Nautilus Members enjoy an ad-free experience. or Join now .

Humans have been trying to predict earthquakes at least since first-century China, when the device of choice was a vessel fitted with metal dragons facing each compass direction. If the ground shook somewhere in the region, the metal ball in the dragon’s mouth would drop out, roughly indicating the direction of the earthquake. Our methods have gotten a bit more sophisticated since, but predicting earthquakes ahead of time remains shaky business.

“Why are earthquakes the last of the natural hazards to be predictable? For one thing,” Paul Silver, the late American seismologist, once said, “the short propagation time means that prediction must be based on the existence of a preparation phase. It is clear that we have yet to detect, on a reliable basis, such a preparation phase.” Terry Tullis, a seismologist with the National Earthquake Prediction Evaluation Council, told Nautilus in July that, “over time, with enough measurements and careful analysis, maybe at some point, someone will stumble across something that has definitive predictive value.”

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Here are some (supposed) earthquake prediction methods—some strange, some useful, and some that even a metal dragon could beat.

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Animal Behavior

When an earthquake struck the ancient Greek cities of Helike and Boura in 373 BC, reports claimed they were devoid of animals: In the days prior to the quake, rats, weasels, snakes, and insects were recorded fleeing the area. These stories of apparently clairvoyant creatures abound, but scientists have had trouble pinning down how animals might be able to sense future quakes. Research on the topic is fraught, since studies usually come after the fact and risk applying retroactive significance. In a 1981 study that controlled for this, researchers found that unusual animal behavior could only be significantly tied with one of the four earthquakes investigated.  As the paper puts forward, not all earthquakes are identical; each earthquake may have its own unique precursors, only some of which may be apparent to animals.

Radon Gas Emissions

After the 1966 Tashkent earthquake in Russia, scientists observed a strange spike in the concentration of radon—a colorless, odorless, and tasteless radioactive gas—in groundwater near the epicenter. They speculated radon had been bubbling into the water for several days before the quake, which could make it a potential, and highly useful, earthquake precursor. Experiments had shown that rocks do emit significantly more radon gas under stress. What’s more, radon’s radioactivity makes it easy to spot, and its short half-life means it cannot diffuse far from its source. Yet after examining 125 radon observations from 86 earthquakes between 1966-2002, the International Commission on Earthquake Forecasting for Civil Protection found no significant correlation between the two. Often, gas releases are only at sites far from the earthquake’s epicenter, and highly monitored faults like the San Andreas have not shown any regular radon releases that would support this hypothesis.

Nautilus Members enjoy an ad-free experience. Log in or Join now .

One paper that utilized a database of over 13,042 earthquakes still could not find a statistically significant correlation with Earth’s tides.

Earthquake Lights

Earthquake stories both ancient and modern have sometimes been accompanied by tales of strange lights: lightning that comes from the ground, floating orbs, even displays that resemble bluish flames. Little surprise that these lights were often seen as religious portents, or even alien visitors. However, recent research has shown that igneous rocks under seismic stress could actually produce the lights. During laboratory experiments, researchers have found that these rocks accumulate electric charges as stress deforms them. It’s theorized these charges could build into a plasma-like state that illuminates in air. Though geophysicists aren’t completely sold, the phenomenon is now more than conspiracy-theory fodder—but because earthquake lights are relatively rare, they have limited usefulness for predicting events.

Electric Signals

Nautilus Members enjoy an ad-free experience. Log in or Join now .

In 1981, a group of Greek seismologists began publishing papers about a new method to predict earthquakes on a short-term basis: the “VAN” method, an amalgam of the researchers’ initials. The team measures electro-telluric signals, electric voltages conducted through the ground, recorded on specially calibrated instruments. Using these tests, the group has claimed many successful earthquake predictions, but their method is still highly disputed; critics have claimed that the VAN method’s successful predictions of earthquakes could be statistically attributed to chance. This hasn’t dissuaded the VAN team, however. The group revised its methods in 2001, and since then has claimed to have predicted 25 of 28 major earthquakes in the Mediterranean region.


In the 1980s, a self-appointed climatologist named Iben Browning gained notoriety when he forecast that a major earthquake would occur on the New Madrid Fault, in Missouri, within 48 hours of December 3, 1990. Browning said he had calculated when the tides were exerting maximum force on Earth’s crust due to its alignment with the moon and sun—an idea known as syzygy, speculated to influence earthquakes since at least the 1700s. Yet December 3 came and went, and no earthquake stuck. Browning wasn’t the first to stymy science and the public alike with syzygy; scientists have long searched for some correlation between tides and earthquake frequency, yet the results have been largely inconclusive. The contribution of tides to earthquakes is likely tiny overall, and restricted to only certain places on the planet; one paper that utilized a database of over 13,042 earthquakes still could not find a statistically significant correlation with Earth’s tides.

Ambient Noise

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Extremely low-amplitude waves arise from transportation and other heavy human activity, as well as natural movements like ocean tides. In 2001, a paper showed that this ambient noise had an unlikely use, says science writer Mark Harris in his Nautilus article, “The Last of the Earthquake Predictors.” When it’s detected by receivers spread out over a sufficient distance, it’s possible to estimate the waves’ travel time, which allows you to infer the material they’re propagating in. A 2005 study took advantage of this to map the ground beneath Southern California—especially slow waves were linked to sedimentary basins, while remarkably fast waves hinted at the igneous mountain cores.

Scientists surmised that ambient noise could likewise be employed to observe geologic ruptures forming deep underground. This lent support to a controversial idea called dilatancy, which holds that fractures widening in stressed subterranean rock will cause it to stretch. This slow opening retards any seismic waves moving along, perhaps indicating that the rock is about to bust. Using ambient noise to predict earthquakes is prohibitively expensive, though, because it would require sensitive instruments to be placed along the lengths of entire fault lines.


Before an earthquake strikes, some scientists believe it’s possible to detect pulses in Earth’s magnetic field. This happened a few years ago, before earthquakes in Perú went off. Friedemann Freund, a physicist associated with the SETI Institute, thinks he knows what might be behind them. In the days leading up to an earthquake, he believes, stressed underground rock produces sizable electric currents that propagate to the crust, distorting Earth’s magnetic field while ionizing the atmosphere and shooting out infrared energy.

Nautilus Members enjoy an ad-free experience. Log in or Join now .

Critics of Freund’s view suggest that these magnetic pulses have some inexplicable human cause. John Ebel, a seismologist at Boston College, has noted that, decades ago, his magnetometers in Boston began to detect a sequence of strange pulses each morning. Eventually, according to Science, “he and his colleagues identified the sources of those gremlins: It was the engineers cranking up Boston’s trolley cars at a rail yard a few kilometers away from the instruments.”

Claudia Geib is an editorial intern at Nautilus. Follow her on Twitter @cm_geib.

The lead photograph is courtesy of orangejon via Flickr.

Nautilus Members enjoy an ad-free experience. Log in or Join now .
close-icon Enjoy unlimited Nautilus articles, ad-free, for less than $5/month. Join now

! There is not an active subscription associated with that email address.

Join to continue reading.

You’ve read your 2 free articles this month. Access unlimited ad-free stories, including this one, by becoming a Nautilus member.

! There is not an active subscription associated with that email address.

This is your last free article.

Don’t limit your curiosity. Access unlimited ad-free stories like this one, and support independent journalism, by becoming a Nautilus member.