RESEARCH
Below is a list of current research directions and ongoing projects. If you would like to collaborate, please contact me at rtwilliams-at-wisc.edu.
How does fluid flow in faults affect seismic and interseismic deformation and fault strength?
In addition to the well-known role of pore pressure changes in promoting or preventing fault slip, fluid-rock interactions locally change the composition and microstructure of fault rocks, affecting mechanical properties, deformational response to fault slip, and permeability.
My recent research in this area is challenging long-held models regarding the mechanisms and rates of post-seismic fault healing, and relies on a combination of field-based, experimental, and theoretical approaches. For example, see my recent paper in Geology, where I challenge the viability of coseismic boiling of pore-fluids as an agent of rapid hydrothermal cementation and healing (Williams, 2019). In another paper published at Geophysical Research Letters, I used a novel U-Th sampling method to quantify the rates of post-seismic fracture healing fault-damage zones, and showed that the rates of fracture healing are heterogeneous in both space and time. Additional works examining the role of fluids in determining fault strength are under review (see my CV). I am also starting a new experimental program examining the hydrothermal reactions governing fault strength deep in the seismogenic crust.
How does deformation affect fluid flow and mass transport in the crust?
The ability of faults to trap or transport fluids is determined by the character and magnitude of fault-zone deformation. Understanding the fluid transport properties of faults is therefore inexorably tied to understanding the mechanical properties that shape their structural evolution.
My current research in this area is examining fault-localized mineral phases as geochemical records of the timing, origin, and magnitude of fault-zone fluid flow. Much of this ongoing work employs a novel application of radiogenic isotope geochemistry to understand fluid budgets and mixing during flow through the crust. For example, see a recent paper in Journal of Structural Geology where I used a combination of Sr, Pb, and U isotope analyses to quantify temporal changes in post-seismic fluid flow through a seismogenic normal fault over a period of >400 ka. A previous paper in Geology examined the role of extension and syntectonic sedimentation in influence fault-zone fluid flow pathways at the basin scale. Ongoing projects in this area include Pb isotope analyses to determine the source of phyllosilicate forming fluids in the San Andreas fault, and U-Pb geochronology of fault-localized silica veins to trace the outflow of fluid and heat from the Valles Caldera over the last 10 ma in New Mexico.
What are the controls on the nucleation and recurrence of large earthquakes?
Earthquakes pose a significant to seismically active regions that are home to millions of people around the world. Although modern seismological analyses are an invaluable tool for assessing earthquake mechanics and physics, these records are necessarily short (10's of years) in comparison to the time scales of large earthquake recurrence (100's or 1000's of years). For this reason, the rock record of earthquake activity is an invaluable tool for understanding fault mechanics over time scales consistent with the seismic cycle.
My current research in this area is focused on examination of the rock record of earthquake processes to better understand the controls on the nucleation and recurrence of large earthquakes. I am particularly focused on understanding the role of fluids in these processes. For example, see a recent paper in Proceedings of the National Academy of Sciences, I used U-Th dating of coseismic calcite veins to document a catalog of 13 distinct ground-rupturing earthquakes on an intraplate fault in New Mexico. In a more recent paper published in Geophysical Research Letters, I developed a new statistical approach and used it to assess the commonality of periodic (as opposed to random) earthquake recurrence, and showed that the majority (58%) of previously published large-earthquake chronologies support an interpretation of quasiperiodic recurrence (probability of random recurrence < 10%).