How Earthquakes Happen: Faults, Seismic Waves, and Rupture Mechanics

A Fault That Stores Centuries of Energy and Releases It in 90 Seconds

The 2011 Tōhoku earthquake, which struck Japan on March 11, occurred along the Japan Trench where the Pacific plate subducts beneath the North American plate at roughly 9 cm per year. Over centuries, this locked subduction zone had accumulated elastic strain equivalent to approximately 1.9 × 10¹⁸ joules — the energy content of ~450 million tons of TNT. When the fault finally slipped, roughly 500 km of fault rupture propagated in about 90 seconds, displacing the seafloor by up to 50 meters horizontally and 7–8 meters vertically, generating a tsunami that killed approximately 19,700 people and triggered the Fukushima nuclear disaster. It was the most powerful earthquake Japan had ever recorded and the fourth most powerful in the world since 1900.

Elastic Rebound Theory

The modern understanding of earthquake mechanics is elastic rebound theory, developed by Harry Fielding Reid following the 1906 San Francisco earthquake. Tectonic forces slowly deform rock on either side of a fault over years and decades. The rocks behave elastically — storing strain energy like a compressed spring — because friction along the fault plane prevents slipping. When accumulated stress exceeds the fault's frictional strength, the fault ruptures. The two sides of the fault snap back toward their undeformed shapes — elastic rebound — releasing the stored strain energy as seismic waves.

Types of Faults

Seismic Waves: The Messengers of Rupture

An earthquake radiates energy in the form of seismic waves, which travel through Earth's interior and along its surface:

• P-waves (Primary waves): Compressional waves; particles oscillate parallel to the direction of wave travel. Travel through solids, liquids, and gases. Fastest seismic wave (~6–13 km/s in the crust and mantle). Arrive first at seismometers. P-waves can propagate through the liquid outer core; P-wave shadow zones reveal the core's existence.
• S-waves (Secondary waves): Shear waves; particles oscillate perpendicular to wave travel direction. Travel only through solids (cannot propagate in liquids). Slightly slower than P-waves (~3.5–7.5 km/s in the crust). S-wave shadow zones on the opposite side of Earth from an earthquake confirmed the existence of the liquid outer core.
• Love waves: Surface waves with horizontal ground motion perpendicular to wave travel. Can cause significant lateral shaking of buildings.
• Rayleigh waves: Surface waves with elliptical retrograde particle motion (combination of vertical and horizontal). Often felt as a rolling motion; cause much of the visible surface damage in large earthquakes.

Measuring Earthquake Magnitude and Intensity

Two types of measurements characterize earthquakes: magnitude (energy released at the source) and intensity (shaking at a specific location).

Moment Magnitude Scale (Mw) — now the standard — is based on the seismic moment M₀ = μ × A × D, where μ is rock rigidity, A is the fault rupture area, and D is average slip. Each integer step on the Mw scale represents ~31.6 times more energy released:

Aftershocks and the Omori Law

Following a major earthquake (mainshock), a sequence of smaller aftershocks follows. Japanese seismologist Fusakichi Omori established in 1894 that the rate of aftershocks decays approximately as 1/t after the mainshock, where t is time since the mainshock. The largest aftershock is typically about one magnitude unit smaller than the mainshock (Båth's law). After the 2011 Tōhoku earthquake (Mw 9.0), aftershocks with Mw ≥ 5 continued to occur at measurably elevated rates for years. The foreshock-mainshock-aftershock sequence reflects the progressive slip and stress transfer along and surrounding the initial rupture zone.

Earthquake Early Warning Systems

Because P-waves travel faster than S-waves and surface waves, and because P-waves cause less shaking, detecting P-waves and broadcasting a warning before the more damaging S-waves arrive is the basis of earthquake early warning (EEW) systems. Japan's nationwide J-Alert system, launched in 2007, can provide up to 30–90 seconds of warning for distant earthquakes — enough time to stop bullet trains, halt surgeries, duck under desks, and take other protective actions. California's ShakeAlert system has been operational since 2019. The warnings are short (often 3–30 seconds) but valuable: death and injury rates decrease significantly with just a few seconds of advance notice.

The Challenge of Earthquake Prediction

Despite decades of research, reliable short-term earthquake prediction — specifying time, location, and magnitude days to weeks in advance — remains beyond current scientific capability. The reasons are fundamental: the Earth's crust is a complex nonlinear system; many faults are hidden; and the nucleation of rupture may be inherently unpredictable below a certain scale. What seismology can do is probabilistic seismic hazard assessment (PSHA): quantifying the probability that shaking of a given intensity will occur at a location within a specified time period, which is the basis for building codes and urban planning in seismically active regions.