Applications of Passive Acoustics to Fisheries Science Scott Holt Listens to Red Drum Sounds from a Fish Pier in Texas

SEPTEMBER 2006 VOL 31 NO 9

Fish News Legislative Update Fisheries Journal Highlights American Fisheries Society • www.fisheries.org Calendar Job Center

LISTENING TO FISH: APPLICATIONS OF PASSIVE ACOUSTICS EVALUATING HUMAN AND BIOLOGICAL OBJECTIVES OF COOPERATIVE RESEARCH Cover: Passive acoustics can be an excellent tool for public education. Here children listen to underwater sounds as they learn about the estuarine environment.

Credit: North Falmouth Congregational Church FEATURE: SCOTT HOLT, UNIVERSITY OF TEXAS AT AUSTIN FISHERIES RESEARCH

Listening to Fish: Applications of Passive Acoustics to Fisheries Science Scott Holt listens to red drum sounds from a fish pier in Texas. Rodney A. Rountree Passive acoustics is a rapidly emerging field of marine biology that until recently R. Grant Gilmore has received little attention from fisheries scientists and managers. In its sim- Clifford A. Goudey plest form, it is the act of listening to the sounds made by fishes and using that information as an aid in locating fish so that their habitat requirements and Anthony D. Hawkins behaviors can be studied. We believe that with the advent of new acoustic tech- Joseph J. Luczkovich nologies, passive acoustics will become one of the most important and exciting David A. Mann areas of fisheries research in the next decade. However, a widespread lack of Rountree is a senior scientist at Marine familiarity with the technology, methodologies, and potential of passive acous- Ecology and Technology Applications, Inc., tics has hampered the growth of the field and limited funding opportunities. Waquoit, MA. He can be reached at Herein, we provide an overview of important new developments in passive [email protected]. Gilmore is a acoustics together with a summary of research, hardware, and software needs to senior scientist, Estuarine, Coastal, and advance the field. Ocean Science, Inc., Vero Beach, FL. Goudey is director, Center for Fisheries Engineering Research, MIT Sea Grant INTRODUCTION College Program, Cambridge, MA. finding the fish in the first place. Often Hawkins is director, Loughine Ltd Listen to fish? Although most fish- a scientist must go to great lengths con- Aberdeen, Scotland. Luczkovich is an eries biologists are generally aware that ducting expensive and time-consuming associate professor at the Institute for some fishes are soniferous (sound pro- biological surveys simply to determine Coastal and Marine Resources, Department ducing), relatively few have stopped to the locations or habitats where a fish of Biology, East Carolina University, consider the potential importance of lis- can be found, before any attempt to Greenville, NC. Mann is an assistant tening to fish sounds to their fields of study its biotic and abiotic interactions professor, University of South Florida, study. In fact, few realize that many can be made. After all, you can't study College of Marine Science, St. Petersburg. important recreational and commercial something you can't find. Any tool that fishery species are highly soniferous, can help scientists to locate fish is there- including many in the cod and drum fore valuable. Joe Luczkovich monitors fish sounds in the families. Fishes produce sounds to com- A new field of passive acoustics is laboratory at East Carolina University. municate with one another while they rapidly emerging in fisheries science and are feeding, mating, or being aggressive marine biology, in which scientists use and also make incidental noises associ- underwater technology to listen in on ated with feeding, swimming, and other the noisy aquatic realm. Passive acous- behaviors (Fine et al. 1977b). Far from tics is distinguished from other types of Jacques Cousteau's The Silent World bioacoustics, because it uses naturally (Cousteau and Dumas 1953), our estuar- occurring sounds to gather information

ies, seas, and oceans are teaming with on fishes and other marine organisms, JOE LUCZKOVICH,, EAST CAROLINA UNIVERSITY the sounds of marine life. rather than using artificially generated Fish are difficult to see and study in sounds. Passive acoustics provides a the ocean. Scuba techniques can help in number of important benefits for fish- shallow waters and a range of active eries research. First, it provides a acoustic and optical techniques can method of non-optically observing fish assist in deep water, but we are still activity and distribution. That is, it can largely ignorant of the distribution and be used to find and monitor fishes (and behavior of the great majority of marine other animals) that produce sound. fish. Possibly one of the greatest chal- Second, passive acoustics is a non-inva- lenges to researchers attempting to study sive and non-destructive observational the behavioral ecology of fishes is that of tool. Third, it provides the capability of

Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 433 continuous or long-term monitoring as Figure 1. Representative calls of sciaenid fishes. Energy distribution patterns associated with well as remote monitoring. Such long- harmonic frequency bands and fundamental frequencies are species specific and can be used to term monitoring can provide important identify species presence based on the call characteristics. information on daily and seasonal activity patterns of fishes and other marine organisms. In addition, passive acoustics can also be used to simultaneously monitor sources of noise pollution, and to study the impact of man's activities on marine communities. Anthropogenic sources include noise generated by boat- ing activity, seismic surveys, sonars, fish finders, depth finders, drilling for oil and gas, and military activities. As recently reviewed by Popper (2003) these anthropogenic noise sources all have potentially important impacts on marine fauna. The ability to listen to fish and other marine life allows scientists to identify, record, and study underwater animals even in the absence of visual information. Coupling passive acoustics with conventional fishery sampling techniques provides a powerful new approach to fisheries research. We seek to promote a better understanding of passive acoustics applications to fisheries among fisheries professionals, and feel that passive acoustics will become a major area of fisheries research in the next decade. Towards this purpose, we provide, herein, an overview of some important developments in passive acoustics applications to fisheries and summarize areas of research and development that we believe are needed to catalyze advancement in the field. Figure 2. Figure 4. Spatial distribution of spawning Representative aggregations of sciaenid fishes as BACKGROUND sciaenid internal GRANT GILMORE, ECOUS determined by soniferous activity in a anatomy revealing the Florida estuary. Spawning locations of Over 800 species of fishes from 109 gas bladder and sonic different species are represented by ellipses families worldwide are known to be muscles of a male with different shading patterns. soniferous (Kaatz 2002), though this is weakfish, Cynoscion likely to be a great underestimate. Of regalis. these, over 150 species are found in the northwest Atlantic (Fish and Mowbray Figure 3. Temporal overlap pattern of soniferous 1970). Amongst the soniferous fishes are activity among sciaenid fishes in a Florida estuary. some of the most abundant and important commercial fish species, including many codfishes, drum fishes, grunts, groupers, snappers, jacks, and catfishes. Some invertebrates with important fisheries also produce sounds, including mussels (Mytilus edulis), sea urchins (Fish 1964), white shrimp (Penaeus setiferus, Berk 1998), spiny lobsters (Moulton 1957; Fish 1964; Patek 2002), American lobster (Homarus americanus, Fish 1966; Henninger and Watson

