Faults can appear at the surface as simple, straight features, but deep below ground they can be complex, and their hidden geometry can play in a role in their hazard.
By Erin Martin-Jones, Ph.D., @Catherine_MMJ
Citation: Martin-Jones, E., 2021, What looking below-ground can reveal about earthquake hazard, Temblor, http://doi.org/10.32858/temblor.227
For decades, seismologists have tried to figure out what makes some earthquakes grow to destructive proportion, whereas others stop short, sparing nearby cities from the strongest shaking. Understanding how fault ruptures spread is key and can help researchers make accurate hazard assessments in complex fault zones. Now, a new study shows that slight variations in the orientation along the length of faults, such as the San Andreas, can affect how big or how damaging an earthquake can be.
Slip follows the easy path
Research on recent and historic earthquakes shows that slip along a fault often follows the path of least resistance. A rupture will continue along a straight section of fault, but often stops when the slip reaches an obstacle — sometimes one that has been mapped out at the surface, such as a bend or step in the fault.
Julian Lozos, a seismologist at the California State University, Northridge, who authored the new study, wanted to know whether hidden features at the base of a fault could also deflect seismic waves and stop an earthquake in its tracks. His model is the first to consider the impact of a fault’s below ground geometry on rupture spreading.
Thinking in three dimensions
Lozos was inspired by his local San Andreas Fault, which is a classic example of a “strike-slip” fault — a vertical (or near-vertical) boundary where two sections of earth slide horizontally past each other. At least, that’s the idealized definition. Scientists know from patterns of past earthquakes and below-ground geophysical imaging that strike-slip faults, though often vertical, sometimes deflect, or “dip,” into the earth at an angle.
That dip angle can also vary along a fault’s length. Back in 2012, a study showed that part of the San Andreas Fault, located east of Los Angeles, has a “propeller” shape, with two adjacent sections dipping in opposing directions.
“Since that 2012 propeller model, I’ve wanted to take an earthquake model and introduce variable dip; I’d never seen it done before,” said Lozos.
The impact of variable dip
Lozos’ model shows that a rupture can stop when it reaches a change in dip along the length of a fault. Large dips that change over short distances are more likely to have this effect.
Lozos says that where the dip changes, there is a significant change in the amount of friction clamping the fault together. At depth, the change in friction has the same decelerating effect on earthquake rupture propagation as similar fault kinks or steps at the surface. Scientists had not been able to test the impacts of this hidden geometry before now.
That deceleration can even stop an earthquake rupture in its tracks. If a propagating rupture reaches a slight change in dip along the fault and stops, then the magnitude of the earthquake will be smaller than if it had continued unabated. As a result, shaking is restricted to a smaller area than if the rupture had continued. Dip changes could therefore result in a lower hazard.
The model also showed that variable fault dip at depth caused secondary effects, including vertical ground displacement normally associated with so-called dip-slip faults, which are boundaries where the blocks mostly shift vertically relative to one another. This ground displacement controls which areas are most affected by shaking, but is often overlooked in hazards assessments for strike-slip faults.
“The question of what causes earthquakes to stop once they have started has been a topic in the earthquake science community for decades, and we still haven’t quite figured out the answer,” explains Ruth Harris, a seismologist at the U.S. Geological Survey (USGS) Earthquake Science Center, who was not involved in the study. “Fault geometry is an appealing solution, especially because it is an Earth feature that we have a chance of imaging.”
Determining hazard
Although Lozos’ model is not tuned to a specific fault, his findings highlight why dip measurements should be included in hazards analysis for zones like the San Andreas. “It’s especially critical if you know the dip changes along strike,” said Lozos, who explains that this will directly impact the shaking and damage to nearby buildings and infrastructure.
Glenn Biasi, a geophysicist at the USGS Earthquake Science Center, says “[Lozos’] model echoes what many of us in the community have suspected from field observations for a while. We now have the physical model results needed to say, ‘look we need to think about this problem in three dimensions.’” Biasi, who was not involved in the study, says that the next step will be to use dip measurements from the San Andreas and other California faults to assess the hazard.
“I hope that my model will inspire both modelers and observationalists alike. This really shows how important it is that models include dip,” says Lozos. “But I also think models can help us better interpret what we see in the real-world.”
Lozos says he hopes his model will help seismologists interpret what they see in the field, whether they are studying recent or historic earthquakes. Past ruptures that stopped in the middle of a straight section of fault might hint to a change in dip direction. Knowledge of why a fault behaved in a particular way in the past, whether it stopped in a particular location or spread to a neighboring fault, is key to understanding what might happen next time.
Further Reading
Lozos, J. C. (2021). The effect of along-strike variation in dip on rupture propagation on strike-slip faults. Geosphere.
Fuis, G. S., Scheirer, D. S., Langenheim, V. E., & Kohler, M. D. (2012). A new perspective on the geometry of the San Andreas fault in southern California and its relationship to lithospheric structure. Bulletin of the Seismological Society of America, 102(1), 236-251.
Biasi, G. P., & Wesnousky, S. G. (2021). Rupture Passing Probabilities at Fault Bends and Steps, with Application to Rupture Length Probabilities for Earthquake Early Warning. Bulletin of the Seismological Society of America.