Earthquake Prediction Remains a Moving Target

John Roach
for National Geographic News
July 14, 2004
According to Max Wyss, when the Beatles first shook the world with hits like "I Want to Hold Your Hand" and "She Loves You," in the early 1960s, scientists were clueless as to why the Earth literally rattles and trembles on occasion.

"What I'm trying to say is this is a very young, recent advance to know why this planet has earthquakes," said Wyss, who is the director of the World Agency of Planetary Monitoring and Earthquake Risk Reduction in Geneva, Switzerland.

By the late 1960s, as the Fab Four were pulling stunts like rocking a rooftop in Santa Monica, California, scientists began piecing together the theory of plate tectonics.

According to that groundbreaking theory, the Earth's surface is divided into about a dozen thin shells of crust that move about as a result of motions in the Earth's interior.

At mid-ocean ridges, which are known as seafloor spreading centers, new crust erupts from Earth's interior and pushes the plates apart. At so-called convergent boundaries, the plates crash into each other, sometimes forming mountains and volcanoes. At transform fault boundaries plates slide horizontally past each other.

"When the plates push against each other, friction results in stress accumulation, which is released in earthquakes," said Carol Raymond, a geophysicist at NASA's Jet Propulsion Laboratory in Pasadena, California.

In the few decades since the theory of plate tectonics gained widespread acceptance, scientists have been trying to understand the physics of earthquakes. They hope that they can accurately predict when earthquakes may occur and thus save lives and property.

Andy Michael, a researcher with the U.S. Geological Survey's Western Earthquake Hazards Team in Menlo Park, California, said that, unlike the physics that control the weather, the physics that control earthquakes are at this time poorly understood.

"If you know the weather in Kansas today, you can pretty much guess areas a few hundred miles east of Kansas will get that weather tomorrow. … Earthquakes happen suddenly, we can't watch the system evolve," he said.

Modern Earthquake Forecasts

Today agencies such as the U.S. Geological Survey issue assessments—based on a region's earthquake history—that a given area will experience shaking beyond a certain force. By knowing the location and the size of past earthquakes, scientists can forecast the probability of future earthquakes in the region.

Michael said these forecasts, or hazard assessments, are broken into three time scales—long, intermediate, and short term. Long- and intermediate-term forecasts are useful for government decisions, such as where to spend money retrofitting buildings to make them secure during an earthquake. Short-term forecasts help with decisions about where to position emergency supplies.

According to Michael, although people tend to want short-term earthquake forecasts, the long-term forecast is actually more important. "It does things like set building codes, so you don't care when it's happening because you're going to be safe anyway," he said.

Wyss said that people need to understand the probability component of earthquake forecasts. He likens an earthquake to a powder keg that is likely to go off soon, but what the exact spark will be that sets it off is unknown.

"Please accept as useful something that states a year and a large area, perhaps. Please accept from us a probability. There really is no statement we can make about the future that doesn't have a probability associated with it," he said.

Earthquake Prediction

In addition to earthquake forecasts such as hazard assessments, scientists also attempt earthquake predictions based on their understanding and analysis of the physics that govern earthquakes.

A "prediction is a statement that a single earthquake will take place, with a very high probability, in a specified region, time frame, and magnitude range," said David Jackson, director of the Southern California Earthquake Center at the University of California, Los Angeles.

For today, at least, Jackson said long-term forecasts are more useful than short-term predictions as long-term forecasts can be tested against the earthquake record. Predictions need a long string of success and failure to evaluate, which at this point does not exist.

But scientists are working on new prediction models. Michael is interested in the concept of predicting a major earthquake based on a pattern of increasing small earthquakes prior to the main event. "It's controversial, but with further investigation it may prove correct," he said.

Even more controversial is the concept of predicting earthquakes based on the strange behavior of animals such as dogs and cats. The animal method is largely dismissed by earthquake researchers as having no scientific basis. "Animal behavior is generally considered not worthy of study," Raymond said.

According to Wyss, since prediction is it based on physics telling scientists something is about to happen, their prediction can only be as good as their grasp on the physics. "We have a problem not knowing the Earth so well," he said.

Scientists hope that, as they develop and deploy new technologies to study the Earth, they will begin to understand the physics better.

Raymond is involved in the study of one such possible new technology—using satellites to study the deformation of the Earth's crust as a possible precursor to earthquakes. We'll learn more about this technology in a future story.

Scroll down for more earthquake news.

© 1996-2008 National Geographic Society. All rights reserved.