The temporal limits of predicting fault failure

TL;DR

This study investigates whether seismic acoustic emissions from laboratory fault experiments contain predictive information about near-future frictional failure, using advanced machine learning models with interpretability features.

Contribution

It demonstrates that AE signals contain near-term predictive information about fault failure, utilizing a transformer-based model to interpret the regions of the signal relevant for prediction.

Findings

AE signals contain near-term predictive information about fault failure.

Prediction accuracy decreases as the forecast window extends further into the future.

Transformer attention maps reveal which parts of the AE signal are most informative.

Abstract

Machine learning models using seismic emissions can predict instantaneous fault characteristics such as displacement in laboratory experiments and slow slip in Earth. Here, we address whether the acoustic emission (AE) from laboratory experiments contains information about near-future frictional behavior. The approach uses a convolutional encoder-decoder containing a transformer layer. We use as input progressively larger AE input time windows and progressively larger output friction time windows. The attention map from the transformer is used to interpret which regions of the AE contain hidden information corresponding to future frictional behavior. We find that very near-term predictive information is indeed contained in the AE signal, but farther into the future the predictions are progressively worse. Notably, information for predicting near future frictional failure and recovery are found to be contained in the AE signal. This first effort predicting future fault frictional behavior with machine learning will guide efforts for applications in Earth.