Validation of fault displacements from dynamic rupture simulations against the observations from the 1992 Landers earthquake

Yongfei Wang, & Christine A. Goulet

Published March 15, 2021, SCEC Contribution #10961

Coseismic fault displacements in large earthquakes have caused significant damage to structures and lifelines located on or near fault lines. Examples include severely damaged bridges, dams and tunnels in the 1999 Chi-Chi earthquake, pancaked buildings in the 1999 Kocaeli earthquake, and ruptured water lines during the 2019 Ridgecrest Earthquake Sequence. Fault displacements, therefore, represent an important seismic hazard, especially for distributed infrastructure systems that may cross faults in multiple locations. Predictive fault displacement models are sparse and poorly constrained partly due to the scarcity of detailed fault displacement observations. Advancements in physics-based dynamic rupture simulation methods make them an attractive approach to address this important problem. Dynamic ruptures construct the earthquake rupture evolution and propagate seismic waves under plausible physical conditions and they can produce fault displacement through physical causative parameters. When used with appropriate constitutive models to characterize the bulk medium around the fault, dynamic rupture models can also capture off-fault distributed inelastic fault displacements. In our project, we first validate a suite of models against well-documented case histories, and we then extend the simulations to predict displacements for scenarios and events we have not yet experienced. Our initial validation is based on the 1992 Landers M7.3 earthquake for which we compare simulation results to on- and off-fault displacement data. We also perform a validation of ground motions against recordings from the event, to ensure that all physics important to modeling have been properly parameterized and that the results are defensible. In the predictive step, we aim to provide a range of expected displacements that capture several aspects of modeling uncertainty. Hence the validation is performed using a similar approach and parameter ranges that are consistent with observations are selected. For the Landers scenario, we developed an ensemble of dynamic rupture models with varying imposed stresses and geometrical fault roughness profiles, resulting in a suite of 48 simulation datasets. We present the validation results and provide insight on the uncertainty of resulting ruptures including their magnitude, length, propagation velocity and displacement distributions.

Citation
Wang, Y., & Goulet, C. A. (2021). Validation of fault displacements from dynamic rupture simulations against the observations from the 1992 Landers earthquake. Bulletin of the Seismological Society of America,. doi: 10.1785/0120210082.