14 September 2011
Just before the recent huge earthquake in Japan, electron counts in the atmosphere high above the epicenter took a surprising turn, a new study indicates. Measurements gleaned from GPS satellites recorded more electrons in the ionosphere over the soon-to rupture fault than expected. A similar uptick occurred before extra-large quakes in Chile in 2010 and Sumatra in 2004, the researcher found.
A tantalizing question for seismologists and atmospheric scientists is whether this high-altitude electron bump, if confirmed by other studies, is a true early-warning signal for devastating earthquakes. Geophysicist Kosuke Heki of Hokkaido University in Sapporo, Japan, who reports the suggestive buildups in a new scientific paper, thinks that it could be.
“The claim that earthquakes are inherently unpredictable might not be true, at least for M9 [magnitude 9] class earthquakes,” Heki writes in an article accepted for publication in Geophysical Research Letters, a journal of the American Geophysical Union.
The study suggests that the total electron content, or TEC, in the ionosphere starts increasing as much as eight percent above background levels prior to massive earthquakes, with the biggest effect above the rupturing fault. The electron buildup before the Japan earthquake started 40 minutes before disaster struck.
The time involved in retrieving and analyzing the TEC data would make it difficult to use the information as a potential earthquake warning, the researcher says. For some scientists, even the notion of an earthquake precursor is controversial. Another earthquake expert, not involved in the study, said the data is interesting and should be studied in the future, but he is not completely convinced that change in TEC is an earthquake precursor.
Days after the devastating magnitude 9.0 Tohoku-Oki earthquake struck Japan, Heki downloaded data from satellites that are part of the GPS Earth Observation Network. He was interested in oscillations of the TEC when acoustic waves echo from the epicenter into the ionosphere.
“I thought I’d see a very strong signal after this earthquake,” Heki says. “And in the course of working on that, I found something strange happened.” The TEC was higher before the expected oscillations occurred than it was after the oscillations, so he took another look at the raw data from the GPS satellites.
When the Tohoku-Oki earthquake struck on March 11, eight GPS satellites were relatively near the epicenter. Normally, satellite measurements of the ionosphere’s TEC vary gradually due to the satellite’s elevation changes. But about 40 minutes before the earthquake, the TEC stopped following the expected pattern, reaching as much as about 8 percent above the background curve, or about 2.3 total electron content units (the equivalent of 23 million billion electrons per square meter). And the satellites close to the ruptured fault showed more of an anomaly than those farther away.
But changes in total electron content aren’t rare, Heki says. Solar flares and other ionospheric disturbances can cause fluctuations. So he looked at TEC prior to other major earthquakes, including the 2010 Chile earthquake.
“I saw almost the same signature as the Tohoku earthquake,” Heki says. And although there were fewer GPS stations operational during the 2004 Sumatra quake, those showed a similar TEC anomaly. Smaller quakes, around magnitude 8.0 and below, don’t appear to have the same TEC increases.
It’s currently not feasible to use total electron content as a warning system for giant quakes, Heki says. There is a dense network of GPS satellites, especially over earthquake-prone areas like Japan, California and Indonesia, but the organization running the GPS network over Japan uploads the data once every few hours. And it takes time to analyze the data and weed out other disturbances to the total electron content like solar flares.
The next step for researchers could be to both determine what causes a change in TEC prior to earthquakes, as well as establish additional measuring stations on the ground to monitor the electric field.
“Whenever the next magnitude 9 earthquake happens, I think we can catch the signal,” Heki says.
Hiroo Kanamori, a professor emeritus of geophysics at the California Institute of Technology who was not involved with the study, says that Heki makes a good argument that something was anomalous in the ionosphere, but he is not ready to call it an earthquake precursor.
“I think it’s worthwhile to document it like this, to see what will happen with the next event,” Kanamori says, “but I can’t be completely convinced.”
One of the reasons, he says, is that there is no physical reason to draw the background reference curve as Heki did, and so it’s hard to say that Heki’s curve is the correct one to use when looking for anomalies in TEC. TEC can naturally be irregular, Kanamori adds, and a magnetic storm around the time of the earthquake could have thrown off the data.
Still, it’s interesting that the deviation from the curve seems to be smaller farther away from the quake, and that similar patterns showed up before the Sumatra and Chile events.
“It’s amazing if something really did happen 40 minutes before,” says Kanamori. “However, there are lots of uncertainties.”
– Kate Ramsayer, AGU science writer