Transfer learning: Improving neural network based prediction of earthquake ground shaking for an area with insufficient training data

TL;DR

This paper explores transfer learning to enhance CNN-based earthquake ground shaking predictions in data-scarce regions, demonstrating improved accuracy and potential for early warning systems.

Contribution

It introduces transfer learning approaches to improve CNN predictions of earthquake ground shaking in areas with limited data, extending previous methods to new regions.

Findings

Transfer learning improves prediction accuracy and reduces outliers.

Adding station position information enhances model performance.

The method shows promise for real-time earthquake early warning applications.

Abstract

In a recent study (Jozinovi\'c et al, 2020) we showed that convolutional neural networks (CNNs) applied to network seismic traces can be used for rapid prediction of earthquake peak ground motion intensity measures (IMs) at distant stations using only recordings from stations near the epicenter. The predictions are made without any previous knowledge concerning the earthquake location and magnitude. This approach differs from the standard procedure adopted by earthquake early warning systems (EEWSs) that rely on location and magnitude information. In the previous study, we used 10 s, raw, multistation waveforms for the 2016 earthquake sequence in central Italy for 915 events (CI dataset). The CI dataset has a large number of spatially concentrated earthquakes and a dense station network. In this work, we applied the CNN model to an area around the VIRGO gravitational waves observatory sited near Pisa, Italy. In our initial application of the technique, we used a dataset consisting of 266 earthquakes recorded by 39 stations. We found that the CNN model trained using this smaller dataset performed worse compared to the results presented in the original study by Jozinovi\'c et al. (2020). To counter the lack of data, we adopted transfer learning (TL) using two approaches: first, by using a pre-trained model built on the CI dataset and, next, by using a pre-trained model built on a different (seismological) problem that has a larger dataset available for training. We show that the use of TL improves the results in terms of outliers, bias, and variability of the residuals between predicted and true IMs values. We also demonstrate that adding knowledge of station positions as an additional layer in the neural network improves the results. The possible use for EEW is demonstrated by the times for the warnings that would be received at the station PII.