Le fluage de la glace

Abstract

An ice monocrystal can undergo plastic deformation by slip along basic crystallographic planes, without strain hardening occurring. The strain rate then varies as the cube of stress (for the deviatoric stress of the order of 1 bar observed in glaciers). The creep rate variation with temperature is due to activation energy of about 16 kcal/mol. These relationships do not apply to slip along planes running parallel to the optical axis, which is much less common. The rate of creep of polycrystalline glacier ice under very low stress must depend on the latter type of slip, but in more highly stressed ice it depends on recrystallization. In addition, pressuremelting and re-freezing effects are apt to take place above -10°C, resulting in greater apparent activation heat (25-43 kcal/mol). The basic glacier dynamics assumptions are: 1) Hydrostatic pressure has no effect on ice creep (confirmed experimentally) ; 2) Nor have finite strains and discontinuities (e.g. faults) ; 3) Deviatoric stress is proportional to creep rate at a given point (fairly doubtful) ; 4) Viscosity only depends on temperature and the second invariant of the stress deviator (probably incorrect, at least for the polar ice caps).