Earthquake: unpredictability is its middle name

Published in The Hindu on March 24, 2011

Fukushima DaiichiHow safe are nuclear plants located in countries that have faults cutting across them or located close to tectonically active regions? How reliable and helpful is seismology (the science of earthquakes) in forecasting the probability of occurrence of damaging quakes?

The 6.3 quake that struck Christchurch in New Zealand on February 22 killed nearly 65 people and caused widespread damage to property. Were scientists expecting a killer quake to strike the city? The simple answer is, no.

Blind fault

The Christchurch quake occurred along a “blind fault.” Blind faults have no surface expression, and scientists are plainly ignorant of their existence. Christchurch is not an isolated case.

The 7 magnitude quake of January 2010 that hit Haiti is another example. Blind faults have played a role in the not so recent past as well.

The 7 magnitude Loma Prieta earthquake that struck the San Francisco Bay Area in 1989, and the 6.7 magnitude Northridge quake of 1994 which hit southern California were caused by release of energy from blind faults. May 2003 saw a 6.7 magnitude tremor from a blind fault striking northern Algeria.

Japan had also witnessed a blind fault ripping apart structures way back in 1995. Kobe city that was struck by a quake will be most remembered for the damages caused (running to hundreds of billions of dollars) and thousands of fatalities.

As scientists continue to discover new blind faults, as in the case of the Puente Hills Fault that runs right under downtown Los Angeles, they have come to realise that the earth’s crust, just a few kilometres below the surface, has an innumerable number of such weak zones.

Far away places hit

If blind faults are a great cause for concern, high magnitude quakes striking one part of the globe can result in smaller intensity quakes in places located as far as a few thousand kilometres away.

Luckily, till date, tremors, triggered by high magnitude quakes, striking places hundreds of kilometres away have been low-intensity ones.

Scientists never believed these tremors were possible till the 7.9 quake that rocked Alaska in November 2002, and the 7.3 magnitude earthquake of June 1992 that struck the town of Landers in California, set off jolts thousands of kilometres away.

If the Alaskan quake triggered tremors as far as 3,200 km away in the Yellowstone National Park in the U.S., the Landers quake led to smaller ones, again, in the Yellowstone National Park.

Scientists found that the 2004 Sumatra earthquake triggered quakes even on the opposite side of the earth in Ecuador.

In fact, according to a study published in May 2008 in the Nature Geoscience journal, 12 of the 15 major tremors (between 1992 and 2006) greater than 7 magnitude caused quakes even thousands of kilometres away.

Quake clustering

If places far away from major quakes are jolted, there is plenty of evidence to show that quakes come in clusters following a giant tremor. For instance, the 9.1 magnitude Sumatra quake of 2004 set off a series of nearby quakes, including one five years later. These are distinctly different from aftershocks.

But do giant quakes come in clusters? Though statistically significant evidence of a number of giant quakes occurring in clusters is not available, it is a fact that within a time interval of less than seven years there have been three giant quakes — December 2004 Sumatra quake of 9.1 magnitude, February 2010 Chile quake of 8.8 magnitude, and now the 9 magnitude quake off Sendai.

Similarly, six of the 16 greatest quakes have occurred between 1950 and 1965.

Another instance of comparably large quakes originating from the same or neighbouring faults is the 8.3 magnitude Kuril Islands event (north of Japan) of November 2006 followed by another one of 8.1 magnitude within two months (January 2007).

Stress loading

It is well known that release of stress during an earthquake can in turn load up the same fault or adjacent faults with stress.

So was that the case even with the Sumatra and Kuril Islands events?

Clustering has been seen even when the initial quake has not been a giant one. A 2009 paper in Nature cites how a series of quakes in 1992 shook California’s Mojave Desert in quick succession.

It started with the 6.2 magnitude Joshua Tree quake of April 1992 followed by two quakes in June 1992 — the 7.3 magnitude Landers quake and the 6.5 magnitude Big Bear quake, and finally the 7.1 magnitude Hector Mine earthquake in 1999.

Are cratons safer?

Cratons, the old and stable parts of the continental crust, which are far away from the plate margins, are generally considered to be free from big quakes.

But the late 1811 and early 1812 quakes of more than 8 magnitude that jolted the New Madrid region in Mississippi, U.S. defied that assumption.

So will the 9 magnitude quake that rocked Japan on March 11 lead to minor jolts in far away places, big quakes in the neighbouring regions, and giant quakes in other regions of the world in the coming years?