Accelerating anisotropic elastic subsurface imaging with dual-pair finite differences
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Anisotropic elastic full waveform inversion based on finite difference modeling on fully staggered grids like the Lebedev grid comes with a high computational cost. We examine using a non-staggered finite difference method instead, as a possibility to accelerate elastic subsurface imaging based on fully staggered grids. Most existing studies on finite difference elastic modeling on non-staggered grids are limited to modeling waves sampled above 4 points per P wavelength. This makes them relatively inefficient for subsurface imaging using P waves, compared to the Lebedev grid which can accurately model waves sampled below 3 points per P wavelength for a suitable stencil width. We show that accurate finite difference wave modeling on non-staggered grids below 3 points per P wavelength can be achieved by a combination of dual pairing,selective filtering, and some adjustments to source injection. Preliminary tests on field data show that the proposed methodology can lead to roughly 3 times faster subsurface elastic imaging using P waves, compared to using the Lebedev grid for the same sampling rate, even below 3 points per P wavelengths.
Fields of Study & Subjects
Seismic Imaging and Inversion Techniques
Seismic Waves and Analysis
Advanced Numerical Methods in Computational Mathematics
Publication details
- Journal / source
- Second International Meeting for Applied Geoscience & Energy
- Volume
- 1
- Pages
- 825–829
- Publication date
- 10 February 2026
- Language
- EN