Earthquakes and Prediction

It's early days yet but it looks as though the damage from the Sichuan earthquake that hit this afternoon isn't too bad. (Later note: numbers are rising as they always do. I guess I was being optimistic, a triumph of hope over experience: I recall the first reports out of Aceh only had a few dozen dead) The size of the quake and the fact that it was felt so far away (the 1970s building that houses our bureau in Beijing swayed giddily for thirty or forty seconds) had led me to fear the worst. It now seems that there was a small (3.9 Richter) quake quake only a few minutes after the Sichuan quake with its epicenter in eastern Beijing. Then a report appeared on the Chinese web and spread rapidly to almost all news portals and bulletin boards. It claimed that the China Earthquake Administration had issued a warning predicting a further quake this evening in Beijing between 10 and 12 of a magnitude between 2 and 6. The report was later denied by the Beijing Seismological Bureau, and was apparently a hoax.

The warning was somewhat credible because China has put much time and energy into earthquake prediction, though with mixed results. In 1975, scientists ordered the evacuation of 1 million people the day before the M7.3 Haicheng earthquake, a decision credited with saving up to 150,000 lives. But the next year, they totally failed to predict the massive, M7.8 Tangshan (which is only 75 miles from Beijing) earthquake, which killed somewhere between 250,000 (official figure) and 650,000 people. And they don't seem to have called the Sichuan quake either.

Of course, there are different kinds of quakes with different precursors, but many seismologists remain deeply skeptical about prediction. I recall traipsing through rice fields after a distinguished seismologist in Sumatra a few years back (he was doing research into the Aceh earthquake and tsunami) and listening to a long harangue that basically put predicting earthquakes on the same level as astrology.

(Here's a link to a Wikipedia entry on earthquake prediction that is fairly balanced).