New Scientist
February 9, 2005

http://www.newscientist.com/article.ns?id=dn6991

Power of tsunami earthquake heavily underestimated

The earthquake that created the devastating Asian tsunami on 26 
December 2004 was three times more powerful than first thought, say 
researchers analysing long-period seismic waves.

  The finding could upgrade the quake to the second strongest ever 
recorded and explain why the tsunami caused such great damage across 
the ocean in Sri Lanka and India.

Earthquakes are classified on the Richter scale by their 
largest-amplitude seismic wave. These seismic waves come in a variety 
of periods, or wavelengths - but only the most powerful quakes pack a 
lot of energy into long-period waves.

Seismologists initially used seismic waves with periods of about 300 
seconds to set the magnitude of the Sumatran earthquake at 9.0 - making 
it the fifth most powerful event on record.

  Now, seismologists Seth Stein and Emile Okal at Northwestern 
University in Evanston, Illinois, US, have scrutinised seismograms 
taken from 7 stations around the world in the week or so following the 
earthquake. They looked for the longest-period waves possible - those 
lasting about 3200 seconds (53 minutes).

"We found, to our surprise, that there was three times more energy out 
there than at the 300-second period," Stein told New Scientist. "It was 
colossal." The new work reclassifies the earthquake on the logarithmic 
Richter scale at magnitude 9.3 - second only to the 9.5-magnitude quake 
recorded in Chile in 1960.

  Built-up pressure

The Asian earthquake occurred at the eastern edge of the Indian Ocean 
where, over millions of years, the Indian tectonic plate has been 
disappearing beneath the Burma plate. This "subduction" zone had been 
locked for perhaps 200 years before the built-up pressure was finally 
released in the slippage of 26 December.

The Burma plate rebounded upwards by about 10 metres at the quake's 
epicentre - setting the deadly tsunami waves in motion. And the process 
continued along the border between the two plates, causing the earth to 
rupture along the fault line - running from south to north. But 
seismologists are not sure exactly where the rip stopped.

Some think the rupture only made it through the southern third of the 
1200-kilometre-long zone that was rocked by aftershocks. "But if the 
earthquake is three times more powerful then previously believed, 
that's telling you the fault area is three times bigger," says Stein. 
"We think the entire aftershock zone ruptured." The northern two-thirds 
of the zone may have taken longer to slip, which is why its energy was 
released in longer-period waves.

This could be actually be positive news for survivors living near the 
zone. Having released such a large amount of energy, Stein thinks it 
will take another few hundred years for the zone to build up the strain 
necessary for another huge earthquake.

  Localised tsunamis

But he warns that smaller earthquakes could still occur, perhaps 
spawning localised tsunamis. Furthermore, other locked sections of the 
fault - further to the south, near Java, for example - could still 
rupture catastrophically.

If the rupture did indeed occur along the whole length of the 
aftershock zone, it could explain why some distant regions were so 
devastated by the tsunami. While the lower third of the zone directed 
tsunami waves to the southwest, the upper portion has a different 
orientation and sent waves due west - straight towards hard-hit Sri 
Lanka and southern India. However, other factors, such as the 
topography of the sea floor, may also explain why the waves gathered so 
much force in those regions.

Other seismologists have also calculated that the Asian earthquake was 
significantly larger than initially thought. Teh-Ru Alex Song, at the 
California Institute of Technology in Pasadena, California, US, and 
colleagues used long-period waves from about 20 seismometers around the 
world to confirm that the earthquake was two to three times more 
powerful than magnitude 9.0 .

But he says it is not clear yet how fast or slow the slip proceeded 
along the fault. The group arrived at their preliminary result on 
Sunday and will continue to refine their analysis.

Song hopes seismologists will develop a technique to analyse and convey 
the magnitude of any earthquakes that could spawn tsunamis as they 
actually happen - information that could come from waves with periods 
of 200 to 500 seconds.

  "You need that kind of index so everybody around the world knows the 
magnitude and you can issue a tsunami warning to local people," he told 
New Scientist.