How Do Earthquakes Work?

In February 2023, a magnitude 7.8 earthquake struck southern Turkey and northern Syria in the early morning hours. It killed more than 50,000 people, destroyed or damaged hundreds of thousands of buildings, and set off a humanitarian crisis that would last months. The earthquake lasted about 90 seconds. In those 90 seconds, stress that had been building along a fault line for hundreds of years was released completely. The ground on one side of the fault shifted up to 10 meters relative to the other side, in some places permanently reshaping the landscape.

This is what earthquakes are: the sudden release of stored mechanical energy in the Earth’s crust. The shaking is the energy radiating outward as seismic waves. The damage comes from structures that were not built to withstand that shaking.

The short answer

An earthquake is the ground shaking caused by the sudden rupture and slip along a fault in the Earth’s crust. Tectonic forces slowly build stress in the crust over years or centuries. When the stress exceeds the strength of the fault, the fault ruptures and the two sides slip past each other, releasing energy as seismic waves. Scientists measure earthquakes using magnitude (the total energy released) and intensity (the shaking felt at a specific location). The three main fault types are normal (extension), reverse (compression), and strike-slip (horizontal shear).

The full picture

Why earthquakes happen: plate tectonics and stress accumulation

Earth’s outer layer, the lithosphere, is broken into tectonic plates that move relative to each other at a few centimeters per year. This is slow by human standards, but it is relentless. At plate boundaries, where plates collide, separate, or slide past each other, the movement is not smooth. Friction locks the rocks together where they are in contact. As the plates continue to move, stress builds up in the locked section, like pressure building in a spring.

When the stress exceeds the strength of the locked section, the fault ruptures. The two sides of the fault slip suddenly, releasing the stored energy as seismic waves. This is an earthquake, as described by the USGS earthquake science program.

The process of building stress and releasing it is called the earthquake cycle. A fault may not move for centuries while stress accumulates, then rupture in seconds, then lock again and begin accumulating stress anew. This is why major earthquakes in a given region tend to recur at intervals of hundreds or thousands of years.

Fault types

The direction of movement along a fault determines what kind of fault it is and what kind of stress caused it.

Normal faults occur where the crust is being extended, like at mid-ocean ridges where plates pull apart. The hanging wall (the block above the fault plane) drops down relative to the footwall. This creates features like the Basin and Range province in the western United States, where parallel normal faults have created a series of alternating mountain ranges and valleys.

Reverse faults, also called thrust faults, occur where the crust is being compressed. The hanging wall is pushed up and over the footwall. The Himalaya Mountains and the Subduction zones where one tectonic plate dives beneath another are characterized by reverse faulting. The 2011 magnitude 9.1 Tohoku earthquake in Japan occurred on a reverse fault at a subduction zone, where the Pacific Plate is being pushed beneath Japan.

Strike-slip faults occur where two blocks of crust slide horizontally past each other. There is no vertical displacement, only horizontal. The San Andreas Fault in California is the most famous example, where the Pacific Plate slides northwest relative to the North American Plate. The 1906 San Francisco earthquake was a strike-slip event on the San Andreas Fault, with the two sides of the fault offset by up to 6 meters.

Seismic waves: how the energy travels

When a fault ruptures, it releases energy in the form of seismic waves that radiate outward in all directions. There are three main types.

P-waves (primary waves) are the fastest. They are compression waves, similar to sound waves, where material compresses and expands as the wave passes. P-waves travel through both solids and liquids. They arrive at seismometers first, which is how scientists know they are primary.

S-waves (secondary waves) are shear waves. They move material perpendicular to the direction the wave is traveling, like a rope being snapped. S-waves are slower than P-waves and can only travel through solids, not liquids. This is how scientists know Earth’s outer core is liquid: S-waves do not pass through it.

Surface waves travel along the Earth’s surface. They are slower than P-waves and S-waves but typically cause the most damage because they have larger amplitudes and longer durations. Love waves move the ground side to side horizontally. Rayleigh waves make the ground roll up and down, like ocean waves. Both are surface phenomena that decay with depth.

The shaking you feel during an earthquake is mostly surface waves. The P-wave arrives first as a sharp jolt. The S-wave follows with a more prolonged shaking. The surface waves arrive last and are usually the most destructive.

Measuring earthquakes: magnitude and intensity

When you hear that an earthquake was magnitude 7.0, what does that mean?

Magnitude is a measure of the total energy released by an earthquake, calculated from the amplitude of seismic waves recorded at seismometers. The scale used today is the Moment Magnitude Scale (Mw), which is calculated from the area of the fault that slipped, the amount of slip, and the stiffness of the rocks involved. This replaced the older Richter scale, which was less reliable for large earthquakes and could only be applied to local events.

Each whole number increase in magnitude represents roughly 32 times more energy released. A magnitude 7.0 earthquake releases about 32 times more energy than a magnitude 6.0, and about 1,000 times more than a magnitude 5.0. The largest earthquake ever recorded was the 1960 Chile earthquake at magnitude 9.5, which released more energy than all nuclear weapons ever tested combined.

Magnitude is a single number for the whole earthquake. Intensity, on the other hand, varies by location. Intensity measures the shaking experienced at a specific place, influenced by distance from the earthquake source, local geology, and building construction. Soft soils like reclaimed land amplify shaking more than solid rock. The Modified Mercalli Intensity Scale rates shaking from I (not felt) to XII (total destruction). A magnitude 6.0 earthquake might produce Intensity IV in a city 50 kilometers away and Intensity VIII closer to the source.

