Whistles as potential indicators of stress in bottlenose dolphins
By H. Carter Esch, MSc student and Laela S. Sayigh, PhD, University of North Carolina, Wilmington

The welfare of an organism, or the state of an animal in relation to how it attempts to cope with its environment, is often related to the stress that it experiences (Broom and Johnson 1993). The diversity and limited observation of stress responses among marine mammals makes it difficult to develop a comprehensive diagnostic protocol. Typically, stress hormone profiles are produced from blood samples drawn from restrained animals (stranded, temporary capture-release). Behavioral measures are becoming an increasingly valuable component in the evaluation of mental and physical state, particularly when used in combination with quantitative physiological measurements (Frohoff et al. 2004). A systematic methodology for identifying and interpreting behavioral symptoms of stress does not exist for marine mammals. One potential approach might include acoustic monitoring of vocalization rates. Marler et al. (1992) found that the intensity and structure of vocalizations may vary throughout a period of separation, and that these changes may reflect an altered motivational state. The current project is designed to quantify vocal rates and whistle parameters (maximum and minimum frequency, number of loops, and loop duration) of bottlenose dolphins in a variety of contexts. We predict that (a) whistle rates and number of loops will be greater at the beginning of a capture-release session than at the end; (b) whistle rates and number of loops will be greater during an individual’s first capture-release session than during later capture-release sessions; (c) whistle rates and number of loops will be greater when a mother is caught with a dependent calf than without a dependent calf; and (d) whistle rate and number of loops will be greater during capture-release than in normal, free-ranging conditions. Selected recordings made during brief capture-release events near Sarasota, FL (from 1975-2005) were analyzed using Avisoft sound analysis software. Vocal rates were calculated and whistle parameter measurements were measured for at least 20 whistles from each dolphin; sample size was hypothesis dependent.

Preliminary results indicate that, within a single capture-release session, vocal rates appear to be correlated with location (in water versus on the veterinary examination boat). There does not appear to be a significant decrease in vocal rate that can be associated with habituation during a single capture-release event. No clear patterns have emerged indicating that vocal rates are depressed in dolphins that have prior capture-release experience. Females vocalize significantly more when captured with a calf than without a calf. Finally, vocal rates are significantly higher during capture-release events than in undisturbed conditions. Whistle parameter analyses are ongoing.

Evaluating stress responses in bottlenose dolphins is useful in a variety of contexts including live stranded, temporarily captured, captive, and free-swimming animals. The development of an effective tool with which to evaluate stress in bottlenose dolphins, that did not require capture-release, would allow for quicker, more efficient assessments of animals that may be at risk. Reliable indicators of stress could be used to monitor dolphins exposed to stressors such as anthropogenic noise. This project was funded by Harbor Branch Oceanographic Institute, with additional support from Dolphin Quest, NOAA Fisheries, and Disney’s Animal Programs.

References:

Broom, D.M. and K.G. Johnson. 1993. Stress and Animal Welfare. Chapman and Hall, London, UK.

Frohoff, T.G. 2004. Stress in Dolphins. Pages 1158-1164 in Encyclopedia of Animal Behavior (Marc Bekoff, Ed.). Greenwood Press, Westport, Connecticut. 1274 pp.

Marler, P., C.S. Evans, and M.D. Hauser. 1992. Animal signals: motivational referential, or both? In:Papousek, H., Jurgens, U., and Papousek, M. (Eds), Nonverbal Vocal Communication. Cambridge: Cambridge University Press, p. 66-86.