Content uploaded by Renata Santoro Sousa-Lima
Author content
All content in this area was uploaded by Renata Santoro Sousa-Lima on Dec 05, 2014
Content may be subject to copyright.
Aquatic Mammals 2013, 39(1), 23-53, DOI 10.1578/AM.39.1.2013.23
A Review and Inventory of Fixed Autonomous Recorders
for Passive Acoustic Monitoring of Marine Mammals
Renata S. Sousa-Lima,1, 2, 5 Thomas F. Norris,3
Julie N. Oswald,3, 4 and Deborah P. Fernandes5
1Bioacoustics Research Program, Cornell Lab of Ornithology, 159 Sapsucker Woods Road, Ithaca, NY 14850, USA
E-mail: RSL32@cornell.edu
2Programa de Pós-Graduação em Ecologia, Conservação e Manejo de Vida Silvestre, Instituto de Ciências Biológicas,
Universidade Federal de Minas Gerais, Avenida Antônio Carlos 6627, Belo Horizonte, MG 31270, Brasil
3Bio-Waves, Inc., 144 W. D Street, Suite #205, Encinitas, CA 92024, USA
4Oceanwide Science Institute, PO Box 61692, Honolulu, HI 96839, USA
5Laboratório de Bioacústica e Programa de Pós Graduação em Psicobiologia,
Departamento de Fisiologia, Centro de Biociências, Universidade Federal do Rio Grande do Norte,
Caixa Postal 1511, Campus Universitário, Natal, RN 59078-970, Brasil
Abstract
Fixed autonomous acoustic recording devices
(autonomous recorders [ARs]) are defined as
any electronic recording system that acquires
and stores acoustic data internally (i.e., without a
cable or radio link to transmit data to a receiving
station), is deployed semi-permanently underwa-
ter (via a mooring, buoy, or attached to the sea
floor), and must be retrieved to access the data.
More than 30 ARs were reviewed. They varied
greatly in capabilities and costs, from small,
hand-deployable units for detecting dolphin and
porpoise clicks in shallow water to larger units
that can be deployed in deep water and can record
at high-frequency bandwidths for over a year, but
must be deployed from a large vessel. The capa-
bilities and limitations of the systems reviewed
herein are discussed in terms of their effectiveness
in monitoring and studying marine mammals.
Key Words: passive acoustic monitoring, fixed
systems, marine mammals, acoustic monitoring,
mitigation, autonomous recorders
Introduction
Marine mammals live most of their lives under the
ocean surface, out of view of humans. The diffi-
culties inherent in studying the effects of human
activities on these animals can be overcome only
through the application of technology (Samuels
& Tyack, 1999). Most species of marine mam-
mals are acoustic specialists that rely on sounds
for communication and navigational purposes.
Scientists and engineers have developed passive
acoustic-based technologies to detect and record
sounds produced by marine mammals to more
effectively study them.
In 1880, Pierre and Jacques Curie (1880a,
1880b) discovered that when mechanical pressure
was exerted on a quartz crystal, an electric poten-
tial is produced. This finding enabled the devel-
opment of the first device capable of listening to
sounds underwater—passive acoustic monitoring
(PAM), which was utilized during World War I.
Since then, the development of PAM technology
has made it possible for researchers to listen to,
record, store, and analyze marine mammal sounds.
However, up until the turn of the century, limita-
tions in PAM technologies and methods available,
as well as high costs, inhibited the development
and application of passive acoustics for marine
mammal monitoring. In addition, the technical
expertise required to develop and apply these
technologies was typically beyond that of most
field biologists. The development of fixed autono-
mous underwater sound recorders (ARs) in the
early 1990s greatly reduced the costs and exper-
tise required to monitor marine mammal sounds
for extended time periods. An AR is defined as
any electronic recording device or system that
acquires and stores acoustic data internally (i.e.,
without cable or radio links to a fixed platform or
receiving station) on its own, without the need of
a person to run it; is deployed semipermanently
underwater (i.e., usually via a mooring, buoy, or
attached to the sea floor); and is archival (i.e.,
must be retrieved after the deployment period to
access the data).
Today, ARs can be easily deployed on the
ocean bottom to record acoustic data for days,
weeks, or even months at a time. These archival
ARs must then be retrieved to download data
24 Sousa-Lima et al.
for post-processing and analysis. This approach
allows ARs to be deployed and retrieved by
field personnel with a relatively limited amount
of training or expertise, which frees up valuable
time, resources, and funding.
ARs are most cost effective when used in
extreme or remote locations where access is lim-
ited or difficult—for example, polar regions, the
deep sea, and locations where travel distances are
great or environmental conditions exist that are
too harsh to conduct surveys from aboard research
vessels (Mellinger & Barlow, 2003; Munger et al.,
2005; Sirovic et al., 2009). ARs are also useful
for detecting marine mammals in areas where
the occurrence of animals is infrequent, or where
ship-based surveys have a very high cost per
detection (Mellinger & Barlow, 2003). The cost
savings in the use of ARs is achieved because of
their autonomous nature—that is, their operation
is independent of the presence of a human opera-
tor. The disadvantage is that these instruments
must be recovered to access the data; therefore,
real-time monitoring is not possible. If archival
data are useful, such as for acoustic prospecting
efforts (i.e., during pilot studies), ARs should be
considered as a cost-effective approach. In gen-
eral, set-up and infrastructure costs are lower for
ARs than they are for other types of PAM systems
(e.g., fixed-cabled hydrophones, towed-hydro-
phone arrays, and real-time radio- or satellite-
linked hydrophones; Mellinger et al., 2007a). In
addition, ARs are more flexible in their configu-
ration, timing, and location of deployment, and
they are less obtrusive to both animals and vessel
traffic when compared to other types of PAM
systems. Still, acoustic data bandwidth and col-
lection capabilities are usually higher for these
other types of PAM systems than they are for ARs
(Mellinger et al., 2007a; Van Parijs et al., 2009).
These trade-offs must be considered when decid-
ing which type of PAM system to use to reach a
particular goal.
A critical view of state-of-the-industry AR tech-
nology is provided herein, including both “tradi-
tional” autonomous recording devices (i.e., those
designed specifically for recording geophysical
events, underwater noise, and marine animal
sounds) and “nontraditional” recording devices
(e.g., electronic animal tags such as acoustic data-
loggers). A review of the history of AR develop-
ment; their capabilities and constraints with respect
to different application requirements (monitoring
vs mitigation); specific environments in which they
can be used, and, perhaps most importantly, the
species to be monitored; and the biological ques-
tions that are to be addressed are presented herein.
AR capabilities and constraints are discussed with
respect to their use in monitoring marine mammals
in relation to oil and gas exploration and produc-
tion (E&P) activities.
Historical Overview of the Development of
Autonomous Recorders
During the late 1960s, a change in spatial scale
occurred in marine geophysical research when
scientists focused their studies on earthquakes in
smaller areas of the sea floor. This shift required
higher accuracy and more precise geophysical
instrumentation and led to the development of
autonomous instruments called Ocean Bottom
Seismometers (OBSs) for monitoring underwater
earthquakes. OBSs were able to measure move-
ments of the Earth’s crust (Loncarevic, 1977). An
OBS is designed to rest on the ocean floor and
uses a sensor called a seismometer to take mea-
surements. The seismometer is comprised of a
heavy mass suspended on a spring between two
magnets. Seismometers use the principle of iner-
tia: the resistance of an object to a change in its
state of motion. When the earth’s crust shifts, the
seismometer and its magnets move concurrently,
but the heavy mass momentarily remains in its
original position. The relative movements of the
mass through the magnetic field produce elec-
trical currents that are then measured by instru-
mentation in the OBS (Dorman, 2001; Ocean
Instruments, n.d.).
A typical OBS consists of a seismometer, a
data logger, batteries to power the device, weight
to sink it to the sea floor, a remotely activated
(or timed) release mechanism, and flotation to
buoy the instrument back to the surface (Dorman,
2001; Ocean Instruments, n.d.). By 1975, the
OBS became an operational tool used by a dozen
or so research groups in at least seven countries
(Loncarevic, 1977). Since then, OBSs developed
by researchers from France, Japan, Australia,
Germany, Russia, and the U.S. have been used
extensively in geophysical research efforts.
Small ground motions caused by earthquakes
have relatively higher frequencies, so monitor-
ing them requires special short-period OBSs that
can record motions up to hundreds of times per
second (Ocean Instruments, n.d.). These higher
frequency OBSs, originally intended to pick
up the motion of the crust via the motion of the
substrate upon which they rest, typically record
up to 100 Hz and are also capable of recording
low-frequency sounds produced by large baleen
whales (e.g., blue whale [Balaenoptera musculus]
and fin whale [B. physalus])—sounds that con-
tain frequencies below 100 Hz. McDonald et al.
(1995) were the first to use OBS data to study
marine mammals: blue and fin whale calls were
detected and localized in deep waters during a
seismic study on the southern Juan de Fuca Ridge
25 Fixed Autonomous Recorders
off the coast of Oregon. These data were recorded
incidentally during a seismology experiment.
A similar device called an Ocean Bottom
Hydrophone (OBH) is also used by geologists to
study seismic activity in the ocean. An OBH has a
hydrophone instead of (or in addition to) a seismom-
eter. Experiments using both vertical seismometers
and hydrophones have shown a higher signal-to-
noise ratio for large whale low-frequency calls on
seismometers than on hydrophones (McDonald et al.,
1995), even though hydrophones are able to record
higher frequencies than seismometers. Besides
John Hildebrand’s group at the Scripps Institution
of Oceanography Whale Acoustics Lab (McDonald
et al., 1995), Christopher Fox’s group at the National
Oceanic and Atmospheric Administration’s (NOAA)
Pacific Marine Environmental Laboratory (PMEL)
has also used OBSs/OBHs to gather marine mammal
data since the early 1990s (e.g., Stafford et al., 1999;
C. W. Clark, pers. comm., 28 November 2009).
OBSs and OBHs were too expensive for most
researchers to purchase in the quantities needed to
study marine mammals, so during the 1990s, sev-
eral laboratories started developing their own ARs
in an attempt to lower the costs and to modify the
design for their own needs. One example was a rela-
tively small experimental instrument with a single
hydrophone that was developed by John Orcutt at
Scripps Institution of Oceanography (SIO). The low
cost and smaller size of the Low-Cost Hardware
for Earth Applications & Physical Oceanography
(LCHEAPO) was the direct result of the availability
of new, low-power consumer electronics. Different
institutions were then collaborating in the deploy-
ment and testing of these instruments (C. W. Clark,
pers. comm., 28 November 2009). An example was
the collaboration between Peter Worcester’s group
at SIO and Cornell University during the monitor-
ing of the ATOC (Acoustic Thermometry of Ocean
Climate) transmissions on the Pioneer Seamount
during humpback whale (Megaptera novaeangliae)
research off Hawaii (Frankel & Clark, 1998) and
during research on blue and fin whales off south-
ern California (Clark & Fristrup, 1997; Fristrup &
Clark, 1997).
Soon thereafter, John Hildebrand (SIO
Whale Acoustics Lab), Christopher W. Clark
(Bioacoustics Research Program [BRP] at Cornell
University), and Haru Matsumoto (NOAA/PMEL)
were among the first to develop and deploy their
own ARs. Each group designed a different instru-
ment specifically to collect bioacoustic data from
marine mammals (Calupca et al., 2000; Wiggins
& Hildebrand, 2007). Thus began the cultural
transmission of oceanography to bioacoustics as
some of these instruments (such as the Marine
Acoustic Recording Units [MARUs] or pop-ups
from BRP) were the direct result of researchers
and engineers from these two areas of expertise,
and from two different institutions (SIO and BRP),
exchanging technology and knowledge to help
create the initial design of that instrument (C. W.
Clark, pers. comm., 28 November 2009).
More recently, advances in low-power elec-
tronics, high-capacity data storage, computer
processing technology, and power supply units
have enabled the development and use of ARs
capable of monitoring the acoustic environment
and behavior of many species of marine mam-
mals. Improvements in electronic data storage and
battery technologies have allowed data collection
for much longer periods of time and at higher
data-sampling rates than previously possible.
These ARs will be reviewed below with examples
provided related to their use in marine mammal
research and monitoring.
Methods
An inventory of autonomous recorders was con-
ducted between 2009 and 2012 by searching
available systems online using the beta version
of Scientific WebPlus (ISI Web of Knowledge), a
Web-based search engine that is focused on sci-
entific content, recent scientific developments,
and other science-based information selected by
Thomson Reuters editors. A search for the string
in English “autonomous underwater sound record-
ing” returned 149 results. Additional information
included in scientific papers and reports and on
commercial webpages was searched using the
Google search engine (both the regular Web search
and Google Scholar) and all relevant library data-
bases available through Cornell University and the
University of Hawaii. A request for information
was sent to Bioacoustics-L and MARMAM list-
servers, which are commonly viewed by marine
mammal researchers and bioacousticians and other
professionals working on passive acoustic moni-
toring of marine mammals. Conference proceed-
ings and abstracts were also reviewed for relevant
information. Finally, researchers, organizations,
and companies were contacted directly via e-mail
to inquire about specific systems or devices.
Results
Inventory of Current Fixed Autonomous Recorders
Over 40 instruments were identified that fit the
working definition of fixed autonomous acoustic
recording devices used for marine mammal moni-
toring (Table 1). These included miniaturized
recording devices (i.e., data-logger animal tags)
that have been modified or can be implemented
as fixed ARs (Au et al., 2000; Thode et al., 2006;
Akamatsu et al., 2008; Arias et al., 2008).
26 Sousa-Lima et al.
Table 1. Inventory list of fixed autonomous acoustic recording (AR) devices, including acronyms, developers, and sources
of information
Acronym System name Developers References listed by date
AAR on a
MFP or
“Insta-array”
Autonomous Acoustic
Recorder arranged as
sensors in a Portable
Matched-Field
Processing System or
“Insta-array”
Marine Physical Laboratory (MPL),
Scripps Institution of Oceanography
(SIO), Greeneridge Sciences, Inc.,
University of Queensland, and the
Defence Science and Technology
Organization, Defence Department of
Australia
Thode et al., 2006
AcousondeTM
3A (tag)
and
AcousondeTM
3B (tag)
Replaced the Compact
Acoustic Probe (CAP)
or Bioacoustic Probe
(Bio-probe)
Greeneridge Sciences, Inc. Burgess et al., 1998, 2011; Au
et al., 2000; Burgess, 2000; Insley
et al., 2007; Oleson et al., 2007b;
Acousonde, n.d.
ADIOS NA National Oceanic and Atmospheric
Administration (NOAA) and Central
Bering Fishermen’s Association (CBSFA)
Ponce et al., 2012
AARS Autonomous Acoustic
Recording System
National Sun Yat-sen University,
Kaohsiung, Taiwan
Ming-Hao et al., 2007
AHS, AUH,
OBH, or
Haruphone
Autonomous
Hydrophone
System, Underwater
Hydrophone, or
Ocean Bottom
Hydrophone
NOAA Pacific Marine Environmental
Laboratory (PMEL) and Oregon State
University (OSU)
Stafford et al., 1999; Fox et al.,
2001; Fowler, 2003; Mellinger et al.,
2004a, 2004b, 2007; Nieukirk et al.,
2004; Heimlich et al., 2005; Dziak
et al., 2007
AQUAclick NA Aquatec Group Limited, Hampshire, UK Kyhn et al., 2008; AQUATEC, n.d.
AMAR Autonomous Multi-
Channel Acoustic
Recorder
JASCO Research Ltd, Canada JASCO, 2009a, 2009b
AMAR G3 Autonomous Multi-
Channel Acoustic
Recorder Generation 3
JASCO Research Ltd, Canada JASCO, 2012
ARP Acoustic Recording
Package
Scripps Institution of Oceanography
Whale Acoustic Lab (SWAL)
Wiggins, 2003; Gedamke, 2005;
Munger et al., 2005; Gedamke et al.,
2007; Oleson et al., 2007a; Stafford
et al., 2007; Wiggins & Hildebrand,
2007; Širović et al., 2009
A-TAG
(tag)
Acoustic tag Marine Micro Technology, Japan Akamatsu et al., 2000, 2005, 2008,
2011; Wang et al., 2005; Kimura
et al., 2009
AUAR Autonomous
Underwater Acoustic
Recorder
V. I. Ili’chev Pacific Oceanological
Institute in Russia
Acoustics, 2004; Borisov et al., 2008
AULS Autonomous
Underwater Listening
Stations
Clifford Goudey, MIT Sea Grant, Center
for Fisheries Engineering Research
(CFER); Rodney Rountree, University
of Massachusetts/Dartmouth; and
Tony Hawkins, University of Aberdeen,
King’s College, UK
MIT Seagrant, n.d.; Discovery of
Sounds in the Sea (DOSITS), 2011
Fixed Autonomous Recorders 27
Acronym System name Developers References listed by date
AURAL-M2 Autonomous
Underwater Recorder
for Acoustic Listening
Model-2
Multi-Electronique Inc., France (MTE) Simard et al., 2008; Multi-
Électronique Inc (MTE), 2012
AUSOMS-D Automatic Underwater
Sound Monitoring
System
System Intech Co., Ltd, Tokyo, Japan Shinke et al., 2004; Ichikawa et al.,
2006; Tsutsumi et al., 2006
Crittercam
(tag)
Video camera Greg Marshall with support from
National Geographic
Marshall, 1998; Calambokidis et al.,
2007
DASAR Directional
Autonomous Seafloor
Acoustic Recorders
Greeneridge Sciences, Inc. incorporated
DIFAR sensors from Sparton Electronics,
FL, into DASARs.
