Time-Series Forecasting of the Pazarcık Earthquake Using LSTM, Transformer and RNN Models

Abstract

The Earth's internal structure and mitigating seismic hazards are very important for understanding for earthquake prediction and seismic wave analysis. In this study, we studied with different deep learning models for earthquake time series prediction using Broadband Teleseismic Data from the USGS database. This dataset consists of 1000 seismic records in SAC format with long-period seismic waves from global earthquakes. The aim of this study was to test LSTM and RNN models with LSTM Transformer to predict the next time step based on previous seismic waves. In this study, model performances was evaluated with Mean Square Error (MSE), Mean Absolute Error (MAE) and R² Score. In conclusion, the LSTM Transformer+RNN model achieves the lowest error rates and presents its effectiveness in learning both short-term dependencies and long-term correlations in seismic data. At the same time, this study can also provides to the advancement of deep learning applications in seismology and the improvement of the prediction capabilities of earthquake monitoring systems.

Keywords

• Eartiquake Prediction,
• Deep Learning
• LSTM
• Transformer
• RNN
• Seismic Time-Series

LSTM, Transformer ve RNN Modellerini Kullanarak Pazarcık Depreminin Zaman Serisi Tahmini

Abstract

Dünya'nın iç yapısını anlayarak sismik tehlikeleri azaltmak için deprem tahmini ve sismik dalga analizlerinin yapılması büyük önem taşımaktadır. Bu çalışmada, USGS veritabanından elde edilen Geniş Bant Teleseismik Verileri kullanarak deprem zaman serisi tahmini için farklı derin öğrenme modelleri ile çalışıldı. Bu veri seti, küresel depremlerden gelen uzun dönemli sismik dalgaları yakalayan 1000 adet SAC formatında sismik kayıttan oluşmaktadır. Bu çalışmada ki amacımız, önceki sismik dalgaları temel alarak bir sonraki zaman adımını tahmin etmek için LSTM Transformer ile geliştirilmiş LSTM ve RNN modellerinin denenmesini içermektedir. Bu çalışmada model performansları, Ortalama Karesel Hata (MSE), Ortalama Mutlak Hata (MAE) ve R² Skoru kullanılarak değerlendirildi. Sonuç olarak LSTM Transformer+RNN modelinin en düşük hata oranlarına ulaştığını ve sismik verilerdeki hem kısa vadeli bağımlılıkları hem de uzun vadeli korelasyonları öğrenmede etkinliğini gösterdiğini ortaya koymaktadır. Bu çalışma aynı zamanda, sismolojide derin öğrenme uygulamalarının ilerlemesine ve deprem izleme sistemlerinin tahmin yeteneklerinin geliştirilmesine katkıda bulunabilir.

Keywords

• Deprem Tahmini,
• Derin Öğrenme,
• LSTM,
• Transformer,
• RNN
• Sismik Zaman Serisi

References

1. [1] B. Gutenberg and C. F. Richter, Seismicity of the Earth and Associated Phenomena. Princeton University Press, 1954.
2. [2] AFAD, “6 Şubat 2023 Kahramanmaraş Depremleri Raporu,” Afet ve Acil Durum Yönetimi Başkanlığı, 2023. [Online]. Available: https://www.afad.gov.tr
3. [3] USGS, “M 7.8 – Türkiye-Suriye sınır bölgesi,” United States Geological Survey, 2023. [Online]. Available: https://earthquake.usgs.gov.
4. [4] K. Satake, Y. Fujii, T. Harada, and Y. Namegaya, “Time and space distribution of coseismic slip of the 2011 Tōhoku earthquake as inferred from tsunami waveform data,” Bulletin of the Seismological Society of America, vol. 103, no. 2B, pp. 1473–1492, 2013.
5. [5] R. Madariaga, M. Métois, C. Vigny, and J. Campos, “The 2010 Mw 8.8 Maule megathrust earthquake of Central Chile, and its aftershocks,” Geophysical Journal International, vol. 184, no. 1, pp. 1–17, 2011.
6. [6] USGS “M 7.8 - Pazarcik earthquake, Kahramanmaras earthquake sequence” Avaible:https://earthquake.usgs.gov/earthquakes/eventpage/us6000jllz/executive [Accessed : February 09,2024].
7. [7] Qin, Yao & Song, Dongjin & Cheng, Haifeng & Cheng, Wei & Jiang, Guofei & Cottrell, Garrison. (2017). A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction. 10.48550/arXiv.1704.0297.
8. [8] Youru Li. ,Zhenfeng Zhu. ,Deqiang Kong. , Hua Han. ,Yao Zhao.,”EA-LSTM: Evolutionary attention-based LSTM for time series prediction”,2019,01.

9. [9] R. Kail, E. Burnaev and A. Zaytsev, "Recurrent Convolutional Neural Networks Help to Predict Location of Earthquakes," in IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1-5, 2022, Art no. 8019005, doi: 10.1109/LGRS.2021.3107998.
10. [10] Bosilovich, M.G.; Robertson, F.R.; Chen, J. NASA’s Modern Era Retrospective-analysis for Research and Applications (MERRA). U.S. CLIVAR Var. 2006
11. [11] You, Y., Zhang, L., Tao, P., Liu, S., & Chen, L. (2022). Spatiotemporal Transformer Neural Network for Time-Series Forecasting. Entropy, 24(11), 1651. https://doi.org/10.3390/e24111651.
12. [12] Saranya A., Subhashini R., A systematic review of Explainable Artificial Intelligence models and applications: Recent developments and future trends,Decision Analytics Journal,Volume 7,2023,100230, ISSN 2772-6622.
13. [13] “What is explainable AI?” Available :https://www.ibm.com/ [Accessed :June 2020].