LSTM-Based One-Day-Ahead Prediction of Ionospheric TEC Variations Associated with Major Earthquake Events in Japan and China
DOI:
https://doi.org/10.61326/jaasci.v4i1-2.432Keywords:
GNSS observations, Long Short-Term Memory, Time series forecasting, Total Electron ContentAbstract
Earthquakes pose significant threats to human life and infrastructure worldwide, motivating researchers to investigate potential precursory signals that may indicate impending seismic events. This study focuses on evaluating the capability of Long Short-Term Memory (LSTM) neural networks for time series prediction of IONOLAB-Total Electron Content (TEC) variations using Global Navigation Satellite Systems (GNSS) measurements during major earthquake events. We analyze IONOLAB-TEC data from 19 GNSS stations across Japan and China for four significant earthquakes: the 2011 Tohoku earthquake (Mw 9.1), the 2008 Iwate-Miyagi Nairiku earthquake (Mw 6.9), the 2008 Sichuan earthquake (Mw 7.9) and the 2010 Yushu earthquake (Mw 6.9). The LSTM model is trained using 10 days of TEC observations with 2.5-minute temporal resolution (576 observations per day) to forecast TEC values on the 11th day (earthquake day). Model performance is evaluated using three complementary metrics: Mean Squared Error (MSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Results demonstrate high prediction accuracy across most stations, with MAPE values below 2% for 14 out of 19 stations. The best performances are achieved at MIZU station (MAPE: 0.45%) for the Tohoku earthquake and XIAN station (MAPE: 0.62%) for the Sichuan earthquake. Overall MSE values range from 0.0036 to 0.4493 and MAE values range from 0.0420 to 0.5014. The findings demonstrate that LSTM networks can effectively learn and reproduce temporal patterns in IONOLAB-TEC time series, accurately capturing diurnal variations and magnitude fluctuations. This study demonstrates IONOLAB-TEC prediction capability rather than earthquake precursor detection or operational early warning capability.
References
Akhoondzadeh, M. (2012). Anomalous TEC variations associated with the powerful Tohoku earthquake of 11 March 2011. Natural Hazards and Earth System Sciences, 12, 1453-1462. https://doi.org/10.5194/nhess-12-1453-2012
Arikan, F., Erol, C. B., & Arikan, O. (2003). Regularized estimation of vertical total electron content from Global Positioning System data. Journal of Geophysical Research: Space Physics, 108(A12), 1-20. https://doi.org/10.1029/2002JA009605
Arikan, F., Erol, C. B., & Arikan, O. (2004). Regularized estimation of VTEC from GPS data for a desired time period. Radio Science, 39(6), 1-10. https://doi.org/10.1029/2004RS003061
Bouktif, S., Fiaz, A., Ouni, A., & Serhani, M. A. (2018). Optimal deep learning LSTM model for electric load forecasting using feature selection and genetic algorithm: Comparison with machine learning approaches. Energies, 11(7), 1636. https://doi.org/10.3390/en11071636
Budak, C., & Gider, V. (2023). LSTM based forecasting of the next day’s values of ionospheric total electron content (TEC) as an earthquake precursor signal. Earth Science Informatics, 16, 2323-2337. https://doi.org/10.1007/s12145-023-01027-2
Budak, C., Karatay, S., Erken, F., & Cinar, A. (2024). Classification of the ionospheric disturbances caused by geomagnetic and seismic activity with k-nearest neighbors algorithm. Wireless Personal Communications, 134, 1551-1569. https://doi.org/10.1007/s11277-024-10965-z
Cansu, T., Kolemen, E., Karahasan, Ö., Bas, E., & Egrioglu, E. (2023). A new training algorithm for long short-term memory artificial neural network based on particle swarm optimization. Granular Computing, 8, 1645-1658. https://doi.org/10.1007/s41066-023-00389-8
