Draft Protocol for conducting a communication experiment between a captive killer whale and free ranging killer whales of the same natal community.

Prepared by Howard Garrett, Orca Network,
and Kenneth C. Balcomb, III, Center for Whale Research

This communication experiment is designed primarily to investigate behavioral context and responses of killer whales in captive and free-ranging populations to vocalizations which are produced and relayed in real-time between them. This will be an initial attempt to determine how and what these highly vocal animals might be saying to each other. It is a well established fact that killer whales (Orcinus orca) produce underwater calls (Ford, 1987; Moore, et. al, 1988), and that these calls can be used by researchers for differentiating populations, i.e., that distinct sets of calls are used by distinct populations. It is presumed that the calls have some meaning for the whales (Adler, 1996). Ford (1987) noted that there were variations in the vocal repertoires of specific matrilineal associations (pods) and termed them "dialects." He found that in killer whale pods of the Pacific Northwest of North America these dialects consist of a mean of 10.7 discrete calls/pod (range 7-17, n=16). In addition there is a wide range of "abberant calls" that occur during times of intense socializing. There are consistent and often striking differences in the call repertoires of different pods that are highly stable over time. The calls are presumably communicative, as distinguished from pulsed broadband vocalizations which are used for echolocation. Nobody knows what the killer whales within a particular dialect are saying to one another, or whether there is communication between pods with different dialects; but, with the availability of the captive killer whale known as "Lolita" performing daily at the Miami Seaquarium, from a known pod with a known dialect, there is a fascinating opportunity for scientific enquiry into their communications abilities.

Haida, a male killer whale that was captured from J or L pod of the Southern Resident community (Lolita's community) in October 1968 and held in Victoria, British Columbia in a netted area adjadent to the waters where he was captured, would respond vocally and excitedly to calls from whales of the Southern Resident community whenever they swam by his sea pen ten years after his capture. Haida was estimated six years of age at capture and would therefore have been in his mid to late teens when he was communicating with his presumed relatives. Ford (pers. comm.) has stated that killer whales that were old enough when captured to have learned their pod's vocal repertoire will communicate with pod members if offered the opportunity after prolonged captivity.

Our proposed acoustic experiment will examine whether Lolita, captured from the Southern Resident community in 1970, will communicate via state of the art technological devices with her community that continues to inhabit the waters of the Pacific Northwest. If so, we have a unique opportunity to evaluate acoustic details which could be very illuminating for exploring whale communication theory. This offers a very straightforward and exciting opportunity to learn how whales talk, and with concurrent rigorous behavioral studies we may even learn a bit about the meanings of particular discrete calls.


A hydrophone (underwater microphone) and a speaker would be placed in the water in Lolita's tank and calibrated to broadcast at volumes commensurate with her normal hearing range. Eight speakers and eight hydrophones would be placed on floating, anchored sonobouys along about ten miles off the west and north sides of San Juan Island in Washington State, in the primary habitat of the Southern Resident community of orcas (Lolita's extended family), currently numbering approximately 85 orcas made up of three pods. Transmission, analysis and recording devices, both acoustic and video, would be present in both locations.

Battery life per sonobouy is about eight hours, so experimental sessions could take place during eight hour periods. Flexibility in timing of sessions may be needed according to behavioral cues from Lolita and the intensity of any interactions, to avoid interrupting intense acoustic sessions. Acoustic interactions could be broadcast over park loudspeakers for visitors and the experiment could be interpreted for their education and enjoyment.

In scientific method, it is desirable to form hypotheses and then test them in ways that lead to unambiguous results. The following is a first draft of a protocol that might be developed.

HYPOTHESIS #1 to be tested concerns whether or not captive killer whales from known pods recognize and respond differentially to dialects from their own pod versus other pods.

The NULL HYPOTHESIS is that captive killer whales from a particular pod do not respond to that particular pod's vocalizations any differently than they would to vocalizations of unrelated pods, or other acoustic stimuli. To test this, the captive whale (Lolita) would be exposed to an underwater speaker. The speaker would play the following sounds at comparable amplitudes in random order for a series of short trials: test tones, trainer's voice, music, white noise, ambient noise amplified, and real-time transmissions of vocalizations from the wild community. Spontaneous responses would be monitored and recorded.

