In Memorial
Jerry P.
Eaton
Jerry P. Eaton, the
legendary pioneer of telemetered seismic networks for monitoring volcanoes and
active earthquake faults, died April
2, 2004, after a long battle with cancer. His friends and colleagues
knew Jerry as a dedicated scientist and resourceful inventor who was
unfailingly generous in providing help and encouragement to younger scientists.
Jerry's impact on
seismology, volcanology, and the programs of the U.S. Geological Survey (USGS)
was enormous. More than anyone else, it
was his pioneering vision of instrumental networks designed to observe active
processes in the near field that advanced microearthquake seismology to the
forefront in studies of earthquake tectonics and volcanology.
Through his reading knowledge of French,
German, Russian, and Spanish, he was familiar with seismological research
published in these languages as well as that published in English. His
knowledge of early seismology papers was encyclopedic, and he could recall
details of observations he had made a half-century ago. Just a week before his
death, he was at his workstation in his office at the USGS, Menlo Park, timing earthquakes, as
he had done virtually every day since his retirement in 1995.
Background
and Education
Jerry was born on December
11, 1926, on a Central Valley farm, near Fresno, California. When Jerry was about
five years old, his father responded to the bad economic times by building a
well rig from salvaged materials and moving it and his family to Woodland, near Sacramento, where he eked out a
living drilling water wells. Jerry’s
mother, a former school teacher, was the bookkeeper for the business. For several years she grew crops of tomatoes
to augment the family income. On
occasion, Jerry would tell his colleagues of his early life in the farming
area, a time when he and his brothers learned to be self-reliant in solving
problems as they arose.
Jerry attended the University of California, Berkeley, earning a
B.A. with honors in physics (1949) and a Ph.D. in geophysics (1953). An early
indication of his ability and resourcefulness was that he completed his Ph.D.
thesis while his professor, Perry Byerly, was on sabbatical leave. At Berkeley, Jerry met his wife,
Nancy, a fellow student studying zoology. Nancy remarked later that, “We
were both kind of shy people, and we just fit perfectly”. They have four children, Marian, Jeffrey,
Dana and Carol.
At the
Hawaiian Volcano Observatory (HVO)
Immediately after
receiving his Ph.D., Jerry accepted a position with the USGS and was stationed
at HVO on the Island of Hawaii where he spent the next
8 years developing the modern science of volcano monitoring. Jerry’s first challenge
at HVO was the modernization of the seismic network. He designed and installed a telemetered,
electronic seismic network that significantly improved earthquake detection and location accuracy
compared to the previous methods of analyzing seismic data from mechanical
seismographs. Like T. Asada in Japan and P. G. Gane in South Africa before him, Jerry
realized that to capture the numerous, smaller earthquakes, a dense array of
seismometers of high magnification was required. To establish a common timing
reference for individual seismometers in the 1950’s, signals from these
seismometers in the field had to be telemetered to a central recording site.
At this time seismic
equipment was quite costly, so Jerry and a machinist at HVO built their own
seismometers, amplifiers, recorders, and timing systems. Jerry was able to
raise the peak magnification to 40,000 at 5 Hz, an order of magnitude higher
than other recording systems available at that time. Because radio telemetry
was still in its infancy, Jerry solved the telemetry problem by laying out
miles of wires, which he obtained free as military surplus. Because these wires had been previously used,
relatively short lengths of them first had to be stretched out and spliced
together. Jerry then laid them out by hand in rough volcanic terrain to relay
seismic signals from the seismometers in the field to the headquarters of
HVO.
Jerry developed methods
for calibrating the seismometers (Eaton, 1957; Eaton and Byerly, 1957) and
locating the earthquakes, which sometimes occur at Kilauea and Mauna Loa by the hundreds per
day. He soon defined the primary regions
where earthquakes occur throughout the Island of Hawaii, outlined the basic
plumbing system of Kilauea volcano, and based on traveltimes of
seismic waves, showed that the crust bends downward under the island apparently
due to the load of the volcanic pile. “The
high-gain vertical component network virtually revolutionized our understanding
of the seismicity of Kilauea volcano, particularly of its summit region. The number of earthquakes recorded increased
nearly 100 times, and it became possible to resolve them into ‘families’ on the
basis of epicenter, depth, temporal pattern of occurrence, etc., and to
associate them with the primary structures of the volcano: summit, rift zones,
south flank, Kaoiki fault zone, etc.”
(Eaton, 1996).
In addition, Jerry
recognized the importance of ongoing ground surface deformation associated with
the slow inflation of Kilauea Volcano's summit region between eruptions and its
rapid deflation during flank eruptions.
