
consequence , sound t ravels m uch grea ter dist ances at hi gher
amplitude levels in water compared to air, thereby enabling long-
distance communication, but also a long-distance impact of noise on
aquatic animals.
Sound levels or sound pressure levels (SPL) are referred to in
decibels (dB). However, the dB is not an absolute unit with a physical
dimension, but is instead a relative measure of sound pressure with
the lower limit of human hearing corresponding to 0 dB in air.
Underwater dB-levels are different from above water dB-levels [95].
Sound pressure levels above water are referenced to 20 mPa, while
underwater they are referenced to 1 mPa. As a consequence, adding
25.5 dB to the airborne dB-level is required to get a comparable
underwater dB-level. Furthermore, related to the much higher
acoustic impedance of water compared to air, another 36 dB
correction is required, making an airborne sound pressure level of
70 dB re 20 mPa comparable to an underwater 131.5 dB re 1 mPa.
Sound pressure levels are based on root-mean-square (RMS)
measures averaged over time. They are useful for relatively long
sounds but less effective for brief sounds such as pile-driving strikes
and echolocation clicks of whales. Peak-to-p eak values in the
amplitude waveform provide an alternative measure, but compar-
isons between peak-to-peak and RMS levels are difficult [96].
Recently, investigators have adopted another so-called Sound
Exposure Level (SEL), which is an alternative measure reflecting the
total acoustic energy received by an organism [13,14]. A final issue of
critical importance for understanding underwater sound as it relates
to fish is the presence of a substantial particle motion component in
the aquatic sound field, along with pressure. Since water is so dense,
particle motion is a component of the sound field at all distances and
fish are adapted to detect this component (see Box 3).
Review
Trends in Ecology and Evolution Vol.xxx No.x
TREE-1243; No. of Pages 9
2
behavioural and physiological impact of human-generated
sounds among very different vertebrate groups.
Third, it is important to stress that some fundamental
insight into underwater acoustics is critical for a proper
understanding of the problem with noise (Box 2).
For exa mple, the world of fish has bee n wrongly
assumed to be quiet, as reflected by the title of Jacques
Cousteau’s 1956 movie The Silent World. The supposi-
tion that the underwater world is silent no doubt
arose because sound transmission from water to air
is poor, and because the air-adapted human ear is a
relatively poor receiver underwater. In thinking about
underwater acoustics, it is also important to realize
that aquatic animals often live in a dark or turbid
environment: even a few metres from the animal the
use of vision to gather information becomes restricted. In
contrast, sound is not restricted by low light levels or
objects in the environment, while many aquatic animals,
including all fish, have more or less advanced abilities to
hear ( Box 3).
Are there noise-dependent fish distributions?
If anthropogenic noise deters fish, or if noise is bad for fish
survival and reproduction, one might predict lower fish
diversity and density at noisy places. At the moment,
however, there are few studies that indicate such negative
correlations between the presence of noise and the pre-
sence of fish. Some studies report an effect of vessel noise
on fish flight behaviour in the context of population assess-
ments and catch rates for commercially important fish
stocks. For example, horizontal and vertical movements
away from vessels have been reported for Atlantic herring
(Clupea harengus) and Atlantic cod (Gadus morhua)
[17,18], presumably in response to ship noise. Another
example concerns effects of nearby boating noise on
blue-fin tuna (Thunnus thynnus) in large oceanic pens.
In the presence of boat noise, tuna schools were less
coherent than when the noise was not present and indi-
vidual fish often swam independently towards the surface
or the bottom [19] . Fish have also been reported to flee from
seismic shooting areas as inferred from decreased catch
Box 3. Fish ears and hearing abilities
All fish studied to date are able to hear sound s [15,97,98].
They have two sensory systems for detection of water motions:
the inner ear (there is no outer or middle ear) and the lateral line
system. The ear serves to detect sound up to hundreds or even
thousands of H z (depending on the species), whereas the lateral
line detects low-frequency sound (e.g. <100 Hz), but is generally
considered to be primarily a detector of water motion relative to
the body.
