Then there was the deadly quake and tsunami off the coast of Sumatra in 2004. Chile was then jolted almost as severely a year ago. Now, Japan is recovering after the country's most powerful earthquake in recorded history.

Three of the world's most violent earthquakes occurred within seven years of each other, prompting some experts to wonder if we are living in a megaquake era.

The Pacific Northwest, where an offshore fault is primed to unleash a quake and tsunami on par with the magnitude 9.0 that ravaged Japan last month, is one of the world's most eager to learn the answer.

The majority of geologists doubt that mega quakes occur in clusters. Some people believe they do, despite the fact that no one can explain how quakes thousands of kilometers apart might be related.

On the surface, the history of extremely large earthquakes appears to be divided into two periods of activity: between 1950 and 1965, when seven earthquakes of magnitude 8.5 or higher occurred; and 2004 to date, when three megaquakes (and two huge aftershocks) have occurred.

According to U.S. Geological Survey scientist Charles Bufe, who spoke at the Seismological Society of America's annual conference in Memphis, Tenn., there had been no major earthquakes in the roughly four decades leading up to the tragedy off the coast of Sumatra (one of Indonesia's main islands).

When Bufe and his USGS colleague David Perkins examined the clustering, they came to the conclusion that it wasn't accidental when it came to the tiny subset of genuinely enormous quakes: those with magnitudes of 9.0 or larger.

"Statistically, it's really important," Bufe added. "We believe we're in a higher-than-normal hazard condition for these massive earthquakes."

If they're correct, another magnitude 9.0 or bigger earthquake will occur somewhere in the planet over the next six years, according to Bufe and Perkins. The risk of giant quakes striking at random is roughly 24%.

Several additional scientists who did more comprehensive statistical studies on the earthquake records support the random interpretation.

"We can't do tests," said Andrew Michael of the USGS Science Center in Menlo Park, Calif. "Seismologists are in a really tough situation." The most they can do is sift the existing data in various ways before comparing it to random scenarios to see if any patterns emerge.

It's common knowledge that earthquakes cause more earthquakes, as Michael pointed out. Since the major incident on March 11, Japan has been shaken by hundreds of aftershocks.

As happened twice following the magnitude 9.1 Indian Ocean quake in 2004, a large quake on one stretch of a fault can prime surrounding portions to rupture. Large quakes can also cause lesser quakes up to 1,000 miles distant. In 2002, a magnitude 7.9 earthquake in Alaska sparked rattles and changed the plumbing in Yellowstone National Park's geyser basins.

But, more than five years later and 9,000 miles away, might a megaquake in the Indian Ocean trigger the rupture of a major fault off the coast of Chile?

Michael isn't convinced.

After accounting for aftershocks, he and his colleagues discovered no pattern in the timing of earthquakes with magnitudes ranging from 7.5 to 8.5 that could not be explained by random variation. Only five magnitude 9 or bigger earthquakes have occurred in the last century, he claimed, far too few to make any conclusions.

Michael also discovered no indication that the number of earthquakes increases in the five years after a major quake.

Aaron Velasco of the University of Texas at El Paso studied earthquakes with a magnitude of 7.0 or larger and discovered that their geographic reach in terms of triggering subsequent damaging quakes is limited. Big quakes do produce moderate-sized daughter quakes, 5.0 or more, but only within 600 miles of the epicenter, he added.
With just around 100 years of data and a far from full knowledge of earthquakes, it's critical to keep asking questions about probable megaquake connections, according to Rick Aster, the seismological society's outgoing president.

The prevailing wisdom can be proven incorrect, as it was when the faults off the coasts of Japan and Indonesia suddenly broke.

"Many of these subduction-zone parts were previously thought to be incapable of magnitude 9," said Aster of the New Mexico Institute of Mining and Technology. "These earthquakes have completely broken that confidence."
Scientists hunting for evidence of ancient quakes in the Pacific Northwest extract sediment cores and dig trenches and detect traces of grouping through time.

The shallow faults that run across the Puget Sound region appear to have gone through phases of activity and inactivity. The Cascadia Subduction Zone, a dangerous fault where the ocean floor sinks beneath the continent off the coasts of Washington, Oregon, and Northern California, is no exception.

On one fault segment, Oregon State University's Chris Goldfinger discovers evidence of three to four earthquake clusters in the previous 10,000 years.

In an email, he said, "The clusters are genuine." "They are not a meaningless product."

He's not sure what it is, or if it has anything to do with worldwide earthquake clusters.

Michael believes that a convincing explanation for how megaquakes connect over long distances would provide scientists with a better starting point for future research.

The Japanese earthquake shook the world and shifted the Earth's axis by many inches.

"We have the Earth ringing like a bell after a very large earthquake... and this might last for days, weeks, or months," Bufe explained.

The movements are minute, yet they are pervasive. Large earthquakes also cause waves in the planet's stony crust's taffy-like mantle layer, which spreads slowly.

Bufe suggested that these little oscillations may further weaken a defect that was already on the verge of rupturing.

Michael and others are still skeptical.

But, as Velasco pointed out, scientists aren't sure how earthquakes like the one in Alaska in 2002 cause lesser quakes as far away as Yellowstone.

"In terms of comprehending the earthquake process, we still have a long way to go."