434 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 2005), and perhaps squid (Iversen et al. able to link the aggressive and spawning conducted by Takemura et al. (1978), 1963). behaviors of some fisheries species to Mok and Gilmore (1983) and Qi et al. Passive acoustics has been used for their sound production using a combina- (1984). A portable hydrophone and over 50 years in fish biology and fish- tion of in situ and tank studies. For recording system was carried via a boat eries surveys (see Fish et al. 1952 and example sounds produced by haddock from one site to another along a mea- Fish and Mowbray 1970 for a summary (Melanogrammus aeglefinus) during sured transect with recordings made of early work) and is being used rou- courtship and mating have been along a preset grid or in a linear series tinely today to determine habitat use, recorded and analyzed in this manner (Mok and Gilmore 1983; Gilmore delineate and monitor spawning areas, (e.g., Hawkins 1986). Once the associa- 2002). Recordings were made for and study the behavior of fishes tion of sounds to specific species and 30–300 seconds at each site depending (Hawkins 1986; Rountree et al. 2003a, behaviors has been established, passive on transect length. Recorded sounds 2003b). Using hydrophones, marine acoustics provides a rapid way of estab- were verified by auditioning of captured ecologists and fishery biologists have lishing the spawning component of specimens. This technique allowed spa- been able to listen to the sounds fishes essential fish habitat (EFH). tial and temporal isolation and produce and identify species specific identification of species-specific sounds EXAMPLES OF FISHERIES (Hawkins and Rasmussen 1978; Myrberg produced by sciaenid fishes, particularly APPLICATIONS and Riggio 1985; Lobel 1998), and even under conditions of high sound attenua- individual specific (Wood et al. 2002), Passive acoustics studies using relation for large group sounds (low signatures using signal processing and tively simple techniques have been frequency, high intensity sounds). Using spectral analysis computer algorithms. successful in locating concentrations of detailed sonographic analyses of field Often these sounds dominate the acous- important fish species, opening the way recordings made on transects, Mok and tic environment where they occur, as in for further, more detailed studies of their Gilmore (1983) described the character- the drum family (Sciaenidae), so much behavior, distribution, and habitat use. istic sounds of black drum (Pogonias so that they interfere with military and Below we provide a brief review of past cromis), spotted sea trout (Cynoscion petroleum prospecting operations that and current research on two groups of nebulosus), and silver perch (Bairdiella involve acoustic monitoring (e.g., Fish soniferous fishes that support large fish- chrysoura, Figure 1). Subsequent to and Mowbray 1970). In other situations, eries, the drumfishes (sciaenids) and these observations, considerable addi- such as damselfishes on coral reefs, the codfishes (gadids). tional work has been done on sound sounds are not loud and require special- characterization in these species, as well ized techniques to detect them (Mann Drum fishes (Sciaenids) as the weakfish (C. regalis) and the red and Lobel 1995a,b). History—Sciaenid fishes have been drum (Sciaenops ocellata). Passive acous- To identify the species of fish produc- known to produce sound for centuries tic transect techniques have been used ing a sound, one first must do “sound (Aristotle 1910; Dufossé 1874a, 1874b) by several investigators to locate spawn- truthing.” There are two main ways this and the association of sciaenid sounds ing sciaenid groups in the field (Saucier has been accomplished: (1) captive fish with spawning has been known nearly as and Baltz 1992, 1993; Connaughton and recordings and (2) in situ (i.e., in the long (Darwin 1874; Goode 1887). For Taylor 1994, 1995; Luczkovich et al. field under natural conditions) record- hundreds of years the Chinese have 1999, 2000). ings. Although acoustic complications located sciaenid spawning sites from Function of Sound Production in in a tank or aquarium, combined with their water craft by listening to drum- Sciaenids—The most predictable and unnatural behavior and sound produc- ming sounds emanating from the water robust sounds produced by many fishes tion, make captive fish recordings through the hull of their boats (Han are those associated with reproduction. problematic, especially for larger fishes, Ling Wu, Shanghai Fisheries Institute, As in many soniferous animals, it is the such problems can be overcome pers. comm.). The location of sciaenid male that must attract a mate and (Okumura et al. 2002). Field recordings, spawning sites using underwater tech- induce her to donate eggs for fertiliza- on the other hand, sometimes suffer nology is recent and dependent on the tion, and, therefore, it is often only the from the difficulty of matching sounds availability of underwater transducers, male that produces sound. Large choral to species and behaviors. Field methods hydrophones, and acoustic recorders aggregations of male sciaenid species are have been particularly successful in used to access and study underwater formed by spotted seatrout, weakfish, tropical waters where environmental sounds (Fish and Mowbray 1970). red drum, and silver perch specifically to conditions allow the use of advanced Hydrophone tape recordings of sounds attract females with which to spawn. scuba and underwater video technolo- produced by large sciaenid aggregations Sciaenid Sound Production gies (Lobel 2001, 2002). Knowledge of during spawning were pioneered by Mechanisms—The most robust and sound source levels is important for cal- Dobrin (1947), Dijkgraaf (1947, 1949), energetic sciaenid sounds are produced culating the detection limits of Knudsen et al. (1948), Protasov and by sonic muscles indirectly or directly hydrophones (e.g., Sprague and Aronov (1960), Tavolga (1960, 1980), vibrating the membrane of the gas blad- Luczkovich 2004). Precise measurement and Fish and Mowbray (1970). der. When a freshly captured, recently of sound source requires knowledge of The location and description of calling, male seatrout is dissected, the the location of the fish and of the soniferous sciaenid aggregations using bright red sonic muscles surrounding the hydrophones’ characteristics. Despite mobile hydrophones moving along a gas bladder can be easily differentiated these difficulties, biologists have been sound transect at spawning sites was from the exterior lateral body muscles

Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 435 (Figure 2). The muscle vibratory rate is seatrout were collected with plankton cycle has been observed in most species directly associated with the fundamental nets at spawning sites during periods of studied to date (Hawkins and frequency of the characteristic seatrout sound production (Mok and Gilmore Rasmussen 1978). Although Atlantic call produced by the gas bladder. 1983; Alshuth and Gilmore 1993, 1994, cod (Gadus morhua) have a limited Most of the 270 species in this family 1995). The seasonal mating calls were sound production repertoire, haddock probably produce sound using sonic directly associated with primary spawn- have been shown to exhibit different muscles associated with the gas bladder. ing activity in east-central Florida calls at different stages of courtship Using the species specific muscle con- sciaenids. Soniferous activity of various (Figure 6; Hawkins et al. 1967; Hawkins traction rates and the gas bladder shape, sciaenid species exhibited distinct sea- and Rasmussen 1978; Hawkins 1986). sciaenids produce sounds that can be sonal patterns, based on an analysis of Both species have specialized sonic mus- used to identify species within the fam- over 300 acoustic transects collected cles that vibrate the swim bladder to ily (Mok and Gilmore 1983), as has between 1978–2002 (Figure 3). These produce sounds for communication been demonstrated in amphibians and same studies of soniferous spawning (Hawkins 1986). The sonic muscle is birds. The characteristic shape of the aggregations have demonstrated long sexually dimorphic in haddock and sciaenid gas bladder is so conservative term stability of spawning site locations Atlantic cod, with males having signifi- that it has been used as one of the pri- (Figure 4), with the principal spawning cantly larger muscles than females mary morphological characters to sites identified by Mok and Gilmore (Templeman and Hodder 1958; classify sciaenids and to determine their (1983) being used for over 20 years Templeman et al. 1978; Rowe and phyletic relationships (Chu 1963; Chao (Gilmore 2002). Hutchings 2004). In addition, the sonic 1978). muscle undergoes seasonal maturation Codfishes (gadids) When and Where Do Sciaenids in concert with the gonad maturation cycle. Evidence for sexual selection of Produce Sound?—Mok and Gilmore A number of gadid species are sonif- sonic muscle size has been reported (1983) demonstrated that sciaenid erous and the trait is likely widespread (Engen and Folstad 1999; Nordeide and sound production was specifically associ- within the family (Figure 5; Hawkins Folstad 2000; Rowe and Hutchings ated with crepuscular and nocturnal and Rasmussen 1978; Almada et al. 2004). courtship and spawning activities. 1996). A strong correlation between Brawn (1961a, b, c) provided the best Pelagic eggs and larvae of the spotted soniferous activity and the spawning description to date of the role of sound communication in courtship and spawn- Figure 5. Representative calls of gadid fishes. Each species can be identified by its unique pulse ing behavior in Atlantic cod based on pattern. For scale, the horizontal bar under each graph represents 100 ms. (Redrawn from laboratory studies. She found that sonif- Hawkins 1986). erous activity is most common during the spawning season, being rare at other times, except for a fall “aggression period.” She hypothesized that the fall soniferous period was related to increased aggressive interactions prior to dispersal to the foraging grounds. In the spawning season soniferous behavior is strongly associated with spawning and both spawning activity and call frequency peaked during the early evening hours. Interestingly, soniferous activity is most frequent at night during the spawning season, but most frequent during the day during the fall “aggression period.” Brawn (1961a, b, c) attributed this to nocturnal spawning in the win- ter, and diurnal feeding interactions during the fall. Unfortunately, Brawn did not provide us with a detailed statistical description of Atlantic cod sound characteristics. However, later studies indicate that Atlantic cod produce sounds predominantly in the frequency range of 80–500 Hz (Fish and Mowbray 1970; Hawkins and Rasmussen 1978; Finstad and Nordeide 2004). Nordeide and Kjellsby (1999) recently recorded sounds of Atlantic cod from the spawn-

