Content uploaded by Julius Piercy
Author content
All content in this area was uploaded by Julius Piercy on Dec 03, 2014
Content may be subject to copyright.
MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 516: 35– 47, 2014
doi: 10.3354/meps10986 Published December 3
INTRODUCTION
The connection between habitats and their sound-
scape has been recognized for many decades (Car-
son 1962, Krause 1987), but the true scope of the eco-
logical function of soundscapes and its importance to
the fitness of the individual has only emerged over
the past 10 yr (e.g. Slabbekoorn & Peet 2003, Simp-
son et al. 2005). Sound is used as a cue for identifying
and locating targets across many taxa, from parasitic
flies that lay eggs on singing crickets (Cade 1975)
and male mosquitoes in search of mates (Göpfert &
Robert 2000) to penguins locating parents and part-
ners amongst the cacophony of a squawking colony
(Aubin & Jouventin 2002). The use of natural sound-
scapes as an orientation cue to locate breeding
grounds and settlement habitats has gained increas-
ing interest over recent years (Slabbekoorn & Bouton
2008). Newts have been found to use the vocaliza-
tions of other anurans to guide them to permanent
ponds in which to breed (Diego-Rasilla & Luengo
2004, Pupin et al. 2007), and birds select suitable
nesting habitats based on the calls of heterospecifics
(reviewed by Mönkkönen & Forsman 2002).
© Inter-Research 2014 · www.int-res.com*Corresponding author: jjbpie@essex.ac.uk
Habitat quality affects sound production and likely
distance of detection on coral reefs
Julius J. B. Piercy1,*, Edward A. Codling1, 2, Adam J. Hill3, David J. Smith1,
Stephen D. Simpson4
1School of Biological Sciences, University of Essex, Colchester CO4 3SQ, UK
2Department of Mathematical Sciences, University of Essex, Colchester CO4 3SQ, UK
3Department of Engineering, University of Derby, Derby DE22 1GB, UK
4Biosciences, College of Life and Environmental Sciences, University of Exeter, Exeter EX4 4QD, UK
ABSTRACT: The interwoven nature of habitats and their acoustic fingerprints (soundscapes) is
being increasingly recognized as a key component of animal ecology. Natural soundscapes are
crucial for orientation in many different taxa when seeking suitable breeding grounds or settle-
ment habitats. In the marine environment, coral reef noise is an important navigation cue for set-
tling reef fish larvae and is thus a possible driver of reef population dynamics. We explored reef
noise across a gradient of reef qualities, tested sound propagation models against field recordings
and combined them with fish audiograms to demonstrate the importance of reef quality in deter-
mining which reefs larvae are likely to detect. We found that higher-quality reefs were signifi-
cantly louder and richer in acoustic events (transient content) than degraded reefs, and observed
that sound propagated farther with less attenuation than predicted by classic models. We discuss
how zones of detection of poor-quality reefs could be reduced by over an order of magnitude com-
pared to healthy reefs. The present study provides new perspectives on the far reaching effects
habitat degradation may have on organisms that utilize soundscapes for orientation towards or
away from coral reefs, and highlights the value of sound recordings as a cost-effective reef survey
and monitoring tool.
KEY WORDS: Underwater soundscape · Anthropogenic impact · Larval fish · Coral reef ·
Settlement habitat · Passive acoustic monitoring
Resale or republication not permitted without written consent of the publisher
This authors' personal copy may not be publicly or systematically copied or distributed, or posted on the Open Web,
except with written permission of the copyright holder(s). It may be distributed to interested individuals on request.
Mar Ecol Prog Ser 516: 35–47, 2014
Recent research has demonstrated the critical role
that coral reef soundscapes play in the recruitment of
settlement-stage larval fish, crustaceans and corals
(Simpson et al. 2005, Montgomery et al. 2006, Arved-
lund & Kavanagh 2009, Vermeij et al. 2010). Eggs
and hatchlings of most tropical and temperate reef
fish and decapod crustaceans are exported from their
coastal habitats and develop as larvae in the open
ocean (Montgomery et al. 2001, Kingsford et al. 2002,
Leis & McCormick 2002). Species abundance and
community structure are largely driven by the supply
of these pelagic larvae (Doherty & Fowler 1994, Lee
& Bruno 2009), and therefore understanding the
environmental factors that regulate the supply of lar-
vae to coral reefs is crucial to conservation and fish-
eries management. Whilst there are still major gaps
in our knowledge concerning the ecology of oceanic
larvae, often referred to as the ‘pelagic black box’
(Leis & McCormick 2002), it is now known that larvae
of many species across taxa use noise generated by
resident animals as one of the cues to identify settle-
ment areas (Simpson et al. 2004, 2005, 2008, 2011a,
Radford et al. 2007, Vermeij et al. 2010). Since reefs
and nearby nursery habitats produce unique signa-
ture sounds (Kennedy et al. 2010, Huijbers et al.
2012), there is potential for larvae to be able to distin-
guish between them to select habitat (Stanley et al.
2010, 2012, Radford et al. 2011a). Studies have
demonstrated that soundscapes can be used to esti-
mate biodiversity in terrestrial habitats (Sueur et al.
2008, Pijanowski et al. 2011a) which has led to the
hypothesis that they can provide information on
habitat quality and integrity (Krause 2002, Dumyahn
& Pijanowski 2011). However, no studies have specif-
ically looked at sound signatures across a gradient of
habitat quality in either terrestrial or marine systems.
Defining the relationship between reef quality and
acoustic signature would allow researchers to formu-
late new predictions on the long-term implications of
environmental degradation on patterns of recruit-
ment by animals using auditory cues, and validate
the use of sound recordings as a cost effective and
rapid assessment tool for reef quality.
Many reef organisms are relatively sedentary dur-
ing their adult life and thus the ‘decision’ to settle at
a specific site for a larva returning from an early
pelagic developmental phase has consequences for
the success and fitness of the individual. Audition,
together with olfaction, has emerged as one of the
key senses that provide larvae with the ability to ori-
entate towards a reef (see review by Arvedlund &
Kavanagh 2009). The value of audition depends on
species’ hearing capabilities and characteristics of
the sound source (Montgomery et al. 2006, Mann et
al. 2007). Sound can provide information on both the
identity and direction of the source over long dis-
tances (reefs can be detected up to 15 km away with
a hydrophone; Cato & McCauley 2002) and, unlike
olfactory cues, is largely unaffected by water cur-
rents. Larval fish and decapod crustaceans, however,
might detect reefs only over shorter distances (in the
range of hundreds of metres; Mann et al. 2007)
because of their low sensitivity and because most fish
only detect the particle motion component of sound
which decays faster than the particle pressure com-
ponent (Kalmijn 1988, Rogers & Cox 1988). Despite
an increase in the number of studies demonstrating
the use of sound cues by fish and crustacean larvae
for orientation towards reefs (see Simpson et al. 2004,
2005, 2008, 2011a, Radford et al. 2007, Stanley et al.
2010, 2012), the distances over which larvae detect
and discriminate between reefs are uncertain (Mann
et al. 2007, Wright et al. 2010, 2011, Radford et al.
2011b).
The importance of distance of detection was first
highlighted in models of larval recruitment, where
simulations demonstrated that the size of the ‘sensory
halo’ around reefs and the differential sensory abili-
ties of larvae strongly affected simulated fish recruit-
ment to a stylized reef (Armsworth 2000, Codling et
al. 2004, Leis 2007, Staaterman et al. 2012). Several
studies have since focused on characterizing reef
sound propagation (Mann et al. 2007, Radford et al.
