HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Figures Only Full Text Full Text (PDF) Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Ferretti, G. Articles by Solarino, S. Search for Related Content GeoRef GeoRef Citation

An Improved Method for the Recognition of Seismic Families: Application to the Garfagnana–Lunigiana Area, Italy

¹ Dipteris
Dipartimento per lo studio del Territorio e delle sue Risorse
Università di Genova
Viale Benedetto XV, 5
16132 Genova, Italy
ferretti{at}dipeteris.unige.it
 (G.F., M.M.)

² INGV Istituto Nazionale di Geofisica e Vulcanologia
Rome
c/o Dipteris, Dipartimento per lo studio del Territorio e delle sue Risorse
Università di Genova
Viale Benedetto XV, 5
16132 Genova, Italy
 (S.S.)

With the aim to find a more objective way to detect seismic families, we applied a series of successive steps to constrain the results of a waveform similarity analysis. The evaluation of similarity was carried out on the waveforms recorded in the period 1999–2003 by the stations operating in the Garfagnana area, located in northern Tuscany (Italy). The algorithm is based on the cross-correlation technique applied in a process that overcomes the limit of one order of magnitude between events to be compared through a bridging technique. In practice, if two couples of events (A, B) and (B, C), each exceeding the correlation threshold, share a common quake (B), then all three events are attributed to the same family even if the match between A and C is below a value chosen as a reference for similarity.

To avoid any subjective choice of threshold for cross-correlation values, the results from the computation algorithm are submitted to a routine that gives increasing reliability to them if they are confirmed by the three components of the seismogram and if the number of families detected by each station is confirmed by more recordings. This latter constraint is made possible by the geometry of the recording network, with interdistances between stations of the order of 40–50 km. The process finally leads to the recognition of 27 families detected and confirmed by, on average, 3 stations that represent 40% of the recording capabilities. Since the performances of the recording network have been very odd in the past, especially in the early years of operation, the reliability of the detection is much higher, as in most cases the stations that detected the families were the only ones to be effectively recording.

The methodology proved to be more efficient than other methods applied in the past; moreover, the results could be probably improved even more if, instead of doing a one-run process, it would be borne as a trial-and-error approach.