|
|
|
AS IMPORTANT AS ALL THIS MAY BE, understanding the mechanics of an earthquake is still not enough.
“Yes, we need to be mapping faults,” says USC engineer Jean-Pierre Bardet, “but the key thing we want to do is to avoid loss of life.”
Bardet’s particular area of interest is the interaction between people and structures when faced with the violent forces unleashed by major earthquakes. “What are the effects of the faults on structures? How far should we set buildings back from faults? How much engineering do we have to put into structures? How much are we willing to pay?” These are the kinds of questions Bardet hoped to explore when he combed through the rubble of Turkey’s and Taiwan’s recent catastrophic events.
The closely-timed temblors provided a good opportunity to compare and contrast. At a magnitude of 7.6, the quake in Taiwan released twice as much energy as the first Turkish 7.4 earthquake, yet it killed fewer people and did less damage to the country’s infrastructure. The official death toll in Turkey stands at 17,000 (though outside agencies place the toll nearer 50,000). In Taiwan, about 2,300 people died. The difference, in Bardet’s view, is explained in part by the Taiwanese government’s superior preparation and quick response.
Physically, the two earthquakes had some striking similarities – and some key differences. In both, large areas were subject to “soil liquefaction,” a phenomenon in which loose and water-saturated soil softens and behaves like a viscous liquid. As the soil liquifies, the ground rolls and stretches like an ill-fitting carpet, leaving in its wake sinking structures, crooked bridges and bulging sidewalks and roads.
In Turkey, the liquefaction was so disastrous that it prompted tsunami researchers Synolakis and Borrero to pay a visit, following up on reports that waves had flooded some buildings in coastal areas. “This was very unusual,” Synolakis explains.
At Izmit Bay, the researchers found, the coastline had actually dropped by about 3 meters. “Buildings by the water settled into the ground during the shaking. The third floors were now at water level,” he says.
TUNGSHIH, TAIWAN This aerial view reveals the conspicuous vulnerabilty of 12-story buildings, which require fewer design reviews under Taiwan’s building codes.
|
It turned out that liquefaction, not a tsunami, was to blame. The sinking occurred in a matter of seconds, Synolakis believes, and people living on the first and second floors almost certainly drowned.
Upon further investigation, however, Synolakis and Borrero determined that at least one tsunami had indeed struck the Izmit Bay coast, killing several people. It occurred after the earthquake triggered an underwater landslide (not an uncommon reaction) somewhere in the Sea of Marmara.
In Taiwan, the September 21 earthquake triggered huge landslides in interior mountains. The slides carried away buildings and cars, smashed across mountain roads, plugged tunnels and barricaded many people. The mountainous interior of Taiwan – which bears a striking resemblance to parts of California – is fortunately not heavily populated.
Another key measure of earthquake activity – ground accelerations – proved twice as violent in Taiwan as in Turkey. Acceleration, Bardet explains, is the force that pushes a driver back into his seat as his car moves forward. The Turkish quake produced ground movement ranging from 35 to 40 percent of the acceleration of gravity; in Taiwan, motion was in the range of a full gravity of acceleration.
In both Turkey and Taiwan, the earthquakes produced long, dramatic “surface fault ruptures” – pulling apart buildings, bridges and roads, even creating new waterfalls on rivers. But the island nation’s building codes and earthquake preparedness helped check the extent of the catastrophe in urban areas.
Bardet saw some buildings located directly over the fault that had remained standing. Others hadn’t fared so well. In both countries, he observed structures whose “soft” first floors – usually containing shops or garages – had given way. This kind of structural weakness is typically the product of a builder’s decision to eliminate shear walls to maximize the number of store entrances or parking spaces.
Sloppy construction also was a culprit in most building collapses.
“One of the main explanations for collapsed buildings in Taiwan was that the re-inforced concrete columns simply did not have enough steel confinement and strength,” Bard
FENGYUAN, TAIWAN A “soft” first floor that failed. When this building’s ground level pancaked, the second story came down on a car’s front end like a sledgehammer.
|
et says.
In large construction projects, he explains, builders typically wrap a piece of steel around the reinforcing rods of a column. The ends of the steel wrap are then bent into the column. In the collapsed buildings in Taiwan, the bends were at 90 degrees instead of 135 degrees – the current practice in California.
“This is a structural detail, but it’s important,” Bardet says.
Height also matters, the USC researcher found. Taiwan’s building code requires fewer design reviews for 12-story structures than for taller buildings. The nation of 21 million inhabitants, consequently, has many 12-story edifices. During the September 21 quake, numerous smaller buildings collapsed, while hardly any larger ones did. Counterintuitive as it may seem, Bardet says, “it would have been safer to buy a condo in a 14-story building in Taiwan than in a 12-story one.”
Despite these flaws in its infrastructure, Taiwan acquitted itself commendably in the face of crisis. Good communications made a vast difference.
“There’s a network of 600 strong-motion sensors deployed around Taiwan so the government in Taipei knew right away what parts of the island had been hit,” Bardet says. “They knew how strong it was and how widespread.”
The Taiwanese government had in-stalled this network at the suggestion of USC earth scientist Ta-Liang Teng. It was money well-spent. Not only did the network help the Taiwanese government respond to a crisis; it also produced invaluable data for Teng and other earthquake researchers.
“The amount of recorded data from large earthquakes available to scientists has quadrupled as a result of this network,” Teng says.
