Abstract
It is an open question whether machine-learning (ML) methods can be trusted in areas where dense and localized seismic networks are in operation, and prompt and accurate detection and location of earthquakes are essential to guide decision-making processes that contribute to seismic-risk-mitigation-strategies, even for very low-magnitude events. To address these concerns, we compare the performance of a widely-used ML phase picker, PhaseNet, integrated with several popular earthquake location methods (included in LOC-FLOW), with the results obtained by the workflow adopted since 2012 by the Collalto Seismic Network, installed to monitor natural and potentially induced microearthquakes nearby an underground gas storage. The tested dataset concerns the most populated microseismic sequence observed so far (374 events, ML⩽2.5, August 2021, Refrontolo, NE-Italy), as its unusual productivity raised some criticalities in the combination of automatic routines, and time-consuming manual revision of phase picks adopted by the standard workflow. LOC-FLOW is able to detect the majority of the events listed in the manually revised catalog, demonstrating its ability to efficiently and accurately build earthquake catalogs from continuous seismic data. We highlight both the advantages and limitations of the ML-picker and recommend the use of template-matching-techniques in the final stage of processing to increase the number of events.
Acknowledgments
This work is done in the frame of the activities that OGS performs for seismic monitoring of industrial sites. The Collalto Seismic Network is managed by OGS on behalf of Edison Stoccaggio S.p.A. in compliance with the requirements of the Italian Ministry of Environment, Land and Marine Protection (MATTM) and in agreement with the Veneto Region. The authors would like to thank Enrico Priolo and the technical and control room staff: Fabio Franceschinel, Marco Santulin, Paolo Bernardi, Peter Klin, Giovanna Laurenzano, Franco Pettenati, Alessandro Rebez, and Angela Saraò.
Author contributions
MS performed conceptualization, data analyses, and representation, original draft and manuscript handling. LP performed data representation. LP, MAR, MG, LM, DS, MPPL and MR contributed to data interpretation, manuscript writing, reviewing and editing.
Data availability statement
The data cited and used in this study (Peruzza et al. Citation2022b) are available in Zenodo Open Access repository, https://doi.org/10.5281/zenodo.7252308.
The continuous data of the RSC seismic network, and of the SMINO Network are available from OASIS Web Services (Priolo et al. Citation2015b), http://oasis.crs.inogs.it.
The following codes are available on github, PhaseNet: https://github.com/wayneweiqiang/PhaseNet; REAL: https://github.com/Dal-mzhang/REAL; FDTCC: https://github.com/MinLiu19/FDTCC; Match&Locate: https://github.com/Dal-mzhang/MatchLocate2; LOC-FLOW: https://github.com/Dal-mzhang/LOC-FLOW, last access April 2023. hypoDD software available at https://www.ldeo.columbia.edu/∼felixw/hypoDD.html; Hypoellipse software available at https://pubs.usgs.gov/of/1999/ofr-99-0023/.
Disclosure statement
No potential conflict of interest was reported by the authors.
Correction Statement
This article has been corrected with minor changes. These changes do not impact the academic content of the article.