Plate Tectonics and Fault Types

Plot every earthquake recorded over a decade on a world map and something striking appears: the dots are not scattered. They form narrow, continuous belts that wrap around the globe. Those belts are the edges of tectonic plates, and this observation was one of the decisive pieces of evidence for plate tectonics in the 1960s.

The plates and how they move

The Earth's rigid outer shell, the lithosphere, is broken into roughly fifteen major plates and dozens of smaller ones. It sits on the asthenosphere, a hotter, weaker layer that deforms slowly over geological time. Plates move over that layer at rates between about 1 and 15 cm per year, driven mainly by slab pull — the weight of cold dense lithosphere sinking at subduction zones — with a contribution from ridge push at spreading centres.

Because the plates are rigid and the Earth's surface is fixed in area, every plate boundary must be one of three kinds: pulling apart, pushing together, or sliding past. Each produces a characteristic fault style and a characteristic earthquake signature.

The three fault types

Normal faults form where the crust is being stretched. The block above the fault plane, the hanging wall, drops down relative to the block below. Extension thins the crust, so normal faulting dominates at mid-ocean ridges and continental rifts such as the East African Rift and the Basin and Range province.

Normal-faulting earthquakes are generally shallow and rarely exceed magnitude 7. The mechanics limit them: extension does not create the enormous locked interfaces that compression does.

Reverse and thrust faults form where the crust is being compressed. The hanging wall is pushed up and over the footwall, thickening the crust. A thrust fault is simply a reverse fault with a shallow dip angle, typically under 45 degrees.

These produce the largest earthquakes on Earth. Compression at a converging boundary generates a broad, shallow-dipping contact surface between the two plates, and a shallow dip means a huge fault area available to rupture in one go. Every magnitude 9 event in instrumental history has been a subduction thrust.

Strike-slip faults form where two blocks slide horizontally past each other with little vertical motion. The San Andreas Fault in California, the North Anatolian Fault in Turkey and the Alpine Fault in New Zealand are the classic examples.

Strike-slip earthquakes can be large — the North Anatolian Fault has produced magnitude 7.8 events — but they are bounded because the fault is essentially vertical, so its rupture width is limited to the thickness of the brittle crust, about 15–20 km. Length can be great, width cannot, so the total area is capped.

Boundary types and the earthquakes they make

Divergent boundaries occur where plates pull apart, mostly along mid-ocean ridges. New crust forms as magma rises. Earthquakes here are shallow, frequent and small — usually below magnitude 6 — because the hot young crust is thin and weak and cannot store much elastic strain.

Convergent boundaries occur where plates collide. What happens depends on what is colliding.

Where oceanic lithosphere meets continental lithosphere, the denser oceanic plate subducts beneath the continent. This creates a deep trench offshore, a volcanic arc inland, and a locked plate interface capable of magnitude 9 earthquakes. The full depth range of seismicity is expressed here, from shallow interface events to deep-focus earthquakes 700 km down inside the descending slab.

Where two oceanic plates meet, one subducts beneath the other, producing island arcs such as the Aleutians, the Marianas and Tonga–Kermadec.

Where two continental plates collide, neither subducts readily because both are buoyant. Instead the crust crumples and thickens. This built the Himalaya, and it produces frequent large shallow earthquakes across a broad zone rather than a single narrow boundary — which is why seismic hazard extends across much of central Asia rather than along one line.

Transform boundaries occur where plates slide past each other, conserving crust rather than creating or destroying it. Earthquakes are shallow and can be destructive precisely because they occur in the crust directly beneath populated areas.

Intraplate earthquakes

Not every earthquake obeys the boundary rule. Intraplate events occur within plate interiors, typically reactivating ancient faults from long-dead tectonic episodes under stresses transmitted from distant boundaries.

They are far less frequent, but they are not harmless. The New Madrid sequence of 1811–1812 in the central United States produced several events estimated in the magnitude 7 range. Because old continental crust is cold, rigid and comparatively unfractured, it transmits seismic energy far more efficiently than the young broken crust at plate boundaries, so intraplate earthquakes are felt over dramatically larger areas than boundary events of the same magnitude.

Their rarity is a hazard in itself: regions that experience a damaging earthquake once every several centuries rarely build for it.

Why this shapes hazard maps

Understanding the boundary type tells you what to expect. A coastal city above a subduction zone faces rare but potentially enormous events plus tsunami risk. A city on a strike-slip fault faces more frequent moderate-to-large shallow events with little tsunami hazard. A city in a stable interior faces very rare events that most residents will never experience but that would be widely felt if they occurred.

That distinction drives building codes, insurance pricing and emergency planning far more usefully than a raw count of how many earthquakes a region records.