Amateur Radio Public Seismic Network (ARPSN)
Earthquake References

Quake Prediction Gets Shake-Up
By Stephen Leahy via 10.06.05 | 2:00 AM

Researchers in Sweden claim to have developed a new computer model for predicting earthquakes that correctly -- retroactively -- forecast the devastating 2004 Indian Ocean earthquake and tsunami that killed 275,000 people.

Using five years of seismological data from the region including records of 624 quakes, researchers from the Swedish Defense Research Agency, or FOI, studied the enormous stress created by the Indian Plate as it grinds into the Australasian Plate near the island of Sumatra.

They were particularly interested in the relative frequency of major and minor quakes, what seismologists call the b-value.

The lower the b-value, the greater the increase in tension in the Earth's crust, which entails a greater risk of major quakes. While this has been known for decades, the FOI model uses the information in a new way to plot the b-value ratios in time and space.

"We found that all of the major tremors were clearly visible in a time perspective. The b-value dropped drastically before the big quakes," said FOI researcher Leif Persson in a statement.

And the biggest drop-off was four months prior to the Sumatra-Andaman earthquake that measured 9.0 on the Richter scale.

The researchers said the model also accurately plotted the location of the quake's epicenter. And it worked for another more recent quake this year off the coast of Sumatra.

"Using this method, major quakes like the one that caused the tsunami could be predicted better, both in terms of time and geographic area," he said.

The model was originally developed to predict tremors following blasting operations in Swedish mines.

"This could be a very simple warning tool that would be much less expensive than the multimillion-dollar Indian Ocean Tsunami Warning System that's being built," Persson added.

However, Ota Kulhánek, a seismologist at Sweden's Uppsala University who helped develop the model, said it's one thing to interpret the data retrospectively but very difficult to predict future events.

"Earthquakes are extremely complex events. I would not dare use our technique alone to make forecasts," he said.

Kulhánek said the new model would have to be used in combination with other techniques and data.

Indeed, seismologists are divided on whether earthquake predictions or forecasts are even possible.

Methods that accurately predict quakes after the fact have nearly always failed to predict future quakes, said seismologist Jim Dewey of the Earthquake Hazards Program at the U.S. Geological Survey.

And those that appear to work once or twice -- like one developed by UCLA's Vladmir Keilis Borok -- fail on their next big prediction.

Dewey summarized decades of effort to predict quakes as "periods of optimism followed by disillusionment."

But earthquake satellites may soon change that.

Earthquake forecasting has long been hobbled by its lack of data. While it may always be difficult to know what's happening 15 miles deep in the Earth's crust, new satellites with interferometric synthetic aperture radar, or InSAR, will illuminate the stresses in the Earth's surface, said John Rundle, director of the Center for Computational Science and Engineering at the University of California, Davis.

InSAR uses low-angle radar images of a landscape to determine changes in the surface over very broad regions to within a couple of inches. It can detect slight deformations in the Earth's crust, which may indicate strain prior to an earthquake.

In 10 years, such satellites could form a Global Earthquake Satellite System that will be a "major leap forward in earthquake forecasting," said Rundle.

Even without InSAR, Rundle's computational analysis of magnitude 3 or 4 quakes has accurately located earthquake hot spots in California. Although he doesn't use b-values, nearly all quakes in California in the past 10 years have been in his designated hot spots, he said.

"Earthquake forecasting is where weather forecasting was a few decades ago," he said.

Weather is no less chaotic or complex than earthquakes, and accurate weather forecasts have become routine, he added.

With InSAR-equipped satellites and better computer models combining a variety of analytical methods, "earthquake forecasting is definitely possible," said Rundle.

Evidence Mounts for Electromagnetic Earthquake Precursors
By Keay Davidson via 12.14.07 | 6:00 PM

SAN FRANCISCO - Scientists revealed data Thursday that an electromagnetic alarm might have preceded a 2007 earthquake in Northern California. The evidence could offer support to a controversial theory that mysterious and little-understood signals might offer fair warning for imminent catastrophic earthquakes.

Scientists detected the signal Oct. 30 near Milpitas, California, 19 hours before a medium-size quake -- with its epicenter in the Alum Rock neighborhood of San Jose -- shook the region, scientists told Wired News Thursday.

"Alum Rock saw a signal that didn't happen at any other site: It was a series of electromagnetic pulses that were drawn out over eight minutes," said Tom Bleier, a researcher with QuakeFinder, a Palo Alto firm. He cautioned, however, that further study is needed to determine if the electromagnetic signal has "some other cause" besides the quake.

