An earthquake is ground shaking caused by the sudden and rapid movement of one block of rock slipping past another along fractures in Earth’s crust, called faults. Most faults are locked except for brief, abrupt movements when sudden slippage produces an earthquake. Faults are locked because the confining pressure exerted by the overlying crust is enormous, causing these fractures in the crust to be “squeezed shut.”

Earthquakes tend to occur along preexisting faults where internal stresses have caused the crustal rocks to rupture or break into two or more units. The location where slippage begins is called the hypocenter, or focus. The point on Earth’s surface directly above the hypocenter is called the epicenter (Figure 1). Although the hypocenter and epicenter are displayed as points in the figure, the slippage on a fault occurs over an area, not a discrete point.

Large earthquakes release huge amounts of stored-up energy as seismic waves—a form of energy that travels through the lithosphere and Earth’s interior. The energy carried by these waves causes the material that transmits them to shake. Seismic waves are analogous to waves produced when a stone is dropped into a calm pond. Just as the impact of the stone creates a pattern of circular waves, an earthquake generates waves that radiate outward in all directions from the zone of slippage (refer to Figure 1). Although seismic energy dissipates rapidly as it moves away from the source of an earthquake, sensitive instruments can detect earthquakes even when they occur on the opposite side of Earth.

Thousands of earthquakes occur around the world every day. Fortunately, most are small enough that people cannot detect them. Only about 15 strong earthquakes (M7 or greater) are recorded each year, many of them occurring in remote regions. The occasional large earthquakes that are triggered near major population centers are among the most destructive natural events on Earth. The shaking of the ground, coupled with the liquefaction of soils, wreaks havoc on buildings, roadways, and other structures. In addition, a quake occurring in a populated area can rupture power and gas lines, causing numerous fires. In the famous 1906 San Francisco earthquake, much of the damage was caused by fires that became uncontrollable when broken water mains left firefighters with only trickles of water (Figure 2).

Discovering the Causes of Earthquakes 

The energy released by volcanic eruptions, massive landslides, and meteorite impacts can generate earthquake-like waves, but these events are usually weak. What mechanism produces a destructive earthquake? As you have learned, Earth is not a static planet. Because fossils of marine organisms have been discovered thousands of meters above sea level, we know that large sections of Earth’s crust have been thrust upward. Other regions, such as California’s Death Valley, exhibit evidence of extensive subsidence. In addition to these vertical displacements, offsets in fences, roads, and other structures indicate that horizontal movements between blocks of Earth’s crust are also common (Figure 3).

The actual mechanism of earthquake generation eluded geologists until H. F. Reid conducted a landmark study following the 1906 San Francisco earthquake. This earthquake was accompanied by horizontal surface displacements of several meters along the northern portion of the San Andreas Fault. Field studies determined that during this single earthquake, the Pacific plate lurched as much as 9.7 meters (32 feet) northward, past the adjacent North American plate. To better visualize this, imagine standing on one side of the fault and watching a person on the other side suddenly slide horizontally about 10 meters to your right.

What Reid concluded from his investigations is illustrated in Figure 4. Over tens to hundreds of years, differential stress slowly bends the crustal rocks on both sides of a fault. This is much like a person bending a limber wooden stick, as shown in Figure 4A,B. Frictional resistance keeps the fault from rupturing and slipping. (Friction inhibits slippage and is enhanced by irregularities that occur along the fault surface.) At some point, the stress along the fault overcomes this frictional resistance, and slip initiates. Slippage allows the deformed (bent) rock to “snap back” to its original, stress-free shape; a series of earthquake waves radiate outward as rock slides (Figure 4C,D). Reid termed this “springing back” elastic rebound because the rock behaves elastically, much as a stretched rubber band does when it is released.

Foreshocks and Aftershocks

Strong earthquakes are often followed by numerous earthquakes of lesser magnitude, called aftershocks, which result from crust along the fault surface adjusting to the displacement caused by the main shock. Aftershocks gradually diminish in frequency and intensity over a period of several months or years following an earthquake. In a little more than a month following the M 7.0 earthquake that devastated Haiti in 2010, the U.S. Geological Survey detected nearly 60 aftershocks with magnitudes of 4.5 or greater. The two largest aftershocks had magnitudes of 6.0 and 5.9, both large enough to inflict further damage. Hundreds of minor tremors were also felt.

Although aftershocks are weaker than the main earthquake, they often trigger the destruction of already weakened structures. For example, in northwestern Armenia in 1988, where many people lived in large apartment buildings constructed of brick and concrete slabs, a moderate earthquake of M 6.9 weakened many structures, and a strong aftershock of M 5.8 completed the demolition.

In contrast to aftershocks, earthquakes called foreshocks often, but not always, precede major earthquakes by days or, in some cases, several years. For example, the 2011 M 9.1 Tōhoku earthquake, which occurred near the northeast coast of Japan, was preceded two days before by a M 7.9 event 40 kilometers (25 miles) to the east. A foreshock cannot be labeled as such until a larger earthquake follows in the same area. Monitoring of foreshocks to predict forthcoming earthquakes has been attempted with only limited success.

Faults and Large Earthquakes 

The slippage that occurs along faults can be explained by the plate tectonics theory, which states that large slabs of Earth’s lithospheric plates are continually grinding past one another. These mobile plates interact with neighboring plates, straining and deforming the rocks along their margins. Faults associated with convergent and transform plate boundaries are the source of most large earthquakes, while divergent margins are associated with numerous small earthquakes.