To further test for recognition of Lolita's natal pod, sessions could include:

1) a vocalization episode from a totally unrelated pod (e.g., since Lolita is believed to be from the L-25 subpod, a real-time vocalization episode from a Northern Resident pod (Bigg, et. al., 1982)) could be transmitted , while simultaneously videotaping behaviors of both captive and wild whales.

2) a vocalization episode from a related, but different pod, e.g.: a real-time vocalization episode could be transmitted from a different Southern Resident pod (e.g., J pod), while simultaneously videotaping behaviors of both captive and wild whales.

3) a vocalization episode from the pod in which Lolita was born, i.e., L-25 subpod, while simultaneously videotaping behaviors of both captive and wild whales.

Responses would be evaluated from the videotape and audiotape recordings of the test episodes with respect to type and intensity of observed behaviors, and duration, type and amplitude of vocalizations produced. In a series of test episodes, the responses would be evaluated with respect to whether there was any significantly different or measurable response to any of the stimuli presented. Statistical tests would be done to determine significance, and the null hypothesis would be rejected if there were significant difference in responses.

HYPOTHESIS #2 to be tested concerns whether or not whales in captivity and whales in the wild will utilize a transducer/electronic link for communications between them, such as humans use in voice communications over the telephone.

The null hypothesis is that captive and wild whales will not utilize a transducer/electronic link for communications between them. This would be tested by making the link available at random intervals for random periods and observing whether the whales respond and use the link more often than would be expected by chance. Responses would be evaluated from videotape and audiotape recordings of test episodes with respect to whether "normal" behaviors would be interrupted or modified to utilize the communications link. The null hypothesis would be rejected if the whales utilized the communications link significantly more than chance would predict.

Other hypotheses could be tested to learn more about the mysterious acoustic abilities of these animals, but it is a moot point if the holders of captive animals of this species will not consider implementation of any such protocol because of discomfort about pressures to release their animals back to the wild. In North American parks which hold killer whales, there are only two animals surviving whose natural families have been well studied (Lolita, at the Miami Seaquarium and Corky, at San Diego Sea World). It would be relatively simple to provide a followup research program to evaluate the success of any release scenario. We would be provided a further opportunity to learn a great deal of information about the potential for reintroduction of individuals back into their native population, about the strength of social bonds, and perhaps about social structures of the natural population.

We want to emphasize that this proposed experiment has been developed because we have a fascinating opportunity to critically examine the communications capabilities of a highly evolved marine mammal species. This opportunity is unfortunately being lost because of discomfort with the concept of reintroduction. The hypotheses guiding this experiment are sound, the technology is available to fully exploit the idea, and there remain two whales alive who are uniquely suited to explore the idea. Pursuit of this idea could offer a fascinating opportunity for exhibit development which could pave the way into the future for marine park exhibits. There is tremendous business and promotional potential for the marine park owner that is first to take the opportunity to examine the state of the art in science and technology available for display.

Equipment required:

9 Lubell Model 3300 Underwater Speaker sets
Frequency Range 200-10,000 Hz
Output to 85db/1 ubar/1 yard

9 Federal PA300 Amplifier sets
Audio Frequency Range 300-10,000 Hz
Voltage output 24-volts rms
Input Sensitivity and Impedence 0.55 volts ems, 2000 ohms

