Occurrence Probability in Seismogenic Areas of DISS3.0.2 by Renewal Process: Importance of the Consistency between Model Assumptions and Input Data

In a previous project the long-term occurrence probability within various time intervals for the seismogenic areas of DISS3.0.2 was estimated on the basis of a nonparametric estimation of the density of the recurrence time between earthquakes of M >= 5.3. The main assumption is that the interevent times are calculated inside areas such that it is reasonable to assume that whenever an event occurs there, the loading process starts over again. This means that all the events recorded in that area are associated with a unique loading process , hence the interevent times follow the same distribution and are independent, whereas the hazard depends on the time elapsed since the last event. These assumptions may be unfulfilled if the sizes of the seismogenic areas are not homogeneous. This issue is taken into account comparing the estimates obtained by applying the same methodology on data sets generated by partitioning in different ways some seimogenic areas.