Norman & Greene, 2000; Greene
et al., 2004; Blackwell & Greene,
2006; Blackwell et al., 2007, 2012;
Greeneridge Sciences, n.d.
DSG-Ocean Ocean Digital
Spectrogram Recorder
Loggerhead Instruments Loggerhead Instruments, 2012
DMON Digital Acoustic
Monitor
Woods Hole Oceanographic Institution
(WHOI)
M. Johnson, pers. comm., 25 August
2008; A. Bocconcelli, pers. comm.,
17 October 2009
DTA G
(tag)
Digital Acoustic Tag WHOI Johnson & Tyack, 2003; Tyack
et al., 2006; Arias et al., 2008
EAR Ecological Acoustic
Recorder
Marc O. Lammers, Oceanwide Science
Institute (OSI), and Kevin Wong, NOAA
Fisheries, Pacific Islands Fisheries
Science Center, Coral Reef Ecosystem
Division (CRED), Hawaii
Lammers et al., 2008
HARP High-frequency ARP SWAL Wiggins & Hildebrand, 2007;
Baumann et al., 2008
LARS-LF and
-HF
Long Term Acoustic
Recording Systems
(Low and high
frequency)
David Mann, University of South Florida
(USF)
DOSITS, 2011
LADC EARS The Littoral Acoustic
Demonstration Center
Environmental Acoustic
Recording System
Naval Oceanographic Office
(NAVOCEANO)
Newcomb et al., 2002, 2009; Ioup
et al., 2009; J. Newcomb & G. Ioup,
pers. comm., 30 November 2009
OAR
(tag)
Onboard Acoustic
Recorder
Stacia Fletcher, Department of Biology
and Institute of Marine Sciences,
University of California at Santa Cruz
Fletcher et al., 1996
OCEANPOD NA LADIN (Laboratório de Dinâmica
e Instrumentação), Universidade de
São Paulo, São Paulo, Brazil
LADIN, 2012
OCEANBASE NA LADIN, Universidade de São Paulo,
São Paulo, Brazil
LADIN, 2012
PAL Passive Aquatic
Listener
University of Washington’s Applied
Physics Laboratory (APL); available
commercially through RSL
(Environmental Remote Sensing
Technologies Ltd.) in Athens, Greece
Nystuen, 1998, 2006; Nystuen et al.,
2007; J. Nystuen, pers. comm.,
15 November 2008; Anagnostou
et al., 2011
28 Sousa-Lima et al.
Acronym System name Developers References listed by date
PANDA Pop-up Ambient Noise
Data Acquisition
Acoustic Research Laboratory (ARL)
of Tropical Marine Science Institute in
National University of Singapore
Koay et al., 2001, 2002
A-PANDA Advanced Pop-up
Ambient Noise Data
Acquisition
ARL of Tropical Marine Science Institute
in National University of Singapore
Koay et al., 2006
Pop-up or
MARU
Marine Acoustic
Recording Unit
Bioacoustics Research Program (BRP)
at the Lab of Ornithology (CLO),
Cornell University
Calupca et al., 2000; Clark et al.,
2000, 2002; Sousa-Lima & Clark,
2008, 2009; T. Calupca, pers.
comm., 14 January 2009
RASP Registratore
Acustico Subacqueo
Programmabile
Nauta Ricerca e Consulenza Scientifica,
Italia
NAUTA, n.d.
RUDARTM Remote Underwater
Digital Acoustic
Recorder
Cetacean Research Technology J. R. Olson, pers. comm.,
1 February 2009; Cetacean
Research Technology, 2012
μRUDARTM Micro Remote
Underwater Digital
Acoustic Recorder
Cetacean Research Technology J. R. Olson, pers. comm.,
20 October 2009; Cetacean
Research Technology, 2012
nRUDARTM Nano Remote
Underwater Digital
Acoustic Recorder
Cetacean Research Technology Cetacean Research Technology,
2012
SM2M Song Meter
Autonomous
Submersible Recorder
Wildlife Acoustics, Inc., USA Wildlife Acoustics, Inc., 2012
SM2M
Ultrasonic
Song Meter
Autonomous
Submersible Recorder,
Ultrasonic Version
Wildlife Acoustics, Inc., USA Wildlife Acoustics, Inc., 2012
SRB-16 SRB-16 Autonomous
Recording Buoy
High Tech, Inc., USA High Tech, Inc., 2005
T-POD,
C-POD, and
DeepC-POD
NA Chelonia Limited, UK Watkins & Colley, 2004; Chelonia
Limited, 2007, 2011, 2012, n.d.;
Kyhn et al., 2008
USR Underwater Sound
Recorder
CMST (Centre for Marine Science and
Technology), Curtin University, Australia
CMST, 2011
UTDRT
(tag)
Ultrasound Time/
Depth-Recording Tags
Peter Madsen, Department of
Zoophysiology, Institute of Biological
Sciences, University of Aarhus, Denmark
Madsen et al., 2002
Fixed Autonomous Recorders 29
The instruments reviewed herein were in vari-
ous stages of development. Some AR systems that
were researched were in early stages of devel-
opment and did not have detailed specifications
available, or in some cases, a response to our
direct attempts at contacting developers for fur-
ther information was not received (e.g., the Digital
Hydrophone from MarSensing Lda. in Portugal
and the Autonomous Acoustic Recording System
developed by Ming-Hao et al., 2007). Thus, it was
not possible to provide complete, or in some cases
any, information on some of these systems.
Instruments researched that have very limited
application for passive acoustic monitoring were
also not included in the inventory. An example
is the system designed by Hayes et al. (2000):
an “inexpensive animal recording and tracking
system” (see also Møhl et al. [2001] for a similar
passive location system). This system used auton-
omous sound-recording buoys deployed at sev-
eral locations simultaneously to produce a sparse
hydrophone array. Each buoy was an instrument
that contained a global positioning system (GPS)
location logger, a portable stereo digital audio
tape (DAT) recorder with a hydrophone connected
to one channel, and a VHF radio signal for time
synchronization connected to the second channel.
The authors point out that the main disadvantage
of the system for PAM applications is the duration
of the recordings. DAT tape recorders are capable
of recording sounds for a maximum of 6 h (using a
90-m tape and setting the recorder to “long-play”
mode at 32 kHz), which is not a long enough
duration for most PAM applications. Note that
the μRUDARTM (Cetacean Research Technology,
2012), although also limited in recording duration
(up to 61 h), uses a compact flash card as storage
media which, along with solid-state hard drives,
have mostly replaced DATs as portable record-
ing devices. Therefore, Hayes et al.’s instrument
is now considered outdated and is not further
reviewed or included in Tables 1 and 2, but the
μRUDARTM is. In a number of cases, newer ver-
sions of the instruments included in this review
are also being developed and are noted as such in
Table 2.
Capabilities of Fixed Autonomous Recorders
ARs provide a cost-effective way to determine
the presence, relative numbers, and distribution
of vocalizing marine mammals in space and time.
The capabilities of ARs that are necessary for
monitoring marine mammals will vary according
to the goals and biological questions, the sound
production behavior of the species of interest, the
environment in which the ARs are to be deployed,
and the ambient noise characteristics. For exam-
ple, monitoring the seasonal occurrence of baleen
whales usually requires deployments of several
months to a year. However, because baleen whales
produce low-frequency sounds with good propa-
gation characteristics, the requirements for spatial
coverage and sample rates are relatively low (usu-
ally less than 1 kHz; Wiggins, 2003) compared
to those that would be required to monitor most
species of odontocetes (at least 48 kHz; Oswald
et al., 2004) over a similar area. Low sample rates
required to record low-frequency sounds also
allow modest power and storage capabilities for
the AR.
In general, odontocetes produce mid- (whis-
tles between approximately 5 and 25 kHz) to
high-frequency sounds (pulsed clicks containing
energy well above 20 kHz) that do not propagate
as well as the sounds produced by baleen whales
(Richardson et al., 1995). This is because higher
frequencies attenuate more rapidly (Bradbury &
Vehrencamp, 2011). In addition, the pulsed sig-
nals produced by odontocetes often have very
narrow beam patterns and so are more difficult
to detect when animals are not on axis relative
to the hydrophone. Because the sounds produced
by marine mammals cover such a large frequency
range, different AR deployment strategies must be
considered for different species. For example, in
order to monitor odontocete echolocation clicks,
an area must be relatively densely covered with
ARs that have high sampling rates (e.g., harbor
porpoises [Phocoena phocoena]; Kyhn et al.,
2008), while a sparse population of ARs that have
low sample rates are required to monitor low-fre-
quency sounds such as those produced by baleen
whales (e.g., Sousa-Lima & Clark, 2008, 2009).
How should a user choose an AR system? First,
the questions and goals to be addressed for a given
study must be clearly defined and considered.
This will, in turn, dictate the requirements of the
AR system. Based on the costs, capabilities, and
specifications of an AR system, as well as deploy-
ment and retrieval issues related to the monitored
area, the user may then consider the options avail-
able. For example, suppose the scientific ques-
tion of interest concerns the effects of oil and gas
E&P activities on the spatial distribution of sing-
ing humpback whales on their breeding grounds
during winter when animals are singing for many
hours continuously. Addressing this question will
require multiple time-synchronized ARs that can
be deployed close enough to each other so that
each AR can record the same sounds for multiple
whales to allow localization and tracking of mul-
tiple animals for 3 to 7 mo, sampling at relatively
low frequencies (1 to 2 kHz). If sampling schemes
are available (i.e., recordings made at predefined
intervals), the AR can be programmed to record on
a duty cycle of 30 min on, 30 min off, for example
30 Sousa-Lima et al.
,
,
Examples of
species studied
vaeangliae
A (tested in
ield with
vaeangliae
,
ostris
htius
ocephalus
Balaenoptera
ata
,
a
,
ostr
ied baleen
gapter
ounga
,
,
angustir
hric
ustus
vaeangliae
Physeter
B. musculus
B. physalus
B. acutor
Eubalaena
glacialis
and other non-
whales
,
Me
no
N
the f
noise)
M. no
no
Mir
NA
Esc
obr
macr
edeni
M.
identif
attletale Model 7
ARM
ersion
ector floating point
.arm.com)
Microprocessor
NA
PC-104 single-board
computer (Celeron
1G)
T
on earlier v
(Bioacoustic probe)
ARM9 with an
v
(VFP) coprocessor
(www
208 MHz
Persistor CF2
NA
iles, depth,
Data format
MT f
tag temperature
2-D acceleration/
tilt
NA
2 acoustic
channels, MT
iles, depth, tag f
temperature
2-D acceleration/
tilt (option)
2 acoustic
channels, MT
iles, depth, tag f
temperature,
3-D acceleration/
tilt, ambient light
elvle
NA
Spectrograms
vailable)
Data storage
1 GB flash memory
Hard disk
8 GB
16 GB internal
memory and 4
MicroSD storage-
card slots;
16 GB internal
memory and 2
MicroSD storage-
esv
card slots
128 GB
4 GB hard dri
A = information not a
wer supply and
gy capacity
Po
ener
(N
Alkaline battery
pack
V lithium cells16
Lithium battery
pack
A-cell lithium
battery pack
D-cell batteries
Alkaline battery
pack
AR system
y (Hz)
w-
Sampling
frequenc
25,793 (lo
232,000 (high-
frequency
y
y
w-
ications of each
2,000
(100-
20,000)
69,000
werpo
channel)
channel)
22 (lo
frequenc
channel)
232,000 (high-
frequenc
channel)
100,000
100
900
yment
Summary of the main capabilities and specif
Maximum
deplo
time
71 h
4.5 h
23 d
(at 2 kHz)
14 d
7 d
1 y
14 d
Maximum
yment deplo
depth (m)
2,000
NA
3,000
3.000
100
4,000
2.5 cm for
Dimensions
each recorder
× 3.2 cm
× 3.2 cm
× 4.2 cm
0.17 m
×
20
NA
22.1
22.1
22.4
12 cm diameter
× 75 cm long
×1.8
TM
TM
TM
UH,
AAR
(tag)
and
Table 2.
Instrument
AARS
Acousonde
Acousonde
3A (tag)
Acousonde
3B (tag)
ADIOS
AHS, A
OBH, or
Haruphone
Fixed Autonomous Recorders 31
,
ca,
Delphinapterus
,
species studied
Examples of
,
, B. physalus
htius
,
,
Phocoena
mysticetus
cinus or
,
phocoena
Odobenus
,
,
vaeangliae
Erignathus
B. musculus
B. physalus
ensis
htius
Balaena
,
hric
ustus,
ustus
Esc
leucas
hric
ob
ob
r
Or
osmarusr
barbatus
NA
B. bonar
M. no
Esc
r
Balaena
mysticetus
Eubalaena
japonica
Microprocessor
NA
NA
Ocean Sensors
.
NA
OS500 data
logger (www
oceansensors.com)
20 MHz
y other
Data format
Click parameters
stored are
occurrence time,
duration, and
iles
elsound lev
iles with
1-2 channel
V fA
W
non-acoustic data
(Temperature and
3-axis orientation
are standard;
high-precision
orientation;
ADCPs; turbidity;
depth; or an
analog, RS-232,
or RS-485 system
on request.)
Acoustic data as
V formatted AW
iles; non-acoustic
data as CSV f
ime series,
f
T
spectra, or
spectrograms
TB
xpandable
Data storage
8 MB standard
memory
Solid-state storage.
128 GB base,
xpandable to 2 e
or more
Solid-state storage.
256 GB, e
to 1,796 GB
72 GB hard disk
esvdri
wer supply and
geable
gy capacity
wer from
Po
ener
NiMh rechar
batteries
Alkaline or lithium
batteries
DC po
battery pack (7 to
Vdc) or PoE 16
three standard
battery packs
ailable (short, va
medium, and long)
Alkaline or lithium
battery pack
y (Hz)
Sampling
frequenc
50,000-
170,000
Up to 128,000
1-150,000
1,000
Maximum
yment deplo
time
Until storage
capacity is
1 y dependent
iguration,
reached
50 d
on input
ycle
channel
conf
duty c
settings, and
attached
battery packs
~ 400 d
Maximum
yment deplo
depth (m)
100
400
250 (shallow
AMAR)
2,500 (deep
AMAR)
Up to 7,000 m
Dimensions
240 mm long
Standard: 71.1
× 17.8 cm;
40.4
36.4 kg
×
× 88 mm
diameter
1.5 m by 1.5
m
132.1
cm; Diameter:
16.5 cm
by 1.5
plus a 10-m
long line and
flotation for the
ydrophoneh
Instrument
AclickUQ
A
AMAR
AMAR G3
ARP
32 Sousa-Lima et al.
)
Examples of
species studied
Neophocaena
phocaenoides
asiaeorientalis
htius
wrence
P.
hric
ustus
er
ocephalus
Esc
ob
r
Fish
Whales in the
St La
vRi
Dugong dugon
All odontocetes
xcept sperm e
whales (
macr
Microprocessor
(PIC18F6620;
Microchip, Detroit,
single
actured by
CPU
MI, USA)
ometheus Pr
board PC/104
computer
manuf
Diamond Systems
Corporation
NA
33 MIPS Dallas
DS89C450 Ultra
High Speed Flash
Microcontroller
NA
Altera MAXII
, Click intensity
al between
ydrophones
Data format
timing, and the
ference in time
iles
iles,
, y
y trend,
ertical, and
v
V f
V f
o h
A
A
elope slope, v
dif
arri
tw
Analog sound data
W
W
temperature and
depth
2 acoustic
channels
Click center
frequenc
frequenc
duration, intensity
(8 bit), bandwidth,
en
angle of the POD
to the v
temperature
e
Records data on
GB flash disk
v
es
GB
vable 4 GB
GB hard dri
"
v
disk 320 GB
Data storage
128 MB flash
memory
then writes it onto a
10 GB hard dri
more
o remo
a 1
160
Compact flash 1
or more and 2.5
hard
or
NA
wT
SD cards
Ah
gy capacity
Power supply and
ener
Lithium battery
cell
Three sealed gel
batteries with a
capacity of 115
Lead-acid gel cells
Alkaline D-cell or
battery pack
NA
Alkaline battery
pack
y (Hz)
Sampling
frequenc
55,000-235,000
15,000
11,000-44,000
256-32,768
44,100
20,000-160,000
Maximum
yment deplo
time
75 d
30 d
57 h
1 y depending
on setting
parameters
NA
4 mo
Maximum
yment deplo
depth (m)
200
50
200
or
1,800 (not
tested)
300
NA
At least 100 m
178 cm
kg without
Dimensions
eight =
21 mm
diameter and
108 mm length
kg
ith 16
W
145
NA
W
batteries: 14.6
90 cm and
20 kg; with 64
batteries: 14.6
120 cm and
32 kg; with
128 batteries:
×
and 49 kg
660 mm length
mm
14.6
90
NA
×
diameter;
2.1
×
×
batteries,
3.55 kg with
batteries
AR **U
ULS
URAL-M2
USOMS-D
Instrument
A-tag (tag)
A
A
A
A
C-POD
Fixed Autonomous Recorders 33
Examples of
All odontocetes
)
species studied
xcept sperm
P.
ocephalus
Balaena
mysticetus
vertebrates,
e
whales (
macr
In
fishes, and
marine
mammals
Fish and
cetaceans
Microprocessor
,
NA
Persistor
Instruments Inc.
