WSL Institute for Snow and Avalanche Research SLF Link zu SLF Hauptseite Link zu WSL Hauptseite

Crystalline orientation of ice grains in the snowpack


Fig. 1: The orientation of the c-axis determines the orientation of the ice crystals in the snowpack.

Schneeprofil in der Antarktis

Fig. 2: Snow profile in Antarctica. Photo: Martin Schneebeli, SLF

Dünnschnitt einer antarktischen Schneeprobe

Fig. 3: Thin section of 100 micron thick cut out from Antarctic snow sample. Photo: Neige Calonne, SLF

As in every other crystal, the building blocks of an ice crystal – in this case the water molecules – are arranged in an ordered fashion and follow a crystal lattice. For ice, the crystal lattice is hexagonal. An ice crystal shows two basal faces and six hexagonal faces. In nature, the hexagonal lattice of ice is revealed in fresh snow crystals, which exhibit often a hexagonal shape that can be seen by bare eyes (fig 1.) The axis perpendicular to the basal faces is defined in crystallography as the axe c (fig 1). Thus, whenever interested in the crystalline orientation of an ice crystal within the snowpack, the orientation of this axe c is given by an inclination (from 0° (vertical) to 90° (horizontal)) and an azimuth (from 0° to 360°).

The crystalline orientation of ice grains within the snowpack has been the subject of much interest among snow researchers for a long time. They suspect that it may influence the physical and mechanical properties of snow, and thus impact the compaction of the snowpack for exemple. Until recently, only one study had produced statistically reliable measurements of crystal orientation. This study, based on cold-laboratory experiments, showed that the distribution of the crystalline orientation of ice grains in snow evolves when snow is subjected to temperature gradients. more

Researchers from the SLF and the Laboratory of Glaciology and Geophysics of the Environment (LGGE) in Grenoble have now, for the first time, measured the crystalline orientation of ice grains in a natural snowpack, from samples collected in the field in Antarctica. Their results seem to show a relationship between the evolution of the distribution of the crystalline orientation of snow and the temperature to which this snow has been subjected. At the Concordia station in East Antarctica, snow temperatures at 10 cm below the surface vary between -25 °C in the summer and ‑70° C in the winter. The researchers collected samples at this site continuously from the surface down to a depth of 3 m. Back in Switzerland, they studied the 3D microstructure of these samples (density, specific surface area and pore size) by X-ray tomography. In addition, they have cut almost 80 thin sections of snow from the collected samples (fig 3.), a delicate operation, and analysed them with the Automatic Ice Texture Analyser (AITA), a device that measures the orientation (inclination and azimuth) of the c-axis of each ice grain contained in the analysed thin section with an accuracy of just a few degrees.

Temperature influences the evolution of the distribution of crystalline orientation of ice grains in snow

The results corroborated the experimental findings : when snow is undergoing temperature gradient, not only the microstructure is evolving, but also the distribution of the crystalline orientation of the ice grains. As a consequence, different types of orientation have been observed (the typical ones being isotropic, vertical, or horizontal) within the studied Antarctic snowpack. Because ice has strong anisotropic mechanical properties (non-basal deformation of ice crystals requires a stress at least 60 times larger than that for a basal slip at the same strain rate), it is possible that the mechanical behaviour of a snow layer does not depend only on the microstructure, but might be also influenced by the crystalline orientation of the ice grains.

In Antarctica, these results might be important for the modelling of the snow and firn densification, and open the way to deeper studies on the role of the crystalline orientation of ice grains in snow. The modelling of snow and firn densification is crucial for the dating of the ice cores extracted at depth in the ice sheet and from which paleo-climates can be reconstructed back to several millennia. Beyond Polar Regions, these results are likely to enhance our understanding of snow's mechanical properties.