JpGU-AGU Joint Meeting 2017

Presentation information

[EE] Oral

S (Solid Earth Sciences) » S-CG Complex & General

[S-CG62] [EE] Dynamics in mobile belts

Wed. May 24, 2017 10:45 AM - 12:15 PM Convention Hall B (International Conference Hall 2F)

convener:Yukitoshi Fukahata(Disaster Prevention Research Institute, Kyoto University), Robert Holdsworth(Durham University), Jeanne Hardebeck(USGS), Hikaru Iwamori(Geochemical Evolution Research Program, Japan Agency for Marine-Earth Science and Technology), Chairperson:Bunichiro Shibazaki(International Institute of Seismology and Earthquake Engineering, Building Research Institute), Chairperson:Satoshi Matsumoto(Institute of Seismology and Volcanology, Faculty of Sciences, Kyushu University)

10:45 AM - 11:00 AM

[SCG62-17] Fast Scalable Finite Element Analysis Method for Crustal Deformation using a 3D Island-scale High-fidelity Model

★Invited papers

*Tsuyoshi Ichimura1 (1.The Univeristy of Tokyo)

Keywords:crust deformation, finite element method, fast scalable parallel computation

Recently, three-dimensional high-fidelity crustal structure data in island-scale domains are becoming available due to the accumulation of high-resolution observation data. However, it has been difficult to use such data to perform crustal deformation analysis in an island-scale domain with quality assurance of the numerical simulation because of its huge analysis cost.
We propose a high-fidelity crustal structure finite element (FE) model construction method using high-fidelity crustal structure data, and a fast FE analysis method to reduce its huge analysis cost. These methods used an automatic FE model generation method for parallel computation, MPI and OpenMP hybrid parallel computation on a distributed memory super-computer, a geometric multigrid, variable preconditioning, and multiple-precision arithmetic. Using the proposed methods, we constructed 10 billion DOF high-fidelity crustal structure FE models including the whole Japanese Islands, and conducted elastic/viscoelastic crustal deformation analysis using this model with quality assurance of numerical simulation.
The proposed method can be applied for larger crust deformation problems and extended to nonlinear/dynamic problems. As illustrative examples, we present a crust deformation analysis with 2.05 trillion DOF and an implicit nonlinear wave analysis with 1.08 trillion DOF on 0.270 trillion unstructured finite elements.