Dr. Glenn Rix (right) and civil engineering graduate student Wesley Spang use a series of receivers to measure the amplitude of seismic waves generated in the soil by a mechanical device.
Because of Emir Macari's birth, his family lives. Dr. Macari, a professor in Georgia Tech's School of Civil and Environmental Engineering, was born in Mexico City on July 22, 1957, six days before a major earthquake rocked the city. His family had planned on moving into a larger apartment complex, but delayed their move because the baby was late in arriving.
"It was a good thing, too," says Macari. "The building we were to move into collapsed during the earthquake. We all could have been killed."
Now, Macari wants to help protect the lives of others living in earthquake-prone areas.
He is not alone in his quest. Across the Tech campus, civil engineers, structural engineers, seismologists and city planners are pursuing a number of programs ranging from earthquake prediction and earthquake engineering to risk assessment and damage mitigation. The programs are diverse, but the goals remain the same: to lessen the risk posed by major earthquakes.
Major earthquakes seem to be occurring with more frightening regularity, says Macari ,and they are causing much more damage and loss of life than in the past. A recent quake in Russia, for example, caused billions of dollars in property damage and claimed 2,000 lives. The earthquake that struck Kobe, Japan, early this year, killed more than 5,000, injured more than 26,000, and left some 300,000 homeless. Damage estimates topped $200 billion.
"We can't stop earthquakes from occurring," says Macari, "but we can come up with effective ways of reducing the risk and mitigating the damage.
"Common geotechnical hazards that occur during earthquakes include soil liquefaction, ground motion amplification and landslides," says Macari. "Soil liquefaction occurs when water-saturated, sandy soil is shaken during an earthquake. The intense vibrations disrupt the normal character of the soil, transforming it from a solid to a fluid state. In essence, liquefaction turns the ground into quicksand, causing building foundations lose their footing and sink."
Ground amplification occurs when the surrounding soil goes into resonance, thereby magnifying the intensity of earthquake-induced vibrations. Mexico City offers a classic example.
"Mexico City was buiit on top of an old lake bed," says Macari. "The underlying sediments can be compared to a huge bowl of gelatin. When you shake this 'bowl' at the bottom, the lake bed can resonate, substantially increasing the vibrational amplitude at the surface. So even a small earthquake can have devastating consequences to the buildings located above."
With funding from the National Science Foundation and the U.S. Geological Survey, Dr. J. David Frost, associate professor in the School of Civil and Environmental Engineering, is developing a framework for identifying and mapping geotechnical hazards through the use of Geographic Information System (GIS) technology. Researchers and practitioners using the advanced computer code will be able more accurately to predict where major damage will occur in a given area and make better plans for mitigating it.
Frost's research focuses on the development of a methodology that evaluates geotechnical earthquake hazards spatially using GIS. The impact of these hazards on other geo-referenced information (such as demographics, infrastructures and life lines) is assessed in a spatial environment, in order to mitigate possible consequences. Currently, this methodology is being used to assess the earthquake hazards in various parts of the United States, including Treasure Island, Calif., Evansville, Ind. and Western Puerto Rico.
Frost and colleagues Mike Rowan, a senior research scientist in the Georgia Tech Research Institute, post-doctoral fellow Ronaldo Luna and graduate assistant Tom Rockaway have spent four years linking GIS, geostatistics and a data visualization software with custom algorithms they wrote to analyze various earthquake hazards.
"We can overlay maps of buildings and roads and see which ones are in areas where we may see liquefaction or amplification during an earthquake," he says. "We also digitized photos of the buildings at several locations and linked them to the map, so users can see which types of buildings are located on potential problem areas."
On top of the maps this data produces, one may superimpose other information such as population centers, transportation networks or even building types and densities. Ultimately, this information will help planning agencies in earthquake-prone areas better assess what types of structures should be built in various locations, and determine which areas should be avoided altogether. The information would also assist emergency response teams in dealing with the aftermath of a disastrous quake.
"This software helps identify where the problems are and where they aren't," Frost continues. "I don't think you can design a completely quake-proof building, but you can certainly design to minimize the impact of an earthquake on structures during and after the event. You can understand the consequences, and then design to minimize the impact."
To better forecast the potential effects of ground amplification at a specific site, Dr. Glenn Rix in the School of Civil and Environmental Engineering is developing a non-invasive field method for measuring an important dynamic soil property, damping. The research is being funded by grants from the National Science Foundation and the U.S. Geological Survey.
"For many years, the standard approach to measuring damping involved going out in the field and bringing back a soil sample to test in the lab," says Rix, an associate professor. "In the lab you have tremendous control over the testing, but the act of removing the soil sample from the ground may fundamentally change the very properties you want to measure. Also, you are left with the nagging question of whether this small sample is really representative of a huge site."
