GNGTS 2017 - 36° Convegno Nazionale

46 GNGTS 2017 S essione 1.1 Strong motion data analysis of Central Italy, 2016 seismic sequence A Gallo 1 , G Costa 1 , R de Nardis 2 , L Filippi 3 , G Lavecchia 2 , E Zambonelli 3 1 Department of Mathematics and Geosciences, University of Trieste, Italy 2 CRUST-DiSPUTer-Universiy of Chieti-Pescara “G. d’Annunzio”, Chieti Scalo, Italy 3 Dipartimento di Protezione Civile, Roma, Italy The strong motion data if appropriately analyzed can provide the best resource to estimate the physical parameters of seismic and engineering interest for a specific area for civil protection purpose. The SeisRaM (Seismological Research and Monitoring) Group of the Department of Mathematics and Geosciences of the University of Trieste is concerned to the seismic monitoring in real time. A procedure has been developed for the accurate determination of source parameters that are computed within few minutes after the earthquakes and rapidly revised. Thanks to a long collaboration with a National Civil Protection we improved the quality of the results and optimizes the analysis for civil defense purpose. The implementation of techniques that are able to determine the characteristics of a given event in near real time is essential for seismology and is of enormous importance to deal with the seismic emergency as quickly and accurately as possible. In this work we describe the development of our monitoring system and source parameters data analysis. Seismic moment and the corresponding moment magnitude are obtained from the spectrum of far-field body waves (Gallo et al. , 2014), as well as radiated energy, the Brune stress drop and apparent stress. Strong motion parameters and engineering parameters are also determined. The recent seismic sequence, occurred in the intra-Apennine extensional fault system of Central Italy, gave the opportunity to further test the near real time system. In fact, records from the seismic sequence following the Amatrice earthquake of 2016, 24 August (Lavecchia et al. , 2016), are analyzed using this automated routine. We analyzed the strong motion waveforms recorded by permanent and temporary stations of Italian strong Motion Network (RAN) associated to more than 2000 small and large events of the Central Italy seismic sequence. The earthquakes are located automatically by Antelope ® software (BRTT, Boulder). Moment magnitude estimates are reasonable agreement with the local magnitude and moment magnitude values calculated by other institutions. We compared our M W with M L calculated by Antelope ® software, with those by INGV reported on web site for the events with Mw > 3.5 and in Pondrelli et al. (2016) work with EMSC estimations. The differences between these estimates ranged between 0.3 and 0.5. Mw and its uncertainty are the median of M W at each stations and the L1-norm, respectively. This choice is due to their robustness compared with the mean and standard deviation. As reported in the statistical theory, the median and L1-norm are resistant to outliers in the data. This may be helpful in M W estimation where outliers may be Fig. 1 - Radiated energy plotted as function of seismic moment for events with MW > 3.5. In black the better fit of data.