Date of Award

Summer 8-2018

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Doctor of Philosophy (PhD)

Department

Earth Sciences

Advisor

Scott Johnson

Second Committee Member

Christopher Gerbi

Third Committee Member

Peter Koons

Additional Committee Members

Senthil Vel

David Okaya

Abstract

While earthquakes represent a major hazard to life and property, there are a number of open questions about how earthquake faults operate at depth, and how the energy released by earthquakes travels as elastic waves through Earth’s complexly deformed crustal rocks. The aims of my dissertation are to explore (a) the extent of co-seismic damage in an ancient earthquake fault exhumed from great depths, (b) the deformation processes and mechanics of the fault at depth during earthquake cycles, and (c) the role of different rock structures in determining the velocities of seismic waves.

When tectonic plates collide, deformation tends to localize into narrow zones: frictional faults in the upper crust and high-temperature viscous shear zones in the lower crust. The transition in material behavior from the upper to lower crust is known as the frictional-to-viscous transition (FVT; ~10–20 km deep). During earthquake cycles, the FVT experiences transient brittle deformation followed by long-term viscous processes. Owing to this complex behavior over the earthquake cycle, the FVT is the most important horizon for understanding earthquake mechanics. Rareness of exposures of ancient earthquake faults at FVT depths has hindered studying of their brittle co-seismic damage structures and rheology of their deep portions during earthquake cycles. From the Sandhill Corner shear zone, a strand of the Norumbega fault system (an ancient seismogenic strike-slip fault at the FVT), I analyze fluid inclusion abundance in quartz as a proxy for transient co-seismic damage using secondary electron image and optical observation, and collect quantitative data of quartz across the shear zone such as grain-size, grain-shape, crystallographic orientation, misorientation, and fabric intensity through electron backscatter diffraction. The results indicate that brittle co-seismic damage occurs up to at least ~90 m in width at the FVT, and the inner shear zone (~40 m wide) experienced cycles of co-seismic microfracture-assisted grain-size reduction followed by post-seismic viscous deformation dominated by grain-size-sensitive processes, whereas the outer shear zone was deformed dominantly by grain-size-insensitive processes during earthquake cycles. My findings have important implications for the strength, or mechanics, of the fault/shear zone system, and may help determine 3-D volume of brittle damage zone. Measuring the extent of damage zone is critical for estimating the potential energy that an earthquake releases because the co-seismic damage zone acts as a dissipative energy sink by creating fracture surface areas.

Earthquakes not only represent hazards but radiate energy as seismic waves. Since the direction-dependent nature of wave propagation velocities (called “seismic anisotropy”) changes in response to rock flow due to preferred orientation of elastically anisotropic minerals, the seismic anisotropy has been used to investigate Earth’s interior structure and deformation processes in tectonically active regions. However, this is a challenge for waves passing through the crust because their anisotropies are profoundly modified by macroscale folds, which are very common structures in ancient and current orogenic belts and shear zones. To evaluate the modification of seismic anisotropy by the deformation structures, I develop a new mathematical methodology for calculating bulk elastic tensors and seismic anisotropy of macroscale folds, assuming the seismic waves are much larger than the fold heterogeneity. The results show that the velocities of seismic waves propagating through macroscale folds in three dimensions are systematically related to fold shape and orientation. Because fold orientations are related to flow directions, it is now possible for real seismic observables to provide information on the directions of flow for actively deforming rocks at depth.

Download

Share