Stick-slip statistics for two fractal surfaces: A model for earthquakes

Bikas K. Chakrabarti,Robin B. Stinchcombe

DOI: https://doi.org/10.1016/S0378-4371%2899%2900146-6

1999-02-11

Abstract:Following the observations of the self-similarity in various length scales in the roughness of the fractured solid surfaces, we propose here a new model for the earthquake. We demonstrate rigorously that the contact area distribution between two fractal surfaces follows an unique power law. This is then utilised to show that the elastic energy releases for slips between two rough fractal surfaces indeed follow a Guttenberg-Richter like power law.

Statistical Mechanics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **Explain the statistical characteristics of elastic energy release in earthquakes, especially the phenomenon of following the Gutenberg - Richter law**. Specifically, the author proposes a new model to explain how the stick - slip behavior between two solids with fractal surfaces leads to the energy release in earthquakes following a power - law distribution.

### Problem Background

1. **Outer Structure of the Earth and Plate Movement**
- The solid outer layer (crust) of the Earth is about 20 kilometers thick and lies above the tectonic plates.
- Tectonic plates move relatively under the high - temperature and high - pressure phase change and convection in the Earth's mantle, at a speed of about several centimeters per year.
- The elastically strained energy accumulated over a long period generates huge stress between the crust and the moving tectonic plates. When the stress exceeds the frictional force, slippage occurs, releasing the stored energy and causing an earthquake.

2. **Gutenberg - Richter Law**
- The statistical characteristics of earthquake energy release usually follow a power - law distribution. That is, the relationship between the number of earthquakes \( N(m) \) and their magnitude \( m \) is:
\[
\ln N(m)=\text{constant}-a m
\]
- The relationship between energy \( \epsilon \) and magnitude \( m \) is:
\[
\ln \epsilon=\text{constant}+b m
\]
- Combining the above two equations, the distribution of the number of earthquakes \( N(\epsilon) \) with energy greater than or equal to \( \epsilon \) is obtained:
\[
N(\epsilon)\sim\epsilon^{-\alpha}
\]
where \( \alpha=\frac{a}{b} \), and usually the value of \( \alpha \) is between 0.7 and 1.0.

3. **Deficiencies of Existing Models**
- Existing models (such as the Burridge - Knopoff model, the self - organized critical model by Bak et al.) have successfully captured the Gutenberg - Richter law, but they have ignored the fractal nature of the contact surface between the crust and tectonic plates.

### New Contributions of the Paper

1. **Introduction of Fractal Surfaces**
- The contact surfaces between the crust and tectonic plates have fractal characteristics, and the roughness index \( \zeta \) is approximately 0.80 - 0.85.
- This fractal characteristic means that the contact surface exhibits self - similarity at different scales.

2. **Core Idea of the New Model**
- The author considers the distribution of contact areas between two fractal surfaces and assumes that the contact area is proportional to the elastic strain energy.
- When one fractal surface slides on another surface, the change in the contact area will lead to the release of elastic strain energy, forming an earthquake.

3. **Theoretical Derivation**
- Calculate the contact area distribution \( \rho(s) \) by the recursive group (RG) method and obtain the power - law distribution of the contact area \( s \):
\[
\rho(s)\sim s^{-\gamma}, \quad \gamma = 1
\]
- The elastic energy \( \epsilon \) is proportional to the contact area \( s \), and the slip frequency is related to the size of the contact area. Finally, the frequency distribution of earthquake energy release is obtained:
\[
n(\epsilon)\sim\epsilon^{-(\gamma+\delta)}, \quad \delta=\frac{1}{D}
\]
where \( D \) is the fractal dimension, usually less than 2.

4. **Result Verification**
- The power - law index \( \alpha=\delta=\frac{1}{D} \) predicted by the model is usually greater than 0.5, which is consistent with the observed values (0.7 to 1.0).

### Summary

This paper through...