This project consists of two linked investigations into the large-scale elastic and anelastic structure of the Earth's mantle. The first study concerns the development of a three-dimensional model of shear attenuation in the Earth's upper mantle. The model is constrained by a large data set of Love and Rayleigh surface-wave amplitude measurements in the period range 50--250 seconds. A new method, which removes extraneous effects on wave amplitude by solving for them as part of the inversion, is used. Preliminary results of this approach applied to the mapping of Rayleigh wave attenuation show that it successfully isolates the signal due to attenuation in the amplitude data. Three-dimensional finite-frequency kernels will be incorporated in the inversion to account for the spatially broad sensitivity of the long-period surface waves considered here. A simultaneous inversion for the anelastic and elastic structure of the upper mantle will both account for and be constrained by the effects on wave amplitude of focusing by velocity heterogeneity. The 3-D attenuation model developed in this study will be used to address still-unresolved but fundamental questions regarding the underlying causes of large-scale elastic and anelastic heterogeneity in the upper mantle, including temperature, composition, water, and partial melt.
The second study is aimed at advancing the mapping of the large-scale anisotropic structure of the Earth's mantle, and developing a regional-scale model of the elastic structure beneath Eurasia. This work expands existing collections of dispersion measurements, cross-correlation travel times, and long-period waveforms, and inversion of the data simultaneously for three-dimensional Earth structure. Work to date has included inversion for a new one-dimensional anisotropic mantle for the upper mantle without a discontinuity at 220 km depth and the development of a new technique to account for the non-linear effects of crustal structure on long-period waveforms. Tomographic models of radially anisotropic three-dimensional structure for the whole mantle are developed, and a careful resolution analysis will be conducted to determine where anisotropy is required by the observations. Building on the global model, a multi-resolution parameterization will be employed to develop a model with greater lateral resolution beneath Eurasia.The tomographic models (elastic and anelastic) developed in this project will be useful for direct interpretation in terms of composition and state of the Earth's mantle. The research will lead to the development and distribution of several data products, including collections of direct measurements and tomographic models, as well as computer programs to make use of these products.