2011b) and have explored interspecific differences in
the auditory thresholds of fish larvae (Wright et al.
2010, 2011). Recent studies have shown identifiable
differences in the sound produced by different tem-
perate (Radford et al. 2010) and tropical (Huijbers et
al. 2012) coastal environments. However, as yet, little
attention has been paid towards characterizing
sound produced within reef habitat, although a study
by Kennedy et al. (2010) revealed that particular
components of the soundscape could be utilized
to discriminate broad reef characteristics. Those
authors found that high-frequency sound correlated
with coral diversity at 11 Panamanian reefs surveyed
in the same year, likely due to increased numbers of
snapping shrimp, which are known to associate with
both coral and coral rubble (Hultgren & Duffy 2010,
Enochs et al. 2011). Their findings also supported sig-
nificant correlations between broad-spectrum root
mean square (RMS) levels and density of the sonifer-
ous damselfish Stegastes flavilatus, one of the most
abundant fish on the reefs studied. Comparisons of
recordings taken in 2007 with a further 30 sites
surveyed 3 yr earlier confirmed these findings and
36
Author copy
Piercy et al.: Reef habitat quality affects sound production
revealed some additional associations between the
reef community and the soundscape.
We investigated the importance of reef quality in
shaping the surrounding soundscape with 2 key
objectives in mind: (1) to characterize sound at reefs
across a gradient of habitat quality utilizing short
‘snapshot’ recordings and (2) to evaluate models of
underwater sound propagation using acoustic
recordings taken at different distances from coral
reefs. Considering the role of sound in recruitment,
we assumed that larvae attempting to locate reefs are
unlikely to rely on sound intensity alone to gauge
their distance from a reef or to distinguish between
reefs if other acoustic information is available. We
therefore considered sound intensity combined with
transient content, a measure of the rate of short
energy bursts in the sound, characteristic of the
clicks produced by snapping shrimp that dominate
many reef soundscapes. Our findings suggest that
reef quality substantially modifies the intensity and
transient content of the reef’s soundscape, with the
magnitude of this effect possibly large enough to
greatly undermine potential distance over which
recruitment-stage larvae can detect impacted reefs.
We suggest that the ‘snapshot’ approach to sound
recordings could be useful as an objective and cost-
effective tool to assess and monitor reef condition.
MATERIALS AND METHODS
Sites
Recordings across a gradient of habitat quality
were taken around 7 small islands off Bohol in the
central Philippines (see Fig. S1 in Supplement 1,
available at www.int-res.com/ articles/suppl/ m516
p035_ supp.pdf), where reefs with a range of habitat
qualities, from coral- to algae-dominated reefs, were
studied. The sites consisted of (1) 3 healthy reefs in
marine protected areas (MPAs; CCEF 2007, UPMSI
2007): Balicasag Island (BCBF), Pamilacan Island
(PMSY) and Bilang-bilangan (BB); (2) 1 recovering
reef recently designated as an MPA: Pangapasan
Island (PG); and (3) 3 impacted reefs dominated by
macroalgae or urchins: BHIN, BHOUT and CBIN (S.
Green pers. comm.; see Table 1 for site abbreviations
and descriptions). Further details including coral
cover and fish density are provided in Table 1.
Our study of sound propagation focused on reefs in
Oman and Indonesia representing extremes in biodi-
versity and bathymetry. The Omani reefs are located
on shallow shelving coastlines in the Masirah Chan-
nel (depth <20 m; Fig. S2 in Supplement 1) and are
characterized by high coral cover dominated by a
single coral species, Montipora foliosa (Claereboudt
2006). In contrast, the fringing reefs of Hoga Island,
in the Wakatobi Marine National Park, south-east
Sula wesi, Indonesia, are surrounded by waters
>100 m deep and are located in the centre of the
Coral Triangle, one of the most biodiverse regions in
the world (Veron 1995, McMellor & Smith 2010;
Fig. S3 in Supplement 1).
Sound recordings
Recordings were made in February 2005 (Oman),
June 2007 (Philippines) and June 2009 (Indonesia)
using a calibrated omnidirectional hydrophone
(HiTech HTI-96-MIN with inbuilt preamplifier, High
Tech) and either a calibrated Edirol R-1 recorder
(Roland Systems Group; Oman recordings) or Zoom
H4 recorder (Zoom Corporation; Philippines and
Indonesia recordings), both at 24-bit, 44.1 kHz sam-
pling rate. The hydrophone was suspended from the
boat, with the cable held by the person making the
recording to dampen out any movements caused by
the boat bobbing (which was negligible due to work-
ing in calm conditions). Depth was measured by
weighted rope when less than 20 m, but otherwise
taken from charts (e.g. Hoga Island). Recordings
were made 5 m below the surface, as this escapes the
very different conditions of the 1 m closest to the sur-
face, and is where fish are likely to swim when seek-
ing habitat (e.g. Leis et al. 1996). For the study on
soundscapes across a gradient of different quality
reefs, two 1 min recordings separated by >1 h were
taken for each of the 7 sites (20 m from the reef) in the
central Philippines (between 09:00 and 13:00 h). The
daytime ‘snapshot’ approach (two 1 min recordings
1 h apart) was adopted so that meaningful compar-
isons with other widely adopted census methods,
such as underwater visual censuses (UVCs), could be
made at these sites.
In the second part of this study where propagation
models were compared with real sound recordings, 3
seaward transects up to 1.5 km away from the reef,
each consisting of up to 5 recording points, were
completed in Oman (Fig. S2a in Supplement 1): one
East of Barr al Hickmann (BAHE; Fig. S2b in Supple-
ment 1) and 2 south of Masirah Island (MIS1 and
MIS2; Fig. S2c in Supplement 1) and 2 seaward tran-
sects were taken at Hoga (Fig. S3a in Supplement 1):
Front Beach (FB) and Pak Kasim’s (PK; Fig. S3b in
Supplement 1). One minute recordings were taken
37
Author copy
Mar Ecol Prog Ser 516: 35–47, 2014
38
Site Location MPA status Site description
Philippines
Balicasag Island 9° 30’ 48” N, MPA and no-take High coral cover (70%) and fish density
‘Black Forest’ 123° 41’19” E zone; established (7208.5 fish 500 m−2); Management Level 2
Fish Sanctuary 1986 rating (established), low fishing beyond
(BCBF) boundary (White et al. 2007)
Bilang-bilangan 9° 59’ 31”N, MPA and no-take Fair coral cover (55.3%), medium fish density
Marine Sanctuary 123° 52’48” E zone; established (1396.6 ± 290.1 fish 500 m−2); Management
(BB) 1999 Level 4 rating (sustained), high fishing pressure
outside protected zone (CCEF 2007)
Pamilacan Island 9° 29’ 16”N, MPA and no-take Fair coral cover (36.8 %), medium fish density
Fish Sanctuary 123° 55’ 58”E zone; established (1285 ± 218.7 fish 500 m−2); Management Level 4
(PMSY) 1985 rating (sustained), medium fishing pressure
outside protected zone (White et al. 2007)
Pangapasan Island 9°59’ 52” N, MPA and no-take Poor coral cover (18.4 %), fair fish density
Fish Sanctuary 123° 56’ 37”E zone; established (428.55 ± 71.7 fish 500 m−2); Management
(PG) 1998 Level 3 (enforced); regular boundary breaching
by fishermen (CCEF 2007)
Reef in the 10° 0’ 43”N, No official Low coral cover, low fish density, macroalgae-
Bohol-Cebu 123° 55’40” E management or dominated reef; historically overfished through
Channel (BHIN) protection status destructive methods; no active management
(S. Green pers. comm.)
