HR: 13:40h
AN: S32C-01    [PDF]
TI: Shallow Anisotropy in the Mediterranean Mantle From Surface Waves
AU: * Marone, F
EM: federica@seismo.berkeley.edu
AF: Berkeley Seismological Laboratory, University of California, 215 McCone Hall, Berkeley, CA 94720 United States
AU: van der Lee, S
EM: suzan@tomo.ig.erdw.ethz.ch
AF: Dept. of Geological Sciences, Northwestern University, 1850 Campus Dr., Evanston, IL 60208 United States
AU: Giardini, D
EM: giardini@seismo.ifg.ethz.ch
AF: Institute of Geophysics, ETH-H{\"o}nggerberg (HPP), Zurich, 8093 Switzerland
AB: In this study, we advance new evidence for the existence of the Love-Rayleigh discrepancy in the Mediterranean region and provide information on the average polarization anisotropic structure of the Mediterranean mantle. Using PWI (Nolet, 1990), we analysed about 1800 Rayleigh and Love waveforms from regional events recorded at 3-component broad band seismic stations of the MIDSEA and other networks in this region. Not one of the 3-component seismograms showing both Rayleigh and Love waveforms with an acceptable quality could be fit with a single 1D isotropic model characterized by a velocity smoothly varying with depth. However, satisfactory fits for individual Rayleigh and Love waveforms could often be obtained using realistic 1D velocity models. These 1D path-average velocity structures have been used in the framework of PWI to derive 3D \textit{S}-velocity models for the Mediterranean region. The uppermost mantle velocity structure retrieved from Rayleigh waveforms is consistently slower than velocities obtained from Love wave data. We suggest that the observed Love-Rayleigh wave incompatibility is most likely caused by radial anisotropy. However, the Love wave data are not only incompatible with the Rayleigh wave data but are also internally less compatible than the Rayleigh wave data set. This difference in internal compatibility and the non-uniqueness of tomographic inversions prevent a meaningful quantitative comparison of the 3D \textit{SV}- and \textit{SH}-velocity models independently derived from Rayleigh and Love wave data, respectively. Instead we study the \textit{SH}-\textit{SV} velocity difference necessary to fit both Love and Rayleigh data with a smoothly varying velocity model, by inverting the Love wave data relative to the 3D \textit{SV}-velocity model EAV03 derived from the Rayleigh wave data set. This procedure leads to remaining Love data residuals that are comparable to those obtained in an independent inversion of the Love wave data and to 3D \textit{SH}-velocity models that have more similarities with the better constrained \textit{SV}-velocity model EAV03. Our preferred 3D \textit{SH}-velocity model has been obtained in a strongly damped inversion towards EAV03, which ensures that it shows only the minimum deviation from EAV03 and thus the simplest anisotropic structure necessary to fit all data. Our results show that Love waves require higher velocities (about 200 m/s) compared to Rayleigh waves in the sub-Moho mantle, between 30 and 120 km depth. Deeper anisotropy could exist, but is not necessary to explain our data. We relate the observed anisotropy to lattice-preferred orientation of crystallographic axes of elastically anisotropic minerals such as olivine, as suggested, throughout the Mediterranean region, by \textit{SKS} splitting measurements. Due to the strongly heterogeneous upper mantle structure and to the complex lateral variation of the lithosphere-asthenosphere boundary in this region, it is not possible to discriminate between frozen-in preferred orientation of olivine in the lithosphere or alignment of olivine crystals in a present-day asthenospheric flow.
DE: 7218 Lithosphere and upper mantle
DE: 7255 Surface waves and free oscillations
DE: 8150 Plate boundary--general (3040)
DE: 8180 Tomography
DE: 9335 Europe
SC: Seismology [S]
MN: 2003 Fall Meeting