A Caltech Library Service

Deep learning for seismic phase detection and picking in the aftershock zone of 2008 M_W 7.9 Wenchuan Earthquake

Zhu, Lijun and Peng, Zhigang and McClellan, James and Li, Chenyu and Yao, DongDong and Li, Zefeng and Fang, Lihua (2019) Deep learning for seismic phase detection and picking in the aftershock zone of 2008 M_W 7.9 Wenchuan Earthquake. Physics of the Earth and Planetary Interiors, 293 . Art. No. 106261. ISSN 0031-9201. doi:10.1016/j.pepi.2019.05.004.

[img] PDF - Accepted Version
See Usage Policy.


Use this Persistent URL to link to this item:


The increasing volume of seismic data from long-term continuous monitoring motivates the development of algorithms based on convolutional neural network (CNN) for faster and more reliable phase detection and picking. However, many less studied regions lack a significant amount of labeled events needed for traditional CNN approaches. In this paper, we present a CNN-based Phase-Identification Classifier (CPIC) designed for phase detection and picking on small to medium sized training datasets. When trained on 30,146 labeled phases and applied to one-month of continuous recordings during the aftershock sequences of the 2008 M_W 7.9 Wenchuan Earthquake in Sichuan, China, CPIC detects 97.5% of the manually picked phases in the standard catalog and predicts their arrival times with a five-times improvement over the ObsPy AR picker. In addition, unlike other CNN-based approaches that require millions of training samples, when the off-line training set size of CPIC is reduced to only a few thousand training samples the accuracy stays above 95%. The deployment of CPIC takes less than 12 h to pick arrivals in 31-day recordings on 14 stations. In addition to the catalog phases manually picked by analysts, CPIC finds more phases for existing events and new events missed in the catalog. Among those additional detections, some are confirmed by a matched filter method while others require further investigation. Finally, when tested on a small dataset from a different region (Oklahoma, US), CPIC achieves 97% accuracy after fine tuning only the fully connected layer of the model. This result suggests that the CPIC developed in this study can be used to identify and pick P/S arrivals in other regions with no or minimum labeled phases.

Item Type:Article
Related URLs:
URLURL TypeDescription
Zhu, Lijun0000-0002-3046-5824
Peng, Zhigang0000-0002-0019-9860
McClellan, James0000-0003-0647-9561
Li, Chenyu0000-0002-0442-6928
Yao, DongDong0000-0002-2280-6617
Li, Zefeng0000-0003-4405-8872
Fang, Lihua0000-0003-2156-4406
Additional Information:© 2019 Published by Elsevier B.V. Received 22 May 2018, Revised 27 April 2019, Accepted 1 May 2019, Available online 17 May 2019. We utilized the PyTorch deep-learning neural network package (Paszke et al., 2017) and ObsPy package (Beyreuther et al., 2010). The seismic data utilized in this study is obtained during the 2017 “Aftershock Detection Artificial-Intelligence Contest” (Fang et al., 2017). We thank PEPI editor Vernon Cormier, Dr. Robert Geller and two anonymous reviewers for their constructive comments/suggestions. This research was supported by the Southern California Earthquake Center (Contribution No. 9046). SCEC is funded by NSF Cooperative Agreement EAR-100087 & USGS Cooperative Agreement G17A00047.
Group:Seismological Laboratory
Funding AgencyGrant Number
Southern California Earthquake Center (SCEC)UNSPECIFIED
Subject Keywords:Earthquake; Detection; Phase picking; Machine learning; CNN; Wenchuan earthquake
Other Numbering System:
Other Numbering System NameOther Numbering System ID
Southern California Earthquake Center9046
Record Number:CaltechAUTHORS:20190517-103813582
Persistent URL:
Official Citation:Lijun Zhu, Zhigang Peng, James McClellan, Chenyu Li, Dongdong Yao, Zefeng Li, Lihua Fang, Deep learning for seismic phase detection and picking in the aftershock zone of 2008 Mw7.9 Wenchuan Earthquake, Physics of the Earth and Planetary Interiors, Volume 293, 2019, 106261, ISSN 0031-9201, (
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:95563
Deposited By: Tony Diaz
Deposited On:20 May 2019 18:12
Last Modified:16 Nov 2021 17:13

Repository Staff Only: item control page