Résumé
Seismic anisotropy in oceanic basin inferred from surface waves shows a controversial discontinuity near the lithosphere-asthenosphere boundary (LAB). Radial anisotropy displays an age independent positive gradient, that may correspond to a shallow discontinuity at approximately 70 km depth. This is at odds with the view of a mechanical and age dependent LAB, expected to roughly follow the isotherms. To model the development of seismic anisotropy in oceanic basins, and its potential implications for the interpretation of the G discontinuity, we use the model of lattice preferred orientation (LPO) evolution D-Rex, coupled with a two dimensional model of a plate-driven flow in a fluid with a viscosity depending mainly on stress and temperature. We perform a systematic investigation of the influence on seismic anisotropy of the parameters controlling olivine LPO development. We find that the fraction of deformation accommodated by dislocation creep relative to diffusion creep, the strength of the slip systems involved in plastic deformation, and the efficiency of dynamic recrystallization are key parameters for the production of seismic anisotropy. For a wide range of parameters, the predicted radial anisotropy displays an age independent positive gradient near the depth of the G discontinuity. We thus conclude that this is an ubiquitus characteristic of the seismic anisotropy produced by the 2-D plate driven flow in oceanic basins. If not excluded, no additional ingredients such as partial melting, or change in water content are thus required to explain the radial anisotropy pattern near the LAB.