Source: Dissertation Abstracts International, Volume: 76-07(E), Section: B.

Adviser: Maureen D. Long.

Access restricted by licensing agreement.

This item must not be sold to any third party vendors.

Knowledge regarding the dynamics of Earth's interior is critical to our understanding of past and present tectonic processes. In subduction zones, the geometry of the mantle flow field and the fine-scale structure of the mantle wedge remain poorly understood. This is a major deficit in our understanding, as the mantle wedge plays an important role in many of the processes associated with subduction, including the transport of water into Earth's deep interior and the generation of arc volcanism. Similarly, in continental interiors, an improved understanding of past tectonic processes is necessary to attain a better understanding of the formation and evolution of continental lithosphere, also poorly understood phenomena.

In this work, I interrogate the character of seismic anisotropy in both the subduction zone mantle wedge and continental interiors to learn about dynamic processes both past and present. In Chapter 1, I measure shear wave splitting of local S-waves beneath Japan. I find that shear wave splitting due to anisotropy in the mantle wedge exhibits strong lateral variations in character, as well as a strong dependence on frequency. This ultimately points to heterogeneous anisotropic structure and a complex mantle flow field. A consequence of this work is the detection of anomalously small shear wave splitting delay times beneath northeast Japan, which implies extremely weak or heterogeneous mantle wedge anisotropy.

In Chapter 2, I use receiver functions as a tool for interrogating mantle wedge anisotropy, specifically beneath the anomalous region of northeast Japan. Results from this work clearly reveal a layer directly above the subducting slab with an anisotropic symmetry axis that is oriented parallel to the convergence direction. There is also evidence for a region of relatively low seismic velocities in the central portion of the mantle wedge and an anisotropic layer beneath the crust of the overriding plate, all of which exhibit a relatively modest strength of anisotropy.

In Chapter 3, I use geodynamic modeling to explore the possibility of small-scale convection in the mantle wedge beneath northeast Japan. Small-scale convection would align perpendicular to large-scale flow, destroy the coherency of the wedge flow field, and result in relatively weak anisotropy. The could potentially explain the small shear wave splitting delay times observed beneath northeast Japan, and has already been proposed to explain other aspects of wedge structure. Results from this study indicate that mantle wedge viscosity plays the most significant role in determining whether small-scale convection will occur and must be sufficiently low (∼ 1018 Pa s).

In Chapter 4, I transition to using seismic anisotropy as an indicator of past dynamic processes in continental interiors. I again use receiver functions to investigate the orientation of anisotropy, this time in the stable cratonic interior beneath the central United States, with a specific focus on improving our understanding of mid-lithospheric discontinuities (MLDs). Results reveal strong evidence for sharp changes in the orientation of anisotropy across multiple MLDs, with an approximately N to NW orientation of anisotropy in the upper mantle lithosphere. The consistency of this signature over a significant lateral distance suggests that the observed anisotropy may be a relic of North American craton formation, while the presence of several distinct anisotropic layers within the cratonic lithosphere supports models for craton formation via stacked subducted slabs and/or a series of underthrusting events.

Finally, in Chapter 5, I explore an important issue regarding the use of receiver functions to infer Earth structure. The commonly used trial-and-error method to generate acceptable models of Earth structure based on receiver function data makes it nearly impossible to explore model space in a systematic and quantitative manner. Here, I explore the potential of using a Markov chain Monte Carlo method with Gibbs sampling for forward modeling of receiver functions. Such an approach will allow us to place more quantitative constraints on Earth structure as inferred from receiver function data.