A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland
SUMMARY Despite advanced seismological techniques, automatic source characterization for microseismic earthquakes remains difficult and challenging since current inversion and modelling of high-frequency signals are complex and time consuming. For real-time applications such as induced seismicity mo...
Published in: | Geophysical Journal International |
---|---|
Main Authors: | , , , , , , |
Other Authors: | , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Oxford University Press (OUP)
2021
|
Subjects: | |
Online Access: | http://dx.doi.org/10.1093/gji/ggab511 https://academic.oup.com/gji/advance-article-pdf/doi/10.1093/gji/ggab511/41810957/ggab511.pdf https://academic.oup.com/gji/article-pdf/229/2/999/42334927/ggab511.pdf |
id |
croxfordunivpr:10.1093/gji/ggab511 |
---|---|
record_format |
openpolar |
spelling |
croxfordunivpr:10.1093/gji/ggab511 2024-04-07T07:53:28+00:00 A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland Nooshiri, Nima Bean, Christopher J Dahm, Torsten Grigoli, Francesco Kristjánsdóttir, Sigríður Obermann, Anne Wiemer, Stefan European Commission Icelandic Centre for Research 2021 http://dx.doi.org/10.1093/gji/ggab511 https://academic.oup.com/gji/advance-article-pdf/doi/10.1093/gji/ggab511/41810957/ggab511.pdf https://academic.oup.com/gji/article-pdf/229/2/999/42334927/ggab511.pdf en eng Oxford University Press (OUP) https://creativecommons.org/licenses/by/4.0/ Geophysical Journal International volume 229, issue 2, page 999-1016 ISSN 0956-540X 1365-246X Geochemistry and Petrology Geophysics journal-article 2021 croxfordunivpr https://doi.org/10.1093/gji/ggab511 2024-03-08T03:02:36Z SUMMARY Despite advanced seismological techniques, automatic source characterization for microseismic earthquakes remains difficult and challenging since current inversion and modelling of high-frequency signals are complex and time consuming. For real-time applications such as induced seismicity monitoring, the application of standard methods is often not fast enough for true complete real-time information on seismic sources. In this paper, we present an alternative approach based on recent advances in deep learning for rapid source-parameter estimation of microseismic earthquakes. The seismic inversion is represented in compact form by two convolutional neural networks, with individual feature extraction, and a fully connected neural network, for feature aggregation, to simultaneously obtain full moment tensor and spatial location of microseismic sources. Specifically, a multibranch neural network algorithm is trained to encapsulate the information about the relationship between seismic waveforms and underlying point-source mechanisms and locations. The learning-based model allows rapid inversion (within a fraction of second) once input data are available. A key advantage of the algorithm is that it can be trained using synthetic seismic data only, so it is directly applicable to scenarios where there are insufficient real data for training. Moreover, we find that the method is robust with respect to perturbations such as observational noise and data incompleteness (missing stations). We apply the new approach on synthesized and example recorded small magnitude (M ≤ 1.6) earthquakes at the Hellisheiði geothermal field in the Hengill area, Iceland. For the examined events, the model achieves excellent performance and shows very good agreement with the inverted solutions determined through standard methodology. In this study, we seek to demonstrate that this approach is viable for microseismicity real-time estimation of source parameters and can be integrated into advanced decision-support tools for controlling ... Article in Journal/Newspaper Iceland Oxford University Press Hengill ENVELOPE(-21.306,-21.306,64.078,64.078) Rapid Point ENVELOPE(-97.552,-97.552,75.868,75.868) Geophysical Journal International |
institution |
Open Polar |
collection |
Oxford University Press |
op_collection_id |
croxfordunivpr |
language |
English |
topic |
Geochemistry and Petrology Geophysics |
spellingShingle |
Geochemistry and Petrology Geophysics Nooshiri, Nima Bean, Christopher J Dahm, Torsten Grigoli, Francesco Kristjánsdóttir, Sigríður Obermann, Anne Wiemer, Stefan A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland |
topic_facet |
Geochemistry and Petrology Geophysics |
description |
SUMMARY Despite advanced seismological techniques, automatic source characterization for microseismic earthquakes remains difficult and challenging since current inversion and modelling of high-frequency signals are complex and time consuming. For real-time applications such as induced seismicity monitoring, the application of standard methods is often not fast enough for true complete real-time information on seismic sources. In this paper, we present an alternative approach based on recent advances in deep learning for rapid source-parameter estimation of microseismic earthquakes. The seismic inversion is represented in compact form by two convolutional neural networks, with individual feature extraction, and a fully connected neural network, for feature aggregation, to simultaneously obtain full moment tensor and spatial location of microseismic sources. Specifically, a multibranch neural network algorithm is trained to encapsulate the information about the relationship between seismic waveforms and underlying point-source mechanisms and locations. The learning-based model allows rapid inversion (within a fraction of second) once input data are available. A key advantage of the algorithm is that it can be trained using synthetic seismic data only, so it is directly applicable to scenarios where there are insufficient real data for training. Moreover, we find that the method is robust with respect to perturbations such as observational noise and data incompleteness (missing stations). We apply the new approach on synthesized and example recorded small magnitude (M ≤ 1.6) earthquakes at the Hellisheiði geothermal field in the Hengill area, Iceland. For the examined events, the model achieves excellent performance and shows very good agreement with the inverted solutions determined through standard methodology. In this study, we seek to demonstrate that this approach is viable for microseismicity real-time estimation of source parameters and can be integrated into advanced decision-support tools for controlling ... |
author2 |
European Commission Icelandic Centre for Research |
format |
Article in Journal/Newspaper |
author |
Nooshiri, Nima Bean, Christopher J Dahm, Torsten Grigoli, Francesco Kristjánsdóttir, Sigríður Obermann, Anne Wiemer, Stefan |
author_facet |
Nooshiri, Nima Bean, Christopher J Dahm, Torsten Grigoli, Francesco Kristjánsdóttir, Sigríður Obermann, Anne Wiemer, Stefan |
author_sort |
Nooshiri, Nima |
title |
A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland |
title_short |
A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland |
title_full |
A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland |
title_fullStr |
A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland |
title_full_unstemmed |
A multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the Hengill Geothermal Field, Iceland |
title_sort |
multibranch, multitarget neural network for rapid point-source inversion in a microseismic environment: examples from the hengill geothermal field, iceland |
publisher |
Oxford University Press (OUP) |
publishDate |
2021 |
url |
http://dx.doi.org/10.1093/gji/ggab511 https://academic.oup.com/gji/advance-article-pdf/doi/10.1093/gji/ggab511/41810957/ggab511.pdf https://academic.oup.com/gji/article-pdf/229/2/999/42334927/ggab511.pdf |
long_lat |
ENVELOPE(-21.306,-21.306,64.078,64.078) ENVELOPE(-97.552,-97.552,75.868,75.868) |
geographic |
Hengill Rapid Point |
geographic_facet |
Hengill Rapid Point |
genre |
Iceland |
genre_facet |
Iceland |
op_source |
Geophysical Journal International volume 229, issue 2, page 999-1016 ISSN 0956-540X 1365-246X |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.1093/gji/ggab511 |
container_title |
Geophysical Journal International |
_version_ |
1795669352908849152 |