Pattern of Destruction : New Lessons From Quake in Mexico

One year after the Mexico City earthquake struck the nation’s capital and left its core in ruins, seismic experts are beginning to unravel several troubling mysteries surrounding the event. Ultimately, the scientists say, their studies will reshape their understanding of the way earthquakes inflict their damage on cities.

Erupting from the floor of the Pacific Ocean, the Mexican earthquake surged into the capital with crippling force after traveling 250 miles through several mountain ranges. Nearly a thousand buildings collapsed in the city, and more than 10,000 persons died. Since modern records have been kept, seismic waves had never wreaked so much damage from so great a distance.

In contrast, coastal cities much nearer the earthquake suffered only moderate damage. Lazaro Cardenas, a small city closest to the rupture, lost only 10 buildings.

Pattern of Destruction

Adding to the mystery was the pattern of destruction within the capital. Older brick or stone buildings--widely regarded as the most dangerous structures in an earthquake--withstood the rigors of the shaking while more modern buildings failed. The great majority of buildings that collapsed were constructed of reinforced concrete.

Now, after some of the most extensive study in seismic history, scientists believe that they have found the answers to the puzzles. Those answers, they add, may contain important lessons for cities in the United States and elsewhere that are in seismic zones.

“This was an historic earthquake,” said Vitelmo V. Bertero, a UC Berkeley engineering professor who specializes in seismic analysis. “No one expected the intensities of motion that were recorded in Mexico City. No one had designed for it, and that is why so many buildings failed.”

Instruments showed that the acceleration of seismic waves hitting buildings in parts of the city were up to five times greater than waves outside the city. The intensity was so strong that one of the instruments designed to measure the force broke under the strain.

Effect of Soft Soil

Mete A. Sozen, an engineering professor at the University of Illinois who visited the disaster area as a member of the National Academy of Sciences assessment team, recalled the collapse of a major hospital where hundreds died. “It was designed to withstand a lateral (sideways) force equal to 6% of its weight, and it probably met those standards,” Sozen said. “The forces exerted on it were much greater than that, and it came down. Only some accidental feature that added strength would have saved it.”

Instrument recordings have shown that the extreme vibrations were produced by soils beneath the city. For many years scientists have known that Mexico City is on top of soft, elastic sediments that tend to amplify some frequencies of shock waves from earthquakes. What they did not know was just how extreme the magnification would be.

“The statistical data on earthquakes is still very small,” Bertero said. “Since the beginning of scientific record-keeping, we have had just a few great earthquakes striking large cities. This one taught us that soft soils can magnify motion to a degree never thought possible.”

Seismologists agree that Mexico City’s location is more dangerous for earthquakes than any found in the United States. The soft soils under the city are part of an ancient lake bed that was filled thousands of years ago with wet clay deposits. The lake bed is surrounded by mountains of bedrock so that the whole thing, in the words of one engineer, reacts to shaking “like a bowl of Jello.”

Scientists have reconstructed this scenario for the disaster:

On Sept. 19, 1985, shock waves from the magnitude 8.1 earthquake began shooting across western Mexico toward the capital. The waves were grouped into two pulses produced by twin ruptures in the Orozco fracture zone off the Mexican coast.

Initially the waves were complex, containing many different frequencies. But over long distances the shorter frequencies were filtered out, leaving only long, smooth waves to strike the lake bed at Mexico City. Waves are measured by their “period,” or the length of time that it takes to complete a cycle. These waves had cycles of 1.5 to 2 seconds, very long by earthquake standards.

Within the lake bed, a transformation began. The elastic soil, saturated with water, amplified the motion four to five times its previous level. The waves struck the surface with undulating regularity that continued for the unusually long time of a minute.

Like Battering Rams

In the city, some buildings began to sway in time with the waves. A few pounded against each other, acting like battering rams. Each sway became worse until the buildings tore themselves apart, the floors falling onto one another in a phenomenon known as “pancaking.” At last count, 954 buildings collapsed inside the city, 1,100 others eventually will be demolished, and thousands of others suffered damage.

Scientists have also come to the unlikely conclusion that the destructive force--in this particular earthquake--was increased rather than diminished by the long distances that the shock waves traveled. In a more local event, they say, the waves would not have been filtered into the regular pattern of the motion that proved so deadly.

“The Mexican earthquake points out the very disastrous effects of a large earthquake at large distances,” Sozen said in a report to Congress. “The damaging effect of low-level ground motion of long duration has not really been appreciated before.”

Seismic experts emphasize that a long-distance event is not more dangerous than a nearby earthquake, just different. A long-distance rupture will tend to destroy different types and sizes of buildings and, contrarily, leave others relatively untouched.

That result was illustrated dramatically in Mexico City where many 19th-Century, unreinforced buildings were left standing while more modern buildings of reinforced concrete collapsed. Engineers soon discovered a similarity in the buildings that failed: Almost all were mid-rise constructions ranging from six to 16 stories.

These buildings had a natural tendency to vibrate at the same rate as the shock waves: one cycle every 1.5 to 2 seconds. Like a child on a swing, the buildings swayed more and more each time a new shock wave pushed them until they collapsed.

