Throughout this project, I investigated what might allow some major faults within California's San Andreas fault system to move together in a single earthquake, and what may prevent an earthquake from even reaching the end of its causative fault. I used physics-based computer simulations of the earthquake process to address these questions, and incorporated geological, observational, and historical data in both the setup and interpretation of those models, to make sure they remained grounded in reality.

One of my major focuses was interactions between the San Andreas and San Jacinto faults, in southern California's densely populated Inland Empire. The San Jacinto may accommodate over half of the movement between the Pacific and North American plates where it overlaps with the San Andreas. This implies not only that both faults pose hazard individually, but also that they have the potential to interact. The question of their interaction is made more complicated by the fact that the San Jacinto is composed of several disconnected strands, and is also disconnected from the San Andreas.

In collaboration with researchers at University of California, Riverside, San Diego State University, and University of Nevada, Reno, I simulated earthquakes on the northern and central San Jacinto, to test how physically plausible it was for an earthquake to span both discontinuous fault strands. My only model earthquakes that spanned the discontinuity produced far more slip on the fault than is supported by geological evidence. These models also produced stronger ground shaking than is supported by the presence of precariously-balanced rocks - fragile structures that only remain standing because earthquake shaking has not been strong enough to knock them down - near the fault discontinuity. We published these simulations, and the interpretation that the northern and central San Jacinto are not likely to participate in a single earthquake, in Bulletin of the Seismological Society of America in 2015.

My San Jacinto Fault simulations were critical background for my investigation of the southern California earthquake of 8 December 1812. The San Andreas Fault, and both the northern and central San Jacinto Fault, have geologic evidence of a large earthquake in the early 1800s. There were two such earthquakes - one in 1800, another in 1812 - but three fault segments. This implies that one of these earthquakes involved two fault segments. Considering my earlier San Jacinto work, I chose to simulate earthquakes on the northern San Jacinto and the San Andreas faults: both to see if it was plausible for these two faults to slip together, and to see if any of these simulated earthquakes were consistent with geological and historical evidence of the 1812 earthquake. I found that not only was it favorable for an earthquake to involve both the San Andreas and San Jacinto faults, but an earthquake that started on the San Jacinto and propagated northward onto the San Andreas fit all the observational datasets best. I published this result as a single-author paper in Science Advances in 2016.

My other major focus during this project was investigating how patches of aseismic creep on a fault may confine the size of the earthquakes that fault can host. Aseismic creep is a slow constant movement of a fault, which is caused by frictional properties that make the fault strengthen as it slips further, and which results in the creeping part of the fault releasing its energy constantly, and not having as much stored for a potential future earthquake. To test these effects, I collaborated with colleagues at the U.S Geological Survey on simulating earthquakes on Northern California's Bartlett Springs Fault: which has both geologic and geodetic data about its pattern and rate of creep, and which poses a significant hazard to power infrastructure. I found that creeping sections of a fault do affect maximum earthquake size, and that larger creeping patches are more likely to stop an earthquake. Because the geologic and geodetic creep distributions for the Bartlett Springs differ significantly, I was not able to make a definitive statement about maximum earthquake sizes on this fault. However, my results on the basic physics of how creeping patches affect rupture are still useful for assessing hazard on partially-creeping faults in general. We published this work in Geophysical Research Letters in 2015.

Lozos, J.C., Harris, R.A., Murray, J.R., & Lienkaemper, J.J. (2015). Dynamic rupture models of earthquakes on the Bartlett Springs Fault, Northern California. Geophysical Research Letters, 42(11), 4343-4349.

Lozos, J.C., Olsen, K. B., Brune, J.N., Takedatsu, R., Brune, R.J., & Oglesby, D D. (2015). Broadband ground motions from dynamic models of rupture on the northern San Jacinto fault, and comparison with precariously balanced rocks. Bulletin of the Seismological Society of America, 105(4), 1947-1960.

Lozos, J.C. (2016). A case for historic joint rupture of the San Andreas and San Jacinto faults. Science advances, 2(3), e1500621.

Last Modified: 10/12/2021
Modified by: Julian C Lozos