Quakes and Bridges: Sensors Help Scientists Span Information Gap

When the Whittier quake struck, the Vincent Thomas Bridge in San Pedro vibrated like a plucked harp for a full minute. And like the different strings of a harp, each of the parts of the gracefully drooping suspension bridge had its own motion.

The two 506-foot end spans had the largest movements: The roadway segments twisted evenly, their edges rising and falling 10 centimeters at just under one cycle per second. The massive, 1,500-foot main span did a smaller twist to a slightly faster beat. And the 336-foot towers swayed back and forth in a jerky, even faster fashion.

An inspection after the quake revealed no apparent damage, but the shaking has given researchers their first opportunity to study the effect of a fair-sized temblor on a long suspension bridge wired with sensitive accelerometers. Accelerometers measure the force exerted by the shaking.

‘Unique Opportunity’

“This is the first real data ever obtained from a modern suspension bridge in the whole world,” said Prof. Ahmed M. Abdel-Ghaffar, who did his Ph.D dissertation at Caltech on the vibrations of the Vincent Thomas Bridge. “The data represents a unique opportunity to learn about the behavior of these flexible structures.”

It may take hundreds of thousands of dollars and months of sophisticated computer computations, but scientists hope to learn how to design bridges more safely, as well as how to make existing structures better withstand the terrible shaking of a major temblor.

Suspension bridges several hundred feet long have been built for more than 150 years and some long-standing spans, including San Francisco’s famed Golden Gate Bridge, are in areas prone to major earthquakes. But until recent years, none has been rigged with instruments to measure the force of an earthquake’s shaking.

In addition to the bridge, about 400 sites throughout California are wired with instruments in the state’s Strong Motion Instrumentation Program, and 100 were activated during the Whittier quake, producing what state officials say is the largest set of data ever recorded from an earthquake.

Besides the bridge, the most important sites include eight dams, the Fox Airfield control tower in Lancaster, the Southern California Edison Co. power plant in Etiwanda and 27 other buildings, where multiple instruments recorded motion in various directions at different structural parts.

The San Pedro bridge was wired in 1981 with 26 sensors, at a cost of $500,000.

“We had to wait six years for this one,” said Marvin Huston, a state seismic technician who clambered all over the bridge setting up the precisely placed sensor system. “An awful lot of gray matter goes into this thing,” he said as he inspected some of the instruments recently.

The accelerometers lie dormant until a large quake starts them measuring the force of the shaking and turns on a special recorder. Then their output, flashed through wires more than 2,000 feet long in some cases, is recorded on a slowly winding five-inch-wide strip of film. A bank of five 12-volt gel-cell batteries is ready to power the equipment for up to a week in case a major quake cuts off electricity.

During the Whittier quake, the maximum acceleration was recorded on the north edge of the end spans. Expressed as a fraction of the force of gravity, or g, the force was .28g, which is enough, were it maintained steadily in the same direction, to accelerate an object from zero to 60 m.p.h. in just under 10 seconds.

In earlier work, Abdel-Ghaffar, who teaches civil engineering at USC, correlated the vibrations predicted from a theoretical model of the bridge with the actual motion of the bridge caused by “micro” earthquakes too small to feel and other sources of vibration, including wind, noise and traffic.

“It was fascinating. This structure was easy to model and we found very good agreement between the theory and the measurements,” he said.

Seeks Further Study

Now he has applied for a $150,000 National Science Foundation grant to compare the larger motions of the bridge during the Whittier quake with theoretical movements.

“Earthquakes pose a particular hazard to long-span, cable-supported bridges, as opposed to most other civil engineering structures, due to their . . . configuration and their relative flexibility,” he wrote in his application. In addition, he said, “Although earthquake-resistant requirements for most bridge types are adequately covered by the design codes, there are no codes” for suspension bridges.

The basic problem is to determine how the vibrational energy imparted to the bridge at the cable anchors and the towers is transmitted to the other parts of the bridge.

A particular area of interest that Abdel-Ghaffar said he wants to study is why the roadway of the end spans settled into a back-and-forth twisting motion instead of “flapping,” the way a blanket moves when being spread on a bed. Another question is why the end spans vibrated more strongly than the main span.

Considerable Interest

There is already widespread interest in analyzing what happened to the Vincent Thomas Bridge.

James Gates, structural mechanics engineer with Caltrans’ division of structures, said he was curious about what safety conclusions can be drawn.

“My job is to think earthquakes 24 hours a day, seven days a week,” he said. “Suspension bridges are rare birds.” He estimated, however, that it would take an earthquake generating vibrations perhaps 10 times stronger than the Whittier Quake to damage the bridge.

In San Francisco, officials who plan to put a sensor system similar to the one on the Vincent Thomas Bridge on the Golden Gate Bridge in the next two years are also watching closely.

Mervin Giacomini, deputy district engineer for the Golden Gate, whose main span is about three times the length of the Vincent Thomas’, said, “Any time you get nice lab testing under actual conditions, it is certainly information we are going to be interested in.”

WHEN THE BRIDGE SHOOK

Vincent Thomas Bridge’s sensors provided data on a quake’s effect on suspension bridges.

Whittier quake struck at 7:42 a.m. on Oct.1 and measured 5.9 on the Richter scale.