The seismic sequence of 30th May-9th June 2016 in the geothermal site of Torre Alfina (central Italy) and related variations in soil gas emissions

In the framework of a medium-enthalpy geothermal exploitation project, seismicity and soil gas emissions have been monitored in the area of Castel Giorgio-Torre Alfina (central Italy) since 2014. A dedicated local seismic network, called ReMoTA, allows deepening of the knowledge of the natural local seismicity in terms of source mechanisms, high-quality event localization and magnitude estimation. From November 2014 to May 2016, ReMoTA recorded 846 seismic events with a magnitude range from Md 0.1 to ML 2.8 and with hypocentres between 4 and 8 km depth. Most of these events occurred in six short lasting clusters. On the 30th May 2016 a Mw 4.3 earthquake occurred near Castel Giorgio, followed by almost 1700 aftershocks. The moment tensor solution for the main shock depicts a WNW-ESE oriented normal fault with small right-lateral strike component. An overview of the epicentral distribution of the different clusters recorded since 2014 highlights that the active tectonic structures are orientated principally along the NE-SW and WNW-ESE directions. The relocation of the 1957 Me 4.9 earthquake suggests that this event occurred in the same fault system as of the 2016 seismic sequence. In the study area, there is only one natural emergence (Solfanare) emitting a CO2 dominated gas, having the same chemical and isotopic composition of the gas contained in a cap at the summit of the geothermal reservoir. Apart from small perturbations, no significant compositional variations were recorded during the 2016 seismic sequence in the gas of the Solfanare vents that was analyzed continuously by an automatic gas-chromatographic station. The diffuse soil CO2 flux is monitored since 2013 in six target areas located around the future production and reinjection wells, in order to assess the level of background natural degassing. In all target areas the maximum value of soil CO2 flux has been recorded during the 2016 seismic sequence. However, values remained relatively low (maximum 112 g ⋅m−2 ⋅d−1) and the values of δ13C of the emitted CO2 (−25.25 to −24.22‰ vs. PDB) indicated a shallow biological origin of the gas (by soil respiration). Only at Solfanare high values of diffuse soil CO2 flux were recorded up to a maximum of 20125 g ⋅m−2 ⋅d−1). All the seismicity of the May-June 2016 sequence is located above the ML4.1 main event and is distributed on small distinct faults (such as at San Lorenzo Nuovo and Acquapendente) and triggered by the main shock. The source mechanism provided by the full moment tensor indicates that rupture processes at depth probably deviate from a pure normal fault. The significant contribution of CLVD and isotropic components suggest a possible opening of fluid cracks below the geothermal reservoir hosted in fractured Mesozoic limestones. In spite of the increase of the CO2 flux, no significant changes have been observed in the chemical and isotopic composition of the Solfanare gas. The seismo-tectonic scenario indicates that the Solfanare fault was not activated. Kinematics and orientation of the activated faults suggest a relationship with the Bolsena caldera collapse.