AI-based anatomy of the continuous seismic wavefield at Sos Enattos (Sardinia, 
Italy) over one year of data

Zerafa, Christopher; Leonard, Seydoux; Giunchi, Carlo; Cianetti, Spina; Naticchioni, Luca; D'Urso, Domenico

doi:10.57757/IUGG23-4332

Item

ITEM ACTIONSEXPORT

Add to Basket

Local TagsRelease HistoryDetailsSummary

Released

Conference Paper

AI-based anatomy of the continuous seismic wavefield at Sos Enattos (Sardinia, Italy) over one year of data

Authors

Zerafa, Christopher
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Leonard, Seydoux
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Giunchi, Carlo
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Cianetti, Spina
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Naticchioni, Luca
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

D'Urso, Domenico
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

External Ressource

No external resources are shared

Fulltext (public)

There are no public fulltexts stored in GFZpublic

Supplementary Material (public)

There is no public supplementary material available

Citation

Zerafa, C., Leonard, S., Giunchi, C., Cianetti, S., Naticchioni, L., D'Urso, D. (2023): AI-based anatomy of the continuous seismic wavefield at Sos Enattos (Sardinia, Italy) over one year of data, XXVIII General Assembly of the International Union of Geodesy and Geophysics (IUGG) (Berlin 2023).
https://doi.org/10.57757/IUGG23-4332

Cite as: https://gfzpublic.gfz-potsdam.de/pubman/item/item_5021766

Abstract

Sardinia is a seismically quiet region: for this reason, it has been proposed as an excellent location to host fundamental physics experiments requiring low seismic ambient noise such as the Einstein Telescope, the third-generation gravitational wave observatory. In the framework of the instrumental deployment to characterise the formerly lead and zinc mine of Sos Enattos, currently dismissed and converted to a low-noise laboratory, we focus on the link between the records from seismic stations in different locations and the meteorological records collected in their vicinity. To do this, we use a scattering network, a convolutional neural network with wavelet filters, to extract relevant spectro-temporal features at different time scales of the signal. We then use a dimensionality reduction algorithm to reduce the features' dimensions and apply a hierarchical clustering algorithm to identify patterns in the continuous seismic data. We choose hierarchical clustering because it allows us to understand the inter-cluster similarity. We finally investigate the link between these clusters and the external meteorological data collected nearby and reveal the mutual information between the two datasets.