CoChat

Seismic arrival-time picking on distributed acoustic sensing data using semi-supervised learning

Open access

Published version of a preprint

The DOI matches its Crossref record; 81 of the 81 references it cites were found in a registry or an index.

Abstract

Distributed Acoustic Sensing (DAS) is an emerging technology for earthquake monitoring and subsurface imaging. However, its distinct characteristics, such as unknown ground coupling and high noise level, pose challenges to signal processing. Existing machine learning models optimized for conventional seismic data struggle with DAS data due to its ultra-dense spatial sampling and limited manual labels. We introduce a semi-supervised learning approach to address the phase-picking task of DAS data. We use the pre-trained PhaseNet model to generate noisy labels of P/S arrivals in DAS data and apply the Gaussian mixture model phase association (GaMMA) method to refine these noisy labels and build training datasets. We develop PhaseNet-DAS, a deep learning model designed to process 2D spatio-temporal DAS data to achieve accurate phase picking and efficient earthquake detection. Our study demonstrates a method to develop deep learning models for DAS data, unlocking the potential of integrating DAS in enhancing earthquake monitoring.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-023-43355-3

Cites

81 of 81 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Examples of noisy picks predicted by PhaseNet and improved picks predicted by PhaseNet-DAS. a–d Four examples with different signal-to-noise ratios. Each sub-panel shows (i) DAS recordings of 30 s and 5000 channels, (ii) the PhaseNet picks, and (iii) the PhaseNet-DAS picks.
Fig. 1 Examples of noisy picks predicted by PhaseNet and improved picks predicted by PhaseNet-DAS. a–d Four examples with different signal-to-noise ratios. Each sub-panel shows (i) DAS recordings of 30 s and 5000 channels, (ii) the PhaseNet picks, and (iii) the PhaseNet-DAS picks.CC BY · PubMed Central
Fig. 2 SNR distributions of detected events across four DAS arrays. a Mammoth north, b Mammoth south, c Ridgecrest north, d Ridgecrest south. The locations of the four DAS array are shown in Fig. 8. The SNR is calculated using two 5-s windows before and after the theoretical P wave arrival time. The PhaseNet-DAS v1 and v2 models are from the first and second iterations of the semi-supervised learning procedures illustrated in Fig. 6.
Fig. 2 SNR distributions of detected events across four DAS arrays. a Mammoth north, b Mammoth south, c Ridgecrest north, d Ridgecrest south. The locations of the four DAS array are shown in Fig. 8. The SNR is calculated using two 5-s windows before and after the theoretical P wave arrival time. The PhaseNet-DAS v1 and v2 models are from the first and second iterations of the semi-supervised learning procedures illustrated in Fig. 6.CC BY · PubMed Central
Fig. 3 Residuals of differential arrival-times picked by PhaseNet-DAS. a P waves and b S waves. We first measure differential arrival-times of PhaseNet-DAS picks (dtphasenet-das) and waveform cross-correlation (dtcross-correlation) from selected event pairs. Then we calculate the residuals between these two differential arrival-times (dtphasenet-das − dtcross-correlation) to evaluate the accuracy of PhaseNet-DAS picks, assuming waveform cross-correlation measurement as the ground truth.
Fig. 3 Residuals of differential arrival-times picked by PhaseNet-DAS. a P waves and b S waves. We first measure differential arrival-times of PhaseNet-DAS picks (dtphasenet-das) and waveform cross-correlation (dtcross-correlation) from selected event pairs. Then we calculate the residuals between these two differential arrival-times (dtphasenet-das − dtcross-correlation) to evaluate the accuracy of PhaseNet-DAS picks, assuming waveform cross-correlation measurement as the ground truth.CC BY · PubMed Central
