EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking
Abstract In this study, we build a multi‐station phase‐picking model named EdgePhase by integrating an Edge Convolutional module with a state‐of‐the‐art single‐station phase‐picking model, EQTransformer. The Edge Convolutional module, a variant of Graph Neural Network, exchanges information relevant...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-11-01
|
| Series: | Geochemistry, Geophysics, Geosystems |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2022GC010453 |
| _version_ | 1797637961329672192 |
|---|---|
| author | Tian Feng Saeed Mohanna Lingsen Meng |
| author_facet | Tian Feng Saeed Mohanna Lingsen Meng |
| author_sort | Tian Feng |
| collection | DOAJ |
| description | Abstract In this study, we build a multi‐station phase‐picking model named EdgePhase by integrating an Edge Convolutional module with a state‐of‐the‐art single‐station phase‐picking model, EQTransformer. The Edge Convolutional module, a variant of Graph Neural Network, exchanges information relevant to seismic phases between neighboring stations. In EdgePhase, seismograms are first encoded into the latent representations, then converted into enhanced representations by the Edge Convolutional module, and finally decoded into the P‐ and S‐phase probabilities. Compared to the standard EQTransformer, EdgePhase increases the precision (fraction of phase identifications that are real) and recall (fraction of phase arrivals that are identified) rate by 5% on our training and test data sets of Southern California earthquakes. To evaluate its performance in regions of different tectonic settings, we applied EdgePhase to detect the early aftershocks following the 2020 M7.0 Samos, Greece earthquake. Compared to a local earthquake catalog, EdgePhase produced 190% additional detections with an event distribution more conformative to a planar fault interface, suggesting higher fidelity in event locations. This case study indicates that EdgePhase provides a strong regional generalization capability in real‐world applications. |
| first_indexed | 2024-03-11T12:56:43Z |
| format | Article |
| id | doaj.art-a05461ad93d44301ac640cd64f4fea34 |
| institution | Directory Open Access Journal |
| issn | 1525-2027 |
| language | English |
| last_indexed | 2024-03-11T12:56:43Z |
| publishDate | 2022-11-01 |
| publisher | Wiley |
| record_format | Article |
| series | Geochemistry, Geophysics, Geosystems |
| spelling | doaj.art-a05461ad93d44301ac640cd64f4fea342023-11-03T17:00:48ZengWileyGeochemistry, Geophysics, Geosystems1525-20272022-11-012311n/an/a10.1029/2022GC010453EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase PickingTian Feng0Saeed Mohanna1Lingsen Meng2Earth, Planetary and Space Sciences University of California Los Angeles Los Angeles CA USAEarth, Planetary and Space Sciences University of California Los Angeles Los Angeles CA USAEarth, Planetary and Space Sciences University of California Los Angeles Los Angeles CA USAAbstract In this study, we build a multi‐station phase‐picking model named EdgePhase by integrating an Edge Convolutional module with a state‐of‐the‐art single‐station phase‐picking model, EQTransformer. The Edge Convolutional module, a variant of Graph Neural Network, exchanges information relevant to seismic phases between neighboring stations. In EdgePhase, seismograms are first encoded into the latent representations, then converted into enhanced representations by the Edge Convolutional module, and finally decoded into the P‐ and S‐phase probabilities. Compared to the standard EQTransformer, EdgePhase increases the precision (fraction of phase identifications that are real) and recall (fraction of phase arrivals that are identified) rate by 5% on our training and test data sets of Southern California earthquakes. To evaluate its performance in regions of different tectonic settings, we applied EdgePhase to detect the early aftershocks following the 2020 M7.0 Samos, Greece earthquake. Compared to a local earthquake catalog, EdgePhase produced 190% additional detections with an event distribution more conformative to a planar fault interface, suggesting higher fidelity in event locations. This case study indicates that EdgePhase provides a strong regional generalization capability in real‐world applications.https://doi.org/10.1029/2022GC010453deep learningseismic phase pickinggraph neural networkedge convolutionSamos earthquakeEdgePhase |
| spellingShingle | Tian Feng Saeed Mohanna Lingsen Meng EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking Geochemistry, Geophysics, Geosystems deep learning seismic phase picking graph neural network edge convolution Samos earthquake EdgePhase |
| title | EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking |
| title_full | EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking |
| title_fullStr | EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking |
| title_full_unstemmed | EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking |
| title_short | EdgePhase: A Deep Learning Model for Multi‐Station Seismic Phase Picking |
| title_sort | edgephase a deep learning model for multi station seismic phase picking |
| topic | deep learning seismic phase picking graph neural network edge convolution Samos earthquake EdgePhase |
| url | https://doi.org/10.1029/2022GC010453 |
| work_keys_str_mv | AT tianfeng edgephaseadeeplearningmodelformultistationseismicphasepicking AT saeedmohanna edgephaseadeeplearningmodelformultistationseismicphasepicking AT lingsenmeng edgephaseadeeplearningmodelformultistationseismicphasepicking |