Earthquake Detection and P-Wave Arrival Time Picking Using Capsule Neural Network

Earthquake detection is an essential step in observational earthquake seismology. We propose to utilize a capsule neural network (CapsNet) to automatically identify and detect earthquakes. CapsNet is the new generation of deep learning architecture. It has the capability of learning with a great generalization performance from a small dataset. We train the CapsNet using 50% of the Southern California seismic data (2.25 million 4-s-three-component seismic windows) and use 222 395 waveforms from different seismic areas to evaluate the CpasNet performance, e.g., western United States, Europe, and Japan. As a result, the CapsNet misses 367 events and detects 217 305 events with an accuracy of 97.71%. Among these picked events, 210 498 events have an arrival time error below 0.2 s (96.86%) and 197968 waveforms with an arrival time error below 0.1 s (91.11%). The CapsNet precision, recall, and F1-score are 97.78%, 99.83%, and 98.79%, respectively. In addition, the CapsNet is tested using 100 000 60-s-three-component seismic noise waveforms. CapsNet shows a low false alarms rate of 1384, which gives the CapsNet an accuracy of 98.61%. In addition, CapsNet is tested using continuous seismic data associated with the 24-hours microearthquakes swarm that occurred in the Arkansas area. Accordingly, the CapsNet detects 221 earthquakes and releases 37 false alarms with a detection accuracy of 85.65%. CapsNet detects many microearthquakes with a small magnitude, as low as −1.3 Ml, and detects earthquakes that have a low signal-to-noise ratio (SNR), e.g., as low as −8.07 dB. The results of the CapsNet are compared to the benchmark methods, e.g., short-time average/long-time average (STA/LTA) and GPD methods. The CapsNet shows the highest picking accuracy and outperforms the benchmark methods.

[1]  G. Beroza,et al.  Temporal variation in the magnitude‐frequency distribution during the Guy‐Greenbrier earthquake sequence , 2015 .

[2]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[3]  Anthony Lomax,et al.  TexNet: A Statewide Seismological Network in Texas , 2019, Seismological Research Letters.

[4]  Koji Inoue,et al.  Automatic Arrival Time Detection for Earthquakes Based on Stacked Denoising Autoencoder , 2018, IEEE Geoscience and Remote Sensing Letters.

[5]  Michaël Gharbi,et al.  Convolutional neural network for earthquake detection and location , 2017, Science Advances.

[6]  S. Mostafa Mousavi,et al.  Seismic Signal Denoising and Decomposition Using Deep Neural Networks , 2018, IEEE Transactions on Geoscience and Remote Sensing.

[7]  Gregory C. Beroza,et al.  Earthquake Fingerprints: Extracting Waveform Features for Similarity-Based Earthquake Detection , 2018, Pure and Applied Geophysics.

[8]  Gregory C. Beroza,et al.  Earthquake transformer—an attentive deep-learning model for simultaneous earthquake detection and phase picking , 2020, Nature Communications.

[9]  S. Mostafa Mousavi,et al.  CRED: A Deep Residual Network of Convolutional and Recurrent Units for Earthquake Signal Detection , 2018, Scientific Reports.

[10]  Zefeng Li,et al.  High-resolution seismic event detection using local similarity for Large-N arrays , 2018, Scientific Reports.

[11]  Yisong Yue,et al.  PhaseLink: A Deep Learning Approach to Seismic Phase Association , 2018, Journal of Geophysical Research: Solid Earth.

[12]  Peng Jiang,et al.  Deep-Learning Inversion of Seismic Data , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[13]  Yangkang Chen,et al.  Fast waveform detection for microseismic imaging using unsupervised machine learning , 2018 .

[14]  Zongben Xu,et al.  An Optimized Deep Network Representation of Multimutation Differential Evolution and its Application in Seismic Inversion , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[15]  R. V. Allen,et al.  Automatic earthquake recognition and timing from single traces , 1978, Bulletin of the Seismological Society of America.

