Late to Post-Variscan tectonics in the Sardinia Einstein Telescope candidate site (Italy): insights from Structural Survey and Electrical Resistivity Tomography

In north-eastern Sardinia, due to the present-day geodynamic quiescence of the island and its very low seismicity and anthropogenic seismic noise, an area has been candidate for hosting the Einstein Telescope (ET). ET is the European third-generation underground interferometric detector of gravitational waves, whose functioning requires a rocky volume virtually devoid of permeable fractures ideally not crossed by main regional faults. Hereby, we present the structural and Electrical Resistivity Tomography (ERT) features of the most relevant brittle structures in the ET candidate site.

The late- to post-Variscan tectonics in Sardinia is accompanied by extensive magmatism giving way to the Corsica-Sardinia Batholith emplacement. This is followed by a dense dyke swarm of bimodal mafic-felsic composition. These dykes are hosted in the batholith and its host metamorphic basement, being their trends reflecting the stress field of the new-formed Variscan crust during early Permian. Field evidence, shows that a ductile to brittle fault network affects both the Variscan metamorphic basement and the late-Variscan plutons. Fault zones are generally NNW-, and WSW-striking and are associated with more altered bedrock and/or occasionally pseudotachylite-bearing cataclastic bands that have been locally injected by hydrothermal fluids, as testified by thick quartz veins and chlorite-rich selvedges. Near two new drilling sites (ca. 250 m total depth), ERT shows a stratified resistivity array, that consists of up to three electrolayers with variable distribution and thickness. As supported by field observation, we have interpreted the more conductive electrolayer as regolith and alluvial units associated with minor faults, while the most resistive electro-layers correspond with the less-fractured granitoids. Overall, the large deep conductive anomalies are bounded by suddenly graded resistivity drops tracing fault systems that are NNW-, N(NE)- and WSW-striking. Upscaling the local results, which provide an accurate estimate of satured fault geometry at depth, we recognize that: i)  the post-Variscan brittle structures mirror the trend of Permian dykes, sills and veins; ii) the main fault zones that underwent strike-slip reactivation were site of later hydrothermal circulation possibly related to Oligocene-Aquitanian tectonics. Further studies are needed to constrain the actual pattern of differential uplift to exclude the presence of neotectonics in the area. Thus, direct dating of faults and dykes and new Global Navigation Satellite System data acquisition could constrain the age of faulting and the differential uplift contribution into the eventual current reactivation of the inherited Variscan structures.