436 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG ing grounds off of Norway, and suggested Hawkins 2003). In an Arctic fjord in European fisheries (Svellingen et al. that passive acoustics could be used to northern Norway, workers from the FRS 2002). study spawning in the field. Marine Laboratory, Aberdeen and the A PRIMER ON TECHNIQUES The soniferous behavior of haddock University of Tromsø have located a is somewhat better described than that spawning ground of haddock. Passive lis- The success and development of pas- of Atlantic cod (Figure 6; Hawkins et al. tening has revealed that this species, sive acoustics applications to fisheries 1967; Hawkins and Rasmussen 1978; previously thought to spawn offshore in depends on high quality recording sys- Hawkins 1986). They have a similar fre- deep water, can also form large spawning tems and analysis software. The quency range, but can be distinguished concentrations close to shore (Hawkins technology should be matched to the by differences in pulse characteristics et al. 2002; Hawkins 2003). questions being asked. For most ques- (Figure 5; Hawkins and Rasmussen Norwegian researchers at the tions, the needed technology exists for advancing the field and the main imped- 1978). Hawkins and his colleagues are Institute of Marine Research have pio- iment is insufficient knowledge on the currently conducting studies aimed at neered the use of remote controlled use of technology. using passive acoustics as a tool to iden- platforms to obtain video and audio data tify spawning habitat of haddock in on the spawning behavior of Atlantic Hydrophones European waters (Hawkins et al. 2002; cod and other gadids important in Hydrophones are the most basic ele- ment of any recording system. They are Figure 6. Different stages of the courtship behavior of haddock, Melanogrammus aeglefinus, can underwater microphones that typically be distinguished by differences in the pulse repetition rate of the call. The time-base, scale bar, convert sound pressure into an electrical for the calls is 50 ms. (Redrawn from Hawkins 1986). signal that can be recorded by a data acquisition system. Many commercially available hydrophones can be used for fish bioacoustics studies. When choos- ing a hydrophone, researchers must consider its sensitivity and frequency response characteristics to ensure it is well suited to their needs. Since most fish sounds range in frequency from 20 Hz for the largest fishes to 4 kHz for the smallest fishes, the hydrophone should cover this frequency range. The hydrophone sensitivity should be such that a loud fish sound produces a signal of about 1 V, and thus, a sensitivity range of around -160 to -170 dBV/␮Pa at 1 m (decibel volts/micro pascal) is good for most fisheries applications. Data storage Data storage devices include analog and digital tape recorders, audio video recorders (Lobel 2003), and computers with sound cards. Digital systems provide obvious advantages over analog systems in terms of greater frequency bandwidth and dynamic range, and will be the most commonly used systems in the future. Which system is chosen will depend on the recording situation. Computer systems that may be practical for recordings made in the laboratory may not be practical in a field situation, because of power, portability, and environmental issues. One important development is the use of remotely operated vehicles (ROVs), telemetry, and underwater listening stations for monitoring sound producing fishes (Mann and Lobel 1995b; Sprague and

Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 437 Luczkovich 2004; Locascio and Mann grammed to record as desired (Mann 2005). These systems will be important Examples of data logger use in passive and Lobel 1995b; Calupca et al. 2000; in characterizing which species produce acoustics. Sakas et al. 2005). No data loggers can which sounds, especially for species that TOP: Autonomous underwater listening be purchased and used directly without are difficult to maintain in a laboratory system (AULS) designed by Cliff either engineering or software program- tank. They will also prove useful for doc- Goudey. ming (usually both). This lack of an off umenting behavior of fish aggregations -the-shelf product has greatly limited MIDDLE: Commercial fishermen John and when multiple individuals call simulta- Moe Montgomery of the F/V Chandelle their use in passive acoustic applications neously. deploying an AULS on Jefferies Ledge to fisheries. One needs to be aware of several in the Gulf of Maine as part of a Telemetry caveats of recording systems. cooperative study of haddock spawning. Telemetry systems broadcast a 1. Data compression: Some recorders hydrophone signal to a boat or shore (such as mini disc and MP3) use data BOTTOM: AULS deployed in the MacGregor Point coral reef in Maui as based receiver. They perform the same compression techniques that alter the part of the Sanctuary Sounds project. function as data loggers in allowing recorded sound frequency and level. recordings over a large area for long These would not be appropriate periods of time. Several types of teleme-

when detailed descriptions of sound ENGINEERING RESEARCH, try systems are available including characteristics are required, but could CENTER FOR FISHERIES sonobuoys (VHF), cell phone systems,

be useful for ecological monitoring of MIT SEA GRANT and short range microwave systems. All temporal and spatial patterns of well of these systems require line of sight known sounds. between the transmitter and receiver, 2. Automatic Gain Control (AGC): and a relatively high level of engineer- Many systems (especially many ana- KATIE ANDERSON, UNIVERSITY ing to set up and maintain. Telemetry is log and digital tape recorders and also capable of delivering video to docu- video cameras) use AGC to keep the OF MASSACHUSETTS AT ment behavior during sound production recorded volume within the same (Svellingen et al. 2002). Satellite sys- range. If a system uses AGC, it will tems generally do not support the not be possible to determine the AMHERST bandwidth needed for transmitting received sound level. acoustic data. At this point, some 3. Bit resolution: Systems that record amount of preprocessing would be with a higher bit resolution will have NORM GARON, HAWAII ISLAND HUMPBACK required, so that limited data on sound a larger dynamic range (the range of WHALE NATIONAL MARINE SANCTUARY characteristics (e.g., root-mean-square the quietest and loudest sounds that amplitude, frequency spectra) could be can be recorded). transmitted. One problem that many researchers Hydrophone Arrays have encountered is boat induced noise on recording systems, either through Fixed and towed hydrophone arrays electrical noise on the boat, or the phys- have been used for determining the ical movement of the boat causing the locations of vocalizing whales in many hydrophone to move. Bungee cords have different situations (e.g., Watkins and been successfully used to decouple boat Schevill 1972; Clark 1980), but have movement and hydrophone movement, only recently been applied to fishes and acceleration canceling hydrophones (Barimo and Fine 1998; Mann and are commercially available. Jarvis 2004). Hydrophone arrays hold software (see below) exists for recording promise in answering questions about Data loggers on a given duty cycle. However, contin- fish distributions that could not be oth- Acoustic data loggers are useful for uous power is rarely available in field erwise obtained with single hydrophone recording over long periods of time in situations where one would like to make recordings (D'Spain et al. 1996), but many locations simultaneously. Data recordings. In these situations, low require a high level of sophistication for loggers provide a way to gather informa- power battery operated acoustic data setting up, operating, and analyzing the tion on the temporal distributions of loggers are required. data. The minimum spacing between sound producing fish that would not be To date all acoustic data loggers that hydrophones in an array needed for possible otherwise without considerable have been used for passive acoustics localization is calculated as the speed of investment of human resources. A good have been engineered by individual lab- sound in water (approximately 1,500 example of this are the pop-up recorders oratories to perform this task. These m/s) divided by the sound frequency. For (Calupca et al. 2000). Computers are include analog tape recorders that have example, to locate a fish calling at 1,000 the best option for recording where con- been modified to record on a particular Hz (e.g., silver perch), a minimum spac- tinuous power is available, such as from duty cycle (Luczkovich et al. 2000), and ing of 1.5 m is required, while a spacing shore or on a large boat. Commercial digital dataloggers that have been pro-