Fig. 4 Magnitude and distance distributions of earthquakes. a PhaseNet-DAS v2, b PhaseNet-DAS v1, and c PhaseNet. The gray dots are earthquakes in standard earthquake catalogs within 3 degrees of the Long Valley DAS array. The red dots indicate the earthquakes that can be detected with more than 500 associated P and S picks. The PhaseNet-DAS v1 and v2 models are from the first and second iterations of semi-supervised learning (Fig. 6). The histogram of earthquake numbers is shown in Fig. S3.
Fig. 4 Magnitude and distance distributions of earthquakes. a PhaseNet-DAS v2, b PhaseNet-DAS v1, and c PhaseNet. The gray dots are earthquakes in standard earthquake catalogs within 3 degrees of the Long Valley DAS array. The red dots indicate the earthquakes that can be detected with more than 500 associated P and S picks. The PhaseNet-DAS v1 and v2 models are from the first and second iterations of semi-supervised learning (Fig. 6). The histogram of earthquake numbers is shown in Fig. S3.CC BY · PubMed Central
Fig. 5 Earthquake locations determined by phase arrival-times measured by PhaseNet-DAS. The black dots are earthquakes in the standard earthquake catalogs. The red dots are earthquakes detected by the DAS arrays and the PhaseNet-DAS v2 model. Only the DAS events corresponding to a catalog event are shown. The results of PhaseNet and PhaseNet-DAS v1 models are shown in Fig. S4.
Fig. 5 Earthquake locations determined by phase arrival-times measured by PhaseNet-DAS. The black dots are earthquakes in the standard earthquake catalogs. The red dots are earthquakes detected by the DAS arrays and the PhaseNet-DAS v2 model. Only the DAS events corresponding to a catalog event are shown. The results of PhaseNet and PhaseNet-DAS v1 models are shown in Fig. S4.CC BY · PubMed Central
Fig. 6 The procedure of semi-supervised learning for training the PhaseNet-DAS model. The PhaseNet model19, which is pre-trained using a large dataset of seismic waveforms, is used to generate pseudo-labels on DAS data. This semi-supervised approach transfers the phase picking capability from PhaseNet to the new PhaseNet-DAS model designed for DAS recordings.
Fig. 6 The procedure of semi-supervised learning for training the PhaseNet-DAS model. The PhaseNet model19, which is pre-trained using a large dataset of seismic waveforms, is used to generate pseudo-labels on DAS data. This semi-supervised approach transfers the phase picking capability from PhaseNet to the new PhaseNet-DAS model designed for DAS recordings.CC BY · PubMed Central
Fig. 7 The neural network architecture of PhaseNet-DAS. We use the U-Net73 architecture to consider spatial and temporal information of 2D DAS recordings. PhaseNet-DAS processes raw DAS data through four stages of downsampling and upsampling and a sequence of 2D convolutional layers and relu activation functions and predicts P and S phase arrivals in each channel of the DAS array, represented by blue and red lines respectively.
Fig. 7 The neural network architecture of PhaseNet-DAS. We use the U-Net73 architecture to consider spatial and temporal information of 2D DAS recordings. PhaseNet-DAS processes raw DAS data through four stages of downsampling and upsampling and a sequence of 2D convolutional layers and relu activation functions and predicts P and S phase arrivals in each channel of the DAS array, represented by blue and red lines respectively.CC BY · PubMed Central
Fig. 8 Four DAS cables used in the training dataset. a Long Valley and b Ridgecrest, CA. The blue and orange lines are the locations of the fiber-optic cables. The black dots are earthquakes in the standard earthquake catalogs used to build the training dataset.
Fig. 8 Four DAS cables used in the training dataset. a Long Valley and b Ridgecrest, CA. The blue and orange lines are the locations of the fiber-optic cables. The black dots are earthquakes in the standard earthquake catalogs used to build the training dataset.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 81 references