Elliott, J. R., Feng, W., Jackson, J. A., Parsons, B. E., & Walters, R. J. (2011). The 2010 MW 6.8 Yushu (Qinghai, China) earthquake: Constraints provided by InSAR and body wave seismology. Journal of Geophysical Research: Solid Earth, 116(B10), 1-16. https://doi.org/10.1029/2011JB008358
Erken, F., Karatay, S., & Cinar, A. (2019). Spatio-temporal prediction of ionospheric total electron content using an adaptive data fusion technique. Geomagnetism and Aeronomy, 59, 971-979. https://doi.org/10.1134/S001679321908005X
Gooya, H. R., Katibeh, H., & Maleki, A. (2024). Forecasting groundwater fluctuations caused by earthquakes using fuzzy logic and AHP Method: A case study from Iran. Earth Science Informatics, 17, 2143-2158. https://doi.org/10.1007/s12145-024-01264-z
Heelis, R. A., & Maute, A. (2020). Challenges to understanding the earth's ionosphere and thermosphere. Journal of Geophysical Research: Space Physics, 125(7), e2019JA027497. https://doi.org/10.1029/2019JA027497
Houdt, G. V., Mosquera, C., & Nápoles, G. (2020). A review on the long short-term memory model. Artificial Intelligence Review, 53, 5929-5955. https://doi.org/10.1007/s10462-020-09838-1
Hua, Y., Zhao, Z., Li, R., Chen, X., Liu, Z., & Zhang, H. (2019). Deep learning with long short-term memory for time series prediction. IEEE Communications Magazine, 57(6), 114-119. https://doi.org/10.1109/MCOM.2019.1800155
Hurvich, C. M., & Deo, R., & Brodsky, J. (1998). The mean squared error of Geweke and Porter-Hudak's estimator of the memory parameter of a long-memory time series. Journal of Time Series Analysis, 19(1), 19-46. https://doi.org/10.1111/1467-9892.00075
Iluore, K., & Lu, J. (2022). Long short-term memory and gated recurrent neural networks to predict the ionospheric vertical total electron content. Advances in Space Research, 70(3), 652-665. https://doi.org/10.1016/j.asr.2022.04.066
IONOLAB. (2006). Ionosphere research laboratory. https://www.ionolab.org/
Jia, K., Zhou, S., Zhuang, J., & Jiang, C. (2014). Possibility of the independence between the 2013 Lushan earthquake and the 2008 Wenchuan earthquake on Longmen Shan fault, Sichuan, China. Seismological Research Letters, 85(1), 60-67. https://doi.org/10.1785/0220130115
Karatay, S., Arikan, F., & Arikan, O. (2010). Investigation of total electron content variability due to seismic and geomagnetic disturbances in the ionosphere. Radio Science, 45(5), 1-12. https://doi.org/10.1029/2009RS004313
Karatay, S., Cinar, A., & Arikan, F. (2017). Ionospheric responses during equinox and solstice periods over Turkey. Advances in Space Research, 60(9), 1958-1967. https://doi.org/10.1016/j.asr.2017.07.038
Karatay, S. (2020a). Detection of the ionospheric disturbances on GPS-TEC using Differential Rate Of TEC (DROT) algorithm. Advances in Space Research, 65(10), 2372-2390. https://doi.org/10.1016/j.asr.2020.01.042
Karatay, S. (2020b). Estimation of frequency and duration of ionospheric disturbances over Turkey with IONOLAB-FFT algorithm. Journal of Geodesy, 94, 89. https://doi.org/10.1007/s00190-020-01416-1
Karatay, S., & Gul, S. E. (2023). Prediction of GPS-TEC on Mw > 5 earthquake days using bayesian regularization backpropagation algorithm. IEEE Geoscience and Remote Sensing Letters, 20, 1001005. https://doi.org/10.1109/LGRS.2023.3262028
Karatay, S., Arikan, F., & Pirti, A. (2025). Spatio-temporal evaluation of ionospheric disturbances before, during and after earthquakes using differential rate of TEC (DROT) from GPS measurements. Environmental Earth Sciences, 84, 222. https://doi.org/10.1007/s12665-025-12216-1
