Geology topics

Terry Tullis

Publications and source records attributed to Terry Tullis.

3 recordsLinked to original sources

USGS2022

The evolution of rock friction is more sensitive to slip than elapsed time, even at near-zero slip rates

For many decades, frictional strength increase at low slip rates has been ascribed to time-dependent contact-area growth across the sliding interface. As a result, phenomenological models that correctly predict contact-area growth, as observed in laboratory experiments, have also been widely assumed to be appropriate descriptors of frictional strength evolution. We present experiments that impose more than 5-orders-of-magnitude slip-rate reductions on granite to show that frictional strength evolution in these rocks unequivocally refutes such models. Instead, the data suggest that, even at subnanometric slip rates, frictional strength dominantly evolves with accrued slip. This remarkable slip-sensitivity of friction requires changes of intrinsic strength of the interface with slip that are absent from popular conceptual models of friction at the microscopic contact scale.

Proceedings of the National Academy of Sciences↗

USGS2022

Apparent age dependence of the fault weakening distance in rock friction

During rock friction experiments at large displacement, room temperature and humidity, and following a hold test, the fracture energy increases approximately as the square of the logarithm of hold duration. While it's been long known that failure strength increases with log hold time, here the slip weakening distance, d h , also increases. The weakening distance increase is large, hundreds of percent change over a few thousand seconds. The initial bare surface and simulated fault gouge experiments were conducted in rotary shear at 25 MPa normal stress, 21 MPa confining stress and at displacements greater than 100 mm. In contrast, initially bare surface experiments at 5 MPa normal stress, unconfined at displacements less than 10 mm show effectively no change in d h . We attribute the difference to the presence of an appreciable shear zone that develops due to wear over significant displacements, confined at elevated normal stress. Prior published studies of sheared simulated fault gouge at short displacement show both acknowledged and unacknowledged increases in d h that may relate to our observations. Since natural faults have well-developed shear zones, the observations have more direct relevance to earthquake nucleation than prior laboratory studies that use short displacement data and focus on frictional strength recovery alone. However, the physics underlying this increase in weakening distance are not known; candidates are compaction (Nakatani, 1998) and delocalization (Sleep et al., 2000). Additional caveats are that these are room temperature and humidity experiments, at a single normal stress that have not yet been reproduced in other laboratories.

Journal of Geophysical Research↗

USGS2014

Laboratory constraints on models of earthquake recurrence

In this study, rock friction ‘stick-slip’ experiments are used to develop constraints on models of earthquake recurrence. Constant-rate loading of bare rock surfaces in high quality experiments produces stick-slip recurrence that is periodic at least to second order. When the loading rate is varied, recurrence is approximately inversely proportional to loading rate. These laboratory events initiate due to a slip rate-dependent process that also determines the size of the stress drop [Dieterich, 1979; Ruina, 1983] and as a consequence, stress drop varies weakly but systematically with loading rate [e.g., Gu and Wong, 1991; Karner and Marone, 2000; McLaskey et al., 2012]. This is especially evident in experiments where the loading rate is changed by orders of magnitude, as is thought to be the loading condition of naturally occurring, small repeating earthquakes driven by afterslip, or low-frequency earthquakes loaded by episodic slip. As follows from the previous studies referred to above, experimentally observed stress drops are well described by a logarithmic dependence on recurrence interval that can be cast as a non-linear slip-predictable model. The fault’s rate dependence of strength is the key physical parameter. Additionally, even at constant loading rate the most reproducible laboratory recurrence is not exactly periodic, unlike existing friction recurrence models. We present example laboratory catalogs that document the variance and show that in large catalogs, even at constant loading rate, stress drop and recurrence co-vary systematically. The origin of this covariance is largely consistent with variability of the dependence of fault strength on slip rate. Laboratory catalogs show aspects of both slip and time predictability and successive stress drops are strongly correlated indicating a ‘memory’ of prior slip history that extends over at least one recurrence cycle.

Journal of Geophysical Research↗