(Bourne, MA,
USA) single-board
computer
model CF1 with a
Persistor Instruments
BigIDEA IDE
controller
dsPIC33F
TMS320VC5509A
DSP
3 channels
vide azimuthal pro
bearings to sound
T32 A
Data format
iles; F
iles (3
sources
V f
NA
AW
ile system that
f
stores latitude,
longitude, depth,
calibrations, and
time stamps
Sound f
independent
acoustic
channels),
temperature,
depth, and
orientation
* * *
w
Data storage
o removable 4 GB
ev
T
SD cards
30 GB disk dri
wo 32 GB SDHC T
cards or one 128 GB
SDHC card
32 GB flash
memory
wer supply and
gy capacity
geable
Po
ener
Alkaline battery
pack
Alkaline battery
pack
Alkaline battery
pack
8 3-D-cell battery
holders
Rechar
Li-Ion battery
y (Hz)
urst
Examples of
three possible
y
Sampling
frequenc
20-160,000
2,400
80,000 b
recordings of up
to 400,000
frequenc
settings:
LF 80,000
MF 240,000
HF 480,000
Maximum
yment deplo
time
4 mo
45 d
Calculated by
proprietary
are softw
during set-up
based on
memory size
and recording
schedule
LF 50 d
MF 180 h
Maximum
yment deplo
depth (m)
At least
m2,000
30
100 m for
PVC housing
and
2,000 m for
aluminum
housing
1,500
Dimensions
680 mm length
100 mm
kg without
diameter;
5.5
batteries, 6.96 kg
×
with batteries
Cylinder of 30
e v
×
kg positi
ater;
kg in air
by 36 cm
11.4 cm
diameter
63.5 cm
PVC: 3.6
kg in air (no
batteries);
e in
w
v
cm)
mm
ater
cm)
2.9
in sea
weight of 3 D
cells = 454 g
Aluminum:
7.8
(no batteries);
~1.4 kg
ati
70 mm
diameter
215
w
gne
sea
(7.1
×
(21.6
Weight (air):
1.5 kg
eight (salt W
ater): 400 gw
Instrument
DeepC-POD
ASAR
D
DSG-Ocean
DMON
34 Sousa-Lima et al.
,
,
Examples of
species studied
sp.
,
ish,
Eubalaena
,
,
ocephalus
ed whales
ostris
ostris
ocephalus
ostris
glacialis
Stenella
siops
P. macr
beak
longir
urT
truncatus
Stenella
,
longir
eponocephala
a
Mesoplodon
dolphins
ounga
P
electr
Fish and
. macr
P
Mir
angustir
Zalophus
californianus
Cetaceans, f
sea state, and
vessels
Microprocessor
, a
NA
Persistor CF2
microprocessor
GB compact flash
card, a Persistor
BigIDEA IDE
adapter
32-bit, 20 MHz
microcontroller from
Motorola (www.
1
motorola.com)
NA
NA
NA
NA
Data format
Sounds, depth,
temperature, and
orientation
iles
AV time series
iles
Binary f
XW
files
NA
NA
Sound f
PCM, MP3
es
Data storage
3.3 or 6.6 GB flash
Flash memory
card periodically
transferred to a hard
e
v
v
disk
ev
16 high-capacity
grated dri
"
es
T recorder)
v
A
memory
dri
inte
electronics (IDE)
laptop disk dri
Flash memory
4 IDE 2.5
dri
Digital
audiotape
(D
32 GB
wer supply and
gy capacity
Po
ener
Lithium battery
Alkaline battery
pack
Alkaline or lithium
battery pack
NA
Alkaline D cells
Alkaline batteries
Alkaline D cells
y (Hz)
Sampling
frequenc
2,000-20,000
2-64,000 (max)
10-200,000
OR
30,000
3,333
44,100
50-192,000
(max)
20-14,500
48 or 96 kHz
Maximum
yment deplo
time
Determined by
its memory
capacity
and audio
sampling rate
y
y
1 y
55 d
OR
1 y
1 y
2 mo
14 d at max
sampling
frequenc
>66 d at
11,700 Hz
sampling
frequenc
6 d
23 d
Maximum
yment deplo
depth (m)
2,000
500
~7,000
NA
3,000
800
70 for PVC
1,000 m for
aluminium
Dimensions
× 12 cm
(attached to a
GPS equipped
y)uo
5
b
10.16 cm
diameter by
60 cm long
cylinder
Six 30.5-cm
diameter
glass spheres
plus a 10-m
long line and
flotation for the
ydrophoneh
NA
21.6 cm in
diameter and
61.9 cm long;
× 12.70
45.5 kg
17.08
× 45 cm
× 6.67 cm
11
G (tag)A
AR (tag)
Instrument
DT
EAR
HARP
LARS-LF
and
LARS-HF
LADC EARS
O
OceanPod
Fixed Autonomous Recorders 35
Examples of
ish,
,
,ca
,
,
,
ensis
vaeangliae
essels
cinus or
bonar
species studied
Cetaceans, f
sea state, and
v
Or
dolphins
NA
NA
B. musculus
B. physalus
B.
M. no
Eubalaena
glacialis
Whales and
dolphins
Cetaceans
Microprocessor
ARM platform
attletale Model 8
-based Persistor
MCU system with
wer
T
processing po
of an
80386 CPU
PC104+ stack that
consist of an Intel
Pentium III based
industrial PC, a
40 GB hard disk,
and a PC104 data
acquisition module
attletale Model 8
and analog to digital
rack
mobile
ersion board
el monitor
v
T
con
made by Onset
ied MicroT
Technologies
Modif
24/96 pocket
recorders with an
original time control
echnology
v
board
Sound T
ST1400ENV
data recorder and
sound le
Data format
PCM, MP3;
non-acoustic
signals in CSV
Binary restored to
time series (sound
f)
iles
VA
ime series
Direction
al v
V or MP3 f
format
bites)
T
of arri
estimations, time
series
Binary restored to
iles (aifsound f
AW
Up to 4
ydrophone h
channels, W
TB
2 GB flash memory
e
e
yments
Data storage
SSD 128 GB
xpandable to 1
v
e
or more
card
12 GB hard dri
40 GB hard disk
v120 GB hard dri
4 GB compact
flash cards (8 GB
yments
vailable)a
Compact flash cards
for short deplo
and hard disks for
longer deplo
Power supply and
gy capacity
wer
geable
iguration
geable
ener
Depending on
s application
Alkaline D cells
lithium video
camera batteries
Custom Li-Ion
ubble
user’
Rechar
battery pack
Alkaline battery
pack
(Double-b
conf
doubles po
Battery pack
ast
capacity.)
geable
NiMH f
rechar
Rechar
Li-Ion batteries
y (Hz)
s application
Sampling
frequenc
Until 100 kHz
according to
user’
or 44, 96, 192
kHz
0-50,000
10-10,000
10,000-150,000
2,000-
64,000 (max)
Up to 96,000
Selectable
sampling rates up
to 192,000
Maximum
yment deplo
time
Until storage
capacity is
reached
1 y
9-10 d
35 h
90 d
184 h
maximum
Depends on
sample rate
chosen
Maximum
yment deplo
depth (m)
70 for PVC
1,000 m for
aluminium
1,000
200
200
2,500
(acoustic
release
dependent)
Up to 6,000
(on moorings)
500
1,500
or
3,500
Dimensions
45 cm
76.2 cm long
15.2 cm in
diameter
30 kg without
anchor
30 cm diameter
70 cm long
Single sphere:
48.3 cm high
and 58.4 cm
ubble:
diameter
Double-b
100 cm high
58.4 cm
diameter
50 cm
17.8 cm,
kg or
×
and
×
9
×
×
9
36.4
45.5 kg with
batteries
Instrument
OceanBase
A
TM
A
U*
R
L
AND
AND
DA
PA
P
A-P
Pop-up or
MAR
RASP
RU
36 Sousa-Lima et al.
NA
Phocoena
mammals, and
ocephalus
Examples of
species studied
Cetaceans
NA
NA
NA
phocoena
Fish, marine
noise
P. macr
Microprocessor
II
ation
vig
grated
M-Audio
rackMicroT
recorder
RS-232 na
ace
digital
NA
NA
NA
interf
Altera MAXII
NA
Maxim Inte
Products, Inc.
DS5000T
V
el, SPL
T32
Data format
A
els
els,
A
v
v
Up to 4
ydrophone
channels, W
RMS lev
e le
vel, SPL
e le
ertical,
ydrophone
or MP3
v
NA
receiv
RMS le
recei
with optional
iles, real
h
Hi-SPL
NA
Click start and
end times, battery
voltage, angle
of the POD to
the v
system noise, and
temperature
1 or 2 h
channels; F
ilesf
Sound f
time and depth
hard
es (60 GB to
ype
Data storage
16 GB industrial-
grade compact flash
memory card
NA
128 GB with SDHC
or 512 GB with
SDXC
128 GB with SDHC
or 512 GB with
SDXC
5 GB recorded on
Exabyte 8500 digital
tape
128 MB memory
T
"o 2.5One or tw
vdisk dri
160 GB) and/or
1 compact flash card
192 MB compact
flash card
Power supply and
gy capacity
geable
Li-Ion batteries
LSD NiMH,
alkaline, or lithium
anese D-cell
anese D-cell
ener
Rechar
NA
mang
batteries
LSD NiMH,
alkaline, or lithium
mang
batteries
Internal battery
pack
Alkaline battery
, 2008)..
pack
NA
NA
v et al
y (Hz)
s
Sampling
frequenc
UAR (Boriso
96,000
96,000
4,000-96,000
4,000-384,000
Dynamic range
to suit user’
application
200,000
1,000-
15,000
62,500
-ATAR: UA
ycle
ycle
*** 64 GB capacity possible in 2010 (M. Johnson, pers. comm., 25 August 2008)
Maximum
yment deplo
time
Up to 61 h
depending on
sample rate
chosen
Up to 26 h
depending on
sample rate
chosen
Up to 104 d
depending on
sampling rate
and duty c
Up to 42 d
depending on
sampling rate
and duty c
72 h and
standby mode
of up to 15 d
1 y OR
until storage
capacity is
reached
2 y
30 min
ersion of the
Maximum
yment
ater:
deplo
depth (m)
100
100
150
150
3,049
Center
opening: 150
Deep w
2,500 (tested),
3,500 (not
tested)
NA
1,100
elopment.v
m
Dimensions
×
10 cm diameter
33 cm long
cells
×
diameter
79.4 cm long
×
kg with
900 mm,
ailable)
NA
16.5 cm
16.5 cm
diameter
79.4 cm long
NA
Length: 0.86
Diameter:
m; 0.90
no batteries;
4.4 kg with 15
alkaline D
×
v
2.8
114
weight = 30 kg
(size options
are a
NA
ven here is for a radio transmitting v
Instrument
TM
TM
AR
AR
T (tag)
UD
UD
μR
nR
SM2M
SM2M
Ultrasonic
SRB-16
ersions or capabilities are in de
UTDR
w v
USR
* Ne
The information gi
T-POD
**
Fixed Autonomous Recorders 37
(e.g., if analysis time is a constraint). This will save
on power, storage, and post-processing require-
ments. The minimum number of units required and
their deployment geometry are related not only to
the species of interest and research question but
also to the sound propagation profile of the area.
The farther sound travels, the fewer the number
of AR units that are needed to cover an area (the
maximum number of units is usually limited by
budget). Humpback whale breeding grounds are
typically shallow (< 100 m); therefore, the depth
rating requirement of the AR can be relatively
modest. Other issues, such as high fishing activity
or the presence of pirates in the area, the type and
availability of deployment/retrieval vessels, and
the amount of funding available, will all affect the
best choice for an AR device.
Choosing an AR system will likely not be
as simple as this example. If the question was
“What is the diversity and relative occurrence
of odontocetes in an area?,” high sampling rates
would be required, which, in turn, would limit
deployment duration for continuous recordings.
There are three main requirements for the data
acquisition electronics to provide long-term con-
tinuous acoustic records of identifiable odontocete
calls using an autonomous instrument: (1) low
power, (2) high-speed digitization, and (3) high-
capacity data storage (Lammers et al., 2008). As
with any battery-powered autonomous instrument,
low-power components are essential for long-
duration deployments. High-speed digitization is
necessary to record broadband odontocete calls
and to provide enough bandwidth for species iden-
tification (upper bandwidth limit of at least 24 kHz
for Delphinus delphis, Stenella attenuata, S. coe-
ruleoalba, and S. longirostris; Oswald et al., 2004).
High-speed digitization coupled with long-dura-
tion recordings requires high-capacity data stor-
age capability (Wiggins & Hildebrand, 2007) and/
or the use of a duty cycle recording schedule. In
most cases, high-capacity data storage is achieved
using multiple hard or flash drives, which require
a microcontroller and firmware dedicated to con-
trolling the data-recording process (e.g., HARPs;
Wiggins & Hildebrand, 2007). These capabilities
are important to understand when choosing the
best AR available for a particular application.
Tradeoffs Among Fixed Autonomous Recorder
Capabilities and Limitations
ARs have self-contained power supplies and data
acquisition and storage electronics. These com-
ponents constrain the design and capabilities of
AR systems because of tradeoffs among power
supply, data storage capacity, required sam-
pling frequency, and instrument size and depth
rating, which, in turn, effect cost and deployment
duration. Each AR developer presents a different
solution to manage these capability tradeoffs, and
the resulting compromises are critical to selec-
tion of an AR system for application during any
passive acoustic study—for example, oil and gas
or other E&P-related activities or aquatic animal
behavior, distribution, or presence research. The
main limitation on deployment duration is sam-
pling frequency, which is directly linked to stor-
age and battery capacity. Increased power require-
ments have a direct effect on the number, and
possibly type, of batteries included in a package,
thereby potentially increasing both instrument
size and flotation requirements. The size of the
package will effect costs related to deployment
and retrieval.
Figures 1 and 2 illustrate the tradeoffs among
AR capabilities and how these influence each
other; for example, given that the size of the
device housing dictates the amount of power,
when going from less to more power (bigger,
more expensive housings), one can increase the
sampling frequency to record higher-frequency
sounds, which requires greater data storage capac-
ity at the expense of deployment and recording
duration. The more hydrophones on a unit (may
enable localization/directionality capabilities, e.g.,
AUSOMS-D and DASAR), the greater the data
storage requirement, which will impact deploy-
ment duration and increase the number of batteries
needed. Systems that can be deployed to greater
depth are usually more expensive because of the
need for special pressure-resistant housings; thus,
Figure 1. Schematic of the tradeoffs among power supply, sampling frequency, deployment duration,
size and deployment, and retrieval costs. Less power supply will limit AR sampling frequency and
deployment duration, but in turn will result in a smaller instrument package and decrease deployment and
retrieval costs.
