Scalar model for frictional precursors dynamics

TL;DR

This paper presents a scalar model for frictional precursors that accurately reproduces experimental contact area dynamics and stress distributions, revealing the quasi-static nature of precursory slip and its implications for earthquake forecasting.

Contribution

The model integrates Archard's and Amontons' laws to simulate frictional precursors, highlighting their dependence on geometry and loading, and distinguishes elastic equilibrium effects from out-of-equilibrium dynamics.

Findings

Precursors depend non-universally on geometry and load conditions.

The model reproduces contact area and stress distributions with high accuracy.

Large Coulomb stress variations explain limitations of current earthquake forecasting methods.

Abstract

Recent experiments indicate that frictional sliding occurs by the nucleation of detachment fronts at the contact interface that may appear well before the onset of global sliding. This intriguing precursory activity is not accounted for by traditional friction theories but is extremely important for friction dominated geophysical phenomena such as earthquakes, landslides or avalanches. Here we simulate the onset of slip of a three dimensional elastic body resting on a surface and show that experimentally observed frictional precursors depend in a complex non-universal way on the sample geometry and the loading conditions. Our model satisfies Archard's law and Amontons' first and second laws, reproducing with remarkable precision the real contact area dynamics, the precursors' envelope dynamics prior to the transition to sliding, and the normal and shear internal stress distributions close to the slider-substrate interface. Moreover, it allows to assess which experimental features can be attributed to the elastic equilibrium, and which are attributed to the out-of-equilibrium dynamics, suggesting that precursory activity is an intrinsically quasi-static physical process. A direct calculation of the evolution of the Coulomb stress before and during precursors nucleation shows large variations across the sample, which helps to explain why earthquake forecasting methods based only on accumulated slip and Coulomb stress monitoring are often ineffective.