The new data, reported here at the American Geophysical Union annual meeting, was met with some skepticism. But the evidence could be a watershed moment in earthquake detection, a field that has a long and perpetually disappointing history. The discovery could strengthen the case of scientists who suggest that big quakes are preceded by strange signals, including one that may have come before the catastrophic 1989 Loma Prieta earthquake in the San Francisco Bay Area.

"There are at least a dozen theories that predict these (electromagnetic precursors) should occur," said Jacob Bortnik, a UCLA space physicist and a consultant for QuakeFinder.

To test the theory that quakes emit advance warning signals, a small team of California scientists funded by the satellite company QuakeFinder, has installed some 70 electromagnetic sensors across California, including some in high schools, in exchange for satellite internet access. The device is a white box, 4 feet tall, which contains an instrument called a search-coil magnetometer, designed to detect the type of signal that researchers theorize acts as a quake alarm.

At about 1:30 a.m. on the day before Halloween, one of the sensors -- located on the property of a plumber near Milpitas -- detected a puzzling series of electromagnetic pulses. Late that same day, a 5.6-magnitude quake occurred nearby, with its epicenter at Alum Rock, just south of the Calaveras earthquake fault.

QuakeFinder engineers are now analyzing data from the sensor, trying to determine whether it gave advance warning of the Alum Rock quake, said Bleier, who is trained in electrical engineering. He cautioned that the data analysis is only 30 percent finished, and it's premature to say whether the signal emanated from the quake or is due to some other cause.

Skeptics believe the electromagnetic signals might not be due to quakes at all. Rather, they could be caused by sources ranging from the natural to the artificial -- say, from solar activity or from electromagnetic fields generated by auto engines.

Scientists at the AGU meeting Thursday argued whether a particularly dramatic electromagnetic alarm preceded the 7.1-magnitude Loma Prieta earthquake of October 1989, which devastated parts of the San Francisco Bay Area. Antony Fraser-Smith, now an emeritus professor of electrical engineering at Stanford University, detected the signal.

According to Fraser-Smith, the 1989 signal began from an unknown source up to two weeks before the Loma Prieta quake. The signal peaked sharply a few hours before the quake struck.

But critics point out that Fraser-Smith's records don't include evidence of natural, day-and-night variations in the Earth's electromagnetic field, which are normally present in such records, said Malcolm Johnston, a geophysicist at the U.S. Geological Survey office in Menlo Park, California.

Fraser-Smith also had some supporters at the conference. David Culp, now a senior at Purdue, presented evidence he gathered as an undergraduate intern in the Stanford Department of Geophysics in the summer of 2007 that Fraser-Smith's 1989 detection. He emphasized the dramatic spike in electromagnetic intensity hours before the quake. Efforts to explain the signal via nonquake mechanisms are "entirely implausible," Culp said.

A long-time proponent of the earthquake-alarm theory, Friedemann T. Freund, a scientist at NASA Ames Research Center in Mountain View, California, also presented data on the possible mechanism underlying the electromagnetic signals.

He reported the result of a lab experiment in which he subjected rock to high pressures, modeling pre-earthquake conditions. It caused the rocks to develop electrical currents, he said. After relieving pressure on the rock, the electrical current slowly faded out -- just like the electromagnetic measurements after quakes.

"Either there is a big devil down there moving magnets back and forth, or there is some kind of physical effect causing (these signals)," Freund said.

Johnston, however, said Freund's electrical currents would be "short-circuited" by the abundant groundwater in underground rock.

Still, with evidence mounting that the signals might be real, some scientists are calling on the federal government to develop a network of electromagnetic sensors to detect such signals before quakes. Even skeptics agree more detection is necessary.

"We need a much more comprehensive (electromagnetism-monitoring) network," Johnston said.

But Bill Ellsworth, a prominent Geological Survey geophysicist, said that in the absence of an infinite amount of federal funding, first priority should go to the development of more seismic-detection networks that -- unlike earthquake alarms -- are based on well-understood physical principles.

Prize-Winning Seismologist Embraces Earth's Little Faults
By Marty Graham via 03.20.08 | 12:00 AM

Seismologists say their science can be easily divided into two eras: before and after Hiroo Kanamori.

