GLACIER FLOW: A REVIEW

Vigorous attack is currently being made on glacier flow by field studies, laboratory investigations, and theoretical analyses. Knowledge of the physical properties and behavior of ice deep within glaciers remains unsatisfactory, because laboratory experiments do not reproduce the relations of time and temperature existing in the natural environment. It is not yet certain whether ice under long-continued stress in glaciers behaves as a plastic substance of fixed yield stress or as an essentially “viscous” material, but it is clear that neither the plasticity nor the “viscosity” is constant. This is particularly important in respect to calculations of glacier movement. Current petrofabric studies on glaciers show that earlier ideas of crystallographic orientation in ice were oversimplified. Strong fabric patterns are found, but they cannot be satisfactorily interpreted on the basis of simple gliding on the basal plane parallel to the direction of shear. Fabrics in surface ice may actually be the product of an ordered recrystallization proceeding from an orientation produced by flow deeper in the glacier. Knowledge of relaxation behavior in ice is required, and controlled laboratory work establishing relations of stress, crystallographic orientation, and mechanisms of yield is needed to provide a satisfactory basis for interpretation of the fabric patterns. Among the several possible mechanisms of glacier flow, intracrystalline gliding is currently favored because of strong fabrics in glacier ice. It remains to discover the means by which the favored crystal orientation is produced, to work out the exact nature of intra crystalline yielding, and to explain the lack of grossly elongated crystals in flowing ice. Slipping on the subglacial floor accounts for as much as 90 per cent of the movementof thin ice resting on steep slopes and for as much as 20–50 per cent of the movement in some valley glaciers. Intergranular shifting can be the predominant process of flow in firn, and it may be effective in some ice bodies, but for most glaciers it is less significant than intra crystalline gliding. Material transfer associated with changes of state can contribute directly and indirectly to glacier flow. It is probably not of major significance but may have been too greatly discounted in recent years. Discrete displacements along shear planes in glaciers are probably largely surficial. Foliation in glacier ice suggests laminar flow, but this may occur by plastic yielding rather than by mechanical slippage on shear planes. Movement in glaciers is inclined obliquely downward from the surface in the accumulation area and obliquely upward in the ablation zone. This enables the glacier to maintain its thickness and surface profile. In a valley glacier, velocity increases from the head to about the firn limit and decreases progressively below that limit to the terminus, more or less in direct relation to the amount of ice available for transport. Annual fluctuations of velocity in a valley glacier probably reflect the influence of pressure near the head of a glacier and of meltwater near the terminus. Some reported diurnal variations in flow velocity may be partly the product of instrumental wandering or refraction of the line of sight. Changes in temperature, radiation, and meltwater may be the cause of bonafide diurnal velocity variations, but only a thin surficial crust can be involved if this is so. Changes of atmospheric pressure are worthy of consideration, although they do not appear quantitatively competent to account for the larger diurnal velocity variations. Serious doubts are thrown on Demorest's concept of extrusion flow by theoretical analyses and field tests. It is in such a shaky position that satisfactory alternatives should be sought. The most promising to date is Nye's extending and compressive flow which accounts for many of the relations heretofore attributed to extrusion flow. Since Nye's concept is based on theoretical analyses necessarily involving major simplifications and assumptions concerning the behavior of ice deep within a glacier, it needs further examination, analysis, and test before being unconditionally accepted.