The most important fact about the San Andreas Fault is also the least satisfying one: scientists can measure rising risk, but they still cannot name the day a giant earthquake will begin. That gap between knowledge and certainty is why researchers focus on warning patterns instead of predictions. The U.S. Geological Survey has stated that “Neither the USGS nor any other scientists have ever predicted a major earthquake.” What they can do is study how the fault stores strain, where it remains locked, how often it has ruptured in the past, and how seismic waves could move through Southern California if a large event starts in the wrong place.
One major clue is slip building on locked fault sections. The San Andreas is part of the boundary between the Pacific and North American plates, which continue to move even when the fault itself is stuck. That mismatch stores energy. Some segments creep a little and release motion gradually, but others stay locked for long stretches. Scientists have used GPS, satellite radar, trench studies, and drilling projects such as SAFOD near Parkfield to map this uneven behavior. Repeating microearthquakes can reveal where small patches are slipping, while nearby zones remain jammed.
A second clue comes from the long record buried in the ground. Paleoseismology studies near the Salton Sea and other parts of the fault show that some sections have gone beyond their average interval between large earthquakes. That does not mean the fault is running late on a schedule. It means the rhythm is irregular, and the southernmost San Andreas has not produced a rupture of similar size in recent recorded history even though evidence suggests major events occurred there centuries ago. A third clue is scale. In the UCERF3 statewide forecast, researchers estimated a 7% chance of a magnitude 8 or larger earthquake striking California within 30 years. That figure is not a countdown. It is a probability model built from fault slip rates, geology, and past earthquakes. It places the conversation in decades, not days. Direction also matters.
Geologists have long warned that a rupture beginning near the Salton Sea could be especially damaging because the fault geometry can send shaking energy toward the Los Angeles Basin. Kate Scharer described that scenario bluntly: “It’s shooting all of that energy straight into the L.A. Basin.” In the same discussion, Lucy Jones explained why the San Andreas stands apart from smaller nearby faults: “With 300 miles of fault all going in the same earthquake, you then have everybody affected at the same time.” That concern is not only about collapsed structures. Passes such as San Gorgonio and Cajon carry highways, rail lines, aqueducts, fuel corridors, and power transmission routes that connect Southern California to the rest of the region.
Recent research has added another layer of caution. A 2025 study examining a long rupture on Myanmar’s Sagaing fault, which researchers called similar to the San Andreas, found that earthquakes do not repeat in exact patterns. Faults can rupture farther than expected, and old scenarios do not capture every possibility. That finding reinforces an older lesson from California: large earthquakes are not clockwork events, even on well-studied faults.
The practical response is not prediction theater but faster alerts and stronger systems. California’s earthquake early warning network processes ground-motion data fast enough to send alerts before strong shaking arrives in some locations, and ShakeAlert can also trigger automated actions such as slowing trains and protecting equipment. Seconds are not a forecast. They are time to brace, stop, shut down, and hold on.
