Numerical Studies of Seismic Patterns and Failure Process in Relation to Heterogeneity of Rocks

Abstract eng:
Seismic pattern is one of the most promising subjects in earthquake prediction studies because it generates information about physical conditions such as the degree of heterogeneity in the source region of an impending large earthquake [1]. According to the statistical studies in nature, there exist three types of seismic patterns: (1) main shock-aftershock type, (2) foreshock-main shock-aftershock type, and (3) swarm type. The foreshock as a precursor can be used to predict a major earthquake. However, what makes the seismic patterns different? Many scientists investigate the seismic patterns through experimental or theoretical method [2]. However, the theoretical models are difficult to deal with the rocks with some discontinuities. Almost no convenient experimental method is available to obtain the stress field information and the progressive failure process image in heterogeneous rocks. Therefore, the physical mechanism of precursory occurrence is difficult to be fully understood. Numerical simulation can provide much information regarding the stress distribution and the failure-induced stress redistribution. In this work, a series of numerical model tests were performed to investigate the effects of the heterogeneity of rocks on the rock failure and induced seismic patterns using the Rock Failure Process Analysis code (RFPA2D). It suggested that rock failure with the different heterogeneity produces different seismic sequence types, similar to the above-mentioned statistical results in nature. That is, type (1) occurs in more homogeneous cases, (2) in moderately heterogeneous cases, and (3) in extremely heterogeneous cases. Besides, the numerical simulation displays that the non-linear behavior of rock deformation and failure mode are clearly influenced by the heterogeneity of rocks. With the increase of the degree of rock homogeneity, the macroscopic failure process presents obvious brittle behavior.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket