Community Code Verication Exercises for Simulations of Earthquake Sequences and Aseismic Slip (SEAS): 3D Effects, Fully Dynamic Ruptures, and Dipping Fault Geometries

Numerical simulations of Sequences of Earthquakes and Aseismic Slip (SEAS) have rapidly progressed over recent decades to address important questions in earthquake physics and fault mechanics. Challenges in SEAS modeling remain in resolving the multiscale interactions between slow slip, earthquake nucleation, and dynamic rupture; and understanding physical factors that control observables such as seismicity and deformation. To advance SEAS simulations with rigor and reproducibility, we pursue community efforts involving researchers around the globe to verify numerical codes in a growing suite of SEAS benchmarks. We achieved an excellent agreement of simulations from 11 modeling groups of quasi-dynamic earthquake sequences on a 1D vertical fault obeying rate-and-state friction in a 2D elastic half-space (Erickson et al., 2020, Seismol. Res. Lett.). This exercise shows that the recurrence intervals and statistics of simulated earthquakes are robustly reproduced only if we use large computational domain sizes and small grid spacing. Here, we highlight code comparison results from three recent SEAS benchmarks that incorporate additional important ingredients: 3D effects, fully dynamic ruptures, and dipping fault geometries. In these more complicated problems, the increased computational cost often prohibits strict convergence tests for many codes; code comparisons thus provide an alternative, valuable way to provide confidence in simulation results. We use simulations from multiple modeling groups to explore how various numerical factors, such as computational domain sizes, boundary conditions, and numerical resolution, affect interseismic fault stressing, earthquake recurrence, and coseismic rupture process. Comparisons of these simulations allow us to assess relative sensitivities of different geophysical observables to numerical factors. These community exercises will guide the development of more realistic SEAS models and their integration with geophysical observations, contributing to a better understanding of earthquake system dynamics.