To document these inflation/deflation episodes accurately, he designed,
built, and deployed a network of water-tube tiltmeters around the summit of Kilauea that are sensitive to
changes of 0.3 microradians. To ensure the uniformity of environmental
conditions under which the measurements were taken, Jerry and his colleagues
took their readings late at night, often in the pouring rain.
In 1960, Eaton and Murata
published a classic paper describing how Kilauea volcano works: magma
from 50 to 60 km depth is fed to a shallow magma reservoir. When the magma pressure in this reservoir
exceeds the strength of the volcano, a fracture breaks upward to feed a summit
intrusion or eruption, or sideways to feed a flank intrusion or eruption. This conclusion was based on eight years of
study on earthquakes, deformation, and the eruptive behavior of the volcano,
including close monitoring of a major eruption in 1959. The model presented in
Eaton and Murata (1960) continues to play a central role in the study and
understanding of volcanism in Hawaii. Jerry set a standard
for volcano monitoring that has been followed and refined by volcanologists all
over the world. He also became a
skillful photographer. The films and
slides taken by Jerry and his HVO colleagues to document the course of the
1954-55 and 1959-60 Kilauea eruptions are still among the best shown
today.
In addition to his volcano research, he also addressed the tsunami
hazard in Hawaii. Long-period seismographs were installed, and coupled with seismograms
from the short-period network, Jerry was able to detect seismic signals that
indicated the occurrence of major earthquakes in the Pacific rim that were
capable of generating damaging tsunamis in Hawaii. “My experience
in Hawaii left me in awe not only of the irresistible destructive power of its
volcanoes but also that of even distant earthquakes. A small group of us from HVO witnessed the
destruction of a large part of Hilo, Hawaii, by the tsunami generated by the great 1960 Chilean earthquake (Eaton,
Richter, and Ault, 1961). We were particularly
distressed by the deaths of more than 60 people in Hilo. Timely interpretation of
widely available seismic and wave height data and implementation of simple but
effective protective measures could have saved them, but such actions were not
carried out because of lack of adequate planning and organization for such an
emergency” (Eaton, 1996).
Although he left Hawaii in 1961, Jerry
maintained an active interest in volcanology throughout his career. In 1987, he and colleagues Don Richter and
Harold Krivoy published a new paper on the cycling of magma between Kilauea’s summit reservoir and
the Kilauea Iki lava lake based on data they had collected during the 1959 Kilauea eruption. It was published in a collection that marked
the 75th anniversary of the founding of the Hawaii Volcano
Observatory (Eaton, Richter and Krivoy, 1987).
At USGS,
Denver
In 1961, Jerry Eaton
joined the USGS Crustal Studies group in Denver, Colorado, led by Lou Pakiser.
There he contributed to the development of new, long-range, seismic refraction
instrumentation and took the lead in running a seismic refraction profile from
the Bay Area across the Sierra Nevada into central Nevada. His analysis of that
profile provided clear evidence for the existence of an asymmetrical crustal
root that extends to at least 50 km beneath the high crest of the range (Eaton,
1963). Jerry’s stay in Denver was brief, however, as
he was soon attracted by another great challenge.
At USGS,
Menlo Park
In 1965, following the
Great Alaskan earthquake of 1964, Jerry moved to Menlo Park, California, where
he became a key player in the group led by Lou Pakiser that laid the groundwork
for establishing the USGS's National Center for Earthquake Research (NCER) and
the National Earthquake Hazards Reduction Program (NEHRP). Jerry's greatest
contributions stem from his conviction that measurements of signals from the
Earth's crust are best done with many instruments of acceptable quality, rather
than a few instruments of outstanding quality as are typically used to study
the Earth's interior. He continually promoted the development of better and
cheaper seismic instruments, always mindful of the scientific objectives and of
the need for careful calibration in order to advance the understanding of earth
processes. The importance of dense networks of instruments is widely understood
today, but Jerry was the first person both to see their importance and to
deploy them within a limited budget. In this way, he set the foundation in the
1960s for subsequent evolution of techniques and instrumentation for studies of
local seismicity, regional strain, and seismic refraction applied to studies of
crustal structure.
Jerry's classic study of the
aftershocks of the Parkfield earthquake of 1966 was a startling demonstration
of the value of dense networks. Applying his HYPOLAYER earthquake location
program (Eaton, 1969) to the Parkfield data, he demonstrated the power of
high-precision studies of microearthquake locations, and established
definitively that earthquakes occur on faults. By precisely locating the
aftershocks, he showed that the slippage had occurred along a 30-km segment of
the near-vertical fault to a depth of 12 km (Eaton, O’Neill and Murdock,
1970). This study led the way to
identifying seismic gaps along strike-slip faults and other key concepts at the
center of studies of earthquake hazards today.
