Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

DeepShake: Shaking Intensity Prediction Using Deep Spatiotemporal RNNs for Earthquake Early Warning

Journal Article · · Seismological Research Letters
DOI:https://doi.org/10.1785/0220210141· OSTI ID:1981025
 [1];  [2];  [3];  [2];  [3]
  1. Electrical Engineering Department, Stanford University, Stanford, California, U.S.A.
  2. Computer Science Department, Stanford University, Stanford, California, U.S.A.
  3. Department of Geophysics, Stanford University, Stanford, California, U.S.A.

Abstract We propose a deep spatiotemporal recurrent neural network, DeepShake, to project future shaking intensity directly from current ground-motion observations. DeepShake is a network-based forecasting model, able to predict future shaking intensity at all stations within a network given previously measured ground shaking. The model is not given any a priori knowledge of station locations; instead, it learns wave propagation amplitudes and delays solely from training data. We developed DeepShake with the 35,679 earthquakes from the 2019 Ridgecrest sequence. Tasked with alerting for modified Mercalli intensity (MMI) IV+ shaking on 3568 validation earthquakes at least 5 s in advance, DeepShake achieves an equal error rate of 11.4%. For the Mw 7.1 earthquake that hit Ridgecrest on 5 July 2019, DeepShake was able to provide targeted alerts to all stations inside the network 5 s prior to the arrival of MMI IV+ waveforms. DeepShake demonstrates that deep spatiotemporal neural networks can effectively provide one-step earthquake early warning with reasonable accuracy and latency.

Research Organization:
Stanford Univ., CA (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
DOE Contract Number:
SC0020445
OSTI ID:
1981025
Journal Information:
Seismological Research Letters, Vol. 93, Issue 3; ISSN 0895-0695
Publisher:
Seismological Society of America
Country of Publication:
United States
Language:
English

References (20)

Real-time earthquake detection and hazard assessment by ElarmS across California journal January 2009
Earthquake Early Warning: Advances, Scientific Challenges, and Societal Needs journal May 2019
ShakeAlert Earthquake Early Warning System Performance during the 2019 Ridgecrest Earthquake Sequence journal July 2020
How low should we go when warning for earthquakes? journal November 2019
Demonstration of the Cascadia G‐FAST Geodetic Earthquake Early Warning System for the Nisqually, Washington, Earthquake journal June 2016
Remarks on Some Nonparametric Estimates of a Density Function book January 2011
Operational real-time GPS-enhanced earthquake early warning journal October 2014
CyberShake: A Physics-Based Seismic Hazard Model for Southern California journal May 2010
Long Short-Term Memory journal November 1997
Numerical Shake Prediction for Earthquake Early Warning: Data Assimilation, Real‐Time Shake Mapping, and Simulation of Wave Propagation journal May 2015
Transfer learning: improving neural network based prediction of earthquake ground shaking for an area with insufficient training data journal December 2021
The Propagation of Local Undamped Motion (PLUM) Method: A Simple and Robust Seismic Wavefield Estimation Approach for Earthquake Early Warning journal January 2018
Earthquake Early Warning ShakeAlert 2.0: Public Rollout journal April 2020
Real‐time inversions for finite fault slip models and rupture geometry based on high‐rate GPS data journal April 2014
Real-Time Performance of the PLUM Earthquake Early Warning Method during the 2019 M 6.4 and 7.1 Ridgecrest, California, Earthquakes journal June 2020
Earthquake transformer—an attentive deep-learning model for simultaneous earthquake detection and phase picking journal August 2020
The transformer earthquake alerting model: a new versatile approach to earthquake early warning journal December 2020
Deep Learning Model for Spatial Interpolation of Real-Time Seismic Intensity journal August 2020
About Feature Scaling and Normalization and the effect of standardization for machine learning algorithms text January 2014
Modified Mercalli intensity scale of 1931 journal December 1931

Similar Records

Geology and ground shaking: The April 25--26, 1992 Cape Mendocino earthquake sequence
Conference · 1993 · Geological Society of America, Abstracts with Programs; (United States) · OSTI ID:5024852