This article has been published in Oceanography published been This has article or collective redistirbution other or collective or means this reposting, of machine, is by photocopy article only anypermitted of with portion the approval The O of

NURC: Celebrating 50 Years of International Partnerships IN oCean Research and Operations , Volume 2, a quarterly 21, Number The O journal of The Interaction of Marine Mammals ceanography S

and Active Sonar ociety. Copyright 2008 by The O By Kendra Ryan

Over the last 50 years, international FOUNDING OF SOLMAR environmental assessment procedures public concern regarding human impact From May 12–15, 1996, NURC (for- be implemented as soon as possible ceanography S

on the environment has increased merly known as SACLANT Undersea with a view to recommending suitable ceanography S drastically. Specific to the maritime Research Centre, SACLANTCEN) mitigation and monitoring protocols. domain, attention has more acutely conducted the Shallow Water Acoustic In addition, the panel determined that ociety. A ociety. S

focused during the last ten years on the Classification (SWAC ’96) research additional research was needed to Permission is reserved. ll rights granted to in teaching copy this and research. for use article Republication, systemmatic reproduction, temporal and spatial relationships of cruise using low- and mid-frequency determine biological and behavioral end all correspondence to: [email protected] Th e O or marine mammals’ strandings to the use (600-Hz and 3-kHz, respectively) characteristics of marine mammals. The of active sonar. The NATO Undersea active sonars off the coast of Greece findings and recommendations of the Research Centre (NURC) has been at (SACLANT Undersea Research 1998 Bioacoustics Panel prompted the the forefront of research regarding the Centre, 1998). Concurrently, 12–13 creation of the SOLMAR program. interaction of marine mammals and Cuvier’s beaked whales (Ziphius active sonar. In 1999, NURC established cavirostris) stranded along the coast SOLMAR Objectives one of the first research programs in of Kyparissiakos Gulf, Greece. NURC (1999–2006)

the world to address this topic, the hosted a bioacoustics panel comprised The SOLMAR program was a multi- ceanography S Sound Ocean Living Marine Resources of subject experts shortly after Frantzis national, multidisciplinary research (SOLMAR) program, now known as (1998) hypothesized an association project with the objective of developing ociety, P the Marine Mammal Risk Mitigation between the SWAC experiment and the the tools and/or procedures with which

(MMRM) program. With the participa- stranding event. The panel concluded an experimenter can ensure there are no O B ox 1931, Rockville, M D 20849-1931, U ox 1931, Rockville, tion of numerous international partners, that an acoustic link was neither clearly marine mammal risk mitigation policies, established nor eliminated as a direct or Lt. Kendra Ryan (USN) ([email protected]. procedures, and technologies were indirect cause of the May 1996 stranding int) was Marine Mammal Risk Mitigation implemented, and continue to be devel- (SACLANT Undersea Research Centre, Project Officer, NATO Undersea Research oped, to minimize the potential impact 1998). Despite the nondefinitive conclu- Centre, La Spezia, Italy, from January 2007 on marine mammals. sion, the panel recommended that to May 2008. SA .

38 Oceanography Vol.21, No.2 Figure 1. Sirena ’99 (left) and Sirena ’00 (right) Sea WIFS chlorophyll a (mg m-3) in the upper layer, chlorophyll a maximum (mg m-3) with fin whale visual sightings in the middle layer, and the depth of the maximum values of chlorophyll a in the lower layer. SACLANT Undersea Research Centre, 2001

