MR015-0002
A new method for determining rate-and-state frictional properties from experimental shear stress oscillations, Application to ice-on-rock sliding

Wednesday, 16 December 2020
Poster
Rob M Skarbek, Lamont-Doherty Earth Observatory, Palisades, NY, United States, Christine Mccarthy, Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY, United States and Heather M Savage, University of California Santa Cruz, Santa Cruz, CA, United States
Abstract:
Rate and state frictional parameters are typically determined using two types of imposed, experimental friction protocalls: velocity steps and slide-hold-slide (SHS) events. In a velocity step, the load point is set in motion at a constant rate until a steady friction coefficient is achieved. The load point velocity is then changed as quickly as possible to a new value. A SHS event also initiates at a steady-state value of friction; then the load point is stopped altogether for some amount of time (the hold), and then started moving again at a new value.

Here we take a new approach to determining frictional properties, by examining the frictional response to controlled, harmonic oscillations in the load point velocity. Under the spring-block model, load point velocity oscillations are equivalent to oscillations in the applied shear stress. Many natural fault systems exhibit slip behaviors that depend on harmonic oscillations in applied tidal stresses. Our new method provides a way to study how frictional properties directly depend on parameters relevant to tidal forcing, such as the frequency and amplitude of the tidal signal.

We apply our new methods to a set of ice-on-rock experiments conducted over a temperature range of -16C to -2C. Values of the frictional stability parameter (a-b) determined from oscillations reveal dominantly velocity-weakening behavior across the entire range of experimental conditions. However, values of (a-b) determined from velocity steps in the same experiments yield velocity-strengthening to velocity-neutral behavior. We also show that the elastic stiffness (which we treat as a fitting parameter) depends strongly on the temperature, which is unlikely to be explained by changes in the elastic properties of ice. We suggest that temperature-dependent changes to the condition of the sliding surface cause changes in the elastic stiffness.