9 Custom hydrophone sets with preamplification Frequency Range 50-15,000 Hz

8 sonobuoy moorings, with anchors and caution flag

Many SSQ57B sonobuoys (<10 kHz) broadcasting on VHF

2 TEAC Tascam recorders

PowerMac G3 Powerbook computer with custom spectogram analysis program

9 anti-aliasing filters/amplifiers

2 Sony portable DAT recorders

6 Sony digital video camcorders, NTSC, EIA standards
stereo audio input -7.5dBs, impedence >47 kilohms

two digital cellular telephones to be used for human voice communications

two Magnavox Nav-com satellite uplinks to be used for whale communications

An overall system frequency response 300-10,000 Hz will be utilized for purposes of conducting this experiment. Killer whales have an extensive repertoire of sounds, with most phonations in the range of 4-5 kHz (Ford, 1984). On the hearing side, Hall and Johnson (1971) presented an audiogram of Orcinus with a sensitivity range from 500 Hz to 31 kHz. In Pacific Northwest Killer whales, most of the discrete calls have primary acoustic energy between 1-6 kHz, with high frequency components occasionally extending to > 30kHz (Ford, 1987). We consider the proposed system frequency response will be sufficient to evaluate the hypotheses presented, however if it seems necessary an improved system frequency response to the known limits of hearing of Orcinus can be developed using additional customized equipment.

(Relevant abstracts included)

Adler, T., 1996. Whale chatter - making sense of marine mammals' clicks and calls. Science News. 149: 328-330.

Au, W.W.L. and D.L. Herzing, 1998. Real-time acquisition of echolocation signals by wild Atlantic spotted dolphin (Stenella frontalis) utilizing hydrophone arrays with simultaneous underwater video. 28 Nov 97. Technical Report.

Awbrey, F.T., J.A. Thomas, W.F. Evans and S. Leatherwood, 1982. Ross Sea killer whale vocalizations: preliminary description and comparison with those of sonic Northern Hemisphere killer whales. Rep. Int. Whal. Commn. 32: 667-670.

Bain, D.E., 1986. Acoustic behaviour of Orcinus orca: sequences, periodicity, behavioural correlates and an automated technique for call classification. In Behavioral Biology of Killer Whales, B.C. Kirkevold and J.S. Lockhart, eds. Alan R. Liss: New York. pp. 335-371.

Baker, M.C., 1982. Vocal dialect recognition and population genetic consequences. Amer. Zool. 22: 361-369.

Balcomb, K.C., J.R. Boran and S.L. Heimlich, 1982. Killer whales in greater Puget Sound. Rep. Int. Whal. Commn. 32: 681-686.

Becker, P.H., 1982. The coding of species-specific characteristics in bird sounds. In D.F. Kroodsma and E.H. Miller (Eds.), Acoustic Communication in Birds. Academic Press, New York, p. 213.

Bigg, M..A.,. 1982. An assessment of killer whale (Orcinus orca) stocks off Vancouver Island, British Columbia. Rep. Int. Whal. Commn. 32: 655-666.

-, M.A., P.F. Olesiuk, G.E. Ellis, and K.C. Balcomb, 1990. Social organization and genealogy of resident killer whales (Orcinus orca) in the coastal waters of British Columbia and Washington State. In: Hammond, et. al. (eds.) Individual recognition of cetaceans: Use of photo-identification and other techniques to estimate population parameters. Int. Whal. Commn. Special Issue 12, Cambridge.

-, M.A., G.E. Ellis, J.K.B. Ford, and K.C. Balcomb, 1987. Killer Whales, a study of their genealogy and natural history in the Pacific Northwest. Phantom Press.

Bowles, A.E., W.G. Young ad E.D. Asper, 1988. Ontogeny of stereotyped calling of a killer whale calf, Orcinus orca, during her first year. In Sigurjonsson and Leatherwood [Eds], North Atlantic killer whales, Rit Fisk. XI: 251-275. Reykjavik.

Byrne, H.W., 1982. Primate vocalizations: structural and functional approaches to understanding. Behaviour 80: 241-285.

Caldwell, M.C., and D.K. Caldwell. (1979). The whistle of the Atlantic bottlenosed dolphin (Tursiops truncatus): Ontogeny. In H.E. Winn and B.L. Olla (eds.) Behavior of Marine Animals. Plenum, 3: 369-410.

Caldwell, M.C., D.K. Caldwell, and P.L. Tyack, 1990. Review of the signature-whistle hypothesis for the Atlantic bottlenose dolphin. In S. Leatherwood and R.R. Reeves, (eds.), The Bottlenose Dolphin, Adacemic Press, pp. 199-234.

