One of the intriguing facts about elephants is that some of their vocalizations are infrasonic, and thus are inaudible to humans. The prominance of very low frequencies in the vocalizations is a defining characteristic of all three species of elephants. Sounds are generally considered to be infrasonic if their frequency is less than 20 Hz (the lower limit of human hearing). Low frequency sounds travel farther than high frequency ones, which make them ideal for long distance communication.
The spectrogram below shows some classic rumble vocalizations by forest elephants.
The frequency (or tone) scale runs from 0 to 420 Hertz (middle C vibrates at about 262 Hz.). The blue line near the bottom is the lower limit of human hearing sensitivity.
[Click to SHOW the frequency range of human voices].
© Melissa Groo
The discovery that elephants use infrasound in communication led from a hunch Katy Payne had when she was observing elephants in the Washington Park Zoo, Portland, Oregon. She was paying attention to their communication, when in addition to the sounds she could hear, she thought she felt, rather than heard, other rumbles. She suspected these were infrasonic rumbles. Further work with William Langbauer, Jr. and Elizabeth Thomas showed that the elephants were indeed making infrasonic calls. Subsequent studies, in association with Joyce Poole, William Langbauer, Cynthia Moss, Russell Charif, Rowan Martin and others, took place in Kenya, Namibia, and Zimbabwe, and led to the conclusion that elephants use their powerful deep calls in long distance communication.
Check out what YOU can hear! Below are pictures of three low frequency sounds, and controls to play them. These sounds were computer-generated and the images show you the wave form of the sounds (the spectrogram would show just a single straight line). Notice the spacing difference - this is the wavelength.
As we consider the size of elephants’ sensory world, the timing as well as the frequency and power of their vocalizations turns out to be important. The propagation of very low frequency sound varies with atmospheric conditions, which change on a diurnal schedule. On a typical dry season evening in the savannah a temperature inversion forms, potentially increasing the listening area of elephants as much as ten-fold -- from 30 km2 at midday to 300 km2 in the same evening (Larom et al. 1997). In light of this fact it is interesting that savannah elephants make most of their loud low-frequency calls during the hours of best sound propagation (Ibid.). We do not know whether this is an innate or opportunistic response to fluctuations in the size of their communication area, but in either case it is clear that as the area shrinks and expands, so does the network of potential associates and mates.
This finding offers a solution to many old mysteries about elephant society, particularly the mystery attending the ability of males to find females for breeding, and the ability of separated family groups to coordinate their patterns of movement for weeks at a time without losing communication or converging on the same scarce resources.
Frequently Asked Questions
What is infrasound?
Infrasound is sound below the level of human hearing. The frequency of a sound is measured in Hertz (Hz) and the infrasonic range is generally considered to be between 1 and 20 Hz.
What sort of frequency spectrum is found in an infrasonic elephant call?
Elephants make a variety of vocalizations including rumbles, screams, and trumpets. Rumbles are low-frequency calls, often falling partially or entirely in the infrasonic range. Most elephant rumbles consist of a fundamental frequency between 5-30Hz with audible harmonics or overtones. With distance, the upper harmonics attenuate at a greater rate than the lower ones. A good working range for capturing elephant rumbles with their harmonics, is 5-250 Hz. The lowest call we have measured for forest elephants was at 5Hz; from savannah elephants, 14Hz.
Is infrasound emitted by all elephants?
Yes. Researchers have recorded infrasonic calls in captive Asian elephants (Payne, Langbauer & Thomas 1986), African savannah elephants (Poole, Payne & Langbauer 1988) and African forest elephants (Elephant Listening Project research in Central and West Africa).
Under what circumstances or conditions do animals emit infrasonic rumbles?
Most elephant rumbles are rich in infrasound; many of these contain frequencies high enough to be audible to humans. Elephants make these calls when coordinating family and larger group behaviors, when competing for resources and/or dominance, and when attracting mates and announcing reproduction (Poole, Payne & Langbauer, 1988). Females with young are most vocal as confirmed by Katy Payneís data from Amboseli park in Kenya and Etosha park in Namibia. The vast majority of infrasonic calling took place in family groups; bull groups were relatively silent. There is, however, still much to be learned about the functions of elephant calls.
How often will an elephant use infrasonic calls?
Females vocalize more frequently than males; the rate of vocalization in both males and females is highly variable and dependent on social circumstances. We know from data gathered in Namibia,the Central African Republic, and Gabon that the number of calls per unit time increases predictably with the number of elephants present (Payne et al, 2003, Thompson et al, 2009, Wrege et al, 2011), but the rate of calling by each elephant remains relatively consistent.
How far do infrasonic elephant calls travel?
The lower the frequency of a sound, the longer its sound wave. Low frequency sounds can therefore travel farther without being absorbed or reflected by the environment. Intensity of elephant calls varies widely from very soft calls made between mothers and their adjacent infants to the very loud calls made by females announcing their availability. Playback experiments demonstrated that savannah elephants responded to each othersí loud vocalizations over distances of at least 2 kilometers. Because playbacks were only broadcast at half the amplitude of the strongest elephant calls in their sample, the authors estimated the actual range as at least 4 kilometers (Langbauer, Payne et al, 1991). The calling area may be expanded by as much as an order of magnitude during temperature inversions in the evening and night (Larom et al, 1997). Preliminary results from our recent work with forest elephants suggest that powerful forest elephant calls can be heard by elephants 7km away through the dense forest. The fact that elephant calls can travel several kilometers enables elephant societies to coordinate movements over large areas.
