Spatiotemporal Evolution of a Landslide: A Transition to Explosive Percolation

Area of Science:

• Geophysics and Earth Science
• Physics of Complex Systems
• Material Science

Background:

• Granular material failure is often preceded by distinct motion patterns.
• Existing methods for forecasting granular failure from motion data are insufficient.
• Advances in remote sensing and imaging generate vast amounts of relevant data.

Purpose of the Study:

• To develop a timely and accurate method for forecasting granular failure using motion data.
• To leverage percolation theory and graph analysis for failure prediction.
• To bridge the gap between laboratory-scale and field-scale granular failure signatures.

Main Methods:

• Mapping granular motion data to time-evolving graphs.
• Analyzing graph evolution through the framework of explosive percolation.
• Validating the approach with simulated granular tests and real landslide ground motion data.

Main Results:

• A critical transition to explosive percolation was identified at the point of imminent failure.
• The emerging connected components in the graphs accurately predicted failure locations.
• Spatiotemporal dynamics were revealed that are consistent across simulations and real-world events.

Conclusions:

• The explosive percolation model provides a robust signature for predicting granular material failure.
• This method offers a promising tool for early warning, forecasting, and mitigation of landslides and other catastrophic granular events.
• The findings demonstrate a unified approach to understanding granular failure precursors from bench to field scales.