Conner, D.A., 1980. Dialects versus geographic variation in mammalian vocalization. Animal Behavior. 30: 297-298.

Dalheim and Heyning, in press. Killer Whale. In Ridgeway and Harrison [Eds] Handbook of Marine Mammals vol 6., Academic Press, London

Deecke, V.B., Ford, J.K.B., Spong, P., 1999. Quantifying complex patterns of bioacoustic variation: Use of a neural network to compare killer whale (Orcinus orca) dialects. Jour. Acoustical Soc. of Amer. 105(4): 2499-2507.

Abstract: A quantitative measure of acoustic similarity is crucial to any study comparing vocalizations of different species, social groups, or individuals. The goal of this study was to develop a method of extracting frequency contours from recordings of pulsed vocalizations and to test a non-linear index of acoustic similarity based on the error of an artificial neural network at classifying them. Since the performance of neural networks depends on the amount of consistent variation in the training data, this technique can be used to assess such variation from samples of acoustic signals. The frequency contour extraction and the neural network index were tested on samples of one call type shared by 9 social groups of killer whales. For comparison, call similarity was judged by 3 human subjects in pairwise classification tasks. The results showed a significant correlation between the neural network index and the similarity ratings by the subjects. Both measures of acoustic similarity were significantly correlated with the groups' association patterns, indicating that both methods of quantifying acoustic similarity are biologically meaningful. An index based on neural network analysis therefore represents an objective and repeatable means of measuring acoustic similarity, and allows comparison of results across studies, species, and time.calls closer to that of the adult repertoire. At the same time the pulse duration and repetition rate begin to more closely approximate those of adultsinformation to support or refute the findings of bioacoustical studies.

Ellis, G., C.J. Matkin, J.D. Hall and S. Leatherwood, 1999. Killer whales (Orcinus orca) photoidentified in Prince William Sound, Alaska. l976-1987, Can. Field Nat.

Evans. WE., A.V. Yablokov and A.F. Bowles, 1982. (Geographic variation in the color pattern of killer whales (Orcinus orca). Rep. Int. Whal. Commn 32: 687-694.

Ford, J.K.B., 1984. Call traditions and dialects of killer whales (Orcinus orca) in British Columbia. Ph.D. Thesis. University of British Columbia. Vancouver. B.C., 435 pp.

-, 1991. Family fugues. Natural History. pp. 68-76

-, 1987. A Catalogue of underwater calls produced by killer whales (Orcinus orca) in British Columbia. Can. Data Rep. Fish. Aquat. Sci. 663: 165 pp.

-, and H. D. Fisher, 1982. Killer whale (Orcinus orca) dialects as an indicator of stocks in British Columbia. Rep. Int. Whal. Commn. 32: 671-680.

-, and H. D. Fisher, 1983. Group-specific dialects of killer whales (Orcinus orca) in British Columbia. In R. Payne (Ed.), Communication and Behaviour of Whales. AAAS Selected Symposia Series 76, Boulder, CO. Westview Press. pp. 129-161.

-, JK.B., G. Ellis and K.C. Balcomb, 1999. Killer Whales, a study of their genealogy and natural history in the Pacific Northwest. UBC Press, Vancouver, Canada.

-, J.K.B. 1991. Vocal traditions among resident killer whales (Orcinus orca) of British Columbia. Can. Jour. of Zoo. 69: 1454-1483.

-, J.K.B., G.M. Ellis, L.G. Barrett-Lennard, A.B. Morton, R.S. Palm, and K.C. Balcomb III, 1999. Dietary specialization in two sympatric populations of killer whales (Orcinus orca) in coastal British Columbia and adjacent waters. Can. Jour. Zoo. 76: 1456-1471.

