Turkey Quake Provides Lessons for California : Research: U.S. experts say the shock wave moved in unexpected ways, which could occur in local valleys.

Just days before this city collapsed, James F. Dolan of USC was here, snapping pictures of homes he was certain would never survive an earthquake.

“I was going to use them in a class on how not to build a building for an earthquake,” said Dolan, a geologist at the Southern California Earthquake Center at USC.

After Turkey’s catastrophic quake, Dolan returned to see his prediction sadly fulfilled. Adapazari and the cities around it lay in ruins, crumbled monuments to weak laws and shoddy construction.

“It’s the same lesson we’ve learned over and over again,” said Dolan, whose class begins Tuesday. “The overriding problem here was bad building practices.”

Poor construction was perhaps the dominant factor that allowed this month’s earthquake to cut such a sweeping path of destruction through the valleys and hills east of Istanbul. Cheap materials, faulty design and a culture of corruption helped build the cities that fell so fast.

But Dolan and the dozens of other scientists who descended on this area after the quake are picking through the rubble for other clues to Turkey’s tragedy that might also be used to understand quakes of the future--including any in Southern California.

One thing nearly all the scientists agree on: In terms of destruction, this quake ranks among the worst seismic disasters of modern times. Virtually every factor that could contribute to damage and death seemed to conspire to produce a truly catastrophic result.

“I doubt I will ever see greater destruction than this,” Tom Holzer of the U.S. Geological Survey in Menlo Park said as he walked between rows of tumbled apartment houses. “How could it be worse?”

Earthquakes are manifestations of a series of seismic phenomena that are still little understood, and they are virtually impossible to predict. Yet one of the remarkable things about Turkey’s quake was that, when it struck at 3:02 a.m. on Aug. 17, few experts expressed surprise. The earthquake hit 11 miles underground along the North Anatolian fault--where the Eurasian plate, moving east, grinds against the Anatolian plate, which is moving west. The North Anatolian fault roughly matches California’s San Andreas fault in both length and the rate at which its plates slip.

Since 1939, seven major earthquakes have hit along the North Anatolian fault, each one farther west. The rearrangement of the continental plates after each quake transferred the point of friction down the line. Some of those quakes were very bad--the one in 1939, for instance, killed more than 30,000 people.

“The earthquakes fell like dominoes, one after the other,” Dolan said.

Yet even with such a precise pattern, most scientists refrained from forecasting the most recent quake. When they did, they were tentative: Ross Stein of the U.S. Geological Survey wrote recently that the city of Izmit faced a 12% probability of being struck by a major quake in the next 30 years.

“That’s as close as you can get to forecasting an earthquake,” said Edward Field of USC, who inspected the disaster area.

Next Big One May Strike Istanbul

If the pattern in Turkey holds, some experts say, the next big quake could strike near Istanbul--a city of 6.6 million people.

Some experts also believe that two earthquakes actually struck south of Izmit on Aug. 17--a main event followed 30 seconds later by a weaker one nearby. The shaking was finished, and the damage done, in about 45 seconds. But the dual eruptions, if that is what they were, were indistinguishable to people in the area. Many reported hearing a rumbling that resembled a great thunderstorm.

With a magnitude of 7.4, the quake released an enormous amount of power. Turks as far away as the capital, Ankara--160 miles to the southeast--felt the tremors. In the quake zone, the ground slipped horizontally as much as 15 feet, curving roads and pushing forests apart.

When an earthquake struck Northridge in 1994, one of the things that limited the damage and loss of life was the direction of the energy released by the temblor. In Northridge, the shock waves shot mostly northward, into the mountains and away from Los Angeles. In Turkey, the shock waves rolled east and west--into the country’s industrial heartland.

Even the nature of the earth beneath the cities worked to make the effects of Turkey’s quake more destructive than the Northridge quake. The towns around Izmit are scattered across valley floors, old riverbeds and alluvial flats. The soil is typically soft and loose--helping to amplify the shock waves as they rumbled through. While the low-lying downtowns of Adapazari and Golcuk were obliterated, many houses higher in the hills--where the rocky surfaces helped muffle the shock waves--remained intact.

“The pattern of destruction is extremely clear,” said Holzer, of the U.S. Geological Survey.

One thing that caught the eye of Field, the USC seismologist, was that much more of the earthquake’s power seemed to enter the valley near Izmit than would in a typical tremor.

Valleys are known to capture shock waves, allowing them to resonate and cause greater damage. Faults often lie far away from valleys, and, as a result, much of the energy they generate never makes it inside the valleys. But in this case, Field believes, the fault cuts right through the valley and sent shock waves directly into it. As a result, the quake was far more destructive.

“The energy from the quake appears to have entered the valley and rolled across it like a wave,” Field said.

Field’s hypothesis could be relevant to the Los Angeles Basin, where faults often form the walls of valleys. If one of those faults ruptured, it could release relatively large amounts of energy into one of the valleys--and have a comparable effect.

“If we can document that this happened here, then we can speculate about what might happen in California,” Field said.

Quake Jumped Over 3-Mile-Wide Lake

Another intriguing thing about Turkey’s earthquake was the way it skipped from one fault to another. It began near Izmit, then jumped over 3-mile-wide Lake Sapanca and activated the Akyazi fault. The quake then set off a third fault called the Duzce.

Until now, scientists didn’t believe that a quake could traverse a body of water so wide and set off another fault.

“Another one of our precious ideas fell,” said Stein, of the U.S. Geological Survey. “We didn’t think a rupture could jump that far.”

Here’s why skipping is relevant: In the San Francisco Bay Area, the Hayward fault stops at San Pablo Bay, and the Rogers Creek fault begins on the other side, about three miles away. Scientists consider them separate faults. Now, the experts are not so sure.

“San Pablo Bay and Lake Sapanca are about the same size,” Holzer said. “If the quake can skip here, it’s possible that it could skip in California.”

The quake also skipped from the North Anatolian fault to the nearby Duzce fault. This has a rough parallel in California, where the San Andreas is shadowed by the San Gregorio fault off the shore of San Francisco Bay.

In 1906, a single rupture on a 300-mile length of the San Andreas caused the great San Francisco earthquake. But that magnitude 8.3 temblor did not set off the San Gregorio.

Still, other faults in California once thought distinct may now have to be considered as being together, scientists said. One of these instances occurs where the San Gabriel fault meets the San Andreas fault in Southern California.

“Every time we get a major earthquake, it humbles us,” Stein said.

Finally, Turkey’s quake was significant in where it stopped--in roughly the same area as two others, one in 1967, the other in 1943. That tends to affirm that faults are divided into segments, and that each one has distinct properties.

That, in turn, is important in the effort to predict the behavior of faults in California that lie beneath major population centers. The Elysian Park fault runs under Los Angeles, and scientists are unsure whether it is one long fault or several distinct segments. A long fault might mean a big earthquake, while several short segments might mean smaller quakes. The same goes for the Hayward fault that runs beneath Oakland.

The behavior of the North Anatolian fault thus presented scientists with a paradox: In some places the faults linked up, in other places they didn’t. How do the scientists reconcile that? The answer is that, in some cases, the scientists don’t know where faults end and where they begin. Finding their boundaries is one of the biggest challenges.

After all, Holzer says, it is the job of scientists to find order in otherwise unexplainable events.

* QUAKE VICTIM FLOWN TO L.A.

Boy, 4, who lost his mother and broke his legs is recovering at a Southland hospital. B1