GNGTS 2013 - Atti del 32° Convegno Nazionale

analysis of the inter-event times (IETs) distribution pattern could represent an important tool to distinguish among sectors affected by different stress fields (e.g. Bell and Kilburn, 2008; Traversa and Grasso, 2010). An IET is defined as the waiting time between two consecutive earthquakes, related to a specific interval of time and a particular threshold of magnitude. The waiting time distribution of global or national size catalogues is usually modelled using a gamma law (Corral, 2003). This single-peaked distribution, typical of tectonic areas, is clearly different from the IET distribution for regional or local catalogues that generally has a bimodal shape deriving from the combination of two distributions, one due to correlated events (which have short inter-event time) and the other due to independent events (which tend to be separated by longer gaps) (Naylor et al. , 2010). Therefore, for a small region, the bimodal shape of IET distribution is heavily influenced by the high percentage of correlated events (aftershocks). IET analysis in the Etnean area was already performed by Sicali et al. (2012) investigating the seismic events occurred during the time interval 1988-2011 (Patanè e t al. , 2004; Gruppo Analisi Dati Sismici, 2011). The authors, aimed to identify the existence of either a periodicity or a stationary behaviour of the seismic activity and tried to correlate it with the volcano-tectonic features of the region. As a result of this study, the presence of different volcano sectors showing specific behaviour was set into evidence. The comparison between the spatial variation of Etna IET distributions with those obtained for Sicily and Italy, showed that at a large scale the IETs are well-modelled by a gamma distribution, whereas at the local Etnean scale the IETs are characterized by a bimodal curve. The two peaks are related to: (i) the background regional stationary seismicity, (ii) the contribution of the local seismic swarms with very short inter-event times, which considerably modify the usual seismic rate. Sicali et al. (2012) concluded that the seismicity taking place at depth shallower than 5 km is almost entirely represented by short IETs and is mainly confined to the Etna summit area. On the other hand, earthquakes deeper than 5 km, appear mainly linked to the regional tectonic setting. In particular, the eastern flank seismicity is influenced by the extensional regional tectonics typical of the eastern Sicily, whereas the western flank seismicity seems consistent with the compressional processes observed at a regional scale. Such findings further support the evidences that Mt. Etna is located at the boundary of two different tectonic domains (Neri et al. , 2005; La Vecchia et al. , 2007; Palano et al ., 2012). In the present study a more thorough analysis, at a scale of the sectors previously identified, is performed in order to investigate about possible correlations between the occurrence of Etnean eruptions and its seismicity. The dataset used is now extended back to the period 1976-1987, matching the date coming from the IIV, Poseidon and INGV-CT seismic networks (Gruppo Analisi Dati Sismici, 2011) with those recorded by the seismic network run by University of Catania. This allow us to analyze a longer and more significant range of time which spans from 1976 to 2011, evaluating both the space-time IET distributions and the earthquake cumulative patterns in different sectors of the volcano. Seismic data and method. The used catalogue consists of 12,645 earthquakes occurred at Mt. Etna from January 1976 to December 2011, which were recorded through the permanent seismic networks run by the University of Catania (1976-1987) and by the IIV-CNR (Istituto Internazionale di Vulcanologia), Sistema Poseidon and INGV-CT (Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Catania) from 1988 up to 2011 (Patanè et al. , 2004; Gruppo Analisi Dati Sismici, 2011). Moreover, in addition to the permanent network, data recorded by several temporary seismic networks, held by different institutions, were added when available in order to reduce the gap in the azimuthal distribution of the recording stations. With the aim of having an homogenous catalogue, all recorded earthquakes were relocated using the HypoEllipse algorithm (Lahr, 1989) and the velocity model proposed by Hirn et al. (1991). The sensitivity of the seismic monitoring network at Mt. Etna (gap, erh, erz, rms, etc.) was gradually enhanced several times between 1976 and 2011 so that, the minimum magnitude of earthquakes located in the study area, varied over the time following the network upgrade. 278 GNGTS 2013 S essione 1.3