Nowcasting Earthquakes: Imaging the Earthquake Cycle in California With Machine Learning

John B. Rundle, Robert Granat, Andrea Donnellan, Geoffrey Fox, & James P. Crutchfield

Published November 24, 2021, SCEC Contribution #10929

We propose a new machine learning-based method for nowcasting earthquakes to image the time-dependent earthquake cycle. The result is a timeseries which may correspond to the process of stress accumulation and release. The timeseries is constructed by using Principal Component Analysis of regional seismicity. We first compute the characteristic spatial patterns for the region at time t using seismicity data for times t'  t in California. The patterns are found as eigenvectors of the cross-correlation matrix of a collection of seismicity timeseries in a coarse grained regional spatial grid (pattern recognition via unsupervised machine learning). The eigenvalues of this matrix represent the relative importance of the various eigenpatterns. Using the eigenvectors and eigenvalues, we then compute the weighted correlation timeseries (WCT) of the regional seismicity. This timeseries has the property that the weighted correlation generally decreases prior to major earthquakes in the region, and increases suddenly just after a major earthquake occurs. As in a previous paper (Rundle and Donnellan, 2020), we find that this method produces a nowcasting timeseries that resembles the hypothesized regional stress accumulation and release process characterizing the earthquake cycle. We propose that this nowcasting timeseries can be used as a proxy for judging the relative maturity of the current state of the California region in the recurring cycle of major earthquakes. We then address the problem of whether the timeseries contains information regarding future large earthquakes. For this we compute a Receiver Operating Characteristic and determine the decision thresholds for several future time periods of interest (optimization via supervised machine learning). We find that signals can be detected that can be used to characterize the information content of the timeseries. These signals may be useful in assessing present and near-future seismic hazard.

Citation
Rundle, J. B., Granat, R., Donnellan, A., Fox, G., & Crutchfield, J. P. (2021). Nowcasting Earthquakes: Imaging the Earthquake Cycle in California With Machine Learning. Earth and Space Science, 8(12). doi: 10.1029/2021EA001757.