Figure 1. Schematic of the tradeoffs among power supply,
sampling frequency, deployment duration, size and deploy-
ment, and retrieval costs; less power supply will limit AR
sampling frequency and deployment duration but, in turn,
will result in a smaller instrument package and decrease
deployment and retrieval costs.
38 Sousa-Lima et al.
as size and complexity of a system are increased,
budgetary demands also generally increase.
Instruments like the HARP (Wiggins &
Hildebrand, 2007; Table 2) provide high sample
rates and good storage capacity (1.92 TB), which
allows for approximately 55 d of continuous sam-
pling at 200 kHz or about 1 y continuously sam-
pling at a lower sample rate of 30 kHz. The HARP
package is quite large because of associated bat-
tery and storage requirements, and it comes with a
depth rating of 7,000 m (requiring a high pressure
capable housing; Table 2). Note that the HARP
package has been reduced in size for other appli-
cations such as deployments on gliders (Wiggins
et al., 2010). To deploy large HARPs, a relatively
large (> 24 m) oceanographic ship or mid-sized
fishing vessel with winch and A-frame is required.
These deployment and retrieval costs must be con-
sidered when planning to use HARPs.
Proportionately, the components that use the
greatest amount of system power in ARs include
the hard disk drive and hard drive controller (e.g.,
on HARPs). The data acquisition rate is indirectly
related to power consumption because it deter-
mines how frequently the hard disk will need to be
accessed and written to; for example, in Cornell
BRP’s pop-up, the digital acoustic data are tempo-
rarily saved to a buffer which, once filled, down-
loads to a hard drive. Data recorded at a sampling
frequency of 2 kHz fill this buffer every 3 min,
requiring access to the hard drives and, therefore,
power consumption each time data are transferred.
The hard drive runs for 6 s every 3 min when data
writing is occurring. The standard battery pack
will keep the unit recording continuously for a
little over 100 d at the 2 kHz sample rate. At a
4 kHz sample rate, the data storage buffer will fill
every 1.5 min, and the drive will have to run twice
as often as at a 2 kHz rate, dropping the standard
battery life to 50 d. At 6 kHz, the buffer fills every
45 s, and the efficiency of shutting down the hard
drive between data writing sessions is lost so that
it runs continuously to record the data flow, drop-
ping battery life to about 22 d (T. Calupca, pers.
comm., 14 January 2009).
Hard drive space may become a limiting factor
in pop-ups at sample rates greater than 20 kHz
(T. Calupca, pers. comm., 14 January 2009; Table 3).
The standard pop-up hard drive stores 120 GB of
data; therefore, at high sampling rates, data storage
capacity, as opposed to power supply, can limit AR
monitoring duration. The shift from hard drives to
high storage capacity flash cards will take care of this
limitation. The HARP, which can sample at 200 kHz,
has a much larger storage capacity (1.92 TB) than
pop-ups; the standard power configuration (esti-
mated at 330 Amp-hours using 192 D-size alkaline
batteries), recording continuously at the maximum
sample rate, fills the hard disks before battery capac-
ity is reached (Wiggins & Hildebrand, 2007).
The type and size of storage media influences
tradeoffs as discussed above and, consequently,
costs. In the last few years, solid-state flash
memory has dropped significantly in price and
increased in capacity, offering an alternative to
hard drives that are bigger and heavier.
Fixed Autonomous Recorders
Figure 2. Schematic of more tradeoffs among capabilities and limitations of AR systems.
Figure 2. Schematic of more tradeoffs among capabilities and limitations of AR systems
Fixed Autonomous Recorders 39
Continuous Recording vs Sampling Schemes (Duty
Cycles)
When determining what type of sampling scheme
or duty cycle to use, it is important to have an idea
about the acoustic behavior of the target species
to be monitored. Notably, being aware of the fre-
quency range of the target species and also timing
of calling activity will inform AR decision(s),
although these acoustic characteristics may not
be known before AR deployment. For many pre-
liminary PAM applications, continuous recording
is desirable because complete information about
the acoustic behavior of animals and their acous-
tic environment is often lacking; the continuous
record provides a more in-depth view of the ani-
mals’ vocal activity in the context of environmen-
tal noise. Initially, medium- to long-term acoustic
prospecting must be completed to determine what
species are present, what types of sounds they
produce, and how often they are produced; how-
ever, the amount of data generated by continuous
recordings is typically excessively large (1 y may
result in over 2 TBs of data at sampling rates under
2 kHz) such that automatic detection or sampling
schemes must be applied during post-processing
and analysis. This tradeoff between minimizing
the nonsampling period and maximizing the time
periods during which data are collected must be
considered in relation to monitoring requirements
and temporal aspects of the acoustic behavior
for species in question. For example, a sampling
scheme of 12 h on and 12 h off per day would not
provide adequate coverage to examine whether
a diel calling pattern is present from a particular
species (Wiggins & Hildebrand, 2007; Lammers
et al., 2008). Alternatively, a duty cycle of 5 min
on/5 min off would be more appropriate to inves-
tigate for diel patterns in vocal behavior; a duty
cycle like this would also reduce power consump-
tion and facilitate longer duration monitoring.
Note that using an intermittent duty cycle is not
well-suited for capturing acoustic signals that are
very infrequent or random, but it is effective for
documenting potential patterns of occurrence for
regularly occurring signals typical of some spe-
cies. For example, humpback whales, which sing
continuously for several hours at a time during
the breeding season, have been monitored using
the EAR (Ecological Acoustic Recorder) at 3.3%
duty cycle (i.e., recording once every 15 min for
30 s; Lammers et al., 2008). Continuous acoustic
recordings can be useful when the aim is to com-
pare other phenomena present in the recordings
over time (e.g., Northern right whale [Eubalaena
glacialis] call characteristics; Parks et al., 2007)
or space (e.g., humpback song comparisons across
regions; Cerchio et al., 2001; Darling & Sousa-
Lima, 2005). Alternatively, triggering algorithms
that only record the sounds of interest or record
any sound at preset time intervals also can be
advantageous. This approach involves periodic
sampling with the ability to turn “on” the record-
ing device when signals of interest occur. This
method is desirable from both cost and data man-
agement standpoints (Lammers et al., 2008); how-
ever, such sampling requires validation to confirm
that signals of interest are not missed by the algo-
rithms used, and that the vocal behavior or call
types of interest are well known.
Some systems (e.g., PAL, T-POD, C-POD,
A-TAG, EAR, and AQUAclick; Tables 1 & 2)
have automatic call detection algorithms that trig-
ger recording when predetermined call types are
detected, or when defined acoustic criteria are
met. The PAL (Nystuen, 1998, 2006; Nystuen
et al., 2007), the EAR (Lammers et al., 2008), and
the DMON (M. Johnson, pers. comm., 25 August
2008) pre-process the data based on knowledge
of the sound of interest, saving storage space and
power. “Plug-in” user-supplied automatic detec-
tion algorithms can be used to automatically
process, extract, and store particular parts of the
sounds of interest (DMON). The PAL works with a
satellite communication system that transfers data
every 3 h to the surface. This includes variables
such as wind speed, rainfall, bubble populations,
whale and human activities, and even geological
activities. Identifying correctly all these different
sound sources is possible due to a classification
algorithm that differentiates spectral and temporal
characters of each detection (Anagnostou et al.,
2011).
Even more specific are the click detectors/log-
gers, such as the AQUAclick (includes a porpoise
channel tuned to 130 kHz and a “dolphin” channel
at 50 kHz; AQUATEC, n.d.), the T-POD (Watkins
& Colley, 2004) and C-POD (Chelonia Limited,
n.d.a), and the A-TAG (Akamatsu et al., 2008),
which do not record sounds but rather capture
information associated with the sounds such as
time of occurrence of high-frequency odontocete
clicks. Nevertheless, if the sounds of interest are
too variable, which is the case for many marine
mammal monitoring applications, this advantage
is diminished.
Future HARP systems are planned to implement
such triggering algorithms in the data loggers,
resulting in much smaller quantities of recorded
data (Wiggins & Hildebrand, 2007). While this
approach seems reasonable, the drawback is that
nontargeted calls and other sounds of poten-
tial interest would go unrecorded. For example,
dolphin and pinniped sounds would not likely
be recorded by an algorithm designed to detect
low-frequency baleen whale calls. Furthermore,
investigating the structure and variability of ocean
40 Sousa-Lima et al.
acoustic noise over various time periods would be
difficult, if not impossible, using event-triggered
acoustic data (Wiggins, 2003). Data compres-
sion schemes provide some means for decreasing
power consumption rates while increasing deploy-
ment duration (Wiggins & Hildebrand, 2007).
These approaches should be thoroughly tested
so that recording fidelity is not compromised and
the chosen algorithms match the objectives of the
deployment.
Capability of Collecting Non-Acoustic
Oceanographic Data
Some ARs integrate additional sensors to collect
non-acoustic oceanographic data (Table 2). For
example, the AMAR (JASCO, 2009a) collects
data on water temperature and 3-axis orientation
but also includes other sensors on request depend-
ing on the research questions under study. A
small, self-contained, external CTD data logger or
sound velocity sensor are planned as new add-ons
to the PANDA (Koay et al., 2002; Tables 1 & 2).
With the combined recordings of conductivity,
temperature, pressure, sound velocity, and acous-
tic signals in a single integrated, compact system,
PANDA is very useful for shallow-water physical
oceanographic studies (Koay et al., 2001). Sound
speed data are important for accurately calculat-
ing sound time-of-arrivals when using multiple
ARs to localize the source of a sound.
Miniaturized electronic devices (animal tags)
can be used as sensors and data loggers in fixed
ARs. Several types of electronic tags have been
used in fixed ARs; for example, Thode et al.
(2006) used slight modifications of an older ver-
sion of the Acousonde (Acousonde, n.d.; Burgess
et al., 1998; Burgess, 2000) in designing the AAR
(Tables 1 & 2). The Acousounde is a sound-
recording animal tag with two acoustic channels
that can sample up to 232 kHz, includes depth
and internal temperature sensors, and can also
contain 2-D acceleration/tilt sensors. The A-TAG
has been used to tag and study finless porpoises
(Neophocaena phocaenoides; Akamatsu et al.,
2008) and is yet another example of tag technol-
ogy that has been used in a fixed configuration
(Wang et al., 2005).
The DTAG (Johnson & Tyack, 2003), a digital
acoustic recording tag, contains an accelerometer, a
magnetometer, and pressure sensors. It is designed to
measure the tagged animal’s orientation and sample
sounds between 2 and 200 kHz (Johnson & Tyack,
2003). The DMON is a fixed AR used by the same
group at Woods Hole Oceanographic Institution
(WHOI) that is also capable of acquiring depth,
temperature, and orientation data (A. Mooney, pers.
comm., 26 November 2012).
Tags provide the capability to record oceano-
graphic data, animal orientation, and other infor-
mation and have been used to study several aspects
of the behavior of a variety of species. All of the
tags listed (Tables 1 & 2) also are able to collect
sound data and could potentially be used in a fixed
AR configuration for PAM applications.
Internal Design of Autonomous Recorders
ARs typically include a robust pressure housing to
protect the electronics, digital recording systems,
and batteries. An ideal AR requires high-quality
sensors and low-noise electronics with a high-res-
olution digital recording system. The AR internal
design and external package configuration should
be based on the specific questions and objectives
the system is built to address.
Electronics—Each AR developer has identified
different solutions in their system designs. Still, all
systems include a single or multiple hydrophone(s)
for sound acquisition, internal electronics to con-
trol the system and for acoustic data conditioning
(e.g., signal amplifiers, anti-aliasing and band-pass
filters, and analog-to-digital converters), storage
(e.g., hard drives, flash memory cards, and solid-
state drives), and, often, additional electronics or
devices (e.g., acoustic release mechanisms) to
allow for recovery of the device.
Pop-up developers designed their system with
the objective of creating a compact device that
could be deployed by a single person to depths up
to 2,500 m using an acoustic release. Therefore,
a pop-up includes additional recovery electronics
to enable retrieval, but also an acoustic command
recognition system, an audio signal communica-
tions system, a fail-safe time-release mechanism,
a radio beacon, and a strobe light (T. Calupca,
pers. comm., 14 January 2009). The pop-up
electronics are distributed on two plates that are
housed within a borosilicate glass sphere, which is
placed within a protective plastic helmet with the
hydrophone and piezoelectric speaker mounted
externally on its side (T. Calupca, pers. comm.,
14 January 2009).
The DMON electronics configuration includes
two circuit boards. The main board contains a
digital signal processor, memory, power supply,
and interface circuits. The sensor board contains
sound acquisition circuits and depth and orienta-
tion sensors. This set of two boards can be used
inside a pressure housing (e.g., a profiling float
or glider that requires hydrophone[s] be wired to
a penetrator) or it can also be used in a pressure-
equalized housing (e.g., sealed in an oil-filled soft
rubber sleeve) that can be deployed alone or in the
wet space of an underwater vehicle (all sensors
can be internal for protection and durability; M.
Johnson, pers. comm., 25 August 2008).
Fixed Autonomous Recorders 41
High-capacity data storage is desirable for
some applications (e.g., long-term, continuous
monitoring of species that emit high-frequency
sounds). Such high-capacity storage is achieved
on AMARs and HARPs. The AMAR electronic
board features eight channels of 24-bit analog-to-
digital conversion and can host up to 16 solid-state
memory modules, each of which has a capac-
ity of 128 GB, for a total of 2 TB of on-board
memory (JASCO, 2009b). Similar high-capacity
data storage on HARPs is achieved using 16 inte-
grated laptop hard drives arranged in a block and
addressed sequentially through a single 50-pin
bus. The 16-drive block can be easily removed and
replaced following instrument recovery (Wiggins
& Hildebrand, 2007).
Monitoring high-frequency sounds of some marine
mammal species can also be achieved by using trig-
gering algorithms. The PAL (Nystuen, 1998, 2006;
Nystuen et al., 2007), the EAR (Lammers et al.,
2008), and the DMON (M. Johnson, pers. comm.,
25 August 2008) are examples of ARs that include
electronics to monitor continuous sound and only
save information on sounds that trigger the detection
algorithms. For example, the EAR has a signal con-
ditioning module that includes circuitry that moni-
tors the input signals for specific types of acoustic
events (Lammers et al., 2008).
Hydrophones—As part of their systems, AR
developers include hydrophones that are off-the-
shelf, customized by other companies for their
specific device, or build their own. For example,
GeoSpectrum Technologies (GTI) is a company
that designs and manufactures custom acoustic
transducers, including directional hydrophones
for AMARs (particle velocity sensors; JASCO,
2009b). DASARs use technology developed for
DIFAR sonobuoys by Greeneridge Sciences and
also include two horizontal, orthogonal direc-
tional sensors (particle velocity hydrophones)
and one omnidirectional pressure sensor. In order
to have information on the DASARs’ reference
direction, it is necessary to perform acoustic cali-
bration transmissions, which can provide precise
bearing data (Greene et al., 2004; Blackwell et al.,
2007). This allows the bearing to the sound source
to be estimated so that animal sounds can be local-
ized and, in some cases, so that animals can be
tracked (Blackwell et al., 2012).
HARP developers designed a low self-noise,
high-gain hydrophone that can pre-whiten (adding
more gain at higher frequencies where ambient
noise levels are lower and sound attenuation is
higher) ocean ambient noise across four frequency
decades (10 Hz to 100 kHz). This is achieved
using two separate stages of signal conditioning,
one for a low-frequency band (10 Hz to 2 kHz) and
another for a high-frequency band (1 to 100 kHz)
(Wiggins & Hildebrand, 2007). These two stages
use different transducers—a single, spherical,
omnidirectional transducer for the high-frequency
stage and six cylindrical transducers connected in
series for the low-frequency stage—and provide
the ability to record both baleen whale low-fre-
quency and high-frequency sounds produced by
odontocetes. The signals from these two stages
are pre-amplified and pre-whitened and are then
added together via a differential receiver (Wiggins
& Hildebrand, 2007).
Some developers adopt standard field practices
that include calibration of all hydrophones before
deployment or on recovery; others calibrate
the entire AR system. Calibration is an impor-
tant issue and must be addressed for many PAM
applications.
Power Supply—ARs operate autonomously
and, thus, must be powered internally by a set
of batteries. AR developers design their systems
aiming for low-power consumption and include
user options to use different battery types, sizes,
and quantities in their systems depending on the
desired capabilities and design. Different batter-
ies are used to power the AR systems reviewed
herein (Table 2). Alkaline batteries are cheap and
relatively safe to dispose of, but they provide less
power and are not ideal for high-drain devices
because they cannot deliver power quickly. On
the other hand, lithium batteries are more expen-
sive (cost at least twice per Amp-hour compared
to alkaline), are relatively more toxic to the
environment, and can explode if short-circuited.
