Q&A: 30 years after the Loma Prieta earthquake

On Oct. 17, 1989, destruction from the Loma Prieta earthquake killed 67 people and injured 3,757. The magnitude 6.9 quake went down in the history of California’s central coast as the most damaging seismic event since 1906. It sent seismic waves from its origin in the Santa Cruz Mountains to San Francisco, the East Bay and beyond. The 20 seconds of shaking knocked down part of the Bay Bridge, collapsed a section of freeway in Oakland and caused more than $5 billion in damages.

Anne Kiremidjian, a professor of civil and environmental engineering, remembers exactly what she was doing 30 years ago when the shaker struck. She was driving in Los Altos southeast of Stanford’s campus around 5 p.m. “As the aftershocks were coming it was amazing to see the cars across from me on Foothill Expressway heave up and then down and, as the seismic wave rolled across, the cars on our side did the same,” said Kiremidjian, who studies the intensity and duration of ground shaking during a quake and estimates the probable structural damage. On the Farm, library collections fell over like dominoes and huge pieces of concrete were shaken from the facades of old buildings. More than 200 campus structures were damaged, some beyond repair. That night, 1,600 students were displaced from their residences.

Loma Prieta caused an underground rift along 22 miles, mostly on a previously unknown fault, whereas the 1906 quake occurred along the San Andreas Fault and ruptured about 300 miles of Northern California with an estimated magnitude of 7.9. But Loma Prieta impacted a larger population and more buildings – and it served as a much-needed wake-up call to residents, city planners, engineers and geophysicists. According to the U.S. Geological Survey, there is a 72 percent likelihood of at least one earthquake of magnitude 6.7 or greater striking somewhere in the San Francisco Bay region before 2043.

“As soon as it had stopped, I went down the hall to an old analog phone – all the others were computer phones and dead – to call Dr. Rob Wesson, PhD ’70, who was the head of the earthquake office at the USGS,” said geophysics professor William Ellsworth, who was working as a research geophysicist at the U.S. Geological Survey (USGS) in his Menlo Park office at the time of the quake. “He was excited to hear me and wanted to talk baseball, at least until I told him that we had just experienced a major earthquake and our lives would be different from now on. How true that proved to be.”

Reflecting on the 30th anniversary of Loma Prieta this week, earthquake experts shared their perspectives on how the event impacted them, the Bay Area and the research community at large. Greg Beroza is the Wayne Loel Professor in the School of Earth, Energy & Environmental Sciences (Stanford Earth), Paul Segall is a professor of geophysics and Gregory Deierlein is the John A. Blume professor in the School of Engineering.

Where were you when Loma Prieta struck?

KIREMIDJIAN: I felt a strong jolt and thought the car behind me had run into me. When I looked back, the car behind me was stopped more than two feet behind me. I wondered what was the problem, and as I was thinking that there was a second strong jolt, the trees started swaying and so did the traffic signals. My first reaction was that there was a strong wind, but when I pulled my window down, I realized that there was no wind and we had just felt a strong earthquake and its significant aftershock. The lights went out at that point and all the drivers proceeded with great caution.

ELLSWORTH: The earthquake had caused the electrical grid to crash and we had no power. The one real-time resource we had was an old black-and-white monitor that displayed the current earthquake detections from the Real Time Processor that monitored the seismic network. The detections were literally flying by on the screen, but I could see just enough to identify the names of the stations that were most common, which put the earthquake in the southern Santa Cruz Mountains.

The building was eerily quiet that night (I took the overnight shift), with only one reporter, Charlie Petit from the San Francisco Chronicle, showing up. We had a long discussion about what had happened and what we knew, which was very little at that point. The quiet continued for about a day longer when the press showed up in large numbers. I think that there was not a time in the next month when there wasn’t a film crew somewhere in the building.

BEROZA: I was a postdoc at MIT, where I got my PhD. I went home to watch the Giants vs. A’s in the World Series and turned on the TV at about 8:30 to hear Al Michaels state that given what had just happened, it’s not surprising that the World Series game was cancelled. I immediately suspected an earthquake, but wondered whether I was biased. There was no more information on that, or any other, TV channel, which were all showing sitcoms. It finally occurred to me to turn on the radio (this was before the World Wide Web), where reports were coming out about the earthquake.

How did Loma Prieta impact you professionally? How did it affect you personally?

ELLSWORTH: The Loma Prieta earthquake redirected my work and that of most of my USGS colleagues. We had just the year before released the first 30-year earthquake forecast for California. This report highlighted the southern Santa Cruz Mountains as one of the more hazardous sections of the San Andreas Fault system. While the Loma Prieta earthquake didn’t perfectly fit the forecast, it was close enough (having half of its length on the San Andreas Fault, but most of its slip on the previously unknown Loma Prieta Fault) that an update of the 1988 forecast was needed. This was completed in 1990. It led to the development of much-improved forecasts for the Bay Area (in 2002) and for the entire state in the following decade. Preparing these forecasts was a broad community effort involving hundreds of geophysicists, geologists and engineers.

SEGALL: This was in the very early stages of using GPS to measure crustal motions. Right after the earthquake, I was able to borrow some early generation receivers from a local company and enlist graduate students to make some surveys of the area. I was concerned that post-earthquake adjustments could stress the part of the San Andreas closer to Stanford, potentially triggering another quake – such sequences had occurred along the North Anatolian Fault in Turkey.

KIREMIDJIAN: Prior to Loma Prieta I had visited many locations that had been affected by significant earthquakes, including 1973 Managua, Nicaragua; 1976 Guatemala City, Guatemala; 1986 El Salvador; and 1988 Spitak, Armenia, earthquakes. While I had experienced some of the aftershocks from these earthquakes, it was the first time in my adult life that I was experiencing a real earthquake.

