Universal Non-Extensive Statistical Physics Temporal Pattern of Major Subduction Zone Aftershock Sequences

Abstract

Large subduction-zone earthquakes generate long-lasting and wide-spread aftershock sequences. The physical and statistical patterns of these aftershock sequences are of considerable importance for better understanding earthquake dynamics and for seismic hazard assessments and earthquake risk mitigation. In this work, we analyzed the statistical properties of 42 aftershock sequences in terms of their temporal evolution. These aftershock sequences followed recent large subduction-zone earthquakes of M ≥ 7.0 with focal depths less than 70 km that have occurred worldwide since 1976. Their temporal properties were analyzed by investigating the probability distribution of the interevent times between successive aftershocks in terms of non-extensive statistical physics (NESP). We demonstrate the presence of a crossover behavior from power-law (q ≠ 1) to exponential (q = 1) scaling for greater interevent times. The estimated entropic q-values characterizing the observed distributions range from 1.67 to 1.83. The q-exponential behavior, along with the crossover behavior observed for greater interevent times, are further discussed in terms of superstatistics and in view of a stochastic mechanism with memory effects, which could generate the observed scaling patterns of the interevent time evolution in earthquake aftershock sequences.

Complexity of Recent Earthquake Swarms in Greece in Terms of Non-Extensive Statistical Physics

“…The cumulative inter-event time and distance distribution functions for each earthquake swarm are well described with the Q -exponential function, indicating asymptotic power–law behavior and long-term correlations in the spatiotemporal evolution of seismicity. In addition, the q -values of q T > 1 and q D < 1 are in agreement with the q -values found for aftershock sequences [ 38 , 40 , 83 , 93 , 94 ], for the Hellenic Subduction Zone [ 91 ], for the temporal properties of seismicity [ 60 , 61 ] and for global seismicity [ 41 , 84 ]. In addition, the concept of NESP describes well both the spatial and temporal behavior of the earthquake swarms in diverse tectonic environments [ 7 , 18 , 26 , 95 ] and in volcanic regions [ 23 , 59 ].…”

Forecasting Strong Subsequent Earthquakes in Greece with the Machine Learning Algorithm NESTORE

“…It is widely known that large earthquakes are followed by other earthquakes, usually smaller and occurring in close proximity, days to years later, and that it takes some time for seismicity to return to normal levels [1][2][3][4]. However, it may happen that the following earthquake magnitude is comparable with the previous one.…”