HR: 0800h
AN: S41A-01    [Abstracts]
TI: Density, Shear and Compressional Velocity Models of the Vrancea Seismogenic Zone
AU: * Tondi, R
EM: rosaria@eost.u-strasbg.fr
AF: IPGS - Institut de Physique du Globe de Strasbourg, 5, rue René Descartes, Strasbourg, 67084, France
AU: Achauer, U
EM: Ulrich.Achauer@eost.u-strasbg.fr
AF: IPGS - Institut de Physique du Globe de Strasbourg, 5, rue René Descartes, Strasbourg, 67084, France
AU: Besutiu, L
EM: besutiu@geodin.ro
AF: Institute of Geodynamics of the Romanian Accademy, 19-21 Jean-Louis Calderon St., sector 2, Bucharest, 020032, Romania
AB: Additional constraints on the geodynamic models for the origin of the intermediate depth Vrancea Seismogenic Zone are given by three-dimensional P and S-wave velocity and density images. The reconstructed physical parameters aim to substantiate or eliminate two contrasting models which explain the Vrancea seismicity: the subduction model and the active continental lithospheric delamination model. For our goal, we apply the tomographic inversion method of sequential integrated inversion proposed by Tondi and de Franco (2006) to shot data collected during the VRANCEA99 (Hauser et al., 2001) and VRANCEA2001 (Landes et al., 2004) seismic refraction experiments, to local earthquake data collected during the CALIXTO (EOS, 1998) experiment and to recent gravity measurements of the studied area. We first locate P and S wave sources of local events with the NonLinLoc location program (Lomax et al., 2000) and then we consider these events as those originated from shots points. The mathematical formulation of the seismic travel time inversion algorithm, which regularizes the solution with the minimization of the first and the second partial derivatives of the functionals describing the velocity parameters, enables us to control the proliferation of caustics and arrivals during iterations, which is a common problem when using ray-tracing techniques with realistic and extensive heterogeneous velocity models. This increases the robustness and efficiency of the method and efficiently handles a seismic data set which is severely affected by scattering effects. Furthermore, the density model parametrization, which uses polyhedral bodies whose density is linearly dependent on the three coordinates (Pohànka, 1998), leads to a perfect match between the density and the velocity model parametrization and takes into account the presence of geological structures characterized by a gradual increase in density with depth. After each iteration, the events are relocated with the updated velocity model until the discrepancies between two subsequent localizations are sufficiently small. The reliability of the reconstructed models, which explain equally well both travel times and gravity data, is quantified through a restoring test and the estimation of travel times and gravity residuals.
DE: 0520 Data analysis: algorithms and implementation
DE: 1219 Gravity anomalies and Earth structure (0920, 7205, 7240)
DE: 3260 Inverse theory
DE: 7270 Tomography (6982, 8180)
DE: 8170 Subduction zone processes (1031, 3060, 3613, 8413)
SC: Seismology [S]
MN: 2007 Joint Assembly