438 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG of 12 m is required to locate a fish call- conventional means (e.g., Woods Hole, ing at 125 Hz (e.g., red drum). Data loggers have been used to study Massachusetts), or that are in the very deep sea fishes in cooperation with heart of the industrial world (the docks Signal Processing Software commercial red crab fishermen. on Manhattan Island, New York; Many packages are available for data TOP: Cliff Goudey demonstrates data Anderson et al. unpublished). How can acquisition and signal processing. Some logger to Daher Jorge of the F/V it be that we know so little about fishes Krystal James. such as MATLAB (Mathworks, Inc.) in these areas that we can not even have a great deal of flexibility and BOTTOM: Juan Espana aboard the F/V identify some of the most frequent and power, but require a high level of pro- Hanna Boden prepares to deploy a widespread fish sounds? As biologists, we data logger attached to the inside of a find the lure of the unknown so close to graming knowledge. Others are targeted red crab pot. specifically at bioacousticians including the doorstep to be a powerful motivator. Although some progress is being made Signal (Engineering Design), Raven in this area, the most pressing needs are: (and its precursor Canary; Cornell University), SASLab Pro (Avisoft 1. an effort to catalogue historic records Bioacoustics), and Adobe Audition (for- of known and unknown sounds; merly Cool Edit). The manuals to these 2. systematic efforts to identify and val- software programs are often the clearest idate sound sources; source of information for learning signal RODNEY ROUNTREE 3. studies to determine the correlation processing techniques and their applica- between sound production and spe- tion. cific behaviors; and 4. studies to determine under what FUNDING PRIORITIES AND behavioral conditions fish/inverte- RESEARCH NEEDS brates make sounds. Research presented at an interna- Efforts are now underway to digitize tional workshop on the “Applications of and catalogue sound archives from labo- Passive Acoustics to Fisheries” in April ratories across North America and in 2002 underscored the great strides that Europe (Rountree et al. 2002; Bradbury have been made in the application of and Bloomgarden 2003). The MaCaulay passive acoustics to fisheries and related Library of the Cornell Lab of issues over the last two decades Research Needs Ornithology has recently made a large (Rountree et al. 2003a, b). The work- collection of fish sounds from these Research needs can be simplistically archives available to the public online shop was followed by a special summarized as the questions: what, (www.birds.cornell.edu/MacaulayLibrary/). symposium on passive acoustics applica- when, where, and how many? Unfortunately, many of these historic tions to fisheries held at the American What?—What fishes and inverte- recordings are poorly documented and Fisheries Society Annual Meeting in brates are making sounds? To date we were made with long outdated technolo- Quebec in August 2003 and organized know of approximately 800 species of gies. More importantly, they are by Luczkovich, Mann, and Rountree fishes worldwide that produce sounds. that again highlighted the rapid More soniferous species are reported on insufficient for the needs of fisheries advances in the field (Luczkovich et al. a frequent basis, but there are no current researchers, as they are only partially complete. Many of our most important unpublished). And most significantly, systematic efforts to catalogue soniferous commercial fishes that are soniferous important new initiatives in passive fishes. The landmark studies by Marie have yet to be recorded. Information on acoustics have begun in many areas of Fish and William Mowbray (1970) soniferous freshwater fisheries species is the United States. Although passive ended over 30 years ago. No one has almost entirely lacking in North acoustics is currently underutilized as a taken up the gauntlet since then. America. research tool, the success of the work- Knowledge of what produces underwater To expand the catalogue of fish and shop and symposium demonstrates how sounds is critical to the success of higher level studies. Currently the number of invertebrate sounds, field and laboratory rapidly the field is emerging and suggests unidentified underwater sounds studies are needed to identify unknown great promise for future research. attributed to fishes is far greater than sounds, and audition fishes and inverte- As a result of discussions and corre- those that can be positively identified. brates for sound production. Systematic spondences initiated at the workshop Rountree and Goudey (unpublished) efforts to identify sounds in each geo- and symposium, we have prepared a recently recorded unknown sounds on graphic region, including estuaries and comprehensive description of areas that the commercial fishing grounds off the tidal fresh waters, are critical to the participating scientists felt needed to be coast of Massachusetts. Even more advancement of passive acoustics. addressed to advance the field of passive remarkably, Rountree and Juanes Evidence that a particular sound is made acoustics in fisheries. These areas can be (unpublished) and their students by, and unique to, a given species is of categorized as research, software, hard- recorded many unknown sounds in areas critical importance. Lack of, or per- ware, and education/outreach needs. that have been extensively studied by ceived lack of, such evidence is the most

Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 439 common criticism encountered in fund- tion them for sound production. We temporal and spatial distribution pat- ing proposals. liken the denial of the usefulness of cat- terns of fishes by sex and maturity There are two major ways to go about alogues of unidentified fish sounds to stages. They can also provide data cataloguing fish sounds in a particular the absurd notion that collections of useful in studies of reproductive ecol- region. The first, is through the system- ichthyoplankton for which identifica- ogy and studies specifically important atic auditioning of animals collected in tions are not currently available are not for fisheries assessment, such as the the field, in aquaculture facilities, and in useful. quantification of fish fecundity. public aquaria. Fish and Mowbray Incidental sounds made by fishes can 3. Quantification of daily and seasonal (1970) used this method to study the be just as important as soniferous behav- patterns in sound production (Breder sounds of over 200 species of fishes. One ior. For example, feeding sounds can be 1968; Fine et al. 1977a; major problem with such efforts is that used to determine foraging times, loca- Connaughton and Taylor 1994). If failure to record fish sounds during audi- tions, and consumption rates (Sartori we know when a fish produces sounds tions does not indicate the species is not and Bright 1973; Mallekh et al. 2003; then studies of the daily and seasonal soniferous, because soniferous behavior Anderson et al. unpublished). patterns of that soniferous activity is often controlled by sex, age, matura- Often investigators have used terms can be correlated with daily and sea- tion stage, and behavior. Sometimes fish such as “grunts,” “knocks,” “snaps,” sonal patterns of that specific are not physiologically ready for sound “pops,” “staccato,” “drumming,” “hum- behavior. For example, if it is known production outside of the spawning sea- ming,” “rumbles,” “percolating,” that a particular sound is only pro- son. Additionally, auditioning programs “purring,” etc., to describe the sounds duced when a fish is spawning, then like that carried out by Fish and heard; the names often being ono- the determination of the daily pat- Mowbray (1970) are sometime criticized matopoeic. Standardizing these sound tern in soniferous activity can be as producing artificial sounds through descriptions would allow rapid commu- used to infer the daily pattern of electrical stimulation; however, we note nication between biologists and other spawning (i.e., to determine what that at least 47 of 150 species regarded observers (Anderson et al. unpublished). time of day the fish spawns). as soniferous by Fish and Mowbray When?—Temporal patterns in sound Similarly, seasonal patterns in sonif- (1970) produced sounds spontaneously. occurrence are needed for two main rea- erous activity can provide an index of The second major method of cata- sons. First, knowledge of when fishes the spawning season period (Figure loguing fish sounds is to conduct field and invertebrates make sounds leads to 3). surveys to identify temporal and spatial knowledge about their biology and ecol- Where?—Identifying where a fish patterns of sound production. Once ogy. Second, we need this information occurs is one of the most fundamentally unknown sounds are identified and their to effectively use passive acoustics as a important topics in fish ecology and temporal and spatial patterns described, tool to locate fish, to identify their habi- fisheries management. It is the first step research can be conducted to determine tat requirements, and to conduct towards identifying essential fish habitat the identity of the fish, including the presence/absence and abundance sur- (EFH) as mandated of fisheries managers auditioning of fish captured in the veys (see below). Some of the most by the reauthorization of the Magnuson appropriate place and time. The advan- important needs here are: tage of this approach is that the field of Stevens Fishery Conservation and potential sound producers is narrowed 1. Studies to determine the relationship of Management Act (Oct. 1996). Essential down, and the researcher knows that the sound production to fish size. Early Fish Habitats are defined as “those target species is at least physically capa- studies have shown that sound char- waters and substrate necessary for fish ble of sound production at the time of acteristics sometimes change with for spawning, feeding or growth to matu- auditioning. However, efforts by fish size (e.g., dominant frequency) rity.” The lowest level criterion for researchers to catalogue underwater and hence the relationship between identification of EFH is simply presence/ sounds in freshwater and marine habi- fish size and sound characteristics absence. At the minimum, passive tats are hampered by the failure of may be used to determine length fre- acoustics surveys can be used to identify funding agencies to recognize the impor- quency data for soniferous fishes the presence of soniferous fishes in habi- tance of such studies. Data on (Fish and Mowbray 1970; Lobel and tats. If the behavior associated with a unidentified fish sounds are especially Mann 1995; Connaughton et al. particular fish sound is known, then difficult to publish, even when extensive 2000). higher levels of identification, such as observations are available. We suggest 2. Studies to determine the relationship of the identification of spawning habitat, that such studies are valuable in their sound production to sex and maturity can be achieved. own right as they provide the basic stage (e.g., Connaughton and Taylor To accurately associate underwater information necessary to systematically 1995). Do both male and females in a sounds with habitats, it is necessary to survey soniferous fishes by providing population produce sounds? Are the know the range at which the sound can data on when and where soniferous sounds the same or different? Do be detected by the survey instruments. activity occurs. Armed with such data, immature fish make sounds? Are the A lack of quantification of sound source scientists can devise studies to deter- sounds made by immature and mature detection ranges is the second most mine the identity of the sound individuals different? The answers to commonly cited criticism encountered producers. The only alternative is to these questions can provide scientists in funding proposals. To determine simply blindly capture fishes and audi- with valuable information on the sound source detection ranges several

