ASA Lay Language Papers

2nd Pan-American/Iberian Meeting on Acoustics

Do Killer Whales Influence the Vocal Behavior of Tropical Dolphins?

Shannon Rankin - shannon.rankin@noaa.gov

Eric Archer - eric.archer@noaa.gov

Southwest Fisheries Science Center

3333 N. Torrey Pines Ct.

La Jolla, CA 92037

Julie N. Oswald - oswald.jn@gmail.com

Oceanwide Science Institute

P.O. Box 61692

Honolulu, HI 96839

Popular version of paper 4aAB12

Presented Thursday morning, November 18, 2010

2nd Pan-American/Iberian Meeting on Acoustics, Cancun, Mexico

Killer whales are the main predator of dolphins and porpoise, and it has been suggested that killer whales may eavesdrop on vocalizing dolphin schools as a hunting strategy (Deecke et al., 2005).  Previous researchers have hypothesized that killer whale predation may affect dolphin vocal behavior, and anecdotal reports support this.  For example, when highly vocal beluga whales in Alaska were exposed to playback of killer whale sounds, they moved downriver from the sound and produced very few vocalizations (Fish and Vania, 1971).  Until now, however, little data existed to test the hypothesis that dolphins change their vocalization patterns to avoid detection by eavesdropping killer whales.  Here we analyze data from shipboard surveys in the tropical Pacific Ocean to rigorously test this hypothesis.

Dolphins produce a variety of sounds, including clicks and whistles (Figure 1).  Dolphins use click sounds for echolocation—using sound to ‘see’ their environment.  Echolocation clicks are short broadband sounds that are highly directional.  These high-frequency sounds do not travel great distances.  Dolphins whistles are social calls that are lower in frequency (<30 kHz) and do travel great distances.  The frequencies of greatest hearing sensitivity of killer whales coincide with that of dolphin whistles (Szymanski et al., 1999). 

Figure 1.  Graphical representation of dolphin and porpoise vocalization types, with the range of killer whale hearing sensitivity shown in yellow (greatest sensitivity shown in orange).  Sound types include whistles (red), dolphin clicks (purple), and porpoise clicks (pink).

For the past ten years, we have been conducting shipboard visual and acoustic surveys of dolphins and whales in the Pacific Ocean from Alaska to Peru and from the farthest reaches of the northwest Hawaiian Islands to the coast of Central America.  This is the largest study of dolphin acoustics in the world, on both a temporal and geographic scale.  Based on our research, we know that most dolphins in the tropical Pacific Ocean are frequently encountered in large, vociferous schools that use whistles to communicate over large areas.  Killer whales are relatively rare in the nutrient-poor tropical Pacific Ocean; however they do occur in this area.  We examined our extensive dataset to test the hypothesis that the dolphins in the tropical Pacific Ocean cease vocalizing in the presence of killer whales.

Figure 2.  Map of study area during surveys in tropical and sub-tropical Pacific Ocean.  Dolphin detections are shown as black dots, and killer whale sightings are shown as green triangles.

Data on the vocal activity of dolphin encounters was collected during 39,520 km of visual and acoustic shipboard surveys in the tropical and sub-tropical Pacific Ocean, including 1,035 dolphin detections of 13 species (Figure 2).  For each dolphin encounter, we used a statistical method called a generalized linear model (GLM) to examine the relationship between vocal behavior (vocal/not vocal) and the distance and time to the nearest killer whale detection.  We found that dolphin vocalizations were inversely correlated with the proximity of killer whales in space and time (p< 0.05, Figure 3).  In other words, despite the relative rarity of killer whales in the tropical Pacific Ocean (n=19 for this study), they appear to have an influence on the vocal behavior of nearby dolphin schools.  Those dolphin schools were more likely to be silent. 

Figure 3.  Logistic regression plot showing the relationship between the probability that a dolphin school was vocal and the distance to the nearest killer whale sighting (solid blue), with 95% confidence intervals (dashed blue).  The positive slope of this regression line shows that the probability that a dolphin school is vocal increases with an increased distance from a killer whale sighting.

Many of the dolphin species encountered in the tropical Pacific Ocean are found in large, spread out schools.  Whistles, which can propagate over several kilometers, may be necessary to retain group cohesion.  This type of long distance communication may also serve as a dinner bell to any killer whales in the area, and our results show that these dolphins cease vocalizing in the presence of killer whales.  Due to the relative rarity of killer whales in the tropics, it is unlikely that these infrequent disruptions to communication significantly impact interactions necessary for survival.   

However, in temperate climates, where increased productivity supports a greater abundance of killer whales, this interruption in communication may have a greater impact on survival.  Dolphins in temperate waters may need to develop alternative anti-predator strategies, such as adapting their vocal behavior to avoid detection by killer whales.  In fact, few dolphin species in the temperate Pacific Ocean whistle, and porpoise produce only click sounds which are entirely above the hearing of killer whales.

References

Deecke, V. B., J. K. B. Ford, P. J. B. Slater.  2005.  The vocal behaviour of mammal-eating killer whales: communicating with costly calls.  An. Behav. 69:395-405.

Fish, J. F. and J. S. Vania. 1971.  Killer whale, Orcinus orca, sounds repel white whales, Delphinapterus leucas.  Fish. Bull. 69(3):531-535.

Szymanski, M. D., D. E. Bain, K. Kiehl, S. Pennington, S. Wong, and K. R. Henry.  1999.  Killer whale (Orcinus orca) hearing: Auditory brainstem response and behavioral audiograms.  J. Acoust. Soc. Am. 106(2):1134-1141.