Continuum mechanics-based numerical simulations are potential methods for evaluating surface fault displacements. We developed a parallel finite element method to evaluate these displacements. In this study, we applied a numerical method to the simulation of the 2014 Nagano-ken-hokubu Earthquake, in which surface faulting was observed. We modeled a 5 km × 5 km × 1 km domain around the northernmost region of the surface faults, which included secondary faults. We applied forced displacements on the bottom surface of the model based on the slip distribution on the primary fault and the elastic theory of dislocations. As the input slip increased, surface slips appeared on the primary fault and a secondary fault. The calculated surface slips were in good agreement with the measured values.

1 Introduction

Since the occurrence of massive earthquakes in Taiwan and Turkey in 1999, there have been growing concerns about the potential damage of various infrastructures and buildings caused by surface fault ruptures. For on-site fault assessment in nuclear power plants, it is important to estimate fault displacements and their impact on the safety functions of the facilities. It is necessary to reliably estimate the possible displacements.

Numerical simulations based on continuum mechanics are potential evaluation methods for surface fault displacements. However, there are major difficulties in simulating the fault rupture process. One difficulty is that it requires a large amount of numerical computations to simulate the fault rupture process of a target area of only a few hundreds of meters. Another difficulty is the loss of stability in the initial boundary value problem to which the numerical analysis is applied. Stability implies that a solution does not change when a small disturbance is added to the problem, and stability loss leads to drastic changes in the solution due to small disturbances. We overcame these difficulties by applying high performance computing methods. We developed a finite element method involving two functions: 1) a symplectic time integration explicit scheme to properly conserve the energy of the system and 2) rigorously formulated high-order joint elements. The finite element method was enhanced with parallel computing capabilities (Sawada et al., 2017).

This content is only available via PDF.
You can access this article if you purchase or spend a download.