STRONG GROUND motion occurs whenever energy, conserved as strain in the Earth’s crust or upper mantle, is released instantaneously. In such an event, commonly called an earthquake, energy radiates from the source of release in a series of waves. As waves propagate through rock and other elastic media, they cause periodic displacements (movements up and down or back and forth) that are experienced as shaking. While the vast majority of earthquakes, or seismic events, release too little energy or occur at too great a distance to cause this experience, a small percentage are perceived as shaking. A smaller number still can cause damage, death, and devastation. Energy transferred through the impact of extraterrestrial bodies, such as meteorites, and from detonations of explosive materials can also cause strong ground motion. In most cases, the energy so released is too slight to effect more than a small area.



The crust of the Earth is in constant motion as new rock, formed when magma rises to the surface of the crust, displaces older crustal rock. To complete a cycle, older rock returns to the Earth’s mantle, replenishing sources of magma. One can imagine a perfect geometry for the Earth’s surface that might accommodate this process without strain—it would resemble the mechanics of an escalator. But the real surface of the Earth, which is spheroidal, is far less than ideal. The creation of new rock requires constant readjustment of the spheroidal surface. Taken together, these processes of creating of new rock, the destruction of older rock, and the readjustment of rock on the spheroidal surface are called PLATE TECTONICS.

Crust at the Earth’s surface is not continuous. The crust is composed of a set of plates, each moving relative to adjacent plates. This movement causes strain wherever movement is impeded or opposed. The resulting strain is expressed in crustal rock as deformation. Deformation is expressed in crustal rocks through several characteristic structures. Folds in the crust are one such structure. Faults—surfaces within the crust that accommodate contrary motions of rock on either side of the surface—are another. While movement along some faults can be nearly continuous, the geometries of other faults can result in discontinuous movement and will store strain. Over time, strain in the rock will increase to a point where the competence of the rock is insufficient to store additional strain. At that point, the fault surface will rupture. The subsequent release of strain in such a rupture is the proximate cause of earthquakes.

The geometries of fault surfaces are numerous and complex. Structural geologists recognize three basic fault types. These are the normal fault, the thrust fault, and the strike-slip fault. Normal faults occur where tectonic movement, resulting in vertical displacements of crust, stretches the crust. Thrust faults also displace crust vertically, but along a lower fault angle and as a result of compression rather than tension. In strike-slip faulting, displacements are horizontal. The San Andreas Fault, running from under the Salton Sea to a point north of San Francisco, is a well-known example of a strike-slip fault. All three of these geometries are generalizations; many faults result in both horizontal and vertical displacement. Moreover, earthquakes generally occur within a fault zone, rather than occurring repeatedly along a single fault plane.

When strain is released in the rupture of a fault surface, energy dissipates as mechanical motion expressed as oscillations, or waves. Seismologists, scientists who observe and study ground motion, have found that the waves generated by fault ruptures take several forms, each with distinct properties. Surface waves generate both vertical and horizontal oscillations at the Earth’s surface. Primary (P) and secondary (S) body waves propagate through crust at different velocities. Primary waves propagate through all layers of the earth, but secondary waves do not propagate through fluids and therefore do not pass through the Earth’s core. The different properties of waves provide seismologists with a variety of techniques for pinpointing the sources of waves (at a focal point on the fault plane, or focus, and at a point on the surface above, known as the epicenter), the characteristics of fault displacement, and a means for studying the Earth’s core and mantle.

Because the crustal rock and sediments at the surface of the Earth are not uniform, wave propagation through the crust is not uniform. Surface waves, which cause most of the damage in earthquakes, exhibit different properties when propagated through different thickness and densities of rock; wave motion can be magnified when surface waves pass through unconsolidated sediments, especially sediments with high ground water content. Often, such sediments experience liquefaction, in which the competence of the sediments is severely diminished. The greatest damage in earthquakes usually occurs in structures built on such sediments, as in the Marina District in San Francisco, which was severally damaged in the Loma Prieta earthquake of 1989.