Reef in the 10° 1’ 10”N, No official Low coral cover, low fish density, macroalgae-
Bohol-Cebu 123° 56’15” E management or dominated reef; historically overfished through
Channel (BHOUT) protection status destructive methods; no active management
(S. Green pers. comm.)
Reef in the 10° 1’ 25”N, No official Low coral cover, low fish density, urchin-
Bohol-Cebu 123° 55’49” E management or dominated reef; historically overfished through
Channel (CBIN) protection status destructive methods; no active management
(S. Green pers. comm.)
Oman
Barr al Hickmann 20°20’ 53” N, No official High coral cover (70−100%), high fish density;
(BAHE) 58° 27’2” E management or monospecific coral reef consisting of tabulate
protection status Montipora foliosa; commercial fishing prohibited
by local fishing right holders of the Hickmani
tribe (Salm 1993, Claereboudt 2006)
Masirah Island 1 20° 9’52”N, No official High coral cover (50−100%), high fish density;
(MIS1) 58° 37’55”E management or offshore monospecific Pocillopora reef with high
protection status densities of urchins and parrotfish; low fishing
activity (Salm 1993, MRME 1996,
Claereboudt 2006)
Masirah Island 2 20° 9’52”N, No official High coral cover (50−100%), high fish density,
(MIS2) 58° 37’31”E management or coral reef dominated by tabulate Montipora
protection status foliosa with some monospecific Pocillopora
stands and Acropora thickets; low fishing
activity with hook and line method (Salm 1993,
MRME 1996, Claereboudt 2006)
Indonesia
Pak Kasim 5° 27’50” S, Marine National Fair coral cover (27−42 %), fair fish density
(PK) 123°45’ 18” E Park (~875 fish 500 m−2); high diversity of coral and
fish; 2001−2012 no active management (Smith &
Jompa 2009, McMellor & Smith 2010)
Front Beach 5°28’20” S, Marine National Fair coral cover (45−55 %), medium fish density
(FB) 123° 45’25” E Park (~1020 fish 500 m−2); high diversity of coral and
fish; 2001−2007 no-take zone; 2008−2012 no
active management (Smith & Jompa 2009,
McMellor & Smith 2010)
Table 1. Geographical location, Marine Protected Area (MPA) status and description of sites used for recording sound for
habitat comparisons and sound propagation models
Author copy
Piercy et al.: Reef habitat quality affects sound production
for each transect point (Oman and Hoga) at 4 m
below the sea surface, and all recordings were made
between 13:00 and 15:00 h during the dry season
(WMO 2013). Distance from the reef was measured
using a handheld GPS navigator (Garmin GPSMAP
60CSx), and satellite images were used to ensure
that no charted reefs were present within 1.5 km
from the sampled reef point. However, at the 2 far-
thest transect points taken from MIS3 (1.2 to 1.5 km),
a reef was clearly audible in the vicinity and so these
were removed from the analyses.
All samples considered in this study were carried
out in Force 1 or 2 conditions (Beaufort scale, wind
speeds <11 km h−1), which cause no wave break and
have negligible effects on background noise (Stewart
2004). However, this meant that the most distant
sampling point for BAHE had to be removed due to
wave break, likely caused by the site having less
protection from the wind and/or local currents. Since
temporal variability in reef sound is predictable
(<2 dB re 1 µPa during the daytime; Lammers et al.
2008), standardized sampling times of between 09:00
and 15:00 h were used to limit within-site variation
(i.e. excluding dawn, dusk and night; Cato 1978,
McCauley & Cato 2000, Cato & McCauley 2002,
Lammers et al. 2008).
Acoustic analyses
Recordings were divided into subsamples (10 s),
and those with obvious anthropogenic noise (passing
boats, waves slapping on the hull of the boat) were
removed from the analyses (as per Kennedy et al.
2010). The remaining recording samples were high-
pass (0.1 kHz) and low-pass (5 kHz) filtered in Adobe
Audition (CS5.5 V4, Adobe Systems) to remove elec-
trical noise from the recording equipment (high-pass
filter) and to limit the analysis to the upper hearing
ranges of coral reef fish (Wright et al. 2010, 2011).
RMS sound intensity and power spectra of each
recording were calculated in Adobe Audition and
calibrated according to the full dynamic range meas-
ured in Avisoft SASLab Pro (Avisoft Bioacoustics).
Transient content, an indicator of the rate of high-
intensity broadband impulsive sounds in the record-
ing, was calculated using a custom designed algo-
rithm in Matlab (v R2010a, The MathWorks; more
details in Supplement 2). The transient content is
independent of the RMS sound intensity and effec-
tively excludes the possible effect of stochastic fish
vocalizations by considering only short broadband
signals.
Modelling sound propagation
Four models of sound propagation were used to
generate predictions that were compared with the
seaward transect recordings (Oman and Indonesia)
taken from the reef up to 1500 m away. Differences
in frequency-dependent propagation loss due to
absorption were not considered since they are negli-
gible (<0.3 dB km−1) within the frequencies (0.1 to
5 kHz) and the distances (<1500 m) considered
(calculations obtained using absorption equation
described by Ainslie & McColm 1998). The models
used were (1) spherical spreading and (2) cylindrical
spreading of sound in water (see Mann et al. 2007 for
a description of both of these models); (3) ‘extended
reef’ model (Radford et al. 2011b), which considers
reefs as extended sources of sound rather than point
sources; and (4) geometric spreading parameterized
by our transect recordings: this model is based on
geometric spreading, where 2 parameters, αand β,
are estimated from the best fit to each of our record-
ings and then averaged between reefs (see Supple-
ment 3).
RESULTS
Spectral analysis of reef noise
Descriptive characterization of the sound across
different frequencies was performed by visual
inspection of the sound spectrograms for all sites
(Fig. 1 and see Fig. S6 in Supplement 4). All sites in
the Philippines had a broadband peak in the range
1.5 to 4 kHz dominated by invertebrate noise, most
likely the ‘clicking’ of snapping shrimp, but each site
also had distinctive power spectra (Fig. 1a). The reef
with highest coral cover and fish abundance (BCBF)
had the highest intensity of sound across all frequen-
cies including those <1 kHz, which consisted of a
variety of fish vocalizations (‘pops’, ‘grunts’, ‘croaks’
and ‘drums’) audible throughout the recordings. The
other 2 healthy sites (BB and PMSY) also had high
levels of higher-frequency invertebrate ‘clicking’
noise (broad peak between 1 and 5 kHz centred on
~3.8 kHz), but relatively lower power in the frequen-
cies associated with fish vocalizations (<1 kHz). The
broad peak in the 0.2 to 0.8 kHz range at PG resulted
from multiple high-energy ‘pops’ and ‘grunts’, indi-
cating a high abundance of vocalizing fish. The 3 his-
torically overfished and now depauperate sites gen-
erally had much lower power across all frequencies,
although BHOUT (one of the macroalgae-dominated
39
Author copy
Mar Ecol Prog Ser 516: 35–47, 2014
reefs) had a sharp peak between 0.1 and 0.2 kHz,
unlikely to be of natural origin and possibly caused
by a distant anthropogenic source.
