Deep learning for fast simulation of seismic waves in complex media
<p>The simulation of seismic waves is a core task in many geophysical applications. Numerical methods such as finite difference (FD) modelling and spectral element methods (SEMs) are the most popular techniques for simulating seismic waves, but disadvantages such as their computational cost pr...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2020-08-01
|
| Series: | Solid Earth |
| Online Access: | https://se.copernicus.org/articles/11/1527/2020/se-11-1527-2020.pdf |
| _version_ | 1818333577746579456 |
|---|---|
| author | B. Moseley T. Nissen-Meyer A. Markham |
| author_facet | B. Moseley T. Nissen-Meyer A. Markham |
| author_sort | B. Moseley |
| collection | DOAJ |
| description | <p>The simulation of seismic waves is a core task in many geophysical applications. Numerical methods such as finite difference (FD) modelling and spectral element methods (SEMs) are the most popular techniques for simulating seismic waves, but disadvantages such as their computational cost prohibit their use for many tasks. In this work, we investigate the potential of deep learning for aiding seismic simulation in the solid Earth sciences. We present two deep neural networks which are able to simulate the seismic response at multiple locations in horizontally layered and faulted 2-D acoustic media an order of magnitude faster than traditional finite difference modelling. The first network is able to simulate the seismic response in horizontally layered media and uses a WaveNet network architecture design. The second network is significantly more general than the first and is able to simulate the seismic response in faulted media with arbitrary layers, fault properties and an arbitrary location of the seismic source on the surface of the media, using a conditional autoencoder design. We test the sensitivity of the accuracy of both networks to different network hyperparameters and show that the WaveNet network can be retrained to carry out fast seismic inversion in the same media. We find that are there are challenges when extending our methods to more complex, elastic and 3-D Earth models; for example, the accuracy of both networks is reduced when they are tested on models outside of their training distribution. We discuss further research directions which could address these challenges and potentially yield useful tools for practical simulation tasks.</p> |
| first_indexed | 2024-12-13T13:53:51Z |
| format | Article |
| id | doaj.art-24edb63e78bc40069c0fd9e9bd40e9ef |
| institution | Directory Open Access Journal |
| issn | 1869-9510 1869-9529 |
| language | English |
| last_indexed | 2024-12-13T13:53:51Z |
| publishDate | 2020-08-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | Solid Earth |
| spelling | doaj.art-24edb63e78bc40069c0fd9e9bd40e9ef2022-12-21T23:42:59ZengCopernicus PublicationsSolid Earth1869-95101869-95292020-08-01111527154910.5194/se-11-1527-2020Deep learning for fast simulation of seismic waves in complex mediaB. Moseley0T. Nissen-Meyer1A. Markham2Department of Computer Science, University of Oxford, Oxford, UKDepartment of Earth Sciences, University of Oxford, Oxford, UKDepartment of Computer Science, University of Oxford, Oxford, UK<p>The simulation of seismic waves is a core task in many geophysical applications. Numerical methods such as finite difference (FD) modelling and spectral element methods (SEMs) are the most popular techniques for simulating seismic waves, but disadvantages such as their computational cost prohibit their use for many tasks. In this work, we investigate the potential of deep learning for aiding seismic simulation in the solid Earth sciences. We present two deep neural networks which are able to simulate the seismic response at multiple locations in horizontally layered and faulted 2-D acoustic media an order of magnitude faster than traditional finite difference modelling. The first network is able to simulate the seismic response in horizontally layered media and uses a WaveNet network architecture design. The second network is significantly more general than the first and is able to simulate the seismic response in faulted media with arbitrary layers, fault properties and an arbitrary location of the seismic source on the surface of the media, using a conditional autoencoder design. We test the sensitivity of the accuracy of both networks to different network hyperparameters and show that the WaveNet network can be retrained to carry out fast seismic inversion in the same media. We find that are there are challenges when extending our methods to more complex, elastic and 3-D Earth models; for example, the accuracy of both networks is reduced when they are tested on models outside of their training distribution. We discuss further research directions which could address these challenges and potentially yield useful tools for practical simulation tasks.</p>https://se.copernicus.org/articles/11/1527/2020/se-11-1527-2020.pdf |
| spellingShingle | B. Moseley T. Nissen-Meyer A. Markham Deep learning for fast simulation of seismic waves in complex media Solid Earth |
| title | Deep learning for fast simulation of seismic waves in complex media |
| title_full | Deep learning for fast simulation of seismic waves in complex media |
| title_fullStr | Deep learning for fast simulation of seismic waves in complex media |
| title_full_unstemmed | Deep learning for fast simulation of seismic waves in complex media |
| title_short | Deep learning for fast simulation of seismic waves in complex media |
| title_sort | deep learning for fast simulation of seismic waves in complex media |
| url | https://se.copernicus.org/articles/11/1527/2020/se-11-1527-2020.pdf |
| work_keys_str_mv | AT bmoseley deeplearningforfastsimulationofseismicwavesincomplexmedia AT tnissenmeyer deeplearningforfastsimulationofseismicwavesincomplexmedia AT amarkham deeplearningforfastsimulationofseismicwavesincomplexmedia |