Seismic images of the Earth?s interior play a central role in the geophysical and geochemical modeling of the planet?s structure and dynamics. The majority of these seismic models contain three-dimensional variations in wave speed. Attenuation, or Q−1, tomography has so far progressed more slowly than velocity tomography because wave amplitudes and spectral content, which are the primary data in attenuation studies, require a more complex interpretation than traveltimes and phase delays. In addition to intrinsic wave attenuation, amplitudes are affected by focusing due to spatial gradients in wave speed, the presence of scatterers, and the earthquake rupture process.Progress in the research of, for example, the nature of the low-velocity zone in the upper mantle, the role of hydrous phases in the mantle, and material flow through the transition zone depends critically on the availablity of models of seismic velocity and attenuation. Because wave speed is influenced by several factors (temperature, composition, melt), robust and unique interpretations of tomographic velocity images require accompanying images of the attenuation structure in the mantle, ideally with comparable spatial resolution.
The main thrust of the proposed work involves the development of a new 3-D global-scale model for wave attenuation in the upper mantle that builds upon our previous modeling. The model will be constrained by new fundamental-mode and higher-mode surface-wave amplitudes. Notably, the overtone data set has not been used in attenuation studies before and will allow resolution of the transition zone. An important component of this research is the investigation into improved theoretical and practical treatments for focusing effects on wave amplitude, including a simultaneous inversion for global shear-velocity and attenuation models. A critical component of ?resolution testing? involves model interrogation with independent data sets of body-wave amplitudes and spectra, and with realistic 3-D synthetics. They will investigate to what extent P and SH spectra, S and SS amplitudes, and regional waveform data can be reconciled with our new attenuation model. They will also examine a number of critical theoretical approximations using SPECFEM synthetics