Transect recordings from Oman had distinct
power spectra in both the higher (>1 kHz) and
lower (<1 kHz) frequency regions compared to sites
in Indonesia (Fig. 1b,c). Despite similarities in the
frequency composition between sites located in
similar geographical regions, the power spectra for
each site was distinct both for the high-frequency
and low-frequency components, suggesting differ-
ent fish and invertebrate assemblages at each site
(Fig. 1b,c and see Fig. S6 in Supplement 4). For
each site, the power spectra were similar across all
ranges for frequencies >1 kHz, decreasing reliably
in power as range increased. There was greater
variability in the pattern of the power spectra at
frequencies <1 kHz, due to a combination of back-
ground noise and vocalizations from non-reef
dwelling fish. However, overall there was a
decrease in power over distance across most of the
lower frequencies as well (Fig. 1b,c and see Fig. S6
in Supplement 4).
Sound intensity and transient content across a
habitat gradient
To quantitatively investigate differences in sound
between reefs across a gradient of habitat qualities,
we adopted 2 simple measures: sound intensity and
transient content. Average RMS sound intensity
between 0.1 and 5 kHz obtained from 10 s subsam-
ples (n = 5, randomly selected from two 1 min record-
ings taken 1 h apart) from the recordings of the 7
Philippine reefs was consistently different between
some sites (1-way ANOVA, F= 350.37, p < 0.001,
Fig. 2a). Two of the high-quality MPAs (BB and
PMSY) and the 2 lowest-quality reefs (BHIN and
CBIN) were each grouped together, while the other 3
sites were significantly isolated from all other groups
(BCBF > BB = PMSY > PG > BHOUT > BHIN = CBIN,
Tukey’s test, p < 0.05).
Transient content decreased significantly (Kruskal-
Wallis [K-W], H= 31.1, p < 0.001) with decreasing
habitat quality (Fig. 2b). Post hoc tests revealed that
the transient contents of the 3 low-quality reefs did
not differ significantly from one another but were
40
65
70
75
80
85
90
95
100
100 1000
Spectrum level (dB re 1 µPa2/Hz) Spectrum level (dB re 1 µPa2/Hz)
aCentral Philippines reefs
PMSY
BCBF
BB
BHOUT
BHIN
CBIN
PG
85
90
95
100
105
100 1000
Frequency (Hz)Frequency (Hz)
bOman reef: Masirah Island 1 (MIS1)
220 m
486 m
764 m
1216 m
1493 m
147 m
305 m
505 m
785 m
1280 m
65
70
75
80
85
90
100 1000
Frequency (Hz)
c Indonesian reef: Pak Kasims (PK)
Fig. 1. Spectrum level of recordings from sites in the Philip-
pines, Oman and Indonesia. Frequency peaks <0.8 kHz
indicate high levels of fish vocalizations at the reef whilst the
broad peaks >1 kHz are indicative of invertebrate noise,
produced mainly by the abundant high-energy ‘snaps’ from
snapping shrimps at all sites. (a) Recordings from 7 reefs in
the central Philippines; darker lines indicate higher-quality
reefs. (b,c) Representative spectrum levels of recordings
taken along seaward transects away from reefs in (b) Oman
and (c) Hoga, Indonesia (see Fig. S6 in Supplement 4 for the
other 3 transects). Darker lines indicate recordings closer to
the reef. See Table 1 for site abbreviations and descriptions
Author copy
Piercy et al.: Reef habitat quality affects sound production
significantly lower than 2 of the high-quality reefs
(BCBF and PMSY > BHIN, CBIN and BHOUT, K-W
all pairwise comparisons by site, p < 0.01; Fig. 2b).
With the exception of one high-quality site (BB),
ranking by transient content was the same as that by
sound intensity. In further investigations, ranking of
reefs by transient content varied little when a range
of window bin sizes and thresholds were applied (see
Figs. S4 & S5 in Supplement 2).
Evaluation of propagation models using transect
sound recordings
All 5 transects performed across Oman and Indo -
nesia displayed characteristic non-linear de clines in
broadband (0.1 to 5 kHz) RMS sound intensity with
increasing distance from the reef (Fig. 3). Source in-
tensities were higher at Omani reefs (mean RMS at
the sites ranged from 123 dB re 1 µPa at BAHE to
133 dB re 1 µPa at MIS1) compared to Indonesian
reefs (mean RMS of 110 and 111 dB re 1 µPa at PK
and FB, respectively) and attenuated more gradually
with distance at Omani reefs (attenuation of mean
RMS at the sites ranged between 5.6 and 8.0 dB re
1 µPa over ~800 m) compared to Indonesian reefs (9.6
to 10.5 dB re 1 µPa over the same distance; Table S1
in Supplement 3). For each site, predictions of sound
prop agation based on spherical spreading, cylindrical
spreading, the ‘extended reef’ and best fit recording-
parameterized geometric spreading models were
compared to measured intensities (Fig. 3). No sig -
nificant dif ference was found between cylindrical
spreading, extended reef or recording-parameterized
geometrical spreading models in their goodness of fit
to sound intensities at different sites but spherical
spreading had a significantly poorer fit for all sites (2-
way ANOVA, F= 64.4, p < 0.001).
Spherical and cylindrical spreading both underesti-
mated sound propagation, most likely due to the sim-
plifying assumption that sound is propagating from a
point source. The extended reef model ex plained
propagation well for the smallest coral reef in this
study (MIS1, 600 m in length), but underestimated
sound intensity at the source for all other sites and
would consequently overestimate sound intensity
over distance if propagated from measured source
levels. The exception to this would be BAHE, where
the sound intensity at 300 m was underestimated by
all 4 models, possibly due to noise from uncharted
reefs. The mean ± SD αβ coefficient calculated for the
geometric spreading model was 8.28 ± 1.70. This coef-
ficient indicates a lower attenuation in sound intensity
compared to cylindrical spreading (coefficient of 10),
but a greater attenuation than predicted by the ex-
tended reef model except at MIS1 for the distances
examined. There was a significant decrease in tran-
sient content levels with distance for all reefs (2-way
ANOVA; MIS1, F= 72. 6, R2= 0.76, p < 0.001; MIS 3,
F= 37.3, R2= 0.84, p < 0.001; FB, F= 96.1, R2= 0.89, p <
0.001; PK, F= 88.0, R2= 0.87, p < 0.001; Fig. 4) with the
exception of 1 reef in Oman (BAHE), where the level
of transient content did not change significantly.
DISCUSSION
The present study demonstrates the influence of
habitat quality on acoustic output in coral reef envi-
ronments. Controlling for abiotic sound sources (e.g.
41
100
105
110
115
120
125
BCBF
(HQ) BB
(HQ) PMSY
(HQ) PG
(MQ) BHOUT
(LQ) BHIN
(LQ) CBIN
(LQ)
BCBF
(HQ) BB
(HQ) PMSY
(HQ) PG
(MQ) BHOUT
(LQ) BHIN
(LQ) CBIN
(LQ)
Mean RMS sound level (dB re 1 µPa)
A
a
b
0
10
20
30
40
50
60
70
Mean transient content (Hz)
A
B
D
C
E
E
B
C
D
Fig. 2. Mean ± SD of (a) root mean square (RMS) sound
intensity and (b) transient content at sites of different quality
in the Philippines. SDs are representative of 10 s subsamples
(n = 5) randomly selected from two 1 min recordings taken
1 h apart. Sites are denoted as high quality (HQ), medium
quality (MQ) or low quality (LQ) based on marine protected
area (MPA) status and habitat type (see Table 1). Groupings
A−E denote ranks of reefs by (a) broadband RMS sound
intensity and (b) transient content. Reefs grouped by letter
were not significantly different (panel a: Tukey’s tests; panel
b: Mann-Whitney U-tests). Site abbreviations as in Table 1
Author copy
Mar Ecol Prog Ser 516: 35–47, 2014
waves and wind Force 1−2 on the Beaufort scale,
wind speed <11 km h−1; Stewart 2004), reefs of
higher quality had higher sound intensity and were
richer in transient content during the daytime (be -
tween 09:00 and 15:00 h). We also evaluated the
validity of different propagation models in 2 different
biogeographic regions. We report that classical mod-
els of sound propagation (spherical and cylindrical
spreading) underestimate the distance over which
sound is broadcast, probably due to the simplifying
assumption that reefs are point sources of sound, as
previously suggested by Radford et
al. (2011b) for temperate reefs.