In the next several days my colleagues, professors [Haresh] Shah, [Helmut] Krawinkler and [James] Gere, our MS and PhD students and I together with the facilities project managers  inspected and assessed the damage to buildings on campus, determining which can be opened and which should remain closed. We also organized a trip for our students to look at the damage to the Cypress Viaduct across the Bay that had collapsed, the areas around the Marina in San Francisco that experienced liquefaction and other locations where there was visible damage. This was a real-life laboratory that provided a tremendous learning experience not only for the students but for us, the faculty, as well.

What might be the consequences of another big earthquake in the Bay Area?

ELLSWORTH: Many people think of Loma Prieta as a Bay Area earthquake. While it is certainly true that there was major damage in San Francisco and Oakland, the earthquake was more of a Monterey Bay earthquake, as the communities of Santa Cruz and Watsonville were heavily impacted by the event. If a similar earthquake struck now in the southern Santa Cruz Mountains, the outcome would be much better than in 1989. Major efforts have been made to improve the seismic safety of our roads and bridges, in particular. Many seismically vulnerable buildings have been retrofit or taken out of service, although there are many problem buildings still out there. Our ability to rapidly identify where strong shaking would be expected to cause damage has also improved markedly, and so I would anticipate a much-improved response, particularly to communities like Santa Cruz and Watsonville.

DEIERLEIN: Since the Loma Prieta earthquake, Stanford has accelerated its programs for seismic risk mitigation on campus, including proactive retrofit of existing buildings, seismic resistant design of new buildings, back-up power, emergency water supplies, etc. So, while the Stanford campus has many more people and buildings today, I think that today it is much better prepared to resist earthquakes. The largest risk to Stanford may well be the surrounding communities, where there are hazardous older masonry, concrete and soft-story buildings that have a greater risk of damage and could impact Stanford faculty/staff/students who live off campus and flow of goods/services to Stanford.

How have earthquake engineering, sensing and forecasting technologies changed in the past 30 years?

KIREMIDJIAN: On the structural response side, we now have a better understanding about the nonlinear behavior of our structures and have developed sophisticated models to capture this nonlinear behavior. The nonlinear behavior of the materials we use in constructing our structures is due to the large deformations imposed by the earthquake vibrations and the physical limitations of these materials. Together with these developments, new materials are being designed to meet some of these requirements of large deformation with increased strength.

ELLSWORTH: The seismic monitoring system has been significantly upgraded from the analog instruments in use in 1989 to modern digital instruments that provide detailed real-time information on earthquakes as they happen. The USGS will be starting statewide alerting of earthquakes (ShakeAlert) this month, which is a major milestone for “early warning” in the state. One major benefit of ShakeAlert that isn’t being discussed enough is the information it produces that will pinpoint the areas of strongest shaking within minutes of the event, leading to much more effective and timely emergency response.

The geologists have not been left behind either. Many new fault investigations within the region have been made that have sharpened our understanding of the frequency of damaging earthquakes, which also feeds directly into the current 30-year forecast.

DEIERLEIN: Since Loma Prieta, state and regional governments have been proactive in retrofitting transportation and water infrastructure. For example, Caltrans has spent billions of dollars seismically retrofitting bridges around the state. And, the San Francisco Public Utilities Commission has done major seismic upgrades to water supply pipelines and infrastructure.

BEROZA: We have much, much more capable computers, which enable much more realistic simulations and much more comprehensive data analysis. As a result, we can get a much clearer view of earthquake processes than before and we can learn more as a result.

What have we learned from Loma Prieta?

KIREMIDJIAN: Loma Prieta exposed the vulnerability of existing structures. It also pointed out how vital they are to our continued functionality and recovery. It showed us that we do not have sufficient knowledge of the behavior of seismic faults and the ground motions that are generated. The field is so broad and interdisciplinary that, although we have made great strides in our understanding and modeling of its various components, I feel that a lot more remains to be done.

ELLSWORTH: Scientifically, Loma Prieta reminded us that large earthquakes will continue to occur without warning on faults that we have not detected. The forecast of 1988 was at best only partially fulfilled. But despite this, the methods of probabilistic seismic hazard analysis provide clear scientific guidance about earthquake probabilities in a framework that can be used to prioritize mitigation measures. We also learned that the near-surface geologic factors that control shaking (and hence damage) can vary significantly over distances of a city block, but that they can also be identified – which has led to major changes in how seismologists and engineers forecast ground motion amplification. The complexity of the rupture and how it affected other faults in the San Francisco Bay region spurred research on fault interaction, which continues today as a major thrust of earthquake research.

SEGALL: There was a significant amount of vertical motion in the earthquake, so we now better understand how the Santa Cruz Mountains are built by repeated earthquake slip. The damage was concentrated not just near the epicenter, but also in areas of fill material that we knew were susceptible to strong shaking: South of Market and the Marina district in San Francisco, and the I-880 Cypress structure in Oakland. The evidence for strong amplification of shaking due to the bay muds in the I-880 Cypress structure was extremely compelling. We learned a good lesson in the many landslides triggered by the earthquake.

It was also an important lesson on how isolated the affected area can be, when roads and other infrastructure are damaged. It’s a lesson I’m afraid we have forgotten.

Deierlein is also director of the John A. Blume Earthquake Engineering Center.

Related

After the Big One: Understanding aftershock risk

Geophysicist Gregory Beroza discusses the culprits behind destructive aftershocks and why scientists are harnessing artificial intelligence to gain new insights into earthquake risks.

Stanford Earth Matters

Media Contacts

Danielle T. Tucker
School of Earth, Energy & Environmental Sciences
dttucker@stanford.edu, 650-497-9541