Deep Learning Forecasts Caldera Collapse Events at Kilauea Volcano

TL;DR

This paper demonstrates that a deep learning graph neural network can predict caldera collapse events at Kilauea with high accuracy using limited early data, revealing underlying physics and outperforming baseline models.

Contribution

The study introduces a GNN-based approach for predicting caldera collapse timing using minimal early data, showing its effectiveness and physical interpretability.

Findings

GNN predicts failure within hours using only 0.5 days of data.

Prediction accuracy improves with more data and high-SNR tilt measurements.

GNN sensing correlates with underlying magma pressure decay physics.

Abstract

During the three month long eruption of Kilauea volcano, Hawaii in 2018, the pre-existing summit caldera collapsed in over 60 quasi-periodic failure events. The last 40 of these events, which generated Mw >5 very long period (VLP) earthquakes, had inter-event times between 0.8 - 2.2 days. These failure events offer a unique dataset for testing methods for predicting earthquake recurrence based on locally recorded GPS, tilt, and seismicity data. In this work, we train a deep learning graph neural network (GNN) to predict the time-to-failure of the caldera collapse events using only a fraction of the data recorded at the start of each cycle. We find that the GNN generalizes to unseen data and can predict the time-to-failure to within a few hours using only 0.5 days of data, substantially improving upon a null model based only on inter-event statistics. Predictions improve with increasing input data length, and are most accurate when using high-SNR tilt-meter data. Applying the trained GNN to synthetic data with different magma pressure decay times predicts failure at a nearly constant stress threshold, revealing that the GNN is sensing the underling physics of caldera collapse. These findings demonstrate the predictability of caldera collapse sequences under well monitored conditions, and highlight the potential of machine learning methods for forecasting real world catastrophic events with limited training data.