Scaling and memory in seismological phenomena

TL;DR

This paper introduces a new method to characterize memory in complex systems, applies it to seismic data, and finds that earthquake energy release is memoryless in event time, while ground acceleration exhibits memory in sampling time.

Contribution

A novel approach for distinguishing memory effects in systems with power-law distributions is proposed and validated on seismic data.

Findings

Earthquake energy release is memoryless in event time.

Ground acceleration shows memory in sampling time.

The method effectively differentiates memory properties in complex systems.

Abstract

The concept of memory is of central importance for characterizing complex systems and phenomena. Presence of long-term memories indicates how their dynamics can be less sensitive to initial conditions compared to the chaotic cases. On the other hand, it is empirically known that the Feller-Pareto distribution, which decays as the power law i.e. the scale-invariant nature, frequently appears as a statistical law generated by the dynamics of complex systems. However, it is generally not a simple task to determine if a system obeying such a power law possesses a high degree of complexity with a long-term memory. Here, a new method is proposed for characterization of memory. In particular, a scaling relation to be satisfied by any memoryless dynamics generating the Feller-Pareto power-law distribution is presented. Then, the method is applied to the real data of energies released by a series of earthquakes and acceleration of ground motion due to a strong earthquake. It is shown in this way that the sequence of the released energy in seismicity is memoryless in the event time, whereas that of acceleration is memoryful in the sampling time.