Effect of Time Dependence on Probabilistic Seismic-Hazard Maps and Deaggregation for the Central Apennines, Italy

We produce probabilistic seismic-hazard assessments for the central Apennines, Italy, using time-dependent models that are characterized using a Brownian passage time recurrence model. Using aperiodicity parameters, alpha of 0.3, 0.5, and 0.7, we examine the sensitivity of the probabilistic ground motion and its deaggregation to these parameters. For the seismic source model we incorporate both smoothed historical seismicity over the area and geological information on faults. We use the maximum magnitude model for the fault sources together with a uniform probability of rupture along the fault (floating fault model) to model fictitious faults to account for earthquakes that cannot be correlated with known geologic structural segmentation. We show maps for peak ground acceleration (PGA) and 1.0 Hz spectral acceleration (SA(1)) on rock having 10% probability of exceedence in 50 yr. We produce maps to compare the separate contributions of smoothed seismicity and fault components. In addition we construct maps that show sensitivity of the hazard for different alpha parameters and the Poisson model. For the Poisson model the addition of fault sources to the smoothed seismicity raises the hazard by 50% at locations where the smoothed seismicity contributes the highest hazard and up to 100% at locations where the hazard from smoothed seismicity is low. For the strongest aperiodicity parameter (smallest alpha), the hazard may further increase 60%-80% or more or may decrease by as much as 20% depending on the recency of the last event on the fault that dominates the hazard at a given site. In order to present the most likely earthquake magnitude and/or the most likely source-site distance for scenario studies, we deaggregate the seismic hazard for SA(1) and PGA for two important cities (Rome and L'Aquila). For PGA both locations show the predominance of local sources having magnitudes of about 5.3 and 6.5, respectively. For SA(1) at a site in Rome, there is significant contribution from local smoothed seismicity and an additional contribution from the more distant Apennine faults having magnitude around 6.8. For L'Aquila the predominant sources remain local. In order to show the variety of impact of different alpha values, we also obtained deaggregations for another three sites. In general, as alpha decreases (periodicity increases), the deaggregation indicates that the hazard is highest near faults with the highest earthquakes rates. This effect is strongest for the long-period (1 sec) ground motions.