Sound can be thought of in terms of both particle motion and
pressure fluctuations. Sensory hair cells in the inner ear and lateral
line(bothofwhichareverysimilartothosefoundinthe
mammalian ear) are stimulated by mechanisms that respond to
particle motion and are responsible for converting these motions
to electrical signals that stimulate the nervous system. The lateral
line system is found along both sides of the body and typically
spreads out over the head region where it plays a dominant role in
the detection of water motion and low-frequency sound at short
distances (one or two body lengths). In contrast, the inner ear also
detects sounds of much higher frequencies and from greater
distances (probably via acoustic pressure since particle motion
declines with distance more rapidly).
Different fish species vary in absolute sensitivity and spectral range
of hearing (Figure I), which relates to an auditory detection continuum
based on presence or absence of specially evolved morphological
structures [15,97,98]. Special features that improve the pressure-to-
motion transduction from the swim bladder may involve gas-filled
cavities reaching the inner ear. There may also be a direct mechanical
connection between the swim bladder and the inner ear through a
series of bones (the Weberian apparatus) such as in a large group of
fish species (Otophysi) that includes goldfish (Carassius auratus) and
catfish. Generally speaking, fish hear best within 30–1000 Hz, while
species with special adaptations can detect sounds up to 3000–
5000 Hz. Some exceptional species are sensitive to infrasound or
ultrasound.
Figure I. Hearing ranges of selected fish and mammal species, reflecting some of
the typical variety in these taxonomic groups (for reviews see Refs [10,15]). The
vertical dashed lines demarcate the human hearing range in air. Each species has
a more restricted range of peak sensitivity within the species-specific limits (not
indicated). From top-to-bottom, red horizontal bars represent: European eel, a
freshwater species spawning at sea with sensitivity to infrasound; Atlantic cod, a
marine species with ‘average’ hearing abilities; and goldfish, representing many
freshwater fishes with specially evolved hearing abilities. For mammals in blue,
we included Californian sea lion (Zalophus californianus), bottlenose dolphin
(Tursiops truncatus), and fin whale (Balaenoptera physalus). The anthropogenic
noise ranges indicate where the majority of sound sources have most of their
energy, although some human-generated sounds exceed these frequencies. At
the bottom of the figure are frequency ranges of low-frequency (USA), mid-
frequency and high-frequency sonar.
Review
Trends in Ecology and Evolution Vol.xxx No.x
TREE-1243; No. of Pages 9
3
rates for both long lines and trawler fisheries [20,21].
However, there is also a study with direct observations
on reef fish that remain close to their territories after
exposure to seismic air-gun shooting [22].
In contrast to the little we know about effects of environ-
mental sound on fish behaviour, a good deal is known about
the potential impact of anthropogenic noise on bird beha-
viour. For a long time, declines in avian diversity and
density associated with highways have been attributed
at least partly to traffic noise [23]. The idea was supported
by a study suggesting that a negative effect was dependent
on the spectral overlap between traffic noise and birdsong
[24] and by several studies showing spectral flexibility in
bird species that do well under noisy conditions [6]. How-
ever, the best evidence for a negative impact of anthropo-
genic noise on birds comes from natural areas around
extraction stations associated with the gas and oil industry
[25,26]. Some extraction stations are noisy and others
quiet, and this subdivision is independent of the above-
ground variation in avian habitat characteristics. Con-
sequently, and in contrast to traditional road-impact
studies, the decline in bird breeding density and diversity
found in these studies can be attributed solely to the
impact of noise.
Whether similar results will be found for fish is not
known, but the avian results are highly suggestive of
questions that must be asked for fish, and they even
suggest ways to explore and answer such questions. For
example, it may be possible to investigate the impact of
noise on fish diversity and density by making use of the
maritime gas and oil industry. Similar to the above water
situation with birds, there are offshore platforms that have
high underwater noise levels due to compressor noise and
human activity, while others serve as more silent satel-
lites. Alternatively, freshwater systems, often being more
accessible, can be explored experimentally by using arti-
ficial noise sources in some locations that can be compared
to quiet, control locations. While data on sound conditions
and fish behaviour at these control locations may provide
insight into the potential for soundscape orientation
[15,16], such an experimental set-up would allow for test-
ing an impact on species community and relative densities
dependent on artificial noise levels and specific noise fea-
tures. The impact of anthropogenic noise on dispersal and
passage of migratory fish can be tested in a similar way in
canal and river systems [27,28].
