Japan earthquake rocked soil stability

GEORGIA TECH (US) — Japan’s March 11 magnitude 9.0 earthquake weakened the subsurface rock and soil by as much as 70 percent, according to a new study.

Understanding how subsurface materials respond could be vital information for engineers and architects designing future buildings that are able to withstand the level of acceleration measured in the Japan quake.

The information will also help seismologists develop new models to predict the effects of these rare and extremely powerful events.

“The degree of nonlinearity in the soil strength was among the largest ever observed,” says Zhigang Peng, associate professor of earth and atmospheric sciences at Georgia Institute of Technology.

[sources]

“This is perhaps not too surprising because the ground shaking generated by this earthquake—acceleration as much as three times the Earth’s gravity—is also among the highest ever observed.”

The findings were reported in the journal Earth, Planets and Space (EPS).

Peng and graduate student Chunquan Wu were among the first scientists to examine data recorded by the high-quality seismometers that are part of the Japanese Strong Motion Network KIK-Net.

The researchers studied data from six stations that have strong velocity contrasts between the surface soil layers and the underlying bedrock. The stations have accelerometers both on the surface and in boreholes located on bedrock far beneath it.

“We were trying to understand the relationship between soil nonlinearity and peak ground acceleration (PGA), which is a measure the ground shaking,” says Wu.  “We want to understand what parameters control this kind of response.”

By comparing data on the acceleration of motion from sensors on the bedrock to comparable information from surface sensors, researchers were able to study how the properties of the soil changed in response to the shaking by computing the spectral ratios of each pair of station measurements, and then using the ratios to track the temporal changes in the soil response at various sites at different levels of peak ground acceleration.

“The shear modulus of the soil was reduced as much as 70 percent during the strongest shaking,” Wu says.

“Typically, near the surface you have soil and several layers of sedimentary rock. Below that, you have bedrock, which is much harder than the surface material. When seismic waves propagate, the top layers of soil can amplify them.”

Nonlinear response from soils is not unusual, though it varies depending on composition. Similar but smaller effects have been seen in other earthquake-prone areas such as California and Turkey.  The shallow layers of the Earth’s upper crust can be complex, composed of varying types of soil, clay particles, gravel, and larger rock layered in sediments.

Because the March 11 quake lasted an unusually long time and generated a wide range of ground motions of greatly varying strengths, it provided an unprecedented data set to scientists interested in studying nonlinear soil behavior.

Beyond the immediate effect of the strongest shock, the researchers were interested in how the soils recover their strength after the shaking stops, a recovery time that  can vary from fractions of a second to several years.

“It is still not clear whether there could be longer recovery times at certain sites,” Wu says.  “This is a function of soil type and other factors.”

If the soils are very porous, water can lengthen the recovery.  “For porous media, the ground shaking could cause water to go into the pores, which will also reduce the shear modulus of the soil.  If water is involved, the recovery time will be much longer.”

The resesarchers also studied soil response to aftershocks, which ranged up to magnitude 7.9 after the main Tohoku earthquake.

Knowing how soils respond to strong shaking is important to predicting how motion deep within the Earth will be translated to structures built on the surface.

“Understanding how soil loses and regains its strength during and after large earthquakes is crucial for better understanding and predicting strong ground motions,” Peng says.

“This, in turn, would help earthquake engineers to improve the design of buildings and foundations, and could ultimately help to protect people in future earthquakes.”

The research was sponsored the National Science Foundation (NSF) and by the Southern California Earthquake Center (SCEC).

More news from Georgia Tech: www.gatech.edu/newsroom/