Rix uses a small mechanical vibrator to generate miniature seismic waves in the field. A series of receivers -- called geophones -- measure the amplitude of the seismic waves at scattered locations, revealing how quickly the energy is being attenuated.
Structural damage from an earthquake is very dependent upon the nature of the ground below, says Rix. Every soil profile has a particular natural frequency, and will tend to amplify motion in certain frequency ranges. If the natural frequency of the soil column coincides with the natural frequency of the buildings above, the structures may resonate and substantial damage could result.
"When a building resonates, it vibrates with an increasing amplitude until the materials cannot withstand the force," says Rix. "Then the building likely suffers substantial damage. No structure can withstand resonance for a long period of time."
Dr. Glenn Rix (right) and civil engineering graduate student Wesley Spang use a series of receivers to measure the amplitude of seismic waves generated in the soil by a mechanical device.
The natural frequency of a structure is highly dependent upon its height, says Rix. "Taller buildings will resonate at lower frequencies. By understanding the damping characteristics of the soil profile at a particular site, we can better determine which buildings are at greater risk."
Architects, city planners and structural engineers would know in advance not to build a certain type of building (a 20-story skyscraper, for example) on a given site. If such a building already existed, the information could be used to design some type of retrofit to either strengthen the structure or in some way counteract the effects of an earthquake.
"Given the fact that earthquakes will continue to occur with little or no warning," says Rix, "the question becomes: 'What can we do to mitigate the damage?' In the field of geotechnical earthquake engineering, we are making some progress in terms of being able to identify in advance those sites which will likely experience damage during an earthquake."
(Sidebar on testing Georgia's 14,000 bridges: "A Bridge Over Troubled Waters?")
On Jan. 17, 1994, the most costly earthquake in U.S. history struck the community of Northridge, Calif. Damage estimates exceeded $20 billion. Fifty-seven people lost their lives, including l6 residents of the three-story Northridge Meadows apartment complex who died when the lower floor collapsed.
"Northridge once again brought home the painful message that we need to improve our building codes," says Dr. Larry Kahn, associate professor of civil engineering.
Kahn is a structural engineer whose major interest is in earthquake engineering: more specifically, in strengthening existing buildings for improved earthquake resistance. In the late 1970s, Kahn explored the rehabilitation of brick-masonry construction. His work, sponsored by the National Science Foundation, helped form the basis of design recommendations for the seismic rehabilitation of unreinforced masonry buildings in the Los Angeles area.
"These recommendations were used in over 8,000 masonry structures within the city of Los Angeles itself," says Kahn, "and most of those buildings survived the Northridge earthquake.... This really showed that even fairly simple techniques can work remarkably well."
To prevent masonry walls from collapsing, Kahn suggests applying a fiber-reinforced coating to the walls. The coating provides a certain amount of ductility, he says. The walls might crack in an earthquake, but they won't fall apart. Another highly useful technique involves securely fastening the walls to the rest of the structure.
"Many masonry walls collapse because they are inadequately connected to the adjoining floors or roofs," says Kahn. "Oftentimes, joists and rafters are simply laid into slots in the masonry and not connected at all. Joist anchors should be used to tie the entire structure together, whether the building is a residence or a warehouse. Such modifications are easy to make and cost very little, but provide a great deal of seismic safety."
To protect a modern office building of steel and concrete, more elaborate and expensive techniques must be employed. Proposed methods range from mounting the building on huge rollers or gigantic shock-absorbing pads, to placing active controllers -- which propel immense weights -- in the upper floors.
Aerospace engineering professor Dr. Jim Craig and civil engineering professor Dr. Barry Goodno have a radically different concept. They want to use the building's external facade, or cladding, to dissipate energy in an earthquake and thereby reduce the building motion and resulting damage.
Cladding, which generally consists of heavy panels of precast concrete or stone veneers, typically offers no structural support to a building.
"At the present time, cladding is a purely decorative building enclosure," says Craig. "It just hangs on a building's superstructure like scales on a fish, offering no structural support whatsoever. We want to take this nice-looking architectural treatment and make it do its part in resisting lateral forces."
Craig and Goodno are developing advanced cladding attachments that could be "sacrificed" in a strong earthquake. The novel energy-absorbing cladding connections could provide an effective means of protecting a building's structural integrity and its economic viability.
"The pattern of earthquake-inflicted damage to buildings is strikingly similar," says Goodno. "There's a lateral motion that's almost always accompanied by a twisting component."
This combined motion can crack walls, cause ceilings to fall, disrupt electrical, plumbing, and ventilation systems, or even weaken the structure, leading to collapse. Even if the building remains structurally sound, it may be a total financial loss due to the high costs of repairing all the nonstructural systems.