Consequences for fish that remain in noisy waters
Notwithstanding the lack of proper monitoring data, fish
sometimes congregate, seeking shelter or food, at places
with artificially high noise levels. Anecdotal observations
on fish under noisy bridges or near noisy vessels indicate
that adverse effects are not necessarily overt and obvious,
but they do not tell us whether fish experience any negative
consequences related to the noise. For example, several
studies in captive fish have shown an increase in secretion
of the stress hormone cortisol during exposure to white
noise or simulated boat noise [29,30, but see 31]. Other
recent studies on potential indicators of stress in captive
fish report noise-related rises in heart rate [32] and
increased motility related to several blood parameters
reflecting increased muscle metabolism [33]. Although
one must be cautious in extrapolating to free-swimming
fish that may be able to leave areas of high stress, these
findings at least suggest that anthropogenic noise could be
a stressor in natural water bodies.
Noise-dependent stress, like other environmental stres-
sors, might affect growth and reproductive processes [34],
but this has hardly been investigated. A relatively old
study, in which the acoustics of the experiments were
poorly controlled and calibrated, suggested lower egg via-
bility and reduced larval growth rates in noisy fish tanks
compared to more quiet control tanks [35]. A more recent
and better study on rainbow trout (Oncorhynchus mykiss),
exposed to realistic noise levels for fish tanks in an aqua-
culture facility [36] showed no impact on growth, survival,
or susceptibility to disease, even over nine months of
exposure [31]. However, given the very limited number
of species investigated, it is not clear whether one can
extrapolate from captive rainbow trout to other species
that may differ in hearing ability and in the extent they
depend on sound for natural activities.
In addition to an impact on growth or reproduction
related to noise-determined physiological stress, anthro-
pogenic noise may also affect populations in a more indirect
way. Data on birds has shown that individuals that vary in
reproductive abilities, related to age, experience, or size,
may not be evenly distributed over noisy and quiet areas of
otherwise suitable habitat [37,38]. The relative absence of
more experienced and typically more productive males in
noisy territories means that habitat productivity for these
species diminished beyond the effect of a reduction in
number of territory holders. These results may be relevant
to fish since many species are territorial and have explicit
age-dependent size classes varying in productivity [39].
However, so far we lack any study looking at distribution of
size classes relative to noise levels.
Population productivity of noisy areas might not only be
affected by lower numbers or lower-quality individuals, but
might also decline due to lowered reproductive efficiency.
Data on frogs has shown, for example, that anthropogenic
noise may either increase or decrease calling activities
[40,41], with possible fitness consequences related to
increased energetic or predation costs or decreased mating
success. While there are no similar data yet for free-living
fish, a relatively old study reports on actual interruption of
spawning in roach (Rutilus rutilus) and rudd (Scardinius
erythrophthalmus) by an approaching fast-moving power-
boat [42]. Although obviously more data are required, it
should be realized that the mere presence of fish in noisy
waters does not necessarily mean that they are part of a
reproductively active population. A better insight can be
generated through studies on the impact of anthropogenic
noise on the rate and nature of reproductive behaviour and
acoustic signalling in free-living fish.
Masking of acoustic communication
A specific noise impact that could lead to lower reproduc-
tive efficiency for fish is masking of communicative sounds.
Over 800 species from 109 families are known to produce
sounds, while many more are suspected to do so [43–45].
The sounds that fish produce are, in most cases, broadband
Review
Trends in Ecology and Evolution Vol.xxx No.x
TREE-1243; No. of Pages 9
4
signals with most energy <500 Hz. Distinct variation in
spectral and temporal characteristics can be related to
species [46,47], populations [48], and gender [49]. Further-
more, graded variation in pitch and duration can be corre-
lated with size [47,50,51] or seasonal fluctuations in
motivation [52]. Such acoustic variation means that
sounds can serve as information carriers in acoustic com-
munication among fish [43,44,53]. That fish communicate
acoustically becomes evident from the contexts in which
the sounds are produced, such as during agonistic inter-
action in territorial fights, when competing for food, or
when being attacked by a predator [54–57].