Abstract: Two forms of killer whale (Orcinus orca), resident and transient, occur sympatrically in coastal waters of British Columbia, Washington State, and southeastern Alaska. The two forms do not mix, and differ in seasonal distribution, social structure, and behaviour. These distinctions have been attributed to apparent differences in diet, although no comprehensive comparative analysis of the diets of the two forms has been undertaken. Here we present such an analysis, based on field observations of predation and on the stomach contents of stranded killer whales collected over a 20-year period. In total, 22 species of fish and 1 species of squid were documented in the diet of resident-type killer whales; 12 of these are previously unrecorded as prey of O. orca. Despite the diversity of fish species taken, resident whales have a clear preference for salmon prey. In field observations of feeding, 96% of fish taken were salmonids. Six species of salmonids were identified from prey fragments, with chinook salmon (Oncorhynchus tshawytscha) being the most common. The stomach contents of stranded residents also indicated a preference for chinook salmon. On rare occasions, resident whales were seen to harass marine mammals, but no kills were confirmed and no mammalian remains were found in the stomachs of stranded residents.

Transient killer whales were observed to prey only on pinnipeds, cetaceans, and seabirds. Six mammal species were taken, with over half of observed attacks involving harbour seals (Phoca vitulina). Seabirds do not appear to represent a significant prey resource. This study thus reveals the existence of strikingly divergent prey preferences of resident and transient killer whales, which are reflected in distinctive foraging strategies and related sociobiological traits of these sympatric populations.

Herman, L.M., R.K. Uyeyama, 1999. The dolphin's grammatical competency: Comments on Kako. Animal Learning & Behavior. 27(1): 18-23

Abstract: Kako (1999) reviews the evidence for syntactic competencies in several animal species exposed to artificial language systems, emphasizing the importance of core syntactic properties such as argument structure and closed-class items. We present evidence from our dolphin studies for the acquisition of the closed-class functionality of demonstratives, prepositions, conjunctions, and locatives. Sensitivity to argument structure is also evidenced by wholly untrained and consistent interpretations of the dolphin to probes of anomalous syntactic structures. These results are generated within our comprehension-based paradigm, which enables us to provide convincing objective evidence for the development and generalization of concepts by the dolphin subject. Demonstrations of animal language competencies may illuminate certain aspects of human linguistic competence by suggesting that the particular modeled subsets can derive from general cognitive mechanisms, rather than language-specific ones.

Hoelzel, A.R. and C.A. Dover, 1987. Molecular techniques for examining genetic variation and stock identity in cetacean species. Int. Whal. Commn, Scientific Comm., April 1987, 66 pp. (unpubl.).

Hoyt, E., 1990. Orca, the whale called killer. Camden House, Camden East, Ontario.

Janik, V.M., P.J.B. Slater 1998. Context-specific use suggests that bottlenose dolphin signature whistles are cohesion calls. Animal Behavior. 57(5): 1173-1173

Leatherwood, S., K.C. Balcomb, C.O. Matkin and G. Ellis, 1984a. Killer whales (Orcinus orca) of South Alaska. Results of field research 1984. Preliminary report. San Diego, Hubbs-Sea World Research Institute Tech. Rep. 84-175. 59 pp.

-, K.C. Balcomb, C.O. Matkin, C. Ellis. and J.D. Hall 1986. Killer whales (Orcinus orca) of south Alaska. Results of field research 1984. Preliminary Report. Rep. Int. Whal. Commn. 37: 504-505 (abstract).

Lyrholm. T., 1985. A photographic and acoustic pilot study of killer whales (Orcinus orca) off the coast of Norway. Thesis. University of Stockholm. 51 pp. (unpubl.).

-, 1988. Photoidentification of individual killer whales. Orcinus orca, off the coast of Norway. 1983-1986. In Sigurjonsson and Leatherwood [Eds], North Atlantic killer whales, Rit Fisk. XI: 89-94. Reykjavik.

-, S. Leatherwood, and J. Sigurjonsson, 1987. Photoidentification of killer whales (Orcinus orca) off Iceland. October 1985. Cetology, 52: 1-14.

Marler, P., 1976. Social organization, communication and graded signals: the chimpanzee and the gorilla. In P. P. G. Bateson and R. A. Hinde (Eds.), Growing points in ecology. Cambridge Univ. Press, Cambridge. pp. 239-280.

