Photograph by Paolo Patrizi, Alamy

Read Caption
A 2007 earthquake-preparedness drill in Tokyo (file photo).

Photograph by Paolo Patrizi, Alamy

Mystery of Slow Earthquakes Solved?

Unfolding over months, the little-understood quakes may help stop the big ones.

For years scientists have known about slow earthquakes—temblors that last days, weeks, or even months.

But do these creeping quakes increase or decrease the threat of "regular" earthquakes—the big, quick kind? A new study may hold a key to the answer. (See earthquake pictures.)

Until now seismologists have been aware of two major types of slow earthquake, deep and shallow. (Video: Earthquake 101.)

Deep slow earthquakes occur about 18 to 24 miles (30 to 40 kilometers) underground and last several minutes to several days. They produce tiny tremors, each no stronger than 1 or 2 on the magnitude moment scale, which measures earthquakes based on the seismic-wave energy they release.

Shallow slow quakes happen at depths of about three miles (five kilometers) and can be as strong as magnitude 3.5 or 4.

Like other earthquakes, these slow quakes release built-up tension in shifting rock layers. Once the layers release that energy, they can "relax" for a while, making additional earthquakes less likely in the near future (quick guide to plate tectonics).

Seismic Mystery in the Middle

What happens in the gap between the deep and shallow quakes, though, has long been unknown, according to geophysicist Chris Goldfinger.

Do slow quakes release tension overall in the region? Or do the deep and shallow quakes transfer tension to the middle, "loading" that section for future big ruptures?

To get to the bottom of the slow-quake mystery, study author Hitoshi Hirose and colleagues studied seismic readings of slow earthquakes in an area around the Bungo Suido, a channel in southwest Japan. Slow earthquakes have been recorded here about every six years since 1997.

Deep and shallow earthquakes measured by the team occurred in tandem, indicating that the deeper and shallower motions are linked, perhaps with one set triggering the other.

This, in turn, suggested that the gap was also moving. But since middle-layer quakes have never registered on seismographs, Hirose's team turned to high-precision GPS instruments—which measure general movement, rather than seismic waves—to confirm the middle slow quakes.

Sure enough, the GPS data revealed slow earthquakes in the gap—really slow.

The movements were "too slow to radiate any seismic waves. Slip on the fault lasts for several days to years," said Hirose, of Japan's National Research Institute for Earth Science and Disaster Prevention.

And though the mid-level quakes were slow, they weren't exactly weak, he added.

Each temblor was about a magnitude 7, roughly the same as Haiti's disastrous 2010 earthquake—the quakes only seemed like nothing because all that energy was released over months, rather than seconds or minutes.

Slow Earthquakes May Signal Buffer Zones

The new find is useful, because it shows that slow earthquakes can release the tension that would otherwise result in damaging earthquakes, said Goldfinger, of Oregon State University, who wasn't involved in the new study.

In the fault zone around the Bungo Suido, at least, the entire segment is releasing a lot of strain, but without knocking down buildings, he added.

It's also important because it means that "the area of slow earthquakes might act as a barrier" to the spread of big quakes.

Similar slow earthquakes have been seen in other parts of the world, suggesting that these quakes also might indicate zones that can help limit big quakes, Hirose said.

"But," he noted, "not all regions have all kinds of slow earthquakes"—deep, mid-level, and shallow—at least as far as we know. If stress is not relieved on all levels of slow-earthquake zones, they might not be effective at reducing strain.

On the other hand, he added hopefully, since the quake-buffering zones are tough to detect, the world may be secretly brimming with them.

The slow-earthquakes study is to be published tomorrow in the journal Science.