Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center
Machineâ€learning algorithms continue to show promise in their application to seismic processing. The U.S. Geological Survey National Earthquake Information Center (NEIC) is exploring the adoption of these tools to aid in simultaneous local, regional, and global realâ€time earthquake monitoring. As...
Published in: | Seismological Research Letters |
---|---|
Main Authors: | , , , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | unknown |
Published: |
Seismological Society of America
2021
|
Subjects: | |
Online Access: | https://doi.org/10.1785/0220200178 |
id |
ftcaltechauth:oai:authors.library.caltech.edu:tzkpz-nkf22 |
---|---|
record_format |
openpolar |
spelling |
ftcaltechauth:oai:authors.library.caltech.edu:tzkpz-nkf22 2024-06-23T07:56:44+00:00 Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center Yeck, William Luther Patton, John M. Ross, Zachary E. Hayes, Gavin P. Guy, Michelle R. Ambruz, Nick B. Shelly, David R. Benz, Harley M. Earle, Paul S. 2021-01 https://doi.org/10.1785/0220200178 unknown Seismological Society of America https://doi.org/10.1785/0220200178 oai:authors.library.caltech.edu:tzkpz-nkf22 eprintid:107938 resolverid:CaltechAUTHORS:20210205-145940821 info:eu-repo/semantics/openAccess Other Seismological Research Letters, 92(1), 469-480, (2021-01) info:eu-repo/semantics/article 2021 ftcaltechauth https://doi.org/10.1785/0220200178 2024-06-12T04:22:29Z Machineâ€learning algorithms continue to show promise in their application to seismic processing. The U.S. Geological Survey National Earthquake Information Center (NEIC) is exploring the adoption of these tools to aid in simultaneous local, regional, and global realâ€time earthquake monitoring. As a first step, we describe a simple framework to incorporate deepâ€learning tools into NEIC operations. Automatic seismic arrival detections made from standard picking methods (e.g., shortâ€term average/longâ€term average [STA/LTA]) are fed to trained neural network models to improve automatic seismicâ€arrival (pick) timing and estimate seismicâ€arrival phase type and sourceâ€station distances. These additional data are used to improve the capabilities of the NEIC associator. We compile a dataset of 1.3 million seismicâ€phase arrivals that represent a globally distributed set of sourceâ€station paths covering a range of phase types, magnitudes, and source distances. We train three separate convolutional neural network models to predict arrival time onset, phase type, and distance. We validate the performance of the trained networks on a subset of our existing dataset and further extend validation by exploring the model performance when applied to NEIC automatic pick data feeds. We show that the information provided by these models can be useful in downstream event processing, specifically in seismicâ€phase association, resulting in reduced false associations and improved location estimates. © 2021 Seismological Society of America. Manuscript received 8 May 2020. Published online 23 September 2020. The facilities of Incorporated Research Institutions for Seismology (IRIS) Data Services, and specifically the IRIS Data Management Center, were used for access to waveforms used in this study. IRIS Data Services are funded through the Seismological Facilities for the Advancement of Geoscience and EarthScope (SAGE) Proposal of the National Science Foundation under Cooperative Agreement EARâ€1261681. Waveform data from 136 ... Article in Journal/Newspaper Seismological Facilities for the Advancement of Geoscience and EarthScope Caltech Authors (California Institute of Technology) Seismological Research Letters 92 1 469 480 |
institution |
Open Polar |
collection |
Caltech Authors (California Institute of Technology) |
op_collection_id |
ftcaltechauth |
language |
unknown |
description |
Machineâ€learning algorithms continue to show promise in their application to seismic processing. The U.S. Geological Survey National Earthquake Information Center (NEIC) is exploring the adoption of these tools to aid in simultaneous local, regional, and global realâ€time earthquake monitoring. As a first step, we describe a simple framework to incorporate deepâ€learning tools into NEIC operations. Automatic seismic arrival detections made from standard picking methods (e.g., shortâ€term average/longâ€term average [STA/LTA]) are fed to trained neural network models to improve automatic seismicâ€arrival (pick) timing and estimate seismicâ€arrival phase type and sourceâ€station distances. These additional data are used to improve the capabilities of the NEIC associator. We compile a dataset of 1.3 million seismicâ€phase arrivals that represent a globally distributed set of sourceâ€station paths covering a range of phase types, magnitudes, and source distances. We train three separate convolutional neural network models to predict arrival time onset, phase type, and distance. We validate the performance of the trained networks on a subset of our existing dataset and further extend validation by exploring the model performance when applied to NEIC automatic pick data feeds. We show that the information provided by these models can be useful in downstream event processing, specifically in seismicâ€phase association, resulting in reduced false associations and improved location estimates. © 2021 Seismological Society of America. Manuscript received 8 May 2020. Published online 23 September 2020. The facilities of Incorporated Research Institutions for Seismology (IRIS) Data Services, and specifically the IRIS Data Management Center, were used for access to waveforms used in this study. IRIS Data Services are funded through the Seismological Facilities for the Advancement of Geoscience and EarthScope (SAGE) Proposal of the National Science Foundation under Cooperative Agreement EARâ€1261681. Waveform data from 136 ... |
format |
Article in Journal/Newspaper |
author |
Yeck, William Luther Patton, John M. Ross, Zachary E. Hayes, Gavin P. Guy, Michelle R. Ambruz, Nick B. Shelly, David R. Benz, Harley M. Earle, Paul S. |
spellingShingle |
Yeck, William Luther Patton, John M. Ross, Zachary E. Hayes, Gavin P. Guy, Michelle R. Ambruz, Nick B. Shelly, David R. Benz, Harley M. Earle, Paul S. Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center |
author_facet |
Yeck, William Luther Patton, John M. Ross, Zachary E. Hayes, Gavin P. Guy, Michelle R. Ambruz, Nick B. Shelly, David R. Benz, Harley M. Earle, Paul S. |
author_sort |
Yeck, William Luther |
title |
Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center |
title_short |
Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center |
title_full |
Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center |
title_fullStr |
Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center |
title_full_unstemmed |
Leveraging Deep Learning in Global 24/7 Real-Time Earthquake Monitoring at the National Earthquake Information Center |
title_sort |
leveraging deep learning in global 24/7 real-time earthquake monitoring at the national earthquake information center |
publisher |
Seismological Society of America |
publishDate |
2021 |
url |
https://doi.org/10.1785/0220200178 |
genre |
Seismological Facilities for the Advancement of Geoscience and EarthScope |
genre_facet |
Seismological Facilities for the Advancement of Geoscience and EarthScope |
op_source |
Seismological Research Letters, 92(1), 469-480, (2021-01) |
op_relation |
https://doi.org/10.1785/0220200178 oai:authors.library.caltech.edu:tzkpz-nkf22 eprintid:107938 resolverid:CaltechAUTHORS:20210205-145940821 |
op_rights |
info:eu-repo/semantics/openAccess Other |
op_doi |
https://doi.org/10.1785/0220200178 |
container_title |
Seismological Research Letters |
container_volume |
92 |
container_issue |
1 |
container_start_page |
469 |
op_container_end_page |
480 |
_version_ |
1802650050213445632 |