Attention-based learning for early earthquake magnitude estimation from strong-motion records
- 1Earthquake Research Institute, The University of Tokyo
- 2Institute of Engineering Mechanics, China Earthquake Administration
- 3Key Laboratory of Planetary Science and Frontier Technology, Institute of Geology and Geophysics, Chinese Academy of Sciences
Abstract
Rapid and reliable earthquake magnitude estimation in the first seconds after rupture initiation is critical for effective earthquake early warning and emergency response. However, the limited information contained in the earliest seismic signals makes this task challenging. Here we develop a deep-learning model, A2MAG, to estimate earthquake magnitude using only the first 3 s of strong-motion acceleration records from the dense K-NET network in Japan. The model is trained and tested on more than 130,000 waveforms from over 12,000 earthquakes recorded between 1996 and 2024. It achieves a mean absolute error of 0.33 magnitude units and performs consistently across different tectonic regions. The model systematically underestimates large earthquakes (M > 6), reflecting the longer rupture duration and limited early energy released by large events. Using the same input waveforms, the approach also provides highly accurate P-wave arrival times with a mean absolute error of 0.07 s. These results demonstrate both the potential and the fundamental limitations of estimating earthquake magnitude from the earliest seismic signals and provide preliminary evidence for the feasibility of the proposed approach in earthquake early warning applications.
Related
- A New Way for Determining Earthquake Magnitude Based on Electromagnetic Radiation
- On dependence of the source deactivation factor on the magnitude of main earthquake shock
- Is Earthquake Replenishment Sufficient for Unbiased b-Value Estimation?
- A New Amphibious Earthquake Catalog for the Alaska Peninsula: Training, Testing, and Applying Machine Learning Earthquake Picking in Offshore and Amphibious Environments
- Deep-learning-based denoising and interpolation of small displacements caused by fault slip using InSAR and GNSS data: Application to the 2019 Mw 6.2 Hyuga-nada earthquake
BibTeX
@article{zhang2026attentionbased,
title = {{Attention-based learning for early earthquake magnitude estimation from strong-motion records}},
author = {Ji Zhang and Aitaro Kato and Huiyu Zhu and Wei Wang},
journal = {npj Natural Hazards},
year = {2026},
month = jun,
doi = {10.1038/s44304-026-00242-3},
url = {https://doi.org/10.1038/s44304-026-00242-3},
note = {CC BY},
}
RIS
TY - JOUR
TI - Attention-based learning for early earthquake magnitude estimation from strong-motion records
AU - Ji Zhang
AU - Aitaro Kato
AU - Huiyu Zhu
AU - Wei Wang
PY - 2026
DA - 2026/06/01
JO - npj Natural Hazards
DO - 10.1038/s44304-026-00242-3
UR - https://doi.org/10.1038/s44304-026-00242-3
AB - Rapid and reliable earthquake magnitude estimation in the first seconds after rupture initiation is critical for effective earthquake early warning and emergency response. However, the limited information contained in the earliest seismic signals makes this task challenging. Here we develop a deep-learning model, A2MAG, to estimate earthquake magnitude using only the first 3 s of strong-motion acceleration records from the dense K-NET network in Japan. The model is trained and tested on more than 130,000 waveforms from over 12,000 earthquakes recorded between 1996 and 2024. It achieves a mean absolute error of 0.33 magnitude units and performs consistently across different tectonic regions. The model systematically underestimates large earthquakes (M > 6), reflecting the longer rupture duration and limited early energy released by large events. Using the same input waveforms, the approach also provides highly accurate P-wave arrival times with a mean absolute error of 0.07 s. These results demonstrate both the potential and the fundamental limitations of estimating earthquake magnitude from the earliest seismic signals and provide preliminary evidence for the feasibility of the proposed approach in earthquake early warning applications.
SN - 2948-2100
ER -