Still, they last longer than other batteries and are
more reliable due to a low rate of shelf discharge
(Bluejay, 2009). Nickel-Metal Hydride (NiMH)
batteries are rechargeable but less reliable (high
shelf-discharge rate and inaccurate voltage read-
ings that can result in sudden discharge) and put
out less voltage than alkaline batteries. Lead-acid
gel cells are also rechargeable but give off poten-
tially explosive gases and are more expensive than
NiMH batteries (Bluejay, 2009).
Package Design and External Configuration,
Deployment, and Retrieval Issues
Another consideration with ARs is external con-
figuration (e.g., shape and size of the package).
The choice depends on (1) the system capabili-
ties required by a specific application; (2) the
environmental conditions and substrate type in
the deployment area (e.g., type of ocean bottom,
presence of strong currents and surface winds,
bathymetry, vessel traffic, bottom fishing activi-
ties, etc.); and (3) deployment logistics, which
include the type of vessel and hoisting equip-
ment available for deployment and retrieval (e.g.,
winches, cranes, A-frames, and divers), all of
42 Sousa-Lima et al.
which have an impact on type and configuration
of instrument (or vice versa).
Depth Rating—AR housings have depth rat-
ings that specify their maximum deployment
depth. Some, like the pop-up, have a depth rating
up to 6,000 m but because of limitations related to
the release mechanism, the depth rating is often
decreased to as shallow as 2,500 m (T. Calupca,
pers. comm., 14 January 2009). In shallow deploy-
ment cases, the depth rating is environmentally
dependent and site specific because it can be
reduced even more due to increased noise from
high sea states, ocean floor topography, variation
in sound speed profile, and natural and anthropo-
genic sources of noise.
Details of the biology of the species of interest,
deployment area, and the scientific question(s) to
be asked all dictate the depth necessary for an AR.
For example, the acoustic behavior of deep-diving
species like beaked whales and elephant seals can
only be monitored effectively with ARs that can
be deployed to depths below ~1,500 m (Johnson
et al., 2006).
Instrument Deployment—ARs can be moored,
deployed using only a small anchor (e.g., pop-ups,
AQUAclick, and C-POD), or be used as a stand-
alone instrument (e.g., ARPs and HARPs). These
configurations can also be modified to address
particular characteristics of a deployment area
(i.e., depth, currents, bathymetry, etc.). Some loca-
tions might have existing infrastructure for ocean
instrumentation in the form of large moorings and
other ocean-bottom instrument packages that can
be used to accommodate fixed ARs. Additionally,
existing and planned seafloor ocean observatories
are capable of providing power for ARs via a junc-
tion box or node and may also provide logistical
support in the form of vessels for deployment and
retrieval of instruments (e.g., the Station ALOHA
Cabled Observatory [ACO], Duennebier et al.,
2008; the Victoria Experimental Network Under
the Sea (VENUS), Dewey, 2009). The possibil-
ity of using existing infrastructures from oil and
gas production platforms as moorings or simple
anchoring sites is an interesting and possibly cost-
effective solution for repeated deployment plans.
Additionally, the availability of auxiliary vessels
and highly trained deep diving crews at oil and gas
production sites can significantly reduce the oper-
ational costs of AR deployment, maintenance, and
retrieval. The costs/benefits comparison between
the use of acoustic releases and having a dive team
to recover ARs should be evaluated for each area.
Moorings can either be large with multiple
components distributed vertically along a line,
wire rope, or chain (e.g., EARS; Newcomb et al.,
2009), or be a small, “homemade” anchor attached
to a line or chain (e.g., T-POD; Watkins & Colley,
2004). Moorings can have surface expression
(attached to buoys visible at the surface) or be
completely underwater (subsurface moorings).
Each mooring type has different advantages and
disadvantages. Attaching ARs to large moorings
with surface expressions provides the following
advantages:
• Reliability and ease of relocating instruments
for retrieval
• Possibility of providing power and data
remotely via radio links (Duennebier et al.,
2008; Dewey, 2009)
Disadvantages include the following:
• Need for specialized and costly deployment
equipment when the size of the mooring is large
(large vessels with trained personnel, A-frames,
or cranes are required)
• Possibilityofbuoybeinganavigationalhazard
• Possibilityofdamageanddestructionofbuoys
in areas with natural drifting hazards (e.g., ice
in polar areas; Greene et al., 2004)
• Highertotalequipment weightandsizeduring
deployment and retrieval
• Highervisibilitytopiratesandvandals
• Highersusceptibilitytotheeffectsofstormsor
other episodic weather events
Surface waves and currents introduce consider-
able drag and tension on the mooring line from
bottom to surface. Nevertheless, moorings can be
configured with large mooring lines and consid-
erable flotation and ballast to provide protection
from fishing operations and heavy weather/cur-
rents, ensuring the mooring maintains its position.
Additional requirements to keep large moorings
in position, such as flotation and ballast, will
make them even larger, thus requiring vessels
with heavy lifting capabilities. Smaller moorings
can be deployed by divers or from a small boat,
and the lifting requirements can be minimized by
handling individual components (i.e., flotation,
data recording electronics, batteries, ballast, and
release system) one at a time (Dudzinski et al.,
2011).
The increasing need for PAM in shallow water
environments, where currents, winds, heavy
vessel traffic, and other conflicting activities
are a concern, will require more reliable moor-
ings. Conventional buoys and mooring systems
can require considerable resources to deploy and
recover. Surface buoys may attract undesirable
human attention, unintended snagging/recovery,
or collateral damage from other marine activities,
such as bottom-fishing, especially in coastal areas.
Fixed Autonomous Recorders 43
Subsurface moorings can overcome some of
these concerns with reduced component size, but
they still consist of multiple physical components
(e.g., anchor-weight, securing line, release, pay-
load, and buoyancy unit) and are therefore not
usually well-suited for deployment from small
vessels with limited manpower (Koay et al., 2002).
Some subsurface ARs (e.g., pop-ups, AQUAclick,
PANDA, and DMON) are easily deployed from
a small boat by a few people without specialized
lifting or hoisting equipment. Nonetheless, there
may be a need to use either a dive search team
or special acoustic equipment (e.g., transponders
and acoustic release mechanisms) during retrieval
operations. In cases in which a surface buoy is not
used, there are higher risks of losing the instru-
ment because there is no marker at the surface.
Some devices may offer an option to have a SPG
locator on the subsurface unit, which can solve
this problem at a reasonable cost.
Anchoring or mooring is an important consid-
eration for shallow water deployments. Proper
anchoring is crucial to avoid equipment loss.
Some AR developers provide extensive advice on
anchoring (e.g., Chelonia Limited, 2007, 2012;
T-POD and C-POD user guides; Pop-up user’s
guide; and in-house training), while others pro-
vide little to none. Prior knowledge of the physi-
cal characteristics of the area for deployment is
invaluable when deciding on the type of anchor
necessary. The choice of anchor/mooring type and
weight depends on anticipated bottom type, depth,
and currents expected at a site (Koay et al., 2002).
A few concrete blocks may not be adequate for a
shallow coastal sandy seabed as this is a dynamic
environment and the instrument package might
move in relation to tidal currents and waves.
Massive concrete anchors, digging metal anchors,
or heavy metal anchors are preferable (Chelonia
Limited, 2007). Anchoring also depends on the
size and weight of the instrument. There is great
variation in the dimensions of the instruments
inventoried here, from very small and easily han-
dled by one or two people (e.g., OCEANPOD,
OCEANBASE, DSG-Ocean, PANDA, pop-up,
AQUAclick, DMON, EAR, SM2M, and mini-
AMAR) to very large instruments (e.g., ARPs and
HARPs) that require lifting equipment.
The risk of loss, especially for fully autono-
mous systems, is high if unreliable mooring sys-
tems are used, especially when using directional
particle velocity sensors, in which case the sus-
pension method is extremely important to ensure
that currents and mooring noise do not affect the
sensor (JASCO, 2009b).
Instrument Retrieval—Selection of the most
appropriate method for AR retrieval is based
on water depth, strength of tidal currents, and
composition of the seabed, as well as on the exter-
nal configuration of the AR deployed (e.g., size,
mooring type, etc.). There are many release sys-
tems available (e.g., mechanical or acoustic), some
of which can be quite expensive and often unreli-
able. The relative cost advantage and reliability
of other methods of instrument retrieval (e.g.,
diver retrieval and grappling) should be consid-
ered, in particular in shallow water deployments.
For example, the DASAR used by Blackwell &
Greene (2006) was retrieved by using a double
grapnel anchor assembly with 6 m of chain towed
perpendicular and across to the line where the
DASAR was moored. Chelonia Limited (n.d.b)
shows how to use a vertical grapple in detail.
When retrieval by grappling or diving is not an
option (or becomes difficult due to weather condi-
tions), a backup release system should be imple-
mented to ensure that a malfunction in the primary
retrieval method does not translate into instrument
and data loss.
An example of a mechanical release mechanism
is on the PANDA; this mechanism will keep the
instrument attached to the anchor but floating at
the surface for retrieval (Fiobuoy® release mecha-
nism; Koay et al., 2001). The PANDA is designed
to leave nothing on the seabed after recovery
and thus provides a system that is ecologically
friendly. Some areas (e.g., Marine Parks and
Marine Protected Areas) require special permits to
deploy permanent or semipermanent instruments
in the ocean or on the sea floor. Many also require
that all components of the anchor/mooring are
removed from the seabed after recovery (“nothing
left behind”). In addition, the PANDA’s release
is equipped with an internal leak detector that
will trigger an immediate emergency-surfacing
sequence in case of leak, avoiding serious damage
to the payload and data. Whereas this system has
a desirable design, its limitation is that it cannot
be used in depths over 200 m (Koay et al., 2002).
AR release systems for deep water deployments
may include an acoustic release that can trigger a
mechanical release mechanism or accelerate the
breakage of a corrodible link, or a timed mech-
anism that can activate the release system at a
preset date and time. Pop-ups, ARPs, and HARPs
have been retrieved by activation of a burn-wire
release mechanism. An acoustic command broad-
cast from the recovery vessel using an underwater
speaker causes the release system to apply a volt-
age between the burn-wire and a saltwater ground,
accelerating corrosion of the wire. The corrodible
wire link releases the device from the weight,
and it floats to the surface where it can be either
seen or found via a self-contained VHF beacon
(e.g., included in the pop-up design; Wiggins,
2003; M. Johnson, pers. comm., 25 August 2008;
44 Sousa-Lima et al.
T. Calupca, pers. comm., 14 January 2009). Two
acoustic release systems can be used on the same
AR to provide redundancy and increase the likeli-
hood of instrument recovery in the event of failure
of one release system (Wiggins & Hildebrand,
2007).
An acoustic release mooring can either discon-
nect the AR from its ballast weight, allowing the
instrument to return to the surface, or it can release
a tethered buoy that returns to the surface, allow-
ing the rest of the mooring to be recovered via the
tether line. This system has been used on AMARs.
JASCO also offers a “nothing left behind” option
to deploy/retrieve AMARs that has several advan-
tages—no anchors remain on the bottom, and the
AMAR remains anchored to the bottom when the
float and release surface (ensuring it does not get
lost after the release is triggered). The third anchor
line also remains on the bottom and can be used
for grappling if necessary (JASCO, 2009b).
Note that the release systems discussed here
are not unique to specific ARs. Dudzinski et al.
(2011) offer a detailed review of deployment and
retrieval options and related issues that may be
very useful for users of ARs.
System Customization
AR systems inventoried herein may be modi-
fied to increase some capability in detriment of
another (see “Tradeoffs” section). An AR capabil-
ity that is usually easily modified is total power
capacity. The total system power capacity (Amp-
hours) information is not always available in the
specifications because most AR system configu-
rations are flexible, allowing the user to either
change the number of batteries (e.g., single- and
double-bubble pop-ups) or battery type (e.g., alka-
line battery packs or longer-lasting lithium battery
packs) to accommodate the requirements of a spe-
cific application.
Some instruments inventoried may offer addi-
tional flexibility in other aspects of their design
and configuration to better address requirements
of each user’s applications and deployment areas.
For example, JASCO (2009b) has developed a
re-usable suite of pressure vessels, suspensions,
anchoring systems, and recovery systems that can
be customized to meet most requirements; there-
fore, AMARs can be deployed in shallow water
using a cement block and be retrieved by a diver,
include an acoustic release system for deep water
applications, and also have localization capabili-
ties (directional AMAR configuration, including
a vertical hydrophone array). Another example is
the NOAA/PMAEL AUH that has been modified
to withstand extreme conditions (~0° C water and
strong currents of the Drake Passage). Dziak et al.
(2007) doubled the strength of the mooring line
and replaced the standard laptop hard drives with
a sealed industrial drive that is rated to -20° C for
that application.
The ability to modify or customize a system
can be advantageous for the user. Adaptations
that have proven reliable in the past should be
used if the application so requires, but system
modification prior to extensive field use is not
recommended as unforeseen problems in system
programming and data management could result
in data loss. A change in the instrument soft-
ware to accommodate recording duty cycles or
an increase or decrease in the sampling rate can
render the system unreliable because of program-
ming errors or limitations of the hardware with-
out sufficient testing. Caution should be used and
pilot tests conducted to ensure system reliability
after custom changes.
Noise Issues
Flow and strum noise caused by water motion
over the hydrophone(s) can be a problem for ARs
in some environments. The DASAR overcomes
flow noise using a latex “sock” secured over an
aluminum cage to shield the hydrophone from
water motion (Greeneridge Sciences, n.d.). Other
solutions to this problem include surrounding the
hydrophone with a perforated PVC tube (e.g.,
pop-ups). Unwanted environmental noise (e.g.,
sea-surface noise) can also be reduced by pre-
conditioning signals via band-pass filters (e.g.,
HARPs and PALs).
In any AR system, the hydrophones should be
free of contact with external objects and the sea
floor and not shielded with acoustically absorbing
or reflecting materials (which would impair sensi-
tivity, especially for high-frequency applications).
High-frequency sounds have short wavelengths
that could be missed if parts of the hydrophone
are covered or shielded by components of the AR.
These sound shadows should be avoided by plac-
ing the hydrophones away from the bulk of the
package.
Self-generated noise is also an important con-
cern. One of the key functional constraints of an
autonomous acoustic recording system is elec-
tronic self-noise (Wiggins, 2003). Instruments
that use spinning hard drives or other moving
mechanical parts can generate undesired signals
in the recordings that can mask sounds of interest.
This can also impair or reduce the effectiveness
of automated detection and classification of calls.
The configuration used in ARPs and HARPs keeps
the hydrophone well away from any electronic
noise generated in the instrument itself (approxi-
mately 8.5 m away; Wiggins, 2003). This design
has solved the issue of noise produced by the
hard drives. The use of nonmoving components
Fixed Autonomous Recorders 45
for electronic data storage (i.e., flash media) is
another effective solution (see additional exam-
ples in Table 2).
Deployment Configuration of Multiple Autonomous
Recorders for Localization, Tracking, and Density
Estimation of Marine Mammals
Performing mammal localizations can be achieved
by using a widely spaced array of omnidirectional,
fixed AR units, but such a protocol requires syn-
chronized timing of all units. It is often difficult
to obtain precise timing with autonomous under-
water recorders, each with its own clock drifting
in time independently (Wiggins & Hildebrand,
2007). AR time-synchronization can be achieved
by recording GPS time-linked signals at deploy-
ment and recovery to time-align the record-
ers. This technology is applicable for localizing
acoustic sources such as vessels, seismic sources,
or marine mammals. The sensors should be
spaced appropriately for the desired spatial cover-
age, signal bandwidths, or time resolution, and the
expected signal to noise levels (JASCO, 2009b).
Cornell’s BRP uses sound-based synchroniza-
tion of multiple pop-ups and proprietary software
alignment to achieve the same goals. Performed
at the beginning and end of deployment, syn-
chronization is a normal procedure when using
multiple pop-ups in an array. It involves gather-
ing all the units close together, producing a set of
sharp tones, and accurately recording the onset
times. This simple procedure allows chronologi-
cal matching of recordings from all units during
the entire duration of the deployment.