Soundscapes across habitat quality
Meaningful heterogeneity in reef
noise found in this study supports the
idea of using ‘snapshot’ sound record-
ings for rapid and cost-effective reef
quality assessment and monitoring
(Simpson 2008). Sound has already
been used in terrestrial environments
to gather biodiversity and abundance
estimates (Sueur et al. 2008, Pija -
nowski et al. 2011a). This method has
surpassed other sampling methods of
assessing individual occurrence in
birds and amphibians for locations
where visibility is poor (e.g. rainforest)
and also because of its potential for
continuous monitoring (Acevedo &
Villanueva-Rivera 2006, Celis-Murillo
et al. 2009, 2012, Marques et al. 2013).
Considering the close correlation
found between certain environmental
characteristics and acoustic finger-
prints of reefs (Kennedy et al. 2010), it
is likely that specific biological indica-
tors of reef quality could also be esti-
mated from acoustic recordings. This
concept has been applied previously
to monitor the presence/absence of
specific fish populations (Lobel 2001,
Mann 2012).
More recently, Staaterman et al.
(2013) recognized the need to con-
sider acoustic monitoring alongside
other assessment methods since the
soundscape plays an important role
in larval recruitment and in other
important ecological functions such
as mating and territorial defence. Similar to Staater-
man et al. (2013), we do not advocate that acoustic
assessment should replace classical assessment
methods. However, we suggest that ‘snapshot’
soundscape measurements could be used as a means
to sample a large number of sites in a cost- and time-
efficient manner. This could help identify priority
locations where more detailed and long-term assess-
ments might be necessary (i.e. to assess fish species
populations which vocalize at specific times of day)
in an unbiased manner.
42
112
114
116
118
120
122
124
0.1 1
c
e
d
ab
10
Barr al Hickmann
122
124
126
128
130
132
134
136
0.1 1 10
Masirah Island 1
120
122
124
126
128
130
132
134
0.1 1 10
Masirah Island 2
97
99
101
103
105
107
109
111
0.1 1 10
Pak Kasim's
96
98
100
102
104
106
108
110
0.1 1 10
Distance from reef (km)
Distance from reef (km)
Front Beach
Sound recording intensity
Parameterized geometric spreading
Spherical spreading
Cylindrical spreading
Extended reef model
RMS intensity (dB re 1 µPa)
Fig. 3. Mean ± SD RMS sound intensity at different distances from the reef at
5 sites in (a−c) Oman and (d,e) Indonesia. SDs are representative of 10 s sub-
samples (n = 5) randomly selected from two 1 min recordings. Fits of 4 propa-
gation models (spherical, cylindrical, extended reef and geometric spreading
parameterized using the mean αβ coefficient) are also displayed for each site.
The source sound is not represented due to geographical complexities which
arise in proximity to the reef (Radford et al. 2011b), and the first recordings are
taken at 150 to 200 m
Author copy
Piercy et al.: Reef habitat quality affects sound production
It is worth noting that although this study does
provide a valuable indication of the relationship be -
tween a reef’s soundscape and its habitat, future
studies would benefit greatly from more detailed
visual assessments. This includes fish biomass and
benthic cover, but also assessments of the cryptoben-
thos which may contribute significantly towards the
reef soundscapes through the snaps of cryptic snap-
ping shrimp and vocalizations of cryptic fish (e.g.
gobies; Tavolga 1956). It would also be of great value
for future acoustic assessments to investigate how
the soundscape changes before and after distur-
bances which greatly affect the reef’s assemblage,
such as tropical cyclones or crown-of-thorns starfish
Acanthaster planci outbreaks (Wolanski 1994). In this
context, it is possible to envisage acoustic monitoring
of the reef as an early warning system for A. planci
outbreaks or as a rapid assessment tool for reef
damage following cyclones.
Lastly, acoustic monitoring may prove to be a use-
ful tool in assessing nighttime assemblages of fish
and invertebrates where visual censuses are chal-
lenging due to the low light conditions and would
likely disturb the wildlife through the use of flash-
lights. The approach of nighttime acoustic monitor-
ing was recently validated by Freeman et al. (2014),
who coupled acoustic recordings with infrared cam-
eras to monitor the activity of invertebrate species on
coral reefs at night. Further studies, however, are
needed to validate its use as an estimator of night-
time fish communities, as acoustic cues are likely
to play a greater role in nocturnal fish than in day-
time fish, when visual cues also play a role in com -
munication.
Comparison of propagation models for coral reefs
In evaluating the fit of different propagation mod-
els to measured sound intensities, the spherical
spreading model (used by Mann et al. 2007) is least
supported as a useful model for predicting reef sound
propagation, and is possibly too conservative, over-
estimating attenuation for extended-source sounds
and shallow water environments. The concept of an
‘extended reef’ (Radford et al. 2011b) for modelling
sound propagation is supported in the present study
above models that assume reefs are a point source.
However, the ‘extended reef’ model only explained
sound propagation from transect recordings most
accurately in the smallest reef (MIS1), of a similar
size to the reef used to test the original model (500 m;
Radford et al. 2011b). Different parameterization to
the extended reef model may be required to charac-
terize the acoustics of reefs >500 m long. The geo-
metric spreading model parameterized using the
recordings in the present study explained sound
propagation well for 2 geographically distinct reef
systems with very different bathymetries (a shallow
shelving coastline in depths <20 m for Omani reefs
and a fringing reef surrounded by waters >100 m
deep for Indonesian reefs) and communities (low fish
and coral diversity in Omani reefs compared to some
of the highest diversity in coral and fish in the oceans
for Indonesian reefs; Claereboudt 2006, McMellor &
Smith 2010). By combining the extended reef model
with the parameterization approach from the present
study, it might be possible to obtain an improved
propagation model for future studies on reef
soundscapes.
43
0
ab
10
20
30
40
50
60
100 1000
Transient content (Hz)
BAHE
MIS1
MIS2
0
10
20
30
40
50
60
100 1000
Ran
g
e (m)
PK
FB
Fig. 4. Mean ± SD transient content values for recordings on seaward transects away from coral reefs in (a) Oman and (b)
Indonesia. The decline in transient content with increasing distance likely indicates homogenization of sound with distance
and a lower signal to noise ratio, with fewer distinguishable transient events due to interference between sound waves. Site
abbreviations as in Table 1
Author copy
Mar Ecol Prog Ser 516: 35–47, 2014
The present study made use of a single hydro -
phone to measure the sound at different distances
from the reef, but it cannot account for the natural
variability in sound at the source. Previously Radford
et al. (2011b) coupled the acoustic measurements
from different distances from a temperate reef with a
fixed hydrophone at the reef. By doing so, they were
able to obtain both spatial and temporal information
on the changes in reef sound during a period of high
temporal variation in sound production (dusk cho-
rus). In addition to this, promising recent develop-
ments have made use of L-shaped hydrophone
arrays to gain directional information on invertebrate
sound sources on coral reefs in the Eastern Pacific
(Freeman et al. 2013). Combined with topographical
mapping of the sea floor, these techniques would
provide us with the clearest picture yet on how sound
propagates from the reef and on what acoustical
information is available to larvae approaching the
reef.
