Napa Valley quake could help forecast 'big one' – 5 seconds early

An earthquake near Napa, Calif., injured more than 200 people. It also provided confirmation that warning systems created by scientists have the potential to give a vital few seconds of warning.

The earthquake that jolted California's Napa Valley in the predawn hours Sunday has left the city of Napa with a large cleanup job – and scientists with increasing confidence in an earthquake early warning system that they say could help save lives when the "big one" comes.

Sunday's magnitude 6.0 quake occurred in a region laced with faults related to California's San Andreas Fault, an 810-mile-long crack in the Earth that defines the boundary between two large, moving patches of crust: the North American and Pacific plates.

Napa city officials reported Monday that while the quake injured 208 people, a few critically, the quake left no fatalities out of a population of nearly 77,000, officials say.

The quake's timing, coming at 3:20 a.m. Sunday, played a key role in the low casualties, says Brad Alexander, a spokesman for the California Governor's Office of Emergency Services.

"We definitely lucked out as far as when this earthquake occurred," he says. Recalling the Loma Prieta earthquake, a magnitude 6.9 event that struck the bay area in 1989, Mr. Alexander notes that while Loma Prieta was more powerful – it released 22 times more energy than the Napa quake – it also struck around 5 p.m., rush hour. People were stuck in traffic on freeways and bridges that collapsed, in addition to being inside buildings that were damaged.

On Sunday, "having folks off the roads and having them in a safe place obviously is going to be helpful" in holding down casualties, he says.

The nature and location of the quake also helped.

The ruptured segment involved a relatively small section of a poorly understood fault about five miles south-southwest of Napa, from which the shaking spread, says Chris Wills, a geologist with the California Geological Survey in Sacramento. By contrast, a magnitude 7 quake would have involved a much longer piece of the fault, he says, "with lots more energy radiating from every patch along the fault.”

In addition, geology helped confine the area of greatest shaking. The quake took place in a valley where "soft sediments amplified the shaking," while the surrounding mountains and their harder rock formations limited it, writes Ingrid Johanson, assistant research seismologist at the University of California at Berkeley's Seismological Laboratory, in an e-mail.

Even so, the temblor led to 90 breaks in the city's water mains, which serve 90,000 people, cut off electricity to 94 customers, and left 33 buildings in the city uninhabitable.

Shortly after the quake, city officials declared a state of emergency. Gov. Jerry Brown followed suit Sunday morning. By some estimates, insured losses from the quake could run between $500 million and $1 billion.

The quake has provided the latest test for an earthquake early warning network researchers have been developing since 2008 – one that it seemed to have passed with flying colors.

The network currently consists of 377 seismic monitoring sites run by a collaborative including US Geological Survey, the University of California at Berkeley, and the Swiss Federal Institute of Technology in Zurich. 

The stations feed their data into three processing centers, which crunch the numbers to provide warnings that an earthquake has happened and quickly compute the quake's magnitude – quickly enough to provide several seconds of warning before the heaviest shaking arrives.

The system alerted the Seismological Laboratory at Berkeley of the quake five seconds before the initial shaking began, and 10 seconds before peak shaking arrived, Dr. Johanson says. The app also displayed the overall intensity of the shaking expected – in Berkeley's case, light shaking.

A few seconds may not sound like much lead time, but it's enough to get people under desks, halt rail traffic, shut down or move away from dangerous machines or chemicals at work, or prevent cars from starting across bridges or into tunnels, the researchers say.

Japan and Mexico have early warning systems in operation, with Japan boasting a network of nearly 1,100 monitoring stations around the country.

Japan's network underestimated the strength of the 2011 Tohoku earthquake, which triggered tsunamis that lead to meltdowns at three of six reactors at the Fukushima Daiichi nuclear power plant. But it did provide precious few seconds for people to duck under desks or under a steel door frame.

The California system has been in a testing mode since 2012. The results have been significant enough that some of the project's industrial partners already have integrated the warning system into their operations, notes Johanson. These range from the Bay Area Rapid Transit System to Google and Southern California Edison.

In order for the system to go public, however, the research team behind it argues that the number of monitor stations needs to be increased significantly. This would fill large geographic gaps in the current network, which was designed for research rather than set up as an operational warning system. And it would allow for a more dense network, since the fastest warnings come in part because the monitor station is near a quake's epicenter.

In a paper published in the journal Seismological Research Letters last December, UC Berkeley seismologist Richard Allen and colleague Serdar Kuyuk recommended that the network expand to include another 2,500 monitoring stations around the state – all of which would need upgrading to match the technological chops of the initial 377 sites. And some would have to be moved. Ideally, the stations should be about six to 12 miles apart, the researchers said.

The concept received a boost last September when Governor Brown signed a bill requiring the establishment of the network. The bill, however, prevented the state's Office of Emergency Services from dipping into the state's general fund to build the network.

Identifying sources of funding remains one of the largest hurdles to setting up the network, Johanson says.

Researchers estimate that it will cost $80 million to set up the system and run it for five years, with the system going public within about two years after the funding sources have been pinned down.

In the meantime, scientists from the US Geological Survey, the University of Washington, Berkeley, and Cal Tech have been working with funding from a private foundation to expand the warning system to cover the entire West Coast.