By Kevin Wilcox
Researchers quantify the collapse threat that aftershocks can pose to damaged structures following large earthquakes.
Buildings hit hard by the 2011 Christchurch earthquake in Australia suffered additional damage from strong aftershocks. New research proves that the closer and stronger the aftershock, the more damage it can do to buildings already compromised by a temblor. Wikimedia Commons/Martin Luff
June 16, 2015—Aftershocks present a significant collapse threat to structures already damaged in a preceding powerful earthquake, and this threat is often overlooked in design codes, which means that some buildings in earthquake zones could have a lower factor of safety than intended by the codes. Those are some of the key conclusions of recently published research by a team from Colorado State University (CSU) and Michigan Technological University (MTU).
The paper, "Effect of aftershock intensity on seismic collapse fragilities," was published in the
International Journal of Reliability and Safety
and written by John van de Lindt, Ph.D., F.ASCE, the George T. Abell Professor in Infrastructure at CSU and the corresponding author; Negar Nazari, Ph.D., who was a graduate student at CSU at the time; and Yue Li, Ph.D., the principal investigator and an associate professor of civil and environmental engineering at MTU. "The thing about aftershocks that a lot of people don't realize is they don't occur in the same place," van de Lindt says. "They typically occur on the same fault, but they can be much shallower, they can be closer to a population center, and even though the Richter [scale] magnitude might be smaller, they can actually have a more devastating effect."
To examine the effect aftershocks have on a structure, the research team began with data from the NEESWood Project, a series of tests conducted through the National Science Foundation's George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES). For the project, a full-scale, two-story townhouse was subjected to a series of simulated earthquakes on shake tables at the University of Buffalo.
The team chose this project because the structure used in the test was more fully complete than most structures set on shake tables, outfitted with interior and exterior finishes and furniture. The researchers focused on data from a test in which the structure was subjected to a simulation of the 1994 Northridge earthquake as recorded at the Rinaldi station, a near-fault recorder.
"We took what was effectively the largest shake, because that produced the biggest response in the structure. It didn't collapse it, but that takes us closest to collapse," van de Lindt says. "What we were able to get from that was a system-level model. So we could calibrate the behavior of our numerical model in a simplified fashion." The team then used a computer model to subject the structure to aftershocks.
The research indicates that larger aftershocks closer to damaged structures present the greatest danger for severe damage or collapse. Although that may seem intuitive, van de Lindt notes that the relationship had never been quantified. Less intuitive is the nonlinear relationship of the initial earthquake intensity to the threat of damage from aftershocks.
"One of the things that we determined—which was perhaps key—was that if the main shock is not large, then the aftershocks don't matter...even if we were to put them in the same place. [The first earthquake] has to be something of at least design-level, a design-basis earthquake or larger," says van de Lindt.
"I think we were a little bit surprised at the relationship between how intense the main shock is and how much the aftershocks matter when you're quantifying the risk and design," he adds. "I thought it would be approximately linear. But it turned out that there is a real fast drop off."
Although researchers have been improving their accuracy in predicting if there will be aftershocks from a given earthquake or not, their location and depth are still elusive. Quantifying the risk to structures in this complex and uncertain environment presented a formidable challenge to the team.
"I can't tell you how many times during the course of the three-year research study we said, 'This explains why no one has worked on it,'" says van de Lindt. "A lot of our study was trying to account for risk before anything has happened. How would you change your design to account for the probability of a collapse in the main shock or—if it survives the main shock—the aftershocks?"
The researchers conclude that designing a building to withstand not only an earthquake, but the aftershocks, requires additional strength. "We are seeing that if you want the safety factor that you think you have, then you should increase the building stiffness and strength by 5 to 7 percent," van de Lindt says.
Van de Lindt hopes to eventually use a large-scale shake table to conduct tests for aftershock effects, using at least two, multistory concrete buildings similar to those built in the 1970s to "see how accurate we are on assessing the collapse risk."
"I've done a little bit at the wall level, but I would like to do it at the system level," he says. "It is something we look forward to, but is probably five years out."