Location: triangulation from seismic waves

Scientists locate earthquakes using seismometer records. A single seismometer tells you how far away an earthquake was but not in which direction. By combining records from three or more seismometers, you can triangulate the location. The point underground where the rupture starts is the focus or hypocenter. The point directly above on the surface is the epicenter.

Modern global networks of seismometers can locate an earthquake within minutes and publish a preliminary magnitude. Within hours, seismologists have refined the location, depth, and mechanism using data from dozens or hundreds of stations.

Effects: beyond the shaking

Ground shaking is the primary cause of earthquake damage, but it is not the only one.

Ground rupture occurs when the fault breaks through to the surface, offsetting roads, buildings, and other structures built across the fault. Building across an active fault is generally a bad idea for this reason.

Soil liquefaction occurs in water-saturated soils when the shaking causes the soil to lose its strength and behave like a liquid. Buildings can tilt or sink into the ground. The 1964 Niigata earthquake in Japan famously toppled several multi-story buildings that settled into liquefied soil, though the buildings themselves remained largely intact.

Tsunamis occur when large submarine earthquakes displace the ocean floor vertically. The water above is lifted and then collapses back, generating waves that travel across the ocean at hundreds of kilometers per hour. The 2004 Indian Ocean earthquake (magnitude 9.1) generated a tsunami that killed more than 230,000 people across 14 countries.

Landslides are common in mountainous areas where earthquakes shake loose unstable slopes. The 2015 Nepal earthquake triggered thousands of landslides in the Himalayan mountains.

Why it matters

Earthquakes are not predictable in the way weather is. We cannot forecast when and where an earthquake will occur, only estimate the probability that one will occur in a given region over a period of years. This probabilistic approach drives building codes, insurance models, and emergency preparedness.

Japan, California, Chile, New Zealand, and Turkey are among the most seismically active regions where large populations live. In these places, earthquake-resistant building design, emergency response systems, and public education about what to do during an earthquake are matters of life and death.

The economic costs are significant. The 2023 Turkey-Syria earthquake caused an estimated $34 billion in damage. The 2011 Tohoku earthquake and tsunami caused more than $200 billion in damage, making it the costliest natural disaster in history at the time.

Understanding earthquakes also tells us something fundamental about how the Earth works. The same plate tectonic forces that create earthquakes also build mountains, create ocean basins, and drive volcanic activity. Earthquakes are a byproduct of a dynamic planet, as explained by Smithsonian’s plate tectonics overview.

Common misconceptions

“Earthquakes are getting more frequent.” The global frequency of earthquakes has remained roughly constant over the past century. What has changed is our ability to detect them. Modern seismometer networks identify millions of small earthquakes each year that would have gone undetected before the 1990s. We are not having more earthquakes; we are simply seeing more of them.

“You can predict earthquakes.” There is no reliable method for predicting the specific time, location, and magnitude of an earthquake. Scientists can calculate probabilities over years or decades, which is useful for building codes and emergency planning. But a specific prediction like “a magnitude 6 earthquake will strike Barcelona next Tuesday” is not possible with current science.

“Earthquakes only happen at plate boundaries.” Most earthquakes do occur at or near plate boundaries, but earthquakes can occur anywhere there is stress in the crust. Induced seismicity from human activity, continental interiors under old tectonic stress, and volcanic regions all produce earthquakes away from active plate boundaries.

Key terms

Focus (hypocenter): The point underground where the earthquake rupture starts. This is the actual location of the energy release, as opposed to the epicenter on the surface directly above it.

Epicenter: The point on the Earth’s surface directly above the focus. This is the location most commonly cited when describing where an earthquake occurred.

Magnitude: A measure of the total energy released by an earthquake. The Moment Magnitude Scale (Mw) is the current standard. Each whole number increase represents roughly 32 times more energy.

Intensity: A measure of the shaking experienced at a specific location. Unlike magnitude, intensity varies by location. The Modified Mercalli Intensity Scale rates shaking from I (not felt) to XII (total destruction).

P-waves: Primary waves. The fastest seismic waves, arriving first at seismometers. They compress and expand the ground in the direction the wave travels, like sound waves through air.

S-waves: Secondary waves. Shear waves that move the ground perpendicular to the direction of travel. Slower than P-waves and unable to travel through liquid.

Surface waves: Seismic waves that travel along the Earth’s surface. Love waves move the ground horizontally; Rayleigh waves make it roll. Slower than P and S waves but typically cause the most damage because they have larger amplitudes.

Normal fault: A fault where the hanging wall drops down relative to the footwall, caused by extension of the crust. Common at mid-ocean ridges and continental rift zones.

Reverse fault: A fault where the hanging wall is pushed up relative to the footwall, caused by compression of the crust. Common at subduction zones and mountain-building regions.

Strike-slip fault: A fault where two blocks of crust slide horizontally past each other with no vertical displacement. The San Andreas Fault in California is a famous example.

Foreshock: A smaller earthquake that precedes the mainshock. Cannot be identified as a foreshock until the larger earthquake occurs.

Aftershock: Smaller earthquakes that follow the mainshock, caused by the crust adjusting to the new stress pattern created by the mainshock. Can continue for weeks to years after a large earthquake.

Induced seismicity: Earthquakes triggered by human activities such as wastewater injection, hydraulic fracturing, or filling large reservoirs. These are called induced earthquakes.