440 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG types of studies are needed. (a) How In the simplest situation, the number multiple sound sources and high lev- loud is it?—Quantification of sound of fish calls can be counted and corre- els of “noise” is essential. source levels for a given lated to the true amount of fish present. 2. Temporal pattern analysis software to species/stage/environment is the first This requires the ability to distinguish track the temporal occurrence pat- step. (b) Sound Propagation—how far individual calls. More rigorously it terns of hundreds to thousands of does sound travel under a given set of would also require the ability to distin- detected calls within long time series environmental conditions. This is the guish individuals that call repeatedly is critical to the efficient determina- study of tomography. Studies in shallow from those that don't. Often, however, tion of daily and seasonal patterns in water are especially needed. (c) Fish fish and calls are so numerous that indi- sound production. Shoals-studies are needed to examine vidual calls cannot be distinguished. In 3. Call characterization software to how the multiple sound production of these cases, studies are needed to deter- speed up the process of measuring call many fish in a shoal combine. How do mine the relationship between sound characteristics such as duration, pulse sound pressure levels vary with shoal size pressure level and fish abundance. For rate, pulse repetition, fundamental and distance from the source? example, Luczkovich et al. (1999) frequency, pulse width, sound level, Passive acoustic mapping of the spa- showed a relationship between sound etc., in large numbers of calls would tial distribution of sounds is one of the pressure level and egg abundance, and increase statistical power in studies of most important potential products of indicated that a relationship should location, behavior, and environmen- passive acoustics technology. Studies to exist between sound production and tal effects on call characteristics. map the distribution of fish sounds, and spawning stock size. 4. Call classification software to aid in hence the distribution of soniferous As mentioned above, soniferous the identification of a new unknown fishes and their essential fish habitat are behavior in many fishes is sex depen- sound through comparison of its call strongly needed. Passive acoustics offers dent. Often, only the males call. In characteristics with the characteris- many advantages over traditional meth- addition, in some cases soniferous activ- tics of calls contained in standard reference libraries of known and ods of mapping fish distribution. It is ity may be related to size and maturation unknown sounds. non-destructive, non invasive, and rela- stage, so that at a given place and time, tively inexpensive over the long run the number of fish calling is a function Localization of sound sources— compared to traditional methods such as of the size and maturation stage distribu- Software and hardware developments trawl surveys. If the “what” and “when” tion of the population. However, are needed to simplify the localization of are known, then passive acoustics can be individual fishes fail to produce sounds sound sources received by multiple used to map EFH based on in a given situation for many reasons. hydrophone arrays. Automatic signal presence/absence for soniferous species. Studies are needed to allow researchers detection and classification software are The “when” is important because spatial to estimate the proportion of soniferous important prerequisites. Currently local- surveys with passive acoustics are only and non soniferous fishes over a given ization is a laborious process of valid if conducted during the window of time period. determining the starting points of the time when fish are soniferous. In many same sound received among multiple cases, fishes restrict their soniferous SOFTWARE NEEDS hydrophones and then using various sta- activity predominately to narrow win- To carry out the research priorities tistical techniques such as time delay dows of time. For example, spotted listed above, software must be developed differences to triangulate on the calls. seatrout (Cynoscion nebulosus) calls pre- in several areas. Programs should be To develop the ability to produce real dominantly from just before sunset into packaged to allow biologists to process time GIS plots of soniferous fishes over- the early evening (Mok and Gilmore sound data to the fullest extent possible. lain on environmental parameters, these 1983), so attempts to use passive acous- Automatic processing of sound steps must be automated. One important tics to map their distribution would recordings—The passive acoustics stud- product of localization, other than habi- have to be conducted within that time ies described above can generate large tat association mapping, is the frame. amounts of acoustic data (often thou- determination of true sound source lev- How many?—Quantification of fish sands of hours of recordings) that are els (see discussion “Where?”) under abundance at a location and time is an laborious to process fully to obtain the varying environmental conditions (e.g., important objective of many fisheries basic biological data sought (e.g., tem- Sprague and Luczkovich 2004). studies and monitoring programs. Once poral activity patterns, spatial HARDWARE NEEDS the what, when, and where questions distribution, etc.). We suggest that soft- have been addressed, additional studies ware designed to address the following Acoustic hardware technology is can quantify abundance through passive processing needs would greatly enhance rapidly evolving; however, there are sev- acoustics techniques. The two most the utility of passive acoustics to fish- eral specific needs that would enhance important study areas are: (1) correla- the development of passive acoustics in eries. tion between the “amount” of sound fisheries. production and the abundance of fish, 1. Automatic signal detection software Data storage—Low cost, high capac- and (2) determination of the proportion used to detect the call of a given fish ity digital recorders are needed to allow of fish that are soniferous at a given species amid long time series of sound field sampling at the rates and intensi- place and time. recordings (thousands of hours) with ties necessary to study temporal and

Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 441 spatial patterns of sound production. recording of acoustic and video during by producing these targeted materials, Devices that can be programmed to shore or small boat based studies. conducting training workshops, and by record at varying sampling rates and Currently investigators have to purchase attracting engineers with a biological time settings, including continuous and separate components and assemble them interest to the field. The workshop par- time lapse would be particularly valu- into a package themselves. Because ticipants identified a strong need to able. Existing low-cost recorders components were not manufactured to educate scientists, managers, and the developed by the music industry are work together, numerous in-line public on the uses of passive acoustics. rapidly becoming scarce due to the adapters often must be used to transfer industry shift from wave file storage for- the acoustic signal among components. Some specific needs include: mats to MP3 file format storage (MP3 is Researchers need to be able to drop a 1. Manuals and other literature intended less desirable because it compresses and hydrophone over the side of a small boat to introduce biologists to the field; distorts the sounds). or dock, together with a small underwa- Coupled “visualization” technolo- ter video camera, and be able to 2. Workshops that bring biologists, acous- gies—To validate the identification of simultaneously record and monitor ticians, and engineers together are vital sound sources in the field, devices that video and acoustic data. Ideally, the as they stimulate the transfer of knowl- couple passive acoustics with other audio and video data would be automat- edge and ideas among the disciplines; technologies that can be used to identify ically converted to digital form for 3. Development of passive acoustic train- a fish are needed. Coupled acoustic storage. Availability of low cost devices ing centers for researchers and optic systems include devices that are of this type would facilitate the feasibil- students; and capable of recording both underwater ity of small-scale research projects for a video and acoustic data are among the larger scientific community. 4. Incorporation of passive acoustics into simplest to use and are highly reliable fisheries curricula. for identification (Lobel 2001). EDUCATION AND Amazingly, most underwater video tech- OUTREACH Passive acoustics technologies provide a unique public outreach potential. nology marketed to researchers lacks Most fish bioacousticians are biolo- Scientists and laymen alike are often acoustic recording capabilities. In our gists first and engineers second. They opinion this is entirely due to a failure of have arrived at fish bioacoustics because fascinated by the phenomenon of under- the manufacturers to recognize the mar- it is a powerful tool for studying fishes. water sounds. Passive acoustics ket potential of such devices, rather This means that engineering and signal technologies are amenable to multime- than to technical issues. Another diffi- processing principles must be learned on dia display via the Internet and have culty of acoustic-optic systems is the the job. Unfortunately, there is no one great potential as public education and need to use artificial light sources for the good source of information about outreach tools. video recording, and resulting dramatic recording and signal processing that is change in fish behavior and probability accessible and practical for the fish bioa- ACKNOWLEDGMENTS of detection. In addition, artificial light coustician. This gap can be bridged both provides a limited range of visibility. We wish to acknowledge the contri- Infrared lights are only effective for a bution of numerous participants in the few feet, and even state-of-the-art low Selected examples of soniferous fishes that Passive Acoustics Workshop and AFS support important recreational or commercial light level cameras are ineffective at Symposium. Numerous discussions at fisheries (fish sketches courtesy of night even at moderate depths. Finally NOAA/NMFS Northeast Fisheries Science these meetings, and in subsequent corre- studies are needed to understand the Center). spondences since, were invaluable to the effect of light of various wavelengths authors in the development of this and intensities on fish behavior to manuscript. The workshop and publica- develop the optimum optic system. tion of the workshop proceedings Coupling passive acoustics with other visualization technologies such as active received major funding from MIT Sea acoustics and laser line-scanning is also Grant, the Office of Naval Research, promising. Active acoustic technology and from the Northeast Great Lakes has the advantage of providing rela- Center of the National Undersea tively long range observation Research Program. Travel for some capabilities compared with video, but workshop participants was funded in requires the input of acoustic energy whole, or in part by: Connecticut Sea into the environment. Laser line-scan- Grant, Florida Sea Grant, Hawaii Sea ning can provide high resolution Grant, Louisiana Sea Grant, North imaging at short ranges (<10 m; Carey Carolina Sea Grant, the South Carolina et al. 2003), but is currently relatively expensive. Sea Grant Consortium, Texas Sea Portable passive acoustic survey Grant, and the Woods Hole Sea Grant devices—Small, self-contained packages Programs. This is SMAST Contribution of acoustic gear would allow mobile No. 06-0601.

442 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 443 REFERENCES _____. 1961c. Sound production by the cod Dobrin, M. B. 1947. Measurements of (Gadus callarias L.). Behaviour 18:239- underwater noise produced by marine life. Almada, V. C., M. C. P. Amorin, E. Pereira, 245. Science 105:19-23. F. Almada, R. Matos and R. Godinho. Breder, C. M., Jr. 1968. Seasonal and diurnal D’Spain, G. L., W. A. Kuperman, L. P. 1996. Agonistic behaviour and sound occurrences of fish sounds in a small Berger, and W. S. Hodgkiss. 1996. production in Gaidropsarus mediterraneus Florida Bay. Bulletin of the American Geoacoustic inversion using fish sounds. (Gadidae). Journal of Fish Biology Museum of Natural History 138:278-329. The Journal of the Acoustical Society of 49:363-366. Calupca, T. A., K. M. Fristrup and C. W. America 100:2665. Alshuth, S., and R. G. Gilmore. 1993. Egg Clark. 2000. A compact digital recording Dufossé, A. 1874a. Rescherches sur les bruts identification, early larval development system for autonomous bioacoustic et les sons expressifs que font entendre les and ecology of the spotted seatrout, monitoring. Journal of the Acoustic poissons d'Europe et sur les organes Cynoscion nebulosus C. (Pisces: Society 108:2582. producteurs de ces phenomenes Sciaenidae). International Council for the Carey, D. A., D. C. Rhoads and B. Hecker. acoustiques ainsi que sur les appareils de Exploration of the Sea, Conference 2003. Use of laser line scan for assessment l'audition de plusieurs de ces animaux. Manuscript 1993/G 28. of response of benthic habitats and Annales des Sciences Naturelles. Zoologie _____. 1994. Salinity and temperature demersal fish to seafloor disturbance. et Biologie Animale Series 5, Vol. 19. tolerance limits for larval spotted seatrout, Journal of Experimental Marine Biology _____. 1874b. Rescherches sur les bruts et les Cynoscion nebulosus C. (Pisces: and Ecology 285-286:435-452. sons expressifs que font entendre les Sciaenidae). International Council for the Chao, L. N. 1978. A basis for classifying poissons d'Europe et sur les organes Exploration of the Sea, Conference western Atlantic sciaenidae (Teleostei: producteurs de ces phenomenes Perciformes). NOAA Technical Report acoustiques ainsi que sur les appareils de Manuscript 1994/L:17. NMFS Circular 415. l'audition de plusieurs de ces animaux. _____. 1995. Egg and early larval Chu, Y. T., Y. L. Lo, and H. L. Wu. 1963. A Annales des Sciences Naturelles. Zoologie characteristics of Pogonias cromis, Bairdiella study on the classification of the sciaenoid et Biologie Animale Series 5, Vol. 20. chrysoura, Cynoscion nebulosus (Pisces: fishes of China, with description of new Engen, F., and I. Folstad. 1999. Cod Sciaenidae), from the Indian River genera and species. 1972 Reprint, courtship song: a song at the expense of Lagoon, Florida. International Council for Antiquariaat Junk, Netherlands. dance? Canadian Journal of Zoology the Exploration of the Sea, Conference Clark, C. W. 1980. A real-time direction 77(4):542 550. Manuscript 1995/L:17. finding device for determining the bearing Fine, M. L., H. E. Winn, L. Joest and P. J. Aristotle. Historia Animalism, IV, 9. Trans. to the underwater sounds of southern right Perkins. 1977a. Temporal aspects of by D. D'A.Thompson. 1910. Clarendon whales, Eubalaena australis. Journal of the calling behavior in the oyster toadfish, Press, Oxford. Acoustical Society of America 68(2):508- Opsanus tau. Fishery Bulletin 75:871-874. Barimo, J. F., and M. L. Fine. 1998. 511. Fine, M. L., H. E. Winn and B. Olla. 1977b. Relationship of swim-bladder shape to the Connaughton, M. A., and M. H. Taylor. Communication in fishes. Pages 472-518 directionality pattern of underwater sound 1994. Seasonal cycles in the sonic muscles in T. Sebok, ed. How animals in the oyster toadfish. Canadian Journal of of the weakfish, Cynoscion regalis. Fishery communicate. Indiana University Press, Zoology 76:134-143. Bulletin 92:697-703. Bloomington. Berk, I. M. 1998. Sound production by white _____. 1995. Seasonal and daily cycles in Finstad, J. L., and J. T. Nordeide. 2004. shrimp (Penaeus setiferus), analysis of sound production associated with Acoustic repertoire of spawning cod, another crustacean-like sound from the spawning in the weakfish, Cynoscion Gadus morhua. Environmental Biology of Gulf of Mexico, and applications for regalis. Environmental Biology of Fishes Fishes 70:427-433. passive sonar in the shrimp industry. 42:233-240. Fish, J. F. 1966. Sound production in the Journal of Shellfish Research Connaughton, M. A., M. H. Taylor, and M. American lobster Homarus americanus. 17:1497-1500. L. Fine. 2000. Effects of fish size and Crustaceana 11:105. temperature on weakfish disturbance calls: Fish, M. P. 1964. Biological sources of Bradbury, J. W., and C. A. Bloomgarden. implications for the mechanism of sound sustained ambient sea noise. Pages 2003. Creating a web-based library of generation. Journal of Experimental 175-194 in W. N. Tavolga, ed. Marine bio- underwater biological sounds. Pages Biology 203:1503-1512. acoustics. Pergamon Press, NY. 103-106 in R. A. Rountree, C. Goudey, Cousteau, J. Y., and F. Dumas. 1953. The Fish, M. P., A. S. Kelsey, Jr., and W. H. and T. Hawkins, eds. Listening to fish: silent world. Perennial Library, Harper Mowbray. 1952. Studies on the proceedings of the international workshop and Row, Publishers, New York. production of underwater sound by North on the applications of passive acoustics to Darwin, C. 1874. The descent of man. 2nd Atlantic coastal fishes. Journal of Marine fisheries. 8-10 April 2002. Dedham, MA. edition. H. H. Caldwell, New York. Research 11:180 193. MIT Sea Grant Technical Report. Dijkgraaf, S. 1947. Ein tone erzeugender Fish, M. P., and W. H. Mowbray. 1970. Brawn, V. M. 1961a. Aggressive behaviour in fisch in Neapler Aquarium. Experimenta Sounds of western North Atlantic fishes. the cod (Gadus callarias L.). Behaviour 3:494. Johns Hopkins Press, Baltimore, 18:107-147. _____. 1949. Untersuchungen uber die Maryland. _____. 1961b. Reproductive behaviour of the funktionen des ohrlabyrinths bei Gilmore, R. G, Jr. 2002. Sound production cod (Gadus callarias L). Behaviour 18:177- meeresfischen. Physiologia Comparata and communication in the spotted 198. Oecoloia 2:81-106. seatrout. Pages 177-195 in S. A. Bortone,

