Abstract
Regional seismic networks often observe artificially induced seismic events such as blasting and collapses. Misclassified seismic events in the earthquake catalog can therefore interfere with assessments of natural seismic activity. Traditional methods rely on the period, and phase features of seismic waves to determine the nature of seismic events. We designed three seismic event classifiers with reference to convolutional neural network structures such as VGGnet, ResNet, and Inception. The designed classifiers were tested and compared using three-channel seismic full-waveform time-series data and spectral data. Our classifiers are shown to only require 60 s of full-waveform seismic event data and first-arrival times for alignment; additional phase labeling or numerical filtering is unnecessary. Rapid classification of earthquakes, blasting, and mine collapses can be achieved within approximately 1 min of an event. As a test case, this study uses 6.4 k observations of actual local seismic events with magnitudes ranging from ML 0.6 to ML 4.5 obtained from 47 broadband seismic stations in the Henan Regional Network of the China Seismological Network Center; these observations include natural earthquakes, blasting, and collapse events. The results indicate that our classifiers can reach a lower classification magnitude limit of ML 0.6 and that their recall and accuracy exceed 90%, outperforming manually performed routine classifications and similar approaches. These findings provide an important reference for the rapid classification of small and medium earthquakes.
Similar content being viewed by others
Code availability
Name of code: Seismic-classifier.
Developer: Luozhao Jia.
Contact address: Henan Earthquake Agency, No. 10 Zhengguang Road, Zhengzhou City, China.
Email: lezhao.jia@gmail.com.
First year available: 2021.
Required hardware: PC or server.
Required hardware for training environment: NVIDIA RTX2080.
The above graphics card, operating system: Windows or Linux.
Programming language: Python 3.7, Tensorflow 2.0.
The source codes are available for download at the link: https://github.com/epnet2018/Seismic-classifier.
References
Bachmann, LM, Puhan MA, Riet GT, & Bossuyt PM (2006). Sample sizes of studies on diagnostic accuracy: literature survey. BMJ, 332
Baumgardt DR, Young GB (1990) Regional seismic waveform discriminates and case-based event identification using regional arrays. Bull Seismol Soc Am 80:1874–1892
Bennett TJ, Barker BW, McLaughlin KL, Murphy JR, 1989. Regional discrimination of quarry blasts, earthquakes and underground nuclear explosions. Final Report, GL-TR-89–0114, S-Cubed, La Jolla, CA, 146 pp
Blandford R (1982) Seismic event discrimination. Bull Seismol Soc Am 72:569–587
Bora SS, Scherbaum F, Kuehn N, Stafford P (2014) Fourier spectral- and duration models for the generation of response spectra adjustable to different source-, propagation-, and site conditions. Bull Earthq Eng 12(1):467–493
Bregman Y, Lindenbaum O, & Rabin N (2020). Array based earthquakes-explosion discrimination using diffusion maps. Pure and Applied Geophysics(2a)
Cho,& Kwang-Hyun.(2014). Discriminating between explosions and earthquakes. Applied Geophysics(04)
Dong LJ, Tang Z, Li XB, Chen YC, Xue JC (2020) Discrimination of mining microseismic events and blasts using convolutional neural networks and original waveform. J Cent South Univ 27(10):3078–3089
Fluss R, Faraggi D, & Reiser B (2005). Estimation of the Youden index and its associated cutoff point. Biometrical Journal, 47(4)
Glorot X, Bordes A, & Bengio Y (2011). Deep sparse rectifier neural networks. J Mach Learn Res (Vol.15, pp.315–323)
Goodfellow I, Bengio Y, & Courville A (2016). Deep learning. JMLR.org
Halevy A, Norvig P, Pereira F (2009) The unreasonable effectiveness of data. IEEE Intell Syst 24(2):8–12
He K, & Jian S (2015). Convolutional neural networks at constrained time cost. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE
He K, Zhang X, Ren S, et al 2016a Deep residual learning for image recognition [C]. Proceedings of the IEEE conference on computer vision and pattern recognition. 770–778
He K, Zhang X, Ren S, & Jian S (2016b). Identity mappings in deep residual networks. European Conference on Computer Vision
He B, Wei M, Watts DR, & Shen Y (2020). Detecting slow slip events from seafloor pressure data using machine learning. Geophysical Research Letters, 47
Huang G, Liu Z, Laurens V, & Weinberger KQ (2016). Densely connected convolutional networks. IEEE Computer Society
Ioffe S & Szegedy C (2015). Batch normalization: accelerating deep network training by reducing internal covariate shift. Proceedings of the 32nd International Conference on Machine Learning in Proceedings of Machine Learning Research 37:448–456
Johnson CW, Y Ben‐Zion, Meng H & Vernon F (2020). Identifying different classes of seismic noise signals using unsupervised learning. Geophysical Research Letters, 47(15)
Kaplan J, Mccandlish S, Henighan T, Brown TB, Chess B, & R Child et al. (2020). Scaling laws for neural language models
Keskar NS, & Socher R (2017). Improving generalization performance by switching from adam to sgd
Kingma D, & Ba J (2014). Adam: a method for stochastic optimization. Computer Science
Koper Keith, D, Hale Mark, J Stein, & Jared et al (2016). Magnitude-based discrimination of man-made seismic events from naturally occurring earthquakes in Utah, USA. Geophysical Research Letters
Kriegerowski M, Petersen GM, Vasyura-Bathke H, Ohrnberger M (2019) A deep convolutional neural network for localization of clustered earthquakes based on multistation full waveforms. Seismol Res Lett 90(2A):510–516
Krizhevsky A, Sutskever I, Hinton GE 2012 ImageNet classification with deep convolutional neural networks [C] Advances in neural information processing systems. 1097–1105
Kuyuk HS, Yildirim E, Dogan E, Horasan G (2011) An unsupervised learning algorithm: application to the discrimination of seismic events and quarry blasts in the vicinity of Istanbul. Nat Hazard 11(1):93–100
Lecun Y, Bottou L (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324
Linville LM (2022) Event-based training in label-limited regimes. J Geophys Res Solid Earth 127:e2021JB022820
Meier MA, Ross ZE, Ramachandran A, Balakrishna A, Nair S, & Kundzicz P et al (2018). Reliable real-time seismic signal/noise discrimination with machine learning. Journal of Geophysical Research: Solid Earth
Mousavi SM (2016) Comment on “Recent developments of the middle east catalog” by Zare et al. J Seismol 21:1–12
Peterson J 1993,Obervations and modeling of seismic background noise. U.S. Geol. Survey Open-File Report 93–322,95 pp.
