Anisotropic velocity structure of the mantle beneath North America

A surface-wave-dispersion dataset of unprecedented size is used to constrain the radially anisotropic shear-wave velocity structure of the upper mantle beneath North America. Love and Rayleigh wave phase-velocity maps for periods in the range 35--150 s are determined in the first part of the study. The development of these maps provides a test for many of the new or refined modeling techniques used in the second part of the study to develop a three-dimensional radially anisotropic shear-velocity model of the upper mantle. The model, which constrains variations in velocity on a lengthscale of a few hundred kilometers within the North American continent, is determined simultaneously with a longer-wavelength model of the upper mantle of the remainder of the globe. Laterally varying velocity-sensitivity kernels are used to account for the dependence of the sensitivity functions on lateral variations in crust and mantle velocity structure. The sensitivity kernels are updated in several iterations to avoid nonlinearities associated with the inverse problem for the determination of mantle structure. Variations in isotropic velocity in the uppermost several hundred kilometers of the mantle are found to correlate well with surface tectonic features.; Variations in radial anisotropy show a clear continent-ocean signature. Strong anisotropy occurs at shallow depths (<100 km) under the continents, with a secondary peak found at a depth of ∼200 km. Maximum anisotropy under the oceans occurs at a depth of ∼125 km, with no secondary maximum. Within the North American craton, the locations of strongest anisotropy are found to correlate with the locations of fastest isotropic velocity.; An empirical relationship between perturbations in shear-wave velocity and perturbations in density is determined using the three-dimensional velocity model developed here and a simple model of the cooling of the oceanic lithosphere. A density model based on this relationship is used to make a forward prediction of the gravity field. The result confirms the importance of compositional effects on upper-mantle density. The scaling relationship is then used to make a rough estimate of the relative importance of thermal and compositional effects on upper-mantle velocity structure under the North American craton.