ASU Research E-Magazine: To Make a Quake

On February 28, 2001, the city of Seattle relocated itself 5.5 millimeters to the north in a magnitude 6.8 earthquake that caused $2 billion in damages and injured 272 people.

On January 13, 2001, a magnitude 7.6 earthquake off the coast of Central America killed 725 people, destroyed more than 100,000 homes, and triggered about 500 landslides across El Salvador.

In September 1999, a magnitude 7.5 earthquake rumbled near the coast of Oaxaca, Mexico. At least 29 people died. Another 100 were injured, and more than 10,000 homes were wrecked.

In 1970, a Peruvian earthquake set off an enormous landslide that buried two villages, along with 18,000 people. At least 50,000 people died as a result of that quake, and more than 800,000 were left homeless.

All of these disasters are related in several ways. First, each earthquake caused significant loss of life and property. Each also belongs to a particular category of earthquakes—a category that is largely ignored by the scientific community because it is considered low risk.

Intermediate depth, or intraslab, earthquakes originate 50 to 300 kilometers beneath the Earth’s surface. They occur in subduction zones, areas where the oceanic crust dives down beneath the edge of a continent.

Intraslab earthquakes are considered less threatening than their shallower counterparts because they occur deep inside the crust. They don’t rupture the surface of the Earth. Yet these earthquakes can and do have devastating effects.

“The seismic risk from intermediate depth earthquakes has been under appreciated,” says Simon Peacock, a professor of geology at Arizona State University. “Typically, the only earthquakes we’ve worried about are those that rupture the surface, the shallow earthquakes.”

Intraslab earthquakes may finally get their due, thanks to research findings from Peacock and Kelin Wang of the Canadian Geological Survey. Their results appeared in the October 29, 1999 issue of the journal Science.

Their results support a recent theory on the causes of intraslab earthquakes. The findings also indicate that certain geographical areas, like the Pacific Northwest, are very much at risk for a strong, intermediate-depth quake.

Peacock says that the cause of these earthquakes has puzzled scientists for a long time.

“Shallow earthquakes represent normal brittle failure of rocks,” he explains. “If you stress rocks enough, they’ll fracture and break. But the intermediate and deep earthquakes are more troublesome, because we don’t think what’s going on there is normal brittle failure. These rocks are under very high pressure because they’re deep inside the Earth. The tremendous weight of the overlying rocks should inhibit slip along a fault.”

In 1996, Stephen Kirby and colleagues at the U.S. Geological Survey proposed that intraslab earthquakes occur when the extreme heat and pressure in a subduction zone cause the rocks to metamorphose, or change into different forms. As they change, the rocks release water, which essentially lubricates the fault, causing it to move.

Peacock compares the process to firing pottery in a kiln.

“If you fire a pot at high temperatures, essentially you are metamorphosing the pot. Some of those pots break in the kiln, because water was driven off too fast. That water vapor accumulated in small fractures in the pot and exploded.”

The water could come from one of two places. It could come from wet clay, if the pot wasn’t dried correctly before firing. Or it could come from metamorphic changes in the clay brought on by heat.

“The firing process actually changes the mineral structure of the clay,” Peacock explains. “The clay minerals break down, vitrify into a glassy material, and make other minerals that have less water.”

To test Kirby’s theory, Peacock and Wang studied subduction zones in Japan, one of the most seismically active areas on the Earth. The country also has some of the world’s best seismograph records.

The researchers focused their efforts on two Japanese subduction zones, one in the southwest and one in the northeast. They examined the relationship between processes going on inside the Earth and the depth of earthquakes in each area. In northeast Japan, earthquakes occur at depths as great as 200 kilometers (124 miles). In contrast, earthquakes in southwest Japan do not exceed 65 kilometers (40.3 miles) depth.

For Kirby’s theory to work, metamorphic processes would have to occur much deeper in the northeast, triggering earthquakes deeper in the Earth. Since metamorphism requires extreme heat, the scientists to determine the temperature far underground.

To do this, they used mathematical simulations that describe heat transfer. They also studied rocks that originally formed at such depths.

“We actually have rocks at the surface that can tell us something about the conditions at depth in the Earth,” Peacock says. “Like a spacecraft exploring outer space, these rocks have explored inner space and come back to tell us about it. By studying these rocks, we can infer things about the processes going on at depth.”

