By Kevin Wilcox
A computer model finds that a damaging Southern California earthquake in 1812 was likely the result of a rupture of both the San Andreas and San Jacinto faults.
New research suggests that the San Andreas Fault and the San Jacinto Fault ruptured jointly to cause the significant December 8, 1812, earthquake in Southern California—and they could do so again in the future. California Department of Conservation
April 5, 2016—Using a complex computer model and paleoseismological data, a researcher at California State University Northridge has determined that a powerful earthquake that rocked a swath of the state from north of Los Angeles to San Diego in late 1812 was likely caused by a joint rupture of the San Andreas and San Jacinto faults. A similar earthquake today would more directly send strong seismic energy into population centers of the San Bernardino and San Jacinto valleys.
Julian C. Lozos, Ph.D., an assistant professor of geological science at California State University Northridge, was drawn to study the December 8, 1812 earthquake by the unusual observations it produced from the state's residents. The pattern of damage did not make sense for an earthquake propagated by the San Andreas Fault alone. Also, "there was such strong damage really far away from the fault," Lozos says. "San Juan Capistrano is not close to the San Andreas Fault at all," yet it suffered some of the worst damage, he says. Lozos recently published his findings, "A Case for Historic Joint Rupture of the San Andreas and San Jacinto Faults," in the journal
The 1812 earthquake was initially attributed to the smaller Newport-Inglewood Fault because of that fault's proximity to the areas that experienced the greatest damage. Paleoseismologists examining the disruption to tree growth rings further north along the San Andreas Fault discovered anomalies between the spring growth seasons of 1812 and 1813 that enabled them to attribute the quake to the San Andreas.
Lozos used a dynamic rupture model for his research. The model is designed to capture the complex physics of an earthquake, from the fault mechanics to the stress transfer to the wave propagation. He used fault geometry data based on field mapping as well as geologic information from the community velocity model developed by the Southern California Earthquake Center at the University of Southern California. He then estimated the amount of stress required to create the slip that has been observed by paleoseismologists in trenches that cut across the fault.
"I have a good idea of the orientation of the stress," Lozos says. So he asked, "What amplitude of stress do I need to get the amount of slip that we see in the paleoseismic trenches? Having that observational reference was really, really important." Lozos plugged the data into his model and it performed thousands of differential equations of simulated earthquake conditions for thousands of points across the Southern California region.
"Stresses that were enough to make [the correct] amount of slip also made both faults [in the model] go off at once," Lozos says. "Any time that I had an earthquake that was only on one fault, it didn't have enough slip to match the data."
The model also matched many of the unusual observations from the actual earthquake, including an exceptionally strong wave of ground motion that destroyed a mission in San Juan Capistrano, reportedly killing 40.
Lozos says his research isn't the first to suggest the possibility of such a joint San Andreas-San Jacinto earthquake, but it does provide physics-based verification of such an event with results that match the real-world observations of a historical earthquake, thus refining scientific understanding of this fault interaction. The results also indicate that the San Jacinto Fault, which extends through populated areas, is capable of stronger earthquakes—magnitude 7.5—when acting in concert with the San Andreas than the 6.5 magnitude earthquakes that standard hazard assessments typically suggest, Lozos says.
"When you have an earthquake from the San Jacinto to the San Andreas, we can have a much bigger earthquake. The implications are pretty important," Lozos says. "We are talking about a 7.5 on both faults. I think the 1812 earthquake started in the south and went north, but the model shows that it's physically plausible for an earthquake to start in the north and go south," Lozos says.
And a stronger earthquake will affect the region significantly because of the composition of its soils. "There is evidence in the 1812 earthquake that San Bernardino had a serious liquefaction problem," Lozos explains. "Certainly San Bernardino and Redlands and Loma Linda are on a sedimentary basin. The ground there is going to shake really strongly. Because [those areas are] between the faults, the faults are actually stretching the land there. [The land] is dropping down and filling up with settlement from the mountains. And the water coming in from the mountains soaks into the sediment and gets stopped up against the faults. You basically have this bowl of sediment."
Lozos notes that his research focused on low-frequency ground motions; he would next like to work with other researchers to develop a broader model that would simulate ground motion at the higher frequencies that pose the greatest problems for infrastructure.
With his current model validated by the 1812 earthquake observations, he is now running a series of forward-focused model scenarios to determine how hypothetical earthquake scenarios would behave along the two faults, which could better inform local planning efforts.