Research works printed but not linked: 11

  1. Li, Z. & Zhan, Z. Pushing the limit of earthquake detection with distributed acoustic sensing and template matching: a case study at the brady geothermal field. Geophys. J. Int. 215, 1583–1593 (2018b). · DOI
  2. Zhu, W. & Beroza, G. C. PhaseNet: a deep-neural-network-based seismic arrival-time picking method. Geophys. J. Int. 216, 261–273 (2019).
  3. Su, J. et al. High resolution earthquake catalog building for the 21 May 2021 Yangbi, Yunnan, M S 6.4 earthquake sequence using deep-learning phase picker. Chin. J. Geophys. 64, 2647–2656 (2021).
  4. Wilding, J. D., Zhu, W., Ross, Z. E. & Jackson, J. M. The magmatic web beneath Hawai‘i. Science 0, eade5755 (2022).
  5. Hernández, Pablo D., Ramírez, Jaime A. & Soto, Marcelo A. Improving earthquake detection in fibre-optic distributed acoustic sensors using deep-learning and hybrid datasets. In 2022 European Conference on Optical Communication (ECOC) 1–4 (Optica Publishing Group, 2022).
  6. VanDecar, J. C. & Crosson, R. S. Determination of teleseismic relative phase arrival times using multi-channel cross-correlation and least squares. Bull. Seismol. Soc. Am. 80, 150–169 (1990).
  7. Zhu, W., Mousavi, S. M. & Beroza, G. C. Seismic signal augmentation to improve generalization of deep neural networks. Adv. Geophys. 61, 151–177 (2020). · DOI
  8. Luo, W., Li, Y., Urtasun, R. & Zemel, R. Understanding the effective receptive field in deep convolutional neural networks. In Advances in Neural Information Processing Systems (Curran Associates, Inc., 2016).
  9. Ioffe, S. & Szegedy, C. Batch normalization: accelerating deep network training by reducing internal covariate shift. In Proc. 32nd International Conference on Machine Learning 448–456 (PMLR, 2015).
  10. Glorot, X., Bordes, A. & Bengio, Y. Deep sparse rectifier neural networks. In Proc. 14h International Conference on Artificial Intelligence and Statistics 315–323 (JMLR Workshop and Conference Proceedings, 2011).
  11. Biondi, E., Wang, X., Williams, E. F. & Zhan, Z. Geolocalization of large-scale das channels using a GPS-tracked moving vehicle. Seismol. Soc. Am. 94, 318–330 (2023).

Cited by

See all 159

Metadata

Authors
Zhu, W., Biondi, E., Li, J., Yin, J., Ross, Z. E., & Zhan, Z.
Affiliations
California Institute of Technology (US)
Corresponding author
Zhu, Weiqiang
Venue
Nature Communications
Published
2023-12-11
Volume
14
Issue
1
Pages
8192-8192
ISSN
2041-1723 (ISSN-L)
OpenAlex
W4389560491
Cited by
130
References
78
Citations by year
2026: 53 · 2025: 44 · 2024: 30 · 2023: 3
FWCI
12.25
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • National Science Foundation (1848166)
  • Department of the Interior | United States Geological Survey | Wyoming-Montana Water Science Center (G22AP00067)
  • Gordon and Betty Moore Foundation
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Seismology and Earthquake Studies
  • Physical Sciences › Earth and Planetary Sciences › Geophysics › Seismic Waves and Analysis
  • Physical Sciences › Earth and Planetary Sciences › Geophysics › Seismic Imaging and Inversion Techniques
Keywords
distributed acoustic sensing, semi-supervised learning, PhaseNet, earthquake detection, seismic monitoring, ground coupling

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Seismic arrival-time picking on distributed acoustic sensing data using semi-supervised learning. https://app.cochat.ai/papers/10.1038/s41467-023-43355-3
The paper
Zhu, W., Biondi, E., Li, J., Yin, J., Ross, Z. E., & Zhan, Z. (2023). Seismic arrival-time picking on distributed acoustic sensing data using semi-supervised learning. Nature Communications. https://doi.org/10.1038/s41467-023-43355-3

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.