marine mammals near a sonar source marine mammal populations in the and ammonias), fluorescence, turbidity, prior to and during its use (D’Amico et Mediterranean Sea and environmental and phytoplankton (chlorophyll), were al., 2003b). At the inception of the pro- parameters in relationship to these gathered concurrently from an Italian gram, little was known about the regions population densities navy hydrographic vessel and from of high and low cetacean density within • Acoustic and visual tools to detect various other vessels provided by partner the areas where NURC experiments were cetaceans organizations. Sea surface temperature, conducted. The first tasks of the project To accomplish these last two objectives, surface currents, surface roughness, and were therefore to develop: a series of sea trials, entitled SIRENA, ocean color were collected in real time at • A network of scientists with knowl- were conducted. NURC using satellite remote sensing. edge of marine mammal distribution, behavior, bioacoustics, and sonars SIRENA SIRENA Results • A database containing all documents The SOLMAR program conducted The SIRENA cruises provided a vast and data related to marine mammal SIRENA sea trials annually in the quantity of data that enabled scientists distribution, behavior, and associated Ligurian Sea from 1999 to 2003. to answer some of the primary questions bioacoustics information Bounded by Italy, mainland France, regarding Mediterranean Sea cetaceans • A risk mitigation instruction based and Corsica, the Ligurian Sea was and their habitat preferences. During on available knowledge that contained designated the first International Marine SIRENA ’99, a total of 869 individuals policy and protocols to ensure that Sanctuary in the Mediterranean Sea was observed during 146 marine mam- active acoustic research was planned and was the common habitat of eight mal sighting events (D’Amico et al., and performed in an environmen- cetacean species. The NATO research 2000). Analysis of the data from 1999 and tally sensitive manner (SACLANT vessel, R/V Alliance, was the primary those from SIRENA ’00 showed that the Undersea Research Centre, 2004) platform for conducting visual and fin whale (Balaenoptera physalus) and Upon completion of these tasks, the acoustic monitoring of cetaceans. sperm whale (Physeter macrocephalus) project shifted focus to development of: Oceanographic measurements, including generally preferred cold, deep, nutrient- • A database containing information on temperature, salinity, nutrients (silicates rich portions of the basin (Figure 1;

Oceanography June 2008 39 SACLANT Undersea Research Centre, 2001) while the Cuvier’s beaked whales preferred a submarine canyon (the Genoa Canyon) where there is a frontal influence and steep bathymetry (D’Amico et al., 2003b). Also during SIRENA ’00, a sperm whale was tagged with a noninvasive digital recording device (Johnson and Tyack, 2003) to gather data regarding the whale’s diving behavior and vocalization activity. The data indicated that the aver- age depth of this sperm whale’s foraging dives was 900 m. At this foraging depth, the acoustic behavior of the sperm whale Figure 2. Three-dimensional track of a tagged sperm whale. Surface positions marked with an “x” indicate a visual sighting. Tyack and changed, and the data supported the Zimmer, 2002 hypothesis that sperm whales echolocate on prey at a constant horizontal layer (Zimmer et al., 2003). Such data advance knowledge of normal sperm whale behavior so that behavioral responses combined with visual and passive acous- application of digital recording devices to anthropogenic activities may be tic methods to enable three-dimensional to sperm whales continued, and opera- better understood. tracking of the whale’s movements and tions included exploring the feasibility The objectives of the SIRENA ’01 foraging events (Figure 2; Tyack and of a whale-finder sonar. Using defined sea trial were to continue collection Zimmer, 2002). Analysis of this informa- protocols, the feasibility test involved of oceanographic and species data for tion supported conclusions regarding controlled transmission of an active population of the developing database sperm whale habitat preference that were sound source in the vicinity of tagged and to evaluate techniques for effective derived from previous SIRENA cruise animals. Analysis of the passive acoustic risk mitigation (SACLANT Undersea data (SACLANT Undersea Research data confirmed that sperm whales produce multipulse clicks centered around 15 kHz, with source levels of more than 230 dB re 1µPa @ 1 m, and (NURC) has been at the forefront of a narrow, forward-oriented sonar beam (Møhl et al., 2003; Zimmer et al., 2005b). research regarding the interaction of Additionally, in 2002, the SIRENA sea marine mammals and active sonar. trial focused on determining the specific habitat of the Cuvier’s beaked whale. Based on Cuvier’s sighting data from previous years, a survey of the Genoa Research Centre, 2000). During the Centre, 2002). In addition, visual sight- Canyon (~10,600 km2) was undertaken. cruise, tag data (e.g., vocalizations, ings and acoustic data from a towed Analysis of the oceanographic and pres- dive depths, and whale orientation) array contributed to the population of ence data determined that all Cuvier’s transmitted from a digital recording the marine-mammal-sighting database. sightings were in water depths greater device attached to a sperm whale were During SIRENA ’02 and ’03, the than 1000 m. These areas may provide a