A California Institute of Technology professor emeritus and newly christened Kyoto Prize honoree, Kanamori led the earthquake research that resulted in the moment-magnitude scale, a better way to measure quakes than the Richter scale it replaced in the 1970s. The moment-magnitude scale is not a pure instrument measurement. It accounts for the area of land that ruptures in a quake and how far it moves.

Today Kanamori is interested in a different kind of moment: the first moments of an earthquake, when long-period waves can give warning of the intensity and damage to come. Kanamori sees an opportunity to use those few seconds of lead time to protect lives and property from catastrophe. He now commits much of his time to developing and raising money for early-warning systems in places where systems are inadequate -- like California and the rest of the United States.

Kanamori sat with for an interview as part of the celebration of the Kyoto Prize, which honors scientific advances, arts and the human spirit.

Wired: What do you see as your greatest accomplishment?

Hiroo Kanamori: Finding ways to understand and better measure earthquakes. Determining the size was the first step. Before the 1970s, seismic measurements were based on seismology readings, not the energy of the quake. We could not measure the long period of the wave of activity. If we couldn't measure it, we couldn't understand the intensity and the damage of these quakes.

The worst earthquakes have a wave that is slower and lasts a much longer period. You don't feel them the way we can feel an earthquake where the wave is condensed into a short time period. But the long, slow wave produces the greatest damage. And tsunamis are produced by the long-period wave, often on the sea floor. So we have a new scale that can measure the tsunami-genic potential of an earthquake.

Wired: We can't feel this long-period wave?

Kanamori: They are very slow. We feel the shorter waves as shocks. Tall buildings are more sensitive to them -- the effect of a seismic wave on structure depends on the size of the wave, and long-period waves are far more serious.

Wired: Why can't we predict earthquakes?

Kanamori: We can, but not in a way that is useful by the minute or the day. We can see where energy is building up, we can see where there is likely to be a rupture in the next thousand years. But we can't tell you it will be this week. And we can't predict all of them.

Wired: Your focus now is on early-detection-and-warning systems for tsunamis.

Kanamori: The most important thing we can do is send out information very quickly to transportation, utility and emergency systems to slow down trains and contain chemicals, to keep the damage at a minimum. Japan has early-detection-and-warning systems where seismic sensors send out warnings. During an earthquake in 2004, the system was able to slow the bullet train that was going 150 miles per hour. It still crashed, but at a much slower speed than the 150 miles per hour. Even a small amount of time gives people in elevators, construction workers, people on bridges a chance to be safer.

Wired: You've crossed the barrier between academia and application with those systems.

Kanamori: In seismology we've studied these things for two or three decades, but the information wasn't utilized in technology and engineering, particularly in this country. In Japan, there are 40 smart buildings now that connect to the warning systems and shut off the gas and electricity, that are built on rollers that get the elevators to the next floor and stop so people aren't trapped in them.

Wired: What do you do with the Pacific Tsunami Warning Center?

Kanamori: Our goal is to really develop seismic instrumentation. We aren't up to that stage of being very useful yet, because we don't have the funding. And we are now implementing tsunami-warning systems, though we knew they were possible before [the 2004] Sumatra earthquake and tsunami. It would have been much nicer if we had it before then. But we know the current system isn't very good for this kind of event.

Wired: We Californians like to think our little quakes, our 3s and 4s, are keeping us safe from the Big One. Is there any truth to that?

Kanamori: Not really. The energy released by a 3.5 is so small the total energy is not enough to dissipate the built-up tensions that produce an 8 or 9.

In California, the main fault structure is the San Andreas. Whenever we have a main we have secondaries nearby, and they are very difficult to find because they are under the ground. Every place has faults, so there's no point in worrying about it.

The big event is very rare. Even the Sumatra event is a once-every-3,000-years event. To live where there are no earthquakes would mean living in the middle of the Sahara Desert: It means having no mountains, no scenery, no ocean. It would be very boring.

Earthquakes are a very rare event. The thing is, a single occurrence can wipe you out. The government has to think about it -- they can't ignore it because it's rare -- because of public safety. But what people should do, it's all in the front pages of a phone book.

Wired: Do you have three days of food and water stashed?

Kanamori: Yes. And my house is bolted to the foundation, and my bookcases are anchored to the wall. But after that, I don't think about it anymore. My home is on a fault, Cal Tech is on the Raymond Fault, a serious fault. It's one of the worst places to live. But you can't spend your time worrying about it.

Valid CSS!Spacer Valid XHTML 1.0 Transitional SpacerLink to