Sub-surface operations for energy production such as gas storage, fluid reinjection or hydraulic fracking may modify the physical properties of the rocks, in particular the seismic velocity and the anelastic attenuation. The aim of the present study is to investigate, through a synthetic test, the possibility of using empirical ground-motion prediction equations (GMPEs) to observe the variations in the reservoir. In the synthetic test, we reproduce the expected seismic activity (in terms of rate, focal mechanisms, stress drop and the b value of the Gutenberg-Richter) and the variation of medium properties in terms of the quality factor Q induced by a fluid injection experiment. In practice, peak-ground velocity data of the simulated earthquakes during the field operations are used to update the coefficients of a reference GMPE in order to test whether the coefficients are able to capture the medium properties variation. The results of the test show that the coefficients of the GMPE vary during the simulated field operations revealing their sensitivity to the variation of the anelastic attenuation. The proposed approach is suggested as a promising tool that, if confirmed by real data analysis, could be used for monitoring and interpreting induced seismicity in addition to more conventional techniques.

No abstract

Abstract. The purpose of this work is to study the subsoil structure of the Campi Flegrei area using both spectral ratios and array techniques applied to seismic noise. We have estimated the dispersion curves of Rayleigh waves by applying the Frequency–Wavenumber (f–k hereinafter) and Modified Spatial Autocorrelation (MSPAC) techniques to the seismic noise recorded by the underground short period seismic Array “ARF”, by the broadband stations of the UNREST experiment and by the broadband stations of the seismic monitoring network of INGV – Osservatorio Vesuviano. We have performed the inversion of a dispersion curve (obtained averaging the f–k and MSPAC dispersion curves of seismic noise and single phase velocity values of coherent transient signals) jointly with the H∕V spectral ratio of the broadband station CELG, to obtain a shear wave velocity model up to 2000 m depth. The best-fit model obtained is in a good agreement with the stratigraphic information available in the area coming from shallow boreholes and deep wells drilled for geothermal exploration. In active volcanic areas, such as Campi Flegrei, the definition of the velocity model is a crucial issue to characterize the physical parameters of the medium. Generally, a high quality characterization of the medium properties helps to separate the contributions of the volcanic source, path and site in the geophysical observables. Therefore, monitoring possible variations in time of such properties in general can help to recognize anomalies due to the volcano dynamics, i.e. fluid migration connected to the volcanic activity.

Abstract-We study the influence of the resistivity frequency dispersion effects on the magnetotelluric (MT) response. Impedivity is the term used to indicate the frequency dependent resistivity in rocks. The impedivity functions, used in this paper, have been derived from the general solution of the motion equation of a charge carrier, discussed in a previous paper. A 1D three-layered earth section, with the second layer assumed to be dispersive, is considered to analyze the distortions due to dispersion on the modulus and phase of the MT responses on the earth's free surface. The MT responses of the section, where the dispersive layer is attributed an impedivity function describing at first a positive, then a negative and finally a resonant dispersion model, are computed for various combines of the dispersion parameters. A general conclusion is that the dispersion effects can strongly influence the MT response either in recognizable or in subtle forms. In the former case, the distortions appear as either steeply rising and/or descending curve branches or spike-like deltas, not compatible with a dispersion-free section. In the latter case, instead, the MT curves preserve the typical behavior for a dispersion-free section, and may thus erroneously be modeled by a section, where the dispersive layer is totally suppressed. In both case, disregarding the distortion effects may lead to misleading conclusions as to the physical properties of the surveyed structures.