GNGTS 2013 - Atti del 32° Convegno Nazionale

accelerometer at each floor and real-time transmission of the data streaming with acquisition at the Istituto Nazionale di Oceanografia e Geofisica Sperimentale (OGS). During the most active part of the seismic sequence it was thus possible to give, on request, information to local Civil Protection authorities about the damage state of the building that never reached level 1. Beside the practical application of monitoring, it was possible to compare ambient noise evaluation of fundamental periods with the variations observed during weak and strong motions to study some debated phenomena. Many recent experimental studies (Gallipoli et al. , 2009; Ditommaso et al. , 2013 and reference therein) have demonstrated the capability of the seismic ambient noise signal to estimate the main frequencies of building vibration in the elastic domain. Then the important aspect is how the frequency drop frequently observed during earthquake recordings on buildings could be considered a variation in elastic domain or a real decrement due to damage. During moderate earthquakes (PGA between 0.1 and 1 m/s 2 ), still in the elastic domain the opening of cracks induces non-linearities, that produce a recoverable frequency decrease reaching 35%. On the other hand, it seems that a 60% drop in frequency is a limit before the collapse according to data compiled by Calvi et al. (2006). So, it is very important to distinguish temporary vs permanent period elongation. In other Italian earthquakes it was possible to observe significative temporary elongation in periods that did not correspond to an increase of damage (see for L’Aquila 2009 earthquake, Mucciarelli et al. 2011) while in other cases the frequency shift was permanent and accompanied to observable damage (Molise, 2002 see Mucciarelli et al. , 2004; L’Aquila 2009, see Ditommaso et al. , 2013; Emilia, 2012 see Masi et al. , 2013). Description of building monitoring and data analysis. On September 2011, we installed at the second floor of the building (see Fig. 1) one ETNA-Kinemetrics accelerometer in stand- alone configuration with local data storage (geographic coordinates: Lat = 39.94938 ˚N; Lon = 16.04157 ˚E). The data were recorded at 200 sps. The NS and EW components of the accelerometer were oriented along the transversal and the longitudinal directions with respect to the main axes of the structure, respectively. We collected in this first acquisition stage 6 foreshocks with 3.0 ≤ Ml ≤ 3.6, the mainshock (Ml = 5.0), and 7 aftershocks with 2.8 ≤ Ml ≤ 3.3, as reported in Tab. 1. Tab. 1 – Events recorded by ETNA-Kinemetrics accelerometer. The mainshock is highlighted in grey. Origin Time (UTC) Latitude (˚N) Longitude (˚E) Depth (km) Magnitude Ml 2012-10-28 13:52:18.000 39.920 15.987 8.9 2.9 2012-10-28 03:37:46.000 39.925 16.007 8.8 3.1 2012-10-28 03:09:17.000 39.912 16.015 9.6 2.8 2012-10-28 00:30:44.000 39.932 16.004 9.4 2.9 2012-10-27 02:42:20.000 39.932 16.025 8.5 2.8 2012-10-26 02:25:09.000 39.920 16.032 6.6 2.9 2012-10-25 23:16:01.000 39.895 16.012 8.3 3.3 2012-10-25 23:05:24.000 39.881 16.009 6.3 5.0 2012-10-23 10:40:24.000 39.906 16.021 9.2 3.0 2012-10-18 02:51:57.000 39.887 16.034 7.8 3.5 2012-10-02 00:08:57.000 39.906 16.019 7.4 3.3 2012-10-01 21:27:51.000 39.903 16.010 7.9 3.3 2012-10-01 20:28:28.000 39.901 16.013 8.1 3.6 2012-09-28 05:56:46.000 39.912 16.087 3.0 3.0 442 GNGTS 2013 S essione 2.3