His study of the 1983 Coalinga earthquake demonstrated the existence of
a shallow, blind thrust-fault within the predominantly strike-slip environment
near the San Andreas Fault (Eaton, 1990; Eaton and Rymer, 1990).
Realtime Seismology
In the early days of
NCER, the research staff ate lunch together every Friday. One day in 1968, Lou
Pakiser, head of NCER, read a memo from an assistant to the President of the United
States to the Director of the USGS inquiring how
long it would take to detect and locate an earthquake. Rapid detection of
nuclear explosions was then a national issue because China became the fifth country
possessing nuclear bombs in 1964. Jerry suggested that the reply should be one
minute, although at that time, a routine earthquake location took more than one
hour. A few weeks later, the same
Presidential assistant ordered that this capability be demonstrated. At that time most of the staff favored an
analog electronic circuit approach for the demonstration, using the signals
telemetered from the local seismic clusters that had just been installed. Willie Lee proposed that it might also be
possible to use a computer simulation to produce the demonstration. Pakiser cautiously decided to pursue both
approaches. He instructed Willie to
visit Jerry at the Stanford University Hospital, where he was recovering
from a back operation.
Although confined to his
bed and in obvious pain, Jerry outlined a scheme for doing realtime earthquake
detection and location. Using the IBM 360-67 mainframe
computer at Stanford University, Willie quickly wrote a
computer program demonstrating that Jerry’s scheme worked and that an
earthquake could indeed be detected and located in about 30 seconds after its
occurrence inside a local seismic network.
Soon thereafter, Sam Stewart joined in the effort. The three of them wrote an unpublished report
to the Defense Advanced Research Projects Agency and obtained funding to implement
a realtime seismic system for local networks using a CDC 1700 mini-computer.
It would be hard to find
three more different personalities than Eaton, Lee, and Stewart, but they
managed to work together in 1968-69.
Jerry’s vision and optimism, Willie’s experience from having worked in a
computer center, and Sam’s meticulousness and perseverance all contributed to
the success of the project that ushered in the era of realtime seismology
(Stewart, Lee and Eaton, 1971).
Jerry Eaton’s Calnet
Jerry frequently stressed
the importance of observing Earth processes at their noise level, of making
continuous recordings, and of systematically cataloging earthquakes and
volcanic phenomena. His goal for NCER
was not just to study choice parts of the San Andreas Fault, but rather to monitor
the system as a whole (Eaton, Lee and Pakiser, 1970). To do so, he not only had to create a modern,
regional seismic network (with Wayne Jackson, Willie Lee, John Roller, Sam
Stewart, Jack Van Schaack and many others) by using telephone and radio
telemetry to carry signals from many remote stations to a central site for
recording and analysis, but also had to find the funds to underwrite it. Beginning with the first two clusters of
about eight stations each in 1967, Calnet grew outward from its initial south
Bay and Hollister regions to cover all of California by 1986 (Hill, Eaton and
Jones, 1990). As it grew, Jerry made
many largely unsung contributions to the high instrumentation standards that
the network maintained. Jerry also had
the foresight and patience to back the development of the automatic seismic
processing systems that we sometimes take for granted today. Jerry summarized his vision and contributions
to local seismic networks in his memoir (Eaton, 1996).
Jerry's many scientific
contributions include studies of numerous important California earthquakes including Kern County, Parkfield, Coalinga, Morgan Hill, and Eureka, to name some notable
studies. In all of these papers his goal
was to use seismicity to elucidate tectonic processes by revealing the
structural elements, crustal structure and stress fields. One of Jerry’s
contributions (Eaton, 1992) was his refinement of duration and amplitude
magnitude scales for short-period stations operating in northern California. This development provides the magnitude of
reference for tens of thousands of earthquake smaller than Mw 3.5 that occur
annually, and allows seismologists to investigate the statistical properties of
earthquake occurrence.
Jerry also contributed greatly to the careers of a
generation of young scientists who worked at the USGS in Menlo
Park. Whether
they were his postdocs, new staff, students or short-term visitors, he was ever
eager to hear their ideas and encourage them.
He also helped set the standard for sharing of seismic data, which today
we take for granted.
A Career of Leadership
Besides his extraordinary
personal contributions to understanding the structure and dynamics of the
Earth's crust, Jerry also held important leadership positions in which he
coordinated the efforts of many investigators.
He was the Scientist in Charge of the Hawaiian Volcano Observatory
(1955-58; 1960; 1961), Chief of the Office of Earthquake Research and Crustal
Studies (1971-73), Chief of the Branch of Seismology (1973-75), and Acting
Chief of Branch of Network Operations (1978-82). He played a major role over many years in
defining and defending the National Earthquake Hazards Reduction Program and
its predecessors.