"Located on the perimeter of the huilding, the cladding is ideally positioned to help control lateral and twisting motions and absorb energy during an earthquake," says Goodno. "This would reduce damage to both the building structure itself and the non-structural systems. And, since the cladding is on the outside of the building, it is much easier and less expensive to repair than structural members buried deep within the building."
Craig and Goodno have designed, built and tested various types of cladding connectors in the laboratory. Using detailed computer models, they have used their laboratory data to simulate the behavior of advanced cladding systems on typical buildings when subjected to strong earthquakes. They have also measured the actual dynamics of typical real buildings to accurately calibrate their computer models.
"Advanced cladding connections take advantage of the relative movement between the architectural cladding panels and the building structure to dissipate energy and provide additional lateral stiffness," says Craig. "Due to this increased damping, the overall building response can be reduced by 25 percent, and displacements and interstory drifts can be maintained within acceptable limits. And, the advanced cladding connections can be applied to both new and retrofitted buildings."
In related work, Craig and Goodno are working with Dr. Tony Calise in aerospace engineering to develop new "hybrid" methods for controlling a building's earthquake response. The new methods involve the simultaneous use of advanced cladding systems along with the added benefits of robust active control.
"In this case, special devices called actuators would be used to introduce controlled forces into the structure during the earthquake that would counter the seismic-induced forces and thereby reduce the dynamic response," explains Craig. "This would combine the passive nature of the cladding system -- which would always be there, no matter how small or large the earthquake -- with a more robust active control system that would respond only to more severe earthquakes."
Like other major earthquakes, the temblor which struck Loma Prieta, Calif., on Oct. 17, 1989, came without warning. Centered in the Santa Cruz mountains near San Francisco Bay, the earthquake rattled 60,000 fans in Candlestick Park waiting for Game 3 of the World Series and wreaked havoc in the Marina District. Across the bay, huge portions of the upper deck of Interstate 880 were shaken loose and collapsed onto the lower roadway. Forty-three motorists were crushed to death.
In an effort to make bridges and elevated highways more earthquake-resistant, Dr. C.-H. Chuang is applying control theory to the active control of bridge supports. Chuang, an assistant professor in the School of Aerospace Engineering, has investigated control applications in aircraft, spacecraft and flexible space structures. He says modern control theory can be applied to reducing the potentially dangerous vibrations of a 200-ton bridge span.
"Bridge spans are not rigidly fastened to their piers," explains Chuang. "Some freedom of movement is built in to allow for thermal expansion, traffic-induced vibrations and seismic disturbances. But if the motion becomes too large -- like during a major earthquake -- the span can slip off its bearings, endangering motorists and necessitating costly repairs."
To prevent the span from moving too far, actively controlled hydraulic actuators could be installed at the support bearings, says Chuang. "During minor disruptions, the oscillations would passively dampen out by themselves. But during a major earthquake, the actuators would apply appropriate forces to counteract and limit the oscillations, and keep the span from slipping off the bearings."
Chuang has recently completed the computer modeling and simulation for his active bridge controllers. His next step will involve constructing physical models and evaluating their performance on a shaker table.
During the winter of 1811-1812, three of the largest earthquakes the United States has ever witnessed struck near present-day New Madrid, Mo. The shaking was so severe, massive tree trunks were snapped in two. The quakes toppled chimneys as far away as Savannah and rang church bells in Boston.
"The only reason why nothing is recorded as to what happened in Atlanta," says Larry Kahn, "is that there wasn't any Atlanta at that time. Another New Madrid-type earthquake could occur in the East at any time, and the damage would be devastating. The danger is very real." (See "The Faults in Earthquake Prediction ")
Kahn is a firm advocate for incorporating earthquake-resistant standards into building codes throughout Georgia and the eastern United States -- if not for every structure, then at least for critical structures like schools, hospitals, high-rise buildings and fire and police stations.
Normally, if a building is designed with some thought given to earthquake resistance, it will survive and a lot of life is protected," says Kahn. "On the other hand, if a structure is designed without regard to earthquakes, there is significant danger of collapse, with huge damage and potential loss of life."
From a civil engineering standpoint, structural engineers are the ones ultimately responsible for public safety, says Kahn. "We are the ones who stamp the drawings and say the buildings are safe. We really need to incorporate more stringent earthquake-resistant standards into our building codes, in order to better protect ourselves from ourselves.
"Perhaps the best statement was made by Professor Vitelmo Bertero, who recently retired from the University of California-Berkeley. Bertero, one of the nation's greatest earthquake engineers, said, 'If we just built buildings the way we know we should, we would have seismic safety in North America.' The problem is, too few building owners are willing to pay the additional cost."
