Armelle Philip

Creep and plasticity of glacier ice: a material science perspective

Major advances in understanding the plasticity of ice have been made during the past 60 years with the development of studies of the flow of glaciers and, recently, with the analysis of deep ice cores in Antarctica and Greenland. Recent experimental investigations clearly show that the plastic deformation of the ice single crystal and polycrystal is produced by intermittent dislocation bursts triggered by long-range interaction of dislocations. Such dislocation avalanches are associated with the formation of dislocation patterns in the form of slip lines and slip bands, which exhibit long-range correlations and scale invariance. Long-range dislocation interactions appear to play an essential role in primary creep of polycrystals and dynamic recrystallization. The large plastic anisotropy of the ice crystal is at the origin of large strain and stress heterogeneities within grains. The use of full-field approaches is now a compulsory proceeding to model the intracrystalline heterogeneities that develop in polycrystals. Ice is now highly regarded among the materials science community. It is considered a model material for understanding deformation processes of crystalline materials and polycrystal modeling.

Identification of strain heterogeneities arising during deformation of ice

Abstract Creep tests carried out on specimens of isotropic ice containing a monocrystalline inclusion allow us to observe some strain heterogeneities that develop during the deformation of polycrystalline ice. Different kinds of heterogeneities, some of them leading to strain localization, are observed and described, and mechanisms are proposed to explain how they arise. However, when the inclusion has a very regular shape with no geometric singularity (e.g. circular shape) and is embedded in a fine-grained isotropic matrix, the observations lead us to assume homogeneous deformation of the inclusion, with no strain localization except that associated with basal glide.

Comparison of finite-element and homogenization methods for modelling the viscoplastic behaviour of a S2–columnar-ice polycrystal

Abstract The main homogenization schemes used to model the behaviour of polycrystalline ice are assessed by studying the particular case of a two-dimensional polycrystal which represents natural S2–columnar ice. The results of the uniform-stress, uniform-strain-rate and one-site self-consistent models are compared to finite-element computations. The comparisons were made using the same model of grain, described as a continuous transversely isotropic medium, in the linear and non-linear cases. The uniform-stress and uniform-strain-rate models provide upper and lower bounds for the macroscopic fluidity which are too far from each other to be useful when a degree of anisotropy relevant to ice is considered. Although the self-consistent model gives a weak representation of the interaction between a grain and its surroundings, due to the strong anisotropy of the ice crystal, the resulting macroscopic behaviour is found to be acceptable when compared to the results from finite-element computations.

Three-dimensional rocking curve imaging to measure the effective distortion in the neighbourhood of a defect within a crystal: an ice example.

A three-dimensional Bragg diffraction imaging technique, which combines rocking curve imaging with ‘pinhole’ and ‘section’ diffraction topography in the transmission case, allows three-dimensional lattice distortion in the bulk of an ice crystal under compression to be measured.

Inception and movement of a `subgrain boundary precursor' in ice under an applied stress, observed by X-ray synchrotron radiation Bragg imaging

Both optical microscopy with polarized light and polychromatic beam synchrotron X-ray diffraction imaging (white-beam topography) are used to study in situ the way an ice single-crystal deforms.

Experimental studies of the viscoplasticty of ice and snow

Polar ice sheets are fundamental elements of the climate system. Their extent and elevation have a direct influence on the global atmospheric circulation (through the albedo and wind intensity and direction) while the fresh water input from ice shelves and icebergs influence the ocean circulation. Furthermore the paleoclimatic records found in the deep ice cores drilled in polar ice sheets are an important source of information on the Earth climate mechanisms. As regards valley glaciers, as found in alpine regions, because they are very sensitive to temperature fluctuations they are nowadays considered as key indicators of present climate change. A good estimate of ice sheets mass balance response to climate change is needed to understand past sea level changes, while the mass balance of alpine glaciers is directly connected to present sea level rise. In this context it is very important to model the slow flow of glaciers and ice sheets, and to do so as well as possible, to improve our knowledge on the mechanical properties of ice. On the other hand, a wide range of engineering problems involve ice and snow mechanics, many of which in the high velocity or/and strain rate regimes (as for instance ice-structure interaction or snow avalanches). Since these later topics are exposed in other contributions to this book (see e.g. E. Schulson and M. Schneebeli) we will restrict ourselves to the very slow deformation of ice and snow in what glaciologist call the “viscoplastic regime”.