1887

Abstract

Summary

Seismic modeling in anisotropic media is computationally intensive, necessitating efficient numerical strategies to meet modern survey demands. This study investigates half-precision (FP16) arithmetic to accelerate 3D elastic wave propagation in elastic anisotropic media. We employed GPU-accelerated finite-difference time-domain (FDTD) method on staggered grids to optimize the 3D elastic wave equation, ensuring numerical stability in FP16 arithmetic. The results demonstrate that FP16 arithmetic reduces the overall computational overhead while preserving sufficient numerical accuracy for resource-intensive applications such as full-waveform inversion (FWI) and reverse time migration (RTM). Future work will validate the framework across diverse anisotropic models and explore adjoint-based FWI and BF16 formats. By leveraging mixed-precision computing and modern hardware, this approach offers a scalable solution for large-scale seismic modeling, balancing efficiency and accuracy for geophysical exploration.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.2025643014
2025-10-06
2026-02-13

  • From This Site

    /content/papers/10.3997/2214-4609.2025643014
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Fabien-Ouellet, G. (2020). Seismic modeling and inversion using half-precision floating-point numbers. Geophysics, 85(3), F65–F76. Available via SEG Library: https://library.seg.org/doi/10.1190/geo2018-0760.1 (institutional access may be required).
     [Google Scholar]
  2. Clapp, R. G., et al. (2010). Bit-width reduction as an optimization strategy for reverse time migration. Likely in Geophysics or SEG Technical Program Expanded Abstracts. Search on SEG Library: https://library.seg.org or Google Scholar: https://scholar.google.com with “Clapp 2010 reverse time migration bit-width”.
     [Google Scholar]
  3. Medeiros, W. E., et al. (2013). Precision reduction in reverse time migration. Likely in Geophysics or SEG/EAGE proceedings. Search on SEG Library: https://library.seg.org or Google Scholar: https://scholar.google.com with “Medeiros 2013 reverse time migration precision”.
     [Google Scholar]
  4. Micikevicius, P., 2009, 3D finite difference computation on GPUs using CUDA: Proceedings of 2nd Workshop on General Purpose Processing on Graphics Processing Units, 79–84.
     [Google Scholar]
  5. Komatitsch, D., G.Erlebacher, D.Göddeke, and D.Michéa, 2010, High-order finite-element seismic wave propagation modeling with MPI on a large GPU cluster: Journal of Computational Physics, 229, 7692–7714, doi: 10.1016/j.jcp.2010.06.024.
     https://doi.org/10.1016/j.jcp.2010.06.024 [Google Scholar]
  6. Sethi, Harpreet et al., “Modeling 3-D anisotropic elastodynamics using mimetic finite differences and fully staggered grids.” Computational Geosciences27.5 (2023): 793–804.
     [Google Scholar]
  7. Virieux, J., 1986, P-wave propagation in heterogeneous media: Velocity-stress finite-difference method: Geophysics, 51, 889–901, doi: 10.1190/1.14421.7.
     https://doi.org/10.1190/1.14421.7 [Google Scholar]
/content/papers/10.3997/2214-4609.2025643014
Loading
Half-Precision Seismic Modeling in 3D Anisotropic Media
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.2025643014
/content/papers/10.3997/2214-4609.2025643014
/content/papers/10.3997/2214-4609.2025643014
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error