GNGTS 2018 - 37° Convegno Nazionale

44 GNGTS 2018 S essione 1.1 For the rock micro-parameters estimation, we consider a method in which the retrieved values of velocities and attenuations are our input data set, following a downscaling procedure. The Pride (2005) poro-elastic theory is used for up-scaling rock parameters to the macro- scale level. By a direct comparison between up-scaled and tomographic macro-parameters, an estimation of host rock micro-parameters is recovered (i.e., porosity, relative saturation of a two-phases fluid and fluid types). Finally, in order to image the temporal variation of elastic properties in the IFZ the 4D seismic tomography was designed (Amoroso et al., 2018). This technique consists in applying the 3D tomography in different time-windows. By analysing the tomographic images obtained for each time-window, after having ensured the same resolution for each of them, it is possible to map out the spatial and temporal changes in elastic/anelastic seismic parameters. The whole catalogue will be divided in consecutive epochs of 6 months with an overlapping of 2 months for the analysis. The model inferred by considering the whole dataset will be used as starting model in the inversion. The model parameterization is the same as for the initial model. We will evaluate the results in terms of Vp and Vs , and their percentage variation with respect to the initial model. For each model the DWS will be computed to determine the well resolved regions. Results. The relocated micro-seismicity, approximately confined within the same volume of the 1980’s Irpinia earthquake faults, is absent in the volume bounded by the first and the second fault segments activated during the event. We hypothesize that this geometrical barrier could have played a key role during the 1980 event, and possibly has controlled the delayed times of activation of the two rupture segments (Fig. 2). The velocity images down to 20 km depth reproduce the main lithological units and discontinuities within this portion of the Apennines chain, showing evidence of a wide and Fig. 3 - Qp (a) and Qs (b) models for parameterization B and earthquake locations projected onto the cross-section located in Fig. 1. White curves delimit the well-resolved regions. (c) Summary of the results obtained from the comparison between the up-scaled macro-parameters curves and the retrieved macro-parameters ranges. The checkmark indicates that all up-scaled curves fall in the retrieved range, otherwise a cross is present. The dash indicates that the value is not reported because the velocity curve falls outside the retrieved range. Modified after Amoroso et al. (2017).