GNGTS 2018 - 37° Convegno Nazionale

GNGTS 2018 S essione 1.1 31 maximum respectively. Hence, the intermediate stress is always horizontal that is we assume that the stress regimes are either normal or reverse. • sum with sign the transferred stress to the regional stress rate * T over a grid of points; estimate the minimum value of T which ensures at each hypocenter that enough regional stress is available so that the total stress is consistent with normal faulting before and after an event in the time history. • this 3D total stress matrix can be projected at each gridpoint on the normal and tangential directions of the fault plane for given strike, dip and rake angles to generate the normal and shear total stress at the grid point. Coulomb stress maps can also be created by assuming a friction coefficient. Deviatoric stress is considered throughout, so that the eigenvalues of the total stress sum to zero. For events available in the CFTI but not in the DISS (hence with no strike/dip/rake angles) the three angles will be inferred from current knowledge. The size of the activated fault is inferred from the magnitude using the Wells and Coppersmith empirical formulas. Results. To estimate the regional stress at an epoch at the location of the L’Aquila 2009 hypocenter we tentatively require that the regional stress at that position and epoch was sufficiently large that the total stress immediately before and after the 2009 event was representative of a normal fault: the tensional axis is horizontal and oriented normal to the strike of the fault, the null axis is horizontal at 90 degrees and the pressure axis is vertical. We show that the minimum regional stress compatible with this requirement is equivalent to a geodetic stress rate multiplied by an empirical temporal scale of at least 2250 years. We verify that with this calibration of the regional stress a reasonable alignment of the principal axes of the stress tensor is maintained across the period 1315-2009 (Fig. 2). Fig. 2 shows that the principal directions of the total stress at the L’Aquila hypocenter rotated in the occasion of the 1315, 1461 and 2009 events, and also in correspondence to the 1703 Montereale event, all of magnitude at least 6. The regional stress was large enough that the tensional state was maintained at all times, as required. Fig. 3 - Coulomb stress map relative to the angles of the 2009 Paganica fault before (left) and after (right) the 2009 event. The Coulomb stress change on the Paganica fault is evident near the lower right corner of the Paganica fault, which was loaded by the stress reservoir to the NE. Coulomb stress after the 2009 event spreads in the neighboring Montereale (NW) and San Pio delle Camere (SE) faults.