Taiwan’s rescue efforts also proved effective. Public calm was promptly restored. Once the government had mobilized engineers to inspect buildings for safety, Bardet noted that refugee camps emptied out quickly. Detailed television reports explained exactly what had transpired, and what emergency measures were underway.
In the Turkish quake’s aftermath, by contrast, confusion reigned. People were left to sleep outside. “In one country there was panic; in the other, people received good information,” Bardet says.
Turkish authorities had a chance all too soon to demonstrate that they had learned from their mistakes. On November 12, Turkey was again wracked by a major earthquake – this time a magnitude 7.2 centered near the town of Duzce. Stung by earlier criticism, the government moved quickly to aid survivors. The death toll reached about 2,000.
Still, the prognosis is not good. Dolan’s work has taken him to many of the world’s seismically active regions, where he repeatedly encounters the deadly combination of growing populations, poor construction and an inability to prepare for inevitable earthquakes.
Place yourself in the position of leaders in these regions, he says: “Are you going to put your money into better sewage systems? Build a hospital, schools and roads? Or bet that the 1-in-500-year earthquake will occur during your administration? Developing countries frequently simply can’t afford to prepare for earthquakes.”
LIKE NUCLEAR FUSION AND A CURE FOR the common cold, earthquake prediction is one of those scientific silver bullets that the public eagerly awaits. Seismologists aren’t nearly as preoccupied with the notion.
It is unlikely that scientists will soon be able to predict the exact time, place and intensity of a future earthquake. However, they are beginning to “forecast” the probability of both future earthquakes and the kind of ground motions likely to result.
SCEC is currently preparing a study of earthquake probabilities in Southern California, says Henyey, that will be the basis for official public policy reports by the U.S. Geological Survey and the California State Division of Mines and Geology. Slated to be completed in two years, the study will predict the percentage chance that, say, during the next 50 years, an earthquake of a given
TEPETARLA, TURKEY A bus was going 70 mph on this overpass – part of the main motorway between Istanbul and Ankara – when it collapsed. Fifteen people died. |
magnitude will occur on a particular fault.
Scientists at USC and elsewhere have also begun exploring the relationships between faults. They are beginning to see how an earthquake on fault A may – 50 years later and 100 miles away – influence the seismic behavior of fault B.
But the main goal of earthquake research, Henyey insists, is not prognostication but preparation: the ability to tell people living in a seismically active area – such as Southern California – what kinds of ground motions and related hazards they can expect from likely future quakes.
“We think we can predict the bigger rolling motions quite accurately,” he says. “Our hope is to ultimately learn to predict motions that are more rapid, those with frequencies of 1 second or 0.5 second. These are more likely to impact the kinds of structures in which we live and work.”
Such information is critical to construction engineers. Say you’re designing a structure with a 50-year predicted lifespan: it would be extremely useful to know what kind of shaking the building is likely to endure during that period. Henyey hopes that in the near future a design engineer will be able to gather such information from a SCEC-sponsored Website. Armed with the knowledge that, hypothetically, there’s a 1 percent chance of an 8.0 earthquake producing strong ground motions, and a 40 percent chance of a 6.5 earthquake producing lesser ground motions on a given building site, the engineer would be in a good position to calculate the costs and benefits of preparing for the larger but less likely earthquake.
To generate this information, Henyey and other SCEC researchers must first identify where the faults lie, estimate what size earthquakes might occur, anticipate how the seismic waves will pass through the earth, and calculate the probability of those quakes occurring over a given span of years.
With these goals in mind, in October SCEC and the U.S. Geological Survey conducted a series of 93 small underground explosions in Southern California, hoping to gain a more accurate picture of active faults and other geological structures underlying the Los Angeles region. Researchers used 1,400 portable seismographs to read the sound waves from the explosions. From these results they plan to map the faults, rock formations, sedimentary basins and other geologic structures buried underground.
“We need to learn more about the configuration of buried faults and other subterranean structures to predict how the ground will shake in future earthquakes,” says Mark Benthien, SCEC’s assistant director for outreach.
STUDYING AN EARTHQUAKE IS NOT ALL science and engineering. There is a human side, too.
JEAN-PIERRE BARDET surveys a drain system trashed by fault ruptures in the small village of Kaynasli, near the epicenter of the Duzce earthquake in Turkey. |
In Taiwan, Bardet spoke to four people rescued from a reinforced concrete edifice buried in a mass of landslide debris. The survivors told how the building had rolled down the mountain for 2 minutes, yet somehow retained enough structural integrity to keep them alive. Thirty-four others trapped inside weren’t so lucky.
Dolan is haunted by Golyaka, a small town near the east end of the rupture where rescue workers were slow to arrive.
“It’s the place we first saw true despair among the Turks,” he says, recalling frenzied efforts by villagers using sledgehammers to tackle collapsed five-story apartments in the absence of heavy equipment. “From a personal standpoint, I hope never to see anything again like what I saw in Turkey. It was truly horrific.”
Realistically, though, there’s a strong probability that he will see it all again before the old memories even start to fade.
Based on his study of the North Anatolian fault, Dolan believes with a fair degree of certainty that a third major quake will soon thrash Turkey. Back in August, he correctly anticipated the second Turkish earthquake, which continued a rupture created by the earlier Izmit Bay quake. One of two remaining unbroken sections of the fault, he believes, will likely give way before long. That quake will probably hit in the vicinity of Bolu, a city of about 150,000 people.
Dolan is bracing himself for the inevitable.
|
