GNGTS 2015 - Atti del 34° Convegno Nazionale

measurement and preventing misinterpretation of low-quality measurements, a reliability classification scheme was developed in the frame of the seismic microzoning activities following the 2009 L’Aquila earthquake by Albarello et al. (2011). This classification, is more conservative than the one proposed by the SESAME group (SESAME 2005), since it includes more criteria. Albarello et al. (2011) defines three classes: • class A: trustworthy and interpretable HVSR curve, it represents a reference measurement that can be considered by itself for the site of concern; • class B: ambiguous HVSR curve, should be used with caution and in case of coherency with other measurements performed nearby; • class C: poor quality HVSR curve (hardly interpretable), to be discarded. The majority of measurements of ambient noise can be grouped in classes A and B while about less than 10% are poor quality HVSR curves. Resonance peaks detection. In order to physically interpret the curve in terms of “absence/ presence” of resonance phenomena, two situations should be distinguished: 1. �������� ���������� ��� ���� ����� �������� �� ����� ��� ����� ���� �� ��� ��������� possible resonance: the HVSR curve presents at least one clear peak in the frequency range where the measurement could be considered reliable; 2. ��� ���� ����� ���� ��� ������� ��� ����� ���� �� ��� ��������� ����� ����� ��� the HVSR curve does not present any clear peak in the frequency range where the measurement could be considered reliable: flat (F) or broad-band (BB) curves. To provide such distinction, an automatic procedure was adopted (Puglia et al. , 2011) to identify the resonances peaks, their reliability and the sharpness of the HVSR curve nearby each of these values. Though the procedure it also to identify the broad-band or flat HVSR curves. In case of resonance, we identify three different frequency – amplification bands: f0≤1 Hz Low frequency (LF) 1 Hz<f0≤4 Hz Intermediate frequency (MF) f0>4 Hz High frequency (HF) and three amplitude – amplification range: 2.7<Af0≤4.2 Small amplitude (LA) 4.2<Af0≤5.7 Medium amplitude (MA) Af0>5.7 Large amplitude (HA) We apply these criteria to the noise measurement set collected in the Itaca database. Tab. 2 summarizes the results. The flat HVSR curves are the majority, however no clear correlation are found with the EC8 soil categories since most of the recording stations are classified on the base of geological information. Tab. 2 - Results of frequency – amplification bands and amplitude – amplification range of HVSR curves collected in ITACA. amplitude – amplification range LA HA MA F 129 frequency – BB 30 27 0 3 amplification LF 65 37 14 14 range MF 56 29 11 16 HF 48 26 11 11 Site classification in ITACA 2.0. The geographic distribution of the 922 recording stations, that will be available in the next release of ITACA (ITACA 2.0), is shown in Fig. 3 (left panel). The majority of the stations (601) belongs to RAN (network IT) operated by DPC, while the remaining 321 are part of other networks. GNGTS 2015 S essione 2.2 125