NEEShub services are beginning to be phased out. Please begin using DesignSafe ( Please submit a support ticket ( if you need any assistance.


Support Options

Submit a Support Ticket

You are here: Home


UCLA Research Engineers Test California Levees’ Earthquake Resiliency

NEES@UCLA researcher Sophia Poulos places sensors on the model levee. in building the mound, the team used clay soil and geo-grid fabric to ensure the model levee would be strong enough to transmit the shaker forces into the peat-soil underlying the levee.

Members of the UCLA research team hammer vertical pipes into the nearby peat soil to visualize propagation of waves away from the levee.

Project PI Scott Brandenberg attaches an acceleration sensor to the base of the model levee.

The model levee with mobile shaker mounted to its crest with electrical inverter that powered the shaker motor in foreground.

The model levee with the mobile shaker in place. in this experiment, the peat soil near the levee was flooded with water to saturate it prior to testing.

NEESR: Levees and Earthquakes: Averting an Impending Disaster, Award #1208170 
Principal investigator: Scott J. Brandenberg, UCLA

In California, the Sacramento-San Joaquin River Delta supplies fresh water to 25 million people in southern and central California and irrigates the mighty breadbasket of the San Joaquin Valley.  In an area of about 1,000 square miles, the Delta’s 1,100 miles of levees protect the region from inundation and serve as a protective lifeline for California agriculture and nearly two-thirds of California’s population.

The aging Delta levees are a known flood risk, and many levees are composed of loose granular soils that are susceptible to liquefaction during earthquakes. (Liquefaction is when saturated soil loses its stiffness and behaves like a liquid.) However, no one knows exactly how the peat underlying many of the levees will behave in an earthquake.

That is now changing, thanks to a team of researchers led by Scott Brandenberg, a professor of civil and environmental engineering at the University of California, Los Angeles (UCLA). Funded by the George E. Brown, Jr. Network for Earthquake Engineering Simulation (NEES), Brandenberg and his team have partnered with the California Department of Water Resources (DWR) to determine how the Delta’s peat soil, which serves as the base of the levees, will affect these earthen structures during an earthquake.

NEES, a National Science Foundation (NSF) grant, is an earthquake-engineering network comprised of 14 equipment sites and a centralized cyberinfrastructure called the “NEEShub,” which is used for earthquake simulations and data sharing. Data from the UCLA tests will be publicly available via the NEEShub.

The state of Delta levees

The Delta was once a vast tidal estuary subject to frequent floods and inundated daily by ocean tides. Early settlers scraped together levees in order to hold back the water so they could farm the rich land. Today the Delta is a jigsaw puzzle of 65 islands surrounded by rivers, sloughs, and other water channels.

Many of the levees are over 100 years old, some constructed as early as the 1850’s. The levees are big mounds of dredged soil: silt, clay, sand, mining waste and peat. Over the years, some areas of the Delta have sunk 25 feet below sea level due to biodegradation and wind erosion of the rich peat soils, leaving the levee embankments even more exposed.

“I think it is crucial we stop relying on an 1,100-mile-long unengineered levee system to provide fresh water to much of the state’s population,” said Brandenberg. “Recent events such as Hurricanes Katrina and Sandy have taught us that levees tend to perform poorly in natural disasters.”

Given that California’s two biggest rivers drain to the Delta and that the region lies near the San Andreas Fault system, the Delta may be more vulnerable to flooding than New Orleans.

Charting new territory

Brandenberg’s UCLA research is the first-of-its-kind for evaluating the seismic response of levees built upon peaty organic soils such as those underlying the Delta levees. Composed of organic plant matter, peat soils are extremely absorbent, soft and compressible -- and therefore not considered subject to liquefaction the way sandy soils are. These qualities make it difficult to predict peat’s behavior during an earthquake.

Although it is generally considered to be non-liquefiable, peat may nevertheless contribute to levee failures in a couple of ways. For example, after an earthquake, peat may undergo volumetric contraction, caused by development of water pressure in the peat due to cyclic deformations. Also, peat may interact with overriding levee fills – the clay, silt, sand, and other materials in the levee -- in a manner that may alter liquefaction potential of the fill soils. Peat soil is very difficult to sample and test in the laboratory because it is fibrous and prone to sample disturbance. In situ field testing of the soil is therefore important.

The research team has conducted full-scale shake tests on model levees as well as lab tests. The shake tests required a portable earthquake simulation device called an “eccentric mass shaker,” which was provided by NEES@UCLA, one of 14 earthquake engineering equipment sites funded by the NSF.

Shaking things up

To conduct the shake tests, Brandenberg and his team built a model-scale levee six feet tall and 40 feet wide on an island located within the Delta. Unlike existing levees, which were constructed in a haphazard, non-engineered manner, the model levee was constructed from clay that was carefully compacted and reinforced with geogrids. Furthermore, the clay was unsaturated and therefore not susceptible to liquefaction like many Delta levees are.