Directional ARs (e.g., DASAR and AMAR
DV) provide the ability to obtain a bearing to
detected sounds without using arrays of record-
ers. By using an array of these sensors spaced
hundreds of meters apart, cross-fix bearings, and
time-delay-of-arrival, data are collected that can
provide localizations for sources that have lower
energy levels. A single directional recorder can
also be used to track bearings of a source. Using
target motion analysis techniques, the bearings can
be converted into a localization and tracked over
time (JASCO, 2009b). Calibrations with known
position sounds can be very important to the suc-
cess of the triangulation approach to sound-source
localization (Greene et al., 2004). The choice also
depends on what species are being targeted, but
even so, if a single or few species that produce
high-frequency clicks are of interest, it may be
more cost-efficient to use sound detectors (e.g.,
AQUAclick, PAL, C-POD, and T-POD) rather
than high-frequency continuous recorders (e.g.,
HARPs). Additionally, if the area of deployment
is deep, the depth rating and retrieval system of
these instruments may limit the choices available.
Instrument Theft and Vandalism
Theft and vandalism can be a serious risk to AR
retrieval in some areas. Example solutions include
obtaining cooperation and advice from local fish-
ermen; using a very small marker with minimal
surface expression or subsurface moorings; using
acoustic transponder releases; using corrodible
links that dissolve after a predetermined time in
the water and then release a recovery buoy from
the bottom; and using divers to deploy and recover
the instruments (Chelonia Limited, 2011, 2012).
Offering a reward might improve the likelihood of
recovering a lost instrument, sometimes even after
long periods of time (R. S. Sousa-Lima, personal
experience with one pop-up found and returned by
a fisherman a year after losing it). The most effec-
tive, reliable solution is usually some combination
of these options.
System Availability to Users
The pricing and availability of PAM systems
inventoried herein varies depending on the type
of organization that provides access to the tech-
nology. Private companies usually have relatively
straightforward lease or purchase options (provid-
ing user support through manuals or staff), while
developers from academia, research, and govern-
ment institutions have customized agreements for
use, lease, or purchase of their AR systems. Many
of these groups also have technical staff to pro-
vide additional data processing services. Pricing
also varies depending on time demands and on the
amount and type of data processing and analyses
provided. For example, some groups tailor pricing
to adapt to a broad customer audience that varies
from small collaborative research efforts (usu-
ally long-term, small scale) to oil and gas indus-
try contracts (short-term, rapid turnaround, high
demand).
High costs restrict the number of units that can
be deployed and thus reduce the system’s useful-
ness as a monitoring tool (Lammers et al., 2008),
especially when array configurations are needed
to estimate relative numbers and distributions.
Some ARs are available for lower costs in order to
make them more accessible for applications that
require multiple units (e.g., EAR; Lammers et al.,
2008). Availability of multiple devices in a timely
manner depends mostly on the technology provid-
ers, but also on their suppliers. Devices that are
requested often must have production capabilities
that meet these demands.
Discussion
Considerations regarding the use of ARs should
include the frequency band of sounds produced
by the species of interest, the areas and scale over
46 Sousa-Lima et al.
which monitoring is intended, background noise
levels, and the specific goals of the study or moni-
toring effort. For example, a study intended to
detect the occurrence of animals near an oil and
gas platform could use several independent ARs,
whereas one intended to localize and track ani-
mals using their calls would likely use an array of
synchronized ARs with spacing that would allow
tracking over ranges of interest.
ARs can be used in every stage of a well-
designed study. ARs are extremely valuable for
the early stages of acoustic prospecting when
information can be gathered before E&P activities
begin. The timing of changes in relative numbers
of animals is important information that can be
used to schedule exploratory activities (e.g., seis-
mic studies), as well as to determine the effects of
production and transportation activities on animal
occurrence and behavior.
The use of ARs is an effective method for
acoustically monitoring marine mammals and
especially for identifying which species are pres-
ent in a given area at a given time (Clark & Charif,
1998; Stafford et al., 1999, 2007; Nieukirk et al.,
2004; Heimlich et al., 2005; Mellinger et al.,
2007b; Širović et al., 2009), locating and track-
ing individuals (Sousa-Lima & Clark, 2008,
2009), identifying sounds associated with dif-
ferent regions (Stafford et al., 1999, 2001), and
determining patterns of distribution and rela-
tive abundance (Mellinger et al., 2004a, 2004b).
Among the main constraints for analyzing and
interpreting acoustic data collected using ARs is
the difficulty of associating the number of sounds
recorded with the number of animals present; the
detection range and location of the sounds; as
well as the seasonal, behavioral, and demographic
variations in the calling behavior of different spe-
cies (Clark & Charif, 1998; Mellinger & Barlow,
2003). The extent to which these types of informa-
tion can be obtained depends on how the study
design takes environmental and biological aspects
into account and on how AR units are deployed
(e.g., the number of units deployed and the geo-
metric spatial arrangement of the ARs).
Fixed PAM, such as ARs, will continue to be
one of the most cost-effective ways to remotely
monitor marine mammal species and their sur-
roundings and to collect data on how human
activities are affecting these dynamic systems.
McDonald et al.’s (1995) study which incidentally
detected whale calls using OBSs, also recorded
noise from seismic air guns and from ship traffic
and is a good example of how ARs can be effec-
tive for monitoring noise produced by oil and gas
E&P activities while also monitoring the occur-
rence, acoustic behaviors, and movements of ani-
mals in the area.
The demand for offshore petroleum and gas
will provide many opportunities to study the
effects of oil and gas E&P activities on marine
mammals. Underwater sounds produced by
E&P activities are superimposed onto an already
dynamic and complex acoustic marine environ-
ment. The world’s ocean can be seen as a mosaic
of areas with different animal acoustic ecologies
and levels of human disturbance. This mosaic of
soundscapes provides opportunities to acousti-
cally compare the effects of noise across different
areas with different levels of disturbance within
a similar habitat (e.g., whale breeding areas in
pristine and disturbed areas), and within a par-
ticular area across time (e.g., before, during, and
after study designs in areas with planned oil and
gas E&P activities). Fixed PAM technologies are
well-suited for these types of investigations. Using
data collected from ARs, statistical models can be
derived to explain the effects of many naturally
occurring and anthropogenic phenomena (e.g.,
Sousa-Lima & Clark, 2008).
The Use of Fixed Autonomous Recorders in Marine
Mammal Monitoring and Mitigation During Oil and
Gas E&P Activities
Seismic Surveys—Some regions in the world that
are important for oil and gas exploration are also
areas of high marine mammal densities. When
baseline data on species occurrence and seasonal-
ity exists, this information should guide the choice
of fixed AR systems used. When such information
is not available, the best approach might be to
deploy a variety of AR systems that can cover a
broad frequency band to gather information about
as many species of marine mammals that might
be present as possible ahead of time (ideally com-
mencing as soon as the area becomes of interest to
the oil and gas industry and continuing for at least
1 y) to facilitate data collection on the seasonality
of species occurrence.
As discussed earlier, a high sample rate equates
to increased power supply and storage capac-
ity demands, which tend to increase the cost of a
system. More sophisticated and costly AR pack-
ages can be used whenever their capabilities, such
as continuous recordings at high-sampling rates
(e.g., HARPs) or greater longevity, are neces-
sary to target specific locations, species, or both
(e.g., for recording beaked whales; Johnston et al.,
2008). For greater coverage or for sampling among
several locations simultaneously, less expensive
equipment can be deployed in a multisensor array
(e.g., EARs; Lammers et al., 2008). The bathym-
etry of the area to be monitored should also be
taken into account. Shallow water propagation
effects may diminish the area over which sounds
can be heard; therefore, more densely populated
Fixed Autonomous Recorders 47
arrays of sensors should be deployed if localiza-
tion capabilities are necessary in areas with prop-
agation issues. Very shallow areas can be moni-
tored using AR systems that are deployed on or
near the ocean bottom (such as the AQUAclick,
pop-ups, and EAR), thus avoiding mooring lines
that can be a hazard to the towed seismic array.
The distribution of seismic exploration activi-
ties should provide ample opportunity to carry out
controlled experiments in collaboration with the
oil and gas industry to identify the effects of seis-
mic activities on the observed behavior or distri-
bution of marine mammals. Opportunistic experi-
ments to determine the effect of seismic surveys
on marine mammal vocal behavior can also be
conducted using ARs while seismic exploration
is ongoing (e.g., Nieukirk et al., 2004; Di Iorio
& Clark, 2010). Both planned and opportunistic
experiments should take into account biologi-
cal and environmental factors that vary spatially
(e.g., bathymetry and water temperature that
affect sound speed profiles and marine mammal
food resources), which may influence the natural
fluctuations in occurrence and vocal activity of
marine mammals.
Construction and Installation of Platforms and
Seabed Production Units—Activities associated
with the construction and installation of platforms
and other production units generate underwater
noise. Blackwell & Greene (2006) determined
the levels, characteristics, and range dependence
of underwater and in-air sounds produced by the
Northstar oil development, located in nearshore
waters of the Alaskan Beaufort Sea. Vessels (i.e.,
crew boat, tugs, and self-propelled barges) were
the main contributors to the underwater sound
field and were often detectable as far as 30 km
offshore. When vessels were not operating, broad-
band noise from the Northstar rig reached back-
ground levels at a distance of 2 to 4 km from the
source. Northstar sound levels showed more vari-
ation during construction of the island than during
drilling and production.
The typical occurrence of multiple platforms in
an oil production area would allow multiple ARs
to be mounted on or close to the platforms, pro-
viding a cost-effective way to deploy an array of
ARs. This could increase the capabilities of the
hydrophone system to allow much greater geo-
graphic coverage and easy maintenance, but also
would allow the possibility of tracking individual
animals and groups. To do this, it is necessary
to have some knowledge of the sound propaga-
tion properties and underwater noise budget of
the deployment area so that hydrophones can be
deployed in a spatial configuration that allows
acoustic coverage of the area of interest and so
that sounds from vocalizing individuals can be
detected on multiple hydrophones (at least three).
Continuous time synchronization of these hydro-
phones could be achieved very effectively by
monitoring the exact locations and times of occur-
rence of acoustic events that are detected by all
hydrophones. These acoustic events need not be
application-specific and could be existing tran-
sient noises generated by the normal activities of
the production platform.
Oil and Gas Transportation—Vessel traffic has
been shown to cause disturbances in the behavior
of several species of marine mammals, including
humpback whales (Sousa-Lima & Clark, 2008,
2009), gray whales (Eschrichtius robustus; Bryant
et al., 1994), blue and fin whales (McDonald
et al., 1995), and belugas (Delphinapterus leucas)
and narwhals (Monodon monoceros) (Finley
et al., 1990), to name a few. Additionally, ship-
ping noise (i.e., background noise from shipping
vessel traffic) is the most important contributor to
the increase in ocean background noise levels over
the last decades (McDonald et al., 2006). Vessel
transportation of commodities (including oil and
gas) is contributing to overall noise pollution in
areas far removed from the production activities
(McDonald et al., 1995). Fixed PAM using ARs
in areas around shipping lanes can be an effective
way to monitor how this source of noise is poten-
tially affecting marine mammals by measuring
how much of their acoustic habitat is being lost
(Clark et al., 2009).
Potential Areas for Further Development
Increased Power Capacity and Low-Power
Electronics—Wiggins & Hildebrand (2007) point
out that as larger capacity disks become available,
longer deployments at higher sampling rates will
be possible. This will require additional batter-
ies which, in turn, means additional weight and
additional buoyancy to compensate. On the other
hand, lower power electronics and faster data
transfer rates from a memory buffer to data stor-
age disks (i.e., disks powered for shorter periods)
or flash memory cards could provide alternative
means for longer deployments with the same or
fewer batteries and lighter components. Several
developers are planning higher capacity systems.
Advances in consumer digital electronics
(e.g., music players, phones, cameras, etc.) have
resulted in dramatic improvements in high-capac-
ity solid-state and low-power processer memory.
Microprocessors with lower power consumption
would allow longer deployment periods and/or
higher sampling frequencies. These advancements
are going to be extremely important in future AR
technologies and will affect all aspects of AR
design and configuration. The result will be lower
data storage costs, lower power requirements, and
48 Sousa-Lima et al.
faster data-transfer and writing rates. Flash stor-
age media are replacing the energy-intensive,
motorized disk drives currently used (e.g., next
generation pop-ups; T. Calupca, pers. comm.,
14 January 2009).
Higher energy capacity batteries (e.g., lithium
chemistry) will likely be used to provide extra
power with the same number of batteries. Until
these higher power batteries are used in ARs, an
alternative approach suggested by Wiggins &
Hildebrand (2007) is to house the extra alkaline
batteries separately from the housing containing
instrument electronics and to jettison the battery
pack during instrument recovery. This would
result in less required buoyancy and smaller
instrument packaging.
The use of solar cells on surface moorings is
also a possibility for increasing deployment dura-
tion without increasing size. Other capabilities,
such as USB interface for data downloading and
rapid battery recharging already implemented in
some systems (e.g., AQUAclick and AMAR),
allow quick download of the collected data with-
out having to open the main housing. This capa-
bility optimizes ship time and reduces deployment
and retrieval costs significantly (Shariat-Panahi
et al., 2008). More efficient ways to accomplish
this should be explored.
Information Networks and Integration—
Underwater networks of acoustic relays using
wireless modems and receivers with networking
capabilities (e.g., AquaNetwork and DSPComm),
uncabled autonomous near-real-time systems, or
acoustic links offer new ways to communicate data
in underwater channels. Ocean observatories are
profiting from these possibilities for integrating
underwater observation systems (Duennebier et al.,
2008; Dewey, 2009). Details on these technologies
are beyond the scope of this review but are fertile
ground for advancements in fixed PAM systems.
Concluding Remarks
The wide range of AR capabilities reviewed
herein is the result of different application needs
that have dictated the design and configuration of
ARs. Original AR applications were not necessar-
ily directed at servicing the oil and gas industry
as they were mostly built for achieving specific
research objectives and for noncommercial pur-
poses. As the research demand increased, devel-
opers expanded AR capabilities to collect acoustic
data for longer periods, in more remote areas, and
covering as many species as possible (C. W. Clark,
pers. comm., 28 November 2009). Monitoring and
mitigation requirements from regulatory institu-
tions that the oil and gas industry must adhere to
will be better achieved as the existing technology
develops.
Details on state-of-the-industry AR technology
are inherently outdated since this technology is
moving forward at a fast pace. Any information
shown herein that appears to be out-of-date is
likely related to a lack of response from develop-
ers or because of websites that were not updated.
Acknowledgments
We thank Whitlow Au and Christopher W. Clark
for their comments, and all the developers and
users for providing unpublished information and
personal communications. Thanks also to the
Joint Industry Program (JIP) for funding this
review, and to FAPEMIG and CAPES (Brazilian
Government) for post-doctoral fellowships to
RSSL and a scholarship to DPF.
Literature Cited
Acousonde. (n.d.). Acousonde™ [Brochure]. Retrieved
4 February 2013 from www.acousonde.com/downloads/
Acousonde3A_Brochure.pdf; www.acousonde.com/
downloads/Acousonde3B_Brochure.pdf.
Acoustics. (2004). Acoustics, Volume 2 (Part of Chapters 5-6
[AUAR application]). Retrieved 28 January 2013 from
www.sakhalinenergy.ru/en/documents/Acoustics%20
Vol2%20(2004)%20Chp%205-6.pdf.
Akamatsu, T., Nakazawa, I., Tsuchiyama, T., & Kimura,
N. (2008). Evidence of nighttime movement of fin-
less porpoise through Kanmon Strait monitored using
a stationary acoustic recording device. Fisheries
Science, 74, 970-975. http://dx.doi.org/10.1111/j.1444-
2906.2008.01614.x
Akamatsu, T., Wang, D., Wang, K., & Naito, Y. (2000). A
method for individual identification of echolocation sig-
nals in free-ranging finless porpoises carrying data log-
gers. The Journal of the Acoustical Society of America,
108, 1353-1356. http://dx.doi.org/10.1121/1.1287841
Akamatsu, T., Kimura, S., Li, S., Dong, L., Wang, K., &
Wang, D. (2011). A stereo acoustic event recorder for
monitoring abundance and movements of dolphins
and porpoises. IEEE Symposium on Underwater
Technology, UT ‘11 and Workshop on Scientific Use of
Submarine Cables and Related Technologies, SSC ‘11.
http://dx.doi.org/10.1109/UT.2011.5774144
Akamatsu, T., Matsuda, A., Suzuki, S., Wang, D., Wang, K.,
Suzuki, M., . . . Oota, K. (2005). New stereo acoustic
data logger for free-ranging dolphins and porpoises.
Marine Technology Society Journal, 39(2), 3-9. http://
dx.doi.org/10.4031/002533205787443980
Anagnostou, M. N., Nystuen, J. A., Anagnostou, E. N.,
Papadopoulos, A., & Lykousis, V. (2011). Passive
Aquatic Listener (PAL): An adoptive underwater
acoustic recording system for the marine environment.
