Running for roughly 750 miles up the length of California, the San Andreas fault marks one of the most closely watched fractures on the planet. It is the boundary where the Pacific and North American tectonic plates grind past one another, and the friction between them stores energy year after year until the rock finally slips. When it does, the release can be violent, which is why the fault sits at the center of California’s long-running effort to prepare for a major earthquake.
Much of that concern focuses on a southern stretch that has stayed locked for an unusually long time. Geologists studying the fault’s history have found that some segments rupture on a rough schedule, and by that reckoning parts of the southern San Andreas have gone longer without a great quake than their past behavior would predict. That gap is the basis for the familiar warning that a large earthquake is, in the language of seismic hazard, overdue.
A plate boundary sliding sideways
The San Andreas is a transform fault, meaning the two sides move horizontally past each other rather than one diving beneath the other. On the whole, the Pacific side is creeping northwest relative to the North American side at an average rate of a few centimeters a year — comparable, over geologic time, to the speed at which fingernails grow. The U.S. Geological Survey traces the system from the Salton Sea in the south to Cape Mendocino in the north, where it heads offshore.
That steady long-term motion hides an uneven reality along the fault’s length. In some places the two sides slide smoothly and continuously; in others they stick fast for centuries, loading up strain that is eventually unleashed in a sudden jolt. It is the sticking sections, not the creeping ones, that produce the largest and most destructive earthquakes.
Locked, creeping, and everything between
Geologists divide the fault into segments that behave very differently. A central section near the town of Parkfield creeps along more or less continuously, relieving stress in small increments and modest quakes. To the north and south, however, long stretches are effectively locked, accumulating strain that will be released only in a major rupture.
The southernmost segment, running through the Coachella Valley and toward San Bernardino, is the one that draws the sharpest attention. It has not produced a great earthquake in the historical record, and paleoseismic studies of ancient ruptures suggest its typical interval between major quakes is shorter than the time that has already elapsed. That mismatch is the technical meaning behind the notion that the section is overdue.
Reading the record of past ruptures
Scientists reconstruct the fault’s history by trenching across it and dating layers of sediment disturbed by prehistoric quakes, a discipline known as paleoseismology. Those records, combined with the historical accounts of large events, allow researchers to estimate how often a given segment ruptures and how much slip each event produces. The picture that emerges is of a fault that releases its accumulated strain in irregular but recurring great earthquakes.
Two events anchor the modern understanding of what the fault can do. The 1857 Fort Tejon earthquake tore a long section of the central and southern fault, and the 1906 San Francisco earthquake ruptured the northern segment, leveling much of the city with shaking and the fires that followed. Both were quakes of roughly magnitude 7.9, and both illustrate the scale of energy the locked sections can hold.
Why forecasts speak in probabilities
Despite decades of study, no method can predict the day or even the year of a particular earthquake. Instead, the U.S. Geological Survey and its partners express the hazard as probabilities — the estimated chance of a damaging quake of a given size somewhere in a region over the coming decades. Those forecasts draw on slip rates, the time since the last rupture, and the mechanical links between neighboring faults, and they consistently rank a large San Andreas earthquake among the more likely major seismic events in the state’s future.
To translate that hazard into preparedness, agencies have modeled detailed scenarios of what a great southern San Andreas rupture might do, estimating the shaking, damage, and disruption to water, power, and transportation that would ripple far beyond the fault line itself. Such exercises underpin the drills and building codes meant to limit the toll when the locked rock finally gives way.
Living alongside the fault
Tens of millions of people live within reach of the San Andreas system, and the practical response to its hazard has been engineering and planning rather than prediction. Stricter building standards, retrofitted structures, and early-warning systems that can send alerts in the seconds before shaking arrives are all designed to reduce casualties from an event whose timing cannot be known in advance.
The fault’s slow, relentless motion guarantees that strain will keep building on its locked segments, and the geologic record makes clear that the strain is eventually released in great earthquakes. The overdue framing is less a countdown than a statement of odds: on a fault that has behaved this way for millions of years, another major rupture is not a question of whether but of when.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
