Statistical moments of power spectrum: a fast tool for the classification of seismic events recorded on volcanoes

Abstract

Spectral analysis has been applied to almost thou-sand seismic events recorded at Vesuvius volcano (Naples,southern Italy) in 2018 with the aim to test a new tool fora fast event classification. We computed two spectral pa-rameters, central frequency and shape factor, from the spec-tral moments of order 0, 1, and 2, for each event at sevenseismic stations taking the mean among the three compo-nents of ground motion. The analyzed events consist ofvolcano-tectonic earthquakes, low frequency events and un-classified events (landslides, rockfall, thunders, quarry blasts,etc.). Most of them are of low magnitude, and/or low maxi-mum signal amplitude, therefore the signal to noise ratio isvery different between the low noise summit stations andthe higher noise stations installed at low elevation aroundthe volcano. The results of our analysis show that volcano-tectonic earthquakes and low frequency events are easily dis-tinguishable through the spectral moments values, particu-larly at seismic stations closer to the epicenter. On the con-trary, unclassified events show the spectral parameters valuesdistributed in a broad range which overlap both the volcano-tectonic earthquakes and the low frequency events. Since thecomputation of spectral parameters is extremely easy and fastfor a detected event, it may become an effective tool for eventclassification in observatory practice.