Aerial view of the San Andreas Fault in California on the Carrizo Plain
New research proposes an explanation for why large faults, like the San Andreas Fault in California, seen here amid the Carrizo Plain, typically experience earthquakes that cluster in time and have a range of magnitudes—with many small events and a few large ones—whereas earthquake sequences on smaller seismic sources are sometimes characterized by periodic earthquakes of fixed size. Credit: John Wiley, CC BY 3.0
Source: Geophysical Research Letters

Many natural and human-made phenomena obey power law distributions. In one of the most well known examples, a power law distribution describes how small earthquakes occur much more frequently than large, potentially destructive ones.

Generally, the power law distribution in earthquake moment holds when considering seismic events over time on multiple faults. However, scientists still puzzle over the distribution of rupture sizes along individual faults. Exceptions to the power law statistics have been observed in rare sequences known as repeating earthquakes. Instead of many small events and a few large ones, these sequences are characterized by periodic earthquakes of fixed size, raising various questions for researchers. Why do these sequences depart from the otherwise ubiquitous power law statistics of earthquake sizes? And what distributions can we expect to occur on faults large enough to produce destructive earthquakes?

In a new theoretical study, Cattania explored the factors controlling earthquake statistics on a single isolated fault. The author used a two-dimensional earthquake cycle model of a simple fault experiencing both earthquakes and slow aseismic slip, or creep, and compared the results to records of earthquakes observed in nature.

The research revealed that although small seismic sources can produce identical and periodic earthquakes, tremors on large faults exhibit different traits, including the power law distributions observed in nature. For bigger faults, the rupture lengths of earthquakes may span several orders of magnitude and cluster in time—instead of spacing out more evenly as they do on small seismic sources. On the basis of straightforward physical concepts related to fault strain and the energy released during fracture formation, the study showed that the transition between these types of behavior is controlled by the ratio of a fault’s size to a length related to the earthquake nucleation dimension.

In essence, the study demonstrated that simple, isolated faults do not necessarily produce regular and periodic earthquakes, especially when the faults are relatively large. The conclusions offer insights into seismic hazard analysis. Although the simplified model used in the study may not adequately represent individual faults found in nature, the theory can be extended to more realistic cases. (Geophysical Research Letters,, 2019)

—Aaron Sidder, Freelance Writer


Sidder, A. (2019), Earthquake statistics vary with fault size, Eos, 100, Published on 21 October 2019.

Text © 2019. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.