40 Oceanography Vol.21, No.2 preferential habitat (one with an abun- dant food source) due to the oceano- graphic conditions that result from the interaction of the Ligurian current and the canyon walls. In addition, the canyon environment can form traps for suspended particulate matter containing high biomass (D’Amico et al., 2001). Prior to 2004, the sperm whale was the animal of interest for tagging due to its ease of acoustic and visual detec- tion as compared to the beaked whale. Although valuable information regard- ing tagging and exposure protocols was obtained from the sperm whale experi- ments, still little was known regarding the particular behavior of the Cuvier’s beaked whale. However, in 2004, analy- ses of data from tags simultaneously applied to two Cuvier’s baked whales resulted in the first estimates of the source level, directivity index, and spec- tral properties of Cuvier’s beaked whale Figure 3. Sample of Z. cavirostris click as received by the tag of a conspecific whale. The top panel shows the time series, the clicks (Figure 3; Zimmer et al., 2005a). middle panel shows the spectrogram, and the bottom panel As a result of the SOLMAR program, shows the relative power spectrum. Zimmer et al., 2005a a thorough database was constructed that contains visual sightings during the SIRENA trials, additional sighting and stranding data from collaborat- ing organizations, and historical data design and implementation of a marine the Marine Mammal Risk Mitigation gathered from literature reviews. This mammal risk mitigation package based (MMRM) project. database contains 5463 sighting data on the information acquired during the Components of the package included points and 4300 stranding data points, previous years of SIRENA sea trials. passive acoustic monitors, a predictive making it one of the most extensive This package provided researchers and habitat model, a sound propagation data sets on marine mammals in the military commanders with aids to assess model, a Web site with training aides, Mediterranean Sea. the potential for a given area of the sea and a guiding policy for the use of active to contain beaked whales and other sonar in experiments at NURC. The MARINE MAMMAL RISK marine mammals that may be impacted passive acoustic monitors implemented MITIGATION by the use of active sonar. The package in the toolkit were the result of evolv- Although some of the initial questions also provided mitigation techniques and ing methods and technologies used in regarding the habitat and characteristics protocols to be implemented during the each successive year of the SIRENA of Cuvier’s beaked whales remained use of active sonar. This shift in project cruises. In the early years of the project, unanswered, ongoing operational and objectives correlated with a change Cuvier’s beaked whales’ vocalizations research requirements necessitated the in the project title from SOLMAR to were unknown. Thus, methodologies