How do you record infrasonic calls?
In order to pick up calls in the infrasonic range (1-20Hz), you must have equipment that is sensitive to these low sounds (microphone, preamplifier and recording device). Many commercial products are not designed to pick up sound outside of the human hearing range and have a severe roll-off in sensitivity on the low-end of the frequency scale. A technical support person from the productís company can usually provide you with information on component sensitivity. We have used a number of different systems over the years as summarized below.
Autonomous Recording Units
We currently use a specialized unit called an ARU (Autonomous Recording Unit) developed by engineers in the Cornell University Bioacoustics Research Program. It consists of a small microphone mounted on a signal conditioning board that connects to a more generalized filter amplification board. The output from the filter amplification board then feeds into a circuit that converts the analog signal into a digital one and then stores the data to a laptop hard drive. The unit runs off of a car or truck battery (or a few lantern batteries). These units allow us to set the sampling rate as we wish (typical CD-quality sound is recorded at a sampling rate of 44.1kHz). Although high sampling rates capture a wide range of frequencies (e.g. birds and elephants on the same recording) large sampling rates also produce large files. For elephant rumbles, we are only interested in the low frequencies and can therefore sample at a much lower rate for much longer amounts of time. We have successfully sampled at 2000Hz for a continuous unmanned 3 months with these units. ARUs are not commercially available at this time.
For more information about the ARUs developed by engineers in the Cornell University Bioacoustics Research Program, click here.
Follow this link to read about ELP's use of ARUs in the field.
Commercially available digital recording system
For situations where we just want to record for a few minutes or hours at a high-sampling rate, we currently use a digital recording system of off-the-shelf components. One bonus of true digital recording is that the recordings can easily be transferred to a computer for analysis and archiving to CD. If you are considering buying a system to record infrasound, make sure to talk with the providerís technical support staff to make sure all components are sensitive to frequencies in the infrasonic range (1-20Hz).
The products listed below are by no means the only effective components available and with the speed of technology development these days, there are probably some superior products out already.
- A pair of Earthworks QTC-1 microphones
- A Sound Devices MP-2 preamplifier
- A Nomad Jukebox digital player/recorder (Another good alternative is using a DAT recorder such as the TASCAM DAP1. The only drawback is that if you later would like to convert the DAT tape to digital files the process involves a real-time transfer.)
Analog recording system
Katy Payneís discovery in 1984 that elephants produced infrasonic calls was made with an analog system including a Nagra IV SJ reel-to-reel recorder and a Bruel & Kjaer (B&K) 4133 microphone.
How do you analyze infrasonic calls if you canít hear them?
One way to discover if you have recorded infrasonic elephant calls is to speed up the recording, raising all the frequencies in the recording to a level that you can hear them. Typically, if you speed up a recording containing infrasonic elephant calls 3 times, you will easily be able to hear them.
Sounds can also be represented visually using spectrograms. Spectrograms graph frequency on the y-axis, time on the x-axis and represent loudness of sound by the darkness of the display. We create spectrograms using the Raven software developed by the Cornell University Bioacoustics Research Program.
Is there a unique feature of infrasonic elephant calls which distinguishes them from other infrasound that may be recorded (for example from wind, other animals etc.)?
The structure of elephant rumbles are quite varied, but readily recognizable. Our team has scanned months of audio data in spectrogram format and clipped over 17,000 elephant calls. Our criteria for detection of an elephant call involved searching for a roughly eyebrow-shaped signal between 1-250 Hz and lasting between 2-10 seconds. The other infrasonic noise we encountered was broadband wind or thunder, which often obscures elephant infrasound.
Langbauer, Jr., W.R., Payne, K., Charif, R., Rappaport, E. and Osborn, F. 1991. African elephants respond to distant playbacks of low-frequency conspecific calls. Journal of Experimental Biology, 157: 35-46.
Larom, D., Garstang, M., Payne, K., Raspet, R., and Lindeque, M. 1997. The influence of surface atmospheric conditions on the range and area reached by animal vocalizations. Journal of Experimental Biology, 200: 421-431.
Payne, K., Thompson, M., and Kramer, L. 2003. Elephant calling patterns as indicators of group size and composition: the basis for an acoustic monitoring system. African Journal of Ecology, 41: 99-107.
Payne, K., Langbauer, Jr., W.R., and Thomas, E. 1986. Infrasonic calls of the Asian elephant (Elephas maximus). Behavioral Ecology and Sociobiology, 18: 297-301.
Poole, J.H., Payne, K., Langbauer, Jr., W.R., and Moss, C.J. 1988. The social contexts of some very low frequency calls of African elephants. Behavioral Ecology and Sociobiology, 22: 385-392.
Thompson, M., Schwager, S.J., Payne, K.B., and Turkalo, A.K. 2009. Acoustic estimation of wildlife abundance: methodology for vocal mammals in forested habitats. African Journal of Ecology, 48: 654-661.
Wrege, P.H., Rowland, E.D., Bout, N., and Doukaga, M. 2012. Opening a larger window onto forest elephant ecology. African Journal of Ecology, 50: 176-183.