Raskutti G,Wainwright MJ, & Yu B (2013). Early stopping and non-parametric regression: an optimal data-dependent stopping rule. J Mach Learn Res, 15
Rodgers AJ, Lay T, Walter WR, Myers S (1997) A comparison of regional-phase amplitude ratio measurement techniques. Bull Seismol Soc Am 87:1613–1621
Saad OM, Shalaby A, & Sayed MS (2019). Automatic discrimination of earthquakes and quarry blasts using wavelet filter bank and support vector machine. J Seismol
Sammut C, & Webb GI (2017). Encyclopedia of machine learning and data mining. https://doi.org/10.1007/978-1-4899-7687-1.
Scarpetta S, Giudicepietro F, Ezin EC, Petrosino S, Del Pezzo E, Martini M, Marinaro M (2005) Automatic classification of seismic signals at Mt. Vesuvius Volcano, Italy, using neural networks. Bull Seismol Soc Am 95(1):185–196
Seydoux L, Balestriero R, Poli P, Hoop MD, Baraniuk R (2020) Clustering earthquake signals and background noises in continuous seismic data with unsupervised deep learning. Nat Commun 11(1):3972
Simonyan K & Zisserman A (2014). Very deep convolutional networks for large-scale image recognition. Computer Science
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from over-fitting. J Mach Learn Res 15(1):1929–1958
Szegedy C, Liu W, Jia Y, Sermanet P, & Rabinovich A (2014). Going deeper with convolutions. IEEE Computer Society
Szegedy C,Vanhoucke V, Ioffe S, Shlens J & Wojna Z (2016). Rethinking the inception architecture for computer vision. IEEE, 2818–2826
Szegedy C, Ioffe S, Vanhoucke V & Alemi AA (2017). Inception-v4, inception-ResNet and the impact of residual connections on learning
Tang L, Zhang M, Wen L (2019) Support vector machine classification of seismic events in the Tianshan orogenic belt. J Geophys Res Solid Earth. https://doi.org/10.1029/2019JB018132
Trani L, Pagani GA, Zanetti J, Chapeland C, Evers LG (2021) DeepQuake -- an application of CNN for seismo-acoustic event classification in The Netherlands
Van Dyk DA, Meng XL (2001) The art of data augmentation. J Comput Graph Stat 10(1):1–50
Shang X, Li X, Morales-Esteban A, Chen G (2017) Improving microseismic event and quarry blast classification using artificial neural networks based on principal component analysis. Soil Dynamics & Earthquake Engineering
Yang XH, Shen P, Liu XQ, & Zheng ZZ (2005). Application of method of spectral component ratio of wavelet-packets to discrimination between earthquakes and nuclear explosions. Chinese J Geophys, 48(1)
Yildinm E, Guelbag A, Horasan G, Dogan E (2011) Discrimination of quarry blasts and earthquakes in the vicinity of Istanbul using soft computing techniques. Comput Geosci 37(9):1209–1217
Zhang Z, & Sabuncu M. (2018). Generalized cross entropy loss for training deep neural networks with noisy labels. Advances in neural information processing systems, 31
Zhou XH ,Obuchowski NA, & Mcclish, DK (2008). 2. Measures of diagnostic accuracy. John Wiley & Sons, Inc
Acknowledgements
The seismic waveform dataset used in this paper was provided by the Henan Regional Center of the China Seismic Network Center. We thank the anonymous reviewers for their useful comments and suggestions. The seismic waveforms used in this paper can be downloaded from https://data.earthquake.cn. The algorithm was implemented using the Python software package Tensorflow (https://tensorflow.org), tflearn (http://tflearn.org), Matplotlib implementation, keras (https://keras.io), and scikit-learn (http://scikit-learn.org).
Funding
This research was supported by the Earthquake Science Spark Program of China Earthquake Administration no. XH20036.
Author information
Authors and Affiliations
Contributions
Luozhao Jia: completed the experiments, program design, and writing.
Hongfeng Chen: contributed to the experimental concept and the analysis of the findings.
Kang Xing: participated in the collection of the data and the production of the illustrations.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Article highlights
• We designed and compared three deep-learning classifiers to distinguish natural earthquakes, blasting, and collapse using convolutional neural network structures.
• The effects of the time-series input and spectrum on the classifiers were compared.
• The designed classifiers do not need to extract waveform features or mark seismic phases in advance and are suitable for moderate and microseismic events.
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Jia, L., Chen, H. & Xing, K. Rapid classification of local seismic events using machine learning. J Seismol 26, 897–912 (2022). https://doi.org/10.1007/s10950-022-10109-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10950-022-10109-5