Hastening of occurrences of earthquakes due to dynamic triggering: The observation at Matsushiro, central Japan

Dynamic stress/strain changes associated with the passage of seismic waves perturb the state of stress of a fault.We hypothesize that this perturbation increases the instability of the fault and that it hastens the occurrence of an earthquake that would otherwise occur somewhat later.To test this hypothesis, we investigate a time interval defined as a time difference between the occurrence of a dynamic strain change and the origin time of the first subsequent earthquake.If the occurrence of an earthquake is hastened by the strain change, the time interval would be shortened, compared with a case of a random occurrence.Here we used two datasets: 1) origin times of earthquakes at Matsushiro, Nagano Prefecture, central Japan, with magnitudes of 1.6 or greater, between November 1984 and December 1994; and 2) strainmeter records of remote earthquakes at the Matsushiro Seismological Observatory over the same period.We applied a statistical test to the distribution of the observed time intervals between dynamic strain changes due to remote earthquakes and the first earthquakes at Matsushiro subsequent to the strain changes, in order to compare it to the distribution of expected time intervals generated by a random (Monte Carlo) simulation.Because of limitations of the statistical test, we could not establish the quantitative relationship between the degree of hastening and the amplitude of the strain changes, but we found a statistically significant decrease of the observed time intervals.We also investigated the number of the earthquakes before and after the strain changes, and found that dynamic triggering has little significant impact on the occurrences of earthquakes at Matsushiro.Therefore, we conclude that dynamic triggering at Matsushiro is weaker than those reported in previous studies and that the time interval might be an important parameter for a statistical study of weak dynamic triggering.