Effects of reduced detection zones on reef animals
Finding an appropriate settlement site is a crucial
step in the life history of reef fish and decapod crus-
taceans (Leis & McCormick 2002, Montgomery et al.
2006). Self-recruitment, the process by which fish
return to their natal reefs, plays a central role in
maintaining supply of larval fish to reefs in MPA net-
works (Berumen et al. 2012, Harrison et al. 2012).
Self-recruitment of larvae may depend on their abil-
ity to detect and orient towards a reef early on in
development or at the time when they become com-
petent to settle (Fisher 2005, Staaterman et al. 2012).
Consequently, smaller detection zones in degraded
reefs could significantly reduce the ability of reef fish
populations to self-sustain and, therefore, compro-
mise the reef’s chances of recovery. If one considers a
propagation loss of 10 dB in sound intensity for every
10-fold increase in distance from the reef, as pre-
dicted by the cylindrical spreading propagation
model, a difference of 14 dB sound intensity at the
source, observed between the highest- and lowest-
quality reef in the present study, would result in a
reduction of the detection zone by over an order of
magnitude. A larger detection zone means that high-
quality reefs may receive higher rates of larval sup-
ply, effectively recruiting more fish from the sur-
rounding ocean, potentially at the expense of nearby
degraded reefs.
Furthermore, differential hearing abilities of larvae
could result in community shifts away from species
with lower hearing sensitivity at degraded sites, fur-
ther decreasing diversity, abundance and resilience
of fish assemblages. This also calls for new research
into the absolute hearing abilities of fish larvae,
which are currently confined to comparative studies
using auditory brainstem responses which are meas-
ured in tanks where the sound field is very complex
(Parvulescu 1967; but see also Akamatsu et al. 2002
for possible solutions to the complexities). The influ-
ence that habitat quality has on biogenic noise
generated on reefs and thus on the likely ability of
larvae to detect settlement sites (Kennedy et al. 2010,
present study) should be a key consideration when
designing and implementing conservation strategies
for degraded sites.
Our study highlights the fact that the acoustic
environment is richer in information than simple
quantitative variations in sound intensity. Transient
content is also a good descriptor of reef quality and
shows a meaningful decrease with distance from
reefs. Both sound intensity and transient content
generally showed similar trends with the exception
of 1 Omani reef (BAHE), where transient content
was variable over distance whilst sound intensity
decreased in a predictable fashion. The reason for
this is currently unclear since the homogenization
of the transient signals is likely to increase with
distance, resulting in a lower measured value of
transient content. However, the fact that in this
case the 2 measures do not appear to be directly
related to one another highlights the potential for
them to be de-coupled and possibly provide orient-
ing larvae with more information than simple
directional gradients. As long as the rate of change
between transient content and sound intensity is
independent, larvae could use the transient content
to sound intensity ratio to tease apart the distance
and quality of a reef (e.g. discern between close
but poor quality, and far but good quality reefs
with similar received sound intensities but contrast-
ing transient content). This theory warrants further
investigation to explore the potential for integrating
different types of acoustic information in animal
orientation.
It would be of great value and interest for future
investigations to explore both the geographical (as
per Staaterman et al. 2013 for the Caribbean) and
seasonal variation in soundscapes for these regions.
These investigations could be coupled with UVCs,
studies on larval recruitment patterns and experi-
ments on larval attraction to site- and time-specific
soundscapes. It is worth noting that if different
patterns of diel variation in reef noise between
44
Author copy
Piercy et al.: Reef habitat quality affects sound production
sites were present (as found between different
geographical regions by Staaterman et al. 2013),
this could also affect the distance from which fish
can detect different habitats. Different patterns of
diel variation in reef noise could potentially exac-
erbate (if sound intensity increases at a higher rate
for high-quality sites at night) or dampen (if sound
intensity increases at a lower rate or decreases at a
higher rate in high-quality sites at night) the effect
of habitat quality on the distance larval recruits
can detect a reef. This is especially relevant in the
context of larval recruitment to reefs since studies
to date have shown that fish larvae may respond
negatively (i.e. swim away) to reef sound during
the daytime but are attracted towards it at night
(Leis et al. 2003, Tolimieri et al. 2004). How these
factors interplay and the effect they may have on
recruitment and avoidance of reef noise remains to
be explored. Together with threats from ocean
acidification (Simpson et al. 2011b) and rising tem-
perature (O’Connor et al. 2007), our findings add
to an increasing number of studies highlighting
anthropogenic effects on larval recruitment, and
support the need to consider and preserve the
intrinsic value of soundscapes (Pijanowski et al.
2011b).
Finally, in the present study, sound intensity and
transient content were found to be reliable indicators
of reef quality and distance. By combining quantita-
tive measures (e.g. sound intensity) with qualitative
information of the signal (e.g. transient content) it
should be possible for larvae to exploit sound not
only for detecting settlement sites but also for choos-
ing between them and thus maximizing their success
in later life stages. Investigating the level to which
larvae can detect and integrate this complex array of
information presents some exciting new directions to
explore in our quest to shine light on the ‘pelagic
black box’.
Acknowledgements. We thank Gilles Lecaillon and Stuart
Green for logistical support and Mark Priest and Adel
Heenan for field assistance in the Philippines; Andrew Hal-
ford and Jennifer McIlwain for field assistance in Oman; and
Pippa Mansell and Operation Wallacea for logistical support
in Sulawesi. We thank Alan White, Rafa Martinez and
Augustus Montebon for valuable site information in the
Philippines. We are very grateful to Simon Wilson, who pro-
vided essential information on Omani reefs. We also thank
Emily Walker, Craig Radford, Marc Holderied, Andy Rad-
ford and Sophie Holles for valuable discussions regarding
our acoustic analyses. This work was funded by the Natural
Environment Research Council UK through a Postdoctoral
Fellowship and KE Fellowship to S.D.S. and a PhD Stu-
dentship to J.J.B.P.