(Sidebar on a new view of quake causation: "The Fault in Earthquake Prediction")
Fortunately, risk assessment has improved dramatically over the past 20 years. Elaborate vulnerability models have been developed that can estimate the amount of damage likely to occur in a given area. By knowing in advance the nature and extent of the damage likely to be caused by an earthquake, local planning agencies can develop mitigation policies and plan how to respond to emergencies.
"Since the 1971 San Fernando earthquake, the state of California has required all cities and counties to include seismic safety as a part of their comprehensive planning process," says Dr. Steve French, director of Tech's graduate program in city planning. "The plans should seek to decrease both the amount and the vulnerability of development in the most hazardous areas."
Because the 1994 Northridge earthquake affected many of the same areas that were heavily damaged in the 1971 quake, researchers can now accurately evaluate how effective land use planning has been in mitigating damage. With sponsorship provided by the National Science Foundation, French and his students are working on one such project.
"We have collected the comprehensive plans for 22 jurisdictions in the San Fernando Valley," says French. "We have evaluated those plans on how well they addressed seismic safety. We have also digitized the seismic hazard information out of those plans onto a Geographic Information System, so we have a graphic representation of where the different hazards are located. And, we have plotted the locations of over 100,000 buildings damaged during the Northridge earthquake, along with information concerning when each building was built and the nature and extent of the damage."
The tedious and time-consuming task of data entry is nearly completed, says French, and soon he and his students will concentrate upon analyzing the data. They anticipate preliminary results in a few months.
"We will be able to look at a community and see how well its plan dealt with seismic hazards," explains French. "Then we will examine how well the plan was actually implemented, in terms of how much development went into hazardous areas. Lastly, we'll take a close look at the resulting damage. This information will help us understand and evaluate how well land use planning in California has worked in terms of a mitigation strategy, and how it might be improved."
In another project, French and his students are developing a GIS model that estimates the societal impacts of earthquake-related damage to urban infrastructure. The model links physical components of the water supply system in Memphis, Tenn., with population and economic data from the U.S. Census.
The model, which is being developed through a grant provided by the National Center for Earthquake Engineering, operates at three distinct levels. First, a simulation module allows specific damage to the system to be input. Then, an assessment module presents the impacts of the damage in terms of selected demographic data. And last, a repair module suggests optimal repair strategies.
"All pipes are not created equal," says French. "Some pipes are more important in the configuration of the network and have larger service populations. To decide which pipe should be repaired first, however, the model allows the user to select which particular characteristic of the population should be optimized."
Although the current model deals only with issues related to water supply, a similar approach could be taken with gas lines, electric power distribution, transportation and other elements of a city's infrastructure. Such a method could be used to allocate emergency response resources in the most effective manner and to set priorities for hazard mitigation efforts.
"Infrastructure damage modeling and other types of risk analysis procedures can be applied not only to existing land use patterns, but also to future land use scenarios" says French. "In this way, we can estimate the type and amount of damage that would be associated with alternative land use patterns. This information is extremely valuable in making decisions about the location of new development."
In today's highly technological society, people and businesses have grown highly dependent upon life lines -- those critical elements of a city's infrastructure which sustain life, society and the local economy. As a result, we have become increasingly susceptible to greater economic damages from natural disasters, says Georgia Tech President Wayne Clough. With a background in Civil Engineering, Clough has pursued research interests in geotechnical engineering, including measuring the liquefaction response of coastal soils. (See "Engineering Georgia Tech's Future" in the Winter 1995 RESEARCH HORIZONS.)
"It is vitally important to keep our life lines open and operating, or we risk financial ruin," says Clough. "For example, the city of San Francisco was saved from real financial disaster in the days following the Loma Prieta earthquake largely because the Bay Area Rapid Transit (BART) system had survived intact."
When a span on the Oakland Bay Bridge collapsed during the quake, commuter traffic was disrupted for well over a month, he explains. Although many people could not drive to work, they could ride BART.
"The tunnel across the bay had been specifically designed to withstand an earthquake, and escaped unscathed," says Clough. "BART kept running, so the city's commerce kept running."
Another prime example of a critical life line is communications, says Clough. "In the past, news of a major earthquake might have taken weeks to traverse the globe. Today, a business might not survive in a globally competitive market if it loses communications for even a day.
"We are a society now that is much more tightly tied together, and when those knots are loosened -- during a major earthquake, for example -- we are much more susceptible to severe economic loss. That's a natural consequence of this incredible, incredibly connected global society in which we live."
As the recent earthquake in Russia demonstrates, these natural disasters will continue to disrupt both our lives and our livelihoods. Ultimately, the best earthquake hazard mitigation program will consist of a blend of strategies which include accurate hazards mapping, better structural design, tougher building codes, sensible land use planning and appropriate emergency response preparation. Only in these ways will we truly be able to lessen the immense risk posed by major earthquakes.
Table of
Contents
Send all questions and comments to Webmaster@gtri.gatech.edu