Paleoseismology

Paleoseismology is the study of the evidence of past earthquakes. By studying the physical features of previous earthquakes, scientists gain knowledge of the forces operating within the Earth, knowledge that may be used to develop a means of predicting future large earthquakes.

Ancient Earthquakes

Seismology is the science that deals with earthquakes and movement within the Earth. Seismologists and geologists use historical and prehistoric information about earthquakes to better calculate the probability of earthquakes occurring in a particular area. However, historical evidence from earthquakes—information acquired from observation and instrumental measurements—is short and incomplete. Paleoseismology is a relatively new branch of earthquake science that uses archaeological techniques to find physical evidence of large ancient earthquakes and thus expand the Earth's seismic history.

In the 1960s, the theory of plate tectonics was introduced. According to this theory, the Earth's outer crust comprises sections or plates in constant motion. When these plates collide or spread apart, the resulting forces produce deformations, or tectonics, of the Earth's surface identified with earthquakes. When plates collide, and one plate moves beneath the other, the resulting fault is called a “subduction zone.” Subduction zones are located underwater and cause changes in land at the shoreline. Earthquakes at the subduction zones often cause tsunamis that devastate coastal areas. When the plates rub against one another, the resulting fault is known as a “plate interaction.” The San Andreas fault in California is a well-known example of a plate-interaction fault.

The idea of elastic rebound is used to explain how earthquakes originate. Under normal circumstances, the rocks at a fault boundary move a few centimeters per year. If the rock cannot move, the force of the normal movement becomes stored as “strain energy.” When the force becomes too great, the fault slips to release energy. As the plate boundaries shift or rupture, earthquakes occur. By studying the historical record, scientists have found that the plate boundaries rupture in segments until the entire section is broken, and the cycle begins again. Scientists believe future earthquakes are more likely to occur where the segments have not recently ruptured. These places are called “seismic gaps.”

With the discovery of plate tectonics, scientists believed that long-term and short-term prediction of earthquakes would be possible. Long-term prediction is based on the probability that an earthquake will occur. Paleoseismic evidence of past earthquake recurrence periods allows scientists to determine when a large earthquake is “overdue” for a particular area.

Short-term prediction involves specifying the future place, magnitude, and time of the earthquake. Short-term prediction relies on data collected from monitoring equipment on known faults as well as precursory signals such as an increase in fault movement, change in deep-well water levels, changes in surface elevations, an increase in minor earthquake activity, the occurrence of a moderate foreshock, or an increase in the release of radioactive radon gas. Since other environmental factors can cause some of these signals, there are no definitive predictors of an imminent earthquake.

The advantages of knowing when, where, and how strong an earthquake will be are numerous. The cost of an earthquake in terms of loss of human life and property can be tremendous. In addition to the damage caused by the actual ground shaking from the earthquake, there may be loss of life and property caused by structural collapse and secondary effects such as landslides, fires, and tsunamis. The September 19, 1985, earthquake that struck Mexico City killed more than eight thousand people, injured thirty thousand people, destroyed or severely damaged more than one thousand multistoried buildings, and caused an estimated $5 billion in damage.

Study of Earthquakes

Scientists use several methods to study earthquakes. One of the methods is based on recorded observations. The earliest known scientifically collected observational data about earthquakes came from a survey taken after a strong earthquake, known as the All Saints Day earthquake, struck Lisbon, Portugal, on November 1, 1755. Local officials sent a list of questions to the affected areas asking about the duration of the shock, the times and intensities of aftershocks, the number of people killed, the structures destroyed, and the extent of damage from related fires. Firsthand accounts of death and structural damage, as well as obvious changes in the physical features of the land, allow scientists to judge the severity of an earthquake. This information is useful in identifying seismic gaps. Probability statistics indicate that an area is more likely to have a large earthquake if a large earthquake has occurred there before.

Another method for studying earthquakes is through instrumentation. Tiltmeters and strainmeters are used to measure changes in the crustal movement in earthquake areas. Hidden faults are located using a device that relies on how sound waves are reflected. Satellites, aerial photographs, and surveying equipment also provide clues to the location of underground faults. Seismic activity is monitored worldwide with seismographs, which record the occurrence, direction, severity, and duration of earthquakes. Modern seismographs are extremely sensitive and can detect ground motions of very small earthquakes. Scientists estimate that approximately eighty thousand small tremors and fifteen to twenty severe earthquakes occur annually throughout the world.

In the late 1970s, Kerry Sieh, a geologist at the California Institute of Technology, pioneered paleoseismology at Pallet Creek on the San Andreas fault. Scientists could now gather additional information about earthquakes to supplement and expand the existing observational and instrumentation knowledge. With the prehistoric evidence of earthquakes, scientists can better calculate the probability of another large earthquake.

