Small and Large Earthquakes Don’t Play By the Same Rules

New research from seismologist Alice Gabriel of UC San Diego’s Scripps Institution of Oceanography suggests that the physics governing the behavior of small and large earthquakes are distinct from one another. The findings could have implications for understanding how earthquakes start and cascade into big seismic events, and ultimately aid in disaster preparedness.

By combining real-world earthquake measurements, key updates to mathematical models describing earthquake behavior and 3D simulations using supercomputers, Gabriel and her co-authors developed a unified theory capable of explaining minor jolts and destructive quakes involving multiple faults. 

“Fundamentally, this is about providing a unified theory of earthquake physics,” said Gabriel. “But this could also open the door to better understanding how earthquakes start, with smaller fractures that break more easily cascading into a large seismic event.”

Published July 26 in Science, the paper’s findings could help seismically active regions better prepare for earthquakes by updating hazard maps and more accurately assessing the strength of shaking their local fault systems can produce.  

The measurable shaking that earthquakes produce is only one part of the energy they release. The rest of an earthquake’s energy goes into processes inside the fault such as overcoming friction and breaking various types of rock. Because scientists can directly measure only the portion of an earthquake’s energy that gets translated into shaking, the remainder can only be estimated, making it challenging to fully understand earthquake mechanics.

Historically, seismologists assumed that earthquakes of all sizes followed the same fundamental processes and physics. An earthquake’s energy, known as its fracture energy, was thought to steadily increase with the distance the fault slipped and the dimensions of the fault rupture. But these assumptions broke down when it came to small and large seismic events. Modern seismic measurements suggested that minor tremors must behave differently than major earthquakes. 

Beginning in 2019, Gabriel and her co-authors were looking to understand how small fractures interacted with bigger faults. In particular, she had in mind the surprising 2019 Ridgecrest, California earthquakes and, as the study neared its completion, the devastating pair of quakes that struck Turkey and Syria in 2023. These earthquake sequences involved so-called cascades of ruptures that propagated across multiple faults in unexpected ways.

“We realized it was impossible to model earthquakes across multiple scales in the current framework because it assumes that the physics of small and large earthquakes are the same,” said Gabriel. For example, she said extending the physics of small earthquakes to larger fault zones led to estimated earthquakes that were “way too big and fast compared to what scientists have actually measured.”

To see if they could reconcile these issues with modeling small and large earthquakes, the researchers created new mathematical models of earthquake physics to describe the mechanics of their propagating cracks and fractures. One key update to these models involved the friction that earthquakes must overcome as they rupture a fault.

“Previous models treated earthquakes as though they didn’t move very fast,” said Gabriel. “But laboratory evidence shows some earthquakes can cause the two sides of a fault slip past each other fast enough that friction almost vanishes, which would mean the fault would have very little resistance to rupture further. Our new models account for that dramatic weakening.” 

The new models also include an effect called stress undershoot, in which a ruptured fault’s frictional stress at the end of an earthquake is higher than the frictional stress during the seismic slip. 

All in all, the new mathematical models allowed the researchers to capture more of the intricate physical processes that real faults experience during earthquakes.

The team then used these new models to reevaluate the fracture energy from measurements of actual earthquakes spanning magnitudes from 1.9 to 9.2. With these fracture energy estimates and their new mathematical models, the researchers were able to derive equations that described the relationship between the size of a fault rupture and the energy released by its fracturing. 

“We found a very intuitive scaling relationship between the minimum energy released by each earthquake and the size of the fault,” said Gabriel. 

But when the researchers parsed the output of their models they noticed a break in the scaling at around one centimeter of slippage (0.4 inches). For small earthquakes with limited fault slip, the fracture energy is linked to the size of the rupture area, with bigger faults requiring more energy to rupture. But beyond one centimeter of slippage fracture, energy began continuously increasing with the amount of sliding in the fault. Gabriel noted that the one centimeter threshold is not universal, and could vary depending on the forces and energy of a given earthquake.

Last, the researchers used all this information to run realistic 3D earthquake simulations across more than 700 intersecting fractures and a nearby larger fault on supercomputers. The simulations demonstrated in detail that the relationships between fracture energy, fault slip and fracture size that the researchers described could help explain the interactions between small fractures and nearby larger faults. In particular, this helps elucidate cascading multi-fault ruptures like those seen in the Turkey-Syria quakes of 2023, in which smaller ruptures appeared to set off the larger fault and fractures jumped from one fault to another. 

The findings could allow researchers to hunt for conditions in which small fractures would have the potential to grow into a large earthquake, or not, which could improve disaster preparedness in places like Southern California by updating hazard maps and estimating maximum earthquake magnitude.

Next, Gabriel said she and her colleagues are searching for measurable qualities in real-world earthquakes that could support or conflict with the relationships outlined in the study. She is also hopeful this new theory of earthquakes may help seismologists understand what happens to faults between earthquakes when stresses are accumulating on longer time scales.

This research was supported by the European Union’s Horizon Research and Innovation Programme, the National Aeronautics and Space Administration, the National Science Foundation, the Gauss Centre for Supercomputing, the Natural Sciences and Engineering Research Council of Canada, the Southern California Earthquake Center, the Swiss Data Science Center and the King Abdullah University of Science and Technology.

In addition to Gabriel, Dmitry Garagash of Dalhousie University, Kadek Palgunadi of the Swiss Seismological Service and P. Martin Mai of King Abdullah University of Science and Technology co-authored the study.