For most organisms, the effects of earthquakes are minimal compared to the effects of other natural catastrophes such as volcanic eruptions, floods, and large storms. Tall trees growing near a fault may be “topped” by side-to-side motion in an earthquake of sufficient intensity, and domesticated animals have been observed to alter their behavior as a precursor to an earthquake. Liquefaction of soils may cause extensive damage to forests. Nevertheless, the effects to most organisms are benign.

The same cannot be said for most modern human populations in cases of earthquakes of moderate to high magnitude. The Modified Mercalli Scale of earthquake intensity, also known in revised form as the European Macroseismic Scale, attests to the effects that very strong ground motion can have for built environments and the assortment of objects that we create and use.

An earthquake at level V in the 12-level scale, for instance, is “felt by nearly everyone; many awakened. Some dishes, windows broken. Unstable objects overturned. Pendulum clocks may stop.” At level XII, damage is “total. Lines of sight and level are distorted. Objects thrown into the air.”

Historically, earthquakes have posed severe challenges for human inhabitations, involving loss of life, destruction of inhabitations and property, and—in some cases—political upheaval. Earthquakes have been recorded throughout human history, and there are records of earthquakes for most great civilizations. The Minoan Civilization on Crete, for example, is believed by many historians to have faltered or collapsed due to earthquakes and tsunamis during the second millennium, B.C.E.

In November 1755, a large earthquake devastated the city of Lisbon, in Portugal, killing 40,000 individuals and destroying most structures. The earthquake was felt throughout the Iberian Peninsula and elsewhere in Europe, and was recorded in a number of contemporary accounts. A fictionalized account of the earthquake serves as a climactic event in François Marie Arouet Voltaire’s literary work, Candide.

Within the contiguous United States, the largest event in recorded history was a cluster of three large earthquakes in 1811 and 1812 centered near New Madrid, MISSOURI. The largest of these, estimated at greater than 8.0 on the Richter scale, occurred in December 1811. The earthquake cluster caused changes in the course of the MISSISSIPPI RIVER and devastated large areas of forest.

Over approximately the past 100 years, large earthquakes have caused extensive damage in San Francisco (1906 and 1989); Anchorage, ALASKA (1964); and in the LOS ANGELES area (1933, 1971, and 1994). In religious cultures, the causes of earthquakes were generally ascribed to acts of the gods, or of God. The roots of western philosophy, however, looked to secular causes; Aristotle explained earthquakes as winds trapped and released from subterranean caves. The Chinese developed a rudimentary (but elegant) seismograph. With the scientific revolution and the subsequent Enlightenment, the search for causes of earthquakes increased. But it was the San Francisco earthquake of 1906 that spurred greater concentrations of scientific interest in earthquakes, resulting in the “elastic rebound” theory, devised by H.F. Reid in a report on that earthquake.

In the 1920s, the newly founded California Institute of Technology and its Seismology Laboratory, with a network of seismographs created to monitor and study earthquakes along California’s San Andreas Fault and ancillary faults, became a magnet for seismologists. There, Charles Richter and Beno Gutenberg devised what has become known as the Richter scale, a logarithmic scale of wave motion amplitudes that yields a measure of the magnitude of an earthquake. Also working in the Seismo Lab, Hugo Benioff developed an interest in studying deep focus earthquakes. Results of his studies were important for the later development of plate tectonics; Benioff’s observed patterns of deep-focus earthquakes were later explained in plate tectonics as part of the subduction process.

In the aftermath of the 1964 Anchorage earthquake, policy makers became interested in seismologists’ hopes that earthquakes might be predictable. This interest may have been related to Cold War politics; a science and technology for predicting earthquakes could lead to evacuations and might save many lives, demonstrating the capacity in the United States for surviving a nuclear attack of comparable destructive force. By the 1980s, primary interest in policy shifted from prediction to preparation and mitigation: engineering better buildings and infrastructure, and strengthening building codes.

Interest in earthquake prediction has not abated. In the 1980s, the U.S. Geological Survey created an extensive system of monitoring devices in Parkfield, California, a site of generally dependable movements along the San Andreas Fault. Data collected in the September 28, 2004, earthquake in this location is presently being analyzed for evidence of precursor activities that may someday permit reliable predictions.