For centuries, scientists and amateur seismologists have tried to use the behavior of both wild and domestic animals to predict earthquakes.
Unusual behaviors in animals reported ahead of seismic events presumably occurred because of some sensory input before the ground starts shaking.
The most striking case occurred in 1975, when, in addition to other precursor events, domestic and wild animals in and around the city of Haicheng, China, behaved in an extremely unusual way. Indeed, it is considered to be the only major earthquake ever to have been predicted.
The city was partially evacuated and many thousands of lives were saved.
That aside, no predictions of earthquakes have withstood careful scrutiny.
So, what, if anything, had the animals detected?
For the first time, researchers in the United States have offered an explanation of how animals might detect an earthquake in advance and why inconsistencies were likely in such predictions.
Researchers Michael Garstang and Michael Kelley propose that precursor earthquake crustal movements produce a sound signal that is detectable by animals.
The pair, writing in the journal Animals, have outlined what they describe as a logical but complex sequence of geophysical events triggered by movements in the Earth’s crust ahead of earthquakes.
“The sound heard by animals occurs only when metal or other surfaces (glass) respond to vibrations produced by electric currents induced by distortions of the earth’s electric fields caused by the crustal movements,” the pair argued.
“While the intrinsic value of such warnings is immense, we show that the complexity of the process may result in inconsistent responses of animals to the possible precursor signal.”
However, they took the view that a combination of existing measurement systems combined with more careful monitoring of animal response could nevertheless be of value, particularly in remote locations.
The pair said animals were capable of hearing a wide range of sounds, with frequencies ranging from ultra- to infrasound (kHz to Hz).
Animals generated and detected sounds over a wide range of frequencies and for many purposes. Bats, for example, use ultra sound (20–200 kHz) to navigate and capture prey. Elephants use infrasound (under 20 Hz) for long-range communication.
Garstang has done extensive work on the use of infrasound by elephants.
He has suggested in a scientific paper that elephants in Thailand and Sri Lanka, up to 1000km away from the epicenter of the magnitude-9.0 earthquake that hit 160km off the coast of Sumatra on Boxing Day, 2005, had detected the threat and moved to higher ground.
He believes what the elephants had detected was the low-frequency sound of the tsunami generated by the quake hitting the coast of Sumatra. The sound set off back across the ocean at 1260kmh, eventually passing the tsunami which was spreading out from the epicenter at 700kmh.
The sound arrived in Sri Lanka and Thailand ahead of the tsunami. Evidence from GPS tags worn by elephants showed they responded to the sound 38 minutes before the arrival of the devastating tsunami. They moved inland to higher ground.
No elephants were lost in the tsunami that hit Thailand. In contrast, a significant number of people lost their lives.
The sounds dealt with in their latest study are abiotic sounds − that’s from non-living parts of the ecosystem − not commonly considered within the biological acoustic literature.
The pair described events that they believed created the sounds that animals may respond to ahead of earthquakes.
Crustal movements in the earth’s near-surface did, indeed, produce sound, they said. However, this sound was not a direct product of the crustal releases.
“Instead, a sequence of events are set in motion, starting with crustal failure, occurring prior to an earthquake,” the pair wrote.
These signals are transferred at the speed of light, guided by the earth’s magnetic field, to the ionosphere. Here the crustal signature is detected by the Slant Total Electron Content, a measure used for the Earth’s ionosphere.
“These STEC fluctuations of the ionosphere are transferred to the earth’s surface and are detectable by ground-based GPS.
“At this point, they are not in the form of sound waves but trigger vibrations in metal, glass and other surfaces.
“This latter process, known as electrophonics, generates audible sounds with frequencies ranging between 20 Hz and 20,000 Hz. These are considered to be the sounds that many animals respond to.”
Garstang and Kelley say the lead time between the precursor crustal movements and any subsequent earthquake is indeterminate, but can probably range from tens of minutes to many hours.
“The ultimate signal, however, represents definitive evidence of the imminent earthquake, an achievement long sought by humans.”
They said the sequence of events they described provided both a testable framework for detecting a precursor earthquake signal and a basis for understanding the inconsistencies observed in animal behavior before an earthquake.
“It is also likely that the severity of the earthquake, magnitude of the audible response and the behavioral reaction of animals are all related.
“The understanding of this process coupled with the technological ability (GPS), to monitor the magnitude and occurrence of the signals should provide the basis for earthquake prediction.
“Reliance upon and the use of animal responses to the potential earthquake can thus be refined and improved. Reliable response to the adherent animal behavior could still prove useful, particularly in poorly monitored remote areas.”
The relationship of the precursor crustal movements to the intensity of the earthquake remained uncertain, they said.
“Nevertheless, we present a plausible predictive approach which can be subjected to both post and future analysis.”
Issues relating to how predictions of earthquakes were used went beyond scientific inquiry, they said.
Garstang is with the Department of Environmental Sciences at the University of Virginia in Charlottesville, and Kelley is with School of Electrical and Computer Engineering at Cornell University in Ithaca, New York.
Understanding Animal Detection of Precursor Earthquake Sounds
Michael Garstang and Michael C. Kelley
Animals 2017, 7(9), 66; doi:10.3390/ani7090066