Japan Geoscience Union Meeting 2023

Presentation information

[J] Online Poster

S (Solid Earth Sciences ) » S-TT Technology & Techniques

[S-TT43] Creating future of solid Earth science with high performance computing (HPC)

Tue. May 23, 2023 10:45 AM - 12:15 PM Online Poster Zoom Room (17) (Online Poster)

convener:Takane Hori(Japan Agency for Marine-Earth Science and Technology), Yuji Yagi(Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba), Katsuhiko Shiomi(National Research Institute for Earth Science and Disaster Resilience), Takanori Matsuzawa(National Research Institute for Earth Science and Disaster Resilience)

On-site poster schedule(2023/5/22 17:15-18:45)

10:45 AM - 12:15 PM

[STT43-P04] Development and application of large scale numerical simulations for earthquake generation, ground motion and non-linear soil amplification

*Takane Hori1, Tsuyoshi Ichimura2, Kohei Fujita2 (1.Japan Agency for Marine-Earth Science and Technology, 2.Earthquake Research Institute, The University of Tokyo)

Keywords:earthquake hazard, nonlinear dynamic finite element method, comprehensive analysis

Quantitative evaluation of earthquake hazards is necessary to prepare for earthquakes occurring directly beneath the Tokyo metropolitan area and in the Nankai Trough. In this project, we will construct an integrated prediction system to carry out such evaluations by using extremely large-scale computations at Fugaku, implement it in the process of national evaluations of seismic hazards, and introduce the concept of quality assurance into numerical simulations for earthquake generation forecasting to lay the foundation for its use in policy decisions. To realize an earthquake simulation using an ultra-high-resolution fault-structure system model, we developed an implicit solver that scales up to 152K computation nodes and conducted an ultra-large-scale analysis using nonlinear dynamic finite element method with three-dimensional low-order unstructured elements. The performance of the proposed method on Fugaku was 25.45 times faster than the conventional state-of-the-art method previously developed for the K computer. The difference in analysis capability of the two methods was 1,070 times when hardware performance was also taken into account. This has made it possible to perform a comprehensive analysis including faults to structures, which is one of the dreams of earthquake engineering and could not be achieved even using the K computer.