Earthquakes Along San Andreas Linked With Shifts in Earth's Magnetic Field

Detectable changes in the Earth's magnetic field can be observed a few days before major earthquakes in California.

According to a paper published in the journal JCR Solid Earth, two to three days before an earthquake, scientists can measure a significant shift in the magnetic field. This was measured across 19 earthquakes in California.

"It's a modest signal," Dan Schneider, director of QuakeFinder, an earthquake research department in Stellar Solutions, and co-author of the paper, told EOS. "We are not claiming that this signal exists before every earthquake, but it is very intriguing."

Earthquakes are caused by the sudden shifting of tectonic plates against one another, resulting in large amounts of energy being released as seismic waves through the Earth.

Earthquakes can be incredibly destructive and result in mass loss of life, often due to collapsing buildings. Being able to predict when an earthquake is about to hit would be incredibly helpful, as it would allow some time for evacuations from especially dangerous areas. However, until now, there has been no reliable way to predict when or where an earthquake would occur, except that they are more frequent near tectonic fault lines.

"What distinguishes this worthy effort from many previous "earthquake prediction" studies is that the authors used rigorous machine learning techniques and statistics to explore many earthquakes and a vast amount of data, rather than selectively picking and choosing," Roland Burgmann, an earth and planetary Science professor at the University of California, Berkeley, told Newsweek.

Having some degree of warning for a coming earthquake is of particular importance in California, which sits on the San Andreas fault line between the Pacific and North American tectonic plates. According to a 2006 Scripps study, the fault is due another "big one." Being able to predict and evacuate for a high-magnitude earthquake would potentially save many lives.

"Depending on the nature of the warning (i.e., how accurate is it in predicting 1) the size, 2) the place, and 3) the timing), it could conceivably save hundreds, if not thousands of lives," David D. Oglesby, a professor of geophysics at University of California, Riverside, told Newsweek.

"A very large earthquake on the San Andreas in Southern California, for example, could be devastating, causing possibly over a thousand immediate deaths, the isolation of the Los Angeles Area from the rest of the country, and long-term impacts nationwide. A warning 2-3 days in advance would give ample time for numerous safety measures to be implemented. Of course, proper preparation of people and infrastructure well ahead of time is crucial for a warning to be effective; there would need to be very specific plans for what to do in advance, with a mechanism to carry them out."

While these magnetic field shifts are a significant step forward, Burgmann urges that this does not surmount to a reliable predictor for a coming earthquake.

"Nonetheless, my first impression is that, despite the great effort, the results are still not conclusive about there being a clear electromagnetic precursor," Bergmann said. "The authors emphasize that 'the observed effect size is not directly useful for earthquake prediction.' So, there is still no indication that a reliable earthquake-prediction method is around the corner, but the results may well encourage renewed efforts into finding and trying to understand interesting phenomena that happen just before some large earthquakes."

Barbara Romanowicz, a professor of the Graduate School at University of California, Berkeley, agrees: "The fact that there might be some magnetically detectable changes in the rock properties before an earthquake is conceivable, but even if some of the observations turn out to be valid, this does not mean that they can be used to predict earthquakes. For that to be useful it would have to be reliable, i.e. one would need to be able to recognize the precursory signal among myriads of other causes for "magnetic shifts" that have nothing to do with earthquakes," she told Newsweek.

According to Romanowicz, the best predictor for earthquakes that we have at the moment are "earthquake early warnings," which mean that the earthquake is already happening, and the first jolts have been picked up by seismographs located right above the hypocenter where the tectonic slip occurs.

"Because the damaging 'shear' earthquake waves travel slower than signals sent over phone lines or radio, you can issue a warning that shaking is coming. It can give a couple of seconds of warning. The farther you are from the earthquake the better it works. It is used to stop trains before they derail (for example, the SF Bay Area transit train, BART, is connected to such a system operated across California), shut down critical facilities (e.g. prevent fire truck garage doors from closing etc...) or simply allow people to jump undercover. Such early warning systems have been deployed for a long time in Japan, and more recently in California."