Journal Article

Detection of Low Frequency Seismicity at Mt. Vesuvius Based on Coherence and Statistical Moments of Seismic Signals

Di Maio, Rosa, Manzo, Roberto, La Rocca, Mario, Galluzzo, Danilo, Nardone, Lucia

2023

Overview

Mt. Vesuvius is a high-hazard active volcano surrounded by a densely populated area. Since human activities generate high levels of seismic noise, recognizing low-amplitude seismic events in the signals recorded by the local seismic monitoring network operating at Vesuvius is very difficult. Here, we describe an automatic procedure applied to continuous data with the aim of finding low-amplitude–low-frequency events hidden in the recorded signals. The methodology is based on the computation of two spectral parameters, central frequency Ω and shape factor ẟ, at selected sites, and the coherence of the seismic signal among different sites. The proposed procedure is applied to 28 months of recordings from 2019 to 2021, tuning the search parameters in order to find low-frequency signals similar to those occasionally observed in the past at the same volcano. The results allowed us to identify 80 seismic events that have the spectral features of low-frequency earthquakes or tremor. Among these, 12 events characterized by sufficiently high signal-to-noise ratio have been classified as deep low-frequency earthquakes, most of which are not reported in the catalog. The remaining events (more than 60) are characterized by similar spectral features but with an extremely low amplitude that prevents any reliable location of the source and definitive classification. The results of this work demonstrate that the low-frequency endogenous activity at Mt. Vesuvius volcano is more frequent that previously thought.

• Publisher: MDPI AG
• Subject: Classification / coherence analysis / Earthquakes / low-frequency earthquakes / Neural networks / signal detection / spectral parameters / Surveillance / volcano seismicity / Volcanoes