A NEW LOOK AT THE CLASSIC HUBBERT-RUBEY BEER CAN EXPERIMENT: LABORATORY ANALOGUE TO FAULT DISPLACEMENT MECHANISMS

In 1959, Hubbert and Rubey presented their now famous “beer can experiment” to illustrate the hypothesis that the movement of low-angle faults may be facilitated by high fault zone fluid pressure beneath thick sedimentary overburden. We expanded this experiment by modeling an unstable system where velocity weakening was the dominant form of frictional sliding, representing stick-slip behavior. Empty chilled beer cans were placed on a glass plane inclined at an angle of 1º and coated the glass with a thin film of beer. Multiple experiments were conducted with cans of different sizes and beer with varied alcohol content, viscosity and body. As the cold air within the can equilibrated to room temperature, the air expanded which caused the internal pressure to increase. Increased pressure allowed the can to overcome normal force by reducing the frictional resistance to sliding, which created a motion history resembling the Prandtl model for stick-slip deformation. In the presence of excess fluids, the beer can displayed failure similar to catastrophic sliding; when excess fluids were not present, the beer can displayed characteristic stick-slip behavior. In some instances, aseismic slip occurred between discrete sliding events. The amount of time for a can to reach equilibrium varied between experiments, depending on the rate of temperature increase of the beer film at the base of the can. Results showed that the temperature and displacement of the can rapidly increased in the first few seconds of the experiment; over time, as the can temperature equilibrated with room temperature, the rate of change in these variables decreased. Can size and composition also influenced motion history and the type of failure. This work supports the idea that friction at the base of a thrust fault may be reduced if fluids at sufficiently high pressures exist along a décollement. Thus, pore fluid pressure allows thrust faults to slide along shallower angles than predicted. This model may be used to predict how stick-slip deformation affects fault characteristics and earthquake magnitude in different tectonic regimes. In a broad extrapolation, we suggest this model as an analogue for the Heart Mountain Detachment in Wyoming and the South Tibetan Detachment System in the Greater Himalaya.

Additional authors: David R. Lageson, Hannah Susorney.