Conveners: Brittany Erickson (U of Oregon), Junle Jiang (Oklahoma), Valere Lambert (UCSC)
Dates: November 2, 2021 (online)
SCEC Award: 21139

SUMMARY: Our group ​is working to advance ​computational methods for simulating Sequences of Earthquakes and Aseismic Slip (SEAS) by conducting a suite of code verification exercises. ​In SEAS models, the goal is to capture the interplay of aseismic fault slip—that ultimately lead to earthquake nucleation—and dynamic earthquake ruptures themselves, in an effort to understand which physical factors control the full range of observables such as aseismic deformation, earthquake initiation locations, ground shaking during dynamic rupture, recurrence times and magnitudes of earthquakes.

A total of 60 people participated in the online SCEC SEAS Workshop on November 2, 2021, including scientists from the U.S.A., Sweden, France, Switzerland, China, Germany, England, Japan, Canada, Brazil, and New Zealand. Almost half of our workshop participants were graduate students or postdocs.

The goal of this workshop was to share some recent progress in SEAS modeling that considers fluid effects and increased frictional and geometric complexity and to introduce our next benchmark problems BP6 and BP7, which focus on fluid and 3D effects, respectively. Results were summarized from the most recent benchmarks BP3-QD, BP1-FD, and BP5 which focused on dipping fault geometries, full dynamic and 3D effects, respectively. The group also discussed research targets and plans for SEAS in the near and longer term for the group, following scientific talks presented by SEAS modelers.

Figure 1. Coseismic time series for BP3-QD across all modeling groups show qualitative (and in some cases quantitative) agreements with sufficiently large computational domain size. For volume-based codes, we found that results are sensitive to choices in boundary conditions truncating the semi-infinite domain, as well as computational choices in grid stretching.

SESSION 1. Brittany Erickson kicked off the workshop with a brief history of SEAS activities since its initiation in 2018 and an overview of the workshop schedule and goals. All participants were then invited to introduce themselves and briefly describe their interest in SEAS and the workshop. Brittany then provided an update on the current status of model results for the two 2D benchmarks: BP1-FD (the first fully dynamic problem) and BP3-QD (the first plane strain problem with a dipping fault). Quantitative agreements have been made for BP1-FD (Figure 1) for several modeling groups, while other participating modeling groups continue to update/upload new results. For BP3-QD, the on-fault time series matched well across codes, and discrepancies in off-fault surface stations have been addressed by resolving an ambiguity in the benchmark description. A draft of the paper presenting these results will be circulated among co-authors and modelers by the end of 2021.

Figure 2. Rupture contours for BP4-QD (whole space problem) show excellent quantitative agreement at suggested resolution (0.5 km) across participating codes. The high computational demand limits participating models at the suggested resolution (BEM/SBEM models), however good agreement is achieved for more models (BEM/SBEM/FEM/FDM) at 1 km resolution (not shown).

Next, Junle Jiang summarized highlights and lessons from the code comparison exercises for 3D SEAS benchmarks. A total of 10 modeling groups together explored how various computational and physical factors affect simulation outputs for all phases of multiple seismic cycles. The excellent agreement among best-resolved simulations lend confidence to participating numerical codes, offering best-practice examples to improve numerical simulations (Figure 2). These comparisons also reveal different sensitivities of model observables, which are important for integrating models with geophysical observations. The manuscript for 3D benchmarks is in review as of December 2021.

SESSION 2-3. Five science talks presented by Stacy Larochelle, Yuyun Yang, Sylvain Barbot, Pierre Romanet, and Duo Li (4 of which are early career scientists) focused on using SEAS modeling to understand the effects of fluids, friction, complex fault geometries and full dynamics on fault slip.

SESSION 4. During the afternoon session, Valère Lambert gave an overview of two new sets of benchmark problems targeted for 2022 (Figure 3). The first, BP6-QD-A/S/C, is a 2D quasi-dynamic problem of a 1-D fault with either constant or velocity-strengthening friction exhibiting an aseismic slip transient arising due to changes in effective normal stress from fluid injection and along-fault pore fluid diffusion. The second, BP7-QD/FD-A/S, is a 3D problem focusing on the interplay between aseismic slip and earthquake nucleation on a circular velocity-weakening asperity. BP6 and 7 have been modified based on feedback from group discussions, which includes using a smooth nucleation procedure for the initial rupture in BP7 and suggesting a finite domain size for volume-based methods.

Figure 3. Problem set-ups for new benchmarks. Left: BP6-QD-A/S/C is a 2D problem of fault with velocity-strengthening or constant friction with fluid injection and along-fault pore fluid diffusion. Right: 3D problem with a circular velocity-weakening asperity.

SESSION 5. To conclude the meeting, the workshop conveners led the planning discussions for SEAS activities in 2022, including possible future benchmark problems, how to best maximize participation in SEAS benchmarks, and the importance of high-performance computing in the advancement of code. The group also discussed possible collaborations between SEAS and other groups (e.g. RSQSim) in future activities.

The final discussion highlighted three general problems to address for future benchmarks: 1) a 2D antiplane problem like BP1 but including bulk viscoelastic deformation, 2) a 2D dipping fault problem like BP3 with depth-variable normal stress, and 3) a 2D problem including a low-velocity damage zone. Looking forward, the group discussed options for revamp of the online platform to improve specific tools and functionality.

Presentation videos and presentations may be viewed by clicking the links below. PLEASE NOTE: Files are the author’s property. They may contain unpublished or preliminary information and should only be used while viewing the talk. Only the presentations for which SCEC has received permission to post publicly are included below.

TUESDAY, NOVEMBER 2, 2021
All times below are Pacific Daylight Time (PDT or UTC-7).

It is SCEC policy to foster harassment-free environments wherever our science is conducted. By accepting an invitation to participate in a SCEC-supported event, by email or online registration, participants agree to abide by the SCEC Activities Code of Conduct.

PARTICIPANTS