Peter L. Tyack, Douglas P. Nowacek, and Mark P. Johnson, Co-Principal Investigators
The goal of this work is to establish factors relating to acoustic environment and behavior that increase the risk of ship collision with right whales. The experiment combiped a new non-invasive acoustic recording tag with controlled sound playbacks to establish the response of right whales to the sound of approaching vessels as a function of behavioral context. We have completed 1 year of a projected 3-year project with funding from CICOR / U.S. National Marine Fisheries Service. In year 1 we built , and calibrated 4 digital acoustic recording tags (DTAGs ); and conducted a 4-week field experiment with the tags. We are currently analyzing the resulting data. The tags built and fielded performed according to specifications, and we successfully completed year 1 of our field effort in the Bay of Fundy (BF). The Hannah T successfully attached the tags (Figure 1), and our suction cup attachment (Figure 1 ) attained acceptable attachment durations for our work (Table 1 ). We are currently analyzing data collected by the DTAG and the concurrent behavioral observations conducted from the observation vessel (OV). We were in the field in the BF from 22 July 2000 -20 August 2000, conducted field operations during 12 of those days , attached 5 tags of ≥ 1 hour longevity, and conducted four playbacks to three different tagged whales. More tags than expected were knocked off during social activity. We feel strongly that it is acceptable to have tags knocked off by normal inter-whale activity, and tags more firmly or invasively attached could have caused injury in the same situation. We remain committed to our non-invasive tagging approach, although due to the knock-off rate we may need to attach more tags than initially anticipated to obtain the necessary sample size . With the DTAG we successfully documented subtle sub-surface responses to our playbacks of naturally occurring sounds, data previously unobtainable for free-ranging cetaceans. Other studies have recorded acoustic reactions to playbacks (1), but the DTAG recorded behavioral changes (e.g. heading) specifically in response to our playback of whale sounds as well to sounds produced by other whales in the area. While we have not yet found any response to vessel playbacks or approaches , establishing that our tag can document subtle responses to experimental conditions is a very significant result . The basic goal of the entire project Is to determine the conditions under which whales do not produce an effective avoidance response to oncoming vessels, so the lack of responses to vessel approaches is consistent with the problem we are investigating.
Work we have completed using the results from tagging efforts in the BF include: i) analyses of the swimming and diving behavior of foraging right whales ; ii} auditing and analyses of acoustic data 1ecorded by the DTAGs; iii ) assembling identification photos for the members of the New England Aquarium Right Whale Research Program, who are in the process of identifying the whales we tagged in 2000; iv) comparison of locomotor behavioral data recorded by the tag with that documented by observers on the OV; and v) analyses of vocalization patterns of focal (i.e., tagged) vs. non-focal whales in collaboration with the International Fund for Animal Welfare (IFAW) to assess the feasibility of an acoustic detection and localization system to passively track movements of right whales in high risk areas.
SUMMARY OF RESULTS
A total of 15 right whales were approached for suction-cup tagging, and tags were attached to 9 (^0°Io) of those animals (Table 1). All tags deployed functioned without
fail, and all were successfully recovered. Tagged whales were tracked on the eastern side of and east of the `whale sanctuary' that lies to the southeast of Grand Manan, New Brunswick. Some areas the whales used overlapped with the outbound shipping lane from St. John, NS, part of which is contained in the `whale sanctuary'. Tracks of tagged whales can be easily accessed through our website: http:/ldtag.whoi.edu. Tags were deployed by the Hannah T with assistance from Michael Moore's blubber thickness measurement team, which reduced the number of boats operating around the whales. Unfortunately, neither Moore's team nor we feel that this mode of operation is tenable in future years due to the amount of time required to attach tags. In past years each tag deployment required <30 minutes, but due to a different pattern of whale behavior this task required up to 4x as long in 2000. This increase in time required for tagging and blubber thickness measurements necessitates that we have our own boat devoted to tagging. This boat will be active only to attach tags, and after tagging will be brought on board a larger ship or towed.
Swimming and Diving Behavior of Focal Whales
The high-speed orientation sensors on the DTAG provide a detailed view of sub-surface behavior previously unavailable. The 3-axis accelerometers in the DTAG, for example, record the pitch, roll, and yaw of the focal whale. In addition, at the position of tag attachment to the whale, dorsal and approximately midway between the blowholes and caudal peduncle, the pitch angle of the animal contains a significant signal due to undulations of the body during swimming and so can provide an indication of when the animal is stroking as well as the stroke rate (Fig. 2) (5). Consequently, while a DTAG is attached to a focal whale it is simultaneously recording the animal's acoustic environment, its pitch, roll, yaw, heading, and fluke stroke rate as well as the depth and temperature of water in which the whale is swimming. Moreover, the sensor sampling rate (> 20 Hz) and resolution are sufficient to record fast or subtle movements of the whale, e.g. transient direction changes or cessation of swimming. In Figure 2, a typical dive from a whale in the BF data set illustrates how fluke stroke rate and gliding can be determined from the pitch record. Other swimming behavior of interest to our project is a whale's heading, and we have been able to successfully ground-truth the magnetometer readings from the tag with sightings from the OV (Figure 3).
