Broadband Simulations of Ground Motions for the M6.5 Norcia Earthquake, 2016-2017 Central Italy Earthquake Sequence

The use of simulated seismograms may improve the seismic hazard and earthquake risk assessment when an accurate modelling of both source and propagation is implemented. To build earthquake rupture scenarios, the absolute ground motion level, frequency content, signal duration and distribution of simulated ground motion are highly dependent on the input parameters adopted in the calculations. Therefore ground motion simulations of occurred events are needed to test and improve the efficiency and performance of our procedures to estimate realistic time series.

Recently, a series of large earthquakes, M>6, occurred at near the town of Amatrice and Norcia in the Central Apennines, during the 2016-2017 central Italy earthquake sequence. These events are recorded over a dense seismic network, providing relevant observational evidence of seismic complexities both in time and space. In this study we benefit these dataset and analyze the ground motion characteristics of these events, in particular the M6.5 30 October 2016 Norcia event, comparing with both empirical and physics-based ground motion models.

Broadband seismograms are obtained by merging the three components high-frequency seismograms calculated throughout a stochastic finite-fault model approach that includes P, SV and SH soil amplification transfer functions (Ruiz et al, 2018) with synthetic low frequency waveforms to cover the entire frequency band of engineering interest. Low frequency seismograms are inferred by using the heterogeneous slip rupture model of Scognamiglio et al (2018). In order to predict earthquake-induced ground motions in the area, we adopted region-specific attenuation and source scaling parameters derived by Malagnini et al (2011).

Ground motion parameters including peak ground acceleration, velocity and spectral accelerations at different periods computed at different sites were displayed in order to highlight the azimuthal variations that depend on the fault geometry and rupture model. The hybrid broadband horizontal component time histories and Fourier and response spectra are presented for the selected sites. The performances of the hybrid ground motion simulations in the study region are discussed assessing the residuals and the level of fit between our simulations, observed and empirical ground motion models.