Coastal tsunami prediction in Tohoku region, Japan, based on S-net observations using artificial neural network

• Published in: Earth, planets, and space Vol. 75; no. 1; pp. 154 - 11
• Main Authors: Wang, Yuchen, Imai, Kentaro, Miyashita, Takuya, Ariyoshi, Keisuke, Takahashi, Narumi, Satake, Kenji
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2023; Springer; Springer Nature B.V; SpringerOpen
• Subjects: 2016 Fukushima tsunami; 2022 Tonga volcanic tsunami; 4. Seismology; Algorithms; Artificial neural networks; Bottom pressure; Data mining; Denoising autoencoder model; Earth and Environmental Science; Earth Sciences; Earthquake prediction; Earthquakes; Express Letter; Forecast accuracy; Gauges; Geology; Geophysics/Geodesy; Japan; Machine learning; Mathematical models; Model accuracy; Modelling; Neural networks; Offshore bottom pressure gauge; Pressure gages; Pressure gauges; Seismic activity; Tsunami early warning; Tsunami warning systems; Tsunamis; Waveforms; Japan; Tonga; Denoising autoencoder model; 2022 Tonga volcanic tsunami; 2016 Fukushima tsunami; Offshore bottom pressure gauge; Tsunami early warning
• ISSN: 1880-5981; 1343-8832; 1880-5981
• DOI: 10.1186/s40623-023-01912-6

We present a novel method for coastal tsunami prediction utilizing a denoising autoencoder (DAE) model, one of the deep learning algorithms. Our study focuses on the Tohoku coast, Japan, where dense offshore bottom pressure gauges (OBPGs), called S-net, are installed. To train the model, we generated 800 hypothetical tsunami scenarios by employing stochastic earthquake models (M7.0–8.8). We used synthetic tsunami waveforms at 44 OBPGs as input and the waveforms at four coastal tide gauges as output. Subsequently, we evaluated the model’s performance using 200 additional hypothetical and two real tsunami events: the 2016 Fukushima earthquake and 2022 Tonga volcanic tsunamis. Our DAE model demonstrated high accuracy in predicting coastal tsunami waveforms for hypothetical events, achieving an impressive quality index of approximately 90%. Furthermore, it accurately forecasted the maximum amplitude of the 2016 Fukushima tsunami, achieving a quality index of 91.4% at 15 min after the earthquake. However, the prediction of coastal waveforms for the 2022 Tonga volcanic tsunami was not satisfactory. We also assessed the impact of the forecast time window and found that it had limited effects on forecast accuracy. This suggests that our method is suitable for providing rapid forecasts soon after an earthquake occurs. Our research is the first application of an artificial neural network to tsunami prediction using real observations. In the future, we will use more tsunami scenarios for model training to enhance its robustness for different types of tsunamis. Graphical Abstract