LITERATURE CITED
Acevedo MA, Villanueva-Rivera LJ (2006) Using automated
digital recording systems as effective tools for the moni-
toring of birds and amphibians. Wildl Soc Bull 34:
211−214
Ainslie MA, McColm JG (1998) A simplified formula for vis-
cous and chemical absorption in sea water. J Acoust Soc
Am 103: 1671−1672
Akamatsu T, Okumura T, Novarini N, Yan HY (2002) Empir-
ical refinements applicable to the recording of fish
sounds in small tanks. J Acoust Soc Am 112: 3073−3082
Armsworth PR (2000) Modelling the swimming response of
late stage larval reef fish to different stimuli. Mar Ecol
Prog Ser 195: 231−247
Arvedlund M, Kavanagh K (2009) The senses and environ-
mental cues used by marine larvae of fish and decapod
crustaceans to find tropical coastal ecosystems. In:
Nagelkerken I (ed) Ecological connectivity among tropi-
cal coastal ecosystems. Springer, Dordrecht, p 135−184
Aubin T, Jouventin P (2002) How to vocally identify kin in a
crowd: the penguin model. Adv Stud Behav 31: 243−277
Berumen ML, Almany GR, Planes S, Jones GP, Saenz-
Agudelo P, Thorrold SR (2012) Persistence of self-recruit-
ment and patterns of larval connectivity in a marine pro-
tected area network. Ecol Evol 2: 444−452
Cade WH (1975) Acoustically orienting parasitoids: fly
phonotaxis to cricket song. Science 190: 1312−1313
Carson R (1962) Silent spring. Houghton Mifflin, New York,
NY
Cato DH (1978) Marine biological choruses observed in
tropical waters near Australia. J Acoust Soc Am 64:
736−743
Cato DH, McCauley RD (2002) Australian research in ambi-
ent sea noise. Acoust Aust 30: 13−20
CCEF (Coastal Conservation and Education Foundation)
(2007) Coastal Conservation and Education Foundation
MPA database. CCEF, Cebu City. www.coast.ph/our-
work/mpdatabase
Celis-Murillo A, Deppe JL, Allen MF (2009) Using sound-
scape recordings to estimate bird species abundance,
richness, and composition. J Field Ornithol 80: 64−78
Celis-Murillo A, Deppe JL, Ward MP (2012) Effectiveness
and utility of acoustic recordings for surveying tropical
birds. J Field Ornithol 83: 166 –179
Claereboudt MR (2006) Reef corals and coral reefs of the
Gulf of Oman. Historical Association of Oman, Muscat
Codling EA, Hill NA, Pitchford JW, Simpson SD (2004) Ran-
dom walk models for the movement and recruitment of
reef fish larvae. Mar Ecol Prog Ser 279: 215−224
Diego-Rasilla FJ, Luengo RM (2004) Heterospecific call
recognition and phonotaxis in the orientation of the
marbled newt, Triturus marmoratus. Behav Ecol Socio-
biol 55: 556−560
Doherty P, Fowler T (1994) An empirical test of recruitment
limitation in a coral reef fish. Science 263: 935−939
Dumyahn SL, Pijanowski BC (2011) Soundscape conserva-
tion. Landsc Ecol 26: 1327−1344
Enochs IC, Toth LT, Brandtneris VW, Afflerbach JC,
Manzello DP (2011) Environmental determinants of
motile cryptofauna on an eastern Pacific coral reef. Mar
Ecol Prog Ser 438: 105−118
Fisher R (2005) Swimming speeds of larval coral reef fishes:
impacts on self-recruitment and dispersal. Mar Ecol Prog
Ser 285: 223−232
45
Author copy
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
Mar Ecol Prog Ser 516: 35–47, 2014
Freeman SE, Freeman LA, Lammers MO, Buckingham MJ
(2013) Locating invertebrate sound sources, including
hermit crabs, on shallow water reefs in the Northwestern
Hawaiian Islands using an L-shaped array of
hydrophones. J Acoust Soc Am 134: 4148
Freeman SE, Rohwer FL, Gerald LD, Friedlander AM,
Gregg AK, Sandin SA, Buckingham MJ (2014) The ori-
gins of ambient biological sound from coral reef ecosys-
tems in the Line Islands archipelago. J Acoust Soc Am
135: 1775−1788
Göpfert MC, Robert D (2000) Nanometre-range acoustic
sensitivity in male and female mosquitoes. Proc R Soc
Lond B Biol Sci 267: 453−457
Harrison HB, Williamson DH, Evans RD, Almany GR and
others (2012) Larval export from marine reserves and the
recruitment benefit for fish and fisheries. Curr Biol 22:
1023−1028
Huijbers CM, Nagelkerken I, Lössbroek PAC, Schulten IE,
Siegenthaler A, Holderied MW, Simpson SD (2012) A
test of the senses: fish select novel habitats by respond-
ing to multiple cues. Ecology 93: 46−55
Hultgren KM, Duffy JE (2010) Sponge host characteristics
shape the community structure of their shrimp associ-
ates. Mar Ecol Prog Ser 407: 1−12
Kalmijn AJ (1988) Hydrodynamic and acoustic field detec-
tion. In: Atema J, Fay RR, Popper AN, Tavolga WN (eds)
Sensory biology of aquatic animals. Springer-Verlag,
New York, NY, p 83−130
Kennedy EV, Guzman HM, Holderied MW, Mair JM, Simp-
son SD (2010) Spatial patterns in reef-generated noise
relate to habitats and communities: evidence from a
Panamanian case study. J Exp Mar Biol Ecol 395: 85−92
Kingsford MJ, Leis JM, Shanks A, Lindeman KC, Morgan
SG, Pineda J (2002) Sensory environments, larval abili-
ties and local self-recruitment. Bull Mar Sci 70: 309−340
Krause B (1987) Bioacoustics, habitat ambience in ecological
balance. Whole Earth Rev 57: 14−18
Krause B (2002) Wild soundscapes: discovering the voice of
our natural world. Wilderness Press, Berkeley, CA
Lammers MO, Brainard RE, Au WW, Mooney TA, Wong KB
(2008) An ecological acoustic recorder (EAR) for long-
term monitoring of biological and anthropogenic sounds
on coral reefs and other marine habitats. J Acoust Soc
Am 123: 1720−1728
Lee SC, Bruno JF (2009) Propagule supply controls grazer
community structure and primary production in a ben-
thic marine ecosystem. Proc Natl Acad Sci USA 106:
7052−7057
Leis JM (2007) Behaviour as input for modelling dispersal of
fish larvae: behaviour, biogeography, hydrodynamics,
ontogeny, physiology and phylogeny meet hydrography.
Mar Ecol Prog Ser 347: 185−193
Leis JM, McCormick MI (2002) The biology, behavior, and
ecology of the pelagic, larval stage of coral reef fishes.
