Seismic interrogation of two highly dynamical earth boundaries: The 410-km discontinuity low-velocity layer and the inner-outer core boundary

Recently, the Transition Zone Water Filter (TZWF) hypothesis of Bercovici and Karato, 2003, highlighted the importance of the 410-km discontinuity in interpreting the Earth's chemical dynamics. This model hypothesizes a localized layer of hydrous melt atop the density increase associated with the 410 km discontinuity. The hydrous melting is predicted to result from a water solubility contrast between the transition zone mineral assemblage dominated by wadsleyite and the upper mantle olivine-dominated mineral assemblage. Thermodynamic calculations and experiments suggest that the hydrous solidus of the olivine dominated assemblage would be exceeded when transition zone material having >1.0% weight percent water is fluxed upwards across the 410 km phase transformation. This hydrous melting thus partitions and sequesters incompatible elements into a melt layer that ponds atop the 410, leaving the upwelling residuum MORB-like with respect to its trace-element inventory. In this model, the geochemical surface observables (mi-oceand ridge basalt and ocean-island basalt geochemistry) are linked to the chemical dynamics at the 410-km discontinuity. If the melt layer is thick enough, it may be detected as a layer of lowered seismic velocity.;In Chapters one and two, I test the melt-layer prediction of the TZWF model by searching for lowered seismic velocities atop the 410 beneath four seismic broad-band arrays in the Rocky Mountain region. Aside from available dense seismic arrays, 150 million years of Farallon slab subduction beneath North America provides (1) a possible mechanism to hydrate the transition zone, and (2) downward flow which may provide enhanced upwelling rates in the study region. Both conditions are necessary for operation of the TZWF; thus the western United States is a natural location to test the TZWF model. Low-velocity layers are found by isolating P-S converted waves, and layer thicknesses and velocity reductions are robustly characterized by a Bayesian statistical analysis of forward modeling results. My results demonstrate that beneath the western U.S. pervasive and thick (24-32 km) low-velocity layers exist atop the 410 with 5-8% shear velocity reductions. These seismic constraints are interpreted as manifesting a ubiquitous occurrence of melt layers as predicted by the TZWF model. Extrapolation of upper-mantle melt-velocity scaling relationships suggests a 2.4-4% melt porosity in the 410 km discontinuity melt layers. In Chapter 2, comparison of our melt-layer constraints from the RISTRA seismic array with high resolution tomographic velocity images that extend to 600 km depth suggests that the 410 km melt layer may be shedding wet blobs or melt-volumes, via Rayleigh-Taylor-like instabilities, into the upper mantle that ultimately impact at the base of the lithosphere.;Seismic characterization of the ICB seismic structure is difficult. First and foremost, due to their small reflection amplitude, P-waves reflecting from the inner-core boundary (PKiKP) are rarely visually identified in seismograms. In fact, my findings show that the peak ground velocity amplitude of a PKiKP arrival from a Guatamalan earthquake recorded by the EarthScope transportable array is only about 18 nm/sec. Yet, the PKiKP phase is a fundamental observable that provides important constraints on the density contrast at the inner-core boundary. Furthermore, this ICB density contrast explicitly determines the dominant power available (Pg = Deltarho ICB·2·1011 W) to drive magneto-hydrodynamic convection in the fluid core, thus powering the geodynamo.;In Chapter 3, I analyze 323 EarthScope Transportable Array (TA) stations for PKiKP arrivals and estimate the density contrast at the inner-core boundary beneath northern Mexico as 0.77 g/cm3, consistent with previous studies and sufficient to drive the geodynamo. Two regions of enhanced reflectivity are found via visually picked waveforms which may be related to ICB dynamics. Analysis of the entire dataset is performed to determine the fraction of stations at which PKiKP energy may be detected in the seismograms. These results suggest where future densely spaced seismic arrays may be located to efficiently target the inner-core boundary and resolve details of its seismic scattering and attenuation structure. (Abstract shortened by UMI.)...