Caltech: Caltech’s Seismo Lab Celebrates 100 Years at the Forefront of Earthquake Science
On November 12, the Caltech Seismological Laboratory—the Seismo Lab, for short—will celebrate its centennial, marking an unparalleled century at the forefront of earthquake science and geophysics.
Founded in 1921 by prominent seismologist Harry Wood, the Seismo Lab actually reached the centennial milestone last year, but restrictions on gatherings to combat the COVID-19 pandemic delayed the official celebration until now. Wood began his seismological studies at the Mount Wilson Observatory, with funding initially supplied by the Carnegie Institution for Science.
In 1924, Caltech and Carnegie agreed to jointly support and manage a new lab, built in the mountains above Pasadena, which became the Kresge Seismological Laboratory. Kresge was located in the mountains to ensure that its seismometers could be placed on bedrock, ideal for making accurate seismic measurements. Later, as space grew tight, the Seismo Lab expanded to a nearby 40-room house on North San Rafael Avenue, which was remodeled and named the Reuben H. Donnelley Seismological Laboratory to honor the father of one of the donors who made it possible. In 1974, the Seismo Lab moved to Caltech’s campus.
At the time of the Seismo Lab’s creation, the U.S. government was not in the business of monitoring earthquakes, and arrays of seismometers did not yet exist. Wood envisioned the construction of a network of seismographs around the seismically active southern California in order to understand earthquakes locally as well as what they could reveal about the deep Earth.
“Modern earthquake seismology was born at Caltech in the 1920s and 1930s,” says Mike Gurnis, the John E. and Hazel S. Smits Professor of Geophysics and current director of the Seismo Lab.
At the Seismo Lab, Wood worked with Mount Wilson Observatory astronomer John Anderson (a pioneer of instrumentation) to develop the Wood–Anderson torsion seismometer, which was capable of detecting short-period seismic waves, which provide information about nearby earthquakes. This seismometer, originally completed in 1922 became the standard for the field, and examples remain in use to this day.
“As of 1920, seismometers were these behemoth, big, heavy instruments that mostly recorded what we call teleseismic waves, so waves you’d never feel from large earthquakes around the world,” said Sue Hough, a Pasadena-based scientist in the earthquake hazards program at the United States Geological Survey (USGS), who has written numerous books on earthquakes and the history of seismology. (Hough was speaking in a 2022 Caltech Heritage Project interview that was excerpted for a book on the Seismo Lab’s history). “A magnitude 8 happens in Japan, and the waves are a very, very slow, long period, and these seismometers would detect them. But they weren’t good at recording local earthquakes. The Seismo Lab was founded initially to look at local earthquakes, and they realized that there really wasn’t an instrument that fit that bill.”
The Seismo Lab soon attracted future leading lights in the Earth sciences, including seismologist Hugo Benioff (PhD ’35) in 1924 and physicist Charles Richter (PhD ’28) in 1927. It also attracted the already-prominent German geophysicist Beno Gutenberg in 1930.
An expert on sound, Benioff constructed advanced instruments for detecting seismic waves, including the Benioff seismograph in 1932 and the Benioff strain instrument for detecting the stretching of the Earth’s crust.
Meanwhile, Richter and Gutenberg collaborated to develop what came to be called the Richter Scale for measuring the magnitude of an earthquake’s shaking. Prior to that, the way to quantify the size of an earthquake was the Italian Mercalli–Cancani–Sieberg (MCS) scale, which became the “modified Mercalli intensity scale of 1931” (MM31) when translated to English by Wood and Frank Neumann. It relied on fairly subjective observations of the impact to structures in the area of a quake. By contrast, the Richter Scale, first laid out in a paper published in 1935, used a seismograph to measure the actual size of an earthquake by taking into account both the intensity of shaking and the distance from a quake’s epicenter. It became the world standard until it was replaced by the moment magnitude scale, developed in 1979 by Thomas Hanks (PhD ’72) of the USGS and Hiroo Kanamori, now Caltech’s John E. and Hazel S. Smits Professor of Geophysics, Emeritus.
Caltech took full charge of the Seismo Lab’s administration in 1937 through an agreement by which the Carnegie Institution relinquished co-responsibility. Benioff and Richter, who had been employees of Carnegie at the Seismo Lab, both became assistant professors at Caltech around this time. Gutenberg, who had accepted a position as professor of geophysics at Caltech in 1930, became the second director of the Seismo Lab from 1946–57, and helped to lead the Seismo Lab to prominence in the study of both local earthquakes and the use of seismometers to study the Earth’s deep interior. Frank Press, who later served as chief science advisor to President Jimmy Carter and then as president of the National Academy of Sciences (NAS) for 12 years, assumed leadership of the Seismo Lab in 1957, two years after he was recruited to come to Caltech.
