Asthenospheric rheology beneath mainland China inferred from mantle flow simulation and shear-wave splitting measurements

Mantle rheology is an essential property to understand the dynamics and evolution of the Earth. Estimates of mantle viscosity date from Haskell's canonical estimate of 10(21) Pa s. For the asthenosphere, as a mechanically weak layer beneath the lithosphere, estimates have varied over a large viscosity range from 4 x 10(17) to 1 x 10(22) Pa s. In this paper, we use global and regional tomography-based geodynamic models with lateral viscosity variations in both the lithosphere and the mantle to predict the asthenospheric flow field, then use D-Rex to calculate transverse isotropy axis which is taken as the proxy of the fast polarization direction of shear-wave splitting, and finally determine the asthensopheric viscosity beneath Mainland China under the constraints of seismic anisotropy examined by a shear-wave splitting technique. Our inference, 1.27-5.58 x 10(20) Pa s, lies in the range of previous estimates, and is comparable to most global estimates inferred from global convection-related observables, glacial isostatic adjustment data, global long wavelength geoid anomalies or their combination, but slightly larger than asthenospheric viscosity in Africa (3.5 x 10(19) Pa s) and global estimates (3-10 x 10(19) Pa s) inferred from the shear-wave splitting measurements. These findings show asthenospheric viscosity inferred from shear-wave splitting data has a small spatial change (< one order of magnitude).