Oceanography June 2008 41 that had previously recorded other the predictive habitat model, a tool to modified until the animals are no longer cetaceans in low-frequency (< 1 kHz) determine the presence or absence of detected in the area. This model and and mid-frequency (1–10 kHz) ranges species of interest in a certain geographi- additional information about the marine were implemented. Instrumentation cal area. The basis of the habitat model protected areas and the most common used included a 32-element passive is the quantification of species and their species of the Mediterranean Sea are array with real-time beamforming in relationship to environmental factors available on the project Web site (http:// frequencies from 750 Hz to 4.1 kHz using sighting and environmental data www.solmar.nurc.nato.int). during SIRENA ’99 and SIRENA ’00, from past SIRENA cruises. From these and the use of a 128-element array to data, distinct environmental variables ENVIRONMENTAL POLICY One of the initial objectives of the SOLMAR project was to develop a risk mitigation instruction that contained policies and protocols to ensure that ...marine mammal risk mitigation policies, active acoustic research was planned procedures, and technologies were and performed without negative impact implemented...to minimize the to human divers and marine mammals (http://solmar.nurc.nato.int). This potential impact on marine mammals. instruction, entitled “NATO Undersea Research Centre Human Diver and Marine Mammal Risk Mitigation Rules” (NATO Undersea Research Centre, provide frequency coverage from 200 Hz were determined to correlate with a 2006), was created shortly after the 1998 to 16 kHz (Zimmer et. al., 2003). These particular species (Azzelino et al., 2001). Bioacoustics Panel convened. It was arrays were suitable for detecting the The model contains predictor variables originally based on knowledge available broadband echolocation clicks of sperm for fin whales, sperm whales, Cuvier’s at the time, but is continually updated by whales; however, the range of the devices beaked whales, and striped dolphins. the MMRM program to reflect the latest was insufficient to detect the Cuvier’s Such a model offers the possibility of research in this field. The instruction beaked whale. It was not until beaked focusing field activities in much smaller is multilayered, with the main points whale clicks were recorded on a digital areas and developing mitigation options, stating that all Centre sea trials must recording tag that the frequency range such as avoiding active sonar operations perform an environmental scoping in which these animals vocalize was in areas where there is high probability study prior to any experiment, that a revealed. The following year, analysis of of encountering sensitive species. maximum received sound-pressure data obtained from digital recording tags The third component of the marine level or energy-density level must not applied to two Cuvier’s beaked whales mammal risk mitigation package is a be exceeded during the experiment, and revealed vocalizations with a center sound propagation tool. Prior to con- that risk-mitigation techniques must frequency of 42 kHz (Zimmer et al., ducting an experiment, characteristics be applied during the experiment. In 2005a). New arrays and compact passive of active sources that will be used in the addition, the instruction was modified acoustic monitors (CPAMS) that include experiment and the geographic area of to include reference to the Marine this higher frequency will be deployed in the experiment are input to the model Mammal Crisis Action Team, now called SIRENA ’08 in an effort to acquire more to predict propagation loss. From this the Human Diver and Marine Mammal data on the vocalization behavior of prediction, visual monitoring ranges are Incident Action Team. The purpose Cuvier’s beaked whales. estimated. During the experiment, if of this team is to respond to marine Another component of the marine animals are sighted within these moni- mammal stranding incidents or diver mammal risk mitigation package is toring ranges, active sonar use must be accidents that may be temporally or spa-

