Numerical calculations by finite elements show that the variation of horizontal velocity with depth in the vicinity of a symmetric, isothermal, non-slipping ice ridge deforming on a flat bed is approximately consistent with prediction from laminar flow theory except in a zone within about four ice thicknesses of the divide. Within this near-divide zone horizontal shear strain-rate is less concentrated near the bottom and downward velocity is less rapid in comparison to the flanks. The profiles over depth of horizontal and vertical velocity approach being linear and parabolic respectively. Calculations for various surface elevation profiles show these velocity profile shapes are insensitive to the ice-sheet geometry.

Walder recently analyzed the initial instability of water flow in a sheet under a glacier that is produced by greater heat production in the flowing water and thus the larger rate of ice melting where a perturbation has increased the water sheet thickness. We have looked at the problem from the other presumed final state. We assume that instabilities have finally caused all the water to flow in channels (R-channels) at the bed. We investigated whether these channels can collect enough of the water that is produced by the geothermal heat and the heat sliding to remain in existence. When a basal shear stress is present, the distance out to which a channel can collect water is not that much greater than the channel radius itself. It is concluded that it is not likely for the channel to be able to collect appreciable amounts of water that is produced at the bed. Hence despite the indication that an instability initially might grow in a water sheet, this instability either cannot develop to the point when the water flow is primarily in channels or else the water flow alternates cyclically between a state primarily of sheet flow and a state of channel flow. It would appear that the channels that do exist under a glacier have their primary origin in the melt water from the upper surface that pours down moulins and thus is already channelized by the time it reaches the bed.

The first extensive and far-reaching investigations of the Sino-German Joint Expedition to the Ch’ing-hai–Hsi-tsang (Qinghai–Xizang) Plateau in the glacierized region of the Amne Machin (Anyemaqen) started work in the area and compiled the glacial inventory in that region. In the past 15 years scientists have found that the majority of glaciers in the Amne Machin (Anyemaqen) have been advancing whereas most glaciers in Ch’i-lien Shan (Qilian Shan), adjacent to Amne Machin (Anyemaqen), have been retreating during the same period. This is a problem worthy of study by glaciologists. Through investigation and analysis, it is also found that Ha-lung (Halong) glacier has the character of a surging glacier. The author tentatively forecasts the next surge of Ha-lung (Halong) glacier and the tendency of variations of glaciers in Ch’i-lien Shan (Qilian Shan) in the near future.

Observations obtained in a long-term monitoring program on Lewis Glacier, Mount Kenya, include measurements of net balance and velocity at a stake network over the period 1978–82, and airborne mapping of the glacier surface topography at the beginning and end of this interval. Horizontal and vertical flow components and mass continuity are evaluated along the longitudinal axis of the glacier.

The largest horizontal velocity at the surface is about 5 m a−1, while the submergent vertical velocity at the surface in the upper, and the emergent component in the lower glacier are of the order of 1–2 m a−1. Horizontal and vertical flow components in the interior of the glacier are calculated from the surface observations under plausible assumptions. Isotachs and streamlines plotted in a longitudinal–vertical cross-section permit the estimation of representative residence times of ice within Lewis Glacier. The largest values for the longest trajectories with submergence at the surface in the uppermost area of the glacier and emergence near the snout are of the order of a few centuries. However, such estimates are compromised, in particular, by the continuing non-equilibrium conditions of Lewis Glacier.

Mass continuity is directly related to the vertical motion field. In the upper area of the glacier the submergence velocity exceeds the rise of glacier surface observed with respect to the stake network, while in the lower glacier the emergent flow component falls short of the local negative net balance. This differing behavior of altitudinal zones is consistent with the fact that actual lowering of the glacier surface topography documented by the repeated airborne mappings is larger than expected from the stake network in the upper but less in the lower glacier. This differing behavior of altitude zones appears mitigated by the tendency for confluence in the upper and diffluence in the lower glacier.

