Interacting faults in california and hindu kush

Abstract

We study seismic fluctuations in California and Hindu Kush using Diffusion Entropy Analysis (DEA), a technique designed to detect the action of crucial events in time series generated by complex dynamical systems. The time distance between two consecutive crucial events is described by an inverse power law distribution density with a power index $\mu$ close to the value $\mu = 2$, corresponding to an ideal $1/f$ noise. DEA was used in the recent past to study neurophysiological processes that in the healthy condition are found to generate $1/f$ noise and $\mu$ close to $2$. In this paper we find that the seismic fluctuations in both California and Hindu-Kush of extended areas implying the action of many faults, $\mu \approx 2.1$, while in regions involving the action of only one fault, or of a very small number of faults, $\mu \approx 2.4$. This observation led us the conjectures that the seismic criticality is due to the interaction of many faults. To support this conjecture we adopted a dynamical model for fault dynamics proposed by Brown and Tosatti and we have extended it to describe the interaction between many faults. The DEA applied to surrogate sequences generated by this dynamical model, yields $\mu = 2.1$ for a single fault and $\mu = 2,4$ for many interacting faults, in a good agreement with the observation of real seismic fluctuations. This result supports our conjecture and suggests interesting applications to neurophysiological and sociological processes.