New studies put potential megaquake closer to Seattle

Pacific Northwesterners have learned to live with the knowledge that a monster earthquake will wallop the region one day.

But for Seattle-dwellers, that harsh reality has always been tempered by the fact that the fault called the Cascadia Subduction Zone is most likely to snap offshore — nearly 100 miles from the urban corridor.

Now scientists are beginning to suspect the jolt could come much closer.

The latest evidence is an upcoming study in Geophysical Research Letters that concludes the danger zone could lie within 50 miles of Seattle and the Interstate 5 corridor.

"That's a lot closer to where everybody lives," said co-author Timothy Melbourne, of Central Washington University.

The closer the quake, the heavier the damage in Seattle, said Thomas Heaton, professor of engineering seismology at California Institute of Technology. Heaton, who was not involved in the study, conducted computer simulations that show the city's tall buildings would be particularly vulnerable.

"If that earthquake happens, it will hammer the place," Heaton said. "It will hammer harder if it's closer."

The researchers hope their work will help spur tougher building codes.

"There's no reason to panic at the moment," Heaton said. "But on the other hand, people shouldn't just walk away and say: 'Oh, let's not think about it.' "

The 600-mile-long subduction zone is the place where the ocean floor is being shoved, or subducted, under the continent.

It's one of the world's most treacherous faults, capable of unleashing megaquakes and tsunamis on a par with the 2004 Sumatra disaster.

Fortunately, such magnitude 9 quakes are rare in the Cascadia zone, averaging one every 400 to 500 years. The most recent occurred in 1700.

To figure out where the fault will snap, Melbourne and co-author James Chapman analyzed data from 15 "silent earthquakes" that occurred beneath the Olympic Peninsula.

Scientists believe the nearly imperceptible rumbles of these silent quakes reflect deep slippage on the fault, in a zone where heat and pressure render the rocks plastic.

But that deep slip increases pressure on the danger zone: The locked portion of the fault closer to the surface.

"Eventually the fault will rupture and unload that spring," Melbourne said.

The key to knowing where the rupture will occur is knowing where the plates are locked.

Early analyses, based on a basic understanding of temperature and rock behavior, put the locked zone just off the Washington coast.

"You try to calculate where the rocks slide freely and where the rocks will lock," said John Vidale, head of the Pacific Northwest Seismic Network at the University of Washington. "It was really kind of a wild guess that we had before."

Melbourne and Chapman took a more systematic approach, using the "silent quake" data to calculate strain on the fault. They then looked at data from a network of GPS receivers across the Northwest that show how different parts of the region are being differentially squeezed and shoved by the slow-motion force of the colliding geologic plates: Pacific Beach on the Washington coast moves northeast about 0.8 inches a year, while Whidbey Island shifts a scant 0.2 inches a year.

"What we measure is where the crust is deformed above the fault, and from that we can back out where the fault itself is locked," Melbourne said.

The best explanation is that the locked portion of the fault actually extends well beneath the Olympic Peninsula, he said.

The UW's Vidale and his colleagues reached the same conclusion from seismic studies that show the silent quakes originate in the same area.

But Herb Dragert, at the Geological Society of Canada's Sidney, B.C., branch, says analysis of a similar fault in Japan suggests there might be a kind of transition zone where the plates are not fully locked, and therefore not as dangerous.

"The question is still open," Dragert said.

No matter where the fault breaks, Heaton and his student Jing Yang concluded the damage in Seattle would be severe. The ground would shake violently for three to five minutes. Wooden houses and modern, low-rise buildings would be relatively unscathed.

But many high-rises built before 1994 could be in serious trouble, largely because of welds used in place of rivets in their steel frames.

"These giant Cascadia earthquakes are certainly very different from anything that was used to formulate the codes," Heaton said.

"What we need to do is follow up on these studies and try to come up with a more rational way to decide what's the appropriate design for our buildings."

Sandi Doughton: 206-464-2491 or sdoughton@seattletimes.com