-, and R. Tenaza, 1977. Signaling behaviour of apes. In T.A. Sebeok (Ed.), How animals communicate. Indiana Univ. Press, Bloomington. pp. 969-1033.

Matkin, C., 1994. The Killer Whales of Prince William Sound. Prince William Sound Books, Valdez, AK.

McCowan B., and D. Reiss, 1995. Whistle contour development in captive-born infant bottlenose dolphins (Tursiops truncatus): Role of learning. Jour. of Comp. Psych. 109 (3): 242-260.

Miller, PJ; Tyack, PL., 1998. A small towed beamforming array to identify vocalizing resident killer whales (Orcinus orca) concurrent with focal behavioral observations. Deep-sea Research Part II-Topical Studies on Oceanography, 45:7 p1389-1405.

Moore, S.E., J.K. Francine, A.E. Bowles, J.K.F. Ford, 1988. Ontogeny of calls of killer whales Orcinus orca, from Iceland and Norway. In J. Sigurjonsson and S. Leatherwood (eds.), North Atlantic killer whales, Rit Fisk. XI:15-21. Reykjavik.

Morton, A.B., J.C. Gale and R.C. Prince, 1986. Sound and behavioural correlations in captive Orcinus orca. In B.C. Kirkvold and J.S. Lockhart (eds.), Behavioural Biology of Killer Whales, Alan R. Liss, pp. 303-333.

Murray, S.O., Mercado, E., & Roitblat, H.L., 1998. Characterizing the graded structure of false killer whale (Pseudorca crassidens) vocalizations. Journal of the Acoustical Society of America, 104, pp. 1679-1688.

Abstract: The vocalizations from two captive false killer whales (Pseudorca crassidens) were analyzed. The structure of the vocalizations was best modeled as lying along a continuum with trains of discrete, exponentially damped sinusoidal pulses at one end and continuous sinusoidal signals at the other end. Pulse trains were graded as a function of the interval between pulses where the minimum interval between pulses could be zero milliseconds. The transition from a pulse train with no inter-pulse interval to a whistle could be modeled by gradations in the degree of damping. There were many examples of vocalizations that were gradually modulated from pulse trains to whistles. There were also vocalizations that showed rapid shifts in signal type, for example, switching immediately from a whistle to a pulse train. These data have implications when considering both the possible function(s) of the vocalizations and the potential sound production mechanism(s). A short-time duty cycle measure was developed to characterize the graded structure of the vocalizations. A random sample of 500 vocalizations was characterized by combining the duty cycle measure with peak frequency measurements. The analysis method proved to be an effective metric for describing the graded structure of false killer whale vocalizations.

Murray, S.O., Mercado, E., & Roitblat, H.L., 1998. The neural network classification of false killer whale (Pseudorca crassidens) vocalizations. Journal of the Acoustical Society of America, 104, pp. 3626-3633.

Olesiuk, P.F., M.A. Bigg and G.M. Ellis, 1990. Life history and population dynamics of resident killer whales (Orcinus orca) in the coastal waters of British Columbia and Washington State. In: Hammond, et. al. (eds.): Individual recognition of cetaceans: Use of photo-identification and other techniques to estimate population parameters. In: Hammond, et. al. (eds.), Individual recognition of cetaceans: Use of photo-identification and other techniques to estimate population parameters Rep. Int. Whal. Commn. (Special Issue 12), Cambridge. pp. 209-243.

Overstrom, N.A., 1983. Association between burst-pulse sounds and aggressive behavior in captive bottlenosed dolphins (Tursiops truncatus). Zoo. Biol. 2: 93-103.

Perrin. W.F. (Ed.), 1982. Report of the workshop on identity, structure and vital rates of killer whale populations. Rep. Int. Whal. Commn. 32: 617-631.

Tyack, P., 1990. Use of a telemetry device to identify which dolphin produces a sound. In K. Pryor and K. Norris, (eds.), Dolphin societies. Discoveries and puzzles. Univ. of Calif. Press, pp. 319-344.




©Orca Network
Please contact
Orca Network to inquire
about educational use of any materials on this site.