Li, T., Gao, Y., Chen, C.-H., Sun, Y.-Y., Zhang, X., Chum, J., Zhou, G., Wen, J., & Chen, X. (2025). Simulation of acoustic-gravity waves generated by an earthquake and explanation of the ionospheric disturbance observed during 2023 M 7.7 Turkey earthquake. Journal of Geophysical Research: Space Physics, 130(7), e2025JA033711. https://doi.org/10.1029/2025JA033711
Lindemann, B., Müller, T., Vietz, H., Jazdi, N., & Weyrich, M. (2021). A survey on long short-term memory networks for time series prediction. Procedia CIRP, 99, 650-655. https://doi.org/10.1016/j.procir.2021.03.088
Liu, J. Y., Chen, Y. I., Jhuang, H. K., & Lin, Y. H. (2004a). Ionospheric foF2 and TEC anomalous days associated with M >= 5.0 earthquakes in Taiwan during 1997-1999. Terrestrial Atmospheric and Oceanic Sciences, 15(3), 371-383. https://doi.org/10.3319/TAO.2004.15.3.371(EP)
Liu, J. Y., Chuo, Y. J., Shan, S. J., Tsai, Y. B., Chen, Y. I., Pulinets, S. A., & Yu, S. B. (2004b). Pre-earthquake ionospheric anomalies registered by continuous GPS TEC measurements. Annales Geophysicae, 22(5), 1585-1593. https://doi.org/10.5194/angeo-22-1585-2004
Manglik, A., Suresh, M., Babu, M. D., & Pavankumar, G. (2024). Pre- and coseismic electromagnetic signals of the Nepal earthquake of 03 november 2023. Earth, Planets and Space, 76, 173. https://doi.org/10.1186/s40623-024-02108-2
Meenalakshmi, M., & Valarmathi, K. (2025). Multigate long short-term memory-based stress detection from multimodal signal. Cognitive Computation, 17, 98. https://doi.org/10.1007/s12559-025-10444-y
Mienye, I., Swart, T., & Obaido, G. (2024). Recurrent neural networks: A comprehensive review of architectures, variants and applications. Information, 15(9), 517. https://doi.org/10.3390/info15090517
Mousavi, S. M., & Beroza, G. C. (2023). Machine learning in earthquake seismology. Annual Review of Earth and Planetary Sciences, 51, 105-129. https://doi.org/10.1146/annurev-earth-071822-100323
Namegaya, Y., Okamura, Y., Satake, K., & Shishikura, M. (2012). Challenges of anticipating the 2011 Tohoku earthquake and tsunami using coastal geology. Geophysical Research Letters, 39(21), 1-6. https://doi.org/10.1029/2012GL053692
Nathasarma, R., & Roy, B. K. (2023). Physics-informed long-short-term memory neural network for parameters estimation of nonlinear systems. IEEE Transactions on Industry Applications, 59(5), 5376-5384. https://doi.org/10.1109/TIA.2023.3280896
Nayir, H., Arikan, F., Arikan, O., & Erol, C. B. (2007). Total electron content estimation with Reg-Est. Journal of Geophysical Research: Space Physics, 112(A11), 1-11. https://doi.org/10.1029/2007JA012459
Nikolopoulos, D., Cantzos, D., Alam, A., Dimopoulos, S., & Petraki, E. (2024). Electromagnetic and radon earthquake precursors. Geosciences, 14(10), 271. https://doi.org/10.3390/geosciences14100271
Ogata, Y., & Toda, S. (2010). Precursory seismic anomalies and transient crustal deformation prior to the 2008 Mw = 6.9 Iwate-Miyagi Nairiku, Japan, earthquake. Journal of Geophysical Research: Solid Earth, 115(B10), 1-19. https://doi.org/10.1029/2010JB007567
Peng, Z., Meng, X., Liu, Q. Y., Guo, B., & Li, S. C. (2018). Evolution and distribution of the early aftershocks following the 2008 Mw 7.9 Wenchuan earthquake in Sichuan, China. Journal of Geophysical Research: Solid Earth, 123(9), 7775-7790. https://doi.org/10.1029/2018JB015575
Pirti, A., & Karatay, S. (2025). Multi-parameter GNSS analysis reveals atmospheric, ionospheric and crustal precursors seven days before the 2016 Mw 7.8 Ecuador megathrust earthquake. Earth Science Informatics, 18, 537. https://doi.org/10.1007/s12145-025-02037-y
