Detection of records of weak local earthquakes using neural networks

Manual processing of large volumes of continuous observations produced by local seismic networks takes a lot of time, therefore, to solve this problem, automatic algorithms for detecting seismic events are used. Deterministic methods for solving the problem of detection, which do an excellent job of detecting intensive earthquakes, face critical problems when detecting weak seismic events (earthquakes). They are based on principles based on the calculation of energy, which causes multiple errors in detection: weak seismic events may not be detected, and high-amplitude noise may be mistakenly detected as an event. In our work, we propose a detection method capable of surpassing deterministic methods in detecting events on seismograms, successfully detecting a similar or more events with fewer false detections.

1. Akram J., Ovcharenko O., Peter D. A robust neural network-based approach for microseismic event detection // SEG Technical Program Expanded Abstracts. – 2017. – P. 2929–2933, doi: 10.1190/segam2017-17761195.1.

2. Coppens F. First arrivals picking on common-offset trace collections for automatic estimation of static corrections // Geophysical Prospecting. – 1985. – Vol. 33 (8). – P. 1212–1231, doi: 10.1111/j.1365-2478.1985.tb01360.x.

3. Dergach P.A., Loginov G.N. An algorithm for detecting signals from weak local earthquakes based on a convolutional neural network: an example of real data processing and comparison with the classical method // Modern methods of seismic hazard assessment and earthquake prediction: Russian scientific conference with international participation (Moscow, November 27–28, 2019): Abstracts and conference program. – Moscow, 2019. – P. 40–41.

4. Earthquakes and microseismicity in problems of contemporary geodynamics of the East European Platform, in: Sharova N.V., Malovichko A.A., Shchukina Yu.K. (Eds.). Book. 2: Microseismicity. – Petrozavodsk, Karelian Scientific Center of the Russian Academy of Sciences, 2007. – 96 p.

5. Emanov A.F., Emanov A.A., Leskova E.V., Fateev A.V., Semin A.Yu. Seismic activations during coal mining in Kuzbass // Physical Mesomechanics. – 2009. – Vol. 12 (1). – P. 37–43.

6. Guo C., Zhu T., Gao Y., Wu S., Sun J. AEnet: Automatic Picking of P-Wave First Arrivals Using Deep Learning // IEEE Transactions on Geosciences and Remote Sensing. – 2020. – Vol. 59 (6). – P. 5293–5303, doi: 10.1109/TGRS.2020.3010541.

7. Loginov G.N., Duchkov A.A., Litvichenko D.A., Alyamkin S.A. An example of using convolutional neural networks in real 3D seismic data processing // Interexpo GEO-Siberia: XV Intern. Scientific. Congr. (Novosibirsk, April 24–26, 2019): Intern. Scientific. conf. "Subsoil use. Mining. Directions and technologies of prospecting, exploration and development of mineral deposits. Economy. Geoecology": Collection of materials in 9 volumes. – Novosibirsk, 2019. – Vol. 2, No. 3. – P. 147–153.

8. Puzyrev N.N. Methods and objects of seismic investigation. Introduction to general seismology [in Russian]. – Novosibirsk: SB RAS, 1997. – 301 p.

9. Srivastava N., Hinton G., Krizhevsky A., Sutskever I., Salakhutdinov R. Dropout: a simple way to prevent neural networks from overfitting. Journal of Machine Learning Research. – 2014. – Vol. 15 (1). – P. 1929–1958.