Earthquakes big and small: same physics, different boundary conditions

TL;DR

This paper demonstrates that rupture tip acceleration during earthquake initiation is self-similar across different fault sizes, with implications for earthquake magnitude prediction and early warning systems.

Contribution

It reveals that rupture acceleration during the start of dynamic rupture scales with fault size, extending the self-similarity concept to rupture tip dynamics.

Findings

Rupture tip acceleration scales with fault size during initiation.

Larger initial rupture patches lead to slower acceleration and potentially larger earthquakes.

Initial moment rate growth is slower for larger faults, aiding magnitude forecasting.

Abstract

Self-similarity indicates that large and small earthquakes share the same physics, where all variables scale with rupture length $L$. Here I show that rupture tip acceleration during the start of dynamic rupture (break-out phase) is also self-similar, scaling with $L_c$ in space and $L_c/C_{lim}$ in time (where $L_c$ is the breakout patch length and $C_{lim}$ the limiting rupture velocity in the subsonic regime). Rupture acceleration in the breakout phase is slower for larger initial breakout patches $L_c$. Because small faults cannot host large breakout patches, a large and slower initial breakout may be indicative of a potentially large final earthquake magnitude. Initial moment rate $\dot{M}_o$ also grows slower for larger $L_c$, therefore it may reflect fault dimensions and carry a probabilistic forecast of magnitude as suggested in some Early Warning studies. This result does not violate causality and is fully compatible with the shared fundamental, self-similar physics across all the magnitude spectrum.