[16]  Weiqiang Zhu,et al.  PhaseNet: A Deep-Neural-Network-Based Seismic Arrival Time Picking Method , 2018, Geophysical Journal International.

[17]  M. Leonard,et al.  Multi-component autoregressive techniques for the analysis of seismograms , 1999 .

[18]  Yangkang Chen,et al.  Automatic high-resolution microseismic event detection via supervised machine learning , 2019, Geophysical Journal International.

[19]  Yangkang Chen,et al.  Convolutional neural networks for microseismic waveform classification and arrival picking , 2020 .

[20]  Zachary E. Ross,et al.  P Wave Arrival Picking and First‐Motion Polarity Determination With Deep Learning , 2018, Journal of Geophysical Research: Solid Earth.

[21]  Omar M. Saad,et al.  Automatic waveform-based source-location imaging using deep learning extracted microseismic signals , 2020 .

[22]  Yue Wu,et al.  DeepDetect: A Cascaded Region-Based Densely Connected Network for Seismic Event Detection , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[23]  Yara Rizk,et al.  A Deep Transfer Learning Framework for Seismic Data Analysis: A Case Study on Bright Spot Detection , 2020, IEEE Transactions on Geoscience and Remote Sensing.

[24]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[25]  Stephen P. Horton,et al.  Characteristics of induced/triggered earthquakes during the startup phase of the Guy–Greenbrier earthquake sequence in north–central Arkansas , 2016 .

[26]  S. Mostafa Mousavi,et al.  STanford EArthquake Dataset (STEAD): A Global Data Set of Seismic Signals for AI , 2019, IEEE Access.

[27]  F. Viégas,et al.  Deep learning of aftershock patterns following large earthquakes , 2018, Nature.

[28]  Wei Chen,et al.  Deep Learning for Regularly Missing Data Reconstruction , 2020, IEEE Transactions on Geoscience and Remote Sensing.

[29]  Bohan Jia,et al.  DE-CapsNet: A Diverse Enhanced Capsule Network with Disperse Dynamic Routing , 2020 .

[30]  Geoffrey E. Hinton,et al.  Dynamic Routing Between Capsules , 2017, NIPS.

[31]  Tohru Kohda,et al.  Seismic noise study for accurate P-wave arrival detection via MODWT , 2013, Comput. Geosci..

[32]  Anthony Lomax,et al.  An Investigation of Rapid Earthquake Characterization Using Single‐Station Waveforms and a Convolutional Neural Network , 2019, Seismological Research Letters.

[33]  Clara E Yoon,et al.  Earthquake detection through computationally efficient similarity search , 2015, Science Advances.

[34]  Chao Zhang,et al.  Automatic Time Picking for Microseismic Data Based on a Fuzzy C-Means Clustering Algorithm , 2016, IEEE Geoscience and Remote Sensing Letters.

[35]  Rinat Mukhometzianov,et al.  CapsNet comparative performance evaluation for image classification , 2018, ArXiv.

[36]  Liguan Wang,et al.  PickCapsNet: Capsule Network for Automatic P-Wave Arrival Picking , 2021, IEEE Geoscience and Remote Sensing Letters.

[37]  Yangkang Chen,et al.  Deep learning for seismic lithology prediction , 2018 .

[38]  Thomas H. Heaton,et al.  Generalized Seismic Phase Detection with Deep Learning , 2018, Bulletin of the Seismological Society of America.

[39]  Yangkang Chen,et al.  Automatic microseismic event picking via unsupervised machine learning , 2020, Geophysical Journal International.

[40]  Sergey Fomel,et al.  FaultNet3D: Predicting Fault Probabilities, Strikes, and Dips With a Single Convolutional Neural Network , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[41]  Stephen P. Horton,et al.  Disposal of Hydrofracking Waste Fluid by Injection into Subsurface Aquifers Triggers Earthquake Swarm in Central Arkansas with Potential for Damaging Earthquake , 2012 .