Tectonic plates reverse movement before major earthquakes

These earthquakes are among the 10 strongest ever measured: the 2011 Tohoku-oki seaquake off the coast of Japan registered 9.0 on the moment magnitude scale, triggering the devastating tsunami that destroyed the Fukushima Nuclear Power Plant. It was the fourth-strongest quake in the last 100 years. The 2010 seaquake off the coast of the Maule region of Chile, occupies sixth place on the list.

Researchers from the Geoforschungszentrum Potsdam (GFZ) along with collaborators from Chile and the United States have now discovered astonishing similarities in the months before both quakes: multiple strange reversals of ground motion — an extremely slow "wobbling" of the continental plate, so to speak.

Lead author, geophysicist Jonathan Bedford, along with a team of geodesists, geologists and seismologists, evaluated the movements of ground stations of the Global Navigation Satellite System (GNSS).

These ground stations continuously track the distances to multiple GNSS satellites that orbit the earth in elliptical orbits. After these distances have been established, and given the estimates of the satellite positions, geodesists are able to define the motion of points on the Earth within a terrestrial reference frame.

In Japan, there is a very dense network of ground stations. Although the network in Chile was not as dense in 2010, it was sufficient to obtain the data.

Read more: Moving hotspots: Scientists explain mystery bend in Hawaii-Emperor volcano chain

Whole continental plate 'wobbles'

Bedford and his colleagues analyzed how the ground stations in Japan and Chile had moved in the five years before the two quakes. They noticed that the motion of the continental plate on which the stations are located had been reversed several times in the last five months before the quake (in the case of Japan) and seven months (in the case of Chile). The researchers published their results in the scientific journal Nature.

Both plate boundaries are subduction zones. Subduction is the process of one plate diving under another. In both study regions, the oceanic plates are subducting underneath the continental plates and meet at a place called the trench, which for both cases is submerged under the ocean. Normally, the continental plate is pressed by the oceanic plate and thus pushed away from the trench. However, the geophysicists have now discovered that this movement was first reversed in the direction of the trench, then away from the trench, then back towards the trench again.

Read more: Volcanoes and earthquakes: The Pacific Ring of Fire

Same movement over thousands of kilometers

They call this phenomenon "wobbling," due to its appearance in the GNSS time series. The amplitude of this movement is not particularly great — it was only between 4 and 8 millimeters —  but Bedford points out that this is significant compared to the relative plate motion that can be a few centimeters per year. Furthermore, the spatial extent of the signal extended thousands of kilometers along the plate boundaries.

"It is a common assumption that deeper subduction proceeds at a fairly constant speed in between large earthquakes," says Bedford. "Our study shows that this assumption is an oversimplification. In fact, its variability might be a key factor in understanding how the largest earthquakes nucleate." 

With global satellite tracking now becoming better and better, and with accurate data available for the first time in decades, earthquake researchers have an ever-increasing ability to make such observations.

"We can now trace movements back decades," says Bedford. "In the next stage, we'd like to monitor the changes in near real time."

Not suitable as an early warning system

Until now, seismologists have been more able to say where large earthquakes are likely to occur — less able to say when. By simply calculating the magnitude of the last large earthquake in a region and knowing the average relative plate velocity, one can estimate when that fault will be mature enough to sustain a repeat event, although there are great uncertainties in this approach. This is because sometimes a fault will rupture only over a smaller area (e.g. with a "large" magnitude 8 event) and other times it will rupture over many magnitude 8 regions all at once (e.g. in a "mega" magnitude 9 event).

Could an observed untypical plate movement therefore give us a better warning of an imminent earthquake? Not really. 

"It would not be wise for a geophysicist to issue such a warning," is Bedford's sobering answer. "The observed signals of this study are not necessarily precursory movements of a major quake."

More research is necessary and as a matter of principle, people in known earthquake areas should not let their guard down. 

"The general public should always be prepared," Bedford warns.