Early Warning for Great Earthquakes From Characterization of Crustal Deformation Patterns With Deep Learning.

Although infrequent, large (Mw7.5+) earthquakes can be extremely damaging and occur on subduction and intraplate faults worldwide. Earthquake early warning (EEW) systems aim to provide advanced warning before strong shaking and tsunami onsets. These systems estimate earthquake magnitude using the early metrics of waveforms, relying on empirical scaling relationships of abundant past events. However, both the rarity and complexity of great events make it challenging to characterize them, and EEW algorithms often underpredict magnitude and the resulting hazards. Here, we propose a model, M‐LARGE, that leverages deep learning to characterize crustal deformation patterns of large earthquakes for a specific region in real‐time. We demonstrate the algorithm in the Chilean Subduction Zone by training it with more than six million different simulated rupture scenarios recorded on the Chilean GNSS network. M‐LARGE performs reliable magnitude estimation on the testing data set with an accuracy of 99%. Furthermore, the model successfully predicts the magnitude of five real Chilean earthquakes that occurred in the last 11 years. These events were damaging, large enough to be recorded by the modern high rate global navigation satellite system instrument, and provide valuable ground truth. M‐LARGE tracks the evolution of the source process and can make faster and more accurate magnitude estimation, significantly outperforming other similar EEW algorithms. This is the first demonstration of our approach. Future work toward generalization is outstanding and will include the addition of more training and testing data, interfacing with existing EEW methods, and applying the method to different tectonic settings to explore performance in these regions. Plain Language Summary: Great earthquakes are infrequent but devastating natural disasters. To mitigate their effects, earthquake early warning (EEW) systems aim to provide advance warning of strong shaking and tsunami. However, many of the most sophisticated EEW algorithms operating globally have a difficult time characterizing large earthquakes quickly and accurately enough to issue a meaningful warning—this is most evident from the failure of EEW during the 2011 M9 Tohoku Oki, Japan earthquake. Here, we propose a model, M‐LARGE, that learns earthquake's surface deformation patterns from millions of simulations for a specific region, and then applies it to unseen events in the same region. Our model shows a high accuracy of 99% performing on the testing data set and accurately estimates the magnitude of five real large historical events in Chile. The M‐LARGE outperforms currently operating similar EEW algorithms. This is the first demonstration of our approach, we note that for actual EEW operation, future work including the addition of more training and testing data, interacting with existing EEW methods, and testing the method in alternative tectonic settings are necessary prior to real‐time operations. Key Points: We use synthetic data to train a deep learning model to predict magnitude from crustal deformation patterns in simulated real‐timeThe model, M‐LARGE, has an accuracy of 99% and accurately estimates the magnitude of five real large events, outperforming other methodsM‐LARGE's rapid and accurate magnitude prediction suggesting that significant warning times are possible during real large earthquakes [ABSTRACT FROM AUTHOR]