Sei qui: Home > BANCHE DATI > Amiata DS

Amiata DS

Seismicity map
Map of the seismicity over the Amiata area in the time interval between March 2016 and December 2020. Red dots represent the location of the events considered for the Amiata DS compilation; blue triangles are the stations of the local temporary network (TV), and yellow triangles are the stations of the Italian National Seismic Network (IV).
Examples of waveform samples extracted from the Amiata DS. Panels a,b,c have an (0≤SNR<4)dB; d,e,f have a (4≤SNR<7)dB, and g,h,i have a SNR>7. In all the plots, waveforms are expressed in counts, and the x-axis represents the time in 10-2s. The blue and orange vertical lines represent P- and S- manually-revised picks, respectively.
Distributions of different metadata parameters of the Amiata DS: a magnitude of the events, b hypocenter of the events from the recording station, c depth of the quake, and d backazimuth. The x-labels correspond to the names of the Amiata DS metadata plotted.

Creative commons license: Attribution 4.0 International (CC BY 4.0)

The Amiata DS (dataset) is a collection of waveforms and associated parameters (metadata) related to 1515 events that occurred in the Mount Amiata area (Tuscany, central Italy) between March 2016 and December 2020. The earthquakes have local magnitude -0.2≤ML≤3.6. Most of the data have been recorded by 12 stations (velocimeters) of a local, temporary network deployed in the area by INGV, with interstation distances ranging between 3km and 21km. Data is complemented by the registrations of other 15 stations of the Italian National Seismic Network (IV) installed farther away.

The dataset has been designed to train Machine Learning algorithms that can process waveforms and metadata. However, its reduced dimension makes it more appropriate for applications such as transfer learning, i.e., improving the performance of already trained ML models. The dataset is organized to be readily accessed by the Seisbench API (Woollam et al., 2022).

The dataset numbers:

  • 1515 earthquakes with -0.2≤ML≤3.6
  • 13721, 3-component waveforms in counts
  • 72 metadata for each 3-component waveform, providing information about the station, trace, source, path and quality
  • 12 stations of a local, temporary network + 15 stations of the Italian National Seismic Network

How to cite the article journal

S. Gaviano, Machine learning analysis of a local seismic network in Mt. Amiata (Italy), Nuovo Cimento 2022, doi:10.1393/ncc/i2022-22082-y

How to cite the dataset

Gaviano, S., Michelini, A., Cianetti, S., Giunchi, C., Jozinović, D., Lauciani, V., Biagini, D., Bruni, R., Capello, M., D’Ambrosio, M., Piccinini, D., & Saccorotti, G. (2023). AMIATA-DS – A dataset with microseismicity from Mount Amiata area. Istituto Nazionale di Geofisica e Vulcanologia (INGV).


This work has been supported by the project INGV Pianeta Dinamico 2021 Tema 8 SOME (grant no. CUP D53J1900017001) funded by the Italian Ministry of University and Research “Fondo finalizzato al rilancio degli investimenti delle amministrazioni centrali dello Stato e allo sviluppo del Paese, legge 145/2018”