The model levee was deliberately constructed to be strong enough to transmit loads from the shaker into the underlying peat – whose qualities were the focus of the research.

Researchers then placed the eccentric mass shaker on top of the levee, which shook the model levee with ground motion intensity similar to a moderate earthquake in the region. Shaking began with low intensity motions, and gradually increased with the shaker eventually imposing 40,000 pounds of dynamic force at a frequency of 3 Hz. Over 100 sensors were placed under, around, and on the levee to measure the soil’s reaction, including accelerations of the model levee and surrounding soil, pore water pressure of the soft saturated peat underlying the levee, and settlement of the levee.

During the field tests, the researchers observed that the levee rocked significantly, unlike a levee built atop stiffer soil. Waves propagating out from the levee were also clearly visible in the peat soil surrounding the embankment.

Current model predictions for liquefaction in levees do not account for stress concentrations caused by such rocking movement. Dr. Brandenberg expects that data from his experiment will allow for more accurate prediction of liquefaction potential for levees on soft peat soil.

In addition to Delta field tests, Dr. Brandenberg’s team also conducted extensive laboratory testing of peat samples, which revealed that peat may be susceptible to volumetric contraction following an earthquake, a previously unknown quality of the soil.

In the next phase of the research, the team plans a series of centrifuge tests at the NEES facility at UC Davis featuring liquefiable levee fills resting atop peat foundation soils.

“We will place water against one side of the levee to replicate current conditions in the Delta and shake the model with realistic earthquake ground motions consistent with seismic hazard in the Delta,” Brandenberg explained. He anticipates the simulated levee will fail during shaking.


New data for decision-making

Data from Brandenberg’s experiments will lead to new techniques for understanding the complex interaction between the levee soil and its peat base during earthquakes, and the information will allow engineers to predict levee settlement caused by the reconsolidation of the peaty organic soils underlying the levees.

Specifically, the research identified several unrecognized mechanisms of seismic deformation potential that are not currently included in most seismic levee evaluation procedures.

These mechanisms include contraction of the peat soil after an earthquake, and levee-peat interaction (i.e., rocking and translation of the levee atop the peat) that may result in stress concentrations in levee fills and in the soils beneath the levees. These stress concentrations may exacerbate liquefaction of granular soils and post-earthquake contraction of peat.

The project team is currently analyzing the experimental data, which is publicly accessible through the NEEShub, the NEES network’s data repository. Brandenberg and his team will be presenting their findings to DWR and other stakeholders about the seismic response of Delta levees. The data and results are freely available to researchers around the world studying seismic response of levees.


Effecting change in the Delta

Experts say there is a 60 percent chance of a significant earthquake in the area within the next 30 years. And in 2008, the Delta’s state monitoring agency estimated that a moderate, 6.5 magnitude earthquake could cause widespread levee failures and cost the state more than $15 billion in economic losses -- and disrupt the area’s fresh water supply for as long as 28 months.

Although earthquake-caused breaches in the Delta levees are a natural disaster waiting to happen, California has struggled to implement policies and infrastructure to avert it.

Implementing change to avert disaster is complicated by the Delta’s unique ecological setting, and potential solutions to divert water supplies in a way that bypasses the Delta’s fragile levees are staunchly opposed by local interests. They worry that tunnels or canals could hurt their farms by depriving the Delta land of fresh water and by increasing salinity in Delta channels.

In fact, Brandenburg was startled to learn how many Delta residents think the earthquake risk is overstated. “The most surprising aspect of my study was that a significant number of people with local Delta interests are skeptical of the seismic hazard for Delta levees,” he said. “From my perspective, the seismic risk to Delta levees cannot reasonably be doubted. I had never considered the possibility that so many people would voice such doubts.”

Indeed, a significant impediment to implementing a solution is that seismic risk is misunderstood and often openly doubted by Delta residents who view the State’s infrastructure plan as an unnecessary expense that would deprive the Delta of fresh water.

Brandenberg and his team will help enlighten the public on this poorly-understood issue.  They are creating models that can predict the seismic response of peat, and they are taking action to disseminate their results to stakeholders, who include Delta residents, government agencies, and lawmakers.

Dr. Brandenberg’s research will help the state making key decisions regarding the Delta levees and the protection of the state’s water supply. Later this year, he and his team will advise California’s DWR on the earthquake risk to the existing levees.

The researchers also plan to coordinate an outreach effort with the DWR to Delta residents.

“Addressing the general misunderstanding of the area’s seismic hazard perhaps will be the biggest impact of our outreach plan,” Brandenberg added.

Images courtesy Teresa Morris, Purdue University