Damage from the 2001 Nisqually earthquake in Seattle. Photo: Kevin Galvin via FEMA.
Damage from the 2001 Nisqually earthquake in Seattle. Photo: Kevin Galvin via FEMA.

The magnitude 6.0 earthquake in Napa, California on August 24 produced a new level of interest in developing earthquake early warning systems (EEWs) for the West Coast.

ShakeAlert, the early warning pilot system at UC-Berkeley, worked as planned during the Napa quake, delivering up to 10 seconds of pre-notice to system enrollees in the San Francisco Bay Area before the shaking began for them. (See report: Quake alerts’ success spurs effort to complete early-warning system)

shakealert11In the Northwest, and the rest of the U.S., we’re still getting used to the idea that an earthquake, like a tornado, might be, if not quite predictable, at least something that we can prepare for in the seconds of warning given in the gap between its initiation and the time when it impacts us.

Japan, Mexico, Taiwan, Turkey, and Italy have rolled out full-fledged earthquake early warning systems and used them to shut down industrial and medical processes, avert transportation disasters (imagine an airplane halting its takeoff or landing attempt because the pilot knew a quake was about to hit), and give ordinary people an early chance to take cover from a quake.

With all this in mind, John Vidale, head of the Pacific Northwest Seismic Network (PNSN) headquartered at the University of Washington in Seattle, answered some questions about how EEW systems work and their benefits to the countries that are using them.

What has been the experience in Japan and other countries with warning systems, in terms of how behavior changes when the system sends a message and people know they’re about to experience an earthquake? For that matter, does the general mentality toward earthquake preparedness change when people know they might get early warnings?

Vidale: “What we know is that the Japanese much prefer to have EEW at their fingertips than not.  In Mexico, the warning systems that have been installed also appear well received.  Quantifying these reactions is difficult. Ann Bostrom here at the UW has looked at some polls about actual actions during earthquakes and attitudes, but I’m not qualified to describe them myself.”

What are the biggest value things—in terms of casualties and property damages—that could be done in the seconds a warning system gives society to prepare for an earthquake?

John Vidale
John Vidale

“My personal opinion is that forecasting what is likely to go wrong is a business fraught with uncertainty.  The Japanese thought their reactors were safe.  Neither the Japanese nor the Sumatrans were much aware of the possibility of a magnitude 9 quake where each had one strike.

Another example is that the dam that failed to nearly flood 80,000 people in the 1971 San Fernando earthquake was not thought to be so weak. As individual lives are assessed at $6 million each by insurance companies and others in life-saving fields, even modest reductions in loss of life in earthquakes would justify the system. So these are opinions rather than documented facts.

There are three areas where big savings and public demand warrant an EEW system:

  1. Loss of life.  People in vulnerable places can gain some protect – on trains that might hit landslides, planes attempting to takeoff land, cars that might collide, lose control, or be on stretches of roadway that can collapse, people on ladders and in cranes, people who can prevent fires or move away from where fires might start, parents who can move to protect their children, there are endless possibilities with some chance of helping forestall death and injury.
  1.  Loss of expensive equipment, operations, and ongoing transactions.  Chemical plants, chip manufacturers, computer operations – most would prefer to temporarily suspend operations and shift fragile equipment out of harm’s way, given even a couple of seconds of warning.
  1.  Peace of mind.  Just knowing that one will hear in advance how much shaking is coming and when will be the worst of it is what modern folks demand.  If it can be known, they want to know immediately.
Recent earthquakes in the  NW.
Recent earthquakes in the NW.

Is there a decent estimate of how many lives were saved and how much damage was averted by early warning systems in the Japan 9.0 magnitude earthquake and tsunami on 3/11/11?

Vidale: “It is clear that early warning saved very many lives in Tohoku.  20,000+ people were killed by the tsunami, only a few percent of the population living in the inundated area, and every second of warning probably saved at the very least 10 or more lives.  If the system had been more accurate, even fewer would have died.  I know of no estimate of damage saved.”

What are the different amounts of alert times given for different types of quakes?

Vidale: “Basically, if an instrument is right next to the earthquake, one gains about a second of warning for every 3 kilometers away one is from where the earthquake starts.  So at 10 kilometers distance, one might get 3 seconds warning.  If the seismometers are farther away, at a similar distance to the vulnerable sites, one gains only 1 second per 10 kilometers of distance, because then one must first observe the P waves, which take time to travel, then predict the S waves. (“Similar distance” means that the seismometers are about as far from the quake epicenter as the area receiving the warning, but not necessarily on the same side of the epicenter as the area receiving the warning.)

The real case is more complicated. It takes a second or two to compute the warning, but big earthquakes take time to build to maximum power, sometimes tens of seconds.  Sometimes we can probabilistically estimate which direction an earthquake rupture might go, that can allow quite a bit of warning time.

Roughly, subduction quakes allow considerable warning time, up to 5 minutes, Nisqually-type events 5-10 seconds, and crustal faults times from 0 to 5-10 seconds for the closest, most shaken sites.”

Many of the countries with early warning systems deployed them only after experiencing a devastating earthquake. It seems that for many types of natural hazards—the Oso landslide for example—people have to experience the disaster in order to be convinced that they need to respond to the threat. I suppose that’s your greatest ongoing worry: that we’ll wake up to the need for an early warning system only after a Big One hits.

Vidale: “Correct, we’d like to implement sensible measures now rather than waiting to have their need made crystal clear by catastrophes that could have been softened by earthquake early warning.”

Editor’s note: Here’s a short documentary film on early warning systems, created by University of Washington grads Ty Kelly and Mahesh Punyamurthula. In the film, Vidale estimates that an early warning system for the West Coast would cost $100 million to $200 million, and cost about $10 million annually to maintain and operate. In the Northwest, the estimates are that a system could give a few seconds to three to four minutes of warning.

Early Warning from jsmags on Vimeo.

Arne Christensen writes about Pacific Northwest earthquakes on his blog. You can follow him on Twitter @NWquakes.

Like what you're reading? Subscribe to GeekWire's free newsletters to catch every headline

Job Listings on GeekWork

Find more jobs on GeekWork. Employers, post a job here.