Hibrid broadband Ground-Motion Simulations for the 2016 Amatrice Earthquake, Central Italy, and Sensitivity of Ground-Motion to Earthquake Source Parameters
On 24th August 2016 at 01:36 UTC a MW 6.0 earthquake struck several villages in central Italy, among which Accumoli, Amatrice and Arquata del Tronto. It caused 299 fatalities, major destruction and extensive damage in the surrounding area (up to 11 intensity degree). The earthquake was recorded by 350 digital accelerometers be- longing to the National Accelerometric Network (RAN) of the Italian Department of Civil Protection, to the National Seismic Network (Rete Sismica Nazionale, RSN) of the Istituto Nazionale di Geofisica e Vulcanologia (INGV), and to other local net- works. This earthquake ruptured a NW–SE oriented normal fault, according the prevailing extensional tectonics of the area.The maximum accelera- tion was observed at Amatrice station (AMT) with epicentral distance of 15 km, reaching 916 cm/s2 and 445.6 cm/s2 on E-W and N-S components, respectively. Motivated by the high levels of observed ground motion and damage, we have computed synthetics broadband time series for engineering purposes. To produce high-frequency seismograms, we have used a stochastic finite-fault model approach based on dynamic corner-frequency.
Source characteristic of 2000 small earthquakes nucleating on the Alto Tiberina fault system (Central Italy)
The Alto Tiberina Fault (ATF) is a 60 km long east-dipping low-angle normal fault, located in a sector of the Northern Apennines (Italy) undergoing active extension since the Quaternary (Chiaraluce et al. 2007). The ATF has been imaged by analyzing the active source seismic reflection profiles, and the instrumentally recorded persistent background seismicity. The present study is an attempt to separate the contributions of source, site, and crustal attenuation, in order to focus on the mechanics of the ATF, as well as the syn- and antithetic structrure related on the ATF hanging-wall (i.e. Gubbio fault and Umbria Valley fault). In order to compute source spectra, we perform a set of regression over the seismograms of ~ 400 small earthquakes (0.5 < ML < 3.0) recorded between 2010 and 2014 at 50 permanent seismic stations deployed in the framework of the Alto Tiberina Near Fault Observatory project (TABOO; Chiaraluce et al., 2014), three of which located in shallow boreholes. Because we deal with some very small earthquakes, we maximize the signal to noise ratio (SNR) with a technique based on the analysis of peak values of bandpass-filtered time histories, in addition to the same processing performed on Fourier amplitudes. We rely on Random Vibration Theory (RVT, Cartwright and Longuet-Higgins, 1956) to completely switch from peak values in the time domain to Fourier spectral amplitudes. So far, highly accurate, stable source spectra have been used to compute moment magnitudes (Mw) of all the events in the present data set, whereas in future developments the same data will be used to gain insights into the underlying mechanics of faulting and the earthquake processes.