Passive seismic and infrasonic monitoring at the Mefite d’Ansanto deep-CO2 degassing site (Southern Apennines, Italy).

The role of fluids in the preparatory phase of major earthquakes and in the evolution of aftershocks and swarms in space and time is well-documented. In particular, numerous studies evidence the primary role that mantle-derived fluids play in the generation of large upper crustal earthquakes in extensional domains, where crustal-scale faults act as preferential hydraulic pathways.  

We focus on the Mefite D'Ansanto degassing site, the largest low-temperature non-volcanic CO₂ emission in the world, located at the northern tip of the Mw6.9 1980 Irpinia faults. The study area experienced strong historical earthquakes (1702, 1732 and 1930 M6+ earthquakes) but it is characterized by a relatively low background seismicity rate with respect to the nearby Sannio and Irpinia regions.    

To collect high-quality microseismicity data in this key sector of the southern Apennine extensional belt and investigate the relationship among seismicity, crustal fluids, and physical-hydraulic properties of the crust, we installed in July 2021 (up to May 2023) a temporary network composed of 10 stations equipped with short-period velocimeters (5 sec). The temporary network covers an area of approximately 30x30 km² surrounding the Mefite d’Ansanto site and integrates with the numerous permanent stations of the INGV and ISNet networks located at the boundary of the survey area. 

Within the Mefite area, we also deployed a temporary seismo-acoustic dense array to study two CO₂ vents. The seismo-acoustic array is composed of 5 infrasonic stations equipped with IST-2018 broadband microphones developed by The ISTerre (Université Savoie Mont Blanc, France), in addition to one seismo-acoustic station equipped with a co-located digital broadband seismometers (120s). The array is positioned approximately at the vertices of a star, with an aperture of about 50 meters. The deployment lasted for 1 week at the end of May 2022, allowing us to sample the emission site during “dry” weather conditions. 

We show first results of the analysis of seismicity recorded by the temporary network applying both standard (STA/LTA) detection algorithms or innovative enhanced techniques such as cross-correlation based template-matching algorithms and/or Deep-Learning-Phase-Recognition methods.

The activities are developed in the framework of the multidisciplinary project FURTHER (https://progetti.ingv.it/en/further).