HR: 0830h
AN: U31B-0007 [PDF]
TI: Observation of a Rayleigh Wave Induced by Infrasonic Elephant Vocalizations: a Possible Communication
Mode?
AU: * Gunther, R
AF: Department of Geophysics, Stanford Univeristy, Stanford, CA 94305-2215 United States
AU: O'Connell-Rodwell, C E
EM: ceoconnell@stanford.edu
AF: Department of Geophysics, Stanford Univeristy, Stanford, CA 94305-2215 United States
AU: Klemperer, S
AF: Department of Geophysics, Stanford Univeristy, Stanford, CA 94305-2215 United States
AU: Rodwell, T C
AF: Department of Geophysics, Stanford Univeristy, Stanford, CA 94305-2215 United States
AU: Haines, S
AF: Department of Geophysics, Stanford Univeristy, Stanford, CA 94305-2215 United States
AU: Goldman, M
AF: US Geological Survey, 345 Middlefield Rd, MS-977, Menlo Park, CA 94025 United States
AU: Evans, J R
AF: US Geological Survey, 345 Middlefield Rd, MS-977, Menlo Park, CA 94025 United States
AB:
A variety of animals such as arthropods, amphibians, reptiles, fish and rodents communicate by creating and sensing ground
vibrations rather than, or in addition to, sound waves. There is evidence that this may be the case for elephants as well.
We set out to characterize the Rayleigh wave generated by near-source coupling during elephant low frequency rumble
vocalizations (25 Hz lasting 3-7 seconds), using standard engineering-scale seismology equipment.
We used a 60-channel GeometricsT seismograph to record data from vertical and horizontal geophones and from microphones,
placed along a 168-m cable near Salinas, CA. Seismic wave-speed for body waves (1400 m/s) and surface waves (440 m/s) and
the air-wave velocity (340 m/s) were established using a sledgehammer source. Trained elephants vocalized on command at one
end of our seismic recording spread. The vocalization was strongest at 25 to 28 Hz (with strong higher harmonics), with a
duration of 3 to 4 seconds, and repeated multiple times with separations of 2 to 5 seconds. Unlike an explosive seismic
source, the duration of the elephant vocalization is tens of times longer than the characteristic period of the source,
lasting far longer than the total propagation time along our seismic recording spread (less than 500 ms), so that different
propagating modes cannot be separated by different arrival times. Unlike a Vibroseis$^{TM}$ sweep, the elephant rumble is
relatively monotonic with no characteristic onset, ruling out the use of deconvolution techniques to recognize the signals.
Using a semblance technique applied to linear moveouts on narrow-bandpass-filtered data, coupled with forward modeling, we
demonstrate that the complex waves observed are the interference of an air wave and a Rayleigh wave traveling at the
appropriate velocities. The Rayleigh wave appears to be generated at or close to the elephant, either by coupling through the
elephant's body or through the air near the body to the ground.
In our experiments, the amplitudes of both the elephant-coupled Rayleigh wave and the elephant-driven airwave had decayed to
almost ambient noise levels at the end of our 168-m-long recording spread. This was most likely due to the high
ambient-noise levels during our experiment. Free-ranging African elephants have been shown to respond to low-frequency calls
of other elephants at ranges of 2 km with an ideal outer limit of 10 km. Because a surface wave decays at only 1/r, we
speculate that wild elephants may detect the Rayleigh waves of other elephants via bone conduction or somatosensory reception
or both, and hence may communicate at greater distances than possible using infrasonic calls transmitted through the
atmosphere.
UR: http://biox.stanford.edu/iip_klemperer.html
DE: 0400 Biogeosciences
DE: 0689 Wave propagation (4275)
DE: 3200 MATHEMATICAL GEOPHYSICS (New field)
DE: 3210 Modeling
DE: 7255 Surface waves and free oscillations
SC: U
MN: 2003 Fall Meeting