The older buildings, though inherently weaker, were also stiffer because of the masonry construction, and their natural vibration patterns did not match the shock waves. They survived.

Debate in U.S.

Studies of the Mexican experience has produced a debate in the United States about parallel dangers in this country. While there are some disagreements about the degree of geological similarity between Mexico City and urban areas here, most seismic experts believe that disaster agencies should heed the warnings of Mexico.

“In California we do not have cities sitting on old lake beds with clay sediments,” said Richard Eisner, director of the Bay Area Regional Earthquake Preparedness Project. “We do have sections of cities sitting on soft soils that will amplify shock waves, and we do have the potential for long-distance earthquakes.”

Only the San Andreas Fault, which curves through the southern and central portions of the state, is believed capable of producing earthquakes of the magnitude that struck Mexico City. Various predictions have placed the chance of that fault rupturing over the next 30 years at greater than 50%.

If and when the rupture occurs, it could send long-distance shock waves headed toward Los Angeles and San Diego from distances of 75 to 150 miles. In such an event, the most vulnerable areas would be neighborhoods built on artificial fill, such as Marina del Rey in Los Angeles and Mission Bay in San Diego.

Alluvial Soils

Other susceptible areas would be regions with alluvial soils that have high water tables, such as some sections of San Bernardino County near the Santa Ana River and Mission Valley in San Diego.

In San Francisco, the San Andreas Fault runs very close to populated areas and hence would not likely produce conditions similar to Mexico City. But scientists now realize that the same earthquake could wreak damage on Sacramento, 80 miles away, where there is substantial development along the alluvial plains of rivers.

Some seismic experts say equal dangers lie outside California. One of the largest earthquakes on record occurred in New Madrid, Mo., in 1811. Sozen at the University of Illinois said a repeat of that event could produce catastrophe in Memphis, some 120 miles away.

“A city like Memphis has not prepared itself for an earthquake the way California cities have,” Sozen said. “A lot of buildings would collapse; there would be some carnage.”

Even California cities cannot take consolation from the notion that their building codes are substantially more restrictive than Mexico City’s, the engineers say. Although the Mexican capital is in the process of upgrading its codes, the old standards were described by one expert as “roughly parallel” to this state’s.

“I am sure that someone could find a few buildings that collapsed (in Mexico City) because of poor construction,” Bertero said. But “it would be a mistake to generalize from that. Mostly, those buildings were good buildings.”

Disaster officials in California say this state contains thousands, and perhaps tens of thousands, of reinforced concrete buildings similar to those that failed in Mexico. While building codes here have been tightened in recent years and retrofit programs have been targeted for masonry buildings, little or nothing has been done to improve the seismic safety of reinforced concrete structures.

Construction Prohibition

Bertero believes that one of the most important steps that governments could take here would involve prohibitions on the construction of buildings that vibrate at the same frequency as the underlying soils. It was that deadly union, he said, that caused most of the destruction in Mexico City.

“You don’t put buildings with two-second periods on loose soils that amplify two-second waves; that is the lesson,” he said. “You put those buildings on bedrock where the waves will be much shorter. And on the loose soils you put very short buildings or much taller ones. You make sure that the soil and the buildings will not fall into sync.”

The process of mapping soils to determine their vibration frequency is called “micro-zonation” and has been begun by Mexico City under a grant from Japan. After the mapping has been completed, the government will face the formidable task of converting the data into its building codes.

Some disaster agency officials say other lessons from Mexico City may be equally crucial. Analysis of the rescue efforts has shown that the government first failed to comprehend the extent of damage and because of that failure it delayed in requesting aid. Consequently, disaster officials say, some of those trapped in the rubble died before they could be rescued. One unofficial report has estimated that 80% of the deaths in Mexico came not from trauma but from exposure.

Richard Andrews, assistant director for the state Office of Emergency Services, said the Mexican government’s problems grew out of its reliance on aerial reconnaissance for damage assessments. “We learned that you could not tell much from the air. You’ve got to have people on the ground, and you have to have them there fast,” Andrews said.

To speed up damage assessments, Andrews has shrunk the time required for establishing a state emergency operations center from 36 hours to four hours following a major earthquake.

Disaster officials also concluded that many died in Mexico City because of the lack of heavy equipment to lift concrete slabs in rescue efforts. An attempt is being made to inventory available equipment in Southern California so it can be obtained quickly.

Neighborhood Volunteers

Finally, rescue agencies say they would like to develop a plan to make full use of neighborhood volunteers who will almost certainly appear at disaster sites. “For trapped persons, there appears to be a ‘golden 24 hours’ in which they can be rescued,” Andrews said. “After that, many will die. In Mexico, most of the successful rescues were made by neighborhood people because they were there in time. The international rescue teams saved very few in comparison.”

In spite of all the analysis of the Mexico event, Andrews said the next large earthquake in California will contain many surprises. “Earthquakes are capricious; each one has its own signature,” he said. “We will learn our lessons from Mexico, but when the next big one hits here, we will rewrite the book again.”