Ele-Comm: Short for Elephant Communication

Caitlin O'Connell-Rodwell is a Stanford University professor whose research focuses on how elephant vocalizations travel through the ground for great distances, and how other elephants can understand them, just as they understand acoustic sound, which travels through the air. O'Connell-Rodwell is the author of The Elephant's Secret Sense. You can see videos of some of the elephant communication experiments she describes in the podcast on her Utopia Scientific site. The links to the videos are on this page.

Early in her research, O'Connell-Rodwell noticed behavior that indicates elephants are listening to acoustic (airborne) sounds by putting their ears out and orienting toward the sound's source. At other times, she also noticed a more puzzling behavior: Several elephants would freeze simultaneously, sometimes in mid-stride, and would press their front feet into the ground. They might also roll a foot forward so that only their toes touched the ground. At other times, they would lift a front leg. The behavior reminded her of the behavior she saw in insects that communicate seismically.

She began a series of experiments that eventually found that:

  • Low-frequency elephant vocalizations, which are below the threshold of human hearing, travel through the ground in the same waveform as they do in the air. The ground vocalization can travel faster or more slowly than acoustic sound, depending on soil conditions, but has the potential of travelling further as there is no outer limit to how far sounds can travel through the earth.
  • When she played a recorded elephant vocalization through the ground only (not through the air), other elephants detected the vocalization.
  • Elephants understood the ground-borne vocalizations. For example, they responded appropriately to an alarm call from another elephant by assuming their defensive posture of bunching and freezing. They also responded only to alarm calls of elephants living in the area rather than those made from elephants elsewhere.

Further research revealed that there are two ways elephants 'hear' sound waves traveling through the ground:

  • They have an enlarged malleus, a middle ear bone that plays an important role in hearing. Animals that communicate seismically often have an enlarged malleus as it also facilitates bone conducted detection of vibrations.
  • Elephants can close their middle ear canal, forming a closed acoustic tube which enhances bone conduction and blocks out acoustic sound, helping the elephant focus on the vibration pathway.
  • They have an acoustically designed foot, with a thick fat pad that perhaps helps in the transmission or conduction of vibrations.

O'Connell-Rodwell was interviewed in LifeLines podcast episode 25. In the first segment, she discusses how elephants communicate seismically during migration. In the second segment, she describes how they sense these seismic vibrations using their feet and trunks. Click here to listen to the full podcast.

More like this

I thought the idea was refuted on the grounds that long-frequency sound is absorbed at long distances. What is the up-to-date knowledge about elephant infrasound communication range?

By Birger Johansson (not verified) on 08 Sep 2010 #permalink

Dear Dr. O'connel-Rodell:
Congratulations on your scientific research. As you know the majority of the public has turned away from science toward "diety worship". Are you working on any practical applications of your research? Can elephants hear hunters or poachers coming to kill them? How far can these low frequancy sounds be heard through the ground? What about the nature-nurture discussion? Using sensitive audio equipment can you build a library of sounds and perhaps someday interpret what they mean? Do different herds use different low frequancy sounds, in other words can one herd communicate with another? At approximately what age do young elephants acquire this ability as well? Must each young elephant be taught to use this underground form of communication or is it genetically linked? Are you working with Forest Elephants or other kinds.
Best of luck with your future research. When you publish the results of your reaearch will you please alert me on this Podcast? My granddaughter's favourite animal is an elephant after my daughter took her to the Toronto Zoo.

Rob Evans, B.Sc., M.Sc.
151 Cliffwood Road
Toronto, Ontario
Canada, M2H 2E3

Hello Birger,

Thanks for your question. It is high-frequency sounds that attenuate quickly over long distances, not low-frequencies. Low frequency sounds, in the range of 20 Hz (the fundemental frequency of an elephant rumble), have the capacity to propagate over several kilometers, depending on wind and temperature conditions. Under inversion conditions, lower temperatures allow low frequencies to travel further as the air molecules within an inversion are more densly distributed and thus facilitate maintaining the integrity of the signal for longer. The published outer limit for acoustic transmission is 10 km (Snell's Law dictates that sounds refract back into the atmosphere after 10 km, but under inversion conditions, it is possible for sounds to travel further). In theory, there is no outer limit for how far these low frequency sounds propagate through the ground. In our studies, we have shown that seismic signals propagate at least in the range of airborn signals (several kilometers), but probably further under ideal conditions. In acoustic playback studies, researchers have shown elephants to respond at the distance of about 2 km, and collared elephants move within the range of acoustic transmission of up to about 4 km.

I hope this helps.

Dear Rob,

Wow, that's a lot of questions! Thanks for your enthusiasm for my research. To answer your questions in order, I do believe that elephants use the ground to assess danger, whether it be the vibrations of a vehicle or footfalls of a human. We have done a study to show that human footfalls are measureable and unique from elephants (look at Wood et al., J. Applied Ecol. 2005). Unfortunately, we haven't been able to measure vibrations in the ground in a truely quiet background (without human generated noise) so that we would know just how far they travel uninterrupted, but our models predict tens of kilometers. I am not sure what the nature-nurture question refers to but would be happy to answer if you clarify.

As for a sound library, many are working on this. It has been difficult because there are very subtle differences between some call types. Usually, length of the call and the amount of repetition of a call is a signal in and of itself, as is the case for estrus calls. Frequency modulation is also important in distinguishing classes of calls (alarm calls have more modulation than contact calls, for example). Check out Joyce Poole's work on wild elephants and the work of Joe Soltis and his colleagues on captive elephants.

Yes, family groups communicate within an extended social network and can determine those that they recognize and those that they do not through vocalizations. Karen McComb has done the most definitive work in this area. What we had done to build on this work was to take two different callers producing the same call type (an alarm call) and play it back seismically to a group that would be either familiar or unfamiliar to the caller. Elephants at my field site only responded to seismic alarms produced from familiar elephants.

A group at the San Diego Wild Animal Park, under Matt Anderson, are looking at the question of what age an elephant can first produce a rumble vocalization as they have a number of calves of different ages. They don't have the answer yet, but stay tuned. I'm sure they give updates on their website.

Elephants are born with the ability to detect vibrations, but it is their life experience that determines whether they focus on vibrations as signals. For example, an elephant born in a zoo would most likely not be communicating with other elephants at a distance, and there would be so many man-made vibrations in the ground of no meaning to the elephant that there is a good chance that captive elephants do not focus on the ground as a communication channel, but may use it to detect large-scale rare events like an earthquake. I think of the captive versus wild elephant experience similar to a person with normal hearing versus one that has a hearing impairment. To the nonhearing human, vibrations are a very important tool to assess the environment and for communication, whereas, for the hearing human that can rely on detecting noises (the phone ringing) or voices, do not use vibration detection in the same way.

I mostly work with savannah elephants in southern Africa, as well as in captivity.

I will certainly let APS know when we have some new results. Glad your granddaughter chose such a great favorite animal!

All the best.

Thanks for bringing insight into these wonderful animals. I've been fascinated by elephants for as long as I can remember. This adds a whole new dimension to them, one I had no idea about. I imagine this ties in to their intuitive senses- knowing when to run from a natural disaster before those around them know anything is wrong. It would make sense that they pick up those vibrations through the ground as well!

Dear Carrie O'Connell,
OK this is not on topic, at all, but my husband and I were wondering where on the ear flap the opening to elephant's ear canal is... is it in front or behind. It make more sense to me for it to be in the front, but he seems to think it is behind... We tried to google this but got more hits for the Elephant ear plant than any thing else. Thought you should know. Thanks :)