GNGTS 2014 - Atti del 33° Convegno Nazionale

GNGTS 2014 S essione 1.1 91 purpose is to increase the number of detected earthquakes and to improve the completeness of the NEI catalogue; possible false events will be checked and deleted by a final manual revision. A similar study was done by Garbin and Priolo (2013) who tested a semiautomatic offline procedure useful to analyse the microseismicity in the Trentino area. Performing the procedure on data recorded in year 2011 validates our configuration; it is worth to evidence that the final manual revision is anyway necessary to check and to validate the automatic results. Methodology. �������� Antelope ® is a system of software modules that implement the acquisition, transport, buffering, processing, archiving and distribution of environmental monitoring information. Antelope ® was constructed with an open systems design criterion that could be easily implemented by the users; automated real time processing of the data is performed and all data and information are automatically merged into long-term information system archives. The core of the system is the ORB (Object Ring Buffer) that buffers, processes and transports the data to another external ORB; seismic processing modules are provided byAntelope ® which implement all of the functions necessary for real-time automated detection, picking, association, location, magnitude estimation and archiving. We are interested to the algorithms that recognize and locate the seismic event, called the detection and the association modules. The detection is done by the STA/LTA algorithm on the continuous waveforms applying different bandpass filters; if a trigger threshold is exceeded the detection is declared open and it is closed when the STA/LTA ratio becomes lower than a detrigger threshold. These detections are utilized to match the estimated time computed on a 3D grid nodes for hypothetic origin time and fixed velocity model; the declaration of the earthquake is set by specific parameters as the minimum number of stations and the temporal time window utilized to associate the phases. Different filters in the detection correspond at different 3D spatial grids utilized in the association; for example the local events require the detection at higher frequencies and a search grid set up on a local area with an adequate resolution. At the end the solution obtained by the ‘trial and error’ association on the 3D grid is used as starting solution to perform the GENLOC inversion (Pavlis et al. , 2004) and to obtain a stable earthquake location. The GENLOC library (Pavlis et al. , 2004) uses the Gauss-Newton method (Lee and Stewart, 1981) proposed by Geiger (1910) with the partial derivatives similarly to HYPOELLIPSE algorithm (Klein, 1978) (see Tabs. 1 and 2 for further details about the configuration parameters). Fig. 1 – The earthquakes of the NEI bulletin of year 2011 recognized by our Antelope® procedure.