Seismic anisotropy of the shallow crust in the Umbria–Marche (Italy) region

Abstract

Small magnitude earthquake data, recorded during the 1997 Umbria–Marche (Italy) seismic crisis, have been used to study the shear wave splitting in the central Apennine. Using the method of diagonalization of the covariance matrix, a set of 1234 S waves impinging the Earth at an angle less than the critical angle has been selected. The cross-correlation method was used to estimate the splitting time between the two horizontal S components and the direction of the symmetry axis of the anisotropic structure. A further analysis, based on visual inspection of corrected traces, allowed us to estimate the anisotropic parameters for 884 S waves. An average splitting time 〈 τ 〉 = 32.3 ∓ 4 ms has been estimated. Splitting times do not generally show a clear pattern with distance, which is an indication of possible spatial variations of seismic anisotropy. A minimum 2 km depth of the anisotropic layer is estimated. The trend of normalized splitting time vs. depth roughly indicates a maximum 4–5 km depth of the anisotropic layer. The distribution of the fast symmetry axis indicates a main symmetry axis oriented along the direction of the maximum compressive stress in the area and a secondary symmetry axis almost orthogonal to the first one. The stereographic plots of fast axis direction seem exclude a depth dependence of anisotropic parameters. A clear spatial correlation between the directions of the fast symmetry axes at the considered stations and the orientations of the nearest fault systems and thrust fronts is found, which confirms the spatial variability in the anisotropic properties of the area, as inferred by the analysis of splitting times. Based also on previous studies, results are interpreted in terms of the extensive dilatancy anisotropy (EDA) model. The estimated crack density is always smaller than the critical coalescence value, enforcing the hypothesis that fluid-filled separated crack systems control the polarization of S waves.