Turkish earthquake: A wobbly domino falls 

 The tremor that turned life upside down in northwest Turkey last week has helped to put an earthquake theory on more solid footing, according to seismologists who anticipated the disaster.

 The magnitude 7.4 quake occurred near the western end of the North Anatolian fault, a 1,200-kilometer-long tear extending across the northern part of Turkey. Unrest along this fault stems from the slow-motion collision between Arabia and Eurasia, located on either side of Turkey. As these two giant pieces of Earth's outer shell crash together, they force the diminutive Anatolian plate carrying Turkey out of the way.

 During quakes, the bulk of Turkey moves westward relative to land north of the fault. "It's kind of like a watermelon seed being squeezed," says Lynn R. Sykes of Columbia University's Lamont-Doherty Earth Observatory in Palisades, N.Y.

 After several centuries of building pressure, the seed started slipping in 1939. In that year, an earthquake estimated at magnitude 7.9 ruptured the crust along the eastern third of the fault, killing some 30,000 people. Between 1942 and 1967, the fault generated six large shocks progressing westward toward Istanbul like a line of falling dominoes. Last week, the next patch of the fault toppled beneath the city of Izmit.

 Well aware of the pattern, seismologists have long warned about the possibilities of earthquakes in this area. In recent years, scientists started exploring why this series of shocks has stepped so consistently down the fault.

 In 1997, a team of U.S. and Turkish researchers proposed that each tremor along the North Anatolian fault helps set off the next one. Scientists had developed this idea, called stress triggering, while studying progressions of California quakes.

 When the two sides of a fault jerk in opposite directions during a tremor, it relieves stress that had built up over decades or centuries. At the same time, however, it adds stress to some neighboring patches of the fault, says Ross S. Stein of the U.S. Geological Survey in Menlo Park, Calif. He and his colleagues calculated that 9 out of 10 large recorded earthquakes on the North Anatolian occurred in areas where previous shocks had increased stress. Their analysis also pinpointed the two most worrisome parts of the fault, one of which caused last week's quake.

 Stein notes that this work did not predict the shock. The researchers gave only a 12 percent probability that the Izmit section would go by 2026. "In a sense, we said there was a 88 percent probability it would not occur," admits Stein.

 Nonetheless, the earthquake supports the hypothesis of stress triggering, says Gregory C. Beroza of Stanford University. Researchers are currently using this theory to determine how past earthquakes have raised and lowered seismic risk in different parts of California.

 U.S. seismologists see parallels between the Turkish and the San Andreas faults. In both, land moves horizontally during quakes. The faults are the same length, and each splits into branches. The earthquake beneath Izmit happened on the north fork of the fault. "That behavior of the fault splitting up is much like what happens in the San Francisco Bay area of the San Andreas," says Sykes.

 The quake also offers sobering lessons about how ineffective scientists are in influencing building practices, says Nick N. Ambraseys of Imperial College in London. "From the point of view of reducing damage and cutting down the death toll, science alone can do absolutely nothing."

 Turkish building codes accounted for the seismic risk, but the government did not enforce such codes, says Ambraseys. "The important thing is complacency. It doesn't apply only to Turkey or Japan. It also applies to California."

 Earthquakes marched west along the North Anatolian fault this century. These shocks created stress, in red, in the region of last week's quake (bottom). (Stein/from Geophysical Journal International