“Now, don’t think that when I took over, I had an agenda,” Press said in a 1983 interview. “The science moved ahead, and I just simply—using the resources of the lab and also the resources that I developed in the government—brought in new people and new technology. This meant computers; this meant new types of seismographs; this meant more field exploration. My attitude toward graduate students was different, I must say. I used graduate students as colleagues: I gave them the best problems to work on, and I encouraged them.”
While Press was at the helm of the Seismo Lab, the two largest earthquakes ever recorded occurred: the magnitude-9.4–9.6 Valdivia earthquake in Chile in 1960 and the magnitude-9.2 Alaska earthquake in 1964. The Valdivia quake allowed Benioff, Press, and Stewart Smith (PhD ’61) to unequivocally observe and measure the “ringing” of the earth, so-called free oscillation, for the first time. The idea that large earthquakes could cause the planet to ring like a bell had been around since the 19th century, but the Valdivia quake provided the first unambiguous measurement of the phenomenon.
“This is one of the most seminal achievements in geophysics,” says Gurnis, as it allowed for observations of the large-scale structure of the earth. “This discovery led to the first direct constraints on how density varies in the core and mantle of our planet and in turn allowed geophysicists to determine more precisely what the planet is made of.”
During the early-to-mid 20th century, a debate sprang up in the geosciences around the new theory of plate tectonics: the idea that the rocky outer crust of the Earth was broken up into segments called plates that moved over the mantle below, which is solid but is under such intense heat and pressure that it behaves like a fluid in geologic time. Though the pioneers of the theory were not located at the Seismo Lab, their work had a tremendous impact on the future of the science, and by extension, the Seismo Lab’s work.
Caltech alumnus Robert Geller (BS ’73, MS ’75, PhD ’77), professor emeritus of the University of Tokyo, remembers the excitement sparked among the Seismo Lab staff when they were reviewing a paper by researchers at MIT about the driving forces of plate tectonics: “[M]any people at the Seismo Lab were really interested in the topic. So, these authors’ theories were a lunchtime discussion topic for two or three weeks. … That kind of discussion was very educational for a young scientist. Lots of hot scientific issues were always being discussed like that while I was at the Seismo Lab,” Geller said in a 2022 Caltech Heritage Project interview.
The Seismo Lab moved onto Caltech’s campus in 1974 with the construction of the Seeley G. Mudd Building of Geophysics and Planetary Science, currently known as South Mudd.
“Our operation has grown so much … There was no question that moving was a necessity,” Don Anderson, then-director of the Seismo Lab and professor of geophysics, said at the time. “It has always been a part of the Institute’s overall plan for us to be located on the campus, and it’s most helpful to be near the rest of the Division of Geological and Planetary Sciences [GPS]. Still, leaving the old building evokes feelings of regret. There are many fond memories of the old hilltop home with its solitude, its muraled ceilings, its landscaped grounds and tennis court, and its residential atmosphere.”
Robert Sharp (BS ’34, MS ’35), chair of what was, at the time, Caltech’s Division of Geological Sciences, recalled in a 1981 interview that space was continually a concern at the San Rafael facility: “One day Frank Press came to me and said, ‘We’ve got to have still more room.’ We had all been unhappy that the seismological group was so separated from us on the campus. So I said to Frank, ‘I can’t get you more room out there, but if you will come to the campus, I will try to get you a building here.’ He finally bought that idea, and that’s how we got them down there into South Mudd.”
A few years later, the Seismo Lab and the USGS began a cooperative program that developed into the Southern California Seismic Network (SCSN), which is currently made up of more than 400 seismic stations throughout Southern California. Until this time, the federal government had largely left earthquake monitoring to universities. But in 1977, it launched the National Earthquake Hazards Reduction Program (NEHRP), making the USGS the tip of the spear for federal seismic monitoring.
“USGS–Caltech network operations began in the late ’70s,” Hough told the Caltech Heritage Project. Hough joined the USGS Pasadena office—a house directly across Wilson Avenue from South Mudd—in 1992. “At that point, the Pasadena office, its whole reason for existence was operating a seismic network in collaboration with Caltech.”
While Anderson was leading the Seismo Lab during the late 1970s the U.S. government began funding efforts to determine whether it was possible to predict earthquakes, spurred in part by the apparently successful prediction of an earthquake in northeastern China in 1975. Chinese officials reportedly ordered evacuations in Yingkou on the morning of February 4, 1975. A magnitude-7.3 earthquake struck the region that evening.
At the time, there was little communication between scientists in the U.S. and China, so the event was shrouded in mystery. “We were told, this is your job: predict earthquakes. And nobody knew how to do it,” Thomas Heaton (PhD ’78), professor of engineering seismology, emeritus, said in 2021. “We’d have arguments about whether it was even possible.”
The Seismo Lab’s Clarence Allen (MS ’51, PhD ’54), a professor of geology and geophysics who passed away in 2021, helped to establish the California Earthquake Prediction Evaluation Council (a committee of earthquake experts tasked with reviewing earthquake predictions and advising the state government), but remained agnostic about the possibility of predicting quakes. “We had some people who were true believers, and some who were true skeptics, but Clarence was just, ‘let’s just look and see where the data is,'” Heaton says.
Lucy Jones, a career-long USGS seismologist and now a visiting associate in geophysics at Caltech, remembers pursuing the hope of prediction early in her career. “I went to China [four times between 1979 and 1983] and realized that their work was all based on foreshocks, and eventually there was this big shift away from the optimism of predictions. By the late ’80s, people really gave up,” she says.
The attempts to predict earthquakes attracted the attention of Hollywood: In 1974, Charlton Heston starred in a move titled Earthquake, about a fictional temblor that wreaks havoc in Los Angeles. “The guys who wrote the movie based the scientists on the Caltech Seismo Lab,” Heaton recalls. “At the beginning, for dramatic effect, they had a professor modeled after Clarence exploring geology in the field across the San Andreas Fault. And there’s a foreshock, and a trench collapses and kills this professor. So, according to Hollywood, Clarence died in ’74,” Heaton says with a laugh. (Earthquake is among the films that will be reviewed at the “Shaking in Our Seats” event on November 12 to commemorate the Seismo Lab’s centennial.)
Hollywood was not the only form of outreach for the Seismo Lab; since its inception, the lab has cultivated a relationship with the media and has become the principal source in the region for seismological expertise, particularly following an earthquake in Southern California.
In 1992, Caltech and the Los Angeles Times Foundation collaborated to build a media center at the Seismo Lab, with dedicated studio lights and hookups for TV crews to connect (via antennas on the roof) with their studios. “The media center at the Seismo Lab is a classroom and the students are the members of the news media who would come at a moment’s notice when the earth shook,” says KNBC’s Conan Nolan, a Los Angeles-based broadcaster who has been reporting the news since the early 1980s. “They’re all looking for answers. Earthquakes are the kind of calamities that have a mystery to them that you don’t see anywhere else. It takes place deep beneath the earth’s surface, so you don’t see it, you just see the effects of it.”
Nolan knows his way around the Seismo Lab’s media center as well as anyone. “I spent two weeks there every day following [the 1994 magnitude-6.7] Northridge quake. And I grew to really truly appreciate several things: the dedication of the people who work there; the commitment to the science and to getting it right,” he says.
Gurnis notes that the connection with the media goes all the way back to Charles Richter, who enjoyed a strong relationship with the press. “Back then, if there was a shaking, no one knew where it was, but Richter and his colleagues could determine the epicenter quickly in the lab. The radio stations would call him up and talk to him, and he loved to talk to them.”
The period of time before the media center was built was chaotic, with news crews showing up and wandering the halls of South Mudd, hoping to corner seismologists for an interview, Nolan remembers. “Once the earthquake starts, it’s a race to get [to the Seismo Lab],” he says.
Nolan describes the media center itself as “a physical manifestation of what everyone was seeking after an earthquake: clarity, understanding, and a sense of control.” Given the seismically active nature of Southern California, the media has acted as a bridge between seismologists and the public, resulting in a rare, decades-long bond between scientists and reporters.
Jones, a media favorite at the Seismo Lab, vaulted into the public eye in the aftermath of the 1992 magnitude-6.1 Joshua Tree earthquake, which struck late in the evening. She and her husband Egill Hauksson, Caltech research professor of geophysics, lived near campus with their two children, so they immediately rushed to campus—children in tow. As she conducted endless on-camera interviews that day, Jones held her youngest son, then 19 months old, who slept on her shoulder.
“Only in L.A. are seismologists really rockstars,” Nolan says.
Kanamori became director of the Seismo Lab in 1990 and expanded the lab’s instrumentation. “For the period between 1985 to 1998, thanks to the efforts of Egill Hauksson and Tom Heaton, the Seismo Lab built a wonderful broadband seismic network in Southern California,” Kanamori said in a 2020 interview.
The network’s broadband capabilities meant it could monitor a wider range of frequencies than traditional seismic networks, yielding a higher resolution understanding of the Earth’s interior. “Although the network was effectively used for long-wavelength applications, it had not been used much for research using short-period waves. I am sure everyone thought that Don’s leadership in this field could promote good science, making good use of the modern network. As we have seen, it worked well indeed,” Kanamori added
Kanamori also helped to develop an algorithm for a new earthquake early warning (EEW) system—a pivot away from trying to predict earthquakes and instead mitigating the havoc they can wreak. Together with Heaton, Hauksson, and colleagues at the USGS, Kanamori is among the founders of ShakeAlert, an EEW system that consists of a network of sensors near faults that transmit signals to data-processing centers when shaking occurs. These data-processing centers use algorithms to rapidly determine an earthquake’s location, magnitude, and the fault-rupture length, and send out an alert for earthquakes of sufficient magnitude—a notification that can provide seconds or even minutes of warning to facilities and the public. Paired with automated responses that will shut off gas lines before shaking starts, ShakeAlert could help prevent the fires that typically damage cities after a major earthquake.