Jerry was a Fellow of the
American Geophysical Union, was President of the Seismological Society of
America (1966), was a member of the California Governor's Earthquake Prediction
Evaluation Committee (1972-92), and was Chairman of the Earthquake Prediction
Commission of the International Association of Seismology and Physics of the
Earth's Interior (1975-79).
Jerry Eaton was truly a
pioneer, for he went not where the path leads, but where there was no path. He
left a trail for others to follow and encouraged many colleagues to excel. He was a dedicated scientist, a wonderful
colleague, and he leaves a rich legacy for future generations of
seismologists.
Acknowledgements
We thank Bob Decker,
Nancy Eaton, Marian Eaton, Donna Eberhart-Phillips, Jack Healy, Dave Hill, Fred
Klein, Bob Koyanagi, John Lahr, Dave Oppenheimer, Jack Van Schaack, Bob
Wallace, Pete Ward, and Rob Wesson for their comments and suggestions on the
manuscript.
References
Eaton, J. P.
(1957). Theory of the electromagnetic seismograph, Bull. Seism. Soc. Am., 47,
37-75.
Eaton, J. P.,
and P. Byerly (1957). Calibration of the short-period Sprengnether seismograph,
Bull. Seism. Soc. Am., 47, 155-166.
Eaton, J. P.,
and K. J. Murata (1960). How volcanoes grow, Science, 132, 925-938.
Eaton,
J. P., D. H. Richter, and W. U. Ault
(1961). The tsunami of May 23, 1960, on
the Island of Hawaii, Bull. Seism. Soc. Am., 51, 135-157.
Eaton, J. P.
(1963). Crustal structure from San
Francisco, California, to Eureka, Nevada, from seismic-refraction
measurements, J. Geophys. Res., 68, 5789-5860.
Eaton, J. P.
(1969). HYPOLAYR, a computer program for determination of hypocenters of local
earthquakes in an earth consisting of uniform flat layers over a half space, U.S. Geol. Surv. Open-file Report, 155 pp.
Eaton, J. P.,
W. H. K. Lee, and L. C. Pakiser (1970). Use of microearthquakes in the study of
the mechanics of earthquake generation along the San Andreas fault in central
California, Tectonophys., 9, 259-282.
Eaton, J. P.,
M. E. O'Neill, and J. N. Murdock (1970). Aftershocks of the 1966 Parkfield-Cholame, California earthquake: A detailed
study, Bull. Seism. Soc. Am., 60, 1151-1197.
Eaton, J. P.,
D. H. Richter, and H. L. Krivoy (1987).
Cycling of magma between the summit reservoir and Kilauea Iki lava lake
during the 1959 eruption of Kilauea volcano. In Volcanism
in Hawaii, Vol II, R. W. Decker, T. L. Wright, and P. H. Stauffer,
eds., U.S. Geological Survey Paper 1350, 1307-1335.
Eaton, J. P.
(1990). The earthquake and its aftershocks from May 2 through September 30, 1983. In The Coalinga, California, earthquake of May 2, 1983, M. J. Rymer and W. L.
Ellsworth, eds., U.S. Geological Survey Professional Paper 1487, 113-170.
Eaton, J. P.,
and M. J. Rymer (1990). Regional seismotectonic model for the southern Coast Ranges, in the Coalinga, California, earthquake of May 2, 1983. In The Coalinga, California, earthquake of May 2, 1983, M. J. Rymer and W. L.
Ellsworth, eds., U.S. Geological Survey Professional Paper 1487, 97-111.
Eaton, J. P.
(1992). Determination of amplitude and duration magnitudes and site residuals
from short period seismographs in northern California, Bull. Seism. Soc. Am., 82,
533-579.
Eaton, J. P.
(1996). Microearthquake seismology studies in USGS volcano and earthquake
hazards studies: 1953-1995, USGS
Open-File Report 96-54, 144 pp.
Hill, D. P.,
J. P. Eaton, and L. M. Jones (1990). Seismicity, 1980-1986. In The San Andreas Fault System, California, R.
E. Wallace, ed., U.S. Geological Survey
Professional Paper 1515, 115-151.
Stewart, S.
W., W. H. K. Lee, and J. P. Eaton (1971). Location and real-time detection of
microearthquakes along the San Andreas fault system in central California, Bull. Royal Soc. New Zealand, 9, 205-209.
William H. K. Lee and William L.
Ellsworth
U.S. Geological
Survey, MS 977
Menlo
Park, CA 94025