A comprehensive series of dye tracer tests was undertaken over two summers at Pasterzengletscher, Austria, and from these a number of important conclusions can be drawn. A lateral division of drainage on the lower tongue was proved, but this was not determined by the direction of oblique crevasses. Water from many moulins passed into two or more drainage courses, as shown by repeated patterns of multiple peaks on the time/concentration curves. The spatial pattern of travel times from different moulins on the tongue was complex, although it remained stable over time. Repeated tracer tests showed that throughflow velocities varied by up to an order of magnitude during a single day. Maximum velocities occurred at the time of peak discharge, and probably the large diurnal velocity variations are caused by changes in the proportion of the length which has channel-full pressure flow. However, seasonal velocity variations were much smaller, although changes in outflow discharge were larger than in the diurnal case. This gives strong support to the concept that channels adjust in size to the average water discharge passing through, and that these changes occur over a matter of days or a few weeks.

18O and deuterium content was measured in snow and ice samples collected from two crevasses on temperate glaciers high in the Andes and in the Himalaya. The results show that, due to intense exposure to sun, a significant part of the annual accumulation of snow is removed by melting and sublimation, the latter being predominant. There is strong empirical evidence that in natural conditions which are characteristic of both the glaciers studied sublimation of the snow cover can be approximated by the thermodynamic equilibrium process. The short-term fluctuations, though slightly moderated, and longer-term variations can still be observed in the deeper parts of the glaciers in spite of firnification of the snow.

The coefficient of thermal expansion of NaCl ice and natural sea ice is theoretically shown to be equal to the coefficient of thermal expansion of pure ice.

The method of Brillouin spectroscopy has been used to measure the dynamic elastic moduli of local homogeneous regions in ice samples representing four different environments of formation. These included artificial ice frozen from distilled water, clear monocrystalline glacial ice, bubbly lake ice, and sea ice. The samples studied were found to have identical local elastic properties. Accordingly the elastic properties of homogeneous monocrystalline ice have been found not to vary with sample age, with impurities present at the time of freezing, or with crystal quality. The bulk elastic properties of ice remain, of course, subject to modification by different crystal grain textures and the presence of inclusions of various sorts. Because the elastic constants obtained in the present work are subject to smaller overall uncertainty than values measured previously, it is believed that they are the most reliable obtained to date. The values at −16 °C were determined to be c11 = 139.29 ± 0.41, c12 = 70.82 ± 0.39, c13 = 57.65 ± 0.23, c33 = 150.10 ± 0.46, c44 = 30.14 ± 0.11 (units of 108 N m− 2 or kbar). A full range of derived elastic parameters for monocrystalline ice and for homogeneous isotropic polycrystalline ice has been calculated.

Drumlin forms are described from maps and air photographs of a part of the Athabasca Plains, northern Saskatchewan. Three major forms, spindle, parabolic and transverse asymmetrical are recognized. These forms, which may show superimposed minor elements, depart from classical descriptions of drumlins, but are similar to moulds of erosional marks created by separated fluid flows. Assemblages of drumlins also show characteristics similar to those of erosional marks. The form analogy between drumlins and moulds of erosional marks is carried to a conclusion that drumlins may be formed by the infilling of erosional marks created on the under-side of glaciers by separated, subglacial melt-water flows. Estimates of specific discharge are obtained by means of an expected range of Reynolds number. Geomorphological evidence is given for large-scale erosion by subglacial melt water. A discussion of the sedimentology, stratigraphy, and deformational structure of the interiors of drumlins shows that they may be explained by the erosional-mark hypothesis. This paper emphasizes the importance of melt water as a geomorphic agent and may have broad implications for ice-sheet dynamics and profiles, rates of deglaciation, and the occurrence of bedrock thrusting by ice.

