Generalized numerical techniques to absorb artificial grid boundary reflections for seismic wave modelling in arbitrary elastic, anisotropic media

  • Moosoo Woo
  • , Bing Zhou
  • , Stewart Greenhalgh
  • , Xu Liu

Research output: Contribution to conferencePaperpeer-review

1 Scopus citations

Abstract

In seismic wave modelling, the perfectly-matched layer (PML) technique is a popular numerical approach to absorb the artificial reflections from the edges of the modeling domain because it places no limitation on the nature of the medium. However, the time-domain PML technique requires recursive convolutional computations at each time step to obtain all the 1 st -order spatial derivatives. For the higher order spatial derivatives, it is not straightforward. In this paper, two alternatives to the PML technique are presented. One is a generalized version of the so-called 'Stiffness Reduction Method', and the other is named the 'Mass Reduction Method'. We show that both approaches simply employ complex-valued wavenumbers rather than complex-valued spatial coordinates as in the PML technique, so that seismic waves decay in the extensional zones and the artificial reflections from the computational edges are suppressed. We show that these two techniques require no additional calculations for the spatial derivatives, are applicable for both 1 st - and 2 nd -order wave equations and arbitrary elastic media, and are much easier to implement in computer coding than the PML.

Original languageBritish English
Pages4010-4014
Number of pages5
DOIs
StatePublished - 2019
Event88th Society of Exploration Geophysicists International Exposition and Annual Meeting, SEG 2018 - Anaheim, United States
Duration: 14 Oct 201819 Oct 2018

Conference

Conference88th Society of Exploration Geophysicists International Exposition and Annual Meeting, SEG 2018
Country/TerritoryUnited States
CityAnaheim
Period14/10/1819/10/18

Fingerprint

Dive into the research topics of 'Generalized numerical techniques to absorb artificial grid boundary reflections for seismic wave modelling in arbitrary elastic, anisotropic media'. Together they form a unique fingerprint.

Cite this