42 Oceanography Vol.21, No.2 tially associated to Centre activities, to program collaborated with academia, recording tag for measuring the response of wild marine mammals to sound. IEEE Journal of assist national authorities analyzing the government, and environmental organi- Oceanic Engineering 28:3–12. event, and to assess the oceanographic zations of partner nations to collect data Møhl, B., M. Wahlberg, P.T. Madsen, A. Heerfordt, and conditions related to the incident. Since and generate guiding principles for the A. Lund. 2003. The monopulsed nature of sperm whale clicks. Journal of the Acoustical Society of the initial policy was developed in 1999, safe conduct of active sonar experiments America 114:1,143–1,154. no known marine mammal strandings in the presence of marine mammals. By NATO Undersea Research Centre. 2006. Human Diver and Marine Mammal Risk Mitigation Rules and have occurred in spatial or temporal adopting a proactive approach, NURC Procedures. NURC Special Publication 2006-008, concurrence to Centre experiments. has not only been able to successfully 21 pages. Available online at: http://solmar.nurc. nato.int/ (accessed May 21, 2008). conduct its program of work, but the SACLANT Undersea Research Centre. 1998. INTERNATIONAL Centre is also now recognized as a leader Bioacoustics Panel and Marine Mammal COLLABORATION in this area and a good steward of the Environmental and Mitigation Procedures Panel. A. D’Amico, ed. NATO Undersea Research Centre, The proactive approach adopted by environment. Italy, SACLANTCEN M-133, 84 pp. NURC includes engaging academia, SACLANT Undersea Research Centre. 2000. Scientific ACKNOWLEDGEMENTS Programme of Work 2001. NATO Undersea government, and environmental organi- Research Centre, Italy, 98 pp. zations into its MMRM program. Over All approaches to the animals for tagging SACLANT Undersea Research Centre. 2001. Annual 20 participants from nine countries have were made following the conditions Progress Report 2001. NATO Undersea Research Centre, Italy, 140 pp. contributed resources including soft- of NMFS Scientific Research Permit SACLANT Undersea Research Centre. 2002. Annual ware, technologies, data, and personnel 981-1578-00. This research was approved Progress Report 2002. NATO Undersea Research Centre, Italy, AE-2001, 145 pp. to the planning, execution, and analyses by the Woods Hole Oceanographic SACLANT Undersea Research Centre. 2004. Annual of sea trials. In 2005 and 2007, NURC Institution Animal Care and Use Progress Report 2003. NATO Undersea Research Centre, Italy, AE-2003, 211 pp. organized two intergovernmental confer- Committee. Tyack, P., and W.M.X Zimmer. 2002. Presentation at ences entitled “The Effects of Sound in the Undersea Defence Technology Europe confer- the Ocean on Marine Mammals.” Over REFERENCES ence, Italy, June 19, 2002. Zimmer, W.M.X., M.P. Johnson, A. D’Amico, and P.L. 175 individuals from 11 countries par- Azzellino, A., J.F. Borsani, A. D’Amico, D. Demer, D. McGehee, N. Portunato, and V. Teloni. 2001. A Tyack. 2003. Combining data from a multisensory ticipated in these events, which provided G.I.S. integrated database to investigate cetacean tag and passive sonar to determine the diving behavior of a sperm whale (Physeter macrocepha- a forum for military personnel and distribution in the Ligurian Sanctuary. (Sirena ‘99 and ‘00 operations). In Abstracts of the 14th lus). IEEE Journal of Oceanic Engineering 28:13–29. government-funded scientists to discuss Biennial Conference on the Biology of Marine Zimmer, W.M.X., M.P. Johnson, P.T. Madsen, and P.L. the status of current research regarding Mammals, Vancouver, Canada, November Tyack. 2005a. Echolocation clicks of free-ranging 28–December 3, 2001. Cuvier’s beaked whales (Ziphius cavirostris). the potential impacts of sonar on marine D’Amico, A., F. Mineur, A. Alvarez, C. Mori, M. Journal of the Acoustical Society of America. mammals. In addition, current mitiga- Podestà, N. Portunato, M. Ballardini, and B. 117:3,919–3,927. Zimmer, W.M.X., P.L. Tyack, M.P. Johnson, and P.T. tion methods employed by national Nani. 2001. Oceanographic correlations with the distribution of Cuvier’s Beaked Whales (Ziphius Madsen. 2005b. Three-dimensional beam pattern navies were shared. These conferences cavirostris) in the Ligurian Sea. Poster presented of regular sperm whale clicks confirms bent-horn provided opportunities for discussion of at the 14th Biennial Conference on the Biology of hypothesis. Journal of the Acoustical Society of Marine Mammals, Vancouver, Canada, November America 117:1,473–1,485. coordinated efforts and future collabora- 28–December 3, 2001. tions to ensure that maximum benefit D’Amico, A., A. Azellino, C. Misic, M. Podesta, M. Carron, and N. Portunato. 2003a. Presentation at results from limited research funding the Symposium on Environmental Consequences and to eliminate duplicative efforts. of Underwater Sound, San Antonio, Texas, May 12–16, 2003. D’Amico, A., A. Bergamasco, P. Zanasca, S. Carniel, CONCLUSION E. Nacini, N. Portunato, V. Teloni, C. Mori, and R. NURC has maintained a robust science Barbanti. 2003b. Qualitative correlation of marine mammals with physical and biological parameters program for the last 10 years designed in the Ligurian Sea. IEEE Journal of Oceanic to address the topic of marine mammals Engineering 28:29–43. Frantzis, A. 1998. Does acoustic testing strand whales? and anthropogenic noise. The SOLMAR Nature 392:29. program and the following MMRM Johnson, M.P., and P.L. Tyack. 2003. A digital acoustic

Oceanography June 2008 43