The late Donald Helmberger, Smits Family Professor of Geophysics, was director of the lab from 1998 until 2003, helping to manage the large volume of data that the lab continued to collect, and also strengthening its relationship with the USGS. An eminent scientist in his own right, Helmberger “discovered more about the deep interior of the Earth in the last half of the 20th century than any other researcher in the world,” Gurnis said in 2020.
“The impact he has had over the years has been enormous,” Heaton added. “It can’t be overstated. He, and the students he worked with, completely transformed seismology.”
In 2003, Jeroen Tromp (now at Princeton University) became director of the Seismo Lab, and he developed, with Gurnis, the lab’s computational capacity, combining its access to vast amounts of data from the SCSN with the computational power needed to translate that data into a model of the crust beneath Southern California. In 2008, Gurnis took the helm of the lab.
Today, the Seismo Lab is a buzzing hive of 80 staff and faculty who are pursuing several avenues of next-generation seismological research. Gurnis sees a few key pillars of the Seismo Lab’s current priorities. Firstly, the lab is focused on dense instrumentation—for example, the use of fiber-optic cables by professor of geophysics Zhongwen Zhan (MS ’08, PhD ’13) and colleagues to detect earthquakes at high resolution. Zhan employs laser emitters that shoot beams of light through the cables, which have tiny imperfections every few meters that reflect back a minuscule portion of the light to the source, allowing them to act as individual seismometers that give researchers the ability to observe the motion of seismic waves as they cause the cables to expand and contract.
“We may be the only regional network trying to integrate that kind of a signal into our traditional dense broadband seismic network through different research projects with faculty,” Gurnis says.
Secondly, Gurnis says, the lab has prioritized the impact of big data and machine learning. For example, work by Hauksson and Zachary Ross, assistant professor of geophysics and William H. Hurt Scholar, uses machine learning to identify tiny earthquakes (between negative magnitude 2.0 (-2.0) and 1.7-magnitude) that are too small to be felt but crucial for understanding how earthquakes work. The efforts of Ross and Hauksson expanded the earthquake catalog for Southern California between 2008 and 2017 by a factor of 10.
A third pillar is the integration of seismology with geodesy, which is the precise measurement of the Earth. The Jet Propulsion Laboratory (JPL), which Caltech manages for NASA, is a leader in geodesy, and its NASA-ISRO Synthetic Aperture Radar (NISAR) mission, scheduled for launch in early 2023, will dramatically improve imaging of earth deformation, Gurnis says. SAR, or synthetic-aperture radar, uses the movement of a radar antenna over an object (usually as the antenna is mounted to plane or spacecraft) to generate two- and three-dimensional images in high resolution.
The proximity of JPL to campus and the robust collaborations already in place positions the Seismo Lab to make significant advances in this field, Gurnis says.
NISAR will launch from India’s Satish Dhawan Space Center in Sriharikota, India, into a near-polar orbit, which will provide near-global coverage. Mark Simons, the John W. and Herberta M. Miles Professor of Geophysics and JPL chief scientist, and colleagues will use radar data gathered from the mission to study multiple Earth processes, including the flow rates of glaciers and ice.
Ice is an important barometer of climate change, and using the geophysical tools that have been honed at the Seismo Lab for the past hundred years, seismologists are adding it to their field of study. “We’re drawing on the same strengths, but applying them to ice,” Gurnis says. “The scientific community is poised to make major advances in understanding how ice caps are moving and collapsing using geophysical methods,” Gurnis says.
Meanwhile, mineral physicists like Jennifer Jackson, the William E. Leonhard Professor of Mineral Physics, are providing granular insight into Earth’s interior—and potentially the interiors of other planets—by decoding the atomic-level properties of materials buried far beyond where they could ever be observed directly. Among the tools Jackson uses are diamond anvils, which allow her to simulate the intense pressures of the Earth’s interior by squeezing mineral samples between a pair of diamonds, and offer direct evidence for how those pressures alter various minerals.
“The future of the lab looks incredibly bright,” Gurnis says. “The combination of the dramatic increase in the fidelity and resolution of geophysical instrumentation with computational pathways using machine learning and robust simulations of natural processes and experiments on the nature of terrestrial materials, positions Caltech well in our journey to understand how the Earth works.”