In: Sale PF (ed) Coral reef fishes: dynamics and diversity
in a complex ecosystem. Academic Press, London,
p 171−199
Leis JM, Sweatman HPA, Reader SE (1996) What the pelagic
stages of coral reef fishes are doing out in blue water:
daytime field observations of larval behavioural capabil-
ities. Mar Freshw Res 47: 401−411
Leis JM, Carson-Ewart BM, Hay AC, Cato DH (2003) Coral-
reef sounds enable nocturnal navigation by some reef-
fish larvae in some places and at some times. J Fish Biol
63: 724−737
Lobel PS (2001) Fish bioacoustics and behavior: passive
acoustic detection and the application of a closed-circuit
rebreather for field study. Mar Technol Soc J 35: 19−28
Mann DA (2012) Remote sensing of fish using passive
acoustic monitoring. Acoust Today 8: 8−15
Mann DA, Casper BM, Boyle KS, Tricas TC (2007) On the
attraction of larval fishes to reef sounds. Mar Ecol Prog
Ser 338: 307−310
Marques TA, Thomas L, Martin SW, Mellinger DK and oth-
ers (2013) Estimating animal population density using
passive acoustics. Biol Rev Camb Philos Soc 88: 287−309
McCauley RD, Cato DH (2000) Patterns of fish calling in a
nearshore environment in the Great Barrier Reef. Philos
Trans R Soc Lond B Biol Sci 355: 1289−1293
McMellor S, Smith DJ (2010) Coral reefs of the Wakatobi:
abundance and biodiversity. In: Clifton J, Unsworth RKF,
Smith DJ (eds) Marine research and conservation in the
Coral Triangle: the Wakatobi National Park. NOVA Sci-
ence Publishers, Hauppauge, NY, p 11–26
Mönkkönen M, Forsman JT (2002) Heterospecific attraction
among forest birds: a review. Ornithol Sci 1: 41−51
Montgomery JC, Tolimieri N, Haine OS (2001) Active habi-
tat selection by pre-settlement reef fishes. Fish Fish 2:
261−277
Montgomery JC, Jeffs A, Simpson SD, Meekan MG, Tindle
C (2006) Sound as an orientation cue for the pelagic
larvae of reef fish and crustaceans. Adv Mar Biol 51:
143−196
MRME (Ministry of Regional Municipalities and Environ-
ment) (1996) Results of fieldwork in support of National
Coral Reef Management Plan. Ministry of Regional
Municipalities and Environment, Oman, p 11–26
O’Connor MI, Bruno JF, Gaines SD, Halpern BS, Lester SE,
Kinlan BP, Weiss JM (2007) Temperature control of larval
dispersal and the implications for marine ecology, evolu-
tion and conservation. Proc Natl Acad Sci USA 104:
1266−1271
Parvulescu A (1967) The acoustics of small tanks. In:
Tavolga WN (ed) Marine bioacoustics. Pergamon Press,
Oxford, p 7−14
Pijanowski BC, Villanueva-Rivera LJ, Dumyahn S, Farina A
and others (2011a) Soundscape ecology: the science of
sound in the landscape. Bioscience 61: 203−216
Pijanowski BC, Farina A, Krause B, Dumyuhn S, Gage SH
(2011b) What is soundscape ecology? Landsc Ecol 26:
1213−1232
Pupin F, Sacchi R, Gentilli A, Galeotti P, Fasola M (2007) Dis-
crimination of toad calls by smooth newts: support for the
heterospecific attraction hypothesis. Anim Behav 74:
1683−1690
Radford CA, Jeffs AG, Montgomery JC (2007) The direc-
tional swimming behaviour of five species of crab post-
larvae in response to reef sound. Bull Mar Sci 80: 369−378
Radford CA, Stanley JA, Tindle CT, Montgomery JC, Jeffs
AG (2010) Localised coastal habitats have distinct under-
water sound signatures. Mar Ecol Prog Ser 401: 21−29
Radford CA, Stanley J, Simpson SD, Jeffs AG (2011a) Juve-
nile coral reef fish use sound to locate habitats. Coral
Reefs 30: 295−305
Radford CA, Tindle CT, Montgomery JC, Jeffs AG (2011b)
Modelling a reef as an extended sound source increases
the predicted range at which reef noise may be heard by
fish larvae. Mar Ecol Prog Ser 438: 167−174
Rogers PH, Cox M (1988) Underwater sound as a biological
stimulus. In: Atema J, Fay RR, Popper AN, Tavolga WN
46
Author copy
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
Piercy et al.: Reef habitat quality affects sound production 47
(eds) Sensory biology of aquatic animals. Springer-
Verlag, New York, NY, p 131−149
Salm RV (1993) Coral reefs of the Sultanate of Oman. Atoll
Res Bull 380: 1−83
Simpson SD (2008) The sounds of the reef: Can we learn to
listen before it is too late? Bioacoustics 17: 28−29
Simpson SD, Meekan MG, McCauley RD, Jeffs A (2004)
Attraction of settlement-stage coral reef fishes to reef
noise. Mar Ecol Prog Ser 276: 263−268
Simpson SD, Meekan MG, Montgomery JC, McCauley RD,
Jeffs A (2005) Homeward sound. Science 308: 221
Simpson SD, Meekan MG, Jeffs A, Montgomery JC, Mc -
Cauley RD (2008) Settlement-stage coral reef fish prefer
the higher-frequency invertebrate-generated audible
component of reef noise. Anim Behav 75: 1861−1868
Simpson SD, Radford AN, Tickle EJ, Meekan MG, Jeffs AG
(2011a) Adaptive avoidance of reef noise. PLoS ONE 6:
e16625
Simpson SD, Munday PL, Wittenrich ML, Manassa R, Dix-
son DL, Gagliano M, Yan HY (2011b) Ocean acidification
erodes crucial auditory behaviour in a marine fish. Biol
Lett 7: 917−920
Slabbekoorn H, Bouton N (2008) Soundscape orientation: a
new field in need of sound investigation. Anim Behav 76:
e5−e8
Slabbekoorn H, Peet M (2003) Birds sing at higher pitch in
urban noise. Nature 424: 267
Smith & Jompa (2009) Wakatobi field report, June–August,
2009. Kementerian Riset dan Teknologi (RISTEK),
Jakarta
Staaterman E, Paris CB, Helgers J (2012) Orientation behav-
ior in fish larvae: a missing piece to Hjort’s critical period
hypothesis. J Theor Biol 304: 188−196
Staaterman E, Rice AN, Mann DA, Paris CB (2013) Sound-
scapes from a Tropical Eastern Pacific reef and a Carib-
bean Sea reef. Coral Reefs 32: 553−557
Stanley J, Radford CA, Jeffs AG (2010) Induction of settle-
ment in crab megalopae by ambient underwater reef
sound. Behav Ecol 21: 113−120
Stanley JA, Radford CA, Jeffs AG (2012) Location, location,
location: finding a suitable home among the noise. Proc R
Soc Lond B Biol Sci 279: 3622−3631
Stewart RH (2004) Introduction to physical oceanography.
Texas A&M University, College Station, TX
Sueur J, Pavoine S, Hamerlynck O, Duvail S (2008) Rapid
acoustic survey for biodiversity appraisal. PLoS ONE 3:
e4065
Tavolga WN (1956) Visual, chemical and sound stimuli as
cues in the sex discriminatory behavior of the gobiid fish
Bathygobius soporator. Zoologica 41: 49−64
Tolimieri N, Haine O, Jeffs A, McCauley R, Montgomery J
(2004) Directional orientation of pomacentrid larvae to
ambient reef sound. Coral Reefs 23: 184−191
UPMSI (University of the Philippines Marine Science Insti-
tute) (2007) University of the Philippines Marine Science
Institute MPA database, UPMSI, Quezon City, www.
mpa. msi. upd. edu.ph/
Vermeij MJA, Marhaver KL, Huijbers CM, Nagelkerken I,
Simpson SD (2010) Coral larvae move toward reef
sounds. PLoS ONE 5: e10660
Veron JEN (1995) Corals in space and time: the biogeogra-
phy and evolution of the Scleractinia. Cornell University
Press, Ithaca, NY
White AT, Meneses A, Tesch S, Maypa A, Stockwell B,
White E, Martinez R (2007) Summary field report: coral
reef monitoring in Bohol, Philippines, March 26–April 3,
2007. The Coastal Conservation and Education Founda-
tion, Inc. and Fisheries Improved for Sustainable Harvest
Project, Cebu City
WMO (World Meteorological Organization) (2013) World
Meteorological Organization Standard Normals data-
base. Available at www.wmo.int/
Wolanski E (1994) Physical oceanographic processes of the
Great Barrier Reef. CRC Press, Boca Raton, FL
Wright KJ, Higgs DM, Cato DM, Leis JM (2010) Auditory
sensitivity in settlement-stage larvae of coral reef fishes.
Coral Reefs 29: 235−243
Wright KJ, Higgs DM, Leis JM (2011) Ontogenetic and inter-
specific variation in hearing ability in marine fish larvae.
Mar Ecol Prog Ser 424: 1−13
Editorial responsibility: Ivan Nagelkerken,
Adelaide, South Australia, Australia
Submitted: October 21, 2013; Accepted: August 3, 2014
Proofs received from author(s): November 26, 2014
Author copy
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤
➤