However, the most common context in which fish are
known to produce sounds is in spawning aggregations
[58,59] and courtship interactions [60,61]. Although often
not explicitly demonstrated, sounds could serve in aggre-
gating reproductive groups, in which they may contribute
to synchronization of male and female gamete release [62].
At a more individual level, sounds could attract potential
mates to a specific place for courtship or egg shedding [63].
Recent experimental evidence has unequivocally shown
that sounds can modify mate choice decisions in fish.
Female haplochromine cichlids (Pundamilia nyererei ) pro-
vided with a choice between two males, matched in size and
colour, preferred to interact with the male associated with
playback of conspecific sounds [47]. An acoustic impact on
sexual preferences was also inferred for Atlantic cod in
which the male drumming muscle mass was correlated
with mating success [64]. Although these examples
strongly suggest acoustic communication occurs in fish,
there is a substantial lack of insight into the distribution
and nature of the phenomenon across species and across
habitats (from shallow waters to the deep sea).
Clearly, however, if fish sounds serve a communicative
function in a reproductive context, problems of detection
and recognition due to the presence of anthropogenic noise
[65–67] could have fitness consequences. It should be clear
that fish have not evolved in a quiet environment, and
natural noise levels can also become loud, for example
during fish choruses [58,59]. Nevertheless, playback of
field recordings under laboratory conditions, at natural
spectral content and level, confirmed experimentally that
noise generated by a cabin-cruiser type of boat can signifi-
cantly increase detection threshold levels for conspecific
sounds in both brown meagre drums (Sciaena umbra ) and
Mediterranean damselfish (Chromis chromis) [68]. Based
Figure 2. (a) Spatial distribution of sender communication space for uniformly distributed right whales listening to a 200 Hz conspecific call from a male individual in the
centre of the space in the presence of noise from a ship with a source level of 172 dB re 1 mPa. Blue dots indicate receivers for which the whales are likely to detect the sound
produced by the focal individual, while red dots indicate receivers for which the noise of the ship exceeds the signal beyond detectable levels. Left: the ship is approaching
the calling whale from the northeast causing a 6% decrease in the sender’s communication area. Right: the ship is within 2 km to the northwest of the caller causing a 97%
decrease in the sender’s communication area [87]. (b) Loss of communication range due to a rise in anthropogenic noise relative to historical conditions. Received sound
levels for 20 Hz fin whale calls are depicted against a background of natural noise levels (in blue), yielding an audible range of 1000 km, and elevated noise levels due to
anthropogenic influences (in red), leading to a reduced audible range of only 10 km. The graph is modified, with permission, after a model by Peter Tyack. He incorporated
realistic signal attenuation during propagation through the ocean in his model and assumed a 90 dB re 1 mPa noise floor for the pre-industrial ocean, currently elevated by
20 dB from shipping (which is still a conservative estimate likely to be met at many places).
Review
Trends in Ecology and Evolution Vol.xxx No.x
TREE-1243; No. of Pages 9
5
![Figure 2. (a) Spatial distribution of sender communication space for uniformly distribu centre of the space in the presence of noise from a ship with a source level of 172 dB re 1 produced by the focal individual, while red dots indicate receivers for which the noise of the calling whale from the northeast causing a 6% decrease in the sender’s communicat decrease in the sender’s communication area [87]. (b) Loss of communication range du levels for 20 Hz fin whale calls are depicted against a background of natural noise level anthropogenic influences (in red), leading to a reduced audible range of only 10 km. The realistic signal attenuation during propagation through the ocean in his model and assu 20 dB from shipping (which is still a conservative estimate likely to be met at many pla](/figures/figure-2-a-spatial-distribution-of-sender-communication-atnyk42q.webp)