Responses to playbacks and other sounds
Social activity of North Atlantic Right Whales in the BF has been documented (3), and southern right whales have been observed to respond to playbacks of sounds made by conspecifics (6, 7). We decided to use social sounds as a positive control for our playback experiments, i.e. we expected the whales to show some acoustic or motor response to these sounds. We did, in fact, record a motor response from focal whales to our playbacks (Fig. 3), and moreover we recorded a similar motor response from focal whales when the sounds of nearby whales (i.e., not our playback) occurred as recorded by the DTAG. This result demonstrates our ability to measure a response by a focal whale to a sound exposure, and confirms our choice of natural sounds as a positive control. Having a functioning positive control provides a context within which responses, or lack thereof, to experimental stimuli can be interpreted. In addition, this result constitutes the first documented, sub-surface swimming (i.e. motor behavior) response to controlled sound exposure in afree-ranging cetacean.
Sounds recorded by DTAGs attached to whales
The DTAG acoustic records from the BF contain a broad range of identifiable sounds including those produced by the focal whale and other nearby animals. The tag also recorded the sounds of many vessels. Samples of sounds recorded by the tag can be accessed on our world wide website, http://dta^.whoi.edu. Sounds from the focal whale can be distinguished from non-focal sounds by a combination of three indicators: received level, the absence of reverberation, and the presence of significant high frequency harmonic components. Knowing with confidence that the focal Whale produced a particular sound is important for at least two reasons. The first motive is to chart any acoustic response by the focal to our playbacks or other sounds in the environment. Secondly, we and other researchers would like to know whether the vocal actlvlty of right whales is conducive to passive ac(ltlstlc detection and localization schemes. If feasible, such systems could provide information as to the presence and hopefully location of right whales in areas of high risk for vessel collision. Our results are very important in this pursuit as the DTAG documents three essential types of Information that have direct bearing on the appropriateness of a sound for passive detection and localization: i) the type of sound produced, i.e. some types of sound are easier to detect and localize; ii) the depth at which a wh^lle is swimming when it produces a sound, which has implications as to how accurately a sound can be detected/localized; and iii) the duty cycle of sound production, i.e. a regular, high duty cycle is much more conducive to accurate detections. Figul-e 4 shows tl^e pattern of sound production recorded from a whale tagged on 5 August, 2000.
SUMMARY OF PRODUCTS AND DISSEMINATION OF RESULTS
Results from Year 1 have contributed to 7 conference pl-esentations: 4 at the Right Whale Consortium Meeting in October in Boston, l at the meeting of the Acoustical Society of America in December in Long Beach, CA, and 2 at the meeting of the European Cetacean Society to be held in May 2001. In addition, our results are included in 4 manuscripts: one in preparation comparing behavior recorded by the DTAG with that visible from surface observations; one documenting the positive buoyancy of right whales as evinced in their swimming/diving behavior in press in T1ze Proceediyzgs of the Royal Society B: Biological Sciences; one describing the DTAG architecture, technology, and products submitted to the Journal of Ocean Engineerit7g; and finally a contribution to a paper documenting the vocalization patterns of right whales that is being submitted by IFAW to the Journal of Cetacean Research and Management. Our results are also included in an article published in the Gulf of Maine News in an issue focused on right whale research. In addition to these formal channels of dissemination, we have made and continue to make our data available to the right whale research community including, but not limited to, posting the data to our world wide website: http://dtag.whoi.edu.
FUTURE RESEARCH PLANS
We have applied for funding to continue this project. In the summers of 2001 & 2002 we Hope to tag approximately twice as many whales per year and conduct further playback experiments. Due to the number of tags knocked-off by whale-whale contact, again a loss we are comfortable to assume, we might need to attach more tags than previously planned. As explained above, we will unfortunately be unable to collaborate with the Hannah T again in future years due to the significantly reduced data rate suffered by both projects. We will, however, minimize the number of boats operating around the whales by having the tagging boat operate only when actively attaching tags. In addition, we will use this same boat for the playback experiments, further reducing the total number of boats. operating and transiting near the whales.
1, P. J. 0. Miller, N. Biassoni, A. Samuels, P. Tyack, Nature 405, 903 (2000).
2. M. T. Weinrich, R. H. Lambertsen, C. S. Baker, M. R, Schilling, C. R. Belt, Report to the International Whaling Commission Special Issue 13, 91-97 (1991).
3. S. D. Kraus, J. J. Hatch, Journal of Cetacean Research and Management (in press).
4. J. D. Goodyear, Journal of Wildlife Management 57, 503-513 (1993),
5. R, C. Skrovan, T. M. Williams, P. S. Berry, P. W. Moore, R. W. Davis, The Journal of Experimental Biology 202, 2749-2761(1999).
6. C. W. Clark, in Communication and Behavior of Whales R. Payne, Ed. (Westview Press, Boulder, 1983) pp. 163-198,
7. C. W. Clark, J. M. Clark, Science 207, 663-665 (1980).