The last time Southern California’s San Andreas fault ruptured near Los Angeles, Abraham Lincoln was six years old. That was 1857 — and in the nearly 170 years since, the fault has been loading in silence.
Millions of people now live above two of the most stressed fault systems in North America, with pressure building invisibly in the crust beneath them. A new physics-based model, reconstructing a full millennium of earthquake history, has found that stress levels along both the San Andreas and San Jacinto faults have now surpassed every recorded peak of the past thousand years.
At the heart of the findings is a single fault junction northeast of Los Angeles that researchers call the “earthquake gate.”
A thousand years of seismic history, compressed into one model
The team, led by Liliane Burkhard at the University of Bern, built a physics-based earthquake cycle model and fed it a millennium of geological evidence — radiocarbon dating, tree-ring anomalies, and historical records of ground ruptures along both fault systems.
The model reconstructs cause and effect across centuries. Every past earthquake shifted stress onto neighboring fault segments, and every quiet interval allowed that stress to accumulate further. Large ruptures triggered slow relaxation in the deeper crust, sometimes for decades afterward. The simulation tracks all of it — a continuous chain of mechanical consequences stretching back a thousand years.
The specific target was Cajon Pass, a tectonically complex junction northeast of Los Angeles where the San Andreas and San Jacinto faults draw close together. Estimating present-day stress loading at that junction was the central goal.
Stress levels that have never been seen before — in a millennium
The results were unambiguous. Tectonic stress along the fault system has reached — and in some cases exceeded — the highest values recorded anywhere in the entire 1,000-year simulation.
The San Jacinto–Bernardino section now sits at 3.6 MPa, surpassing every prior peak in the model. The neighboring Mojave South section of the San Andreas fault registers 2.8 MPa, similarly elevated. Both numbers represent the upper range of what the simulation has ever produced.
The explanation is straightforward, if sobering. Since the 1857 Fort Tejon earthquake — a magnitude 7.9 event — neither fault has produced a comparable release. That silence has persisted for nearly 170 years, stress loading continuously with nothing large enough to interrupt it.
The ‘earthquake gate’ at Cajon Pass
Cajon Pass is not simply a geographic feature. In this model, it functions as an “earthquake gate“ — a junction that can either contain a rupture within one fault system or allow it to cross both simultaneously.
History shows both outcomes are possible. The 1857 Fort Tejon earthquake terminated at Cajon Pass and never involved the San Jacinto fault. The 1812 Wrightwood earthquake did the opposite, rupturing straight through the junction and crossing both systems in a single event.
Burkhard describes the gate not as a wall but as a switch. It doesn’t simply block or redirect earthquakes — it responds dynamically to stress conditions that have been shifting for centuries. Whether it opens or closes depends entirely on the state of the system at the moment a rupture arrives.
Why synchronized stress on both faults is the critical warning sign
The decisive variable is not the total stress on either fault in isolation. It is how closely aligned the stress levels on both faults are at the same time.
When both fault systems rise toward similarly high stress levels in concert, the model associates that configuration with large joint ruptures that cross the gate. Stress evolving out of step between the two systems, by contrast, tends to produce ruptures that stop at the junction rather than continue through it.
Right now, both the San Jacinto and San Andreas segments are highly stressed — and stressed to roughly similar degrees. That combination, Burkhard notes, is what makes the current moment especially significant. Record highs in synchrony is the specific pattern the model links to past joint ruptures.
What this means for millions of people — and for science
A joint rupture crossing both fault systems would be far more severe than anything confined to a single fault. The affected corridor includes greater Los Angeles, San Bernardino, Riverside, and the Coachella Valley — some of the most densely populated terrain in the United States. Major highways, railroads, and energy infrastructure run directly through Cajon Pass.
The researchers draw a clear line: this is not a prediction of when an earthquake will occur. It is a physics-based hazard assessment defining the range of scenarios planners and emergency managers should be actively preparing for.
The framework also extends beyond California. Burkhard notes that the modeling approach applies to complex fault junctions worldwide — anywhere intersecting fault systems create the same kind of conditional, stress-dependent behavior. What follows the science is a question of preparation.