Convergent Plate Boundaries

Most of Earth’s strongest earthquakes occur along large faults associated with convergent plate boundaries. Along convergent boundaries where one continent is colliding with another, the resulting compressional forces slice Earth’s crust along numerous large thrust faults, discussed in more detail later this semester. The 2015 Nepal earthquake is one example of an earthquake generated along a thrust fault. The epicenter of the quake was located about 80 kilometers (50 miles) north of Kathmandu, where the Indian plate is advancing into the Eurasian plate at a rate of 4.5 centimeters (about 2 inches) per year, driving the uplift of the Himalayas.

When convergence entails the subduction of oceanic lithosphere under another plate, the area of contact between the two plates forms an extensive fault zone, called a megathrust fault, that can be several thousand kilometers long (Figure 5). An example of a megathrust fault is the Cascadia Subduction zone that stretches 1000 kilometers (620 miles) from Cape Mendocino, California, to Northern Vancouver Island, Canada. Along most subduction zones these megathrust faults remain locked for decades or even centuries. As the subducting plate slowly descends, it drags and bends the leading edge of the overlying plate, sometimes producing a bulge on the ocean floor. Once the frictional forces between the two stuck plates are exceeded, the overriding plate snaps back to its original shape. This snapping back generates an earthquake whose magnitude depends largely on the size of the zone of slippage.

Megathrust faults have produced the majority of Earth’s most powerful and destructive earthquakes, including the 2011 Japan earthquake (M 9.1), the 2004 Indian Ocean (Sumatra) earthquake (M 9.1), the 1964 Alaska earthquake (M 9.2), and the largest earthquake yet recorded, the 1960 Chile earthquake (M 9.5).

Transform Plate Boundaries

Faults in which the dominant displacement is horizontal and parallel to the direction of the fault trace (the line where the fault intersects Earth’s surface) are called strike-slip faults (discussed in more detail later this semester). Recall from our discussions about plate tectonics that transform plate boundaries, or simply transform faults, accommodate this type of motion between two tectonic plates. For example, the San Andreas Fault is a 1200-kilometer-long (745-mile-long) transform fault that lies between the North American plate and the Pacific plate (Figure .6). The East Anatolian Fault, the site of two catastrophic earthquakes in 2023, is a 700-kilometer (435-mile) strike-slip fault, accommodating movement between the Anatolian and Arabian Plates. Most large transform faults are not perfectly straight or continuous; instead, they consist of numerous branches and smaller fractures that display kinks and offsets (refer to Figure 6). Earthquakes can occur along any of these branches.

Fault Rupture and Propagation 

By studying earthquakes around the globe, geologists have learned that displacement along large faults occurs along discrete fault segments that often behave differently from one another. Some sections of the San Andreas Fault, for example, exhibit slow, gradual displacement known as fault creep and produce little seismic shaking. Other segments slip at relatively closely spaced intervals, producing numerous small to moderate earthquakes. Still other segments remain locked and store elastic energy for a few hundred years before they break loose or rupture. Segments that have been locked for a hundred years or longer usually result in large ruptures and, thus, major earthquakes.

Geologists have also discovered that slippage along large faults, such as the San Andreas Fault, does not occur instantaneously. The initial slip occurs at the hypocenter and propagates (travels) along the fault, either in all directions or in a single direction. As each section slips, it puts strain on the next section, causing it to slip, as well. Slippage propagates at 2 to 4 kilometers (1 to 2.5 miles) per second—faster than a rifle shot. Rupture of a 100-kilometer (60-mile) fault segment takes about 30 seconds, and rupture of a 300-kilometer (200-mile) segment takes about 90 seconds. As rupturing progresses, it can slow down, speed up, or even jump to a nearby fault segment. Earthquake waves are generated at every point along the fault as that portion of the fault begins to slip.

• The sudden movements of large blocks of rock on opposite sides of faults cause most earthquakes. The location where the rock begins to slip is called the hypocenter, or focus. During an earthquake, seismic waves radiate outward from the hypocenter into the surrounding rock. The point on Earth’s surface directly above the hypocenter is the epicenter.

• The ultimate cause of earthquakes is differential stress that gradually bends Earth’s crust over tens to hundreds of years. Up to a point, frictional resistance along the fault keeps the rock from rupturing and slipping. Once that point is reached, the fault slips, allowing the bent rock to “spring back” to its original shape, generating an earthquake. The springing back is called elastic rebound.

• Convergent plate boundaries and associated subduction zones are marked by megathrust faults. These large faults are responsible for most of the largest earthquakes in recorded history. Megathrust earthquakes may also generate tsunamis.

• The San Andreas Fault in California is an example of a large strike-slip fault that forms a transform plate boundary capable of generating destructive earthquakes.

• aftershocks: Earthquakes of lesser magnitude that follow a strong earthquake.

• earthquake: Ground shaking produced by the rapid release of energy from slipping rock blocks along faults.

• elastic rebound: The sudden release of stored strain in rocks that results in movement along a fault.

• epicenter: The location on Earth’s surface that lies directly above the focus of an earthquake.

• faults: Fractures between blocks of rock along which movement has occurred.

• fault creep: Displacement along a fault that is so slow and gradual that little seismic activity occurs.

• focus: The location and depth of thin Earth where displacement begins in an earthquake. Also called the hypocenter.

• foreshocks: Lesser magnitude earthquakes that often, but not always, precede a major earthquake.

• megathrust fault: The contact zone between two converging tectonic plates, where an oceanic lithospheric plate subducts under an overlying plate; capable of producing very strong earthquakes.

• seismic waves: Waves of elastic energy generated by an earthquake that travel within Earth and along the surface.