Nuclear Instruments in Physics Research, A, 626-627,
S94-S98. http://dx.doi.org/10.1016/j.nima.2010.04.140
Fixed Autonomous Recorders 49
AQUATEC Group Ltd. (AQUATEC). (n.d.). The AQUAclick
porpoise click logger. Retrieved 28 January 2013 from
www.aquatecgroup.com/index.php/products/aquaclick.
Arias, A., Johnson, M., Aguilar Soto, N., Madsen, P. T.,
Tyack, P., & Møhl, B. (2008). Acoustic detection of
beaked whales from autonomous recording buoys. The
Journal of the Acoustical Society of America, 123, 3207.
http://dx.doi.org/10.1121/1.2933378
A-tag. (2012). A-tag, a multi-platform acoustic data logger.
Retrieved 28 January 2013 from http://cse.fra.affrc.go.
jp/akamatsu/A-tag/index.html.
Au, W. W. L., Mobley, J., Burgess, W. C., & Lammers,
M. O. (2000). Seasonal and diurnal trends of chorusing
humpback whales wintering in waters off western Maui.
Marine Mammal Science, 16, 530-544. http://dx.doi.
org/10.1111/j.1748-7692.2000.tb00949.x
Baumann, S., Hildebrand, J. A., Wiggins, S. M., &
Schnitzler, H-U. (2008). Species identification and mea-
surement of activity in odontocete species of Palmyra
Atoll by acoustic monitoring. The Journal of the
Acoustical Society of America, 123, 3099. http://dx.doi.
org/10.1121/1.2932958
Blackwell, S. B., & Greene, C. R., Jr. (2006). Sounds from
an oil production island in the Beaufort Sea in summer:
Characteristics and contribution of vessels. The Journal
of the Acoustical Society of America, 119, 182-196.
http://dx.doi.org/10.1121/1.2140907
Blackwell, S. B., Richardson, W. J., Greene, C. R., Jr., &
Streever, B. (2007). Bowhead whale (Balaena mysti-
cetus) migration and calling behaviour in the Alaskan
Beaufort Sea, Autumn 2001-04: An acoustic localization
study. Arctic, 60, 255-270.
Blackwell, S. B., McDonald, T. L., Nations, C. S., Thode,
A. M., Kim, C. H., Greene, C. R., Jr., & Macrander, M. A.
(2012). Bowhead whales and airgun pulses: Detecting
a threshold of behavioral reaction. The Journal of the
Acoustical Society of America, 132, 1949. http://dx.doi.
org/10.1121/1.4755172
Bluejay, M. (2009). Battery guide. Retrieved 28 January
2013 from http://michaelbluejay.com/batteries.
Borisov, S. V., Kovzel, D. G., Rutenko, A. N., &
Uschipovsky, V. G. (2008, October). Transmitting
autonomous underwater acoustic recorder – SHELF-07.
XX Session of the Russian Acoustical Society, Moscow.
Bradbury, J. W., & Vehrencamp, S. L. (2011). Principles
of animal communication (2nd ed.). Sunderland, MA:
Sinauer Associates.
Bryant, P. J., Lafferty, C. M., & Lafferty, S. K. (1984).
Reoccupation of Laguna Guerrero Negro, Baja
California, Mexico, by gray whales. In M. L. Jones,
S. L. Swartz, & S. Leatherwood (Eds.), The gray whale,
Eschrichtius robustus (pp. 375-387). Orlando, FL:
Academic Press. http://dx.doi.org/10.1016/B978-0-08-
092372-7.50021-2
Burgess, W. C. (2000). The bioacoustic probe: A general-
purpose acoustic recording tag. The Journal of the
Acoustical Society of America, 108, 2583. http://dx.doi.
org/10.1121/1.4743598
Burgess, W. C., Oleson, E. M., & Baird, R. W. (2011). A
hydrodynamic acoustic recording tag for small ceta-
ceans and first results from a pantropical spotted dol-
phin. Poster presented at the 19th Biennial Conference
on the Biology of Marine Mammals, Tampa, FL.
Burgess, W. C., Tyack, P. L., Le Boeuf, B. J., & Costa, D. P.
(1998). A programmable acoustic recording tag and
first results from free-ranging northern elephant seals.
Deep-Sea Research Part II, 45, 1327-1351. http://dx.doi.
org/10.1016/S0967-0645(98)00032-0
Calambokidis, J., Schorr, G. S., Steiger, G. H., Francis,
J., Bakhtiari, M., Marshall, G. J., . . . Robertson, K.
(2007). Insights into the underwater diving, feeding,
and calling behavior of the blue whales from a suction-
cup-attached videoimaging tag (Crittercam). Marine
Technology Society Journal, 41(4), 19-29. http://dx.doi.
org/10.4031/002533207787441980
Calupca, T. A., Fristrup, K. M., & Clark, C. W. (2000). A
compact digital recording system for autonomous bio-
acoustic monitoring. The Journal of the Acoustical
Society of America, 108, 2582. http://dx.doi.org/10. 1121/
1.4743595
Centre for Marine Science and Technology (CMST). (2011).
Underwater sound recorders. Retrieved 28 January 2013
from http://cmst.curtin.edu.au/products/usr.cfm.
Cerchio, S., Jacobsen, J. K., & Norris, T. F. (2001). Temporal
and geographical variation in songs of humpback whales,
Megaptera novaeangliae: Synchronous change in Hawaiian
and Mexican assemblages. Animal Behaviour, 62, 313-329.
http://dx.doi.org/10.1006/anbe.2001.1747
Cetacean Research Technology. (2012). Remote Underwater
Digital Acoustic Recorders nRUDAR™, µRUDAR™,
mRUDAR™, RUDAR™. Retrieved 28 January 2013
from www.cetaceanresearch.com/hydrophone-systems/
rudar/index.html.
Chelonia Limited. (2007). T-POD user guide. Cornwall,
UK: Chelonia Limited, Cetacean Monitoring Systems.
Retrieved 17 January 2013 from www.chelonia.co.uk.
Chelonia Limited. (2011). Deep C-POD user guide. Cornwall,
UK: Chelonia Limited, Cetacean Monitoring Systems.
Retrieved 17 January 2013 from www.chelonia.co.uk.
Chelonia Limited. (2012). C-POD user guide. Cornwall,
UK: Chelonia Limited, Cetacean Monitoring Systems.
Retrieved 17 January 2013 from www.chelonia.co.uk.
Chelonia Limited. (n.d.a). About the T-POD. Retrieved
17 January 2013 from www.chelonia.co.uk/about_the_
tpod.htm.
Chelonia Limited. (n.d.b). Vertical grapple. Retrieved
17 January 2013 from www.chelonia.co.uk/vertical_
grapple.htm.
Chelonia Limited. (n.d.c). C-POD specification. Retrieved
17 January 2013 from www.chelonia.co.uk/cpod_
specification.htm.
Clark, C. W., & Charif, R. A. (1998). Acoustic monitor-
ing of large whales to the west of Britain and Ireland
using bottom-mounted hydrophone arrays, October
1996-September 1997 (Report No. 281). Aberdeen,
Scotland: Joint Nature Conservation Committee.
50 Sousa-Lima et al.
Clark, C. W., & Fristrup, K. (1997). Whales ‘95: A com-
bined visual and acoustic survey of blue and fin whales
off southern California. Reports of the International
Whaling Commission, 47, 583-600.
Clark, C. W., Charif, R., Mitchell, S., & Colby, J. (1996).
Distribution and behavior of the bowhead whale,
Balaena mysticetus, based on analysis of acoustic data
collected during the 1993 spring migration off Point
Barrow, Alaska. Reports of the International Whaling
Commission, 46, 541-552.
Clark, C. W., Gillespie, D., Moscrop, A., Fowler, T.,
Calupca, T., & Fowler, M. (2000). Acoustic sampling
of right whale vocalizations in the Great South Channel
using sea-floor pop-up recorders (Technical Report).
Boston: Right Whale Consortium.
Clark, C. W., Ellison, W. T., Southall, B. L., Hatch, L.,
Van Parijs, S. M., Frankel, A., & Ponirakis, D. (2009).
Acoustic masking in marine ecosystems: Intuitions, anal-
ysis and implication. Marine Ecology Progress Series,
395, 201-222. http://dx.doi.org/10.3354/meps08402
Curie, J., & Curie, P. (1880a). Développement, par pres-
sion, de l’électricité polaire dans les cristaux hémièdres
à faces inclinées [Development through pressure of
the electrical polarity in hemiedral inclined face crys-
tals]. Comptes Rendus de l’Académie des Sciences, 91,
294-295.
Curie, J., & Curie, P. (1880b). Contractions et dilatations
produites par des tensions dans les cristaux hémièdres à
faces inclinées [Contractions and dilations produced by
straining in hemiedral inclined face crystals]. Comptes
Rendus de l’Académie des Sciences, 93, 1137-1140.
Darling, J. D., & Sousa-Lima, R. S. (2005). Songs indicate
interaction between humpback whale (Megaptera novae-
angliae) population in the western and eastern South
Atlantic Ocean. Marine Mammal Science, 21, 557-566.
http://dx.doi.org/10.1111/j.1748-7692.2005.tb01249.x
Dewey, R. (2009). The VENUS ocean cabled observatory.
CMOS Bulletin, 37(3), 77-82.
Di Iorio, L., & Clark, C. W. (2010). Exposure to seismic
surveys alters blue whale acoustic communication.
Biology Letters, 6, 51-54. http://dx.doi.org/10.1098/rsbl.
2009.0651
Discovery of Sounds in the Sea (DOSITS). (2011). Acoustic
datalogging systems. Retrieved 28 January 2013 from
www.dosits.org/technology/observermarineanimals/
acousticdataloggingsystems.
Dorman, L. M. (2001). Seismology sensors. In J. H. Steele,
K. K. Turekian, & S. A. Thorpe (Eds.), Encyclopedia of
ocean science (pp. 2737-2744). San Diego: Academic
Press. http://dx.doi.org/10.1006/rwos.2001.0334; http://
dx.doi.org/10.1016/B978-012374473-9.00334-9
Dudzinski, K. M., Brown, S. J., Lammers, M., Lucke, K.,
Mann, D. A., Simard, P., . . . Eriksen, N. (2011). Trouble-
shooting deployment and recovery options for various
stationary passive acoustic monitoring devices in both
shallow- and deep-water applications. The Journal of
the Acoustical Society of America, 129, 436-448. http://
dx.doi.org/10.1121/1.3519397
Duennebier, F., Harris, D., & Jolly, J. (2008). ALOHA
cabled observatory will monitor ocean in real time. Sea
Technology, 49(2), 51-54.
Dziak, R. P., Park, M., Matsumoto, H., Bohnenstiehl,
D. R., Haxel, J. H., Mellinger, D. K., . . . Won Sang
Lee. (2007). Hydroacoustic monitoring of the Bransfield
strait and Drake Passage, Antarctica: A first analy-
sis of seafloor seismicity, cryogenic acoustic sources,
and cetacean vocalizations. Proceedings of the Tenth
International Symposium on Antarctic Earth Sciences,
Santa Barbara, CA.
Finley, K. J., Miller, G. W., Davis, R. A., & Greene, C. R.,
Jr. (1990). Reactions of belugas, Delphinapterus leucas,
and narwhals, Monodon monoceros, to ice-breaking
ships in the Canadian High Arctic. Canadian Bulletin of
Fisheries and Aquatic Science, 224, 97-117.
Fletcher, S., Le Boeuf, B. J., Costa, D. P., Tyack, P. L., &
Blackwell, S. B. (1996). Onboard acoustic recording
from diving northern elephant seals. The Journal of the
Acoustical Society of America, 100, 2531-2539. http://
dx.doi.org/10.1121/1.417361
Fowler, M. (2003). AUHs monitor submarine volcanoes
and gentle giants of the deep. Retrieved 28 January 2013
from http://oceanexplorer.noaa.gov/explorations/03fire/
logs/feb22/feb22.html.
Fox, C. G., Matsumoto, H., & Lau, T. A. (2001). Monitoring
Pacific Ocean seismicity from an autonomous hydro-
phone array. Journal of Geophysical Research, 106(B3),
4183-4206. http://dx.doi.org/10.1029/2000JB900404
Frankel, A. S., & Clark, C. W. (1998). Results of low-
frequency m-sequence noise playbacks to humpback
whales in Hawai’i. Canadian Journal of Zoology, 76,
521-535. http://dx.doi.org/10.1139/cjz-76-3-521; http://
dx.doi.org/10.1139/z97-223
Fristrup, K., & Clark, C. W. (1997). Combining visual and
acoustic survey data to enhance density estimation.
Reports of the International Whaling Commission, 47,
933-936.
Gedamke, J. (2005, Spring). Sounds in the “silent world.”
Australian Antarctic Magazine, 9, 14-15.
Gedamke, J., Gales, N., Hildebrand, J., & Wiggins, S.
(2007). Seasonal occurrence of low frequency whale
vocalizations across eastern Antarctic and southern
Australian waters, February 2004 to February 2007
(Report for the IWC SC/59/SH5). Cambridge, UK:
International Whaling Commission.
Greene, C. R., Jr., McLennan, M. W., Norman, R. G.,
McDonald, T. L., Jakubczak, R. S., & Richardson, W. J.
(2004). DIFAR sensors in seafloor recorders to locate
calling bowhead whales during their fall migration. The
Journal of the Acoustical Society of America, 115, 2346-
2357.
Greeneridge Sciences. (n.d.). Technology. Retrieved 28 January
2013 from www.greeneridge.com/technology.html.
Hayes, S. A., Mellinger, D. K., Croll, D. A., Costa, D. P., &
Borsani, J. F. (2000). An inexpensive passive acoustic
system for recording and localizing wild animal sounds.
Fixed Autonomous Recorders 51
The Journal of the Acoustical Society of America, 107,
3552-3555. http://dx.doi.org/10.1121/1.429424
Heimlich, S. L., Mellinger, D. K., & Nieukirk, S. L. (2005).
Types, distribution, and seasonal occurrence of sounds
attributed to Bryde’s whales (Balaenoptera edeni)
recorded in the eastern tropical Pacific, 1999-2001. The
Journal of the Acoustical Society of America, 118, 1830-
1837. http://dx.doi.org/10.1121/1.1992674
High Tech, Inc. (2005). SRB-16 autonomous recording
buoy. Retrieved 29 July 2012 from www.hightechincusa.
com/taab.html#SRB-16.
Ichikawa, K., Tsutsumi, C., Arai, N., Akamatsu, T., Shinke,
T., Hara, T., & Adulyanukosol, K. (2006). Dugong
(Dugong dugon) vocalization patterns recorded by
automatic underwater sound monitoring systems. The
Journal of the Acoustical Society of America, 119, 3726-
3733. http://dx.doi.org/10.1121/1.2201468
Insley, S. J., Robson, B. W., Yack, T., Ream, R. R., &
Burgess, W. C. (2007). Acoustic determination of activ-
ity and flipper stroke rate in foraging northern fur seal
females. Endangered Species Research, 4, 147-155.
http://dx.doi.org/10.3354/esr00050
Ioup, G. E., Ioup, J. W., Pflug, L. A., Tashmukhambetov, A.,
Sidorovskaia, N. A., Schexnayder, P., . . . Ekimov, A.
(2009, May). EARS buoy applications by LADC: I.
Marine animal acoustics. Proceedings of MTS/IEEE
Oceans ‘09, Brennen, Germany.
JASCO Applied Sciences (JASCO). (2009a). AMAR:
Autonomous Multi-channel Acoustic Recorder [Brochure].
Hallifax, Nova Scotia: JASCO Systems Division.
JASCO. (2009b). Information package provided by JASCO
applied sciences to biowaves: JASCO’s AMAR systems
capabilities. Hallifax, Nova Scotia: Martin, B.
JASCO. (2012). AMAR G3 Autonomous Multichannel
Acoustic Recorder – Generation 3 brochure. Victoria,
Nova Scotia: JASCO Systems Division.
Johnson, M. P., & Tyack, P. L. (2003). A digital acous-
tic recording tag for measuring the response of wild
marine mammals to sound. IEEE Journal of Oceanic
Engineering, 28(1), 3-12. http://dx.doi.org/10.1109/JOE.