In 1890–91 John Hardcastle presented his views on glacier motion and the climatic significance of loess deposits. He should be recognized as a significant pioneer in the study of glacier-flow mechanisms, and more particularly, as the initiator of the study of loess stratigraphy and the multiple-event approach to the formation of loess deposits, especially glacial loess.

The shear frame is a simple in situ device for indexing the shear strength of thin weak layers. The index is sensitive to shear-frame geometry, rate-of-pull, and shear-frame mass. It is time-consuming to carefully align the device on the Gleitschicht (shear failure plane) in a slab avalanche zone. The ratio shear frame index/shear stress of the Gleitschicht has a high variance, and may not be a fundamental measure of slab avalanche stability. Corrections for the normal stress on the Gleitschicht reduce the variance only slightly. Despite these limitations, the shear frame is a useful tool for gathering statistical data on strength distributions and anisotropies of the Gleitschicht until a more fundamental technique is developed.

The flow of glacier ice is mapped using high-resolution photography and non-coherent-light speckle interferometry. Young’s fringe patterns result when a double-exposed photoplate image of the straining surface is illuminated by a narrow beam of coherent light. Geometry gives a relationship between the ice surface displacement vector and the interference fringe patterns. This displacement vector is corrected for rigid-body (camera) movement and projected onto the ice surface using topological maps. The strain during the time-lapse interval is thus known. Comparison with data acquired by surveying techniques at Nisqually Glacier, Washington, U.S.A., is limited because of small overlap of the surface studied. In the areas for which results can be compared, our experiments yield a flow of 0.6 m/d where conventional methods yield about 0.4 m/d.

Studies were made of the correlation between the crystallographic axes and the shape of an individual single crystal of a crystalline aggregate in a temperate glacier and an Antarctic deep core, both of which have the typical fabric pattern of the four-maxima type, by approximating the crystal shape to an ellipsoid and then measuring misorientations between the axes of the ellipsoid and the crystallographic axes of the crystal. The result shows that the crystallographic axes are correlated with the axes of the ellipsoid; that is, in most cases the longest and the shortest axis of the ellipsoid are coincident with one of the a-axes and the c-axis of a crystal, respectively.

The mass concentration of falling snow ρs can be estimated from the snow-fall rate (accumulation rate) qv if there is no significant wind. Limited data show only a weak relation between fall velocity ut and qv (with qv in g/cm2 h). Consequently there is a strong correlation (r2 = 0.97) between ρs and qv (with qv in g/cm2 h). A simple relation of this kind is of practical value for certain technical purposes, and more data would be welcome.

Decisions on snow stability are strongly influenced by the presence of fresh avalanches. An analysis of 18 pairs of controlled avalanche paths in Colorado indicates most behave independently. The study supports the hypothesis that spatial variations in slope stability are common in unstable snow.

In the Nahoni Range of the northern Ogilvie Mountains, Yukon Territory, Canada, dissection channels are most markedly developed on valley sides facing the north and north-east. The extent of slope channelling is controlled by intensity of insolation and its influence on persistence of snow and ice. Upon the cooler north- and north-east-facing slopes, snow and ice remain well into midsummer; melt water is released over a long period and its discharge is restricted to narrow rivulets. In contrast, snow and ice on the warmer west- and south-facing slopes melt rapidly during the late spring months. The scree on these valley sides moves down-slope as a more or less uniform mass of water-saturated detritus.

The upper part of a Permo-Carboniferous glacial valley fill along the northern margin of the Karoo Basin includes glacio-lacustrine sediments. During the last glacier advance into the lake, a bedded heterogeneous diamictite facies was deposited and, on glacier retreat, a sequence of deformed siltstones with diamictite lenses and sandstone beds, varved shale and rhythmite shale was laid down. Black carbonaceous mud was deposited during the subsequent marine transgression. According to varve counts, the glacier receded from the valley over a period of 500 to 1 000 years and it is concluded that the overall ice-retreat rate during the Permo-Carboniferous deglaciation was relatively high.