Paleoseismic techniques may involve digging a trench across a fault and examining the rock strata, or layers, for evidence of previous ruptures. Scientists also look for evidence of flooded forests or sand sheets along coastlines, indicating tsunami activity. The evidence can be direct, such as fault offsets, or indirect, such as landslides or sand deposits, which occur with strong ground shaking. Scientists can determine how much the fault has slipped by measuring the distance separating the two rock segments in a fault offset. The samples can be radiocarbon-dated to determine when the earthquakes occurred. When the time gap between the earthquakes is determined, the recurrence period and average interval can be established. The recurrence period is a range of time between earthquakes; the interval is the average time between earthquakes. For example, if earthquakes happened in a particular region in 1700, 1775, 1825, and 1925, the recurrence range is fifty to one hundred years, with an average interval of seventy-five years.

Paleoseismic Data

Seismologists need information about past earthquakes to study and prepare for future earthquakes. Because of the short historical record, paleoseismology has been very useful in North America. Additionally, the techniques have helped establish the accumulation of stress in faults worldwide.

Paleoseismic data enable scientists to determine higher earthquake hazard levels in a geographic area. Engineers and public officials use the information from scientists to revise building codes and redraw maps of earthquake-shaking hazard zones. The US Uniform Building Code, which includes nationwide design standards, defines six earthquake-shaking hazard levels. By convincing people to reinforce existing structures and adopt earthquake-resistant construction standards, loss of life and property can be greatly reduced.

In California, a 1971 earthquake in the San Fernando Valley killed sixty-five people and caused an estimated $500 million in damages. California has had earthquake-resistant building criteria since 1933, updated several times. In contrast, a 1972 earthquake of similar magnitude in Managua, Nicaragua, killed more than five thousand people and caused economic losses equal to the country's gross national product for that year. Managua had few buildings designed under earthquake-resistant standards.

Paleoseismology helps answer questions about faults, folds, secondary effects of prehistorical earthquakes, length of an earthquake cycle, the cycle's regularity or irregularity, and similarity of characteristics of an area's earthquakes. Scientists believe the past earthquake processes will probably be the same as those in the near future. The longer the record of earthquake occurrences in a particular area, the greater the predictive value of the knowledge. Predicting future earthquakes' likelihood and possible effects enables people to take precautions to protect themselves against injury, death, and economic losses.

Study of Shorelines

Paleoseismology initially studied active faults to help establish the rates of movement. The technique was then applied to interpreting earthquake crustal deformations or shaking effects. Although paleoseismology is typically used to examine surface faults, the technique is also used to examine subduction-zone earthquakes indirectly. Since subduction zones are underwater faults, scientists examine changes in the land at the shoreline for evidence of past activity. The sudden lowering of a shoreline or a series of terraced beaches may be associated with major earthquakes.

In Alaska, scientists examined terraced beach lines near an island in Prince William Sound. This area was the site of the March 27, 1964, “Good Friday” earthquake, the second-strongest earthquake recorded. The earthquake killed 114 people and caused $350 million in property damage. The coastlines of nearby landmasses bowed by as much as 2.5 meters, flooding them with seawater and killing many square kilometers of forest. Scientists discovered that large earthquakes have occurred in the area an average of once every 850 years during the past 5,000 years. Because of this investigation, scientists believe it is unlikely that a major earthquake will occur in this area soon.

Until the mid-1980s, scientists believed that the area off the coast of Washington, Oregon, and Northern California had a low probability of a massive earthquake since there was no historical record of such an earthquake. Paleoseismic studies, however, show evidence of a sudden lowering of the shoreline associated with major earthquakes. In addition, scientists found sheets of sand deposited by tsunamis and sand-filled cracks caused by strong earthquake shaking. Researchers have thus concluded that three massive earthquakes have struck this area in the past three thousand years; the most recent great earthquake struck about three hundred years ago. Based on this knowledge, measures have been taken to reinforce existing structures and toughen earthquake design standards. Without the paleoseismic data, a powerful earthquake would have taken the area by complete surprise.

Scientists have been gathering much information about faults at plate boundaries since an estimated 95 percent of all earthquakes occur at these boundaries. The probability of occurrence for some earthquakes is more challenging to predict because they do not occur near plate boundaries. Often, these inner-plate earthquakes are devastating because they occur so infrequently that there may be no historical record of an earthquake in that area.

On September 30, 1993, ten thousand people in Killari, India, died when an earthquake devastated the area. Before the earthquake, the area was included in the lowest level of the earthquake-hazard map of India. The seismologists were wrong at Killari because the hazard map relied mostly on the historical record of earthquakes. In India, this record is only about 150 years old. Investigation following the earthquake suggested that the area had been seismically quiet for 65 million years or more. Scientists believe that the earthquake may have been triggered by an artificial, or human-constructed, reservoir that changed the stress on a fault. Understanding how such construction affects the stress on the Earth's crust that triggers earthquakes allows more considered and intelligent decisions about building nuclear power plants, dams, reservoirs, and cities.