Based on their calculations and observations, Peacock and Wang determined that northeast Japan is a very cold subduction zone, while southwest Japan is much warmer.

“There’s a striking contrast in temperatures between the two subduction zones. A contrast of 300 degrees Celsius (572 degrees Fahrenheit) at 100 kilometers (62 miles) depth,” says Peacock. “Metamorphic reactions occur much deeper beneath northeast Japan.”

The geologists also studied how seismic waves travel through the two areas. “That tells us something about the rocks through which those waves propagate,” explains Peacock. “In turn, this tells us something about the structure of the Earth.”

In northeast Japan, there is a thin layer extending about 150 kilometers (93.2 miles) deep that has a low seismic velocity. Seismic waves travel through this layer very slowly.

In southwest Japan, however, the low velocity layer only extends about 60 kilometers deep.

All of these observations support the theory that the release of water from rocks causes intermediate depth earthquakes.

“The answer is all tied up in the metamorphism of the oceanic crust as it’s subducted. Because the temperatures are so different, the metamorphic reactions are different,” Peacock says.

In northeast Japan, the cold layer extends much deeper than in the southwest, he says. This explains why earthquakes occur deeper in the northeast than in the southwest.

“Water releasing reactions occur at deeper depths in northeast Japan than southwest Japan because they’re colder, and many of those reactions depend on temperature. Until the rocks reach, say, 500 degrees Celsius (932 degrees Fahrenheit) they’re not going to release a lot of water.”

This also explains why the low velocity layer extends deeper in the northeast, Peacock says.

“As this rock gets buried, it transforms into a red and green rock called eclogite. That’s a very dense rock, which has very high seismic velocity. So where you see the low velocity layer, the oceanic crust has not yet transformed into eclogite. That transformation is delayed in the cold subduction zone.”

These characteristics also help to explain differences in another phenomenon—volcanoes. Most subduction zones, including northeast Japan, have lots and lots of volcanoes. However, they are strangely absent from Japan’s southwestern region.

“The volcanoes we see in subduction zones largely reflect melting of the mantle triggered by water,” Peacock explains. “What triggers melting of rocks is actually the release of water at great depth. That water migrates out of the rocks, travels upwards, and enters very hot rocks above. The water entering hot rocks will trigger melting.”

In southwest Japan’s warm subduction zone, water is released at shallow levels. As a result, it never reaches the hottest rocks deeper down.

Although Peacock’s research focused on Japan, he says the findings apply to the United States, as well.

“Northeast Japan is analogous to Alaska. Southwest Japan is analogous to the Pacific Northwest, beneath Northern California, Oregon, and Washington. The study of Japan certainly gives us insight into processes occurring in our own backyard,” Peacock says.

On average, intermediate depth earthquakes tend to be less intense than their shallower counterparts. However, they are more likely to occur near large population centers, according to the ASU geologist.

“If you want to weigh the seismic risk…you might get as much if not more damage from a magnitude 7 [earthquake] that’s 30 miles down beneath your feet than a magnitude 8 that’s a couple hundred miles away off the coast. This is a major concern in the Pacific Northwest. Seattle is probably more at risk from intermediate depth earthquakes than from a larger earthquake on the coast.”

Peacock’s findings are helping scientists to understand what causes earthquakes, but will they allow us to predict the events? Not in our lifetimes, he says.

“We are a long way from predicting earthquakes. I think it’s something that will be extremely difficult, but that ultimately it’s possible,” Peacock says. “If this model of what causes intermediate depth earthquakes is correct, ultimately, we might be able to detect when water pressure is building up. That might suggest an earthquake is imminent.”

Still, the new information can be useful. Peacock says we need to take intraslab earthquakes more seriously. In 1999, there were a dozen magnitude 7 or higher earthquakes around the world. At least seven of those were of the intermediate depth variety.

The ASU scientists suggests that governing entities focus their efforts on strengthening and enforcing building codes so that structures will better withstand earthquakes.

Every effort counts. A major earthquake can have effects that shake the whole world.

“As the population of the planet grows, many of the major cities are built in seismically active areas: Los Angeles, Tokyo, Athens, Istanbul, Manila,” says Peacock. “A major earthquake beneath Tokyo or Los Angeles could conceivably cause the world to go into a recession.” If the “big one” ever does level Los Angeles, researchers estimate that damage costs could exceed $1 trillion dollars.—Diane Boudreau