Please use this identifier to cite or link to this item:
http://hdl.handle.net/2122/16767
Authors: | Falsaperla, Susanna* Ferrari, Ferruccio* Langer, Horst* Spampinato, Salvatore* |
Title: | Using machine learning for the classification of Seismic Signals at Vulcano, Italy | Issue Date: | Jul-2023 | Publisher: | GFZ German Research Centre for Geosciences | URL: | https://gfzpublic.gfz-potsdam.de/pubman/item/item_5015999 | DOI: | 10.57757/IUGG23-0436 | Keywords: | seismic activity machine learning events classification Vulcano Aeolian Islands VLP seismicity |
Subject Classification: | 04.06. Seismology 04.08. Volcanology 05.06. Methods |
Abstract: | A Vulcanian eruption is described as an eruptive style with strong explosive characteristics. The name derives from the island of Vulcano in Italy, the first place in which it was observed during the last eruptive activity between 1888 and 1890. In this paper we analyze the seismicity recorded at Vulcano during a seismic unrest starting in September 2021 and still present as of November 2022. The distinctive feature of this seismicity is the presence of a variety of signals, most of which have a very long period (\textasciitilde0.5 s) signature. Low frequency content is interpreted as due to fluid involvement. Therefore, the high occurrence rate of VLP seismicity is a potential indication of pressure buildup within the volcanic system, and may herald phreatomagmatic activity (usually the first stage of a Vulcanian eruption), with serious consequences for inhabitants and tourists.Our analyses exploit machine learning procedures, with particular reference to pattern classification, at the aim of identifying varying classes of seismic events and trace their evolution over time. This classification can be useful for surveillance purposes contributing, along with other early warning methods, to reduce the devastating consequences of eruptions for people and property. |
Appears in Collections: | Conference materials |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Using machine learning for the classification of Seismic Signals at Vulcano, Italy __ GFZpublic.pdf | Abstract | 120.1 kB | Adobe PDF | View/Open |