2002.808212
Johnson, M. P., Madsen, P. T., Zimmer, W. M. X., Aguilar
de Soto, N., & Tyack, P. L. (2006). Foraging Blainville’s
beaked whales (Mesoplodon densirostris) produce dis-
tinct click types matched to different phases of echolo-
cation. The Journal of Experimental Biology, 209, 5038-
5050. http://dx.doi.org/10.1242/jeb.02596
Johnston, D. W., McDonald, M., Polovina, J., Domokos,
R., Wiggins, S., & Hildebrand, J. (2008). Temporal pat-
terns in the acoustic signals of beaked whales at Cross
Seamount. Biology Letters, 4, 208-211. http://dx.doi.
org/10.1098/rsbl.2007.0614
Kimura, S., Akamatsu, T., Wang, K., Wang, D., Li, S., &
Dong, S. (2009). Comparison of stationary acoustic
monitoring and visual observation of finless porpoises.
The Journal of the Acoustical Society of America, 125,
549-553. http://dx.doi.org/10.1121/1.3021302
Koay, T. B., Potter, J. R., Johansson, T., & Venugopalan, P.
(2001, November). PANDA: A self-recovering shal-
low water acoustic logger. OCEANS 2001 MTS/IEEE
Conference and Exhibition, Honolulu, HI. http://dx.doi.
org/10.1109/OCEANS.2001.968132
Koay, T. B., Potter, J. R., Venugopalan, P., & Tan, E. T.
(2002). PANDA: A rapidly deployable, self-recov-
ering shallow water acquisition platform. Retrieved
28 January 2013 from http://arl.nus.edu.sg/twiki/pub/
ARL/BibEntries/Koay2002a.pdf.
Koay, T. B., Seeking, P. J., Chitre, M., Tan, S. P., & Hoffmann-
Kuhnt, M. (2006). Advanced PANDA for high speed auton-
omous ambient noise data collection and boat tracking –
System and results. Oceans 2006 Conference, Asia Pacific.
http://dx.doi.org/10.1109/OCEANSAP.2006.4393919
Kyhn, L. A., Tougaard, J., Amundin, M., Stenback, J.,
Teilmann, J., & Wennerberg, W. (2008). Validating pas-
sive acoustic monitoring data loggers by visual observa-
tions. The Journal of the Acoustical Society of America,
123, 3208. http://dx.doi.org/10.1121/1.2933380
Laboratório de Dinâmica e Instrumentação (LADIN).
(2012). Sensoriamento acústico [Acoustic sensing].
Retrieved 28 January 2013 from www.ladin.usp.br/
ACUSTSUB.html.
Lammers, M. O., Brainard, R. E., Au, W. W. L., Mooney,
T. A., & Wong, K. B. (2008). An Ecological Acoustic
Recorder (EAR) for long-term monitoring of biological
and anthropogenic sounds on coral reefs and other marine
habitats. The Journal of the Acoustical Society of America,
123, 1720-1728. http://dx.doi.org/10.1121/1.2836780
Loggerhead Instruments. (2012). Shop: DSG-ocean acous-
tic datalogger. Retrieved 1 February 2013 from http://
loggerheadinstruments.com/shop/dsg-ocean-acoustic-
datalogger.
Loncarevic, B. D. (1977). Introduction to the OBS review.
Marine Geophysical Research, 3, 5. http://dx.doi.org/10.
1007/BF00309791
Madsen, P. T., Payne, R., Kristiansen, N. U., Wahlberg, M.,
Kerr, I., & Møhl, B. (2002). Sperm whale sound produc-
tion studied with ultrasound time/depth-recording tags.
The Journal of Experimental Biology, 205, 1899-1906.
Marshall, G. J. (1998). Crittercam: An animal-borne imag-
ing and data logging system. Marine Technology Society
Journal, 32(1), 11-17.
McDonald, M. A., Hildebrand, J. A., & Webb, S. C. (1995).
Blue and fin whales observed on a seafloor array in the
northeast Pacific. The Journal of the Acoustical Society of
America, 98, 712-721. http://dx.doi.org/10.1121/1.413565
McDonald, M. A., Hildebrand, J. A., & Wiggins, S. (2006).
Increases in deep ocean ambient noise in the northeast
Pacific west of San Nicolas Island, California. The
Journal of the Acoustical Society of America, 120, 711-
718. http://dx.doi.org/10.1121/1.2216565
Mellinger, D., & Barlow, J. (2003). Future directions for
acoustic marine mammal surveys: Stock assessment and
habitat use (NOAA OAR Special Report, NOAA/PMEL
Contribution No. 2557). La Jolla, CA: National Oceanic
and Atmospheric Administration.
52 Sousa-Lima et al.
Mellinger, D. K., Stafford, K. M., & Fox, C. G. (2004a).
Seasonal occurrence of sperm whale (Physeter
macrocephalus) sounds in the Gulf of Alaska, 1999-
2001. Marine Mammal Science, 20, 48-62. http://dx.doi.
org/10.1111/j.1748-7692.2004.tb01140.x
Mellinger, D. K., Stafford, K. M., Moore, S. E., Dziak,
R. P., & Matsumoto, H. (2007a). An overview of fixed
passive acoustic observation methods for cetaceans.
Oceanography, 20, 36-45. http://dx.doi.org/10.5670/
oceanog.2007.03
Mellinger, D. K., Stafford, K. M., Moore, S. E., Munger, L.,
& Fox, C. G. (2004b). Detection of North Pacific right
whale (Eubalaena japonica) calls in the Gulf of Alaska.
Marine Mammal Science, 20, 872-879. http://dx.doi.
org/10.1111/j.1748-7692.2004.tb01198.x
Mellinger, D. K., Nieukirk, S. L., Matsumoto, H., Heimlich,
S. L., Dziak, R. P., Haxel, J., . . . Miller, H. V. (2007b).
Seasonal occurrence of North Atlantic right whales
(Eubalaena glacialis) vocalizations at two sites on the
Scotian Shelf. Marine Mammal Science, 23, 856-867.
http://dx.doi.org/10.1111/j.1748-7692.2007.00144.x
Ming-Hao, C., Chau-Chang, W., Jin-Yuan, L., & Chia-Wei,
C. (2007, April). Design and application of autono-
mous underwater acoustic recorder. Symposium on
Underwater Technology and Workshop on Scientific Use
of Submarine Cables and Related Technologies, Tokyo,
Japan. http://dx.doi.org/10.1109/UT.2007.370949
MIT Seagrant. (n.d.). CFER project on passive acoustic
applications in marine fisheries. Retrieved 28 January
2013 from http://seagrant.mit.edu/cfer/passiveacoustics/
paresearch.html.
Møhl, B., Wahlberg, M., & Heerfordt, A. (2001). A large-
aperture array of nonlinked receivers for acoustic posi-
tioning of biological sound sources. The Journal of the
Acoustical Society of America, 109, 434-437. http://
dx.doi.org/10.1121/1.1323462
Multi-Électronique Inc. (MTE). (2012). AURAL M-2
(Autonomous Underwater Recorder for Acoustic
Listening-Model 2). Retrieved 28 January 2013 from
www.multi-electronique.com/pages/auralm2en.htm.
Munger, L. M., Mellinger, D. K., Wiggins, S., Moore, S. E.,
& Hildebrand, J. A. (2005). Performance of spectro-
gram cross-correlation in detecting right whale calls in
long-term recordings from the Bering Sea. Canadian
Acoustics, 33(2), 25-34.
NAUTA. (n.d.). RASP: Programmable underwater acoustic
recorder. Retrieved 1 February 2013 from www.nauta-
rcs.it/english/page117/page25/page26.
Newcomb, J., Fisher, R., Field, R., Rayborn, G., Kuczaj, S.,
Ioup, G., . . . Turgut, A. (2002, October). Measurements
of ambient noise and sperm whale vocalizations in the
northern Gulf of Mexico using near bottom hydrophones.
Proceedings of MTS/IEEE Oceans ‘02, Biloxi, MS.
Newcomb, J., Tashmukhambetov, A. M., Ioup, G. E.,
Ioup, J. W., Sidorovskaia, N. A., Stephens, J. M., . . .
Summerfield, P. (2009, May). EARS buoy applications
by LADC: II. 3-D seismic airgun array characterization.
Proceedings of MTS/IEEE Oceans ‘09, Brennen,
Germany.
Nieukirk, S. L., Stafford, K. M., Mellinger, D. K., Dziak,
R. P., & Fox, C. G. (2004). Low frequency whale and
seismic airgun sounds recorded in the mid-Atlantic
Ocean. The Journal of the Acoustical Society of America,
115, 1832-1843. http://dx.doi.org/10.1121/1.1675816
Norman, R. G., & Greene, C. R., Jr. (2000). An autono-
mous acoustic recorder using a directional sensor for
locating calling bowhead whales. The Journal of the
Acoustical Society of America, 108, 2582. http://dx.doi.
org/10.1121/1.4743596
Nystuen, J. A. (1998). Temporal sampling requirements
for autonomous rain gauges. Journal of Atmospherical
and Oceanic Technology, 15, 1253-1260. http://dx.doi.
org/10.1175/1520-0426(1998)015<1253:TSRFAR>
2.0.CO;2
Nystuen, J. A. (2006). Marine mammals monitoring for NW
fisheries (NOAA NWFSC Final Report Award N00024-
02-D-6602, Task #0054). Seattle, WA: National Oceanic
and Atmospheric Administration.
Nystuen, J. A., Hanson, M. B., & Emmons, C. (2007, July).
Listening for killer whales in the coastal waters of the
NE Pacific Ocean. 3rd International Workshop on the
Detection and Classification of Marine Mammals using
Passive Acoustics, Boston, MA.
Ocean Instruments. (n.d.). Ocean bottom seismometers.
Retrieved 28 January 2013 from www.whoi.edu/
instruments/viewInstrument.do?id=10347.
Oleson, E. M., Wiggins, S. M., & Hildebrand, J. (2007a).
Temporal separation of blue whale call types on a southern
California feeding ground. Animal Behaviour, 74, 881-
894. http://dx.doi.org/10.1016/j.anbehav.2007.01.022
Oleson, E. M., Calambokidis, J., Burgess, W. C., McDonald,
M. A., Le Duc, C. A., & Hildebrand, J. A. (2007b).
Behavioral context of call production by eastern North
Pacific blue whales. Marine Ecology Progress Series,
330, 269-284. http://dx.doi.org/10.3354/meps330269
Oswald, J. N., Rankin, S., & Barlow, J. (2004). The effect of
recording and analysis bandwidth on acoustic identifica-
tion of delphinid species. The Journal of the Acoustical
Society of America, 116, 3178-3185. http://dx.doi.org/
10.1121/1.1804635
Parks, S. E., Clark, C. W., & Tyack, P. L. (2007). Short- and
long-term changes in right whale calling behavior: The
potential effects of noise on acoustic communication.
The Journal of the Acoustical Society of America, 122,
3725-3731. http://dx.doi.org/10.1121/1.2799904
Ponce, D., Thode, A. M., Guerra, M., Urbán, J. R., &
Swartz, S. (2012). Relationship between visual counts
and call detection rates of gray whales (Eschrichtius
robustus) in Laguna San Ignacio, Mexico. The Journal
of the Acoustical Society of America, 131, 2700-2713.
http://dx.doi.org/10.1121/1.3689851
Richardson, W. J., Greene, C. R., Jr., Malme, C. I., &
Thomson, D. H. (1995). Marine mammals and noise.
San Diego: Academic Press.
Fixed Autonomous Recorders 53
Samuels, A., & Tyack, P. L. (1999). Flukeprints: A history
of studying cetacean societies. In J. Mann, R. C. Connor,
P. L. Tyack, & H. Whitehead (Eds.), Cetacean societies:
Field studies of dolphins and whales (pp. 9-44). Chicago:
The University of Chicago Press.
Shariat-Panahi, S., Bermudez, A., Ambros, M., Manuel, A., &
Sallares, V. (2008). An Ocean Bottom Seismometer (OBS)
built for mid-term marine active refraction seismology.
Geophysical Research Abstracts, 10, EGU2008-A-09738.
SRef-ID: 1607-7962/gra/EGU2008-A-09738.
Shinke, T., Shimizu, H., Ichikawa, K., Arai, N., Matsuda,
A., & Akamatsu, T. (2004). Development of auto-
matic underwater sound monitoring system version 1.
Proceedings of the 2004 FY Annual Meeting of the
Marine Acoustics Society of Japan.
Simard, Y., Roy, N., & Gervaise, C. (2008). Passive acous-
tic detection and localization of whales: Effects of ship-
ping noise in Saguenay–St. Lawrence Marine Park. The
Journal of the Acoustical Society of America, 123, 4109-
4117. http://dx.doi.org/10.1121/1.2912453
Širović, A., Hildebrand, J. A., Wiggins, S. M., & Thiele,
D. (2009). Blue and fin whale acoustic presence around
Antarctica during 2003 and 2004. Marine Mammal
Science, 25, 125-136. http://dx.doi.org/10.1111/j.1748-
7692.2008.00239.x
Sousa-Lima, R. S., & Clark, C. W. (2008). Modelling the
effect of boat traffic on the fluctuation of humpback
whale singing activity in the Abrolhos National Marine
Park, Brazil. Canadian Acoustics, 36(1), 174-181.
Sousa-Lima, R. S., & Clark, C. W. (2009). Whale sound
recording technology as a tool for assessing the effects
of boat noise in a Brazilian Marine Park. Park Science,
26(1), 59-63.
Stafford, K. M., Nieukirk, S. L., & Fox, C. G. (1999).
Low frequency whale sounds recorded on hydrophones
moored in the eastern tropical Pacific. The Journal of the
Acoustical Society of America, 106, 3687-3698. http://
dx.doi.org/10.1121/1.428220
Stafford, K. M., Nieukirk, S. L., & Fox, C. G. (2001).
Geographical and seasonal variation of blue whale calls
in the North Pacific. Journal of Cetacean Research and
Management, 3, 65-76.
Stafford, K. M., Moore, S. E., Spillane, M., & Wiggins, S.
(2007). Gray whale calls recorded near Barrow, Alaska,
throughout the winter of 2003-04. Arctic, 60(2), 167-
172.
Thode, A. M., Gerstoft, P., Burgess, W. C., Sabra, K. G.,
Guerra, M., Stokes, M. D., . . . Cato, D. H. (2006). A
portable matched-field processing system using passive
acoustic time synchronization. IEEE Journal of Oceanic
Engineering, 31(3), 696-710. http://dx.doi.org/10.1109/
JOE.2006.880431
Tsutsumi, C., Ichikawa, K., & Arai, N. (2006). Feeding
behavior of wild dugongs monitored by a passive
acoustical method. The Journal of the Acoustical
Society of America, 123, 1356-1360. http://dx.doi.
org/10.1121/1.2221529
Tyack, P. L., Johnson, M. P., Aguilar Soto, N., Sturlese, A.,
& Madsen, P. T. (2006). Extreme diving of beaked
whales. Journal of Experimental Biology, 209, 4238-
4253. http://dx.doi.org/10.1242/jeb.02505
Van Parijs, S. M., Clark, C. W., Sousa-Lima, R. S., Parks,
S., Rankin, S., Risch, D., & van Opzeeland, I. C. (2009).
Management and research applications of real time and
archival passive acoustic sensors over varying temporal
and spatial scales. Marine Ecology Progress Series, 395,
21-36. http://dx.doi.org/10.3354/meps08123
Wang, K., Wang, D., Akamatsu, T., Li, S., & Xiao, J. (2005).
A passive acoustic monitoring method applied to obser-
vation and group size estimation of finless porpoises.
The Journal of the Acoustical Society of America, 118,
1180-1185. http://dx.doi.org/10.1121/1.1945487
Watkins, H., & Colley, R. (2004). Harbour porpoise Phocoena
phocoena occurrence Carmarthen Bay – Gower Peninsula
– Swansea Bay (December 2002–February 2004 Report).
South Wales: Gower Marine Mammals Project.
Wiggins, S. (2003). Autonomous acoustic recording pack-
ages (ARPs) for long-term monitoring of whale sounds.
Marine Technology Society Journal, 37(2), 13-22. http://
dx.doi.org/10.4031/002533203787537375
Wiggins, S. M., & Hildebrand, J. A. (2007). High-frequency
Acoustic Recording Package (HARP) for broad-band,
long-term marine mammal monitoring. Symposium on
Underwater Technology and Workshop on Scientific Use
of Submarine Cables and Related Technologies, Tokyo,
Japan. http://dx.doi.org/10.1109/UT.2007.370760
Wiggins, S., Manley, J., Brager, E., & Woolhiser, B. (2010).
Monitoring marine mammal acoustics using Wave
Glider. Oceans 2010 Conference, Seattle, WA. http://
dx.doi.org/10.1109/OCEANS.2010.5664537
Wildlife Acoustics, Inc. (2012). Marine monitoring.
Retrieved 28 January 2013 from www.wildlifeacoustics.
com/products/marine-monitoring.