444 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG ed. Biology of the spotted seatrout, CRC Hawkins, A. D., and K. J. Rasmussen. 1978. ______. 2001. Fish bioacoustics and behavior: Press, Boca Raton, Florida. The calls of gadoid fish. Journal of the passive acoustic detection and the Goode, G. B. 1887. American fishes: a Marine Biological Association of the application of a closed-circuit rebreather popular treatise upon the game and food United Kingdom 58:891 911. for field study. Marine Technology Journal fishes of North America with special Henninger, H. P., and W. H. III., Watson. 35(2):2-19. reference to habits and methods of 2005. Mechanisms underlying the ______. 2002. Diversity of fish spawning capture. L. C. Page Co., Boston. production of carapace vibrations and sounds and the application of passive Hawkins, A. D. 1986. Underwater sound associated waterborne sounds in the acoustic monitoring. Bioacoustics 12:286- and fish behaviour. Pages 114-151 in T. J. American lobster, Homarus americanus. 289. Pitcher, ed. The behaviour of teleost The Journal of Experimental Biology ______. 2003. Synchronized underwater audio-video recording. Pages 136-139 in fishes. Groom Hellm, London. 208:3421-3429. R. A. Rountree, C. Goudey, and T. ______. 2003. The use of passive acoustics to Iversen, R. T. B., P. J. Perkins, and R. D. Hawkins, eds. Listening to fish: identify a haddock spawning area. Pages Dionne. 1963. An indication of proceedings of the international workshop 49-53 in R. A. Rountree, C. Goudey, and underwater sound production by squid. on the applications of passive acoustics to T. Hawkins, eds. Listening to fish: Nature 199(4890):250-251. fisheries. 8-10 April 2002. Dedham, MA. proceedings of the international workshop Kaatz, I. M. 2002. Multiple sound producing MIT Sea Grant Technical Report MITSG on the applications of passive acoustics to mechanisms in teleost fishes and 03-2. fisheries. 8-10 April 2002. Dedham, MA. hypotheses regarding their behavioural Lobel, P. S., and D. A. Mann. 1995. MIT Sea Grant Technical Report MITSG significance. Bioacoustics 12:230-233. Spawning sounds of the damselfish, 03-2. Knudsen, V. O., R. S. Alfred, and J. W. Dascyllus albisella (Pomacentridae), and Hawkins, A. D., L. Casaretto, M. Picciulin, Emling. 1948. Underwater ambient noise. relationship to male size. Bioacoustics and K. Olsen. 2002. Locating spawning Journal of Marine Research 7:410-429. 6(3):187-198. haddock by means of sound. Bioacoustics Lobel, P. S. 1998. Possible species specific Locascio, J. V., and D. A. Mann. 2005. 12:284-286. courtship sounds by two sympatric cichlid Effects of Hurricane Charley on fish Hawkins, A. D., C. J. Chapman, and D. J. fishes in Lake Malawi, Africa. chorusing. Proceedings of the Royal Symonds. 1967. Spawning of haddock in Environmental Biology of Fishes Society of London Biological Letters captivity. Nature 215:923 925. 52:443-452. 1:362 365.

National Marine Fisheries Service (NMFS) / Sea Grant Joint Graduate Fellowship Program in Population Dynamics and Marine Resource Economics Description United States or its territories • fellowships for highly qualified Ph.D.-level graduate students • prospective Marine Resource Economics Fellows must be in interested in careers in: (1) population dynamics of living marine process of completing at least two years of course work in Ph.D. resources and development and implementation of quantitative program in natural resource, marine resource, or environmental methods for assessing their status, and (2) economics of economics or related field conservation and management of living marine resources • support for up to three years for Population Dynamics fellowships, Award and up to two years for Marine Resource Economics fellowships • grant or cooperative agreement of $38,000 per year awarded to • approximately two fellowships awarded each year in each local Sea Grant program/host university discipline, with overall maximum of 12 Fellows at any time • 50% of funds provided by NMFS, 33 1/3% provided by National • fellows work closely with mentors from NMFS Science Centers or Sea Grant Office (NSGO), and 16 2/3% provided by university as Laboratories and may intern at NMFS facility on thesis research or required match of NSGO funds related problem • disbursement of award for salary, living expenses, tuition, health insurance, other fees, and travel determined by university Program goals • encourage qualified applicants to pursue careers in and increase Relevant dates available expertise related to: (a) population dynamics and • application deadline—early February 2007 (see Sea Grant website assessment of status of stocks of living marine resources, or for details—www.seagrant.noaa.gov/funding/rfp2006.html) (b) economic analysis of living marine resource conservation • fellowship start date: 1 June 2007 and management decisions • foster closer relationships between academic scientists Contact and NMFS • Dr. Terry Smith • provide real-world experience to graduate students National Sea Grant College Program and accelerate their career development 1315 East-West Highway Silver Spring, MD 20910 Eligibility 301/713-2435 • must be United States citizen [email protected] • prospective Population Dynamics Fellows must be • any state Sea Grant program— admitted to Ph.D. program in population dynamics or www.nsgo.seagrant.org/SGDirectors.html related field (applied mathematics, statistics, or • any participating NMFS facility— quantitative ecology) at academic institution in www.nmfs.noaa.gov/science.htm

Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG 445 Luczkovich, J. J., M. W. Sprague, S. E. Journal of Experimental Biology _____. 1993. Spawning site selection by Johnson, and R. C. Pullinger. 1999. 205:2375-2385. spotted seatrout, Cynoscion nebulosus, and Delimiting spawning areas of weakfish Popper, A. N. 2003. Effects of anthropogenic black drum, Pogonias cromis, in Cynoscion regalis (Family Sciaenidae) in sounds on fishes. Fisheries 28(10):24-31. Louisiana. Environmental Biology of Pamlico Sound, North Carolina using Protasov, V. R., and M. I. Aronov. 1960. On Fishes 36:257-272. passive hydroacoustic surveys. the biological significance of sounds of Sprague, M. W., and J. J. Luczkovich. 2004. Bioacoustics 10:143-160. certain Black Sea fish. (In Russian). Measurement of an individual silver perch Luczkovich, J. J., J. D. Hall, III., M. Biofizika 5:750 752. Bairdiella chrysoura sound pressure level in Hutchinson, T. Jenkins, S. E. Johnson, Qi, M., S. Zhang, and Z. Song. 1984. Studies a field recording. Journal of the Acoustic R. C. Pullinger, and M. W. Sprague. on the aggregate sound production of most Society of America 116(15):3186-3191. 2000. Sounds of sex and death in the sea: species of croaker (sciaenoid fishes) in Svellingen, I., B. Totland and J. T. Oevredal. bottlenose dolphin whistles suppress Bohai Sea, Yellow Sea and East China 2002. A remote-controlled instrument mating choruses of silver perch. Sea. Studia Marina Sinica, Peking platform for fish behaviour studies and Bioacoustics 10:323-334. 21:253-264. sound monitoring. Bioacoustics 12:335- Mallekh, R., J. P. Lagardere, J. P. Eneau, Rountree, R. A., P. J. Perkins, R. D. 336. and C. Cloutour. 2003. An acoustic Kenney, and K. R. Hinga. 2002. Sounds Takemura, A., T. Takita, and K. Mizue. detector of turbot feeding activity. of western North Atlantic fishes: data 1978. Studies on the underwater sound Aquaculture 221:481-489. rescue. Bioacoustics 12(2/3):242 244. VII: Underwater calls of the Japanese Mann, D. A., and S. M. Jarvis. 2004. Rountree, R. A., C. Goudey, T. Hawkins, J. marine drum fishes (Sciaenidae). Bulletin Potential sound production by a deep-sea Luczkovich and D. Mann. 2003a. of the Japanese Society of Scientific fish. Journal of the Acoustical Society of Listening to fish: passive acoustic Fisheries 44:121-125. America 115(5):2331-2333. applications in marine fisheries. Sea Grant Tavolga, W. N. 1960. Sound production and Mann, D. A., and P. S. Lobel. 1995a. Passive Digital Oceans. Massachusetts Institute of underwater communication in fishes. acoustic detection of fish sound Technology Sea Grant College Program. Pages 93-136 in W. E. Lanyon and W. N. production associated with courtship and Tavolga, eds. Animal sounds and MITSG 0301. Available at: spawning. Bulletin of Marine Science communication. American Institute of http://web.mit.edu/seagrant/ 57(3):705-706. Biological Sciences, Washington, D.C. aqua/cfer/acoustics/PD9x9FINAL.pdf _____. 1995b. Passive acoustic detection of ______. 1980. Hearing and sound Rountree, R. A., C. Goudey, and T. sounds produced by the damselfish, production in fishes in relation to fisheries Hawkins. Editors. 2003b. Listening to Dascyllus albisella (Pomacentridae). management. Pages 102-123 in J. E. fish: proceedings of the international Bioacoustics 6:199-213. Bardach, J. J. Magnuson, R. C. May, and J. workshop on the applications of passive Mok, H. K., and R. G. Gilmore. 1983. M. Reinhart, eds. Fish behavior and its use acoustics to fisheries. 8-10 April 2002. Analysis of sound production in estuarine in the capture and culture of fishes. Dedham, MA. MIT Sea Grant Technical aggregations of Pogonias cromis, Bairdiella International Center for Living Aquatic Report MITSG 03-2. Available at: chrysoura, and Cynoscion nebulosus Resources Management Conference http://web.mit.edu/seagrant/aqua/cfer/ (Sciaenidae). Bulletin of the Institute of Proceedings 5, Manila, Philippines. acoustics/PAprocBrFINAL.pdf Zoology, Academia Sinica 22:157-186. Templeman,W., and V. M. Hodder. 1958. Rowe, S., and J. A. Hutchings. 2004. The Moulton, J. M. 1957. Sound production in Variation with fish length, sex, stage of the spiny lobster Panulirus argus function of sound production by Atlantic sexual maturity, and season in the (Latreille). Biological Bulletin 113:286- cod as inferred from patterns of variation appearance and volume of the drumming 295. in drumming muscle mass. Canadian muscles of the swimbladder in the Myrberg, A. A., and R. J. Riggio. 1985. Journal of Zoology 82:1391-1398. haddock, Melanogrammus aeglefinus L. Acoustically mediated individual Sakas, C. J., C. Goudey and R. A. Journal of the Fisheries Research Board of recognition by a coral reef fish Rountree. 2005. Sanctuary sounds Canada 15:355-90. (Pomacentrus partitus). Animal Behavior monitoring underwater sounds in the Templeman, W., V. M. Hodder, and R. 33:411-416. National Marine Sanctuaries. Oceans Wells. 1978. Sexual maturity and Nordeide, J. T., and I. Folstad. 2000. Is cod 2005 Marine Technology Society/Institute spawning in haddock, Melanogrammus lekking or a promiscuous group spawner? of Electrical and Electronics Engineers, aeglefinus, of the southern Grand Bank. Fish and Fisheries 1:90-93. Inc. Conference Proceedings. MTS. The International Commission for the Nordeide, J. T., and E. Kjellsby. 1999. Sound Sartori, J. D., and T. J. Bright. 1973. Northwest Atlantic Fisheries Research from spawning cod (Gadus morhua L.) at Hydrophonic study of the feeding Bulletin 13:53-65. the spawning grounds. ICES Journal of activities of certain Bahamian parrot Watkins, W. A., and W. E. Schevill. 1972. Marine Science 56:326-332. fishes, family Scaridae. Hydro-Laboratory Sound source location by arrival-times on Okumura, T., T. Akamatsu, and H. Y. Yan. Journal 2:25-56. a non-ridid three-dimensional 2002. Analyses of small tank acoustics: Saucier, M. H., and D. M. Baltz. 1992. hydrophone array. Deep-Sea Research empirical and theoretical approaches. Hydrophone identification of spawning 19:691-706. Bioacoustics 12:330-332. sites of spotted seatrout Cynoscion Wood, M., L. Casaretto, G. Horgan, and A. Patek, S. N. 2002. Squeaking with a sliding nebulosus (Ostichthys: Sciaenidae) near D. Hawkins. 2002. Discriminating joint: mechanisms and motor control of Charleston, South Carolina. Northeast between fish sounds a wavlet approach. sound production in palinurid lobsters. Gulf Science 12(2):141-145. Bioacoustics 12:337-339.

446 Fisheries • VOL 31 NO 9 • SEPTEMBER 2006 • WWW.FISHERIES.ORG