Seismic velocity anisotropy and heterogeneity beneath the Mantle Electromagnetic and Tomography Experiment (MELT) region of the East Pacific Rise from analysis of P and S body waves

[1] We use teleseismic P and S delay times and shear wave splitting measurements to constrain isotropic and anisotropic heterogeneity in the mantle beneath the southern East Pacific Rise (SEPR). The data comprise 462 P and S delay times and 18 shear wave splitting observations recorded during the Mantle Electromagnetic and Tomography (MELT) Experiment. We estimate the mantle melt content (F) and temperature (T) variation from the isotropic velocity variation. Our results indicate that the maximum variation in F beneath our array is between zero and 1.2%, and maximum variation in T is between zero and 100 K. We favor an explanation having partial contributions from both T and F. We approximate the seismic anisotropy of the upper mantle with hexagonal symmetry, consistent with the assumption of two dimensionality of mantle flow. Our new tomographic technique uses a nonlinear inversion of P and slow S polarization delay times to simultaneously solve for coupled VP and VS heterogeneity throughout the model and for the magnitude of anisotropy within discrete domains. The domain dimensions and the dip of the anisotropy are fixed for each inversion but are varied in a grid search, obtaining the misfit of the models to the body wave delay data and to split times of vertically propagating S waves. The data misfit and the isotropic heterogeneity are sensitive to domain dimensions and dip of anisotropy. In a region centered beneath the SEPR the best average dip of the hexagonal symmetry axis is horizontal or dipping shallowly (<30 ) west. Given the resolution of our data, a subaxial region characterized by vertically aligned symmetry axes may exist but is limited to be <80 km deep. We infer that the mantle flow beneath the SEPR is consistent with shallow asthenospheric return flow from the direction of the South Pacific superswell.