The seismic quiescence hypothesis asserts that the number of small earthquakes decreases in and around the focal area of a great earthquake near to its time of occurrence. Discussion has continued as to whether this hypothesis is correct since it was proposed in 1906 by a Japenese seismologist, Fusakichi Omori. Various methods have been proposed to investigate the idea, often through statistical searches for spatial and temporal changes in seismicity. In 2017, Kei Katsumata investigated the seismicity in and around the Japan subduction zone from 1975 to 2012, but failed to find any evidence of seismic quiescence associated with the great earthquake which occurred on 15 November 2006 in the central Kurile Islands region. This event was a low-angle megathrust interplate earthquake on the upper surface of the Pacific Plate subducting beneath the North American or the Okhotsk Sea Plate.
As LML External Fellow Jiancang Zhuang and Katsumata note in a new study, there are several reasons for this failure using standard methods. To rectify these issues, they develop and apply a new method – the Poisson probability map (PMAP) method – based on calculating the probability for no earthquake to occur within a particular period. The longer the period when the earthquake does not occur, the smaller the probability; therefore, this probability becomes a P-value measure of seismic quiescence.
Using the PMAP method, the authors do find evidence for long-term seismic quiescence located very close to an asperity that was ruptured by the main shock of the 2006 Kurile Islands earthquake. They also introduce a p-value to measure the rarity of the seismic quiescence, with a smaller p-value implying lower probability. For the Kurile Islands earthquake, they identify a p-value of 9.0 x 10-5 for an interval beginning around 1990 and lasting for 15.4 years. The quiescence occurred within a circular area of radius 26 km centred at 153.8 E, 47.1 N. This seismic quiescence has not previously been recognized using any other method. The authors suggest that this method should be employed in a conditional intensity model, such as the stress release model, to provide better forecasts of the risk of large earthquakes.
The paper is available here.