On the Influence of Fault Geometry on Earthquake Ruptures and Predictability
This study demonstrates through numerical simulations of the Sagaing Fault system that while earthquake ruptures can be predicted using slip-deficit distributions, the accuracy of such forecasting diminishes significantly in complex fault geometries compared to simpler ones, highlighting the critical role of fault structure in earthquake predictability.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Predicting when and where the next major earthquake will strike remains one of the most difficult challenges in geology. For decades, scientists have understood that the Earth's crust is constantly under pressure, building up stress along massive cracks in the ground known as faults. When this stress becomes too great, the rock snaps, releasing energy in the form of an earthquake. The hope has always been that by measuring how much stress has built up since the last quake, we could forecast the next one. However, the real world is messy. Faults are rarely straight lines; they twist, bend, and split into branches, creating a tangled network where stress behaves in unpredictable ways. This complexity often turns the process of earthquake prediction into a game of chance, where the timing and size of future ruptures seem chaotic rather than orderly.
A new study by researchers at the California Institute of Technology explores exactly how this tangled geometry of faults influences our ability to predict earthquakes. The team focused on the Sagaing Fault in Myanmar, a massive geological feature that runs through the country and recently produced a devastating magnitude 7.7 earthquake in 2025. Using powerful computer simulations, the researchers recreated thousands of years of earthquake history on this fault, testing how different shapes of the fault line change the behavior of the quakes. They found that while simple, straight faults produce regular, predictable patterns of shaking, the complex, branching nature of real-world faults introduces a level of chaos that makes forecasting significantly harder.
The researchers built a digital model of the Sagaing Fault system, programming it with the laws of friction that govern how rocks slide past one another. They ran eleven different simulations to see how the fault would behave under various conditions. In the simplest scenario, where the fault was modeled as a perfectly flat, straight line, the earthquakes occurred with remarkable regularity. These simulated quakes were large, striking every 150 to 200 years, and they released a consistent amount of energy. It was a clockwork system where the past reliably predicted the future. However, as soon as the researchers introduced the real-world features of the fault—its sharp bends and the smaller branches that split off from the main line—the pattern broke down. The earthquakes became irregular, varying in size and timing, and the rupture paths became fragmented, often stopping at the bends or jumping onto the smaller branches.
To test if these chaotic sequences could still be forecasted, the team developed a method based on the concept of "slip deficit." Imagine a fault that is supposed to move steadily at a rate of 20 millimeters per year due to the slow drift of tectonic plates. If an earthquake happens, the ground slips and relieves that built-up motion. The time between quakes is when the fault is "stuck," and the difference between how much it should have moved and how much it actually moved is the slip deficit. The researchers hypothesized that if they tracked this deficit over a long enough period, they could predict when the fault would finally give way again. They applied this method to their simulations, looking for areas where the accumulated deficit became large enough to trigger a rupture.
The results showed that this forecasting method works well, but only under specific conditions. When the fault geometry was simple, the method was highly accurate, correctly predicting the timing and location of large earthquakes about 90% of the time. However, in the simulations that included the complex, branching geometry of the real Sagaing Fault, the accuracy dropped significantly. The method could still identify some large events, but it missed many others and produced false alarms. The researchers found that the predictability of an earthquake depends heavily on how long the data used for the forecast covers. To get a reliable prediction for the complex fault, the model needed to look back at least 150 years, which is roughly the average time between major quakes on that fault. Looking back for shorter periods, such as 20 or 50 years, resulted in poor predictions because the irregularity of the complex fault meant that short-term patterns were misleading.
The study also revealed that the complexity of the fault directly influences the size of the earthquakes it produces. In the simpler, straight sections of the simulated fault, the earthquakes tended to be larger, averaging around magnitude 7.5. In the complex, branching areas, the ruptures were more likely to break apart into smaller segments, resulting in slightly smaller average earthquakes of about magnitude 7.1. This segmentation is a key reason why prediction fails in complex zones; the stress does not build up in one giant, predictable package but is instead distributed across many smaller, interacting pieces. The researchers noted that their simulations, while sophisticated, still simplified reality by assuming the fault surface was smooth and ignoring some of the inelastic deformation that happens in the rock surrounding the fault. They suspect that including even more physical details would likely make the system even more chaotic and harder to predict.
Ultimately, the work suggests that the shape of the fault is a dominant factor in whether we can foresee a major earthquake. While the idea of predicting earthquakes based on the time elapsed since the last one is appealing, this study indicates that such a simple approach is insufficient for complex fault systems. The researchers propose that future forecasting efforts must move beyond simple statistical models and embrace physics-based simulations that account for the intricate geometry of the fault network. By combining long-term geological data with advanced computer models that respect the true, messy shape of the Earth's crust, scientists may eventually improve their ability to anticipate these powerful natural events, even if perfect prediction remains out of reach. The findings underscore that in the complex dance of tectonic plates, the geometry of the fault line is not just a backdrop; it is a central character that dictates the rhythm of the shaking.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.