In the United States, the area east of the Rocky Mountains has faults that are buried deep and accumulate stress slowly. Earthquake risk estimates for the eastern United States are not as advanced as those for the West, where the faults break the surface, and earthquakes occur more often. Eastern faults may be quiet for hundreds, thousands, or millions of years.

However, one of the most destructive earthquakes in American history occurred in 1811 and 1812 in New Madrid, Missouri. At least one earthquake per day took place in the area for over a year, with three extremely powerful ones; these quakes were felt over approximately 2.6 million square kilometers and caused tremendous physical changes to the land. Destruction was severe, but casualties and economic losses were low because there were few settlements in the affected area. The earthquakes triggered landslides, submerged islands, opened fissures in the ground, flooded lowlands, and caused land areas to rise or sink.

Large earthquakes have also occurred in Charleston, South Carolina, in 1886; Newfoundland, Canada, in 1929; and New York State in 1988. Numerous small earthquakes have occurred over the past two hundred years, with at least twenty damaging earthquakes in the central Mississippi Valley alone. Eastern earthquakes are puzzling because they do not generate visible surface faults. Because of this lack of physical evidence and the infrequent occurrences of eastern earthquakes, determining the probability and location of another large earthquake in the region is very difficult.

In the twenty-first century, paleoseismology advances, offering valuable information about past earthquakes and informing understanding of the potential for future earthquake events. Improved mapping and imaging techniques, such as high-resolution digital elevation models and advanced remote sensing methods, allow scientists to create maps of fault zones. The accuracy of dating ancient earthquake events has been enhanced by novel geochronological dating methods, including 14C geochronology, uranium-series dating, luminescence geochronology, and soil chronosequences. Paleoseismology is an advancing scientific field incorporating multiple disciplines that have allowed scientists to garner information about past earthquakes and the effects of ones yet to occur. 

Principal Terms

earthquake: a shaking or trembling of the Earth caused by a break or rupture in the rocks of the Earth's outer crust

elastic rebound: the process of the buildup and release of geologic stress; the release of the strain in the rocks results in earthquakes

fault: a fracture of the Earth's crust along which rocks move

fault offset: a feature such as a road, creek bed, or tree line across a fault that separates and becomes misaligned by the fault movement

fault scarps: a steep cliff or slope created by movement along a fault

plate tectonics: the theory that the Earth's hard outer crust (lithosphere) is composed of sections, or plates, that are in constant motion

recurrence period: the range of time between successive earthquakes

seismic gap: a fault region known to have had previous earthquakes but not within the area's most recent recurrence period

tsunami: a large ocean wave caused by an earthquake in the ocean floor

Bibliography

Adams, John. “Paleoseismology: A Search for Ancient Earthquakes in Puget Sound.” Science, vol. 258, 4 Dec. 1992, pp. 1592-1593.

Best, David M., and David B. Hacker. Earth's Natural Hazards. New York: Kendall/Hunt, 2010.

Bolt, Bruce A. Earthquakes and Geological Discovery. New York: Scientific American Library, 1993.

Bull, William B. Tectonic Geomorphology of Mountains: A New Approach to Paleoseismology. Malden, Mass.: Wiley-Blackwell, 2007.

Chester, Roy. Furnace of Creation, Cradle of Destruction. New York: AMACOM Books, 2008.

Guidoboni, Emanuela, and John E. Ebel. Earthquakes and Tsunamis in the Past: A Guide to Techniques in Historical Seismology. New York: Cambridge University Press, 2009.

Lay, Thorne, and Terry C. Wallace. Modern Global Seismology. New York: Academic Press, 1995.

Main, Ian. “Long Odds on Prediction.” Nature, vol. 385, 2 Jan. 1997, pp. 19-20.

McCalpin, James P., editor. Paleoseismology. 2d ed., New York: Academic Press, 2009.

Moseley, Charles. “Slow Progress in Earthquake Prediction.” Editorial Research Reports, vol. 2, 15 July 1988, pp. 354-363.

Rizza, Magali, et al. "Tectonic Geomorphology and Paleoseismology of the Sharkhai Fault: A New Source of Seismic Hazard for Ulaanbaatar (Mongolia)." Solid Earth, vol. 13, no. 3, 2022, pp. 761-777, doi.org/10.5194/se-13-761-2022. Accessed 1 Aug. 2024.

Sintubin, Manuel, et al., editors. Ancient Earthquakes. Boulder, Colo.: Geological Society of America, 2010.

Worsnop, Richard L. “Earthquake Research.” CQ Researcher, vol. 4, 16 Dec. 1994, pp. 1105-1128.

Yeats, Robert S., Kerry Sieh, and Clarence R. Allen. The Geology of Earthquakes. New York: Oxford University Press, 1997.

Zierler, David. “Kerry Sieh, Paleoseismologist - Heritage Project.” Caltech Heritage Project, 26 Sept. 2022, heritageproject.caltech.edu/interviews-updates/kerry-sieh. Accessed 1 Aug. 2024.