Pulinets, S., Kotsarenko, A., & Pulinets, L. C. (2007). Special case of ionospheric day-to-day variability associated with earthquake preparation. Advances in Space Research, 39(5), 970-977. https://doi.org/10.1016/j.asr.2006.04.032
Qian, L., & Meng, G. (2022). Early postseismic deformation of the 2010 Mw 6.9 Yushu earthquake and its implication for lithospheric rheological properties. Geophysical Research Letters, 49(15), e2022GL098942. https://doi.org/10.1029/2022GL098942
Samaei, M., & Miyajima, M. (2022). Finite-fault stochastic simulation of the 2008 Iwate-Miyagi Nairiku, Japan, earthquake. Natural Hazards, 114, 1985-2012. https://doi.org/10.1007/s11069-022-05456-y
Sardón, E., Rius, A., & Zarraoa, N. (1994). Estimation of the transmitter and receiver differential biases and the ionospheric total electron content from Global Positioning System observations. Radio Science, 29(03), 577-586. https://doi.org/10.1029/94RS00449
Sezen, U., Arikan, F., Arikan, O., Ugurlu, O., & Sadeghimorad, A. (2013). Online, automatic, near-real time estimation of GPS-TEC: IONOLAB-TEC. Space Weather, 11(5), 297-305. https://doi.org/10.1002/swe.20054
Takyi-Aninakwa, P., Wang, S., Zhang, H., Li, H., Xu, W., & Fernandez, C. (2022). An optimized relevant long short-term memory-squared gain extended Kalman filter for the state of charge estimation of lithium-ion batteries. Energy, 260, 125093. https://doi.org/10.1016/j.energy.2022.125093
Tian, C., Ma, J., Zhang, C., & Zhan, P. (2018). A deep neural network model for short-term load forecast based on long short-term memory network and convolutional neural network. Energies, 11(12), 3493. https://doi.org/10.3390/en11123493
Tsai, H. F., & Liu, J. Y. (1999). Ionospheric total electron content response to solar eclipses. Journal of Geophysical Research: Space Physics, 104(A6), 12657-12668. https://doi.org/10.1029/1999JA900001
Uchida, N., & Bürgmann, R. (2021). A decade of lessons learned from the 2011 Tohoku-Oki earthquake. Reviews of Geophysics, 59(2), e2020RG000713. https://doi.org/10.1029/2020RG000713
USGS. (1879). United States geological survey. https://www.usgs.gov/
Wen, X., & Li, W. (2023). Time series prediction based on LSTM-Attention-LSTM model. IEEE Access, 11, 48322-48331. https://doi.org/10.1109/ACCESS.2023.3276628
Xiong, P., Zhai, D., Long, C., Zhou, H., Zhang, X., & Shen, X. (2021). Long short-term memory neural network for ionospheric total electron content forecasting over China. Space Weather, 19(4), e2020SW002706. https://doi.org/10.1029/2020SW002706
Yang, L., Wang, S., Chen, X., Chen, W., Saad, O. M., & Chen, Y. (2023). Deep-learning missing well-log prediction via long short-term memory network with attention-period mechanism. GEOPHYSICS, 88, D31-D48. https://doi.org/10.1190/geo2020-0749.1
Yang, T. Y., Lu, J. Y., Yang, Y. Y., Hao, Y. H., Wang, M., Li, J. Y., & Wei, G. C. (2025). GNSS–VTEC prediction based on CNN–GRU neural network model during high solar activities. Scientific Reports, 15, 9109. https://doi.org/10.1038/s41598-025-93628-8
Zhang, Q., Wang, H., Dong, J., Zhong, G., & Sun, X. (2017). Prediction of sea surface temperature using long short-term memory. IEEE Geoscience and Remote Sensing Letters, 14(10), 1745-1749. https://doi.org/10.1109/LGRS.2017.2733548
Zhang, X., Liu, J., Santis, A. D., Perrone, L., Xiong, P., Zhang, X., & Du, X. (2023). Lithosphere-atmosphere-ionosphere coupling associated with four Yutian earthquakes in China from GPS TEC and electromagnetic observations onboard satellites. Journal of Geodynamics, 155, 101943. https://doi.org/10.1016/j.jog.2022.101943
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Secil Karatay, Seyma Avci, Buse Bayram
This work is licensed under a Creative Commons Attribution 4.0 International License.
