An experimental study of the motion of ice past obstacles by the process of regelation

The results of regelation experiments, in which a single object is pulled through ice, cannot be applied directly to the problem of basal sliding in glaciers because the two systems have different geometries. When the force applied to a single object is small, impurities trapped in the regelation water-layer around the object inhibit the regelation process. At larger forces, above the Drake-Shreve transition point, impurities are shed in a trace behind the object. However, when ice moves over a series of obstacles a trace may exist above and below the transition point. The regelation velocity below the transition point is not reduced by the effect of trapped impurities. In an experiment in which brass cylingerrs of various cross-sections rotate in ice, the ratio between the expected regelation velocity, calculated using the basal-sliding theory of Nye, and the measured regelation velocity is 8±2, both above and below the transition point. The same ratio has been obtained by other workers with wires of similar thermal conductivity above the transition point. Measurements of température differences indicate that supercooling cannot be the main source of the unexpectedly low regelation velocities above the transition point.