Community Code Verification Exercises for Simulations of Earthquake Sequences and Aseismic Slip (SEAS): Dynamic Effects and Dipping Fault Geometries

Brittany A. Erickson, Junle Jiang, Mohamed Abdelmeguid, Martin Almquist, Jean-Paul Ampuero, Ryosuke Ando, Sylvain D. Barbot, Camilla Cattania, Alexandre Chen, Luca Dal Zilio, Eric M. Dunham, Ahmed E. Elbanna, Alice-Agnes Gabriel, Tobias Harvey, Yihe Huang, Yoshihiro Kaneko, Jeremy E. Kozdon, Valere R. Lambert, Nadia Lapusta, Duo Li, Meng Li, Chao Liang, Yajing Liu, So Ozawa, Casper Pranger, Paul Segall, Yudong Sun, Prithvi Thakur, Carsten Uphoff, Ylona van Dinther, & Yuyun Yang

Submitted August 14, 2021, SCEC Contribution #11412, 2021 SCEC Annual Meeting Poster #145

The use of numerical modeling to advance our understanding of earthquake dynamics and inform seismic hazard assessment relies fundamentally on credible, reproducible model results. Single-event dynamic rupture simulations have been extensively used to explore earthquake rupture behavior; however, they are inherently limited to the short timescales of wave propagation. Earthquake simulators were developed to model earthquake sequences on millennial timescales and large-scale fault networks, but to make such simulations computationally tractable, adopt simplification of some physical effects that could potentially dominate earthquake and fault interaction. An alternative and complementary modeling framework are simulations of Sequences of Earthquakes and Aseismic Slip (SEAS). SEAS models focus on smaller, regional-scale fault zones with the goal of understanding what physical factors control observations of slow slip, microseismicity, the initiation, rupture, magnitudes, and recurrence intervals of large earthquakes, as well as ground shaking and aseismic movement.

Here, we present results from two new benchmark problems that consider SEAS models addressing fundamental aspects in realistic earthquake source modeling: Full elastodynamics for BP1-FD and dipping faults in BP3-QD. A total of 9 and 10 modeling groups participated, respectively. Both problems consider a planar, rate-and-state frictional fault embedded in a homogeneous, linear elastic half-space with a free surface. BP1-FD considers antiplane shear motion with full inertial effects and BP3-QD is a quasi-dynamic plane-strain problem for dipping thrust and normal faults. These benchmarks required many modeling groups to further develop their codes to handle the computational challenges introduced by these added complexities. For example, many codes for BP1-FD were faced with the challenge of how to couple vastly different quasi-static and dynamic numerical solvers to efficiently simulate the interseismic and coseismic periods. For BP3-QD, we found that, in addition to resolving normal stress changes associated with dipping fault geometries, exploration of the domain-size dependency of simulations came at a greater computational cost. As we worked to overcome these challenges, however, we were able to find good quantitative agreement across codes in long-term time-series, recurrence times, and co-seismic rupture features, providing confidence in SEAS model outcomes in more realistic settings.

Citation
Erickson, B. A., Jiang, J., Abdelmeguid, M., Almquist, M., Ampuero, J., Ando, R., Barbot, S. D., Cattania, C., Chen, A., Dal Zilio, L., Dunham, E. M., Elbanna, A. E., Gabriel, A., Harvey, T., Huang, Y., Kaneko, Y., Kozdon, J. E., Lambert, V. R., Lapusta, N., Li, D., Li, M., Liang, C., Liu, Y., Ozawa, S., Pranger, C., Segall, P., Sun, Y., Thakur, P., Uphoff, C., van Dinther, Y., & Yang, Y. (2021, 08). Community Code Verification Exercises for Simulations of Earthquake Sequences and Aseismic Slip (SEAS): Dynamic Effects and Dipping Fault Geometries. Poster Presentation at 2021 SCEC Annual Meeting.


Related Projects & Working Groups
Fault and Rupture Mechanics (FARM)