An earthquake in Christchurch, New Zealand, destroyed this home in 2011; now researchers are testing the best and most economical ways to strengthen the city’s sandy soils so that the safe reconstruction of such low-rise buildings can begin. Wikimedia Commons/ Schwede66
A team of engineers examines the best ways for the reconstruction to move forward in Christchurch, New Zealand, on soils prone to liquefaction.
April 22, 2014—A team of researchers and engineers has field-tested several soil-improvement methods to determine which will best enable the earthquake-devastated city of Christchurch, New Zealand, to rebuild with greater confidence on soils that are still susceptible to liquefaction.
Christchurch experienced powerful earthquakes in September 2010, February 2011, June 2011, and December 2011. Although the February 2011 earthquake was of a lesser magnitude—6.2 on the open-ended Richter scale—it created some of the most intense ground shaking ever recorded in an earthquake. The costs to rebuild from the earthquakes is approaching NZ$40 billion (U.S.$34.3 billion), which amounts to nearly 25 percent of the country’s 2013 gross domestic product.
The extensive damage can be traced to the soils beneath the city. A thick layer of sand, varying from 10 to 30 m in depth, underlies the city—and it is highly prone to liquefaction. That coupled with the reflection and refraction of earthquake waves contributed to the intensity of the shaking during the recent temblors.
Approximately half of the reconstruction costs that the city faces are focused on the city center, where high-rise office buildings were heavily damaged. The remainder will go to repairing or replacing the more than 60,000 residential structures that have been affected. Approximately 20,000 of those structures are heavily damaged and 10,000 have been or will be demolished.
Large buildings built and rebuilt in Christchurch following the earthquakes will be founded on deep piles or piers. But those foundations are not economically feasible for homes and low-rise structures. So engineers are developing and testing soil improvements that can be used beneath these structures to augment their slab foundations, according to Ken Stokoe, Ph.D., P.E., D.G.E., M.ASCE. Stokoe is the holder of the Jennie and Milton Graves Chair in Engineering in the Department of Civil, Architectural, and Environmental Engineering at the University of Texas at Austin.
The goal, he says, is to develop a “crust” approximately 3 m thick beneath lower-height buildings. “When another earthquake occurs, and some liquefaction is created, the triggering in the localized area beneath that structure will be minimized,” Stokoe explains. “The structure will go through significant shaking, and settle, but the damage it will sustain now will be repairable.”
He compares the reaction to a boat floating on water; the building “floats” on the liquefied soil during the earthquake and then settles once the shaking has stopped. Stokoe says that to achieve this, the ground improvement solution must be cost-effective, allow for competition among contractors, be ready for implementation quickly, and, most significantly, be verifiable at each project location.
“The solutions must be verified at each construction site—tens of thousands—before insurance companies will pay,” Stokoe says. “We’ve never seen this need before.” This requirement is the result of a provision in earthquake insurance issued in New Zealand—but uncommon in other parts of the world—that specifies that the soil itself must be returned to prequake conditions.
The team has tested four ground improvement methods: rapid-impact compaction (RIC), rammed-aggregate piers (RAP), low-mobility grouting (LMG), and horizontal beam construction with either a single row of beams (SRB) or a double row of beams (DRB).
The methods were tested at three sites selected for their differing ground conditions, which are representative of those found throughout Christchurch. The team developed sensors and pushed them into the ground using the massive triaxial vibroseis truck known as T-Rex. (See “T-Rex Pounds the Ground in New Zealand,” on Civil Engineering online.) The sensors’ holes were sealed with bentonite and the wires from the sensors were connected to monitoring systems in an instrumentation vehicle. The T-Rex vehicle was then placed over the monitoring zone and used to simulate ground shaking at a wide range of earthquake levels.
The team found that RIC, RAP, and DRBs performed the best in testing. Although the beams in both DRB and SRB are drilled horizontally and produce minimum disturbance, it was found that the SRB did not offer sufficient projection. The LMG actually made the soil looser at some depths.
“Installation of the horizontal beams actually does a great job of not disturbing anything,” Stokoe says. “But when you don’t disturb anything, you leave all the loose soil. A single row of beams didn’t work, where the double row will be a much better isolator. The two beams still aren’t disturbing much, but they do form a much better mat—or boat—to protect the structure.
“The low-mobility grout hydrofracks the soil,” Stokoe notes. “Because they push [the grout] in under pressure, it goes in every direction. You have no idea where it goes. It seems that grouting never forms that beautiful column that is drawn in the contractor’s sketch. But nobody checks. When it pushes out sideways, it tends to hydrofrack the shallower soil. It lifts the soil and then the soil [spreads] out in a looser state.”
The project was funded by the National Science Foundation—via the George E. Brown Jr. Network for Earthquake Engineering Simulation (NEES)—and the New Zealand Earthquake Commission (EQC). The NEES consortium, led by Purdue University, has 14 sites across the United States, including UT, where it is led by Stokoe.
The research team included Brady Cox, Ph.D., P.E., A.M.ASCE, an assistant professor at the University of Texas; Sjoerd Van Ballegooy, Ph.D., a senior engineer for Tonkin & Taylor, Christchurch; and UT graduate students Julia Roberts and Sungmoon Hwang. “This work represents the most comprehensive study ever performed for residential structures and lowrise buildings,” Stokoe says. “These structures are founded on thin—9 to 24 inches thick—slabs at or near the ground surface. The reason the study has never been done before is that it is generally a dispersed, uninsured problem and T-Rex did not exist.”
The team is just beginning to publish its findings, which Stokoe says will be applicable far beyond Christchurch. “What we have done has some applications everywhere there are big earthquakes,” Stokoe says. “Everywhere they have earthquakes, our results will be used.”