Non-Double-Couple Seismic Sources and Active Intrusions in Mountain Chains: an Example from a Moderate Seismic Sequence in Southern Apennines, Italy

We investigate an anomalous deep seismic sequence characterized by low-frequency bursts of earthquakes (maximum magnitude 5) that occurred between December 2013 and January 2014 in the southern Apenninic chain, Italy. Previous studies (Di Luccio et al., 2018) have shown evidences of fluid involvement in the earthquake nucleation process and identified thermal anomaly in nearby aquifers where CO₂ of magmatic origin dissolves. Seismic source parameters reveal important information about the rupture mechanisms and stress field and their relation with the geological-tectonics processes. It is commonly assumed seismic source as pure shear dislocation described by a double-couple model. When volumetric changes occur, we need to consider the non-double couple source component in the description of the rupture process, as in geothermal and volcanic systems where fluids play an important role. In this study we analyze the 2013-2014 seismic sequence (earthquakes larger than 3) through a full moment tensor (FMT) inversion by using the HybridMT code (Kwiateck et al., 2016). The FMT is based on computing the integral of the first P-wave ground displacement pulse that is proportional to the seismic moment. Uncertainties of estimated FMTs are expressed by the normalized root-mean-squares between theoretical and observed amplitudes. The FMT technique is done on the vertical components of the seismograms, using a detailed 1D velocity model and accurate locations of the events. After a visual inspection of the waveforms based on the signal-to-noise ratio, we compute the displacement to estimate P-wave pulse polarities and the area beneath the P-pulse for each event and each station within epicentral distance comparable to the focal depth. The inversion procedure provides Mw, seismic moment and P, T and B axis orientation. Our results show high percentages of non-double couple components that vary over time and do not depend on earthquake magnitude. The stress axis orientations are in agreement with the regional crustal stress regime. The comparison of the obtained source parameters with petrological and geological data will allow us to better understand the emplacement mechanisms of intrusive bodies and the seismicity in mountain chains.